The Latin Dialect of The Ager Faliscus (Faliscan)

The Latin Dialect of the Ager Faliscus
150 Years of Scholarship

The publication of this book is made possible by a grant from the Machtelt Bolkestein
Fonds.
Lay out: Gabriël C.L.M. Bakkum, Amsterdam

Cover design: René Staelenberg, Amsterdam
ISBN 978 90 5629 562 2

e-ISBN 978 90 4850 875 4
NUR 630
© G.C.L.M. Bakkum / Vossiuspers UvA – Amsterdam University Press, 2009
All rights reserved. Without limiting the rights under copyright reserved above, no part of this
book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any
form or by any means (electronic, mechanical, photocopying, recording or otherwise) without
the written permission of both the copyright owner and the author of the book.
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam,
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het College voor Promoties ingestelde
commissie, in het openbaar te verdedigen in de Agnietenkapel
op dinsdag 10 februari 2009, te 14.00 uur
door
Gabriël Cornelis Leonides Maria Bakkum

geboren te Amsterdam
Promotiecommissie:
Promotor: prof. em. dr. H. Pinkster
Overige leden: prof. dr. Ph. Baldi

prof. dr. M. Gnade
prof. dr. C. Kroon
prof. dr. P.Poccetti
dr. R. Risselada
Faculteit der Geesteswetenschappen

—————
MEMORIÆ DVLCISSIMÆ
INÆ VAN CAMERIJK

AMICÆ DILECTISSIMÆ
DICATVM
—————
CONIVGIO · NOSTRO · SPATIVM · BREVE · CONTIGIT · ÆVI

NON · CAPIVNT · LONGAS · GAVDIA · MAGNA · MORAS
—————
Copyright notice with regard to fonts:
For the Italic, Etruscan, Phoenician, and IPA fonts, as well as for various other symbols and
signs, I have made licensed use of the Alphabetum© font package created by J.J. Marcos of
the University of Plasencia, Spain.
For the Greek font, I have used the shareware version of the Greek Old Face© font created
by R. Hancock and the Son of WinGreek© keyboard program by N. Beshoori.
Contents
Contents I
Preface IX
Bibliography XI
1. Corpora, catalogues, and other publications referred to by sigla XI
2. Publications referred to by author XII
3. A note on sources and editions LI
4. Conventions in the representation of epigraphic texts LII
Chapter 1. Introduction 1
1.1. Introductory remarks 1
1.2. Dialect and language 3
1.3. Languages in contact 6
1.3.1. Language contact in ancient Italy 6
1.3.2. Interference and borrowing 7
1.4. The Faliscan material 9
1.4.1. ‘Faliscan’, ‘Latin’, ‘Sabellic’ 9
1.4.2. The material 10
1.4.3. The dating of the material 11
1.4.4. When is a text Faliscan? 12
1.4.5. The reliability of the material 13
1.5. A short survey of Faliscan studies: 150 years of scholarship 14
Map of the ager Faliscus 18
Chapter 2. The ager Faliscus and its inhabitants 19
2.1. The ager Faliscus 19
2.1.1. Extent 19
2.1.2. Towns 20
2.1.3. Roads and routes 22
2.2. The inhabitants of the ager Faliscus as an ‡dion œqnoj 24
2.2.1. A distinct people 24
2.2.2. The Faliscans as an ethnos 25
2.3. What constituted a ‘Faliscan identity’? 27
2.3.1. Faliscan material culture 27
2.3.2. Faliscan society 28
2.3.3. Faliscan magistracies 30
2.3.4. Faliscan religion 32
2.4. The early history of the ager Faliscus 35
2.4.1. Mythical origins 35
2.4.2. The history of the ager Faliscus until the fifth century BCE 35
2.5. Falerii, Veii, and Rome in the fifth and fourth centuries BCE 37
2.5.1. The Fidenate wars and the siege of Veii 37
2.5.2. The wars of the fourth and early third century BCE 39
2.6. The war of 241 BCE and its consequences 41
2.6.1. The war of 241 BCE 41
I
CONTENTS
2.6.2. The division of the territory and the founding of Falerii Novi 42
2.6.3. The ager Faliscus after 241 BCE 44
2.7. Sociolinguistic factors influencing language preservation or loss 45
2.8. Linguistic evaluation: Faliscan and its neighbours 49
2.8.1. Faliscan and Etruscan 49
2.8.2. Faliscan and the Sabellic languages 50
2.8.3. Faliscan and Latin 50
Chapter 3. Phonology 53
3.1. Methodological issues 53
3.1.1. Method and material 53
3.1.2. The concept of Proto-Italic 54
3.1.2. The PIE phonemic inventory 55
3.2. Proto-Italic developments 57
3.2.1. From Proto-Italic to Proto-Latin 57
3.2.2. Presumed merger of the palatal and the velar series 57
3.2.3. Developments of the laryngeals 58
3.2.4. Voicing of /t#/ and subsequent drop of /i#/ 61
3.2.5. Merger of */e / with */o / 61
3.2.6. Drop of intervocalic // and subsequent contractions 63
3.2.7. The developments of */ / 64
3.2.8. Developments of the voiced aspirates (I): Proto-Italic 64
3.2.9. Proto-Italic voicing of intervocalic /s/? 68
3.2.10. */p...k / → */k ...k / 69
3.3. Proto-Latin developments 70
3.3.1. Proto-Latin 70
3.3.2. The development of */ / 70
3.3.3. Developments of the voiced aspirates (II): Proto-Latin and Faliscan 71
3.3.4. Other possibly Proto-Latin developments 76
3.4. The Faliscan phonemic system 77
3.5. The Faliscan consonants 78
3.5.1. The consonant system of Faliscan 78
3.5.2. The Middle Faliscan development /#fV/ → /#hV/ 79
3.5.3. Rhotacism and allophones of /s/ 83
3.5.4. /t/ realized as [d]? 86
3.5.5. Palatalizations 86
3.5.6. Minor developments 88
3.5.7. Omission of syllable- and word-final consonants 89
3.6. The Faliscan vowels 95
3.6.1. The vowel system of Faliscan 95
3.6.2. Middle and Late Faliscan variation of i and e 97
3.6.3. Middle and Late Faliscan variation of u and o 98
3.6.4. Variation between i and u 100
3.6.5. Anaptyxis 100
3.6.6. Developments of vowels due to stress 100
3.7. The Faliscan diphthongs 104
II
CONTENTS
3.7.1. General 104

3.7.2. The development of /o / 104
3.7.3. The development */lo β-/ → /loβ-/ 105
3.7.4. The development of /a / 106
3.7.5. The developments of /e/, /o/, and /ō/ 107
3.7.6. The developments of /a/ and /ā/ 109
3.7.7. A diphthong /�/? 114
3.8. Summary of §§3.2-7 115
Chapter 4. The nominal and pronominal inflections 117
4.1. The nominal and pronominal inflections: methodological issues 117
4.2. The first nominal declension 118
4.2.1. The first-declension nominative singular 118
4.2.2. The first-declension genitive singular 120
4.2.3. The first-declension dative singular 122
4.2.4. The first-declension accusative singular 123
4.2.5. The first-declension ablative singular 124
4.2.6. The first-declension nominative plural 124
4.3. The second nominal declension 124
4.3.1. The second-declension masculine nominative singular 124
4.3.2. The second-declension dative singular 126
4.3.3. The second-declension accusative and neuter nominative singular 127
4.3.4. The second-declension ablative singular 127
4.3.5. The second-declension vocative singular 128
4.3.6. The second-declension nominative plural 128
4.3.7. The second-declension genitive plural 128
4.4. The second-declension genitive singular 129
4.4.1. The problem 129
4.4.2. The Early Faliscan genitive in -osio 129
4.4.3. The Middle and Late Faliscan genitive in -i 130
4.4.4. The Middle and Late Faliscan genitive in -oi 131
4.4.5. The alleged Middle and Late Faliscan genitives in -io and -oio 132
4.4.6. The second-declension genitive singular in Latin 132
4.4.7. The second-declension genitive singular in the Sabellic languages 134
4.4.8. The second-declension genitive singular in other languages of ancient Italy 135
4.4.9. The relation between /-oso/ and /-ī/ 136
4.4.10. The ending /-ī/ as a phonological development of /-oso/ 137
4.4.11. The ending /-ī/ as a reflex of PIE */-iĦ/ (*/-iħ2/?) 138
4.5. The third nominal declension 141
4.5.1. The nominative singular of the consonant- and i-stems 141
4.5.2. The third-declension genitive singular 143
4.5.3. Other consonant- and i-stem endings 145
4.6. The fourth and fifth nominal declensions 146
4.6.1. The fourth-declension genitive singular 146
4.6.2. The fourth-declension dative singular 146
4.6.3. The fourth-declension accusative singular 147
III
CONTENTS
4.6.4. The fifth declension 147

4.7. The personal pronouns 148
4.7.1. The nominative of the first person singular 148
4.7.2. The accusative of the first person singular 148
4.7.3. The nominative of the second person plural 150
4.8. The demonstrative pronoun 151
4.9. The relative pronoun 152
4.10. Summary of §§4.2-9 152
Chapter 5. The verb 155
5.1. The verb: methodological issues 155
5.2. The Faliscan verb 155
5.2.1. Conjugational system 155
5.2.2. Formation of the tenses 156
5.2.3. Formation of the subjunctive 157
5.2.4. The endings 157
5.3. The attested Faliscan verbs 161
5.3.1. Verb forms 161
5.3.2. Nominal forms 173
5.4. Summary of §§5.2-3 174
Chapter 6. The lexicon 177
6.1. The lexicon: methodological issues 177
6.2. The epigraphically attested lexicon 179
6.3. The Faliscan lexicon and its Latin and Sabellic equivalents 200
6.3.1. The Faliscan lexicon in comparison 200
6.3.2. Lexical subsets 201
6.4. Theonyms 202
6.5. Toponyms, potamonyms, and ethnonyms 205
6.6. Faliscan glosses 209
Chapter 7. The onomasticon 213
7.1. The onomasticon: methodological issues 213
7.1.1. Names and language 213
7.1.2. The problem of abbreviations 215
7.2. Names in the Early Faliscan inscriptions 216
7.2.1. The Early Faliscan onomastic formulas 216
7.2.2. The names attested in the Early Faliscan inscriptions 218
7.3. Middle and Late Faliscan male onomastic formulas 220
7.4. Female onomastic formulas 227
7.4.1. The formula of women’s names 227
7.4.2. Adding the husband’s name: the marital formula 230
7.5. The formula of filiation 231
7.5.1. FILIATION 231
7.5.2. FATHERGEN [SON/DAUGHTER] and the patronymic adjective 232
7.6. The names of freedmen and freedwomen 236
7.7. The praenomina 237
IV
CONTENTS
7.7.1. The praenomina attested from the Middle Faliscan period onward 237
7.7.2. The origins of the Faliscan praenomina 254
7.7.3. Types of Faliscan praenomina 256
7.8. The gentilicia 257
7.8.1. The gentilicia attested from the Middle Faliscan period onward 257
7.8.2. The origins of the Faliscan gentilicia 277
7.9. The cognomina 280
7.9.1. Attestations of cognomina in the Faliscan inscriptions 280
7.9.2. The chronology of the Faliscan cognomen 282
7.10. The Faliscan onomasticon and the question of identity 283
7.10.1. Names as markers of identity 283
7.10.2. The onomastic formula and social status 284
7.10.3. Gentilicia and the question of ethnic origin 285
7.10.4. Praenomina and the family 286
7.10.5. Praenomina and ethnic identification 288
Chapter 8. Syntax and text structure 291
8.1. Syntax and text structure: methodological issues 291
8.2. Nominal morphosyntax 292
8.2.1. Cases and numbers 292
8.2.2. Syntactic functions of the cases 293
8.2.3. The nominative and the genitive in isolation 294
8.2.4. Elliptic sentences, isolated datives, and isolated accusatives 296
8.3. Verbal morphosyntax 297
8.4. Constituent order 299
8.4.1. The order of the arguments 299
8.4.2. The position of adjuncts 300
8.5. Word order within word groups 301
8.5.1. The position of the adjective 301
8.5.2. The position of the attributive genitive 301
8.5.3. The position of appositions 302
8.5.4. Prepositional phrases 303
8.6. Coordination and subordination 303
8.7. Word order typology 304
8.8. Besitzerinschriften 305
8.8.1. OWNERNOM, OWNERGEN, RECIPIENTDAT 305
8.8.2. Besitzerinschriften of the iscrizioni parlanti-type 306
8.9. Signatures 308
8.9.1. MAKERNOM, MAKERGEN 308
8.9.2. Signatures of the iscrizioni parlanti-type 309
8.10. Sepulchral inscriptions 310
8.10.1. DECEASEDNOM and DECEASEDNOM hec cupat 310
8.10.2. ‘X [made this grave] for Y’ 312
8.10.3. OWNERGEN cella and the ius sepulcrale 313
8.11. Dedications 314
8.11.1. OWNERNOM, OWNERGEN, RECIPIENTDAT 314
V
CONTENTS
8.11.2. Other types of dedicatory inscriptions 316

8.12. Official inscriptions 317
8.13. Summary of §§8.8-12 318
Chapter 9. Language contact 319
9.1. Language contact in the ager Faliscus: methodological issues 319
9.2. Faliscan and Etruscan 322
9.2.1. The nature of Faliscan-Etruscan language contact 322
9.2.2. Etruscan features in Faliscan inscriptions 323
9.2.3. Etruscan-speaking groups in the ager Faliscus 328
9.3. Faliscan and the Sabellic languages 330
9.3.1. The nature of Faliscan-Sabellic language contact 330
9.3.2. Sabellic features in Faliscan and Capenate inscriptions 331
9.4. Faliscan and Latin 334
9.4.1. The nature of Faliscan-Latin language contact 334
9.4.2. Latin features in Faliscan inscriptions 336
9.5. Summary of §§9.2-4 339
9.6. A note on Faliscan outside the ager Faliscus 340
Chapter 10. Conclusion: Faliscan as a Latin dialect 341
10.1. Language or dialect? 341
10.1.1. Drawing conclusions 341
10.1.2. The sociolinguistic definition 341
10.1.3. Synchronic comparison 343
10.1.4. The diachronic perspective 346
10.2. Recent views on the position of Faliscan 350
10.2.1. Recent views 350
10.2.2. Campanile: an independent Faliscan 350
10.2.3. Solta: convergence with the Sabellic languages? 352
10.2.4. G. Giacomelli: a Latin Faliscan with Sabellic influence 353
10.2.5. Joseph & Wallace: not a Latin dialect 355
10.2.5. R. Giacomelli: level-distinction 357
10.3. Faliscan, a Latin dialect 358
Chapter 11. The epigraphical material 361
11.1. The Faliscan epigraphical material 361
11.1.1. General 361
11.1.2. The material divided according to provenance 362
11.1.3. The material divided according to period/alphabet category 364
11.1.4. The material divided according to type 368
11.1.5. The material divided according to alphabet 375
11.2. Alphabet and orthography 376
11.2.1. The Faliscan alphabet 376
11.2.2. The origins of the Faliscan alphabet 376
11.2.3. Alphabet and orthography of the Early Faliscan inscriptions 380
11.2.4. The alphabet of the Middle and Late Faliscan inscriptions 383
11.3. The use of the Etruscan and Latin alphabets 388
VI
CONTENTS
11.4. A note on the presentation of the inscriptions 398

Chapter 12. The Early Faliscan inscriptions 393
12.1. The Early Faliscan inscriptions as a group 393
12.2. The ‘Ceres-inscription’ [nr. 1] 393
12.3. The two impasto pitchers [nrs. 2-4] 406
12.4. The other Early Faliscan inscriptions [nrs. 5-10] 414
Chapter 13. The inscriptions from Civita Castellana (Falerii Veteres) I 417
13.1. Civita Castellana and its inscriptions 417
13.2. The necropolis of Contrada Le Colonnette [nrs. 11-19] 417
13.3. The necropolis of Contrada Celle [nrs. 20-40] 422
13.4. The Valsiarosa necropolis [nrs. 41-46] 427
13.5. The La Penna necropolis [nrs. 47-65] 430
13.6. The necropolis of Ponte Lepre [nrs. 66-78 437
13.7. The tombs near the Ponte Terrano [nrs. 79-89] 440
13.8. The tomb near Torrente Purgatorio [nrs. 90-108] 447
Chapter 14. The inscriptions from Civita Castellana (Falerii Veteres) II 457
14.1. The inscriptions from the temples 457
14.1.1. The dedications 457
14.1.2. The temples of Colle di Vignale [nrs. 109-112] 457
14.1.3. The temple ruins ‘ai Sassi Caduti’ [nrs. 113-131] 459
14.1.4. The temple in Contrada Celle [nr. 132] 465
14.1.5. The temple of Lo Scasato [nrs. 133-134] 466
14.2. Civita Castellana, origin unknown 466
14.2.1. Inscriptions on tiles containing identifiable gentilicia [nrs. 135-157] 467
14.2.2. Inscriptions on tiles containing identifiable parts of texts [nrs. 158-169] 477
14.2.3. Tiles inscribed in the Latin alphabet [nrs. 170-174] 481
14.2.4. Tiles containing only fragments of texts [nrs. 175-194] 482
14.2.5. Loculus-inscriptions [nrs. 195-198] 486
14.2.6. Inscriptions on pottery [nrs. 199-204] 487
14.3. Inscriptions from the surroundings of Civita Castellana [nrs. 205-212] 489
Chapter 15. The inscriptions from S. Maria di Falleri (Falerii Novi) 493
15.1. Falerii Novi 493
15.2. Inscriptions from the town [nrs. 213-219] 493
15.3. Inscriptions from near the town [nrs. 220-252] 500
Chapter 16. Corchiano and the northern ager Faliscus 519
16.1. The sites of the northern ager Faliscus 519
16.2. Corchiano and surroundings 519
16.2.1. Corchiano 519
16.2.2. The necropoles of the Il Vallone plateau [nrs. 253-271] 520
16.2.3. The Rio Fratta necropolis [nrs. 272-274] 526
16.2.4. Corchiano, specific provenance unknown [nrs. 275-284] 527
16.2.5. The surroundings of Corchiano [nrs. 285-301] 529
16.3. Vignanello [nrs. 302-323] 534
16.4. Fabbrica di Roma [nrs. 324-328] 540
VII
CONTENTS
16.5. Carbognano-Vallerano [nrs. 329-337] 542

16.6. The site at Grotta Porciosa [nrs. 338-355] 546
16.7. The area around Gallese and Borghetto [nrs. 356-359] 552
16.8. Northern ager Faliscus, exact provenance unknown [nrs. 360-362] 553
Chapter 17. The south-eastern ager Faliscus and the ager Capenas 555
17.1. The inscriptions from the south-east 555
17.2. Rignano Flaminio [nrs. 363-375] 555
17.3. S. Oreste [nr. 376] 558
17.4. Ponzano Romano [nr. 377] 558
17.5. Civitella S. Paolo [nrs. 378-384] 559
17.6. Fiano Romano [nr. 385] 561
17.7. Civitucola (ancient Capena) [nrs. 386-430] 561
17.8. The shrine of Lucus Feroniae [nrs. 431-456] 570
17.9. Ager Capenas, provenance unknown [nrs. 457-466] 574
Chapter 18. Inscriptions claimed as Faliscan or Capenate 577
18.1. The inscriptions claimed as Faliscan or Capenate 577
18.2. Inscriptions that are probably or possibly Faliscan or Capenate [nrs.467*-478*] 577
18.3. Inscriptions that are probably not Faliscan or Capenate 583
18.3.1. Inscriptiones originis incertae uel ignotae [nrs. 479†-481†] 583
18.3.2. The ‘Faliscan’ inscriptions from Ardea [nrs. 482†-484†] 586
18.3.3. The Satricum-inscription as Faliscan 590
Chapter 19. The Etruscan inscriptions 593
19.1. The presentation of the Etruscan inscriptions 593
19.2. Narce and the south-western ager Faliscus [nrs. I-XXIV] 593
19.3. Civita Castellana (Falerii Veteres) [nrs. XXV-XXXI] 600
19.4. Corchiano and the northern ager Faliscus [nrs. XXXII-XLII] 602
19.5. The south-eastern ager Faliscus and the ager Capenas [nrs. XLIII-XLV] 605
19.6. Etruscan inscriptions incertae uel ignotae originis [nrs. XLVI-LI] 607
Dutch summary (Nederlandse samenvatting) 609
Concordances 631
1. General 631
2. Giacomelli, La lingua falisca 642
3. Vetter, Handbuch der italischen Dialekte 644
4. Herbig, Corpus inscriptionum Etruscarum II.1.2 646
5. Rix, Etruskische Texte 649
6. Corpus inscriptionum Latinarum I2 650
7. Corpus inscriptionum Latinarum XI 650
8. First editions of the inscriptions published since 1963 651
Index of sources 653
1. Literary sources 653
2. Inscriptions in my edition 655
3. Other inscriptions 669
VIII
Preface
In the spring of 1987, it was decided that Faliscan would become the subject of my research as
Assistent in Opleiding, mainly for the simple reasons that Faliscan was a subject that could be
treated within the range of a Ph.D. research and that no major work on it had appeared for some
time – for 24 years, to be precise. At first, I was only too glad too agree with the phrase of my
predecessor Gabriella Giacomelli that “Faliscan is not an attractive language”: only after some
time I discovered that I regarded Faliscan neither as unattractive, nor as a language. It took me
several years as well as repeated visits to the epigraphic material in Rome, the Vatican, and
Viterbo, to elaborate why this was so.
Little did I suspect then that the untimely death of my own socia kara would lead to a
period of twelve years during which work on the project would cease completely and everyone
but myself assumed that it would never be resumed, let alone finished. Indeed, when the work
was at long last resumed in the autumn of 2007, part of it survived only as a print-out and as files
on some old 5.25″ floppy discs that were retrieved only with some difficulty. Neither did I
suspect that when the work finally was finished, still no major work on Faliscan had appeared in
the meantime – for 45 years, the span of my own lifetime.
Now, 21 years after its beginning, the work is completed: the time it took Ulysses to marry
Penelope and beget Telemachus, fight the war at Troy, and return home through many adventures
– always wishing, but hardly expecting, to see the day of his homecoming. And, like Ulysses’
companions, many of mine have not lived to see that welcome end: I name only my dearly
beloved girlfriend and companion drs G.A. van Camerijk (†1994), my esteemed teachers, prof.
dr A.M. Bolkestein of the Universiteit van Amsterdam (†2001), prof. dr S.L. Slings of the Vrije
Universiteit, Amsterdam (†2004), prof. em. dr C.J. Ruijgh of the Universiteit van Amsterdam
(†2004), and my mother, C.C.M. Bakkum-Spaan (†2006).
If this research has finally been brought to a conclusion, this is the result of other minds and
hands as well as mine. I wish to thank them here for all their help and assistance:
● First and foremost, prof. em. dr Harm Pinkster (Universiteit van Amsterdam), for all his help
and suggestions, for his confidence in my abilities – namque tu solebas meas esse aliquid putare
nugas –, and for his infinite patience with my erratic style of working. It was due to his tutoring
that my interest in the languages of ancient Italy was awoken, and without him the writing of this
thesis would never have been possible.
I also wish to thank:

● the Universiteit van Amsterdam, Faculteit der Letteren (now Faculteit der Geestes-
wetenschappen) and the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO) for
grants that enabled me to visit the Colloquia on Latin Linguistics, the 1991 Arbeitstagung of the
Indogermanische Gesellschaft, the Nyelvtudomanyos Interzete in Budapest, and the Nederlands
Instituut te Rome; and the Machtelt Bolkestein Fonds for a grant towards the expenses of
publishing this thesis;
IX
PREFACE
● the staff of the Nederlands Instituut te Rome, especially dr Th. Meijer † (then director) and dr
Th. Heeres (then head of the Archaeological Department), for facilitating my research;
● dr P. Pelagatti (then Soprintendente per l’Etruria Meridionale), dr F. Buranelli (then Museo

Gregoriano Etrusco Profano, Città del Vaticano), dr F. Poleggi (then Museo archeologico
dell’agro falisco, Civita Castellana), dr A. Emiliozzi (then Museo Civico, Viterbo), for allowing
me access to the collections in their custody;
● prof. dr Ph. Baldi (Pennsylvania State University), prof. dr A.L. Prosdocimi (Università di
Padova), and prof. dr R. Coleman † (then Emmanuel College, Cambridge), whose interest in my
work has been a greater stimulus than they may have realized;
● prof. em. dr G.J.M.J. te Riele, prof. em. dr F. Waanders, and dr J.G.J. Abbenes, for many
helpful suggestions, discussions, and reading of first drafts;
● prof. dr B.D. Joseph (Ohio State University), for sending me his then unpublished article On
the problematic f/h variation in Faliscan, prof. dr P. Poccetti (Università di Roma II Tor Vergata),
for sending me his unpublished article Il vaso iscritto della necropoli di Magliano Sabina, and dr
Jane Stuart-Smith (Oxford University), for sending me her then unpublished thesis The
development of the ‘voiced aspirates’ in Italic;
● prof. dr C. Kroon (now Vrije Universiteit, Amsterdam) and dr R. Risselada (Universiteit van
Amsterdam), for having to endure all my thinking very much aloud during the early years;
● dr W. de Melo (All Souls College, Oxford), for the correction of my often deplorable English,
and for a great number of helpful suggestions and references;
● dr C. Ewigleben (Museum für Kunst und Gewerbe, Hamburg), for answering my inquiries
with regard to the Capenate inscription 475*;
● dr E. van Opstall (now Vrije Universiteit, Amsterdam), for procuring a copy of Civiltà dei
Falisci, dr A. Bertocchi (Università degli Studi, Bologna), for supplying me with photocopies of
Buonamici’s Il dialetto falisco, and drs E. Drenth, for presenting me with a copy of Potter’s The
changing landscape of South Etruria;
● dr R.S. Brink, my old Latin tutor, for correcting the Russian passages, and drs H. Haabraken,
for translating Safarewicz’s article Dwa studia nad językiem Falisków;
● drs R. Elstrodt, for recuperating the ‘lost files’ from the old 5.25″ floppy disks, and J. Heiligers
for his work on the conversions of the texts to PDF format;
● drs M. van der Horst and drs R. Heiligers for their support and help during the final stages;
● and, finally, drs G.A. van Camerijk †, to whom this thesis is dedicated with the greatest love
and respect, for more things than could possibly be named.
Any errors that this work contains are of course entirely my own.
Amsterdam, December 2008
X
Bibliography
1. CORPORA, CATALOGUES, AND OTHER PUBLICATIONS REFERRED TO BY SIGLA

ANRW I2 H. TEMPORINI (ed.), Aufstieg und Niedergang der römischen Welt: Geschichte und
Kultur Roms im Spiegel der neueren Forschung I: Von den Anfängen Roms bis zum
Ausgang der Republik, fasc. 2, Berlin 1972.
AntFal A.M. MORETTI SGUBINI (coord.), C. CARLUCCI, & M.A. DE LUCIA (text), Villa Giulia
Museum: The antiquities of the Faliscans, Roma 1998.
BarrAtl R.J.A. TALBOT, Barrington Atlas of the Greek and Roman World, Princeton 2000.
CIE O.A. DANIELSSON, G. HERBIG, & B. NOGARA, Corpus inscriptionum Etruscarum,
volumen II, sectio II, fasciculus 1: Ager Faliscus et Capenas, Lipsiae 1912 (=
G. HERBIG, Tituli Faleriorum veterum, linguis Falisca et Etrusca conscripti, diss.
Lipsiae 1910).
CII A. FABRETTI, Corpus inscriptionum Italicarum antiquioris aevi ordine geographico
digestum, et Glossarium Italicum in quo omnia vocabula continentur ex Umbricis
Sabinis Oscis Volscis Etruscis aliisque monumentis quae supersunt collecta et cum
interpretationibus variorum explicantur, Augustae Taurinorum 1867; Primo sup-
plemento alla raccolta delle antichissime iscrizioni italiche, con l’aggiunta di al-
cune osservazioni paleografiche e grammaticali, Torino 1872.
CIL I Th. MOMMSEN & W. HENZEN, Corpus inscriptionum Latinarum I: Inscriptiones
Latinae antiquissimae ad C. Caesaris mortem, Berolini 1863.
CIL I2 E. LOMMATZSCH, Corpus inscriptionum Latinarum I, editio altera: Inscriptiones
antiquissimae ad C. Caesaris mortem, pars posterior, fasciculus 1, Berolini 1918;
fasciculus 2: Addenda, nummi, indices, ed. E. LOMMATZSCH & H. DESSAU, Bero-
lini 1931; fasciculus 3: Addenda altera, indices, Berolini 1943; fasciculus 4: Ad-
denda tertia I: Textus, fasciculus 5: Addenda tertia II: Tabulae, ed. A. DEGRASSI &
J. KRUMMREY, Berolini / Novi Eboraci 1986.
CIL X Th. MOMMSEN, Corpus inscriptionum Latinarum X: Inscriptiones Bruttiorum Lu-
caniae Campaniae Siciliae Sardiniae Latinae: pars 1, Inscriptiones Bruttiorum Lu-
caniae Campaniae, pars 2, Inscriptiones Siciliae et Sardiniae, Berolini 1883.
CIL XI E. BORMANN, Corpus inscriptionum Latinarum XI: Inscriptiones Aemiliae Etruriae
Umbriae Latinae, pars prior: Inscriptiones Aemiliae et Etruriae, Berolini 1888; pars
posterior, fasciculus 1: Inscriptiones Umbriae viarum publicarum instrumenti do-
mestici, Berolini 1901; fasciculus 2: Addenda ad partes priores et indicum capita
tria, Berolini 1925.
CivFal G. MAETZKE, O. PAOLETTI, & L. TAMAGNO PERNA (eds.), La Civiltà dei Falisci: Atti
del Convegno di Studi Etruschi ed Italici (Civita Castellana, Forte Sangallo, 28-31
Maggio 1987), Firenze 1990.
CollCas La collezione Casuccini, vol. 2: Ceramica attica, ceramica etrusca, ceramica falisca
[Monumenta antiqua Etruriae 2], Roma 1996.
CVA refers to the volumes of the series of the Corpus Vasorum Antiquorum.
DÉ A. ERNOUT & A. MEILLET, Dictionnaire étymologique de la langue latine, Paris
19594.
EDL M. DE VAAN, Etymological dictionary of Latin and the other Italic languages [Leiden
Indo-European Etymological Dictionary Series volume 7], Leiden / Boston 2008.
ET H. RIX (& G. MEISER), Etruskische Texte. Editio minor, Band I: Einleitung, Konkor-
danz, Indices [ScriptOralia 23, Reihe A: Altertumswissenschaftliche Reihe, Band
6], Tübingen 1991; Band II: Texte [ScriptOralia 24, Reihe A: Altertums-
wissenschaftliche Reihe, Band 7], Tübingen 1991.
Etr M. CRISTOFANI (ed.), Etruschi: una nuova immagine, Firenze 1984.
XI
BIBLIOGRAPHY
FI II Forma Italiae, serie II: Documenti, 1: G.F. GAMURRRINI, A. COZZA, A. PASQUI, &
R. MENGARELLI, Carta archeologica d’Italia (1881-1897): Materiali per l’Etru-
ria e la Sabina, Firenze 1972; 2: A. COZZA & A. PASQUI, Carta archeologica
d’Italia (1881-1897): Materiali per l’agro falisco, Firenze 1981.
IEW J. POKORNY, Indogermanisches etymologisches Wörterbuch, Bern / München vol. I
1959, vol. II 1969.
III J. ZVETAIEFF, Inscriptiones Italiae inferioris dialecticae in usum praecipue aca-
demicum, Mosquae 1886.
IIM J. ZVETAIEFF, Inscriptiones Italiae mediae dialecticae ad archetyporum et librorum
fidem, Lipsiae 1884; Inscriptiones Italiae mediae dialecticae ad archetyporum et
librorum fidem: Tabulae, Lipsiae 1885.
ILLRP A. DEGRASSI, Inscriptiones Latinae liberae rei publicae, fasciculus prior [Biblioteca
di Studi Superiori 23], Göttingen 1957; fasciculus alter [Biblioteca di Studi Supe-
riori 40], Firenze 1963.
LDIA A.L. PROSDOCIMI (ed.), Lingue e dialetti dell’Italia antica [Popoli e civiltà dell’Italia
antica 6], Roma 1978. (The updates by A. MARINETTI are referred to as MARI-
NETTI 1982 and MARINETTI 1984.)
LEW A. WALDE & J.B. HOFMANN, Lateinisches etymologisches Wörterbuch, Heidelberg
vol. I 19383, vol. II 19543.
MVG G. PROIETTI (ed.), Il Museo Nazionale di Villa Giulia, edRoma 1980.
NDI P. POCCETTI, Nuovi documenti italici: a complemento del Manuale di E. Vetter
[Orientamenti linguistici 8], Pisa 1979.
NRIE M. BUFFA, Nuova raccolta di iscrizioni etrusche, Firenze 1936.
PrItal Prima Italia. Italische kunst van het eerste millennium vóór J.K., Brussel 1980.
RomTar M. CRISTOFANI (ed.) La grande Roma dei Tarquini [Catalogo della mostra : Roma,
Palazzo delle Esposizioni 12 giugno-30 settembre 1990], Roma 1990.
REA refers to the volumes of the Real-Encyclopädie der classischen Altertumswissenschaft
RMR R. BENEDETTO (ed.), Roma medio-repubblicana: aspetti culturali di Roma e del Lazio
nei secoli IV e III a. C., Roma 1973.
Sant G. COLONNA (ed.), Santuari d’Etruria, Milano 1985.
ST H. RIX, Sabellische Texte: Die Texte des Oskischen, Umbrischen und Südpikenischen
[Indogermanische Bibliothek, Erste Reihe] Heidelberg 2002. (= Handbuch der
italischen Dialekte, begründet von Emil VETTER, V. Band).
Tajan Archéologie égyptienne, orientale, greque et étrusque, romaine, préhistorique et
nationale. Paris, mercredi 5 juin, 2002, Tajan auction catalogue, Paris 2002.
TLE M. PALLOTTINO, Testimonia linguae Etruscae, Firenze 19682.
TLL refers to the volumes of the Thesaurus Linguae Latinae
WeltGr B. NEUTSCH, G. HAFNER, H. LUSCHEY, & H. WAGNER, Die Welt der Griechen im
Bilde der Originale der Heidelberger Universtätssammlung: Katalog der
Jubiläumaustellung zur 100-Jahr-Feier der Sammlungen des Archäologischen
Instituts Heidelberg im Sommersemester, Heidelberg 1948.
WOU J. UNTERMANN, Wörterbuch des Oskisch-Umbrischen [Indogermanische Bibliothek,
Erste Reihe] Heidelberg 2000. (= Handbuch der italischen Dialekte, begründet von
Emil VETTER, III. Band).
2. PUBLICATIONS REFERRED TO BY AUTHOR

ADAMS, John Norman
1976 ‘A typological approach to Latin word order’, IF 81 (1976), pp.70-99.
1982 The Latin sexual vocabulary, London 1982.
2003 Bilingualism and the Latin language, Cambridge 2003.
2007 The regional diversification of Latin 200 BC - AD 600, Cambridge 2007.
XII
BIBLIOGRAPHY
ADAMS, John Norman, & Simon Swain

2002 ‘Introduction’, in: J.N. ADAMS, M. JANSE, & Simon SWAIN (eds.) Bilingualism in
ancient society: Language contact and the written text, pp.1-22, Oxford 2002.
ADAMS HOLLAND, Louise (also: HOLLAND, Louise)
1925 The Faliscans in prehistoric times [Papers and Monographs of the American Aca-
demy in Rome 5], Roma 1925.
ADEMBRI, Benedetta
1985 ‘Ceramica falisca ed etrusca a figure rosse: qualche precisazione’, in: Contributi alla
ceramica etrusca tardo-classica [Quaderni del Centro di Studio per l’Archeologia
Etrusco-Italica 10], pp.17-21, Roma 1980.
1990 ‘La più antica produzione di ceramica falisca a figure rosse. Inquadramento stilistico e
cronologico’, in: CivFal, pp.233-244.
ADIEGO LAJARA, Ignacio-Javier
1990 ‘Der Archaismus des Südpikenischen’, HSF 103 (1990), pp.69-79.
AGOSTINIANI, Luciano
1981 ‘Duenom duenas: καλος καλō: mlaχ mlakas’, SE 49 (1981), pp.95-111.
1982 Le ‘iscrizioni parlanti’ dell’Italia antica [Lingue e Iscrizioni dell’Italia Antica 3],
Firenze 1982.
ALFÖLDI, András
1963 Early Rome and the Latins, Ann Arbor 1963.
ALLEN, William Sidney
1978 Vox Latina, Cambridge 19782.
ALTHEIM, Franz
1930 Griechische Götter im alten Rom, Gießen 1930.
1931 Terra Mater: Untersuchungen zur altitalischen Religionsgeschichte [Religions-
geschichte: Versuche und Vorarbeiten XXII.2], Gießen 1931.
1932 ‘Die Anfänge des Vulgärlateins’, Glotta 20 (1932), pp.153-171.
1951 Geschichte der lateinischen Sprache von der Anfängen bis zum Beginn der Literatur,
Frankfurt a. M. 1951.
AMBROSINI, Laura
1995 ‘Sethlans con la ruota di Issione su uno specchio inciso da Corchiano’, SE 61 (1995),
pp. 181-200.
ÁMPOLO, Carmine
1972 ‘Fertor Resius rex Aequicolus’, PP 27 (1972), pp.409-412.
1980 ‘Le condizioni materiali della produzione. Agricoltura e paesaggio agrario’, DArch
n.s. 2 (1980), pp.15-46.
ANDERSON, James M.
1965 ‘A study of syncope in Vulgar Latin’, Word 21 (1965), pp.70-85.
ANDREAE, Bernard
1957 ‘Lucus Feroniae’, in: ‘Archäologische Funde im Bereich der Soprintendenzen von
Rom 1949-1956/57’, JDAI 72 (1957), Archäologischer Anzeiger, cc.110-358, cc.270-
275.
ANDRÉN, Arvid
1940 Architectural terracottas from Etrusco-Italic temples [Skrifter utgivna av Svenska
Institutet i Rom / Acta Instituti Romani Regni Suegiae VI], Lund / Leipzig 1940.
APPEL, René, & Pieter C. MUYSKEN
1983 ‘Buitenlanders en taalbehoud’, TTT 3 (1983), pp.152-166.
1987 Language contact and bilingualism, London / Caulfield / Baltimore 1987.
ARENA, Renato
1967 ‘Das Zeichen í der oskischen Schrift’, in: ‘Oskisches’, Glotta 45 (1967), pp.113-119
(pp.113-116).
1974 ‘L’iscrizione di Vico Equense’, SE 42 (1974), pp.387-390.
XIII
BIBLIOGRAPHY
ASCOLI, Graziadio Isaia

1868 ‘Lateinisches und romanisches, III. Zur lateinischen vertretung der indogermanischen
aspiraten’, HSF 17 (1868), pp.241-281, 321-343.
BABELON, Ernest.
1899 Intailles et camées, donnés aux Département des Médailles et Antique de la Biblio-
thèque Nationale, Catalogue de la Collection Pauvert de la Chapelle, Paris 1899.
BAFFIONI, Giovanni
1967 ‘Sappinates o Capenates?’, SE 35 (1967), pp.127-157.
BAGLIONE, M. Paola
1986 ‘Il Tevere e i Falisci’, in: Archeologia Laziale VII,2: Settimo incontro di studio del co-
mitato per l'archeologia laziale: Il Tevere e le altre vie d’acqua del Lazio antico
[Quaderni del Centro di Studio per l’Archeologia Etrusco-Italica 12], pp.124-142,
Roma 1986.
BAGLIONE, M. Paola, & M. Anna DE LUCIA BROLLI
1990 ‘Nuovi dati sulla necropoli de «I tufi» di Narce’, in: CivFal, pp.61-102.
BAKKUM, Gabriël C.L.M.
1985 Een voorbeeld van taaldood: het Oscisch, unpublished paper, Amsterdam 1985.
1989 ‘Complex s-onsets in Latin and Italic’, in: M. LAVENCY & D. LONGREE (eds.), Actes
du Ve Colloque de Linguistique Latine [Cahiers de l’Institut de Linguistique de Lou-
vain 15.1-4], pp.17-27, Louvain-la-Neuve 1989.
1991 ‘Eitam: Faliscan or Etruscan?’, Glotta 69 (1991), pp.215-218.
1992 ‘Alsmede de Italische dialectologie: kanttekeningen bij enige inscriptiones vetustis-
simae’, in: G. BAKKUM, C. KROON, & R. RISSELADA (eds.), Pentecostalia: Bundel
ter gelegenheid van de vijftigste verjaardag van Harm Pinkster, pp.1-12, Amsterdam
1992.
1994 ‘The second-declension nominative plural in -eis, -is, -es, and the first-declension
nominative plural in -as’, in: J. HERMAN (ed.), Linguistic studies on Latin: Selected
papers from the 6th International Colloquium on Latin Linguistics (Budapest, 23-27
March 1991) [Studies in Language Companion Series, 28], pp.19-40, Amster-
dam/Philadelphia 1994.
1996 ‘Capenate esú CIL I2.476.6 = XI.6707,6, CIL I2.2496,9’, in: R. RISSELADA, J.R. DE
JONG, & A.M. BOLKESTEIN (eds.), On Latin: linguistic and literary studies in honour
of Harm Pinkster, pp.1-6, Amsterdam 1996.
Baldi, Philip
1999 Foundations of Latin, Berlin / New York 1999.
BALDI, Philip, & Ruth JOHNSTON-STAVER
1989 ‘Historical Italic phonology in typological perspective’, in: Th. VENNEMANN (ed.),
The New Sound of Indo-European: Essays in Phonological Reconstruction [Trends in
Linguistics: Studies and Monographs, 41], pp.85-102, Berlin / New York 1989.
BALLESTER, Xaverio
1990a ‘La posición del acento prehistórico latino’, Emérita 78 (1990), pp.33-50.
1990b ‘El acento latino según los antiguos’, Emérita 78 (1990), pp.311-322.
BALLHEIMER, Rudolf
1909 ‘Faliskische Vasen der Sammlung Joh.W.F. Reimers in Hamburg’, JDAI 24 (1909),
Anzeiger cc.29-30.
BARNABEI, Felice
1882 [adunanza dell’Instituto 17.III.1882], Bullettino dell’Instituto di Corrispondenza
Archeologica 1882, pp.71-72.
1894a ‘Degli scavi di antichità nel territorio falisco’, MonAl 4 (1894), cc.5-32.
1894b ‘Dei fittili scoperti nella necropoli di Narce’, MonAl 4 (1894), cc.165-320.
XIV
BIBLIOGRAPHY
BASSI, Stelio
1956 Monumenta Italiae graphica I: La scrittura greca in Italia nell’età arcaica, Cremona
1956.
BAUDRILLART, André
1889 ‘Coupes de Popilius’, MÉFRA 9 (1889), pp.288-298.
BAZZICHELLI, Giosafàt
1881 ‘Carbognano’, NSA 1881, pp.134-136.
1885 ‘Corchiano’, NSA 1885, pp.420-422.
BEAZLEY, John D.
1947 Etruscan vase-painting, Oxford 1947.
BEEKES, Robert Stephen Paul
1969 The development of the Proto-Indo-European laryngeals in Greek [Ianua Linguarum:
Studia Memoriae Nicolai van Wijk dedicata - series practica 42], Den Haag / Paris
1969.
1985 The origins of the Indo-European nominal inflection [Innsbrucker Beiträge zur
Sprachwissenschaft 46], Innsbruck 1985.
1987 ‘On Indo-European ‘wine’’, MMS 48 (1987), pp.21-26.
1990a Vergelijkende taalwetenschap - Tussen Sanskrit en Nederlands, Utrecht 1990.
1990b ‘The genitive in *-osio’, Folia Linguistica Historica 11 (1990), pp.21-26.
1990c ‘De verwantschap van het Etruskisch’, Lampas 23 (1990), pp.5-18.
BEEKES, Robert Stephen Paul, & L.B. VAN DER MEER
1991 De Etrusken spreken, Muiderberg 1991.
BEELER, Madison S.
1952 ‘The relation of Latin and Osco-Umbrian’, Language 28 (1952), pp.435-443.
1956 ‘Venetic and Italic’, in: Hommages à Max Niedermann [Collection Latomus, vol.23],
Bruxelles 1956.
1966 ‘The interrelationships within Italic’, in: H. BIRNBAUM & J. PUHVEL, Ancient Indo-
European dialects [Proceedings of the Conference on Indo-European linguistics held
at the University of California, Los Angeles, April 25-27, 1963], pp.51-58, Berkeley /
Los Angeles 1966.
BECKMANN, G.
1963 Die Nachfolgekonstruktion des instrumentalen Ablativs im Spätlatein und im
Französischen [Beiheft zur Zeitschrift für romanische Philologie 106], Tübingen
1963.
BELOCH, Karl Julius
1925 Römische Geschichte bis zum Beginn der Punischen Kriege, Berlin / Leipzig 1925.
BENDINELLI, G.
1920 ‘Corchiano - Scoperta di antiche tombe falische a camera in contrada «Lista»’, NSA
1920, pp.20-30.
BERANGER, Eugenio Maria, & Patrizia FORTINI
1978 ‘Nepet’, SE 46 (1978), p.355.
BERENGUER [SANCHEZ], José Antonio, & Eugenio Ramón LUJAN [MARTINEZ]
2004 ‘La nueva inscripción falisca de Cavios Frenaios’, ZPE 149 (2004), pp.213-222.
2005 ‘Falisco faced y el perfecto de *dheH1-k- ‘hacer’ en las lenguas itálicas’, Emérita 73
(2005), pp.197-216.
BERNARDINI, Paola
1986 Museo Nazionale Romano - Le ceramiche, V,1: La ceramica a vernice nera dal
Tevere, Roma 1986.
BIVILLE, Frédérique
1988 ‘La pertinence du critère apophonique dans la datation des emprunts au grec et des
faits phonétiques latins’, Glotta 66 (1988), pp.190-210.
XV
BIBLIOGRAPHY
BLANCK, Horst
1990 ‘Ritrovamenti dal cd. Ninfeo Rosa di Falerii Veteres’, in: CivFal, pp.223-230.
BLOCH, Alfred
1960 ‘Kann der lateinisch-keltische Genetiv auf -ī der o-Stämme gleichen Ursprungs sein
wie die altindische Präverbialform auf -ī?’, ZVS 76 (1960), pp.182-242.
BLOCH, Raymond
1952a ‘Nouvelles dédicaces archaïques à la déesse Feronia’, Comptes rendus des séances de
l’Académie des Inscriptions et des Belles-Lettres 1952, pp. 620-628.
1952b ‘Remarques sur quelques inscriptions latines archaïques’, RPh 26 (1952), pp.
179-186.
1955 ‘Découverte d’un habitat étrusque archaïque sur le territoire volsinien’, MÉFRA 66
(1955), pp. 49-70.
1969 Recherches archéologiques en territoire volsinien de la protohistoire à la civilisation
étrusque [Bibliothèque des Écoles Françaises d’Athènes et de Rome 220], Paris 1972.
BLOCH, Raymond, & Giuseppe FOTI
1953 ‘Nouvelles dédicaces archaïques à la déesse Feronia’, RPh 27 (1953), pp. 65-77.
DE BLOIS, Lukas
1976 The policy of the emperor Gallienus [Studies of the Dutch Archaeological and
Historical Society VII], Leiden 1976 (diss. Amsterdam 1974).
BLÜMEL, Wolfgang
1972 Untersuchungen zu Lautsystem und Morphologie des vorklassischen Lateins,
München 1972.
BOLELLI, Tristano
1943 ‘Considerazioni sul genitivo latino della seconda declinazione’, Atti della Reale
Accademia d’Italia serie 7 vol. 4 (1943), pp.49-60.
BOLKESTEIN, A. Machtelt
1980 Problems in the descriptions of modal verbs: an investigation of Latin, Assen 1980
(diss. Amsterdam 1980).
BONAMICI, Marisa
1974 I buccheri con figurazioni graffite [Biblioteca di Studi Etruschi 8], Firenze 1974.
BONFANTE, Giuliano
1935 ‘Osco íív’, RIGI 19 (1935), pp. 182-184.
1966 ‘Il valore della léttera Z in falisco (in màrgine al libro della Dott. G. Giacomelli)’, AGI
51 (1966), pp. 1-25.
1978 ‘La nuova iscrizione di Satricum e il genitivo in -osio’, RAL 33 (1978), pp. 269-272.
LE BONNIEC, Henri
1958 Le culte de Cérès à Rome des origines à la fin de la République [Études et Commen-
taires XVII], Paris 1958.
1980 ‘La fête de Junon au pays des Falisques (Ovide, Am. III, 13)’, in: Actes du colloque
L’élégie romaine: enracinement, thèmes, diffusion (mars 1979) [Bulletin de la Faculté
des Lettres de Mulhouse, fasc. X], pp.233-244. Paris 1980.
BOPP, Franz
1857 Vergleichende Grammatik des Sanskrit, Send, Armenischen, Griechischen, Latein-
ischen, Litauischen, Altslavischen, Gothischen und Deutschen, Berlin 18572.
BORIE, Ferdinand
1898 ‘La mort du Minotaure’, MÉFRA 18 (1898), pp. 51-63.
BORMANN, Eugen
1888 Corpus inscriptionum Latinarum XI: Inscriptiones Aemiliae Etruriae Umbriae
Latinae, pars prior: Inscriptiones Aemiliae et Etruriae, Berolini 1888.
Latinae, pars posterior, fasciculus 1: Inscriptiones Umbriae viarum publicarum ins-
trumenti domestici, Berolini 1901.
XVI
BIBLIOGRAPHY
BORMANN, Eugen (continued)

Latinae, pars posterior, fasciculus 2: Addenda ad partes priores et indicum capita tria,
Berolini 1925.
BOTTIGLIONI, Gino
1954 Manuele dei dialetti italici (Osco, Umbro, e dialetti minori): Grammatica, testi,
glossario con note etimologiche, Bologna 1954.
BRADLEY, Guy
1997 ‘Iguvines, Umbrians and Romans: ethnic identity in central Italy’, in: T. C ORNELL &
K. LOMAS (eds.), Gender and ethnicity in ancient Italy [Accordia specialist studies on
Italy, vol.6], pp.53-68, London 1997.
BRATHER, Sebastian
2004 Ethnische Interpretationen in der frühgeschichtlichen Archäologie: Geschichte,
Grundlage und Alternativen [Ergänzungsbände zum Reallexicon der Germanischen
Altertumskunde, Band 42], Berlin/New York 2004.
BRAUN, Alfonsina
1935 ‘Studi sul dialetto falisco’, RFIC 12 (1935), pp. 433-451.
BREAL, Michel
1881 ‘Épigraphie italique’, MSL 4 (1881), pp. 373-405.
1884 ‘Un génitif latin en -as’, MSL 5 (1884), pp. 228-229.
BREYER, Gertraud
1993 Etruskisches Sprachgut im Lateinichen unter Ausschluss des spezifisch onomastis-
chen Bereichs [Orientalia Lovaniensia analecta, 53], Leuven 1993.
BRIQUEL, Dominique
1972 ‘Sur des faits d’écriture en Sabine et dans l’ager Capenas’, MÉFRA 82 (1972), pp.
789-845.
1974 ‘Le problème des Dauniens’, MÉFRA 86 (1974), pp.7-40.
2003 ‘Una nuova iscrizione falisca su un vaso di tipo Genucilia’, SE 69 (2003), pp.401-403.
BRIZIO, E.
1889 ‘Il nuovo Museo Nazionale delle Antichità in Roma’, Nuova antologia 24 [108]
(1889), fasc. 6, pp.409-444.
BROMLEY, Julian V., & V. I. KOSLOV
1975 ‘Этнические процессы как предмет исследования (предварительные замеча-
ния)’, in: J.V. BROMLEY (coord.), Современные этнические процессы в СССР,
pp.9-23, Moskwa 1975.
BROMMER, Frank
1956 Vasenlisten zur griechischen Heldensage, Marburg (Lahn) 1956.
1980 Göttersagen in Vasenlisten, Marburg (Lahn) 1980.
BRONISCH, Gotthelf
1892 Die oskischen i- und u-Vocale: eine statistisch-descriptive und sprachgeschichtlich-
vergleichende Untersuchung, diss. Leipzig 1892.
BRUHL, Adrien
1953 Liber pater: Origine et expansion du culte dionysiaque à Rome et dans le monde
romain [Bibliothèque des Écoles Françaises d’Athènes et de Rome 175], Paris 1953.
BÜCHELER, Franz
1905 ‘Iugmentum. Offimentum. Detramen’, RhM 60 (1905), pp.317-320.
1921 Anthologia Latina sive poesis Latinae supplementum, pars posterior: Carmina
epigraphica, fasciculus I, Lipsiae 1921.
BUCK, Carl Darling
1928 A grammar of Oscan and Umbrian with a collection of inscriptions and a glossary,
Boston 19282.
1955 The Greek dialects: Grammar, selected inscriptions, glossary, Chicago 19552.
XVII
BIBLIOGRAPHY
BUFFA, Mario
1936 Nuova raccolta di iscrizioni etrusche, Firenze 1936.
BUGGE, Sophus
1878 Altitalische Studien, Christiania 1878.
BUGGE, Sophus / Alf TORP
1909 Das Verhältnis der Etrusker zu den Indogermanen und der vorgriechische Bevölker-
ung Kleinasiens und Griechenlands. Sprachliche Untersuchungen, Straßburg 1909.
BUGLIONE DI MONALE, Alfonso
1887a ‘Delle antichità falische venute alla luce in Civita Castellana e in Corchiano e della
ubicazione di Fescennia’, MDAI(R) 2 (1887), pp.25-36.
1887b [Sitzungsbericht der kaiserlich deutschen archäologischen Instituts zu Rom am 27.
Dezember 1886 und am 7. Januar 1887], Wochenschrift für klassische Philologie
1887, cc.119-120.
BUNDGÅRD, J. A.
1965 ‘Why did the art of writing spread to the West? Reflections on the alfabet of Mar-
siliana’, ARID 3 (1965), pp.11-72.
BUONAMICI, Giulio
1913 Il dialetto falisco, Ímola 1913.
1928 ‘Etruria meridionale’, SE 2 (1928), pp.604-608.
1929a ‘Fiano Romano. Territorio capenate’, SE 2 (1929), pp.504-505.
1929b ‘Varia’, SE 3 (1929), pp.509-513.
1932 Epigrafia etrusca. Saggi e materiali, Firenze 1932.
1934 ‘Rivista di epigrafia etrusca 1933-1934’, SE 8 (1934), pp.353-373.
1935a ‘Agro falisco. Pizzo Piede’, SE 9 (1935), pp.341-345.
1935b [note on CII 2164quater], SE 9 (1935), pp.356-357.
1936 ‘Territorio falisco. Civita Castellana’, SE 10 (1936), pp.424-426.
1938 ‘CII, n.2164 quater’, SE 12 (1938), pp.317-321.
1939 ‘Civita Castellana. Territorio falisco’, SE 13 (1939), p.473.
1940 ‘Tuscania’, SE 14 (1940), pp.393-399.
1941 ‘Agro falisco. Narce’, SE 15 (1941), pp.370-371.
1942 ‘L’interpunzione sillabica e di altre forme nell’etrusco con particolare riguardo al
periodo arcaico (VII-VI sec. a. C.)’, SE 16 (1942), pp.263-344.
BUONARRUOTI, Filippo
1726 Ad monumenta Etrusca operi Dempsteriano addita explicationes et conjecturae,
Florentiae 1726. [Appended to DEMPSTER 1723.]
CALZECCHI-ONESTI, Giorgio
1981 ‘OCR- ed ACR- nella toponomastica dell’Italia antica’, SE 49 (1981), pp.165-189.
CAMPANILE, Enrico
1961 ‘Elementi dialettali nella fonetica e nella morfologia del latino’, SSL 1 (1961), pp.1-
21.
1968 Studi sulla posizione dialettale del latino [Biblioteca dell’Italia antica e di Studi e
Saggi Linguistici], s.l. 1968 (= SSL 31 [8] (1968), pp.16-130).
1986 ‘Pe parai modello di fhe fhaked? (Nota sulla iscrizione della fibula Praenestina)’,
Studi classici ed orientali 36 (1986), pp.13-16.
1991 Rapporti linguistici e culturali tra i popoli dell’Italia antica: Pisa, 6-7 ottobre 1989
[Testi linguistici, 17], Pisa 1991.
1993 ‘Il latino dialettale’, in: Enrico CAMPANILE (ed.), Caratteri e diffusione del latino in
età arcaica, pp. 13-24, Pisa 1993.
CAMPOREALE, Giovannangelo
1956 La terminologia magistrale nelle lingue osco-umbre, Firenze 1956.
XVIII
BIBLIOGRAPHY
CAMPOREALE, Giovannangelo (continued)

1967 ‘Note sulle dediche umbre a Cupra da Colfiorito’, Atti dell’Accademia Nazionale dei
Lincei, serie 8: Rendiconti (classe di scienze morali, storiche e filologiche 22 (1967),
pp.65-72.
1991 ‘L’ethnos dei Falisci secondo gli scrittori antichi’, ArchClass 43 (1991), pp.209-221.
CARLUCCI, Claudia
1995 ‘Il santuario falisco di Vignale: nuove acquisizioni’, ArchClass 47 (1995), pp.169-
101.
CENCETTI, Giorgio
1957 ‘Richerche sulla scrittura latina dell’età arcaica I: Il filone corsivo’, Bullettino
dell’Archivio Paleografico Italiano 2-3 (1956-1957), vol.I, pp.175-205.
CIFANI, Gabriele
1992 ‘Una tegola dipinta dall’area falisca’, ArchClass 44 (1992), pp.263-271.
2002 ‘La posizione di Gallese (VT) nel territorio falisco’, RPAA 74 (2001-2002), pp.19-33.
CIFANI, Gabriele, & Massimiliano MUNZI
1995 ‘Considerazioni sugli insediamenti in area falisca. I periodi arcaico e tardo antico’, in
N. CHRISTIE (ed.), Settlement and economy in Italy 1500 B.C. to A.D. 1500 (Papers
of the Fifth Conference of Italian archaeology, Oxford 1992), pp. 392-393, Oxford
1995.
CLACKSON, James, & Geoffrey HORROCKS
2007 The Blackwell History of the Latin Language, Malden/Oxford/Victoria 2007.
CLUVERIUS, Philippus
1624 Italia antiqua; opus post omnium curas elaboratissimum; tabulis geographicis aere
expressis illustratum. Ejusdem Sicilia Sardinia et Corsica. Cum Indice locupletissimo,
Lugduni Batavorum 1624.
COLEMAN, Robert
1962 ‘Some allophones of Latin /i/’, TPhS 1962, pp.80-104.
1971 ‘Monophthongisation of /ae/ and the Vulgar Latin vowel system’, TPhS 1971,
pp.175-191.
1974 ‘The monophthongisation of Latin ae: a reply’, TPhS 1974, pp.86-92.
1986 ‘The Central Italic languages in the period of the Roman expansion’, TPhS 1986,
pp.100-131.
1990 ‘Dialectal variation in republican Latin, with special reference to Praenestine’, PCPhS
216 [36] 1990, pp.1-25.
COLLINGE, Nigel E.
1985 The laws of Indo-European [Amsterdam Studies in the Theory and History of
Linguistic Science, IV: Current Issues in Linguistic Theory, 35], Amsterdam / Phila-
delphia 1985.
COLONNA, Giovanni
1970 ‘Una nuova iscrizione etrusca del VII secolo e appunti sull’epigrafia ceretana
dell’epoca’, MÉFRA 82 (1970), pp.637-672.
1972a ‘Nuceria Alafaterna’, SE 40 (1972), pp.379-386.
1972b ‘Nepet’, SE 40 (1972), pp.444-446.
1972c ‘Ager Faliscus’, SE 40 (1972), pp.446-447.
1974 ‘Nomi etruschi di vasi’, ArchClass 25-26 (1973-1974), pp.132-150.
1975a ‘Firme arcaiche di artefici nell’Italia centrale’, MDAI(R) 82 (1975), pp.181-195.
1975b ‘Nuovi dati epigrafici sulla protostoria della Campania’, in: Atti della XVII riunione
scientifica dell’Instituto Italiano di Preistoria e Protostoria, pp.151-169, Firenze
1975.
1976a ‘La diffusione della scrittura’, in: G. C OLONNA, G. BARTOLONI, E. COLONNA DE
PAOLO, & F. MELIS, Civiltà del Lazio primitivo, pp.372-376, Roma 1976.
1976b ‘Una nuova iscrizione vascolare falisca’, Epigraphica 38 (1976), pp.117-120.
XIX
BIBLIOGRAPHY
COLONNA, Giovanni (continued)

1976c ‘Lucus Feroniae’, SE 44 (1976), p.251.
1977 ‘Nome gentilizio e società’, SE 45 (1977), pp.175-192.
1979 ‘Duenos’, SE 47 (1979), pp.163-172.
1980a ‘L’aspetto epigrafico’, in: Lapis Satricanus: Archeological, epigraphical, linguistic
and historical aspects of the new inscription from Satricum [Archeologische studiën
van het Nederlands Instituut te Rome: Scripta minora 5], pp.41-52, ’s Gravenhage
1980.
1980b ‘Appendice: Le iscrizioni strumentali latine dal VI e V secolo a. C.’, in: Lapis
Satricanus: Archeological, epigraphical, linguistic and historical aspects of the new
inscription from Satricum [Archeologische studiën van het Nederlands Instituut te
Rome: Scripta minora 5], pp.53-69, ’s Gravenhage 1980.
1980c ‘Graeco more bibere: l’iscrizione della tomba 115 dell’Osteria dell Osa’, in:
Archeologia Laziale III: Terzo incontro di studio del comitato per l’archeologia
laziale [Quaderni del Centro di Studio per l’Archeologia Etrusco-Italica 4], pp.51-55,
Roma 1980.
1980d ‘Sul graffito tecliiam di Nola (Vetter 120)’, SE 48 (1980), pp.429-430.
1983a ‘Un’iscrizione paleoitalica dall’agro Tolfetano’, SE 51 (1983), pp.573-587.
1983b ‘Identità come appartenenza nelle iscrizioni di possesso dell’Italia preromana’
Epigraphica 45 (1983), pp.49-64.
1984 ‘La nuova iscrizione di Satricum’, in: Archeologia Laziale VI: Sesto incontro di studio
del comitato per l’archeologia laziale [Quaderni del Centro di Studio per l’Archeo-
logia Etrusco-Italica 8], pp.104-6, Roma 1984.
1987 ‘Etruria e Lazio nell’età dei Tarquini’, in: M. C RISTOFANI (ed.), Etruria e Lazio
arcaico. Atti dell’Incontro di Studio (10-11 novembre 1986) [Quaderni del Centro di
Studio per l’Archeologia Etrusco-Italica 15], pp.55-66. Roma 1987.
1990a ‘Nuovi graffiti capenati’, SE 56 (1989-1990), pp.462-464.
1990b ‘Corchiano, Narce e il problema di Fescennium’, in: CivFal, pp.111-140.
1992a [‘Anagni’], SE 58 (1992), pp. 321-325.
1992b ‘The inscription on the miniature axe-head’, in: M. GNADE, The southwest necropolis
of Satricum: excavations 1981-1986, pp.125-128, Amsterdam 1992.
1993 ‘Falerii’, SE 59 (1993), pp. 297-299.
1994 [‘Ager Signinus’], SE 60 (1994), pp.298-301.
1995 ‘Anfora viniaria con iscrizione etrusca dalla Castellina del Marangone’, ArchClass 47
(1995), pp.261-266.
1997 [‘Inscriptiones originis incertae, 36’], SE 63 (1997), pp. 410-412.
COMBET FARNOUX, Bernard
1980 Mercure romain: Le culte public de Mercure et la fonction mercantile à Rome de la
république archaïque à l’époque augustéenne [Bibliothèque des Écoles Françaises
d’Athènes et de Rome 238], Roma 1980.
COMELLA, Annamaria
1986 I materiali votivi di Falerii [Archaeologica 63: Corpus delle stipi votive in Italia I:
Regio VII.1], Roma 1986.
CONWAY, R.
1897 The Italic dialects, Cambridge 1897.
CONWAY, R., Joshua WHATMOUGH, & Samuel JOHNSON
1933 The Prae-Italic dialects of Italy, Cambridge 1933.
CORNELL, Tim
1986 ‘The value of the literary tradition concerning ancient Rome’, in: K.A. R AAFLAUB
(ed.), Social struggles in archaic Rome: New perspectives on the conflict of the or-
ders, Berkeley/Los Angeles/London 1986, pp.52-76.
XX
BIBLIOGRAPHY
CORNELL, Tim (continued)

1991 ‘The tyranny of evidence: a discussion of the possible uses of literacy in Etruria and
Latium in the archaic age’, in: M. BREAD e.a., Literacy in the Roman world [Journal
of Roman Archaeology, supplementary series 3], Ann Arbor 1991, pp.7-33.
1995 The beginnings of Rome: Italy and Rome from the Bronze Age to the Punic Wars
(c.1000-264 BC), London 1995.
1997 ‘Ethnicity as a factor in early Roman history’, in: T. CORNELL & K. LOMAS (eds.),
Gender and ethnicity in ancient Italy [Accordia specialist studies on Italy, vol.6],
pp.9-22, London 1997.
CORTSEN, S.P.
1908 Lyd og skrift i Etruskisk I: Konsonanterne, diss. København 1908.
1935 ‘Literaturbericht 1928-1934: Etruskisch’, Glotta 23 (1935), pp.145-187.
COURTNEY, E.
1995 Musa lapidaria: a selection of Latin verse inscriptions [American Philological
Association: American Classical Studies, 36], Atlanta 1995.
COZZA, Adolfo
1886 ‘Corchiano’, NSA 1886, pp.152-156.
1894 ‘Topografia di Narce e della sua necropoli’, MonAl 4 (1894), cc.104-164.
COZZA, Adolfo, & Angelo (Angiolo) PASQUI
1887a ‘Civita Castellana (antica Faleria) - Scavi della necropoli falisca in contrada «la
Penna»’, NSA 1887, pp.170-176.
1887b ‘Civita-Castellana (antica Faleria) - Scavi della necropoli falisca in contrada «Pen-
na»’, NSA 1887, pp.262-273.
1887c ‘Scavi della necropoli falisca in contrada «Valsiarosa»’, NSA 1887, pp.307-319.
CRISTOFANI, Mauro
1972 ‘Sull’origine e la diffusione dell’alfabeto etrusco’, in: ANRW I2, pp.466-489, Berlin /
New York 1972.
1977 ‘L’iscrizione arcaica di Poggio Sommavilla riscoperta e riconsiderata’, Civilta arcaica
dei Sabini nella valle del Tevere III: Rilettura critica della necropoli di Poggio Som-
mavilla, pp.95-108, Roma 1977.
1978 ‘L’alfabeto etrusco’, in: LDIA, pp.401-428.
1984 ‘Demografia e storia’, in: Etr, pp.29-32
1988 ‘Etruschi nell’agro falisco’, PBSR 56 (1988), pp.13-24.
1990a ‘Olla iscritta da Caere’, in RomTar p.101.
1990b [note on two inscriptions from Tarquinii], SE 56 (1989-1990), pp.346-349.
1993 ‘Le prime iscrizioni latine dell’Etruria’, in: E. CAMPANILE (ed.), Caratteri e diffusione
del latino in età arcaica, pp.25-34, Pisa 1993.
DANIELSSON, Olav August
1910 ‘Italica’, in: Sertum philologicum Carolo Ferdinando Johansson oblatum: Festskrift
tillegnad Karl Ferdinand Johansson på hans 50-årsdag den 16 September 1910,
pp.81-104, Göteborg 1910.
DAVIS CHASE, George
1897 ‘The origin of Roman praenomina’, HSCPh 8 (1897), pp.103-184.
DEECKE, Wilhelm
1881 ‘Jahresbericht über die italienischen Sprachen, auch das Altlateinische und
Etruskische für 1879-1881’, Jahresbericht über die Fortschritte der classischen
Alterthumswissenschaft 28 (1881), pp.230-247.
1886 ‘Beiträge zur Entzifferung der mittelitalischen Inschriften’, RhM 41 (1886),
pp.191-202.
1887 ‘Bemerkungen zu den Inschriften in [Wochenschrift für klassische Philologie 1887]
No.4 S.120 f.’, Wochenschrift für klassische Philologie 1887, c.220.
1888 Die Falisker. Eine geschichtlich-sprachliche Untersuchung, Straßburg 1888.
XXI
BIBLIOGRAPHY
DEGRASSI, Attilio
1943 Corpus inscriptionum Latinarum I, editio altera: Inscriptiones antiquissimae ad C.
Caesaris mortem, pars posterior, fasciculus 3: Addenda altera, indices, Berolini 1943.
1957 Inscriptiones Latinae liberae rei publicae, fasciculus prior [Biblioteca di Studi
Superiori 23], Göttingen 1957.
1963 Inscriptiones Latinae liberae rei publicae, fasciculus alter [Biblioteca di Studi
Superiori 40], Firenze 1963.
1965 Inscriptiones Latinas liberae rei publicae - Imagines, Berolini 1965.
DEGRASSI, Attilio, & Johann KRUMMREY
Caesaris mortem, pars posterior, fasciculus 4: Addenda tertia I: Textus, fasciculus 5:
Addenda tertia II: Tabulae, ed. Berolini / Novi Eboraci 1986.
DEMPSTERUS, Thomas (also: DEMPSTER, Thomas)
1723 De Etruria regali libri septem, Florentiae 1723.
DENCH, Emma
1997 ‘Sacred Springs to the Social War: myths of origins and questions of identity in the
Central Apennines’, in: T. CORNELL & K. LOMAS (eds.), Gender and ethnicity in an-
cient Italy [Accordia specialist studies on Italy, vol.6], pp.43-51, London 1997.
2005 Romulus’ asylum: Roman identities from the age of Alexander to the age of Hadrian,
Oxford 2005.
DENNIS, George
1844 [adunanza dell’Instituto 22.III.1844], Bullettino dell’Instituto Archeologico 1844,
pp.91-92.
1845 ‘Civita castellana’, in: ‘Monumenti etruschi’, Bullettino dell’Instituto di Corrispon-
denza Archeologica 1845, pp.137-141 (p.139).
1848 The cities and cemeteries of Etruria, London 1848.
1878 The cities and cemeteries of Etruria, London 18782.
1890 The recently excavated objects from Falerii in the new Etruscan museum at the Villa
Papa Giulio at Rome, Rome 1890.
DEONNA, Waldemar
1915 ‘Variétés au Musée d’Art et d’Histoire de Genève’, RA serie V tome 1 (Janvier - Juin
1915), pp.303-332.2
1919 ‘Notes archéologiques IX, au Musée d’Art et d’Histoire de Genève’, RA serie V tome
IX (Mai - Juin 1919), pp.98-142.
1923 Choix de monuments et de l’art antique, Genève 1923.
DETLEFSEN, Detlef
1861a ‘Alcune osservazioni sulle iscrizioni falische’, Bullettino dell'Instituto di Corrispon-
denza Archeologica 1861, pp.199-205.
1861b ‘Postilla’, Bullettino dell’Instituto di Corrispondenza Archeologica 1861, p.205.
DEVINE, Andrew Mackay
1970 The Latin thematic genitive singular, Stanford University 1970.
DEVOTO, Giacomo
1944 Storia della lingua di Roma [Istituto di Studi Romani - Storia di Roma 23], Bologna
19442, repr. Roma 1983.
1972 ‘Storia politica e storia linguistica’, in: ANRW I2, pp.457-465.
DIEHL, Ernst
1909 Altlateinische Inschriften [Kleine Texte für theologische und philologische Vor-
lesungen und Übungen 38/40], Bonn 1909.
1911 Altlateinische Inschriften [Kleine Texte für Vorlesungen und Übungen 38/40], Bonn
19112.
19303.
XXII
BIBLIOGRAPHY
DIEHL, Ernst (continued)

19755.
DINI, I., G. PIACCIAROTI, F. RICCI, & L. SANTELLA
1985 ‘Analisi delle emergenze storiche e delle «aree a rischio»’, in: Contributi allo studio di
fattabilità della direttrice viaria Civita Castellana - Viterbo, Viterbo 1985.
DIRICHS, J.F.K.
1934 Die urlateinische Reklamestrophe auf dem sogenannten Dresselschen Drillingsgefäß
des sabinischen Töpfers Dufnos (bisher Duenos), Heidelberg 1934.
VAN DOMMELEN, Peter, & Nicola TERRENATO
2007 ‘Introduction: local cultures and the expanding Roman Republic’, in: Peter VAN
DOMMELEN & Nicola TERRENATO (eds.), Articulating local cultures: power and
identity under the expanding Roman Republic [Journal of Roman Archaeology, sup-
plementary series 63], pp.7-12, Portsmouth (Rhode Island) 2007.
DRAGADZE, Tamara
1980 ‘The place of ‘ethnos’ theory in Soviet anthropology’, in: Ernst GELLNER (ed.), Soviet
and Western Anthropology, London 1980.
DRESSEL, Heinrich
1881 ‘Ipogeo falisco presso Carbognano’, Bullettino dell’Instituto di Corrispondenza
Archeologica 1881, pp. 151-159.
1882 [adunanza dell’Instituto 24.III.1882], Bullettino dell’Instituto di Corrispondenza
Archeologica 1882, pp. 97-98.
DUBOIS, Laurent
1989 Inscriptions grecques dialectales de Sicile - Contribution à l’étude du vocabulaire
grec colonial [Collection de l’École Française de Rome 119], Rome 1989.
DUCATI, Pericle
1927 Storia dell’arte etrusca, Firenze 1927.
DURANTE, Marcello
1953 ‘Appunti all’iscrizione della tazza di Narce (CIE 8412)’, SE 22 (1952-1953),
pp.421-423.
1958 ‘Osco hirpo, lupo o capro?’, PP 13 (1958), pp.412-417.
DUVAU, Louis
1889 [review of Deecke 1888], Révue critique d’histoire et de littérature 23 [28] (1889),
pp.9-11.
EICHNER, Heiner
1992 ‘Indogermanisches Phonemsystem und lateinische Lautgeschichte’, in: O. PANAGL &
Th. KRISCH (eds.), Latein und Indogermanisch. Akten des Kolloquium der Indoger-
manischen Gesellschaft, Salzburg, 23.-36. September 1986 [Innsbrucker Beiträge zur
Sprachwissenschaft, Band 64], pp.55-79, Innsbruck 1992.
EMILIOZZI, Adriana
1986 Il Museo Civico di Viterbo: Storia delle raccolte archeologiche [Musei e Collezioni
d’Etruria 2], Roma 1986.
ERASMI, Gabriele
1975 Studies on the language of Livius Andronicus, diss. University of Minnesota 1975.
1979 ‘The Saturnian and Livius Andronicus’, Glotta 57 (1979), pp. 125-149.
ERMAN, Konrad Bessel
1917 ‘Faliskisch efiles’, ZVS 48 (1917), pp.158-159.
ERNOUT, Alfred
1905 ‘Le parler de Préneste d’après les inscriptions’, MSL 13 (1905), pp.293-349.
1909 Les éléments dialectaux du vocabulaire latin, Paris 1909.
1945 Morphologie historique du latin [Nouvelle Collection à l’Usage des Classes 32], Paris
19452.
XXIII
BIBLIOGRAPHY
ERNOUT, Alfred (continued)

1957 Recueil de textes latins archaïques, Paris 19572.
1965 ‘Numina ignota’, RPh 39 (1965), pp.189-199.
ERNOUT, Alfred & Antoine MEILLET
1959 Dictionnaire étymologique de la langue latine, Paris 19594.
EROLI, G.
1875 [adunanza dell’Instituto 9.IV.1875], Bullettino dell’Instituto di Corrispondenza
Archeologica 1875, p.133.
ESKA, Joseph F.
1987 ‘The language of the Latin inscriptions of Pompeii and the question of an Oscan
substratum’, Glotta 65 (1987), pp.146-161.
ETTER, A. (ed.)
1986 O-o-pe-ro-si: Festschrift, für Ernst Risch zum 75. Geburtstag, Berlin / New York
1986.
FABRETTI, Ariodante
1867 Corpus inscriptionum Italicarum antiquioris aevi ordine geographico digestum, et
Glossarium Italicum in quo omnia vocabula continentur ex Vmbricis Sabinis Oscis
Volscis Etruscis aliisque monumentis quae supersunt collecta et cum interpretatio-
nibus variorum explicantur, Augustae Taurinorum 1867.
1872 Primo supplemento alla raccolta delle antichissime iscrizioni italiche, con l’aggiunta
di alcune osservazioni paleografiche e grammaticali, Torino 1872.
FERENCZY, Erdre
1987 ‘Über das Problem der Inschrift von Satricum’, Gymnasium 94 (1987), pp.97-108.
FERGUSON, C.
1959 ‘Diglossia’, Word 15 (1959), pp.325-340.
FILPPULA, Markku
2003 ‘The quest for the most ‘parsimonious’ explanations: endogeny vs. contact revisited’,
in: R. HICKEY (ed.), Motives for language change, pp.161-173, Cambridge 2003.
FIRMANI, Massimo A.S.
1977 ‘Nuovi documenti epigrafici della Sabina Tiberina’, Civilta arcaica dei Sabini nella
valle del Tevere III: Rilettura critica della necropoli di Poggio Sommavilla,
pp.109-116, Roma 1977.
1979 ‘Ricerche nella Sabina Velina e Tiberina’, in: Archeologia laziale II: Secondo
incontro di studio del comitato per l’archeologia laziale [Quaderni del Centro di Stu-
dio per l’Archeologia Etrusco-Italica 3], pp.116-119, Roma 1979.
1985 ‘Panorama archeologico Sabino alla luce di recenti acquisizioni’, in: B. R IPOSATI
(ed.), Preistoria, storia e civiltà dei Sabini [Convegno di Studio del Centro di Studi
Varroniani, Rieti, ottobre 1982], pp.99-124, Rieti 1985.
FIORELLI, Giuseppe
1882 ‘Ardea (Comune di Genzano di Roma)’, NSA 1882, pp.273-274.
1883 ‘Montefiascone’, NSA 1883, pp.431-521 (p.434).
FOTI, Giuseppe
1953 ‘Capena. Iscrizioni rinvenute in località «Bambocci» nella tenuta di Scorano’, NSA
1953, pp.13-17.
FRASCHETTI, Augusto
1975 ‘A proposito di un graffito tardoantico da Lucus Feroniae’, ArchClass 27 (1975),
pp.317-330.
FRANCIA, R.M.
1968 ‘Sobre fonología de la pausa en latín’, in: Actas del III Congreso Español de Estudios
Clásicos (Madrid, 28 de marzo - 1 abril de 1966), vol. III, pp.71-78, Madrid 1968.
DEL FRATE, Oronte
1898 Guida storica e descrittiva della Faleria Etrusca (Civita Castellana), Roma 1898.
XXIV
BIBLIOGRAPHY
FRANCIA, R.M.
1968 ‘Sobre fonología de la pausa en latín’, in: Actas del III Congreso Español de Estudios
Clásicos (Madrid, 28 de marzo - 1 abril de 1966), vol. III, pp.71-78, Madrid 1968.
FREEMAN, Philip
2002 ‘CIL I2.365 and Faliscan language survival’, ZPE 140 (2002), p.252.
GÁBRICI, Ettore
1912a ‘Faleria - Saggi di scavo a Monte Cerreto’, NSA 1912, pp.73-4.
1912b ‘Rignano Flaminio - Saggi di scavo a Monte Casale’, NSA 1912, pp.75-81.
1912c ‘Corchiano - Scavi in contrada Fratta’, NSA 1912, pp.82-83.
GAMURRINI, Gian Francesco
1880 Appendice al Corpus Inscriptionum Italicarum ed ai suoi supplementi di Ariodante
Fabretti, Firenze 1880.
1883 ‘Civita Castellana’, NSA 1883, pp.165-158.
1887a [Sitzungsprotocoll 12.I.1887], MDAI(R) 2 (1887), pp.61-2.
1887b [Sitzungsprotocoll 18.III.1887], MDAI(R) 2 (1887), p.151.
1887c [Sitzungsbericht der kaiserlich deutschen archäologischen Instituts 18.III.1887],
Wochenschrift für klassische Philologie 1887, cc.601-602.
1887d ‘Dell’arte antichissima in Roma’, MDAI(R) 2 (1887), pp.221-234.
1891 ‘S. Maria di Falleri (comune di Civita Castellana)’, NSA 1891, pp.48-50.
1894a ‘Dei fitilli iscritti scoperti nella necropoli di Narce’, MonAl 4 (1894), cc.320-346.
1894b ‘Se Corchiano sia l’antica Fescennio’, RAL serie 5 no. 3 (1894), 145-149.
GARRUCCI, Raffaele (also: GARRUCCIUS, Raphael)
1860 ‘Scoperte falische’, Annali dell'Instituto di Corrispondenza Archeologica 32 (1860),
pp.211-281.
1864 ‘Epigrafi falische anteaugustanee’, in: Dissertazioni archeologiche di vario argomen-
to, pp.59-76, Roma 1864.
1870 De’ canoni epigrafici di Federico Ritschl e di alcune iscrizioni arcaiche finora inedite
(Dissertazione letta alla Pontificia Accademia di Archeologia nelle tornate del 12
Maggio e 16 Decembre del 1869), Roma 1870.
1871a ‘On an archaic Latin inscription in the Faliscan character’, Archaeologia (Miscel-
laneous Tracts relating to Antiquity published by the Society of Antiquaries of Lon-
don) 43 (1871), pp.43-55.
1871b ‘Remarks on a Faliscan inscription’, Archaeologia (Miscellaneous Tracts relating to
Antiquity published by the Society of Antiquaries of London) 43 (1871), pp.259-263.
1877 Sylloge inscriptionum Latinarum aevi Romanae rei publicae usque ad C. Iulium
Caesarem plenissima, Torino 1877.
GASPERINI, Lidio
1989 Iscrizioni latine rupestri nel Lazio I. Etruria meridionale, Roma 1989.
GATTI, G.
1906 ‘Iscrizione arcaica capenate’, BCAR 34 (1906), pp.61-64.
GERHARD, Eduard (bearbeiteit von A. KLÜGMANN & G. KÖRTE)
1843 Etruskische Spiegel I: Allgemeines und Götterbilder, Berlin 1843.
GHIRARDINI, Gherardo
1900 ‘La situla italica primitiva, studiata specialmente in Este. Parte terza: L’ornamenta-
zione zoomorfica’, MonAl 10 (1900), cc.5-222.
GIACOMELLI, Gabriella
1961 ‘Iscrizioni del territorio falisco’, SE 29 (1961), pp.321-326.
1962 ‘Formazioni onomastiche in -aio nelle lingue dell’Italia antica’, SE 30 (1962),
pp.359-367.
1963 La lingua falisca [Biblioteca di Studi Etruschi 1], Firenze 1963.
1963b ‘La questione del lat. infra e una nuova iscrizione falisca’, AGI 48 (1963), pp.52-54.
1965a ‘Questioni di metodo nell’epigrafia’, SE 33 (1965), pp.549-552.
XXV
BIBLIOGRAPHY
GIACOMELLI, Gabriella (continued)

1965b ‘Epigrafi antiche in un codice Vaticano’, in: G. C AMPOREALE (ed.), Studi in onore di
Luisa Banti, pp.153-157, Roma 1965.
1978 ‘Il falisco’, in: LDIA, pp.505-42.
GIACOMELLI, Roberto
1975a ‘Lat. careô, castus e il rotacismo falisco’, RIL 109 (1975), pp.347-352.
1975b ‘Degli esiti falischi del dittongo eu’, RIL 109 (1975), pp.393-398.
1977 ‘Note di epigrafia e linguistica italica’, Paideia 32 (1977), pp.63-69.
1978 Problemi di storia linguistica del latino dialettale I: Ricerche falische [Pubblicazioni
della Facoltà di Lettere e Filosofia dell’Università di Milano, sezione a cura
dell’Istituto di Glottologia 5], Firenze 1978.
1979 ‘Written and spoken language in Latino-Faliscan and Greek-Messapic’, JIES 7
(1979), pp.149-175.
1982 ‘Un altro caso di commutazione automatica diglottica nel latino epigrafico’, IF 87
(1982), pp.82-89.
2006 Nuove ricerche falische, Roma 2006.
GIANNINI, Stefania
1986 ‘Un problema di fonosintassi in latino: la consonante -s finale’, SSL 26 (1986),
pp.111-136.
1987 ‘Per l’interpretazione fonologica delle labio-velari latine’, AION (arch.) 9 (1987),
pp.239-252.
GIGLIOLI, Giulio Quirino
1916 ‘Vignanello - Scavi nella città e nella necropoli’, NSA 1916, pp.37-85.
1924 ‘Vignanello - Nuovi scavi nella città e nella necropoli’, NSA 1924, pp.179-263.
1930 ‘Veio - Rinvenimento di alcune iscrizioni etrusche durante lo scavo del tempio
scoperto in contrada Portonaccio, presso Isola Farnese’, NSA 1930, pp.302-345.
1935 ‘Civita Castellana - Due vasi d’impasto con iscrizioni falische’, NSA 1935, pp. 238-
243.
GIL, Juan
1968 ‘El genitivo en -ī y los orígenes de la declinación temática’, Emérita 36 (1968),
pp.25-43.
GIORDANO, Carlo, & Isidoro KAHN
1979 The Jews in Pompeii, Herculaneum, Stabiae, and in the cities of Campania Felix,
Napoli 1979.
GIRARD, Jean-Louis
1989 ‘Minerva capta: entre Rome et Faléries’, RÉL 67 (1989), pp.163-169.
GNADE, Marijke (with a contribution by M. RUBINI)
2002 Satricum in the post-archaic period: a case study in the interpretation of archaeologi-
cal remains as indicators of ethno-cultural identity, Leuven 2002. (= M. GNADE, Sa-
tricum in the post-archaic period: a case study in the interpretation of archaeological
remains as indicators of ethno-cultural identity, diss. Amsterdam 2000)
GOLDMANN, E.
1928 ‘Ricerche etrusche’, SE 2 (1928), pp.209-286.
GORDON, Arthur Ernest
1975 The inscribed fibula Praenestina: problems of authenticity [University of California
Publications: Classical Studies 16], Berkeley / Los Angeles / London 1975.
GORI, Antonio Francesco
1733 ‘De Etruscorum antiquis privatis aedificiis’, in: Museum Etruscum, volumen III,
pp.69-98, Florentiae 1733.
GORI, Fabio
1864 ‘La via Flaminia fino a Capena ed al fano di Feronia’, Annali dell’Instituto di
Corrispondenza Archeologica 1864, pp.117-135.
XXVI
BIBLIOGRAPHY
GRAN-AYMERICH, Jean, & Dominique BRIQUEL

1997 ‘Originis incertae (Narce?)’, SE 63 (1997), pp. 429-430.
GRANDAZZI, Alexandre
1997 The foundation of Rome: myth and history, Ithaca / London 1997 (translation of La
fondation de Rome: réflexion sur l’histoire, Paris 1991)
GREENBERG, Joseph H.
1966 Universals of language, Cambridge 1966.
GRENIER, Albert
1912 Bologna villanovienne et étrusque, VIIIe-IVe siècles avant notre ère [Bibliothèque des
Écoles Françoises d’Athènes et de Rome, fascicule 106], Paris 1912.
GULINELLI, Paola
1995a ‘Ager Faliscus’, SE 61 (1995), pp. 319-320.
1995b ‘Ager Faliscus (Vignanello)’, SE 61 (1995), pp. 334-335.
1996 ‘L’iscrizione falisca Ve 257’, PP 51 (1996), pp. 225-229.
HAMATTA, Janos
1987 ‘Contribution aux antécédents locaux du latin vulgaire en Pannonie’, in: J. Herman,
Latin vulgaire - latin tardif [Actes du Ier Colloque international sur le latin vulgaire et
tardif (Pécs, 2-5 septembre 1985)], Tübingen 1987.
HAMP, Eric P.
1972 ‘‘Fils’ et ‘fille’ en italique: nouvelle contribution’, BSL 66 (1972), pp.213-227.
1981 ‘Early Italic notes’, Glotta 59 (1981), pp.228-230.
1984 ‘Latin in before dental’, Glotta 62 (1984), pp.73-74.
1988a ‘Miscellanea’, IF 93 (1988), pp.121-123.
1988b ‘Refining Indo-European lexical entries’, ZVS 101 (1988), pp. 79-80.
HARRIS, Martin, & Nigel VINCENT (eds.)
1988 The Romance languages, London / Sydney 1988.
HARRIS, William Vernon
1971 Rome in Etruria and Umbria, Oxford 1971.
HARTMANN, F.
1890 [review of P. VON BRATKE, Beiträge zur Kenntnis der vorhistorischen Entwicklungen
unseres Sprachstammes, Gießen 1888], Deutsche Literaturzeitung 11 (1890),
cc.1830-1832.
1892 [review of H. OSTHOFF & K. BRUGMANN, Morphologische Untersuchungen auf
dem Gebiete der indogermanischen Sprache V, Leipzig 1890], Deutsche
Literaturzeitung 13 (1892), cc.9-11.
HAUGEN, Einar Ingvald
1950 ‘The analysis of linguistic borrowing’, Language 26 (1950), pp.210-232.
HAVERS, W.
1914 ‘Zum Gebrauch des Dativs in den italischen Dialekten’, Glotta 5 (1914), pp.1-8.
HELBIG, G.
1883 Bullettino dell’Instituto di Corrispondenza Archeologica 1873, p.125.
1887 [Sitzungsprotocoll 18.III.1887], MDAI(R) 2 (1887), p.151.
VAN HEEMS, Gilles
2003 ‘¢s²/¢z² (à Volsinies)’, SE 69 (2003), pp.195-219.
HELBIG, Wolfgang
1913 Führer durch die öffentliche Sammlungen klassischer Altertümer in Rom, vol. 2 Die
päpstliche Sammlung im Lateran. Die staatlichen Sammlungen im Thermenmuseum,
Villa Borghese, dem Collegio Romano und dem Museo di Villa Papa Giulio.
Privatsammlungen: Palazzo Spada, Palazzo Barberini, Villa Albani, Leipzig 19133.
XXVII
BIBLIOGRAPHY
HELBIG, Wolfgang / Hermine SPEIER

1969 Führer durch die öffentlichen Sammlungen klassischer Altertümer in Rom III: Die
Staatliche Sammlungen: Museo Nazionale Romano (Thermenmuseum) - Museo
Nazionale di Villa Giulia, Tübingen 19694.
HENZEN, Wilhelm
1844 ‘Epigrafe latina di tomba etrusca’, Bullettino dell’Instituto di Corrispondenza
1864 ‘Scavi capenati’, Bullettino dell’Instituto di Corrispondenza Archeologica 1864,
pp.143-150.
1865 ‘Varietà epigrafiche’, Bullettino dell’Instituto di Corrispondenza Archeologica 1865,
pp.266-267.
HERBIG, Gustav
1904a ‘Vorarbeiten zum Corpus inscriptionum etruscarum. Ein Reisebericht’, Sitzungs-
berichte der kaiserlichen bayerischen Akademie der Wissenschaften 1904, pp.283-
296.
1904b ‘Corchiano’, in: A. TORP & G. HERBIG, ‘Einige neugefundene etruskische Inschrift-
en’, Sitzungsberichte der kaiserlichen bayerischen Akademie der Wissenschaften
1904, pp.489-520 (pp.518-519).
1908 ‘Bericht über eine Reise nach Italien im Frühjahr 1908’, Sitzungsberichte der
kaiserlichen bayerischen Akademie der Wissenschaften 1908, pp.30-34*.
1910 ‘Falisca’, Glotta 2 (1910), pp.83-110, 181-200.
1913a ‘Neue etruskische Funde aus Grotte S. Stefano und Montagna’, Glotta 4 (1913),
pp.165-187.
1913b ‘Altitalische Verbalformen’, IF 32 (1913), pp.71-87.
1914a ‘Die faliskischen Kasusendung -oi’, Glotta 5 (1914), pp.237-249.
1914b ‘Zur Mouillierung des l im Vulgärlateinischen’, Glotta 5 (1914), pp.249-253.
1917a ‘Etruskisches Latein’, IF 37 (1916-1917), pp.163-187.
1917b [review of F.Sommer, Handbuch der lateinischen Laut- und Formenlehre. Eine
Einführung in das sprachwissenschaftliche Studium des Lateins, Heidelberg 19142-3],
IF 37 (1916-1917), Anzeiger, pp.19-40.
1923 ‘Faliskischer Literaturbericht’, Glotta 12 (1923), pp.230-234.
HERBIG, Gustav, O.A. DANIELSSON, & B. NOGARA
1912 Corpus inscriptionum Etruscarum, volumen II, sectio II, fasciculus 1: Ager Faliscus
et Capenas, Lipsiae 1912 (= G. HERBIG, Tituli Faleriorum veterum, linguis Falisca et
Etrusca conscripti, diss. Lipsiae 1910).
HEURGON, Jacques
1961 La vie quotidienne chez les Étrusques, Paris 1961.
1977 ‘Onomastique étrusque: la dénomination gentilice’, in: N. D UVAL, D. BRIQUEL, & M.
HAMIEAUX (eds.), L’onomastique latine : Actes du Colloque International sur
l’onomastique latine organisé à Paris du 13 au 15 octobre 1975 [Colloques interna-
tionaux du Centre National de la Récherche Scientifique, 564], pp.25-34, Paris 1977.
HIERSCHE, Rolf
1965 ‘Der Wechsel zwischen anlautendem f und h im Lateinischen’, Glotta 43 (1965),
pp.103-118.
HIRATA, Ryuichi
1967 L’onomastica falisca e i suoi rapporti con la latina e l’etrusca [Biblioteca di Studi
Etruschi 3], Firenze 1967.
HOFMANN, Johann Baptist
1940 ‘Faliskisches’, in: ‘Altitalische Sprachdenkmäler. 1931-1937’, Jahresbericht über die
Fortschritte der classischen Alterthumswissenschaft 270 (1940), pp.1-122 (pp.119-
122).
XXVIII
BIBLIOGRAPHY
HOLLAND, Louise (also ADAMS HOLLAND, Louise)

1925 The Faliscans in prehistoric times [Papers and Monographs of the American
Academy in Rome 5], Roma 1925.
HÜBNER, Wolfgang
1969 ‘Saurix - ein Vogel ?’, Glotta 47 (1969), pp.266-279.
1971 ‘Sorex und bubo’, Glotta 49 (1971), pp.147-149.
HÜLSEN, Christian
1909a ‘Falerii’, in: Pauly-Wissowa, RE 6.1969-1972, Stuttgart 1909.
1909b ‘Falisci’, in: Pauly-Wissowa, RE 6.1972-1973, Stuttgart 1909.
VAN DER HULST, Harry, & Norval SMITH (eds.)
1983 The structure of phonological representation [Linguistic Models 2-3], Dordrecht /
Cinnaminson 19832, 1983.
INNOCENTI PROSDOCIMI, Ester
1976 ‘Area venetica - Este e territorio’, SE 44 (1976), pp.268-274.
JACOBSOHN, Hermann
1910 Altitalische Inschriften [Kleine Texte für theologische und philologische Vorlesungen
und Übungen 57], Bonn 1910.
1911 [review of Herbig 1910b], Berliner Philologische Wochenschrift 31 (1911), cc.463-
466.
JANSON, Tore
1971 ‘The Latin third declension’, Glotta 41 (1971), pp.111-142.
1979 Mechanisms of language change in Latin [Acta Universitatis Stockholmiensis - Studia
Latina Stockholmiensia XXIII], Stockholm 1979.
JIMENEZ ZAMUDIO, Rafael
1986a Estudio del dialecto peligno y su entorno lingüístico [Acta Salamanticensia, serie
varia, filosofía y letras 173], Salamanca 1986.
1986b Nuevas consideraciones acerta del resultado de la desinencia verbal indoeuropea
*-nt en latin y en las diversas lenguas itálicas [Acta Salmanticensia, serie varia, temas
científicos, literários e históricas 66], Salamanca 1986.
JONES, D.
1949 ‘The relation of Latin to Osco-Umbrian’, TPhS 1949, pp.60-88.
JONES, Geraint Dyfed Barri
1962 ‘Capena and the ager Capenas’, PBSR 31 (1962), pp.116-207.
JORDAN, H.
1881 ‘Faliskisches’, Hermes 16 (1881), pp.510-512.
JOSEPH, Brian D., & Rex J. WALLACE
1987 ‘Latin sum / Oscan súm, sim, esum’, AJPh 108 (1987), pp.675-693.
1991 ‘Is Faliscan a local Latin patois?’, Diachronica 8 (1991), pp.159-186.
1993 ‘On the development of *gh and *g’h in Faliscan. A response to Picard’, Diachronica
10 (1993), pp.144-150.
JOSEPH, Lionel S., & Jared S. KLEIN
1981 ‘A new restoration in the Faliscan Ceres-inscription with notes on Latin molere and its
Italic cognates’, HSPh 85 (1981), pp.293-300.
KAIMIO, Jorma
1969 ‘The nominative singular in -i of Latin gentilicia’, Arctos 6 (1969), pp.32-42.
1972 Studies in the Romanization of Etruria 3: The ousting of Etruscan by Latin in Etruria
[Acta Instituti Romani Finlandiae V], diss. Helsinki, Rome 1972.
KAJANTO, Iiro
1965 The Latin cognomina [Societas Scientiarum Fennica - Commentationes Humanarum
Litterarum XXXVI.2], Helsinki 1965.
1967 ‘Contributions to Latin morphology’, Arctos 5 (1967), pp.67-77.
XXIX
BIBLIOGRAPHY
KAJANTO, Iiro (continued)

1977a ‘On the chronology of the cognomen in the Republican period’, in: N. D UVAL, D.
BRIQUEL, & M. HAMIEAUX (eds.), L’onomastique latine : Actes du Colloque Interna-
tional sur l’onomastique latine organisé à Paris du 13 au 15 octobre 1975 [Colloques
internationaux du Centre National de la Récherche Scientifique, 564], pp.63-70, Paris
1977.
1977b ‘On the peculiarities of women’s nomenclature’, in: N. DUVAL, D. BRIQUEL, & M.
HAMIEAUX (eds.), L’onomastique latine : Actes du Colloque International sur
tionaux du Centre National de la Récherche Scientifique, 564], pp.147-160, Paris
1977.
KAJAVA, Mika
1995 Roman female praenomina: studies in the nomenclature of Roman women Latin [Acta
Instituti Romani Finlandiae vol. XIV], Rome 1994 = diss. Helsinki 1995.
KARO, Georg [Georgius]
1896 De arte vascularia antiquissima quaestiones, diss. Bonnae 1896.
KENT, Roland G.
1913 ‘The Oscan slingshot of Saepinum’, IF 32 (1913), pp.196-202.
1933 [review of A. VON BLUMENTHAL, Die Iguvinischen Tafeln: Text, Übersetzung,
Untersuchungen, Stuttgart 1931], Language 9 (1933), pp.214-218.
KIPARSKY, Paul
1967 ‘A phonological rule of Greek’, Glotta 44 (1966-1967), pp.109-133.
KIRCHOFF, A.
1971 Studien zur Geschichte des griechischen Alphabets, Gütersloh 18874.
KNOBLOCH, Johann
1954 ‘Faliskisch kaisiosio und die Bildungsweise der Possessiv-adjecktiva im Italischen’,
in: ‘Altitalisches’, in: Ammann-Festgabe [Innsbrucker Beiträge zur Kulturwissen-
schaft Band 2, Heft 1 (6)], Innsbruck 1954, pp.34-40 (pp.34-38).
KNOBLOCH, Johann (continued)
1958 ‘Zur faliskischen Ceres-Inschrift’, RhM 101 (1958), pp.134-138.
1966 ‘Zu den faliskischen Inschriften auf dem Schnabelkrüglein aus Altfalerii (Vetter 242A
u. B)’, Sprache 12 (1966), pp.48-49.
KOCH, C.
1934 ‘Bericht über die neuesten Forschungen und Fragestellungen auf dem Gebiet der
etruskischen Religion in Deutschland’, SE 8 (1934), pp.425-30.
KOLBE, Hans-Georg
1979 ‘Emil Braun und die lateinische Epigraphik’, Mitteilungen des Deutschen Archäologi-
schen Instituts, Römische Abteilung 86 (1979), pp.529-543.
KÖRTE, Gustav
1900 ‘Theseus, zum Herakles umgewandelt, vor Minos auf einem etruskischen Spiegel’, in:
Strena Helbigiana sexagenario obtulerunt amici a.d. IIII Non. Febr. a. MDCCC
LXXXXVIIII, pp.164-170, Lipsiae 1900.
KORTLAND, F.H.H.
1980 ‘Albanian and Armenian’, HSF 94 (1980), pp.243-251.
KREBS, J. Ph. / J.H. SCHMALZ
1906 Antibarbarus der lateinischen Sprache; nebst einem kurzen Abriss der Geschichte der
lateinischen Sprache und Vorbemerkungen über reine Latinität, Basel 19067.
KRETSCHMER, Paul
1912 ‘Altlateinische Inschrift von Corchiano’, Glotta 3 (1912), pp.45-6.
1917 ‘Zwei altlateinische Inschriften von Capena’, Glotta 8 (1917), pp.137-139.
1943 ‘Die vorgriechischen Sprach- und Volksgeschichten (Fortsetzung von Gl. XXVII 231-
278)’, Glotta 30 (1943), pp.84-218.
XXX
BIBLIOGRAPHY
KROKER, E.
1888 ‘Necropolis von Falerii’, Berliner Philologische Wochenschrift 8 (1888), cc.514-515.
KRUSCHWITZ, Peter
2002 Carmina Saturnia epigraphica: Einleitung, Text und Kommentar zu den Saturnischen
Versinschriften [Hermes Einzelschriften, 84], Stuttgart 2002.
LANGSLOW, D.R.
2002 ‘Approaching bilingualism in corpus languages’, in: J.N. ADAMS, M. JANSE, & Simon
SWAIN (eds.) Bilingualism in ancient society: Language contact and the written text,
pp.23-51, Oxford 2002.
LANZI, Luigi
1824 Saggio di lingua etrusca e di altre antiche d’Italia, Firenze 18242.
LATTE, Kurt / Peter Allan HANSEN
1953 Hesychii Alexandrini lexicon [recensuit et emendavit Kurt Latte], vol. I: Α-Δ, Hauniae
1953
1956 Hesychii Alexandrini lexicon [recensuit et emendavit Kurt Latte], vol. II: Ε-Ο,
Hauniae 1956.
2005 Hesychii Alexandrini lexicon [post Kurt Latte continuans recensuit et emendavit Peter
Allan Hansen], vol. III: Π-Σ, Berlin 2005.
LATTES, Elia
1895 ‘I giudizi dello Stolz e del Thurneysen contra l’italianità dell’etrusco in relazione colle
fasce della mummia, colla pietra di Lenno, e specialmente coi novissimi fittili di
Narce’, Rivista di Filologia n.s. 1 [23] (1895), pp.449-503.
1896 ‘I giudizi dello Stolz e del Thurneysen contra l’italianità dell’etrusco in relazione colle
fasce della mummia, colla pietra di Lenno, e specialmente coi novissimi fittili di
Narce (continuazione e fine)’, Rivista di Filologia n.s. 2 [24] (1896), pp.1-43.
1904 ‘I fascicoli nono e decimo del nuovo Corpus inscriptionum etruscarum’, Studi italiani
di filologia classica 12 (1904), pp.11-120.
1908 ‘Vicende fonetiche dell’alfabeto etrusco’, RIL serie 3 12 [21] (1908).
1909 ‘Le formule onomastiche dell’epigrafia etrusca’, RIL serie 3 12 [22] (1909).
LAZZARINI, Maria Letizia
1973 ‘I nomi dei vasi greci nelle iscrizioni dei vasi stessi’, ArchClass 25-26 (1973-1974),
pp.341-375.
LAZZERONI, Romano
1962 ‘Le più antiche attestazioni del nom.pl. -ās in latino e la provenienza dei Coloni
pesaresi’, SSL 2 (1962), pp.106-122.
1965 ‘Il dativo ‘sabellico’ in -a. Contributo alla conoscenza della latinizzazione dei
Peligni’, SSL 5 (1965), pp.65-86.
LEHMANN, Christian
1979 ‘Zur Typologie des Lateinischen’, Glotta 57 (1979), pp.237-254.
LEIFER, Franz
1931 Studien zum antiken Ämterwesen I: Zur Vorgeschichte des römischen Führeramts
(Grundlagen) [Klio, Beiträge zur alten Geschichte, Beiheft 23 [X]], Leipzig 1931.
LEJEUNE, Michel
1944 ‘Notes sur la déclinaison latine’, RÉL 21-22 (1943/1944), pp.87-101.
1949 ‘Sur le traîtement osque de *-â final’, BSL 45 (1949), pp.104-110.
1950 ‘Latin et chronologie ‘italique’’, RÉL 28 (1950), pp.97-104.
1951 ‘Notes de linguistique italique, III: L’isoglosse -m/n- dans l’Occident indo-européen;
IV: Effets d’i-Umlaut en latin’, RÉL 29 (1951), pp.86-102.
1952a ‘A propos de trois inscriptions italiques’, RÉA 54 (1952), pp.340-342.
1952b ‘Notes de linguistique italique, V-VII: Les inscriptions de la collection Froehner’,
RÉL 30 (1952), pp.87-126.
XXXI
BIBLIOGRAPHY
LEJEUNE, Michel (continued)

1955 ‘Les correspondants italiques de lat. finxit et de lat. fecit’, in: Corolla linguistica -
Festschrift Ferdinand Sommer zum 80. Geburtstag am 4. Mai 1955, pp.145-153,
Wiesbaden 1955.
1957 ‘Notes de linguistique italique XIII: Sur les adaptations de l’alphabet étrusque’, RÉL
35 (1957), pp.88-106.
1964 [review of GIACOMELLI 1963], BSL 59.2 (1964), pp.89-90.
1966 ‘Notes de linguistique italique XXI: Les notations de F dans l’Italie ancienne’, RÉL 44
(1966), pp.141-181.
1967 ‘‘Fils’ et ‘fille’ dans les langues d’Italie ancienne’, BSL 62 (1967), pp.67-86.
1970 ‘Phonologie osque et graphie grecque’, RÉA 72 (1970), pp.271-316.
1971 Lepontica [Monographies linguistiques I], Paris 1971.
1972a ‘Phonologie osque et graphie grecque’, RÉA 74 (1972), pp.5-13.
1972b ‘*Aisu- «dieu» et la quatrième déclinaison latine’, BSL 67 (1972), pp.129-137.
1974 Manuel de la langue vénète, Heidelberg 1974.
1977 ‘La romanisation des anthroponymes indigènes d’Italie’, in: N. DUVAL, D. BRIQUEL,
& M. HAMIEAUX (eds.), L’onomastique latine : Actes du Colloque International sur
tionaux du Centre National de la Récherche Scientifique, 564], pp.35-41, Paris 1977.
1988 ‘Notes de linguistique italique, XXXVII: Mézence, d’un zêta à l’autre’, RÉL 66
(1988), pp.50-54.
1990 ‘Notes de linguistique italique, XXXVIII: Notes sur la dédicace de Satricum, XXXIX:
Génitif en -osio et génitif en -i’, RÉL 67 (1990), pp.60-77.
1990 ‘Vénètes de Pannonie’, CRAI 1990, pp.628-653.
1991 ‘Notes de linguistique italique, XL: «Bois!» disait ce sicule; «Je boirai» répond ce
falisque’, RÉL 68 (1991), pp.28-30.
LENCHANTIN DE GUBERNATIS, Massimo
1937 Livi Andronici fragmenta [Corpus inscriptionum Paravianum 63], Augustae Taurino-
rum 1937.
LEUMANN, Manu = M. LEUMANN, J. HOFMANN, & A. SZANTYR
1977 Lateinische Grammatik I: Lateinische Laut- und Formenlehre, München 19775.
LIGNANA, Giacomo
1887a ‘Iscrizioni Falische’, MDAI(R) 2 (1887), pp.196-202.
1887b [Sitzungsbericht der kaiserlich deutschen archäologischen Instituts 25.II.1887],
LINDERSKI, Jerzy
1958 ‘Notes on CIL I2 364’, PP 13 (1958), pp.47-50.
LINDSAY, Wallace Martin
1913 Sexti Pompei Festi de uerborum significatu quae supersunt cum Pauli epitome,
Lipsiae 1913.
LINDSAY, W.A.
1897 Handbook of Latin inscriptions illustrating the history of the Latin language, Boston /
Chicago 1897.
LOICQ, Jean
1965 [review of Giacomelli 1963], Latomus 24 (1965), pp.695-697.
LOMAS, Kathryn
1997 ‘Constructing ‘the Greek’: ethnic identity in Magna Graecia’, in: T. CORNELL & K.
LOMAS (eds.), Gender and ethnicity in ancient Italy [Accordia specialist studies on It-
aly, vol.6], pp.31-41, London 1997.
LOMMATZSCH, Ernst
Caesaris mortem, pars posterior, fasciculus 1, Berolini 1918.
XXXII
BIBLIOGRAPHY
LOMMATZSCH, Ernst, & H. DESSAU,

Caesaris mortem, pars posterior, fasciculus 2: Addenda, nummi, indices, ed. Berolini
1931.
LORETO, Luigi
1989 ‘Il conflitto romano-falisco del 241/240 a.C. e la politica romana degli anni successi-
vi’, MÉFRA 101 (1989), pp.717-737.
LUBOTSKY, Alexander
1989 ‘Against a Proto-Indo-European phoneme *a’, in: Th. VENNEMANN (ed.), The New
Sound of Indo-European: Essays in Phonological Reconstruction [Trends in Linguis-
tics: Studies and Monographs, 41], pp.53-66, Berlin / New York 1989.
LUCCHESI, Elisa
2005 ‘Old and new (un)certainties regarding the Lapis Satricanus’, Papers on Grammar
IX.1 [Proceedings of the Twelfth International Colloquium on Latin Linguistics (Bo-
logna, 9-14 June 2003)], pp.161-170, Roma 2005.
DE LUCIA BROLLI, Maria Anna
1991a L’agro falisco, Roma 1991.
1991b Civita Castellana - Il Museo archeologico dell’agro falisco, Roma 1991.
MACURDY, Grace Harriet
1921 ‘The word ‘sorex’ in C.I.L. I2 1988, 1989’, JRS 11 (1921), pp.108-110.
MAGGIANI, Adriano
1972 ‘Aska eleivana’, SE 40 (1972), pp.183-187.
1980 ‘Ager Faliscus’, SE 48 (1980), pp.403-404.
MANCINI, Alberto
1981 ‘Falisco (?) *Akarkelo- = Ocriculum’, SE 49 (1981), pp.368-371.
MANCINI, Marco
1997 Osservazioni sulla nuova epigrafe di Garigliano [Biblioteca di ricerche linguistiche e
filologiche 42.1], Roma 1997.
2002 ‘Tra linguistica ed ermeneutica. Nuove acquisizioni falische’, ILing 25 (2002), pp.23-
46.
2003 ‘Latina antiquissima II: ancora sull’epigrafe del Garigliano’, in : V. ORIOLES (cur.),
Studi in memoria di E. Coseriu [supplemento a Plurilinguismo: contatti di lingue e
culture 10 (2003)], pp.229-252.
2004 ‘Uno scioglilingua da Falerii Veteres e l’etimologia di fal. UMOM’, AGI 89 (2004),
pp.200-211.
MANIET, Albert
1972 ‘La linguistique italique’, in: ANRW I2, pp.522-592.
1975 La phonétique historique du latin dans le cadre des langues indo-européennes, Paris
19755.
MANSUELLI, Guido Achille
1943 ‘Materiali per un supplemento al “corpus” degli specchi etruschi figurati’, SE 17
(1943), pp.487-521.
1947 ‘Gli specchi figurati etruschi’, SE 19 (1946-1947), pp. 9-137.
MARAS, Daniele F.
2002 ‘Ager Faliscus: Corchiano’, SE 65-68 (2002), pp. 467-469.
MARINER, S.
1972 [review of Giacomelli 1963], Emérita 40 (1972), pp.226-227.
MARINETTI, Anna
1981 ‘Il sudpiceno come italico (e ‘sabino’?)’, SE 49 (1981), pp.114-158.
1982a ‘Atta/us : Appius; lat. atta, sabino *appa e sudpiceno apaio-. Sabini a Roma e ‘Safini’
nelle iscrizioni sudpicene’, Res Publica Litterarum 5 (1982), pp.169-181.
1982b ‘Sabina e sudpiceno’, SE 50 (1982), pp.363-369.
XXXIII
BIBLIOGRAPHY
MARINETTI, Anna (continued)

1982c Lingue e dialetti dell’Italia antica: Aggiornamento e indici, Padova 1982.
1984a ‘Il verbo italico: apporti dalle iscrizioni sudpicene’, LEFI - Quaderni di lavoro 2
(1984), pp.25-73.
1984b Lingue e dialetti dell'Italia antica: Aggiornamenti, indici [Popoli e civiltà dell’Italia
antica 6.2], Roma 1984.
1985 Le iscrizioni sudpicene I: Testi [Lingue e iscrizioni dell’Italia antica 5], Firenze 1985.
1991 ‘Sabino hehike e falisco fifiked: nota sul perfetto, tra italico e latino’, ArchClass 43
(1991), pp.597-612.
MARTZLOFF, Vincent
2006 ‘Les syntagmes picéniens povaisis pidaitúpas, me{nt}fistrúí nemúneí, trebegies titúí’,
RPh 80 (2006), pp.63-104.
MASSA, Antonius
1546 De origine, et rebus Faliscorum liber, Romae 1546.
MASSARO, Matteo
2007 ‘Metri e ritmi nella epigrafia latina di età repubblicana’, in: P. KRUSCHWITZ (ed.), Die
metrischen Inschriften der römischen Republik, pp.121-167, Berlin/New Yerk 2007.
MATTHEWS, Peter
1972 ‘Some reflections on Latin morphophonology’, TPhS 1972, pp. 59-78.
MATTIOCCO, Ezio
1986 Centri fortificati vestini, Sulmona 1986.
MAYRHOFER, Manfred
1987 ‘Die Vertretung der indogermanischen Laryngale im Lateinischen’, ZVS 60 (1987),
pp.86-108.
MEID, Wolfgang
1961a ‘Lat. Volcanus - osset. Wärgon’, IF 66 (1961), pp.125-131.
1961b ‘Etrusk. velχans, kret. #elcanoj und die angebliche Herkunft des lat. GN Volcanus aus
dem Etruskischen’, IF 66 (1961), pp.259-266.
MEILLET, Antoine
1894 ‘De quelques difficultés de la théorie des gutturales indo-européennes’, MSL 8 (1894),
pp.277-304.
1922 ‘Du nominatif et de l’accusatif’, MSL 12 (1922), Comptes rendus pp.49-55.
1931 [review of A. DEBRUNNER & J. WACKERNAGEL, Altindische Grammatik III:
Deklination der Nomina. Zahlwörter. Pronomina, vol. 2, Göttingen 1930], BSL 31
(1931), pp.62-65.
1933 Esquisse d'une histoire de la langue latine [Tradition de l’Humanisme 2], Paris 19773
= 1933.
MEISER, Gerhard
1986 Lautgeschichte der umbrischen Sprache [Innsbrucker Beiträge zur Sprachwissen-
schaft 51], Innsbruck 1986.
1987a ‘Pälignisch, Latein und Südpikenisch’, Glotta 65 (1987), pp.104-125.
1987b [review of Marinetti 1985], Kratylos 32 (1987), pp.110-118.
1998 Historische Laut- und Formenlehre der lateinischen Sprache, Darmstadt 20062=1998.
2003 Veni vidi vici: Die Vorgeschichte des lateinischen Perfektsystems [Zetemata,
Monographien zur klassischen Altertumswissenschaft, Heft 113], Heidelberg 2003.
MEISTER, Karl
1916 Lateinisch-griechische Eigennamen I: Altitalische und römische Eigennamen,
Leipzig / Berlin 1916.
MENGARELLI, Raniero
1908 ‘Olla cineraria con iscrizione falisca’, Bollettino d’Arte 2 (1908), pp.101-103
1911 ‘Sulla scoperta del Tempio di Mercurio ai «Sassi caduti» [Falerii]’, BCAR 39 (1911),
pp.62-67.
XXXIV
BIBLIOGRAPHY
VON MERCKLIN, Eugen

1930 Führer durch das Hamburgische Museum für Kunst und Gewerbe II: Griechische und
Römische Altertümer, Hamburg 1930.
1935 ‘Etruskische Keramik im Hamburgischen Museum für Kunst und Gewerbe’, SE 9
(1935), pp.311-324.
MEYER, Jørgen Christian
1983 Pre-Republican Rome: An analysis of the cultural and chronological relations 1000-
500 BC [Analecta Romana Instituti Danici, supplementum XI], Odense 1983.
MICHETTI, Laura M.
2003 La ceramica argentata e a rilievo in Etruria nella prima età ellenistica [Monumenti
Antichi serie miscellanea VIII], Roma 2003.
MICHETTI, Laura M., & Daniele F. MARAS
2003 ‘Falerii’, SE 65-68 (2002), pp. 376-378.
MICHAELIS, Adolf (also MICHAELIS, Adolph)
1862 ‘Inschriften aus Falerii’, Archäologische Zeitung 20 (1862), cc.343*-346*.
MILROY, James
2003 ‘On the role of the speaker in language change’, in: R. H ICKEY (ed.), Motives for
language change, pp.143-157, Cambridge 2003.
MINASI
1894 Civiltà Cattolica 15 (21) (1894).
MOLTONI, Vittoria
1954 ‘Gli influssi dell’osco sulle iscrizioni latine della regio I’, RIL 87 (1954), pp.193-232.
MOMMSEN, Theodor
1850 Die unteritalischen Dialekte, Leipzig 1850.
1860 [Sitzung der philosophisch-historische Klasse vom 16. Juli 1860], Monatsberichte der
königlichen Preußischen Akademie der Wissenschaften zu Berlin 1860, pp.421-460
(pp.451-456).
1865 ‘Lapidi latine del Museo Blacas’, Annali dell’Istituto di Corrispondenza Archeologica
1865, pp.308-314.
1883 Corpus inscriptionum Latinarum X: Inscriptiones Bruttiorum Lucaniae Campaniae
Siciliae Sardiniae Latinae: pars 1, Inscriptiones Bruttiorum Lucaniae Campaniae,
pars 2, Inscriptiones Siciliae et Sardiniae, Berolini 1883.
MOMMSEN , Theodor, & Wilhelm HENZEN
1863 Corpus inscriptionum Latinarum I: Inscriptiones Latinae antiquissimae ad C.
Caesaris mortem, Berolini 1863.
MONTELIUS, Oscar
1904 La civilisation primitive en Italie depuis l’introduction des métaux. Deuxième partie:
Italie centrale, Planches, 2, Stockholm 1904.
MORANDI, Alessandro
1982 Epigrafia italica [Bibliotheca archaeologica 2], Roma 1982.
MORELLI, Giuseppe
1974 ‘Un antico saturnio popolare falisco’, ArchClass 25-26 (1973-1974), pp.440-452.
MORETTI, Mario
1963 The National Museum of Villa Giulia, Roma 1963.
MORETTI, Mario (continued)
1975 Nuove scoperte e acquisizioni nell’Etruria meridionale (Mostra tenuta a Roma nel
1975), Roma 1975.
MOSCATI, Paola
1983 ‘Vignale (Falerii Veteres): Topografia dell’insediamento’, SE 51 (1985), pp.55-89.
1985a ‘La viabilità di una regione: l’Agro Falisco’, in: F. BOITANI, M. CRISTOFANI, P.
MOSCATI, & G. NARDI, Strade degli Etruschi: Vie e mezzi di communicazione
nell’antica Etruria, Roma 1985, pp.89-136.
XXXV
BIBLIOGRAPHY
MOSCATI, Paola (continued)

1985b ‘Studi sul Falerii Veteres I. L’abitato’, RAL 40 (1985), pp.45-74.
1987 ‘Studi su Falerii Veteres II. La necropoli delle Colonnette’, RAL 42 (1987), pp.39-70.
1990 ‘Nuove ricerche su Falerii Veteres’, in: CivFal, pp.141-171.
MÜLLER, Karl Otfried
1828 Die Etrusker, Breslau 1828.
MÜLLER, Roman
2001 Sprachbewußtsein und Sprachvariation im lateinischen Schrifttum der Antike
[Zetemata: Monographien zur klassischen Altertumswissenschaft, Heft 111],
München 2001.
MURRAY-THREIPLAND, Leslie, & Mario TORELLI
1970 ‘A semi-subterranean Etruscan building in the Casale Pian Roseto (Veii) area’, PBSR
37 (1970), pp.62-121.
MUST, Gustav
1953 ‘The genitive singular of o-stems in Germanic’, Language 29 (1953), pp. 301-305.
MUYSKEN, Pieter C.
1981 ‘Spaans en Quechua in Ecuador’, TTT 1 (1981), pp. 124-138.
1982 ‘Halfway between Quechua and Spanish: The case for relexification’, in:
A. HIGHFIELD & A. VALDMAN, Historicity and variation in Creole studies, pp.
52-78, Ann Arbor 1982.
1995 ‘Code-switching and grammatical theory’, in: L. M ILROY & P. MUYSKEN (eds.), One
speaker, two languages. Cross-disciplinary perspectives on code-switching, pp.177-
198, Cambridge 1995.
MURATORIUS, Ludovicus Antonius
1726 Rerum Italicarum scriptores ab anno aerae christianae quingentesimo ad millesium-
quingentesimum, quorum potissima pars nunc primum in lucem prodit ex Ambrosia-
nae, Estensis, aliarumque insignium bibliothecarum codicibus. Cum Indice locupletis-
simo, Mediolani 1726.
NASO, Alessandro
1993 ‘Ager Faliscus: Mazzano Romano’, SE 59 (1993), pp. 263-264.
NEUTSCH, B., G. Hafner, H. Luschey, & H. Wagner
1948 Die Welt der Griechen im Bilde der Originale der Heidelberger Universtätssamm-
lung: Katalog der Jubiläumaustellung zur 100-Jahr-Feier der Sammlungen des
Archäologischen Instituts Heidelberg im Sommersemester, Heidelberg 1948.
NOËL DES VERGERS, A.
1864 L’Étrurie et les Étrusques, ou dix ans de fouilles dans les Maremmes Toscanes, Paris
1862-1864.
NOGARA, Bartolommeo
1909 ‘A Grotte S. Stefano’, Ausonia 4.II (1909), c.196.
1916 ‘Alcuni appunti intorno alle iscrizioni di Vignanello’, NSA 1916, pp.85-86.
NOLL, Rudolf
1932 ‘Eine Gruppe etruskischer Spiegel’, JÖAI 28 (1932), cc.153-160.
NORDEN, Emmanuel
1939 Aus altrömischen Priesterbüchern [Acta Regiae Societatis Humaniorum Litterarum
Lundensis 29], Lund 1939.
NYMAN, Martti A.
1977 ‘Where does Latin sum come from?’, Language 53 (1977), pp.39-60
1977 ‘Mytacism in Latin phonology’, Glotta 55 (1977), pp.111-120.
1990 ‘Latin -īs ‘nom. pl.’ as an Indo-European reflex’, Glotta 68 (1990), pp.216-227.
OGILVIE, Robert Maxwell
1965 A commentary on Livy books 1-5, Oxford 1965.
1979 Early Rome and the Etruscans, [Glasgow] 19792.
XXXVI
BIBLIOGRAPHY
OLMOS ROMERA, Ricardo

2003 ‘Crátera de columnas de Cavios Frenaios’, in: P. C ABRERA, La Colección Valdéz Fisa
en el Museo Arqueológico Nacional: septiembre-novembre, 2003, pp. 430-433, Ma-
drid 2003.
2004 ‘Faliscan column crater’, in: P.G. WARDEN (ed.), Greek Vase Painting: Form, Figure,
and Narrative. Treasures of the National Archaeological Museum in Madrid, pp. 157-
159, Dallas 2004.
OLZSCHA, Karl
1963 ‘Literaturbericht über italische Sprachen 1939-1962’, Glotta 41 (1963), pp.70-138.
1965 [review of GIACOMELLI 1963], Gymnasium 72 (1965), pp.122-124.
1968 ‘Etruskisch lautn und etera’, Glotta 46 (1968), pp.212-227.
ORIOLI, F.
1854 ‘Iscrizioni scoperte a Falerii (lettera al Dott. G. Henzen)’, Bullettino dell'Instituto di
Corrispondenza Archeologica 1854, pp.XXI-XXII.
ORIOLES, Vincenzo
1972 ‘Su alcuni fenomeni di palatalizzazione e di assibilazione nelle lingue dell’Italia
antica’, Studi linguistici salentini 5.1 (1972), pp.69-100.
OSTENBERG, Carl Eric
1969 ‘An Etruscan archaic house-type not described by Vitruvius: Evidence from Luni sul
Mignone and other sites’, in: Opuscula Romana VII [Skrifter utgivna av Svenska In-
stitutet i Rom / Acta Instituti Romani Regni Suegiae XXX], pp.87-107, Lund 1940.
OSTHOFF, Hermann
1884 Zur Geschichte des Perfekts, Straßburg 1884.
NEPPI MÒDONA, A.
1926 ‘Il nuovo monumento epigrafico protoetrusco del museo Metropolitano di New-York
e la questione della provenienza dell’alfabeto in Etruria’, RAL serie 6 vol.2 (1926),
pp.492-520.
PAINO, Emma
1958 ‘Nuova iscrizione sicula’, Κώκαλος 4 (1958), pp.163-168.
PANDOLFINI, Maristella, & Aldo Luigi PROSDOCIMI
1990 Alfabetari e insegnamento della scrittura in Etruria e nell’Italia antica [Biblioteca
di Studi Etruschi, 20], Firenze 1990.
PALLOTTINO, Massimo
1933 ‘Rideterminazione morfologica’, SE 7 (1933), pp.221-243.
1951 ‘Appendice’, SE 21 (1950-51), pp.397-400.
1954 ‘Ager Faliscus et Capenas’, SE 23 (1954), pp.403-404.
1951 ‘Qualche annotazione in margine al CIE (I)’, SE 21 (1951), pp.229-237.
1978a ‘La lingua degli Etruschi’, in: LDIA, pp.429-468.
1978b ‘A proposito della nuova iscrizione latina arcaica di Satricum - interventi sulla
communicazione del dott. Stibbe (I)’, in: Archeologia Laziale: Incontro di studio del
Comitato per l’archeologia laziale [Quaderni del Centro di Studio per l'Archeologia
PARETI, Luigi
1952 Storia di Roma e del mondo romano I: L’Italia e Roma avanti il conflitto con Taranto
(1000 c. - 281 av. Cr.), Torino 1952.
PARIBENI, Roberto
1905 ‘Civitella S.Paolo - Scavi nella necropoli capenate’, NSA 1905, pp.301-362.
1906a ‘Leprignano - Scavi nella necropoli capenate’, NSA 1906, pp.178-9.
1906b ‘Necropoli del territorio capenate’, MonAl 16 (1906), cc.280-490.
1927 ‘Fiano Romano - Vasetto con iscrizione graffita’, NSA 1927, pp.370-371.
PARKER, Holt N.
1988 ‘Latin *sisô > serô and related rules’, Glotta 66 (1988), pp.221-241.
XXXVII
BIBLIOGRAPHY
PASQUI, Angelo (Angiolo)

1887 ‘Civita Castellana (antica Faleria) - Avanzi di un tempio scoperti in contrada «Celle»’,
NSA 1887, pp.92-100.
1889 ‘Fabbrica di Roma - Tomba con iscrizione falisca’, NSA 1889, pp.153-154.
1894 ‘Delle tombe di Narce e dei loro corredi’, MonAl 4 (1894), cc.399-548.
1896 ‘Poggio Sommavilla (frazione del comune di Collevecchio) - Di un’antica necropoli
scoperta a nord dell’abitato’, NSA 1896, pp.476-489.
1900 ‘Ardea (scavi della necropoli ardeatina)’, NSA 1900, pp.53-69.
1902a ‘Mazzano Romano - Scavi del principe Del Drago, nel territorio di questo comune’,
NSA 1902, pp.321-355.
1902b ‘Mazzano Romano - Scavi del principe Del Drago, nel territorio di questo comune’,
NSA 1902, pp.593-627.
PASSERI, Giovanni Battista (also: PASSERIUS, Joannes Baptista)
1767 In Thomae Dempsteri libros de Etruria regali paralipomena, quibus Tabulae eidem
operi additae illustrantur; accedunt dissertationes de re nummaria Etruscorum, de
nominibus Etruscorum, et notae in Tabulas Eugubinas. Lucae, 1767.
PAULI, Carl
1889 [review of DEECKE 1888], Neue Philologische Rundschau 1889, pp.61-63.
1891 Altitalische Forschungen III: Die Veneter und ihre Schriftdenkmäler, Leipzig 1891.
PELLEGRINI, Giovanni Battista
1964 [review of GIACOMELLI 1963], AGI 49 (1964), pp.162-169.
PERUZZI, Emilio
1963a ‘L’iscrizione di Vendia’, Maia 15 (1963), pp.89-92.
1963b ‘Un homo novus di Falerii’, PP 18 (1963), pp.435-446.
1964a ‘Iscrizioni falische’, Maia 16 (1964), pp.149-175.
1964b ‘Le epigrafi falische CIE 8190-92’, PP 19 (1964), pp.139-142.
1964c ‘Etruschi a Corchiano’, PP 19 (1964), pp.227-232.
1964d ‘Note falische’, PP 19 (1964), pp.309-312.
1965 ‘Rapporti di Falischi con la Grecia’, PP 20 (1965), pp.275-280.
1966a ‘La lamina dei cuochi falischi’, AATC 31 (1966), pp. 113-162.
1966b ‘Ferter Resius’, Maia 18 (1966), pp.277-278.
1966c ‘Il latino di Numa Pompilio’, PP 21 (1966), pp.15-40.
1967a ‘Un documento di ius sepulchri falisco’, Klio 48 (1967), pp.189-191.
1967b ‘L’iscrizione falisca delle ‘sociai’’, PP 22 (1967), pp.113-133.
1967c ‘Soresios’, PP 22 (1967), pp.292-294.
1978a ‘I Greci e le lingue del Lazio primitivo’, in: LDIA, pp.487-503.
1978b ‘On the Satricum inscription’, PP 33 (1978), pp.346-350.
1990 ‘Gli Etruschi di Corchiano’, in: CivFal, pp.277-289.
1997 ‘Il voto falisco a Minerva (CIL I2.365)’, in: PP 292 (1997), pp.61-74.
PETERSEN, H.
1962 ‘The numeral praenomina of the Romans’, TAPA 93 (1962), pp.347-354.
PETERSEN, Walter
1930 ‘The inflection of Indo-European personal pronouns’, Language 6 (1930),
pp.164-193.
PETITTI, Patrizia
1990 ‘La preistoria del territorio «falisco». Cenni introduttivi. L’età di bronzo’, in: CivFal,
pp.53-60.
PFIFFIG, Ambros Josef
1965 ‘Zu den etruskischen Inschriften auf einen Stamnos aus Falerii’, JÖAI 47
(1964-1965), Beiblatt cc.101-106.
1969 Die etruskische Sprache: Versuch einer Gesamtdarstellung, Graz 1969
XXXVIII
BIBLIOGRAPHY
PFISTER, Raimund (& Ferdinand SOMMER)

1977 Handbuch der lateinischen Laut- und Formenlehre I: Einleitung und Lautlehre,
Heidelberg 19774.
PFISTER, Raimund
1983 ‘Zur gefälschten Maniosinschrift’, Glotta 61 (1983), pp.105-119.
PICARD, Marc
1993 ‘On the evolution of PIE *gh in Latin and Faliscan’, Diachronica 10 (1993),
pp.139-143.
PINKSTER, Harm
1990 Latin syntax and semantics, London / New York 1990.
PISANI, Vittore
1933a ‘Accusativo per nominativo nel latino volgare?’, AGI 25 (1931-3), pp.138-142.
1933b Studi sulla preistoria delle lingue indoeuropee [Memorie della Reale Accademia
Nazionale dei Lincei, serie VI: Classe di scienze morali, storiche e filologiche, vol. 4,
fasc. 6], Roma 1933.
1934 ‘Zum lat. Gen.Sg. auf -ī’, Glotta 22 (1934), pp.295-296.
1935 ‘Su una iscrizione falisca recentemente scoperta’, in: ‘Epigraphica’, AGI 27 (1935),
pp.153-171 (p.167)
1937 ‘Über drei neue faliskische Inschriften’, in: Mélanges linguistiques offerts à
M. Holger Pedersen [Acta Jutlandica - Aarsskrift for Aarhus Universitet IX.1],
pp.230-245, Aarhus 1937.
1943 ‘Latina’, in ‘Glottica parerga’, RIL 76 (1942-1943), pp.221-263 (pp.259-263).
1946 ‘L’iscrizione falisca detta di Cerere (CIE 8079)’, Athenaeum 24 (1946), pp.50-54.
1949 ‘Der lat. -î-Genitiv und die faliskischen -osio-Bildungen’, RhM 98 (1949), pp.313-
324.
1952 ‘Il gen. sg. dei temi in -o- in veneto e in latino’, Acme 5 (1952), pp.521-530.
1953a Manuele storico della lingua latina IV: Le lingue dell'Italia antica oltre il latino,
Torino 1953.
1953b ‘Sulla iscrizione di Vendia’, SE 22 (1952/53), pp.425-426.
1955 ‘Zur lateinischen Wortgeschichte’, RhM 98 (1955), pp.184-188.
1964 Manuele storico della lingua latina IV: Le lingue dell’Italia antica oltre il latino,
Torino 19642.
1978 ‘Le lingue preromane d’Italia: origini e fortune’, in: LDIA, pp.16-78.
1981 ‘L’iscrizione paleolatina di Satricum’, Glotta 59 (1981), pp.136-140.
VON PLANTA, Robert
1892 Grammatik der oskisch-umbrischen Dialekte I: Einleitung und Lautlehre, Straßburg
1892.
1897 Grammatik der oskisch-umbrischen Dialekte II: Formenlehre, Syntax, Sammlung der
Inschriften und Glossen, Anhang, Glossar, Straßburg 1897.
POCCETTI, Paolo
1979 Nuovi documenti italici: a complemento del Manuale di E. Vetter [Orientamenti
linguistici 8], Pisa 1979.
1983 ‘Eine Spur des saturnischen Verses im Oskischen’, Glotta 61 (1983), pp.207-218.
1997 ‘Etrusco Feluske = Faliscus? Note sull’iscrizione della stele arcaica di Vetulonia’, SE
63 (1997), pp.281-291.
2005 ‘Notes de linguistique italique. Nouvelle série, 2. En marge de la nouvelle attestation
du perfectum falisque faced/facet, le latin de Préneste vhevhaked, et le falisque fifi-
ked’, RÉL 83 (2005), pp.27-35.
2007 ‘Inschriftliche Dichtung in den übrigen Sprachen Altitaliens’, in: P. KRUSCHWITZ
(ed.), Die metrischen Inschriften der römischen Republik, pp.241-259, Berlin/New
York 2007.
XXXIX
BIBLIOGRAPHY
POCCETTI, Paolo (continued)

forthc. ‘Il vaso iscritto dalla necropoli di Magliano Sabina. Contributo ai rapport tra ambiente
falisco e quello sabino archaico’, forthcoming (2008?).
POLIDORI, G.
1977 ‘Ager Nepesinus’, SE 45 (1977), pp.296-297.
POLLAK, L.
1906 Catalogo della collezione oggetti di scavo del fù prof. Prospero Sarti: sculture in
marmo e bronzo, terrecotte, oreficerie, vetri, smalti, camei e pietre incise, avori ed os-
si, ecc. [Galleria Sangiorgi, anno 16, n. 109], Roma 1906.
POLOMÉ, Edgar
1992 ‘Indo-European and substrate languages in the West’, AGI 77 (1992), pp.66-85.
PORZIO GERNIA, Maria Luisa
1986 ‘Latin declension: a theoretical and methodological approach’, in: G. CALBOLI,
Papers on grammar II, pp.1-18, Bologna 1986.
POTTER, Timothy W.
1976 A Faliscan town in South Etruria: Excavations at Narce 1966-71, London 1976.
1979 The changing landscape of South Etruria, New York 1979.
POULTNEY, James
1951 ‘Volscians and Umbrians’, AJPh 72 (1951), pp.113-127.
1959 The bronze tables of Iguvium [Philological Monographs published by the American
Philological Association 18], Baltimore / Oxford 1959.
POUPE, Jean
1963 ‘Les aryballes de bucchero imitant des modèles protocorinthiens’, in: Franz de Ruyt
(red.), Études étrusco-italiques: Mélanges pour le 25e anniversaire de la chaire
d’Étruscologie à l’Université de Louvain [Recueil de travaux d’histoire et de philo-
logie, 4e série, fasc. 31], pp.227-260, Louvain 1963.
PROSDOCIMI, Aldo Luigi
1962 ‘Le iscrizioni venetiche su pietra di Padova e di Vicenza’, AIV 120 (1961-1962),
pp.699-765.
1965 ‘Per una edizione delle iscrizioni della Val Camonica’, SE 33 (1965), pp.575-599.
1974 [appended to Colonna 1974], SE 42 (1974), p.386.
1979 ‘Le iscrizioni italiche. Acquisizioni, temi, problemi’, in: Le iscrizioni pre-latine in
Italia (Colloquio Roma, 14-15 marzo 1977) [Atti dei Convegni dell'Accademia Na-
zionale dei Lincei 39], pp.119-204, Roma 1979.
1981 ‘Studi sul latino arcaico II: saluetod tita di Gabii e l’iscrizione di Vendia’, SE 50
(1981), pp.
1983 [premessa to the 1983 anastatic reprint of DEVOTO 1944], Roma 1983.
1984a ‘Helbig med fefaked? - sull’autenticità della fibula prenestina: riflessioni angolate
dall’epigrafe’, LEFI - Quaderni di lavoro 2 (1984), pp.77-112.
1985a ‘L’etrusco e le lingue dell’Italia antica’, in: Atti del Convegno della Società Italiana di
Glottologia (Pisa, 8-9 dicembre 1984), Pisa [s.a.] (pp.53-68).
1985b ‘Sull’iscrizione di Satricum’, GIF 15 (1985-7), pp.183-230.
1985c ‘I Sabini nelle iscrizioni sudpicene’, in: B. RIPOSATI (ed.), Preistoria, storia e civiltà
dei Sabini [Convegno di Studio del Centro di Studi Varroniani, Rieti, ottobre 1982],
pp.35-56, Rieti 1985.
1985d ‘Tavole Iguvine: Va-b7’, AION (arch.) 7 (1985), pp.37-88.
PROSDOCIMI, Aldo Luigi (continued)
1986 ‘Sull’accento latino e italico’, in: Etter 1986, pp.601-618.
1987 ‘‘Sabinità’ e (pan)italicità linguistica’, DA 1 (1987), pp.53-64.
1990 ‘Vetter 243 e l’imperativo latino tra (con)testo e paradigma’, in: CivFal, pp.292-326.
PULCINI, C.
1974 Falerii Veteres, Falerii Novi, Civita Castellana, Civita Castellana 1974.
XL
BIBLIOGRAPHY
PULGRAM, Ernst
1950 ‘Spoken and written Latin’, Language 26 (1950), pp.458-466.
1958 The tongues of Italy: prehistory and history, Cambridge Mass. 1958.
1976 ‘The Volscian tabula Veliterna: a new interpretation’, Glotta 54 (1976), pp.253-261.
1978 Italic, Latin, Italian, Heidelberg 1978.
QUILICI, Lorenzo
1990 ‘La via di Fantibassi e le vie cave del territorio Falisco’, in: CivFal, pp.197-223.
RADKE, Gerhard
1962 [review of Le Bonniec 1958], Gymnasium 69 (1962), pp.140-144.
1964 ‘Die Erschließung Italiens durch die römischen Strassen’, Gymnasium 71 (1964),
pp.204-235.
1965a ‘Faliskisch loufir - unbezeugt’, Glotta 43 (1965), pp.132-140.
1965b Die Götter Altitaliens [Fontes et commentationes, Heft 3], Münster 1965.
1966 ‘Lateinisch stlimen ‘Kampf, Streit’’, Glotta 44 (1966), pp.34-41.
1994 ‘Die Ceres-Inschrift in Falerii’, Latomus 53 (1994), pp.105-109.
RADT, Stefan
2003 Strabons Geographika, Band 2: Buch V-VIII: Text und Übersetzung, Göttingen 2003.
2007 Strabons Geographika, Band 6: Buch V-VIII: Kommentar, Göttingen 2007.
RASMUSSEN, Tom
1997 ‘The Tarquins and ‘Etruscan Rome’’, in: T. CORNELL & K. LOMAS (eds.), Gender
and ethnicity in ancient Italy [Accordia specialist studies on Italy, vol.6], pp.23-30,
London 1997.
LA REGINA, Adriano
1968 ‘Ricerche sugli insediamenti vestini’, MAL serie 8, 13 (1968), pp.363-446.
1981 ‘Appunti su entità etniche e strutture istituzionali nel Sannio antico’,
AION (arch.) 3 (1981), pp.129-137.
1987 ‘Capestrano’, in: V. CIANFARANI, L. FRANCHI DALL’ORTO, & A. LA REGINA,
Culture adriatiche antiche di Abruzzo e di Molise, pp.308-316, Roma 1978.
RENFREW, Colin
1987 Archaelogy and language: the puzzle of Indo-European origins, New York 1987.
RENOU, Louis
1966 La grammaire de Pāṇini. Texte sanskrit, traduction française avec extraits des
commentaires II: Adhyāya 5 à 8 [Bibliothèque de l’École Française d’Extrème-
Orient], Paris 1966.
RENZETTI MARRA, Stefania
1974 ‘Falisco’, SE 2 (1974), pp.350-357.
1990 ‘Inscrizioni neofalische inedite’, in: CivFal, pp.327-340.
RIBEZZO, Francesco
1914 [review of KENT 1913], Neapolis 2 (1914), pp.108-109.
1918 ‘L’iscrizione di due patere falische’, RIGI 2 (1918), pp.245-251.
1920 ‘Esistenza ed uso del genitivo dedicatorio nel latino preletterario: DEVAS |
CORNISCAS | SACRUM: CIL. I 814 = VI 96’, RIGI 4 (1920), pp.77-81.
1927 ‘Note sul dialetto falisco’, RIGI 11 (1927), pp.148-152.
1929 ‘I testi etruschi CIE 5237 e 4538 (Piombo di Magliano e Cippo di Perugia) rianalizzati
e spiegati’, RIGI 13 (1929), pp.59-104.
1930 ‘Roma delle origini, Sabini e Sabelli (Aree dialettali, iscrizioni, isoglossi)’, RIGI 14
(1930), pp.59-99.
1931a ‘Le iscrizioni di lingua mista egeo-etrusca e veneto-tracia nella stele di Lemno’, RIGI
15 (1931), pp.63-78.
1931b ‘Scoperte ed inedite II : Le nuove iscrizioni etrusche di Veio’, RIGI 15 (1931), pp.92-
99.
XLI
BIBLIOGRAPHY
RIBEZZO, Francesco (continued)

1933 ‘Nota d’appendice. 1. Sull’origine del gen. sg. in -ī. 2. Sull’isoglossa eu > ou’, RIGI
17 (1933), pp.79-80.
1934 ‘Di un caso di diminutivo continuativo nel dialetto falisco del VI sec. a. Cr.’, RIGI 18
(1934), pp.226-227.
1936 ‘Falisci e falisco alla luce delle nuove iscrizioni di Civita Castellana’, RIGI 20 (1936),
pp.143-172.
RICHTER, Gisela
1975 The Metropolitan Museum of Art. Greek, Etruscan and Roman Bronzes, New York
1975.
RIDGWAY, David
1977 ‘Manios faked?’, BICS 24 (1977), pp.17-30.
RIIS, Pål Jörgen
1981 Etruscan types of heads: A revised chronology of the archaic and classical terracottas
of Etruscan Campania and Central Italy [Kongelige Danske Videnskabernes Selskab,
Historisk-filosofiske Skrifter 9.5], København 1981.
RITSCHL, Friedrich
1859 ‘Epigraphische Briefe zu Herrn Professor Th. Mommsen 4: I longa und apex’, RhM
14 (1859), pp.378-388.
1862 Priscae Latinitatis monumenta epigraphica ad archetyporum fidem exemplis litho-
graphis repraesentata [Corpus inscriptionum Latinarum I: Tabulae lithographae],
Berlin 1862.
RIX, Helmut
1957 ‘Sabini, Sabelli, Samnium. Ein Beitrag zur Lautgeschichte der Sprachen Altitaliens’,
BN 8 (1957), pp.127-143
1964 [review of GIACOMELLI 1963], Gnomon 36 (1964), pp.444-449.
1965 Das etruskische Cognomen: Untersuchungen zu System, Morphologie und Verwen-
dung der Personennamen auf den jüngeren Inschriften Nordetruriens, Wiesbaden
1965.
1966 ‘Die lat. Synkope als historisches und phonologisches Problem’, Kratylos 11 (1966),
pp.156-165.
1976 ‘Umbrisch ene ... kupifiaia’, MSS 34 (1976), pp.151-164.
1981 ‘Rapporti onomastici fra il panteon etrusco e quello romano’, in: G. Colonna (ed.), Gli
Etruschi e Roma, pp.104-126, Roma 1981.
1984 ‘La scrittura e la lingua’, in: Mauro CRISTOFANI (ed.), Gli etruschi: una nuova
immagine, pp.200-230, Firenze 1984.
1992a ‘Una firma paleo-umbra’, AGI 77 (1992), pp.243-252.
1992b ‘La lingua dei Volsci. Testi e parentela’, in: S. QUILICI GIGLI (ed.), I Volsci: undice-
simo incontro di studio del comitato per l’archeologia laziale [= Archeologia laziale
XI,1], pp.37-49, Roma 1992.
1993a [review of CivFal], Kratylos 38 (1993), pp.83-87.
1993b ‘Osk. úpsannam - uupsens und Zugehöriges’, in: F. HEIDERMANNS, H. RIX, E.
SEEBOLD (red.), Sprachen und Schriften des antiken Mittelmeerraums: Festschrift für
Jürgen Untermann zum 65. Geburtstag, pp.329-348, Innsbruck 1993.
1994 Die Termini der Unfreiheit in den Sprachen Alt-Italiens [Forschungen zur antiken
Sklaverei, Band 25], Stuttgart 1994.
1996 [review of SCHRIJVER 1991], Kratylos 41 (1996), pp.153-1163.
1998a ‘Eine neue frühsabellische Inschrift und der altitalische Präventiv’, HSF 111 (1998),
247-269.
1998b ‘Teonimi etruschi e teonimi italici’, Annali della Fondazione per il Museo ‘Claudio
Faina’ 5 (1998), pp.207-229.
XLII
BIBLIOGRAPHY
RIX, Helmut (continued)

2002 Sabellische Texte: die Texte des Oskischen, Umbrischen und Südpikenischen
[Indogermanische Bibliothek, Erste Reihe] Heidelberg 2002. (= Handbuch der
italischen Dialekte, begründet von E. VETTER, V. Band).
RIX, Helmut (& Gerhard MEISER)
1991 Etruskische Texte. Editio minor, Band I: Einleitung, Konkordanz, Indices
[ScriptOralia 23, Reihe A: Altertumswissenschaftliche Reihe, Band 6], Tübingen
1991; Band II: Texte [ScriptOralia 24, Reihe A: Altertumswissenschaftliche Reihe,
Band 7], Tübingen 1991.
DE ROSALIA, Antonino
1978 Iscrizioni latine archaiche [Hermes, Collana di testi antichi 8], Palermo 1978.
ROSENKRANZ, Bernhard
1933 ‘Die Stellung des attributiven Genetivs im Italischen’, IF 51 (1933), pp.131-139.
ROSENZWEIG, Irene
1937 Ritual and cults of Pre-Roman Iguvium [Studies and Documents IX], London 1937.
ROSS, Malcolm
2003 ‘Diagnosing prehistoric language contact’, in: Raymond H ICKEY (ed.), Motives for
language change, pp.174-198, Cambridge 2003.
DE ROSSI, Giovanni Battista
1883 ‘Frammenti di fasti di ludi Capenati’, Annali dell'Instituto di Corrispondenza
ROYMANS, Nico
2004 Ethnic identity and imperial power: the Batavians in the early Roman Empire
[Amsterdam Archaeological Studies, vol. 10], Amsterdam 2004.
RUNDGREN, Frithiof
1960 ‘Der Genetivus aestimativus im Lateinischen: einige Bemerkungen’, Eranos 58
(1960), pp.51.
DE RUYT, Claire
1977 ‘Ager Volcentanus: Castro’, SE 45 (1977), pp.288-291.
RYBERG, Inez (also SCOTT RYBERG, Inez)
1940 An archaeological record of Rome from the seventh to the second century B.C.,
London / Philadelphia 1940.
SAFAREWICZ, Jan
1950 ‘La valeur phonologique des diphtonges latines’, Eos 44 (1950), pp.123-130.
1953 [review of PISANI 1953], Eos 46.2 (1952-1953), pp.237-247.
1954 ‘Notes sur les langues italiques’, Eos 47 (1954), pp.101-105.
1955 ‘Dwa studia nad językiem Falisków’, Eos 47.2 (1955), pp.175-190.
1978 ‘A propos du latin sum’, Eos 66 (1978), pp.305-307.
SALMON, Edward Togo
1987 Roman colonization under the Republic, London 1969.
SALOMIES, Olli
1987 Die römischen Vornamen: Studien zur römischen Namengebung [Societas Scientia-
rum Fennicarum - Commentationes Humanarum Litterarum 82], Helsinki 1987.
SCHAUENBURG, Konrad
1960 Perseus in der Kunst des Altertums, Bonn 1960.
SCHERER, Anton
1956 [review of Vetter 1953], Gymnasium 63 (1956), pp.116-118.
SCHINDLER, Jochem
1980 ‘Zur Herkunft der altindischen cvi-Bildungen’, in: M. MAYRHOFER, M. PETERS, &
O.E. PFEIFFER, Lautgeschichte und Etymologie [Akten der VI. Fachtagung der Indo-
germanischen Gesellschaft (Wien, 24.-29. September 1978)], pp.386-393, Wiesbaden
1980.
XLIII
BIBLIOGRAPHY
SCHIPPA, Ferruccio
1980 Officine ceramiche falische. Ceramica a vernice nera nel Museo di Civita Castellana
[Archeologia: Materiali e Problemi 3], [Bari 1980].
SCHMID, Wolfgang P.
1966 ‘Zur Goldfibel von Praeneste’, IF 70 (1965-1966), pp.200-208.
SCHMIDT, Gernot
1992 ‘Indogermanische Ordinalzahlen’, IF 97 (1992), pp.197-235.
SCHMIDT, Johannes
1900 ‘Die kretischen pluralnominative auf -εν und verwandtes’, ZVS 16 [36] (1900),
pp.400-416.
1905 ‘Zur geschichte der langdiphthonge im Griechischen’, ZVS 18 [38] (1905), pp.1-47.
SCHMOLL, Ulrich
1962 ‘Zu den vorgriechischen Inschriften Siziliens und Süditaliens’, Glotta 40 (1962),
pp.54-62.
1968 ‘Neues zu den protosizilischen Inschriften’, Glotta 46 (1968), pp.194-206.
SCHNEIDER, Engelbert
1886 Dialectorum Italicarum aevi vetustioris exempla selecta I: Dialecti Latinae priscae et
Faliscae exempla selecta, Lipsiae 1886.
SCHÖN, Ilse
1971 Neutrum und Kollektivum: Das Morphem -a im Lateinischen und Romanischen [Inns-
brucker Beiträge zur Sprachwissenschaft 6], Innsbruck 1971.
SCHRIJVER, Peter
1991 The reflexes of the Proto-Indo-European laryngeals in Latin [Leiden Studies in Indo-
European 2], Amsterdam / Atlanta 1991.
SCHULZE, Wilhelm
1904 Zur Geschichte lateinischer Eigennamen [Abhandlungen der Königliche Gesellschaft
der Wissenschaften, Philologisch-Historische Klasse, n.F. V.5], Göttingen 1904.
SCOTT RYBERG, Inez (also: RYBERG, Inez)
1940 An archaeological record of Rome from the seventh to the second century B.C.,
London / Philadelphia 1940.
SCULLARD, Howard Hayes
1967 The Etruscan cities and Rome, London 1967.
SFORZINI, Clementina
1985 ‘Nota bibliografica sulla storia del Museo di Villa Giulia’, SE 51 (1985), pp.538-543.
DELLA SETA, Alessandro
1918 Museo di Villa Giulia I: Villa Giulia - Necropoli dell’agro falisco - Templi dell’agro
falisco e del Lazio - Satricum - Capena - Praeneste (Collezione Barberini) - Cista Fi-
coroni, Roma 1918.
SGUBINI MORETTI, Anna Maria
1975 Nuove scoperte e acquisizioni nell’Etruria Meridionale, Roma 1975.
SHIELDS, Kenneth
1980 ‘The Oscan-Umbrian third person plural’, Glotta 59 (1980), pp.68-78.
SHOTTER, David C.A.
1976 ‘Rome, the Faliscans and the Roman historians’, in: POTTER 1976, pp.29-35.
SIEBERT, Gérard.
1978 ‘Signatures d’artistes, d’artisans et de fabricants dans l’antiquité classique’, Ktema 3
(1978), pp.111-131.
SIEBOURG, M.
1897 ‘Italische Fabriken ‘megarischer’ Becher’, MDAI(R) 12 (1897), pp.40-55.
SILVESTRI, Domenico
1993 ‘I più antichi documenti epigrafici del latino’, in: C AMPANILE 1993, pp.97-118.
XLIV
BIBLIOGRAPHY
DE SIMONE, Carlo
1962 ‘Die messapische Sprache (seit 1939)’, Kratylos 7 (1962), pp.113-135.
1968 Die griechischen Entlehnungen im Etruskischen, I: Einleitung und Quellen,
Wiesbaden 1968.
1969 [review of PISANI 1964], IF 74 (1969), pp.246-263.
1970 Die griechischen Entlehnungen im Etruskischen, II: Untersuchung, Wiesbaden 1970.
1972 [review of HIRATA 1967], BN 7 (1972), pp.217-218.
1975 ‘Etruskischer Literaturbericht: neuveröffentlichte Inschriften 1970-1973’, Glotta 53
(1975), pp.125-181.
1978 ‘A proposito della nuova iscrizione latina arcaica di Satricum - interventi sulla
communicazione del dott. Stibbe (I)’, in: Archeologia Laziale: Incontro di studio del
Comitato per l’archeologia laziale [Quaderni del Centro di Studio per l’Archeologia
1979 ‘Il messapico’, in: Atti dei Convegni dell’Accademia Nazionale dei Lincei 39,
Colloquio ‘Le iscrizioni pre-latine in Italia’ (Roma, 14-15 marzo 1977), pp.105-118,
Roma 1979.
1980 ‘L’aspetto linguistico’, in: Lapis Satricanus: Archaeological, epigraphical, linguistic
and historical aspects of the new inscription from Satricum [Archeologische Studiën
1980.
1983 [appended to COLONNA 1983], SE 51 (1983), pp.591-594.
2006 ‘Falisco faced – Latino arcaico vhevhaked: la genuinità della fibula prenestina e
problemi connessi’, ILing 20 (2006), pp.159-175.
SIRONEN, Timo
1989 ‘Una rilettura e un’integrazione di due epigrafi osche di Ercolano (Po 124 e 126)’,
Arctos 23 (1989), pp.185-193.
SITTIG, Ernst
1932 ‘Zu den faliskischen Inschriften CIE 8179 und 8180’, in: ‘Epigraphica’, in: Symbolae
philologicae O.A. Danielsson octogenario dicatae, Uppsala 1932, pp.305-316
(pp.315-316).
SKUTSCH, Otto
1975 ‘Genetival -AI and -AE in Ennius’, Glotta 53 (1975), pp.121-123.
1985 The Annals of Q. Ennius, Oxford 1985.
SKYDSGAARD, Jens Erik
1974 ‘Transhumance in ancient Italy’, ARID 7 (1974), pp.7-36.
SLOTTY, Friedrich
1952 Beiträge zur Etruskologie I: Silbenpuntierung und Silbenbildung im Altetruskischen
[Bibliothek der allgemeinen Sprachwissenschaft, III. Reihe: Darstellungen und
Untersuchungen aus einzelnen Sprachen], Heidelberg 1952.
SMITH, Anthony D.
1986 The ethnic origins of nations, Oxford 1986.
1991 National identity, London 1991.
SOLIN, Heikki
1972 ‘Analecta epigraphica VII. Zur römischen Namengebung auf Delos’, Arctos 7 (1972),
pp.163-167.
1977 ‘Die innere Chronologie des römischen Cognomens’, in: N. D UVAL, D. BRIQUEL, &
M. HAMIEAUX (eds.), L’onomastique latine: Actes du Colloque International sur
tionaux du Centre National de la Récherche Scientifique, 564], pp.103-146, Paris
1977.
1996 Die stadtrömischen Sklavennamen: ein Namenbuch [Forschungen zur antiken
Sklaverei, Beiheft 2], Stuttgart 1996
XLV
BIBLIOGRAPHY
SOLIN, Heikki (continued)

1999 ‘Epigrafia repubblicana: bilancia, novità, prospettive’, in: S. EVANGELISTI, L.
GALLA, M.G. MACCIOCCA, D. NONNIS, & E. ZAPPATA, XI Congresso internazio-
nale di epigrafia greca e latina (Roma, 18-24 settembre 1997), pp.379-404, Roma
1999.
SOLIN, Heikki, & Olli SALOMIES
1994 Repertorium nominum gentilium et cognominorum Latinorum [Alpha-Omega Reihe
A: Lexika, Indizes, Konkordanzen zur klassischen Philologie, LXXX], Hildesheim/
Zürich/New York 1994.
SOLMSEN, Friedrich
1912 ‘Σιληνός Σάτυρος Τίτυρος’, IF 30 (1912), pp.1-47.
SOLTA, Georg Renatus
1974 Zur Stellung der lateinischen Sprache [Österreichische Akademie der Wissenschaften,
philosophisch-historische Klasse, Sitzungsberichte, Band 291, Abhandlung 4.
Veröffentlichungen der Kommission für Linguistik und Kommunikationsforschung,
Heft 2], Wien 1974.
SOMMER, Ferdinand
1902 Handbuch der lateinischen Laut- und Formenlehre: eine Einführung in das
sprachwissenschaftliche Studium des Lateins, Heidelberg 1902.
1914 ‘Der italische Pronominalstamm eo-’, Glotta 5 (1914), pp.253-258.
1918 ‘Oskisch íív’, IF 38 (1917-1920), pp.171-174.
1924 ‘Zur venetischen Schrift und Sprache’, IF 42 (1924), pp.90-132.
SPADEA, Giuseppina
1972 ‘Ager Picentinus: Pontecagnano’, SE 40 (1972), pp.448-450.
1973 [addenda to Spadea 1972], SE 41 (1973), p.358.
SPINAZZOLA, Vittorio
1891 Di alcune iscrizioni e patere falische, Napoli 1891.
(SPRENGER, Maja, &) Gilda BARTOLONI
1977 Die Etrusken - Kunst und Geschichte, München 1977.
STANCO, Enrico Angelo
2001 ‘Falerii Veteres’, SE 63 (2001), pp. 482-483.
STEENSLAND, Lars
1973 Die Distribution der urindogermanischen sogenannten Gutturale [Acta Universitatis
Upsaliensis - Studia Slavica Upsaliensia 12], diss. Uppsala 1973.
STEFANI, Enrico
1958 ‘Capena. Scoperte archeologiche nell’agro capenate. Ricerche archeologiche nella
contrada “le Saliere”’, MonAl 44 (1958), cc.1-204.
DI STEFANO MANZELLA, Ivan
1977 ‘I nomi attribuiti alle due Falerii dalla tradizione letteraria antica e dalle epigrafi’, Atti
della Pontifica Accademia Romana di Archeologia, serie III: Rendiconti 49 (1976-
1977), pp.151-162.
1979 Falerii Novi negli scavi degli anni 1821-1830, con un catalogo degli oggetti scoperti,
un’appendice di documenti inediti, e una pianta topografica [Atti della Pontificia Ac-
cademia Romana di Archeologia, serie III: Memorie 12.2], Roma 1979.
1990 ‘Lo stato giuridico di Falerii Novi dalla fondazione al III secolo d.C.’, in: CivFal,
pp.341-368.
1996 ‘“Furcula Falisca”. Una soluzione per l’enigmatica epigrafe latina rupestre CIL
XI.3161’, ZPE 111 (1996), pp.218-225.
STEINBAUER, Dieter H.
1999 Neues Handbuch des Etruskischen [Subsidia Classica, Band 1], St-Katharinen 1999.
STEPHENS, Laurence
1980 ‘Latin gn-: further considerations’, IF 85 (1980), pp.165-175.
XLVI
BIBLIOGRAPHY
STEPHENS, Laurence
1985 ‘New evidence concerning iambic and cretic shortening in Classical Latin’, CPh 80
(1985), pp.239-244.
STEUART, Ethel
1925 The Annals of Quintus Ennius, Cambridge 1925.
STIBBE, Coenraad Michael
1978 ‘Satricum’, in: Archeologia Laziale: Incontro di studio del Comitato per l’archeologia
laziale [Quaderni del Centro di Studio per l’Archeologia Etrusco-Italica 1], pp.56-59,
Roma 1978.
STOLTE, Erich
1926 Der faliskische Dialect, diss. München 1926.
1928 ‘Die faliskischen Personennamen’, Glotta 16 (1928), pp.287-303.
1929a ‘Zur faliskischen Ceresinschrift’, Glotta 17 (1929), pp.106-113.
1929b ‘Faliskische Bibliographie’, Glotta 17 (1929), pp.113-117.
1930 ‘Nochmals zur Ceresinschrift’, Glotta 18 (1930), pp.199-200.
STOLTENBERG, Hans L.
1956 Die wichtigsten etruskischen Inschriften. Text, Übersetzung und Erläuterung. Gottall
der Bronzeleber von Piacenza. Opferinschriften. Familie und Staat, Leverkusen 1956.
STRZELECKI, Władysław
1954 Cn. Naevii Belli Punici quae supersunt, Lipsiae 1954.
STUART-SMITH, Jane
1991 The development of the ‘voiced aspirates’ in Italic: a new look, diss. Oxford 1991.
2004 Phonetics and phonology: sound change in Italic, Oxford 2004.
STUDNICZKA, Franz
1887 [Sitzungsbericht der kaiserlich deutschen archäologischen Instituts 18.III.1887],
SZEMERÉNYI, Oswald
1952 ‘The development of the Indo-European mediae aspiratae in Latin and Italic’, ALing 4
(1954), pp.27-53.
1953 ‘The future imperative of Indo-European’, RBPh 31 (1953), pp.937-954.
1957 ‘Latin tantus quantus and the genitive of price. With an excursus on quandō and Gk.
πηνίκα’, Glotta 35 (1957), pp.92-114.
1957 ‘The genitive singular of masculine -ā-stem nouns in Greek’, Glotta 35 (1957),
pp.195-208.
1969 ‘Etyma Latina II (7-18)’, in: Studi linguistici in onore di V. Pisani, vol.2, pp. 963-994,
Brescia 1969.
1973 ‘Marked - unmarked and a problem of Latin diachrony’, TPhS 1973, pp.55-74.
1981 ‘Sprachverfall und Sprachtod besonders im Lichte indogermanischer Sprachen’, in:
Y. ARBEITMAN & A. BOMHARD, Bono homini donum: Essays in historical linguis-
tics in memory of J. Alexander Kerns [Amsterdam Studies in the Theory and History
of Linguistic Science 4: Current Issues in Linguistic Theory 16], vol.1, pp. 281-310,
Amsterdam 1981.
1989 Einführung in die vergleichende Sprachwissenschaft, Darmstadt 19893.
TALBOT, R.J.A.
2000 Barrington Atlas of the Greek and Roman World, Princeton 2000.
TAMBRONI, Filippo
1898 Note falische, Bologna 1898.
TAYLOR, Lily Ross
1920 ‘The site of Lucus Feroniae’, JRS 10 (1920), pp.29-36.
1923 Local cults in Etruria [Papers and Monographs of the American Academy in Rome
2], Roma 1923.
XLVII
BIBLIOGRAPHY
TAYLOR, Lily Ross (continued)

1960 The voting districts of the Roman republic [Papers and Monographs of the American
Academy in Rome 20], Roma 1960.
TERRENATO, Nicola
2007 ‘The clans and the peasants: reflections on social structure and change in Hellenistic
Italy’, in: P. VAN DOMMELEN & N. TERRENATO (eds.), Articulating local cultures:
power and identity under the expanding Roman Republic [Journal of Roman Archae-
ology, supplementary series 63], pp.13-22, Portsmouth (Rhode Island) 2007.
THULIN, Carl
1907 ‘Faliskische Inschriften’, MDAI(R) 22 (1907), pp.255-310.
1908 ‘Eine altfaliskische Vaseninschrift’, RhM 63 (1908), pp.254-259.
THURNEYSEN, R.
1887 ‘Lateinischer lautwandel’, ZVS 8 (1887), pp.154-162.
1888 ‘Zur Inschrift von Corfinium’, RhM 43 (1888), pp.347-354.
TILL, Rudolf
1984 Res publica: Texte zur Krise der frührömischen Traditionl, Darmstadt 1984 [=Res
publica: Texte zur Krise der frührömischen Tradition [Bibliothek der alten Welt,
Sammlungen und Anthologien], Zürich/München 1976].
TORELLI, Mario
1966 ‘Vulci’, in: Enciclopedia dell’Arte Antica Classica e Orientale VII, pp.1208-1214,
Roma 1966.
1967 ‘Ager Capenas: Civitella S. Paolo’, SE 35 (1967), pp.535-539.
1974 ‘Feronia e Lucus Feroniae in due iscrizioni latine’, ArchClass 25-26 (1973-1974),
pp.741-750.
1975 Elogia Tarquiniensia [Studi e Materiali di Etruscologia e Antichità Italiche 15],
Firenze 1975.
1981 Storia degli Etruschi, [Roma / Bari] 1981.
1985 Etruria [Guide archeologiche Laterza 3], [Roma / Bari] 19853.
TORP, Alf
1905 Etruscan notes [Videnskabs-Selskabets Skrifter II, Historisk-Filosofisk Klasse, 1905,
No.1], Cristiania 1905.
1906 Etruskische Beiträge. Zweite Reihe, I: Über einige etruskischen Gefässinschriften
[Videnskabs-Selskabets Skrifter II, Historisk-Filosofisk Klasse, 1906, No.8], Christia-
nia 1906.
TOURATIER, Christian
1975 ‘Rhotacisme synchronique du latin classique et rhotacisme diachronique’, Glotta 53
(1975), pp.246-281.
TRUBETZKOY, Nikolai I.
1939 ‘Gedanken über das Indogermanenproblem’, Acta Linguistica 1 (1939), pp.81-9.
TRUDGILL, Peter
1983 On dialect: social and geographical perspectives, Oxford 1983.
1986 Dialects in contact [Language in Society 10], Oxford 1986.
UNTERMANN, Jürgen
1957a [review of G. BOTTIGLIONI, Manuele dei Dialetti Italici (Osco, Umbro, e dialetti
minori), Bologna 1954], IF 58 (1957), pp.186-190.
1957b ‘Oskisches’, IF 58 (1957), pp.241-252.
1961 ‘Zur venetischen Nominalflexion’, IF 66 (1961), pp.105-124.
1964 [review of GIACOMELLI 1963], GGA 216 (1964), pp.171-182.
1968a ‘Zwei Bemerkungen zur lateinischen Perfektflexion’, in: M. MAYRHOFER, F.
LOCHNER-HÜTTENBACH, & H. SCHMEJA (eds.), Studien zur Sprachwissenschaft
und Kulturkunder: Gedenkschrift für Wilhelm Brandenstein [Innsbrucker Beiträge zur
Sprachwissenschaft, Band 14], pp.165-171, Innsbruck 1968.
XLVIII
BIBLIOGRAPHY
UNTERMANN, Jürgen (continued)

1968b [review of PELLEGRINI & PROSDOCIMI 1967], Kratylos 13 (1968), pp.137-145.
VAHLEN, J
1928 Ennianae poesis reliquiae iterates curis, Lipsiae 19282.
VAHLERT, K.
1933 ‘Mutunus Tutunus’, in: REA XVI, cc.979-987, Stüttgart 1933.
VENNEMANN, Theo
1988 Preference laws for syllable structure and the explanation of sound change, Berlin /
New York / Amsterdam 1988.
VERSNEL, H.S.
1980 ‘Historical implications’, in: Lapis Satricanus: Archeological, epigraphical, linguistic
and historical aspects of the new inscription from Satricum [Archeologische studiën
1980.
VETTER, Emil
1925 ‘Zur altfaliskischen Gefäßinschrift CIE 8079’, Glotta 14 (1925), pp.26-31.
1926 ‘Etruskisch ikam und das Verhältnis des Etruskischen zu den indogermanischen
Sprachen’, in: Festschrift für Universitäts-Professor Hofrat Dr. Kretschmer: Beiträge
zur griechischen und lateinischen Sprachforschung, pp.279-288, Wien / Leipzig /
New York 1926.
1932 ‘Literaturbericht 1924-1929. Italische Sprachen’, Glotta 20 (1932), pp.1-45.
1936 ‘Die Herkunft des venetischen Punktiersystems’, Glotta 24 (1936), pp.114-133.
1939a ‘Die neuen faliskischen Gefäßinschriften’, Glotta 27 (1939), pp.145-156.
1939b ‘Etruskisches’, Glotta 27 (1939), pp.157-181.
1940 ‘Literaturbericht 1935-1937. Etruskisch’, Glotta 28 (1940), pp.117-231.
1948 ‘Die etruskischen Personennamen leθe, leθi, und leθia: die Namen unfreier oder
halbfreier Personen bei den Etruskern’, JÖAI 37 (1948), Beiblatt pp.57-112.
1953 Handbuch der italischen Dialekte I: Texte mit Erklärung, Glossen, Wörterverzeichnis,
Heidelberg 1953.
1956 ‘Di Novensides, di Indigetes’, IF 62 (1956), pp.1-32.
VILLAR, F.
1986 ‘El dativo latino epigráfico en -ā’, Emérita 54 (1986), pp. 45-62.
1987a ‘Los diptongos largos en latín y en itálico’, Emérita 55 (1987), pp.43-50.
1987b ‘The latin diphthongs *-ai, *-āi in final syllables’, IF 92 (1987), pp.135-167.
VINE, Brent
1991 ‘Notes on the Corcolle Altar fragments (CIL I2 (4) 2833a)’, Glotta 69 (1991),
pp.219-234.
1993 Studies in archaic Latin inscriptions [Innsbrucker Beiträge zur Sprachwissenschaft,
Band 75] Innsbruck 1993.
WAARSENBURG, Demetrius J.
1994 The northwest necropolis of Satricum: An Iron Age cemetery in Latium vetus, part I.1:
Text, diss. Amsterdam 1994.
WACHTER, Rudolf
1987 Altlateinische Inschriften: Sprachliche und epigrafische Untersuchungen zu den
Dokumenten bis etwa 150 v. Chr. [Europäische Hochschulschriften Reihe 15,
Klassische Sprachen und Literaturen, Band 38], Bern / Frankfurt a. M. / New York /
Paris 1987.
WACKERNAGEL, Jacob
1908 ‘Genetiv und Adjektiv’, in: Mélanges de linguistique offerts à F. de Saussure, pp.
125-152, Paris 1908 (= Kleine Schriften II2, pp.1346-1373, Göttingen 1969).
XLIX
BIBLIOGRAPHY
WALLACE, Rex J.
1982 ‘A note on the phonostylistics of Latin: (s) in Plautus’, Glotta 60 (1982), pp.120-124.
1985 ‘Volscian sistiatiens’, Glotta 63 (1985), pp.93-101.
1988 ‘Dialectal Latin fundatid, proiecitad, parentatid’, Glotta 66 (1988), pp.211-220.
1990 ‘The Palaeo-Oscan Besitzerinschriften from Nocera and Vico Equense’, IF 95 (1990),
pp.171-183.
2004 ‘A Faliscan inscription in the Michael and Judith Steinhardt Collection’, ZPE 153
(2004), pp. 175-182.
WALLACE, Rex J., & Brian D. JOSEPH
1991 ‘On the problematic f/h variation in Faliscan’, Glotta 69 (1991), pp.84-93.
WARD PERKINS, John Bryan, & M. FREDERIKSEN
1957 ‘The ancient road-systems of the central and northern ager Faliscus’, PBSR 25 [12]
1957, pp.67-210.
WARMINGTON, E.H.
1935 Remains of Old Latin I: Ennius and Caecilius, Cambridge (Mass.) / London 1936
[rev. 1967].
1936 Remains of Old Latin II: Livius Andronicus, Naevius, Pacuvius and Accius, Cam-
bridge (Mass.) / London 1936 [rev. 1967].
1938 Remains of Old Latin III: Lucilius - The Twelve Tables, Cambridge (Mass.) / London
1938 [rev. 1979].
1940 Remains of Old Latin IV: Archaic inscriptions, Cambridge (Mass.) / London 1940.
WATKINS, Calvert
1995a How to kill a dragon: Aspects of Indo-European poetics, Oxford 1995.
1995b ‘Greece in Italy outside Rome’, HSCPh 97 (1995), pp.35-50.
WEINREICH, Uriel
1963 Languages in contact: Findings and problems, Den Haag / Paris / New York 19632
[=Publications of the Linguistic Circle of New York 1, New York 1953].
WEINSTOCK, Stefan
1952 ‘Parca Maurtia und Neuna Fata’, in: Festschrift Andreas Rumpf zum 60. Geburtstag
dargebracht von Freunden und Schülern, Krefeld 1952.
WEISS, Michael L.
1993 Studies in Italic nominal morphology, Ann Arbor, 1993.
WHATMOUGH, Joshua
1953 [review of Pisani 1953], Language 29 (1953), pp.192-193.
WHITNEY, William Dwight
1881 ‘On mixture in language’, TAPhA 12 (1881), pp.1-26.
WISEMAN, T.F.
1970 ‘Roman republican road-building’, PBSR 37 (1970), pp.122-152.
WISSOWA, Georg
1902 Religion und Kultus der Römer, München 1092.
ZAMBONI, Alberto
1986 ‘Tra latino e neolatino: L’evoluzione delle medie aspirate indoeuropee e le successive
ristrutturazioni del consonantismo’, IF 91 (1986), pp.205-235.
ZAMBONI, Alberto
1987 ‘Tra latino e neolatino: L’evoluzione delle medie aspirate indoeuropee e le successive
ristrutturazioni del consonantismo (II)’, IF 92 (1987), pp.112-134.
ZAZOFF, Peter
1968 Etruskische Skarabäen, Mainz a. Rh. 1968.
ZIMMERMANN, Jean-Louis
1986 ‘La fin de Falerii Veteres: un témoignage archéologique’, The John Paul Getty
Museum Journal 14 (1986), pp.37-42.
L
BIBLIOGRAPHY
ZMIGRYDER-KONOPKA, Zdzisław, & Kazimierz ROZENBERG

1936 ‘Przyczynek do interpretacji napisu wotywnego kucharzy faliskich (CIL I 2 2, 364)’,
in: Munera philologica Ludovico Ćwikliński bis sena lustra professoria claudenti ab
amicis collegis discipulis oblata, pp.337-346, Posnaniae 1936.
ZOCCA, R.
1996 ‘Inscriptiones Latinae liberae rei publicae Africae Sardiniae Corsicae’, in: M.
KHANOUSSI, P. RUGGIERI, & C. VISMARA (eds.), L’Africa Romana [Atti dell’XI con-
vegno di studio, Cartagine, 15-18 dicembre 1994], vol. III, pp.1425-1489, Cagliari
1996.
ZVETAIEFF, Johannes
1884 Inscriptiones Italiae mediae dialecticae ad archetyporum et librorum fidem, Lipsiae
1884.
1885 Inscriptiones Italiae mediae dialecticae ad archetyporum et librorum fidem: Tabulae,
Lipsiae 1885.
1886 Inscriptiones Italiae inferioris dialecticae in usum praecipue academicum, Mosquae
1886.
3. A NOTE ON SOURCES AND EDITIONS

The names of ancient authors and their works are, where possible, abbreviated according to the
conventions used in the Oxford Latin Dictionary (OLD) or the Thesaurus Linguae Latinae
(TLL), and in Liddel & Scott’s Greek-English Lexicon (LSJ).
Several authors that are quoted with some frequency are referred to as follows:
x the fragments of Livius Andronicus are quoted according to Erasmi’s Studies on the
language of Livius Andronicus (1975) and the numbering, as in this work, follows the edi-
tion of Lenchantin de Gubernatis (1937), abbreviated as L
x the fragments of Ennius are quoted according to Skutsch’s The Annals of Q. Ennius
(1985), but the numbering follows the edition of Vahlen (1928 2), abbreviated as V
x the lexicon of Hesychius is quoted and numbered according to Latte (& Hansen)’s
Hesychii Alexandrini lexicon (1953, 1956, 2005), abbreviated as L or L/H
x the lexicon of Festus and the extracts by Paulus Diaconus are numbered according the
edition by Lindsay (1913), abbreviated as L
x the texts of the Latin grammarians are numbered according to Keil’s Corpus grammati-
corum Latinorum (1855-1880), abbreviated as CGL, although the texts may follow more
recent editions
The inscriptions from the agri Faliscus and Capenas are referred to by their number in my
edition (chapters 12-19). The numbers are preceded by an abbreviation indicating in which
period or alphabet group I have placed them (see §11.1.3):
EF Early Faliscan LtF Latino-Faliscan
MF Middle Faliscan Cap Capenate
MLF Middle or Late Faliscan Lat Latin
LF Late Faliscan Etr Etruscan
The numbering, and, unless otherwise stated, the reading of the inscriptions that are not in my
edition follow wherever possible, the following editions:
x for the Latin inscriptions: Corpus inscriptionum Latinarum (CIL) or Degrassi’s Inscriptio-
nes Latinae liberae rei publicae (ILLRP, 1957, 1963)
x for the Etruscan inscriptions: Rix’s Etruskische Texte (ET, 1991)
x for the Sabellic inscriptions: Rix’s Sabellische Texte (ST, 2002)
x for the Venetic inscriptions: Lejeune’s La langue vénète (Le, 1974)
x for the other languages of ancient Italy: Conway, Whatmough & Johnson’s The Prae-Italic
dialects (PID, 1933) or Vetter’s Handbuch der italischen Dialekte I (Ve, 1953)
LI
BIBLIOGRAPHY
4. CONVENTIONS IN THE REPRESENTATION OF EPIGRAPHIC TEXTS

The following signs have been used in the representation of epigraphic texts:
* An asterisk denotes traces of a letter. The number of asterisks corresponds to the
number of damaged or illegible letters.
ạḅc ̣ Subpunction denotes that a letter is damaged and can be read in different ways.
This is always explained in the text.
[abc] Letters between square brackets denote letters of which no legible trace remains,
but are integrated according to a modern editor of the text.
[...] Points between square brackets indicate a space where no legible trace of letters
remains, but that is assumed to have contained a number of letters that cannot be
integrated, the number of dots corresponding to the number of letters that is as-
sumed to be missing.
[4-5] Numbers between square brackets indicate a space where no legible trace of
letters remains, but that is assumed to have contained a number of letters that
cannot be integrated, as indicated by the numbers. This convention is used in
cases where the number of missing letters is either very large or uncertain.
[---] Dashes between square brackets indicate an unknown number of missing letters:
combined with a question mark, the dashes indicate the possibility that more let-
ters either preceded ([---?]) or followed ([?---]) the preserved text.
‹abc› Letters between pointed brackets are assumed either (a) to have been left out by
mistake or (b) to have been written in error while another letter was intended by
the person who wrote the inscription.
{abc} Letters between accolade brackets are assumed to have been added in error by
the person who wrote the inscription.
 These dots represent the various types of interpuncts as used in the text.
| The long vertical line is used to indicate (a) a regular line end, or (b) lines turning
around a corner of the inscribed object.
|| The double long vertical line is used to indicate a non-regular line division, e.g.
where the text is continued in the line above rather than in the line underneath it.
123
Superscript numbers have been used in sepulchral inscriptions consisting of
several tiles to indicate the arrangement of the text over the separate tiles. If a let-
ter is written partially on one and partially on the other tile, the number follows
the letter.
[1] [2] [3]
Superscript numbers between square brackets have been used in sepulchral
inscriptions consisting of several tiles to indicate the arrangement of the text over
tiles that have not been seen by any editor of the text, but that are assumed to
have originally been part of the inscription.
LII
Chapter 1
Introduction
“On peut se demander s’il y avait lieu d’écrire vraiment un ouvrage d’ensemble sur les
Falisques. Leur histoire militaire n’est guère qu’une petite partie de l’histoire romaine,
prise à rebours. Leur civilisation, assez rudimentaire, n’était qu’un reflet de celle de
leurs voisins du nord, les Étrusques, ou de leurs voisins du midi, les Romains. On la
connaît, assez mal, d’ailleurs, et, comme leur religion, surtout par les témoignages cent
fois cités des écrivains latins. Les inscriptions trouvées sur leur territoire sont ou
étrusques, ou à peu près latines, et ne nous fournissent pour ainsi dire aucun rensei-
gnement: la plupart, d’ailleurs, ne contiennent que quelques lettres à demi effacées ou
des nom propres souvent d’une lecture douteuse.”
1.1. Introductory remarks
With the biased but not wholly unjustified words quoted above, Duvau (1889:9) opened
his review of the first major work on Faliscan, Deecke’s Die Falisker (1888), and
similar pessimism still sounds through in the words of Deecke’s latest successor, G.
Giacomelli (1963:1), “il falisco non è una lingua attraente”. The problems in the study
of Faliscan, however, are no worse than those in the study of any fragmentarily
preserved language, and in the case of Faliscan there are at least three reasons why the
Faliscan material merits the effort of its study: “grande morae pretium ritus cognoscere,
quamuis difficilis cliuis huc uia praebet iter”, as Ovid (Am. 3.13.5-6) tells us.
1. Faliscan is one of the three best documented Latin dialects (the other two being
those of Rome and of Praeneste). The useable Faliscan material consists of c.355
inscriptions, and although in many cases these contain little more than names, they
provide a surprisingly large amount of linguistic data.
2. The Faliscan material is relatively old. Most inscriptions date back to the first
half of the third or the second half of the fourth century, a period for which there
are few documents for most of the other Italic languages, while even for the earliest
period the number of Faliscan inscriptions is comparatively large.
3. The interest of Faliscan is enhanced by the location of the area where it was
spoken. Lying between the areas where Etruscan, Sabellic languages, and Latin
were the native languages, and surviving the domination of the Etruscan culture, as
well as, for a long time, the expansion of Rome, it is of considerable interest for the
study of language contact in ancient Italy.
1
CHAPTER 1
For these reasons many publications have been devoted to the Faliscan material,
including such ‘ouvrages d’ensemble’ as Deecke’s Die Falisker (1888), Herbig’s
Corpus inscriptionum Etruscarum II.2.1 (1912), Buonamici’s Il dialetto falisco (1913),
Stolte’s Der faliskische Dialekt (1926), and G. Giacomelli’s La lingua falisca (1963). It
is therefore perhaps rather the desirability of a new comprehensive study on Faliscan
that could be questioned, especially as La lingua falisca did not receive an unfavourable
press.1 One obvious reason for a new study is that the results of the many linguistic,
epigraphic, and archaeological publications on Faliscan that have appeared in the 45
years since the publication of La lingua falisca have not yet been incorporated into a
major overview, although La civiltà dei Falisci (1990) provides an excellent overview
of the new developments, while the more important linguistic points have been the
subject of the monographs by R. Giacomelli, Ricerche falische (1978) and Nuove
ricerche falische (2006). The same applies to developments in the study of the Italic
languages that bear on the study of Faliscan, most important of which are surely the
publication, in 1978, of the lapis Satricanus (CIL I2.2832a) and the ensuing discussion
about the Italic second-declension genitive singular, the work by Wachter (1987) on the
early Latin inscriptions, and the discussion of the Italic perfects of facio following the
publication of two Faliscan inscriptions by Olmos Romera (2003) and Wallace (2004).
There is, however, another, and, in my view, more important reason for a new
comprehensive study on Faliscan. For a long time Faliscan has been regarded as in
some way related to Latin, and during the last decades more and more authors have
come to see Faliscan as essentially a Latin dialect. An overall critical examination of the
material that starts from this point of view is –or, rather, was– lacking until now. The
aim of this study is therefore to argue that Faliscan is a Latin dialect, and in no respect
a distinct language. In order to do so, the definition of both the term language and the
term dialect must be made clear, and every linguistic feature in the Faliscan epigraphic
material addressed, compared and evaluated in this context.
The remainder of this chapter is reserved for a number of mainly methodological pre-
liminary observations. The rest of this study is divided into two parts.
Part I is a discussion of the linguistic data in their widest sense. It opens with a
discussion of the historical sources and their sociolinguistic implications (Chapter 2).
The next chapters deal with the data on phonology (Chapter 3), the inflexional mor-
phology of the nouns and the pronouns (Chapter 4), the verbs (Chapter 5), the lexicon
(Chapter 6), the onomasticon (Chapter 7), and the syntax (Chapter 8). This is followed
by an assessment of the effects of language contact (Chapter 9). The linguistic part is
rounded off with my conclusions on the linguistic position of Faliscan with regard to
Etruscan, Sabellic, and Latin (Chapter 10).
1
Reviews: Lejeune 1964, Pellegrini 1964, Rix 1964 (critical), Untermann 1964 (critical), Loicq
1965, Olzscha 1965, and Mariner 1972.
2
INTRODUCTION
Part II is a presentation of the epigraphic data on which the discussions in part I

are based. It opens with a discussion of the material and of the Faliscan alphabet and
orthography (Chapter 11). The remaining chapters comprise the edition of the epi-
graphic material, divided into the earliest inscriptions (Chapter 12), the inscriptions
from Civita Castellana (Chapters 13-14), S. Maria di Falleri (Chapter 15), Corchiano
and the northern ager Faliscus (Chapter 16), the south-eastern ager Faliscus and the ager
Capenas (Chapter 17), the inscriptions of unknown or non-Faliscan or non-Capenate
origin (Chapter 18), and the Etruscan inscriptions (Chapter 19).
1.2. Dialect and language
As the main problem of this study is whether Faliscan is a dialect or an independent

language, some discussion of the way I define ‘dialect’ is mandatory. It should be noted
at the outset that I use the term ‘dialect’ here only for a geographical variant of a
language, and not, as is sometimes done, for social or in-group variants (sociolects or
idiolects) as well. The material under discussion here simply does not allow isolating
such variants, however welcome and interesting this would be.
The question of the definition of dialect can, on the whole, be approached from a
sociolinguistic and from a structural point of view. The exclusively sociolinguistic
definition of dialect is perhaps best summarized by the wisecrack about a language
being ‘a dialect with an army’: in other words, it is the politics behind the language that
decide, language being something used by a group that has some measure of political
independence, and dialect, by a group that is an identifiable subgroup of a larger
community but has no independent status. In this view the notions of dialect and
language depend therefore mainly on extralinguistic features, especially the attitudes
towards their dialect or language taken by the speakers of the dialect or language in
question and by those habitually in contact with them. In such an approach the criteria
on which a variant is regarded as a dialect or a language are provided not so much by
those dialects or languages themselves as by sources that shed light on how speakers
perceive the relation of their dialect or language to their cultural, national, tribal, or
ethnic identity. In the case of long dead languages or dialects like Faliscan, this
approach can only have a secondary or explicatory role, and then only if there are
abundant historical or archaeological sources.
For the Faliscan situation, the criteria have to be provided by the structural
definition, where the distinction between dialect and language is made on the basis of
the intralinguistic features of the language or dialect in question. This method, in its
turn, can be subdivided into a synchronic and a diachronic approach. In the synchronic
approach, variants are compared along a number of parameters, traditionally resulting in
a map of isoglosses, and are evaluated accordingly. The parameters should preferably
3
CHAPTER 1
refer to several different layers of the language structure, such as phonology, morphol-
ogy, lexicon, and syntax, and may accordingly be given a different ‘weight’ in deciding
the degree of difference between one or more variants. This method, too, is virtually
impossible to use in the case of such fragmentarily attested languages and dialects as
Faliscan, especially as the documents that are preserved can often not be dated with any
degree of accuracy (cf. §1.4.3). Most Latin dialect studies therefore apply a mixture of
the synchronic and the diachronic approach.2 The diachronic approach starts out from
the assumption, explicitly stated or not, that dialects start off as local variants that
gradually become independent (in a linguistic sense), either by retention or by innova-
tion. The degree to which a variant has or has not reached linguistic independence (i.e.,
to which a dialect differs from the standard or from other dialects of the same language)
can then be measured by the relative frequency of the following four parameters:
1. common retention: The languages or dialects involved have both retained the same
inherited feature.
2. separate retention: The languages or dialects involved have both retained different
inherited features.
3. common innovation: The languages or dialects involved have both replaced the
same inherited feature by the same new feature.
4a. separate innovation (bilateral): The languages or dialects involved have both
replaced the same inherited feature, but each by a different new feature.
4b. separate innovation (unilateral): One language or dialect has replaced an inherited
feature by a new feature, while the other has retained the inherited feature.
(The term ‘innovation’ does not necessarily imply an entirely new feature, but can also
refer to the standardization of one of several co-existing variants.)
Usually, these four parameters can be distinguished without too much trouble, espe-
cially when the variants involved have few contacts: for it is typical of this approach to
look at the variants involved as if they developed in complete isolation from each other,
while disregarding the phenomena related to language contact. It therefore runs the risk
of regarding a feature as e.g. an instance of common innovation, while it may in fact
have been a unilateral innovation followed by spread of the innovation from the
innovating to the non-innovating variant. This weakness is of course compensated by
the fact that it is far more expedient, and, from a structural point of view, theoretically
more justifiable, to study the development of each variant involved independently. The
problems of this method lie therefore not so much in its theoretical basis as in its
application.
2
E.g. Ernout 1905, Campanile 1968, Solta 1974, Wachter 1987:101-277, Coleman 1990, and
Joseph & Wallace 1991. Coleman 1986 is an example of a largely synchronic approach applied
to the Central Italic languages.
4
INTRODUCTION
With regard to the application, there is the problem that not every innovation or
retention is comparable to every other, and that not every innovation or retention is as
significant as every other. For instance, the fact that in both Faliscan and Latin /o / was
monophthongized to /ō ̣/ cannot be treated on a par with the fact that in both Faliscan
and Latin the second-declension genitive singular ending /-oso/ was replaced by /-ī/:
the former is a fairly frequently occurring phonological development that can also be
observed in other Italic languages, while the latter is an apparently unique morphologi-
cal replacement that is of far greater significance in the evaluation of the degree of
difference between Faliscan, Latin, and the Sabellic languages. In the application of this
method, there is a general tendency to regard morphological developments categorically
as more significant than phonological ones (syntax and other parts of language are often
left untreated), and older developments as more significant than recent ones. No doubt
this is partly justifiable: yet it should be stressed that every individual instance of
retention or innovation is to be evaluated for its relevance, and that the relevance given
to any specific instance of retention or innovation will therefore remain a matter of
interpretation.
With regard to the evaluation of the data in this method, the main problem is that
the results of the diachronic method can rarely be quantified, and, as I shall argue
below, the differences between languages and dialects are often gradual rather than
abrupt, and can therefore often be evaluated better by quantitative data than by distinc-
tions of the ‘either … or’ type. Furthermore, it is impossible to say that e.g. a certain
number of shared innovations followed by a certain number of independent innovations
points to a language rather than a dialect: one or two important separate innovations
may outweigh a large number of relatively insignificant common innovations, and vice
versa. The conclusions that can be reached by this method must therefore necessarily be
relative rather than absolute.
With these precautions in mind, the implications for the method followed here are
the following. First, as the outcome of the diachronic method is relative rather than
absolute, the number of languages or dialects compared should preferably be more than
two. To establish the position of Faliscan with regard to (Roman) Latin, it is therefore
not enough to compare Faliscan with (Roman) Latin, but Faliscan should be compared
also with other Latin dialects and with the Sabellic languages.3 Then, if Faliscan is
indeed a Latin dialect, the result of such a comparison would be expected to be, first,
that Faliscan sides with Latin in all cases where Latin has separate retentions or
innovations with regard to Sabellic; second, that Faliscan has a relatively large number
of important common innovations with at least several Latin dialects, and third, that
specifically Faliscan separate innovations would be either recent or of lesser signifi-
cance.
3
Failure to do so gives Joseph & Wallace 1991 a rather lop-sided conclusion.
5
CHAPTER 1
A final point worth noting is that during the period under discussion in this study, that
is, roughly the sixth to the second century, the dialects of the various Latin towns shifted
from being independently coexisting variants of one and the same language (more or
less in the manner of the Greek dialects, although with fewer structural differences)
towards being variants existing alongside a Roman Latin that was gradually becoming
the standard and consolidating its position as such. By the end of the period, if there
were still any regional variants, these differed only in the way or in the degree in which
they diverged from this standard, and can be regarded as local variants of substandard
or ‘rustic’ Latin. This process appears to have been completed by the beginning of the
first century BCE, exactly at the moment when the city-states and peoples of Italy lost
their varying degrees of independence in the Social War and its aftermath: the Social
War was the conclusion, not a catalyst, of this process of ‘substandardization’.
1.3. Languages in contact
1.3.1. Language contact in ancient Italy. In a context like that of ancient Italy,
especially during the period of the Roman expansion, it is reasonable to assume even a
priori the existence of extensive language contact and of bilingual individuals or even
bilingual communities. It is difficult to draw any specific conclusions about the effects
of these contacts, however, as they can be observed almost exclusively through
phenomena occurring in the written material. Sociolinguistic observations can certainly
be derived from archaeological or historical data, and can be used very effectively to
explain phenomena in written material, but such observations are not nearly specific
enough to predict with any degree of accuracy the effects of language contacts in so far
as these do not occur within the written material. Unfortunately, language contact, and
especially its corollaries interference and borrowing, have been and sometimes still are
used more as a device to explain away apparent irregularities in the material than as an
independent fact of language. This is especially so in the case of Faliscan, which many
authors have regarded as heavily ‘influenced’ by the Sabellic languages or Etruscan,
usually without making clear just what was meant by ‘influenced’. At best, they tacitly
placed all interference and borrowing phenomena on one level and equated a high
number of features that could be ascribed to other languages with extensive ‘influenc-
ing’. It is therefore useful to look at some aspects of language contact that are relevant
to the study of Faliscan.
1.3.2. Interference and borrowing. In this section I very briefly touch upon several
features regarding interference and borrowing in language contact: a more detailed
discussion, looking especially at the epigraphic material, may be found in §9.1. First of
all it should be clear that the phenomena that can be resumed under the header of
6
INTRODUCTION
‘linguistic interference’ operate on two different levels. The first is that of the individual
speaker who, using a second language he or she is not completely proficient in, imports
features from one language to the other, usually from his or her first language to his or
her second language. As the proficiency in the second language increases, these imports
gradually become fewer. On another level stand the interference features that have
become part of the language into which they have been imported and are consequently
used by a large part or the whole of the language community, even those that speak this
language as a first language. It is only to this second type of interference that the term
‘borrowing’ is applied. Unfortunately, in the case of languages that are only epigraphi-
cally attested, the difference between interference and borrowing is obscured by the fact
that an epigraphic document represents a language or dialect as used by one speaker at
one specific point in time. It is therefore impossible to establish whether a non-native
feature is due to real borrowing unless it returns in a sufficiently large number of
inscriptions. A second point that should be kept in mind is that, although interference
can take many forms, even in extensive language contacts not just any form of interfer-
ence can acquire the status of a borrowing, as borrowings are embedded in the language
into which they are imported and are therefore constrained by the structure of the host
language. Below, I name a few of the factors that can act as limitations.
(1) Structural difference. The first point is the degree of difference in structure between
the languages or dialects involved: where there is a large degree of structural differ-
ence, there will be less and more limited borrowing than where it is smaller. If the
structure of the language from which the borrowing is taken is similar to that of the host
language on the level where the borrowing takes place, the borrowed feature can be
embedded more easily into the host language, as it can be fitted into a similar place
within the structure. In the case of Faliscan, this is of special interest for the evaluation
of the contacts with the structurally very different Etruscan as compared with the
structurally not dissimilar Umbrian or Sabine. The question is of course of even greater
relevance for the contacts between Faliscan and Latin, for if Faliscan is regarded as a
Latin dialect, the structural differences between the two are assumed to be very small.
The degree of structural difference seems to be of little account as a borrowing con-
straint, however, in a so-called Sprachbund. This concept was originally developed for
the situation in the Balkans, where, in the aftermath of early mediaeval migrations and
the subsequent absorption of the peninsula into the Ottoman Empire, peoples and
minorities in every possible sense coexist in a variety of ways, and languages of various
Indo-European families exist side by side with altogether unrelated languages. In this
situation, languages have been shown to develop along parallel lines even if they are
completely unrelated. The existence of an Italic Sprachbund has been posited by Pisani
(e.g. 1978:39-55, with literature), but I hesitate to adopt this idea. Existing Sprachbund-
situations and the contexts in which they emerged differ greatly from that of pre-Roman
Italy, and I am not convinced that there is any feature in Italic linguistics that can be
7
CHAPTER 1
explained only or better by the assumption of an Italic Sprachbund. As far as I can see,
the Italic Sprachbund has until now usually been invoked to explain unwanted irregu-
larities in the material or in an argumentation (cf. §1.3.1).
(2) The borrowing hierarchy. Even where the structures of the languages in question
are similar, not everything is possible: some parts of language can more easily be
borrowed than others, depending on how deeply they are embedded in the language
structure itself. This concept is often expressed in the form of a ‘borrowing hierarchy’.
Although such hierarchies may differ in details, depending on the languages involved,
they tend to show the same general tendencies. As an example I give the original
borrowing hierarchy as it was long ago proposed by Whitney (1881:19-20), where
elements of language are placed in decreasing order of ‘borrowability’:
nouns > adjectives > verbs > adverbs > prepositions > conjunctions > pronouns >
derivational prefixes or suffixes > inflexional prefixes or suffixes > features of
grammatical distinction
Later publications have not brought significant changes to this model (see e.g. Haugen
1950:210-32, Appel & Muysken 1987:170-2). I regard the borrowing hierarchy as such
a strict constraint on borrowing that in the case of fragmentarily preserved languages or
dialects I have ventured to use it predictively, in the sense that borrowing on the deeper
levels of the hierarchy is indicative of borrowing on higher levels even if this is not
attested in the material: in other words, the presence of e.g. borrowed derivational
suffixes would be a strong indication that borrowing on every higher level of the
borrowing hierarchy took place even if for one ore more of these levels this is not
observable in the material.
As so much of the Faliscan material is onomastic, it should be stressed that
onomastic elements are extremely liable to be borrowed, but have no place on the
borrowing hierarchy as they are not a genuine part of the language structure (§7.1.1). A
large amount of borrowing in the onomasticon is therefore linguistically not significant
except as a very good indicator of language contact and of the way the users of the
languages involved perceive the identity of the communities involved.
(3) Extralinguistic factors. A third important point is that borrowing is not something
that happens by itself: borrowing is the result of a process that is brought about by the
speakers of a language, who select one of various modes of expression offered by
different languages or dialects at their disposal. Of course, this process is in most cases
subconscious, but the fact remains that a choice is made by the speaker, as the result of
a psychological or psycholinguistic process. The factors that decide this choice are often
not intralinguistic, but extralinguistic, the choice normally being made in favour of the
variant that belongs to the language or dialect associated with the community of which
it is attractive to be regarded as a member, e.g. for economical, social, political, cultural,
or perhaps even personal reasons.
8
INTRODUCTION
Borrowing is therefore usually not a two-way process, and especially in the

Faliscan context, where the contacts were primarily with the languages associated with
two dominant political or cultural influences in the area, namely Etruscan (until the fifth
century) and Roman Latin (from the third century onwards), borrowing should be
expected to take the form of import into Faliscan rather than of export from Faliscan.
Note that this implies that even within a strictly structural approach to the definition of
language and dialect, it is therefore still necessary to take the extralinguistic factors into
account when it comes to an evaluation of borrowing, as borrowing is a primarily social
process that manifests itself in a language contact situation.
There is one more point I wish to mention in connection with language contacts in
ancient Italy, namely the possible existence of diglossia, much of which has been made
by R. Giacomelli (e.g. 1978, 1979). Consulting the literature on the subject, from the
seminal article by Ferguson (1959) onward, exposes an inherent contradiction of
studying diglossia from data such as those on Faliscan, for it is a standard feature of
classical diglossia that the low-level language not only is not written, but in the opinion
of its users should not or even cannot be written. It should therefore be impossible to
study diglossia from written documents alone, unless there are strong indications for the
existence of an unwritten low-level language from other sources, which in the case of
Faliscan are not so compelling as to lead to the assumption of diglossia. Furthermore,
diglossia normally requires that the high-level language has a long written tradition
incorporating such fields as law, religion, and literature, as in the diglossia of Classical
and spoken Arabic, or of Greek kaqareÚousa and dhmotik». Such a situation is unlikely
to have existed in pre-Roman Italy.
1.4. The Faliscan material
1.4.1. ‘Faliscan’, ‘Latin’, ‘Sabellic’. In this study the term ‘Faliscan’ is used for a
geolinguistic unit, that is, it denotes the name of a regional variant of Latin spoken in
the area known as the ager Faliscus, whose extent is described in §2.1.1. There are no
Faliscan inscriptions from beyond this areas, although some have been claimed as such
(cf. chapter 18). One consequence of the assumption made in §1.1, that Faliscan is a
Latin dialect, is that the term ‘Latin’ should properly be used here as a hyperonym of
such local variants as Faliscan, Praenestine, etc. Tradition dictates, however, that ‘Latin’
is used exclusively to refer to the Latin of Latium and the colonies and other emigration
areas beyond its confines. I have therefore not without some reluctance maintained this
traditional use of the term ‘Latin’. The term ‘Sabellic’ is used here in the modern wider
sense to refer to Oscan, Umbrian, the Central Italic languages, South Picene, and
Praesamnitic, in short all languages that appear in Rix’s Sabellische Texte (2003), South
Picene probably being a sort of archaic Central Italic koin» (cf. Marinetti 1981,
9
CHAPTER 1
1985:43-4, Prosdocimi 1987, and Meiser 1987a-b). It should be stressed that terms like
‘Latin’, ‘Faliscan’, ‘Etruscan’, etc., are used as glottonyms and therefore refer to
linguistic units, irrespective of whether these coincide or overlap with homonymous
political or cultural units.
‘Proto-Italic’ is defined as the (chronological) stage during which those develop-
ments took place that together set off the Italic languages from the other IE language
families, but that precede any developments that occurred only in the Latin or only in
the Italic branch (see also §3.1.2). The terms ‘Proto-Latin’ and ‘Proto-Sabellic’ are
similarly used to denote the successive stage during which the Latin (or, Latin-Faliscan)
and the Sabellic branches of the Italic languages underwent their respective individual
developments that preceded any developments occurring only in the individual
languages of these branches (see also §3.3.1).
1.4.2. The material. The material for the study of Faliscan consists of inscriptions and
glosses. Virtually all our knowledge of Faliscan is based on the data provided by the
inscriptions: the handful of glosses (discussed in §6.6) add little to this. The number of
inscriptions given by the various authors varies, depending on what is called Faliscan
and what not, as well as on whether all or only the linguistically useful inscriptions are
counted. The lowest count is 100 (Beekes 1990:49, counting only those that he regards
as being of linguistic interest); the highest, 600 (Herbig CIE (1912), publishing exactly
600 texts, including every scrap or trace known to him). If from the 535 inscriptions in
my edition I exclude all that are illegible, consist only of abbreviations, or are clearly
Etruscan, Latin, or a Sabellic language, and count multiple inscriptions on one object
and inscriptions repeated on more than one object separately, I come to c.355 Faliscan
inscriptions, most of which contain features that can in some way or other be used as
linguistic data.
Most of the Faliscan inscriptions are sepulchral, scratched or painted on the tiles
used to close the burial-niches in the rock-cut tombs, or on the wall beside these. They
contain the names of the deceased, often followed by a filiation, and, in the case of
married women, by the word uxor ‘wife’ and the name of the husband in the genitive:
sometimes the words hec cupat/cupant ‘lie(s) here’ are added as well. Others are cut at
the entrance to the tomb, and name its owner, sometimes in the genitive with the word
cela added, ‘the tomb of ...’: a few also contain statements on burial rights. Most of the
remaining inscriptions are on pottery, and are usually either Besitzerinschriften or
potters’ signatures, consisting of names in the nominative or the genitive, though several
of the earliest inscriptions have considerably more content than that. Dedications are
few, and so are official inscriptions: the latter group consists mainly of the names of
magistrates cut in the sides of the hollow roads of the area, although there are a few
bronze inscriptions from the later periods. For an extensive overview of the material,
see §11.1; the formulas used in the various types of texts are discussed in §8.8-12.
10
INTRODUCTION
1.4.3. The dating of the material. The Faliscan inscriptions span a period from the late
seventh or early sixth century to the middle of the second century BCE. The earliest
material can easily be identified both archaeologically and epigraphically, but the main
group, dated loosely between the second half of the fourth and the early second century,
often presents serious dating problems. Dating on archaeological context is difficult, not
only because there may be doubts about the reliability of the archaeological context
provided (cf. below), but also because burial chambers were often reused for centuries,
and, especially in those tombs that were ransacked in antiquity, goods from one burial
may well have become associated with another. As a consequence, a reliable basis for
orthographical dating is lacking, although the use of the Latin alphabet in an inscription
that was clearly written in the ager Faliscus (such as a sepulchral inscription) may be
indicative of the period after 240 BCE (see below). Dating on linguistic features, that is,
on successive stages of diachronic developments, can of course not be used in a
linguistic study because of its obvious methodological disadvantages.
Editors therefore often take the war of 241 BCE (§2.6) as a dating criterion, for it
was a result of this war that the two sites that provide the majority of the inscriptions,
Civita Castellana and Corchiano, were abandoned, while the new Roman Falerii near
S. Maria di Falleri was founded soon afterwards. The inscriptions from Civita Castel-
lana and Corchiano are therefore assumed to date from before 240 BCE and those from
S. Maria di Falerii from after 240 BCE. This division is still applicable, although in the
case of Civita Castellana there are indications that the tombs and temples continued to
be used after 240 BCE (cf. §2.6.2). I have therefore divided the epigraphic material
from the agri Faliscus and Capenas into several age/alphabet groups, further specifying
the categories proposed by G. Giacomelli (1978:510-1). I briefly introduce these
categories here, discussing them further in §11.1.3:
1. Early Faliscan (EF) are the inscriptions between the late seventh and the fifth
centuries. These constitute a group that differs considerably from the rest both in
terms of the contents of the texts and linguistically.
2. Middle Faliscan (MF) are the inscriptions between the late fifth century and the
war of 241 BCE. This group includes virtually all inscriptions from Civita Castel-
lana and Corchiano, as these sites were apparently abandoned soon after 240 BCE,
and few inscriptions from these sites can be shown to be later.
3. Middle or Late Faliscan (MLF) are the inscriptions from the smaller sites in the
ager Faliscus that continued to exist after the war of 241 BCE, and cannot with
certainty be ascribed to either the Middle or the Late Faliscan period.
4. Late Faliscan (LF) are the inscriptions datable after the war of 241 BCE written in
the Faliscan alphabet and showing linguistic features that are consistent with the
Middle Faliscan inscriptions. The inscriptions from S. Maria di Falleri and from the
smaller settlements in the northern ager Faliscus belong to this group.
11
CHAPTER 1
5. Latino-Faliscan (LtF) are the inscriptions, mostly dated after the war of 241 BCE,
that are written in the Latin alphabet but still contain Faliscan dialect features. They
are mostly from the sites at Fabbrica di Roma and Grotta Porciosa.
6. Capenate (Cap) are the inscriptions from the ager Capenas that are written in the
Latin alphabet but still contain Faliscan dialect features. The majority of these in-
scriptions appears to date from before the end of the third century BCE.
Beside these categories, there are the Latin (Lat) inscriptions, that is, the inscriptions
written in Latin alphabet and showing no significant Faliscan features. These are far
more frequent in the ager Capenas (especially at the shrine of Lucus Feroniae), as this
was Latinized at an early date: in the ager Faliscus, they appear to date almost exclu-
sively from the period after 240 BCE. Finally, there are the Etruscan (Etr) inscriptions,
which are defined on the basis of linguistic features as well as of the alphabet.
The criteria on which a text is judged to be Faliscan, Latin, or Etruscan, are
discussed below: where doubt exists and alphabet and language involved seem to be at
odds, inscriptions have been labelled as e.g. Middle Faliscan/Etruscan (MF/Etr).
1.4.4. When is a text Faliscan? Although it is usually not difficult to tell whether an
inscription is Faliscan or not, there are cases where there is doubt as to the language
used. Editors often solve this problem by looking at the alphabet, the presupposition
being that the use of the Faliscan, Latin, or Etruscan alphabet is indicative of the
language used. Although this is to some extent true, it is not a conclusive, and, worse,
not a linguistic argument: I might choose to write this sentence in the Greek alphabet,
but unless I not only transliterate but translate it as well, the sentence remains English.
I have therefore looked first at the lexical, morphological, and phonological data present
in the text (in that order), and have used the alphabet only as an additional criterion.
The decision whether an inscription is Faliscan or Etruscan is usually not a very
difficult one, although even in this case things may turn out to be more subtle than they
appear at first sight. A case in point is umrie (‘Umbrius’) Etr XLIII, a sepulchral
inscription from Rignano Flaminio on the border between the ager Faliscus and the ager
Capenas, and, as a sepulchral inscription, not likely to have been written anywhere else.
I regard the language of this inscription as Etruscan in spite of the fact that it is written
in the Faliscan alphabet, for the cluster mr is Etruscan, and not in accordance with
Faliscan phonology, as is shown by upreciano (‘Umbricianus’) in MLF 363 and 364,
from the same tomb as umrie: the ending -ie(s), although undoubtedly Etruscan, cannot
constitute an argument to call the text Etruscan, as it occurs also as an interferential
form (limited to the onomasticon) in otherwise Faliscan texts. On the other hand, I
regard hermana MF/Etr 264 as Faliscan in spite of the fact that both the name and the
alphabet are Etruscan, because in Faliscan the Etruscan gentilicia in -na were incorpo-
rated into the first declension as borrowings, as they were in Latin: in other words, the
inscription could be Faliscan as well as Etruscan.
12
INTRODUCTION
The differences between Faliscan and Latin are less obvious. An illustrative
instance is med loucilios feced Lat 268, engraved on a fourth-century strigilis found
at Corchiano. I regard the language of this text as Latin, but not primarily because it is
written in the Latin alphabet. In this case the first criterion is morphological, for Latin
feced contrasts sharply with the contemporary Faliscan facet MF 470* and faced MF
471* (since med occurred both in Latin and in Early and Middle Faliscan, it cannot
count as a criterion), and the second, phonological, namely the spelling ou, which at this
date is Latin rather than Faliscan, where the spelling o was already predominant. But
how about the -s in loucilios? After all, omission of word-final -s is generally regarded
as a stock feature of Faliscan. The weakening of /s#/ after a short vowel and its
omission in writing occurred not only in Faliscan, however, but in so many other Latin
dialects as well that at the time it was a feature of Latin and Faliscan in general rather
than of any specific dialect: a point made even more salient by the fact that in early
Latin quantitative poetry /s#/ was treated in a different way than the Greek models for
such poetry prescribed. If the material from the various Latin dialects is evaluated
statistically, however, it appears that in the Faliscan inscriptions word-final -s is omitted
in about 97% of the instances, whereas the percentage of omission in other areas is
noticeably lower (§3.5.7d). It is therefore the frequency of the omission, not the
omission itself, that constitutes a Faliscan dialect feature, and omission of -s counts as a
Faliscan dialect feature only when viewed as part of the whole set of instances of this
omission from the ager Faliscus, and not when viewed in isolation.
As a consequence, when judging whether the language of a text is Faliscan or not,
such a statistically defined feature can count as a Faliscan dialect feature only if the text
is Faliscan in other ways as well (e.g. containing other Faliscan dialect features or
specifically Faliscan names, being written in the Faliscan alphabet, or being from the
ager Faliscus): an important point where texts from other areas are involved. Thus,
titoio 483† from Ardea, has been regarded as Faliscan because of the omission of -s, but
since there is no other indication of its being Faliscan, I can see no reason to regard it as
such (§18.3.2).
1.4.5. The reliability of the material. A last point to be made concerns the reliability of
the Faliscan material. As far as can be ascertained, the inscriptions themselves are
genuine: in spite of the suspicions of some early editors (notably Deecke, who never
saw any inscription himself), falsifications are apparently absent.4 There may be some
doubt, however, with regard to the data on the archaeological context of many of the
inscriptions.
4
Ueiụeto MLF 464 is regarded as a fake by many scholars; cauio | uetulio LF 335 may be an
unintentional double of cauia | uetulia LF 334. I myself must admit to having had initial doubts
with regard to the authenticity of the Ceres-inscription (EF 1), but I have not been able to
substantiate these.
13
CHAPTER 1
First, there is the possibility that chance finds or material from badly documented
private or semi-private excavations at other locations came to be ascribed to Civita
Castellana as the central town of the area, as happened to Manzielli’s excavations near
the Grotta Porciosa site in 1890 (see §16.6) and to finds from the Principe del Drago’s
excavations at Narce in 1891. Second, there is the ‘Villa Giulia scandal’ (see Sforzini
1985:538-9 and Waarsenburg 1994:39 nn.156-7, 159). This affair centred on allegations
made in 1898 by Helbig that the existing documentation on the excavations in the ager
Faliscus was to a great extent added later, which, if true, would seriously have discred-
ited the scientific value of the publications of this material, especially Barnabei’s
overviews in Monumenti Antichi 4 (1894). After a period of intense media attention and
an investigation by a parliamentary committee headed by Pigorini, the affair ended
more or less inconclusively with the temporary closure of the Museo, and with most of
the blame being put on the actual excavators, Benedetti and Mancinelli-Scotti. It is
useless to speculate now on who was to blame for what (or indeed whether anyone was
to blame for anything), but some reservation about the documentation of the finds from
the necropoles of Civita Castellana and Corchiano does not seem unjustified. Fortu-
nately, it would not appear that even in the worst case this would have significant
consequences for the linguistic assessment of the epigraphic material.
I conclude with a short note on the reliability of the epigraphic material as
presented in my edition (chapters 11-19). The scope of this edition was to provide all
the material on which my linguistic history of the agri Faliscus and Capenas is based,
and therefore to include all inscriptions from this area from before the first century
BCE, whether Faliscan, Latin, or Etruscan, to a total of 535 inscriptions. Of the c.440
Faliscan or possibly Faliscan inscriptions in this corpus, including those consisting only
of abbreviations, c.420 are presumably still extant, of which I publish 126 from
autopsy: this is indicated in the bibliography accompanying each inscription, and a full
list is given in §11.1.1. Since many of the inscriptions that I have not been able to see
were seen by G. Giacomelli, and vice versa, the two editions between them present the
majority of the inscriptions from autopsy. Where I have published e prioribus, I can
claim with some confidence to have collated every autoptic description and illustration.
1.5. A short survey of Faliscan studies: 150 years of scholarship
Although the ancient sources are not silent on the ager Faliscus and its inhabitants (cf.
§2.2-6), the data on Faliscan as a language or dialect are very few. They consist only of
a handful of glosses (discussed in §6.6), and of the much-quoted remark from Strabo’s
Geographica (5.2.9), œnioi d'oÙ TurrhnoÚj fasi toÝj Falšriouj, ¢ll¦ Fal…skouj, ‡dion
œqnoj: tinèj dè kaˆ toÝj Fal…skouj pÒlin „diÒglwsson: ‘Some say that the inhabitants of
Falerii are not Etruscans, but Faliscans, a distinct people; and some, too, that the
14
INTRODUCTION
Faliscans are a polis with a distinct tongue.’ It should be noted that as Strabo made this
remark as part of his description of Etruria, expressions like ‡dion œqnoj and pÒlij
„diÒglwssoj are to be taken relatively, as ‘different with regard to the rest of Etruria’,
rather than absolutely.
Although the study of Faliscan antiquities was revived at a very early date with
De origine et rebus Faliscorum (1546) by Massa, the study of Faliscan did not progress
until much later. Cluverius, in his Italia omnis (1624:537-8), established most of the
correct reading of the passage from Strabo, but did not discuss its implications, and
Dempsterus, in his De Etruria regali (1723:2.53), still dismissed Strabo as “in rebus
peregrinis balbutientem”. The first Faliscan inscriptions were noted down already in
1676 (LtF 205, MLF 206-207, MLF/Etr 208-209, and MLF 210), and the first one to
appear in print was published in 1726 (MF 79), but without attracting specific attention:
Lanzi, whose opinion on Faliscan (1824:52) was still no more than a paraphrase of
Strabo’s, in fact republished it as Etruscan (1824:392).5
The first conclusion on Faliscan that was based on linguistic data was drawn by
Mommsen in Die unteritalische Dialekte (1850:364), where he concluded from the few
Faliscan glosses given by the ancient authors that Faliscan was “wahrscheinlich
sabinisch oder umbrisch”, although at the same time drawing attention to the similarity
between the Faliscan glosses and those of the Latin dialects.
Real interest in Faliscan started with the publication, in 1854, of a group of Late
and Latino-Faliscan inscriptions from near S. Maria di Falleri (LF 220-230 and LtF
231-233). This led to Garrucci’s essays ‘Scoperte falische’ (1860) and ‘Epigrafi
etrusche anteaugustanee’ (1864) that mark the beginning of the modern study of
Faliscan. The inscriptions in Latin characters were subsequently included in Ritschl’s
Priscae latinitatis monumenta epigraphica (1862) and Mommsen & Henzen’s Corpus
inscriptionum Latinarum I (1863), while the Faliscan inscriptions, augmented by an
increasing number of new finds, found their way into Fabretti’s Corpus inscriptionum
Italicarum (1867), Garrucci’s Sylloge inscriptionum Latinarum (1877), Zvetaieff’s
Inscriptiones Italiae (1884-1885, 1886), Schneider’s Dialectorum italicarum aevi
vetustioris exempla selecta (1886), and Bormann’s Corpus inscriptionum Latinarum XI
(1888). This period was concluded by the first overall work on Faliscan and the
Faliscans, Deecke’s Die Falisker (1888). Deecke, gathering all data available in his day,
treated Faliscan as influenced by the Sabellic languages, and regarded the ager Faliscus
as at least linguistically entirely separate from Latium. In spite of the fact that his
edition is nowhere based on autopsy, this work provided the basis for the subsequent
publication of the Faliscan inscriptions in Conway’s The Italic dialects (1897).
5
Lanzi’s only Faliscan specimen (1824:459-64) was leipirior santirpior duir for | foveer
dertier dierir uotir | farer uef naratu uef poni | sirtir (= CIL XI, falsae 350*), with the
comment that “a Faleria paese indioglotto [sic] ottimamente conveniva un linguaggio nè latino
nè greco” (1824:461): Faliscum est, non legitur, as one might say.
15
CHAPTER 1
The 1880s and 1890s were fruitful decades for the study of Faliscan, mainly
because of the many excavations connected with the newly-founded Museo di Villa
Giulia. This material has since been gathered in Formae Italiae II.1-2 (1978, 1982).
Even when these were stopped and the Museo temporarily closed in the first decade of
the twentieth century (§1.4.5), a great number of new inscriptions were published,
including the famous Early Faliscan ‘Ceres-inscription’ (EF 1). The wealth of new
material, most of which was first published in Thulin’s seminal essay ‘Faliskische
Inschriften’ (1907), made it quite clear that Faliscan was much less influenced by the
Sabellic languages and stood much closer to Latin than had hitherto been assumed. On
the other hand, Faliscan now came to be regarded as having been extensively influenced
by Etruscan. This theory, already apparent in Jacobsohn’s Altitalische Inschriften
(1910), reached its zenith in Herbig’s Corpus inscriptionum Etruscarum II.2.1 (1912),
a reworking of his Habilitationsschrift, Tituli Faleriorum Veterum (1910). Epigraphi-
cally, this is still the most complete work on the Faliscan inscriptions, with drawings of
virtually all inscriptions then known, and virtually everywhere based on autopsies by
Herbig himself or by Nogara; linguistically, it has the drawback of not containing a
comprehensive evaluation of the linguistic data. Herbig’s work was followed closely by
Buonamici’s Il dialetto falisco (1913), the third work devoted entirely and exclusively
to Faliscan and the Faliscans.
The study of Faliscan then entered a phase in which few new inscriptions were
published, with the important exceptions of the inscriptions from Vignanello (MLF
302-323) in 1916, and, in 1933-1935, of three Early Faliscan inscriptions from Civita
Castellana (EF 2-4). Interest now began to focus on the linguistic interpretations,
resulting in the Inaugural-Dissertation by Herbig’s pupil Stolte, Der faliskische Dialekt
(1926), the first work to concentrate entirely on the linguistic data. Other linguistic
publications were made by Ribezzo (e.g. 1918, 1927, 1930, 1934, 1936), who main-
tained that Faliscan was heavily influenced by the Sabellic languages.
The importance of the Faliscan material now began to be realized outside the
strict field of Faliscan studies, especially in the discussion of the Italic o-stem genitive
singular. Vetter’s Handbuch der italischen Dialekte I (1953:277-331) contained the first
publication of the complete Faliscan corpus since Herbig’s CIE (1912). Although he did
not comment explicitly on the linguistic position of Faliscan, it is clear from his
comments on individual inscriptions that he regards it as at least close to Latin. The
1950s also saw much work in the archaeological field, notably the surveys of the ager
Faliscus and Capenas conducted by the British School, published by Ward Perkins &
Frederiksen (1957) and G. Jones (1962). A number of these new developments were
included in G. Giacomelli’s La lingua falisca (1963), the first publication since Deecke
1888 to include both an edition and a linguistic overview of the material. She published
many inscriptions from autopsy, often for the first time since Herbig’s CIE. As is clear
from the title, G. Giacomelli regarded Faliscan as a language closely related to Latin.
16
INTRODUCTION
After G. Giacomelli’s edition, Faliscan studies went through a revival. A number

of inscriptions were published in the second edition of Pisani’s Le lingue dell’Italia
antica oltre il latino (1964). A great number of publications appeared in the mid-1960s,
of which I name only the many studies by Peruzzi (1963b, 1964a-d, 1965, 1966a,
1967a-b) and Hirata’s monograph on the onomasticon (1967); more extensive studies
are the monographs by R. Giacomelli, Ricerche falische (1978) and Nuove ricerche
falische (2006). A large number of new inscriptions were published, most importantly
the Late Faliscan inscriptions from Pratoro (LF 242-249), published by Renzetti Marra
(1974, 1990) and two Middle Faliscan inscriptions (MF 470*-471*) published by
Olmos Romera (2003) and Wallace (2004). Important, too, was the publication of the
archaeological data gathered in the 1880s and 1890s in Forma Italiae II.1 (1972) and
II.2 (1981). Archaeological work progressed significantly: I name only the studies by
Potter (1976, 1979) and Moscati (1983, 1985a-b, 1987, 1990). A comprehensive
overview of the major developments was provided by the publication of La civiltà dei
Falisci (1990).
The points of view on the linguistic position of Faliscan since the appearance of La
lingua falisca are the following. The view that Faliscan was closely related to Latin was
adopted by most scholars, but the degree of difference between Faliscan and Latin
remains a matter of debate.
G. Giacomelli reiterated her view that Faliscan is a language closely related to
Latin (in her article ‘Il falisco’ in Lingue e dialetti dell’Italia antica (1978), pp.509-
542), but now attributed many features of Faliscan to influence from the Sabellic
languages. Campanile (Studi sulla posizione dialettale del latino (1969), pp.85-92)
pointed mainly to the differences between the two, and seems inclined to award
Faliscan a more or less independent position. Solta (Zur Stellung der lateinischen
Sprache (1974), pp.45-47) stressed the correspondences between Faliscan and Latin,
but also saw a convergence of Faliscan with the Sabellic languages. Wachter (Alt-
lateinische Inschriften (1987), pp.31-2 with n.73) stresses the shared innovations of
Latin and Faliscan, and regards the two as “sprachlich ... nahestehenden Völker-
schaften” (p.32). In the evaluation of the question by Joseph & Wallace (‘Is Faliscan a
local Latin patois?’, Diachronica 8 (1991), pp.159-186), the conclusion is again that
“Faliscan is a separate language from Latin and not a dialect of Latin, though it is the
closest sibling to Latin in the Italic family tree” (p.185): their view is shared by Baldi
(The Foundations of Latin (1999), pp.170-4). R. Giacomelli (first in Ricerche falische
(1978) and recently in Nuove ricerche falische (2006), the most recent contribution on
the subject) treats Faliscan on the whole as a Latin dialect, and I must agree with his
conclusion, although I disagree on many points with his arguments.
In §10.2 the views of these scholars are debated in detail on the basis of the evaluation
of the linguistic data presented in the following chapters.
17
CHAPTER 1
Map of the ager Faliscus

Map of the ager Faliscus, showing most of the principal sites, the rivers and streams dividing the area into
ridges, as well as the course of the three Roman roads (the Viae Cassia, Amerina, and Flaminia)
constructed during the late third and early second centuries BCE. Of the two sites not figured on the map,
Narce was located about 4,5 kilometres to the north-east of Mazzano, and Grotta Porciosa about 2
kilometres to the east of Gallese, close to the place where the Via Flaminia crossed the Tiber. Note the
Monti Cimini that closed off the area to the west and north, and the main route to the west, crossing the
Monti Cimini west of Sutri through the ‘Sutri Gap’. (From Ward Perkins & Frederiksen 1957:68 fig.1.)
18
Chapter 2
The ager Faliscus and its inhabitants
As a prolegomenon to the discussion of the linguistic data in chapters 3-9, I present in this
chapter a survey of the non-linguistic data relevant to the linguistic history of the ager Faliscus.
This consists of brief descriptions of the ager Faliscus (§2.1), its inhabitants and their culture,
regarded from the perspective of ethnic identity (§§2.2-3), and the history of the area (§§2.4-6).
This is followed by an evaluation of the sociolinguistic data relevant to the preservation of
Faliscan and its later disappearance (§2.7), and the general picture that can be drawn from the
data presented in this chapter (§2.8).
2.1. The ager Faliscus
2.1.1. Extent. The geographical and political unit known in antiquity as the ager
Faliscus (see map p.18) is located in what is today the regione Lazio, the southern part
belonging to the provincia of Roma and the northern to that of Viterbo; archaeologi-
cally, the area resorts under the Soprintendenza Archeologica dell’Etruria Meridionale.6
In antiquity, the ager Faliscus was bordered on the east by the Tiber, which seems
to have had no significant crossings between Lucus Feroniae at the southern end of the
ager Capenas and the site near Grotta Porciosa in the north-eastern corner of the ager
Faliscus (cf. Ward Perkins & Frederiksen 1957). The Tiber thus constituted a strong
geographical as well as political and linguistic boundary with the area on the east bank.
To the west, the borders of the ager Faliscus lay on the slopes of the Monti Sabatini in
the southwest and the Monti Cimini in the northwest, which in antiquity were densely
wooded. Livy’s description (9.36.1, cf. also 10.24.5) of this silua Cimina at the end of
the fourth century as “magis tum inuia atque horrenda quam nuper fuere Germanici
saltus, nulli ad eam diem ne mercatorum quidem7 adita”, although obviously meant to
add colour to his story, is confirmed by pollen analyses of samples from the beds of the
Lago di Bracciano, the Lago di Monterosi, and the Lago di Vico, which indicate that the
eastern slopes of the Monti Sabatini were covered by dense oak forests that remained
largely undisturbed until the third and second century (see Potter 1976:6, 1979:96).
6
Archaeological guides to the area are De Lucia Brolli 1991a-b and Torelli 1985:25-48. A
discussion of the area’s history based on the archaeological material is Potter 1979.
7
That is, foreign merchants: the intrepid Fabius who crossed the forest in 310 to avoid the
hostile ager Faliscus was protected not so much by his disguise and his fluent Etruscan as by the
fact “quod abhorrebat ab fide quemquam externum Ciminios saltus intraturum” (Liv. 9.36.6).
19
CHAPTER 2
The northern and southern borders of the ager Faliscus are harder to establish.
Northwards, the ager Faliscus extended to the ager Hortanus to the northeast and the
ager Vulcentanus to the northwest, the border probably running along the upper slopes
of the Monti Cimini. To the southwest, the ager Faliscus bordered on the ager Veienta-
nus. The most obvious natural boundary here is formed by the Monti Sabatini and the
ridge connecting these with Monte Soratte (ancient Soracte), and this may well have
been the original southern boundary of the ager Faliscus. During the fifth century,
however, the political influence of Veii extended northward well beyond this range to
include Sutrium and Nepete (modern Sutri and Nepi), and as these towns subsequently
became colonies in the early fourth century (§2.5.2), they and their territories ceased to
be part of the ager Faliscus at an early date. To the southeast, the ager Faliscus probably
included Monte Soratte, which was ascribed to the Faliscans e.g. by Pliny (NH 7.2.19)
and Porphyrio (in Hor. Carm. 1.9.1): the border with the ager Capenas must have run
somewhere along its southern and south-eastern slopes. It is unclear whether the land
between Monte Soratte and the Tiber belonged to the ager Capenas or the ager Faliscus:
although modern authors tend to assign it to the ager Capenas, the fourth- and third-
century inscriptions from the area to the east of Monte Soratte are virtually without
exception Faliscan instead of (Capenate) Latin (see §17.2-6).
2.1.2. Towns. The main site of the ager Faliscus, at least from the sixth century onward,
was the town called Falerii or Falšrioi by the ancient authors, identified since
Cluverius (1647:544-5) with the site occupied by Civita Castellana, c.50 kilometres
north of Rome.8 It was located at the point where a number of smaller streams flowing
down from the Monti Sabatini joined the Treia, the main river of the ager Faliscus, and
thus the natural centre of the area. The same name was used for the settlement founded
by the Romans after the war of 241 as the new centre of the area, at the place known
today as Faleri, or, in reference to the mediaeval abbey located within the Republican
town walls, S. Maria di Falleri, c.4.5 km to the west of Civita Castellana. Modern usage
therefore refers to the two towns as Falerii Veteres and Falerii Novi respectively.9
The ancient sources sometimes used other names for the two towns, a point
discussed by Di Stefano Manzella (1977). Falerii Veteres is also referred to as Fal…skoj
(Steph. Ethn. 656.24-5 Meineke), Fal…skon (Str. 5.2.9, Diod. 14.96.5; Faliscum Avit.
fr.2.2 apud Prisc. CGL 2.427.2 is a genitive plural = Faliscorum), or Falisca (Solin.
2.7), and Falerii Novi is also called Falšrion (Str. 5.2.9, Ptolem. 3.1.43 Cuntz, Steph.
Ethn. 656.12-3 Meineke: Dionysius (1.21.1-2) uses this name for both Falerii’s).
8
For the older tradition that Civita Castellana was the location of Veii, probably inspired by the
impressive site, see Moscati 1985b:45 n.1.
9
Short descriptions of these sites and references to the literature on their excavations are given
in the introductions to chapters 13-14 (for Falerii Veteres) and 15 (for Falerii Novi).
20
THE AGER FALISCUS AND ITS INHABITANTS
The appellations “[colonia] Falisca quae cognominatur Etruscorum” (Plin. NH

3.5.51) and “colonia Iunonia quae appellatur Faliscos [sic]” (Lib. Col. 217.5) would
seem to refer to Falerii Novi, but this view has been challenged: cf. §2.6.2. Falisci10
appears to be an alternative for Falerii: Ovid (Am. 3.13.1), too, used Falisci as a topo-
nym when speaking of pomiferis ... Faliscis.11 Strabo (5.2.9) and Plutarch (Cam. 2,
9-11, whence Polyaen. 8.7.1) use the two ethnica to distinguish between the inhabitants
of the town (the Falšrioi) and of the ager Faliscus as a whole (the Fal…skoi).
The variants sometimes refer to coexisting towns, providing an argument for a
continued existence of Falerii Veteres after the war of 241 (cf. §2.6.2). Strabo (5.2.9)
names both a Falšrion and a Fal…skon, which Di Stefano Manzella (1977:156) equated
with Falerii Novi and Falerii Veteres respectively: similarly, Stephanus mentions both a
Falšrion (Ethn. 656.12-3 Meineke) and a Fal…skon (Ethn. 656.24-5 Meineke), calling
the latter an ¥poikoj 'Arge…wn, which identifies it as Falerii Veteres (cf. §2.4.1).
Other sources also name an Aequum Faliscum (Str. 5.2.9, Tabula Peutingeriana)
or Aequi Falisci (Verg. A. 7.695, Sil. 8.489; in both these cases the name is used as an
ethnicon, cf. above on Falisci). This place was apparently not identical with Falerii
Novi, for the Tabula Peutingeriana shows the latter as faleros, located correctly at five
miles beyond Nepete on the Via Amerina, and the former as aequo falsico [sic],
unfortunately located erroneously to the east of the Tiber on a road to Spoletium that is
itself located to the west of the Tiber. Strabo’s description (5.2.9) of A„kououmfal…skon
as ™pˆ tÍ Flamin…v Ðdù ke…menon metaxÝ 'Okr…klwn kaˆ `Rèmhj can refer neither to Falerii
Veteres nor to Falerii Novi (nor to the site at Grotta Porciosa, as Ward Perkins &
Frederiksen (1957:189 n.53) suggested). If the town is identical with Falerii Veteres, it
is unclear how the steep site could be called aequus: perhaps aequus is to be understood
as ‘levelled’ (aequus or aequatus solo), referring to the destruction of the town in 241.12
The other town ascribed by the ancient sources to the ager Faliscus is Fescennium
or Faskšnnion: it is mentioned only as the place of origin of the ribald wedding songs
known as carmina Fescennina, and cannot be located with certainty. It has been
identified with Narce (first Barnabei 1894a:22), a major site of the southern ager
Faliscus, whose decline after the sixth century would account for the lack of references.
Habitation at Narce seems to have ceased after the war of 241, however, whereas
10
Faliscos is an accusative (perhaps from a map reference, cf. faleros v on the Tabula Peut-
ingeriana), not a transcription of Fal…skoj, as Di Stefano Manzella (1977:160) took it.
11
Di Stefano Manzella (1977:152-4) furthermore adduced the very ambiguous instances Eutrop.
1.20.1-2 and 2.28, Amm. 23.5.20, and Serv. in Verg. A. 7.695. (The use of Falisci as a toponym
may be due to the existence, in the onomasticon, of Falerius beside the toponym Falerii.)
12
Cf. the Aequimaelium in Rome, thought to mark the site of the razed house of Sp. Maelius
(Liv. 4.16, Dion. 12.4.6, Cic. Dom. 101, Var. L 5.157, V. Max. 6.3.1). Servius’ explanation (in
Verg. A. 7.695) of aequus as ‘just’ because the Faliscans were allegedly the source of the ius
fetiale, although obviously spurious, represents an attempt to address the same problem.
21
CHAPTER 2
Dionysius (1.21.1) speaks of Faskšnnion as an existing Roman town, and Pliny (NH
3.5.51) names Fescennia as a township of the Augustean regio VII. Others have
identified Fescennium with the site near Grotta Porciosa (Dennis 1848:159-62), with
Corchiano (Buglione 1887a:25-6), or with Rignano Flaminio (Taylor 1923:93-4). The
various claims are discussed by Shotter (1976:33-4) and Colonna (1990), who both
decide in favour of Narce (thus also BarrAtl (2007), map 142).
These are only the sites that are named by the ancient authors. The archaeologi-
cal record shows a number of other sites, e.g. at Corchiano and Vignanello in the
north-west, at Gallese and the Grotta Porciosa locality in the north-east, and around
Monte Soratte, with a distinct linguistic interest of their own, especially in the period
after 240, when Falerii Veteres disappeared as the centre of the area and the land itself
was divided into a Faliscan and a Roman part (see §2.6.2). References to the literature
on these sites are given in the sections where the inscriptions from these sites are
discussed in chapters 16 and 17: see also the works mentioned in note 6 (p.19).
2.1.3. Roads and routes. The road-system of the ager Faliscus and its routes to the
surrounding areas are well known thanks to the surveys conducted by the British School
in the 1950s (see Ward Perkins & Frederiksen 1957, G. Jones 1962, and Quilici 1990).
As said in §2.1.1, the ager Faliscus was closed off on the east and west sides by
strong natural boundaries, and major lines of communication in these directions were
few. No ancient Tiber-crossings are known between Lucus Feroniae at the southern end
of the ager Capenas and the site near Grotta Porciosa at the northern end of the ager
Faliscus used by the Via Amerina (cf. Ward Perkins & Frederiksen 1957 passim).
Apparently unlike the next crossing to the north, located c.10 km away near Orte, the
crossing near the Grotta Porciosa site was connected, by way of the valley of the Nar,
with the Sabine interior and Umbria. It therefore constituted an important connection
between South Etruria and the interior, and may have seen extensive traffic not only of
traders but also of transhumance farmers (cf. Skydsgaard 1974:23-8 and Potter
1979:37-41). Since from c.400 to 240 this crossing was the first one to the north of
those controlled directly by Rome, it may have had strategic importance as well.
The same applies to the westward route from the ager Faliscus to coastal South
Etruria through the ‘Sutri Gap’, the saddle between the mountains surrounding the Lago
di Bracciano and those surrounding the Lago di Vico. After the fall of Veii in the early
fourth century, this was the southernmost route through Etruria not under Roman
control, and the Romans were quick to secure it by establishing colonies at Sutrium and
Nepete some years later (§2.5.2). The strategic importance of Sutrium and Nepete as
gateways to Etruria is reflected by the frequent mention, in Livy’s description of the
wars of the fourth and third centuries, of skirmishes for the possession of these towns
(see §2.5.2), especially Sutrium, which commanded the road that in the middle of the
second century would become the Via Cassia.
22
To these routes to the east and west must be added the connection to the north and
south formed by the Tiber. Especially during the floruit of Veii in the sixth and fifth
centuries, when, together with Capena, Veii may have dominated its lower course all
the way down to the river Anio (the modern Aniene), a sizeable amount of traffic must
have passed up and down the Tiber valley (see Baglione 1986). This is reflected by the
antiquity of the Via Tiberina, which followed the west bank for practically the whole
length of the ager Capenas, having been traced even to the north of Fiano Romano
(G. Jones 1962:201). The ager Faliscus was therefore located on the crossroads of two
arteries between coastal South Etruria and the interior, which may explain its independ-
ence to some extent.
The overland routes through the ager Faliscus to the north and south were largely
dictated by the nature of the soil. The streams flowing down from the Monti Sabatini
and the Monti Cimini to the Tiber eroded the soft tuff of the ager Faliscus into deep,
narrow gorges with often nearly perpendicular sides. With heights of c.70 m around
Civita Castellana and c.100 m near Narce, these make a daunting obstacle to the
traveller on foot or on horseback, as can still be glimpsed from Dennis’s accounts
(1848:115-62) of his travels in the area. Such roads as there are tend to follow the ridges
between the gorges, descending only at points where both the gorge and the stream can
conveniently be crossed. In the northern ager Faliscus virtually every stream runs from
the mountains in the west to the Tiber in the east, constituting a severe impediment for
any route to the north. This problem was partly solved by the construction of impressive
bridges (cf. Ward Perkins & Frederiksen 1957:97-101, 144-9), and there were roads
from Falerii Veteres to Corchiano and Horta, and to the site at Grotta Porciosa; another
road may have linked the site of Falerii Novi to Corchiano (cf. Ward Perkins &
Frederiksen 1957:185 fig.29). In the southern half of the ager Faliscus, most of the
streams run northeast, flowing into the Treia near Civita Castellana: accordingly, here
the ridgeways tended to run southwest-northeast, converging at the site of Falerii
Veteres, which thus constituted the natural centre of the area.
The ager Faliscus was therefore best accessible from the south, and had one main
route each to the east, the west, and the north. This is reflected by the way in which the
road-system was restructured by the Romans in the century following their expansion
into the ager Faliscus after 240 (see Ward Perkins & Frederiksen 1957:187-93, Potter
1979:101-9). The Via Amerina, the first stretch of which, from Rome to Nepete, may
already have been constructed when Nepete became a colony in the early fourth century
(§2.5.2), was continued to Falerii Novi, which was thus firmly linked to Rome. From
there, it followed the line of the earlier Faliscan road through the northern ager Faliscus
to Horta, where it crossed the Tiber and continued to Ameria.13 A second road, the Via
13
According to Radke (1964:220-1), the Via Amerina thus took over the function of an earlier
road to Horta leading northeast from Forum Claudii through Sutrium. No trace of such a road
has been found: its presumed route would have taken it straight through the silua Cimina.
23
CHAPTER 2
Flaminia, was constructed in 220: this entered the ager Faliscus near Rignano Flaminio,
and continued almost directly north to the Tiber crossing near the Grotta Porciosa site,
having been joined several kilometers to the south by a road branching off from the Via
Amerina at Falerii Novi. Crossing the Tiber, it lead to Ocriculum, and thence through
the valley of the Nar to Umbria. The Via Flaminia thus bypassed both Capena and
Falerii Veteres, and appears to have been built for long-distance traffic to Umbria and
beyond, being of minor importance for the connections between the ager Faliscus and
Rome. The same is true of the third road in the area, the Via Cassia, constructed in 154,
which branched off from the Via Amerina north of the Baccano crater to take a north-
westerly course along the lines of existing roads to Sutrium and through the ‘Sutri Gap’
without entering the ager Faliscus proper. In the Roman road-system, the ager Faliscus
thus no longer stood on the crossroads of north-south and east-west routes, but consti-
tuted a kind of ‘junction station’ on several long- and middle-distance routes leading
north from Rome.
2.2. The inhabitants of the ager Faliscus as an ‡dion œqnoj
2.2.1. A distinct people. As described in the preceding sections, the ager Faliscus
constituted a fairly well-defined geopolitical unit that was perceived as belonging to
Etruria (thus Strab. 5.2.9, Plin. NH 3.5.51, Ptolem. 3.1.43 Cuntz, Serv. in Verg. A.
7.607, Steph. Ethn. 656.12-3 Meineke). The inhabitants of this area, called Falisci or
Fal…skoi, could therefore be classed as a populus Etruriae (Liv. 5.8.5) or a ciuitas
Tusciae (Serv. in Verg. A. 7.607). Yet they were regarded as in some respects different
from other Etrurian communities, as is expressed by Strabo’s remark (5.2.9), œnioi d'oÙ
TurrhnoÚj fasi toÝj Falšriouj, ¢ll¦ Fal…skouj, ‡dion œqnoj: tinèj dè kaˆ toÝj Fal…skouj
pÒlin „diÒglwsson: ‘Some say that the inhabitants of Falerii are not Etruscans, but
Faliscans, a distinct people; and some, too, that the Faliscans are a polis with a distinct
tongue.’ As noted (§1.5), the context of the remark shows that Strabo is speaking in a
relative sense, ‘different with regard to the other poleis of the area’ (cf. Camporeale
1991:213). Diodorus (14.96.5), too, spoke of Falerii as a town toà Fal…skwn œqnouj, and
Dionysius (1.21.1-2) ascribed features of Faliscan culture to a Pelasgian background
(cf. §2.4.1). Coupled to the linguistic differences with the surrounding areas that can be
observed from the epigraphic material, these statements have often led to the opinion
that the inhabitants of the ager Faliscus were in some sense ‘different’.
In what sense the inhabitants of the ager Faliscus differed from their neighbours is
a very difficult question to answer. This is due in part to the fact that labels like
‘Faliscan’, ‘Etruscan’, or ‘Latin’ are used by different authors for such different entities
as e.g. linguistic, political, or cultural units, which may overlap, but do not necessarily
coincide (§1.4.1), and are often very hard to separate: partly, too, the difficulties derive
24
from the vague genealogical terms in which cultural or linguistic relationships tended
and sometimes still tend to be expressed. Even Pulgram (1958:252-3) used the term
“blood brothers” to refer to the linguistic unity of the ager Faliscus with Rome; Alföldi
(1963:191) classed the Faliscans with the Capenates and Fidenates as “branches of the
same Latin stock”, while Ward Perkins (1970:427) called them “an independent branch
of the same Urnfield peoples as the Villanovans and the Latins”. Such statements are
based wholly on connections that are irrelevant to, say, an assessment of the political
relationship between the inhabitants of the ager Faliscus and the Romans in the third
century, unless these connections were perceived, and perceived as relevant, by the
Faliscans and the Romans themselves.
2.2.2. The Faliscans as an ethnos. In my view, for the purposes of this study the
identity of the Faliscans is best approached by regarding it as an ethnic identity, in the
sense in which this term has come to be used in social anthropology and thence in
sociolinguistics and archaeology. To quote a general but useful definition of this term: 14
“этнос ... может быть определен как исторически сложившаяся на определенной
территории устойчивая совокупность людей, обладающих общими относительно
стабильными особенностями языка и культуры, а также сознанием своего
единства и отличия от других подобных образований (самосознанием),
фиксированным в самоназвании (этнониме).” (Bromley & Kozlov 1975:11)
“Ethnos … can be defined as a firm aggregate of people, historically established on a
given territory, possessing in common relatively stable particularities of language and
culture, and also recognizing their unity and difference from other similar formations
(self-awareness) and expressing this in a self-appointed name (ethnonym).” (translation
from Dragadze 1980:162)
Such a definition is applicable to the inhabitants of the ager Faliscus in so far as they
were ‘historically established’ on the ager Faliscus without major invasions or migra-
tions in the historical period or the centuries closely preceding it (cf. §2.4.1), and as
having ‘relatively stable particularities of language’ and perhaps even ‘relatively stable
particularities of culture’ (see §2.3).
With regard to the ethnonym, the inhabitants of the ager Faliscus do not seem to
have had a real ‘national’ name in the order of Etrusci, Sabini, or Latini, i.e., a name
that referred to a people or an area, but were named Falisci or Fal…skoi after their main
city. However, this name is formed with the suffix that is distinctive of Italic ethnonyms
(e.g. Latin Etrusci, Osci, and Umbrian turskum, iapuzkum TI Ib.17). The only instance
of the use of this ethnonym by the Faliscans themselves is falesce quei in sardinia
14
Although this definition is now much less recent than when I started working on this book,
it is by no means outdated: recent authors on ethnic identity in ancient Italy (e.g. Bradley
1997, Cornell 1997, Dench 1997, 2007) tend to refer to the criteria used by Smith (1986,
1991), which are almost exactly the same, although nowhere expressed as concisely as in this
quotation.
25
CHAPTER 2
sunt, in a late second-century Latin inscription from Falerii Novi (Lat 218), where it is
used for Faliscans that no longer lived in the ager Faliscus: exactly as expected, since
there is no need to use the ethnicon within the group itself except where it is necessary
to contrast members of one’s own group with those of other groups. The same is true
for the possible occurrence of the Faliscan ethnonym in the Etruscan inscription [mi
a]ụṿileś feluskeś tuśnutaḷ[a pa]|panalaś Vn 1.1 (see §2.4.2).
It should also be pointed out that ethnicity may depend on different features at
different times, and that the recognition of certain features as forming part of ethnic
identity is strengthened by conflict or competition. The point that ethnicity is thus a
relative rather than an absolute concept is of some importance for the study of the ager
Faliscus, for it changes the search for features that were specifically and uniquely
‘Faliscan’ to a search for features in which the ager Faliscus could regard itself as
different from the surrounding areas. Thus, the Faliscans may have regarded their
language as distinctive in defining their ethnic identity with regard to the Etruscans,
while on the other hand they may have regarded their traditional political alliances as
distinctive with regard to the Latins.
Ethnicity then consists not in certain features of the culture of a group per se, but
in the way these features are perceived as relevant to the identity of the group by its
members or by those in contact with them. It could be argued that in applying this
concept to the ager Faliscus, the problem of establishing a ‘Faliscan identity’ is
confused rather than solved, for there is no way of knowing what constituted distinctive
‘Faliscan’ features in the eyes of the Faliscans themselves or of their neighbours. Yet I
think it is worthwhile to describe the inhabitants of the ager Faliscus in the terms of
ethnicity: the concept of ethnicity at least provides a framework within which observ-
able differences between the culture of the ager Faliscus and the surrounding areas can
be evaluated for their relevance to a ‘Faliscan identity’, even if the points where the
Faliscans in our view differed most from their neighbours may not at all have been
those the Faliscans themselves or their neighbours perceived as relevant.15
The difficulties of establishing what features did or did not play a role in ethnic
identity are great, especially when the material from which these features must be
derived is almost entirely archaeological. In the case of material objects, especially
those of daily use, it is very hard to establish what role they played in the perception of
ethnic identity: if they were associated with ethnic identity at all, this is often not
because they were produced to be the bearers of such an identity, but because they had
this role thrust upon them according to the historical context in which they were used,
and so assumed a symbolic value that was independent of their intended practical use.
15
The following studies on ethnic identity in the ancient world I have found especially useful:
Cornell 1997 (on early Rome), Gnade 2002 (on the Volscians at Satricum), Roymans 2004 (on
the Batavians), Dench 2005 (on Rome), and Terrenato 2007 (on social change in South Etruria).
26
Of course, the more an object is or becomes linked to non-material purposes, such as

religious or magic ritual, the more easily it may acquire this added symbolic value. The
same is true of objects used in contexts where such identity plays a crucial role or is
contrasted with other identities, such as distinctive armour or weaponry. Yet as this
value is not normally an inherent feature of the object itself, data relevant to ethnic
identity cannot normally be derived from individual objects unless the context in which
and the purpose, symbolic or otherwise, to which the object was used can be interpreted
as having relevance to this ethnic identity.
Further difficulties with regard to establishing the ethnic identity of the Faliscans
are of course the fact that the historical sources on the Faliscans are all written by
‘outsiders’ writing from a Roman perspective that may have influenced or even biased
these authors (cf. §2.7.1b-c), and the fact that these authors usually wrote several
centuries after the events on which they report took place.
2.3. What constituted a ‘Faliscan identity’?
2.3.1. Faliscan material culture. The problems in the interpretation of material culture
mentioned in the preceding section are unfortunately very acute in the case of the
Faliscans. Thus Cato (Agr. 4.1, 14.1) mentions a praesepe Faliscum, but this type of
cattle-stall does not appear to have been limited to the ager Faliscus alone, and in any
case will hardly have been a major feature of Faliscan identity. On the other hand, a
possibly significant feature may have been the distinctive weaponry described by
Dionysius (tîn Óplwn tîn polemisthr…wn kÒsmoj, ¥spidej 'Argolikaˆ kaˆ dÒrata, 1.21.1),
as such apparel may very well have had symbolic value for the combatants: it could in
fact come under the heading of ‘material expressly created to be the bearer of ethnic
identity’. Unfortunately, Dionysius’ account leaves it unclear how specifically Faliscan
these weapons were, and the remark may simply have been made to draw attention to
the alleged Faliscan connection with Argos (cf. §2.4.1 and §2.3.4).
In general, archaeological sources show that the material culture of the ager
Faliscus did not differ greatly from the remainder of South Etruria. In the earlier periods
the area seems to have known some more or less distinctive styles of pottery (see e.g.
Baglione 1986), and also in later periods the Faliscan workshops can be identified by
their own styles (see e.g. Adembri 1985, 1990, Schippa 1980). In contexts of competi-
tion or conflict such as the successive emergence of Veientan and Roman power in
South Etruria, even daily objects may have had a role in stressing or expressing who
presented themselves, or were regarded, as Faliscan, Etruscan, or Latin. There is,
unfortunately, no way of knowing this with any amount of certainty, since the context
of their use is on the whole not specific enough (as opposed to e.g. the material from
Satricum on which Gnade’s (2002) study of ethnic identity is based).
27
CHAPTER 2
More relevant to Faliscan identity may have been the material culture related to
burial rites and religious cults, for these present several distinctive features. Thus,
Faliscan tombe a pozzo were often provided with a loculus, a custom for which parallels
are known only from Veii (Baglione 1986:129), while the seventh-century tomba a
fossa 24/XLII of the La Penna necropolis at Civita Castellana furnishes an instance of a
tree-trunk burial with impasto discs or ‘shields’, which has parallels only at Veii and in
Latium (cf. Baglione 1986:136-9). The use of the rock-cut tombs that today form such a
conspicuous feature of the landscape also seems to have been more or less confined to
the agri Faliscus and Capenas. The custom was continued at Capena, Sutrium, and
Nepete after these towns became colonies in the early fourth century, as well as at
Falerii Novi, and may well have been regarded as a distinctive Faliscan feature by the
neighbouring areas. For the inhabitants of the ager Faliscus this type of burial may have
been due in part to the nature of the terrain and the availability of existing tombs rather
than to feelings of ethnic identity, but family tradition must have played an important
role, and such a tradition is a potent feature in defining one’s identity. With regard to
cultic objects, there are the testimonies of Dionysius (1.21.2) and Ovid (Am. 3.13), who
both described the paraphernalia of the cult of Juno, but they appear to have been more
interested in noting similarities with the cult of Hera at Argos. There are indications that
the cult itself may have played a part in Faliscan ethnic identity: see §2.3.4.
Somewhere between material and non-material culture lie Faliscan literacy and
literature. The Faliscan alphabet was developed independently at a very early date (see
§11.2.1), and although there are no ancient sources to mention it, this alphabet will have
made written Faliscan clearly distinguishable from documents in the Latin or Etruscan
alphabet. Whether there existed any kind of Faliscan literature is unknown. The
carmina Fescennina were ascribed to a Faliscan origin (cf. §2.1.2), and so, perhaps, was
the metrum Faliscum ( ), although Terentianus Maurus (CGL
6.385.1992) and apparently also Servius (CGL 4.465.5) ascribed this to an unidentified
Serenus. The Faliscans were also credited with several supplements to the Lex XII
Tabularum (Serv. in Verg. A. 7.695). Neither the carmina nor the leges of the Faliscans
need necessarily have been committed to writing, however, and the existence of the
infamous Faliscan schoolmaster (see §2.5.1) is evidence of literacy rather than of
literature.
2.3.2. Faliscan society. Close to nothing is known of the structure of Faliscan society.
The inscriptions furnish only familial appellations like pater (only in the theonym [die]s
pater MF 62) and mater, filius and filia, and uxor (for attestations see §6.2.55, 45, 24-
25). The inheritance of the gentilicium through the male line and the use of patronymic
rather than metronymic filiations show that in these aspects at least Faliscan society was
patriarchal, and thus did not differ in this respect from the societies of the surrounding
areas, except perhaps from the Etruscan, where metronymic filiations are found.
28
On the subject of the names themselves, much has been made of gentilicia (see
§7.1.1, §7.10.3), but what is largely disregarded is that the ager Faliscus shows several
praenomina that were unique to the ager Faliscus or exceedingly rare elsewhere (§7.7.2,
§7.10.4-5). Examples of this are Volta and Iuna: Volta and Iuna are all the more
surprising since they are male praenomina in -a, a category absent from the Latin or
indeed the Italic onomasticon. Gauius and Gauia, although well-known elsewhere,
occur in the ager Faliscus with a far greater frequency than anywhere else. It is unclear
in how far any of this was relevant for feelings of ethnic identity, but someone called
Iuna may, when in Rome, have been immediately recognizable as a Faliscan, in which
case the name could be part of an ethnic identification: see §2.3.4 for a possible local
significance of this name.
A point which to my knowledge has never been raised is how Faliscan society
was stratified. Etruscan and Latin society seem to have differed in this respect espe-
cially where the status of slaves and freedmen was concerned (see Heurgon 1961:74-94,
W.V. Harris 1971:114-29, and Rix 1994 passim), and the point could therefore provide
an interesting insight into the nature of the Faliscans’ status as a distinct cultural unit
within Etruria. Unfortunately, there are no data to show whether the position of the
Faliscan slaves resembled that of the Latin serui or that of the Etruscan servile class,
which consisted at least partly of freeborn men, described by Dionysius (9.5.4) as
penšstai ‘serfs, bondsmen’. It is remarkable, but probably not significant, that Zonaras
(8.18.1, from Cassius Dio) uses tÕ douleàon rather than toÝj doÚlouj or toÝj o„kštaj to
describe the slaves of the Faliscans.
Neither is it clear whether the status of the Faliscan freedman was more like that
of the Latin libertus or that of the Etruscan lautni.16 Falerii Veteres has yielded two
Middle Faliscan sepulchral inscriptions apparently naming libertae. The first, MF 41,
names a l]oifirtạ called loụṛia, who was interred in the same loculus as the apparently
freeborn fasies  c[ai]sia, but the implications of this are unclear. A Late Faliscan
sepulchral inscription from Falerii Novi, LF 221, gives a second instance, a loferta
called uipia  zertenea, the mother of the homo nouus Marcius Acarcelinius. Other
instances are ṭị [ ] ṭịria lo[?---]|l[e]a  cs  f MF 155 and [---]*i  u[o]ltiai lo MF 165.
There are, however, also two instances of a double gentilicium in Middle
Faliscan uel[ ]uisni olna MF 82 and Middle or Late Faliscan m tito tulio
uoltilio hescuna MLF 346, which appears to have been a typical designation of the
Etruscan freedman (Rix 1965:376-8, 1994:97-111): see also §7.6. This could imply
that Faliscan society resembled that of the Etruscan communities at least in this
respect.
16
In South Etruria, lautni only occurs in Ta 1.182: as it is very frequent elsewhere, especially in
the ager Saenensis, at Perusia, and at Clusium, this may not be coincidental. Rix (1994:107),
however, thinks otherwise.
29
CHAPTER 2
2.3.3. Faliscan magistracies. We are likewise badly informed on the structure of the
Faliscan magistracies. The literary sources name no Faliscan magistracies except the
political priesthood of the fetiales described by Servius (in Verg. A. 7.695), and
apparently by Dionysius (ƒero… tinej ¥ndrej ¥noploi prÕ tîn ¥llwn „Òntej spondofÒroi,
1.21.1). Fetiales are also mentioned for several Latin towns, and were as such not
distinctive of the ager Faliscus. (Servius’ reference to Aequi Falisci as the place of
origin of the Roman fetiales is a fabrication to explain the word aequus in this
toponym, cf. §2.1.2.) The inscriptions likewise provide few data on Faliscan magis-
tracies. Roadside inscriptions that probably name the magistrates responsible for the
building and the maintenance of these roads (MLF 206, 207, 210, LtF 205, 290, Lat
291, perhaps also MLF/Etr 356-357?) give only names, not magistracies.
The only magistrates named in the Middle Faliscan inscriptions are the efiles
(MF 113-117), whose functions may at least partly have corresponded to those of the
Roman aediles (Vetter 1953:292-3, Combet Farnoux 1980:137-42), and the rex. Note
that G. Giacomelli (1963:243, 1978:521, 530) has suggested that efiles is a calque on
Latin aediles, a suggestion adopted e.g. by Rix (1994:96 n.36). It is noteworthy that
both these offices bear Latin names and by and large correspond to offices known
from Latin towns, while there is no mention of Etruscan magistracies at all (I do not
adopt Vetter’s reading m]ạrọ in MF 91).
The rex is the only office found both at Falerii Veteres (MF 90, and perhaps MF
91) and at Falerii Novi (LF 249 and LtF 231, cf. fig.2.1). He probably had a sacral
function, like the Roman rex sacrorum. The fact that the rex occurs in cursus may
suggest that the office of the Faliscan rex, unlike that of the Roman rex sacrorum, was
not permanent: its place at the end of the cursus may not be due to rank, but to the
fact that it was not part of the normal Roman cursus. Perhaps the rex performed a
periodically returning sacral function that required some equivalent of imperium, like
the Roman dictator clavi figundi causa.17
The inscriptions from Falerii Novi mention other magistracies (cf. fig.2.1), all
reflecting the Roman organization of the town: it cannot be ascertained if comparable
magistracies existed in the area in the period before 240, and by what names they
were known. Two public inscriptions mention two collegiate preṭ[ores (LF 213), and
a pretod acting on behalf of a (presumably local) senate, de | zenatuo sententiad
(LF/LtF 214). The cursus honorum of the sepulcral inscriptions name the quaestor
(LF 242-243, 245, 247, Lat 237-238, 219), the praetor18 (LF 242-243, 247-248, Lat
240), the duouiri (LF 243, 247-249, Lat 237, 240), and the censor (LtF 231-232).
17
The rex could, for instance, have played a role in the ƒerÕj g£moj that according to Taylor
(1923:65) may have constituted a feature of the Faliscan cult of Juno.
18
As the praetor occurs here in cursus honorum, it is unlikely that it is a Latin rendering of a
local supreme magistracy such as the Etruscan zilaθ: a point worth making, as even in the
early Empire praetor could still be used to render Gallic vergobret (cf. Roymans 2004:64).
30
quaestor praetor duouir censor rex
MF 90a rex
MF 91a ṛẹx ?
LF 213b pret[ores
LF 214b pretod
LF 242 cụestod pi pretod pis
LF 243c cues[tor p[reto]|r duum]uiru
LF 245 c]ues[tor
LF 247 cue[stor pret[or duum[uiru
LF 248 pre]tor ii duu[muiru
LF 249 duu]ṃuiru rẹ[x]
LtF 231 cen]|so rex
cen]|sor
LtF 232
censo
LtF 233 [---]or (quaest]or? praet]or? cens]or?)d
Lat 237 bis q duouir
Lat 238 q ii e
Lat 240 prae[tor duum]uir
Lat 219a q
a
) From Falerii Veteres: all other inscriptions are from Falerii Novi. b) Public inscription: all others are
sepulchral. c) The text has the honores in decreasing order. d) As [---]or stands at the end of the
inscription, perhaps cens]or, but ux]or is also possible. e) Doubtful reading, although q is certain.
Fig.2.1. Honores in the inscriptions from Falerii Veteres and Falerii Novi.
31
CHAPTER 2
2.3.4. Faliscan religion. There are more sources, both literary and epigraphic, on
aspects of Faliscan religion, and these data could well be relevant to describing
Faliscan ethnic identity, as a distinct religion is quite often the focus of any form of
group-identity. A survey of the material is given by Taylor (1923:60-93) and, more
briefly, in Sant (1985), passim. I briefly discuss here the cults that can be traced to
Falerii Veteres or Republican Falerii Novi.
The major cult of Falerii Veteres seems to have been that of Juno: Ovid in fact
coined the word Iunonicolae to provide the Faliscans with a suitable epithet (Fast.
6.49), and the whole point about the mythical founding of Falerii by the Argives may
well have originated from the fact that Argos was famous for its cult of Hera. The
temple of Juno has been identified with the temple in Contrada Celle at Civita
Castellana (see §14.1.4 and Sant pp.110-3); the cult, which may have included a
consort pater Curris (Tert. Apol. 24), was thought to have derived from the cult of
Hera at Argos (Dion. 1.21.2, Ovid. Fast. 6.45-9). Worship at the Celle temple
continued after 241, and the temple was in fact completely restructured in the second
century BCE (cf. Potter 1979:100, Moscati 1985b:70-1). Both Dionysius (1.21.2) and
Ovid (Am. 3.13) describe the cult as existing in their own day, although it is unclear
whether this was a continuation of the original rites. Local worship appears to have
persisted for a long time, witness the much later attestations of a pontifex sacrarius
Iunonis Curritis in CIL XI.3100 and 3125. The Roman cult of Juno Curritis is thought
to have originated from an euocatio of the Faliscan deity in 241 (Taylor 1923:68).19
This cult of Juno may very well have played a part in the ethnic identity of the
Faliscans, for several reasons. On the one hand, not only do the sources treat it as an
almost emblematic feature of Faliscan culture, even using it to give a context to other
features of Faliscan culture such as descent, foundation-myth, and weaponry, but the
cult was also the subject of an euocatio, a ritual aimed at least partly at removing the
deity at the very core of the enemy’s religion. On the other hand, the worship of Juno
at Falerii Veteres continued after the fall of Falerii in 241: whatever else was de-
stroyed of the Faliscan culture or even of Falerii Veteres itself, people kept coming to
the old temple, which, as said, was restructured in the second century BCE. The cult
as described by Dionysius (1.21.2) and Ovid (Am. 3.13) either still continued the
ancient cult in some way, or the cult as it was remembered in the time of these authors
had been considered important enough to be worthy of a revival. In such a context, the
popularity of the Faliscan man’s name Iuna (§7.7.1.29), even if etymologically
unconnected with Iuno, may well have been due to a perceived etymological connec-
tion between the two.
19
The only evidence for this seems to be that Juno as worshipped at Falerii was sometimes also
referred to as Curritis. Ogilvie (1965:674) pointed to Ovid. Fast. 3.843, but that text refers to
the euocatio not of Juno but of Minerva.
32
Another famous cult ascribed to the Faliscans is that of the worship by the Hirpi
Sorani on Mount Soracte:
“Soractis mons est Hirpinorum20 in Flaminia conlocatus. in hoc autem monte cum
aliquando Diti patri sacrum persolueretur (nam diis manibus consecratus est) subito
uenientes lupi exta de igni rapuerunt. quos cum diu pastores sequerentur, delati sunt ad
quandam speluncam, halitum ex se pestiferum emittentem, adeo ut iuxta stantes ne-
caret: et exinde est orta pestilentia, quia fuerant lupos secuti. de qua responsum est,
posse eam sedari, si lupos imitarentur, id est rapto uiuerent. quod postquam factum est,
dicti sunt ipsi populi Hirpi Sorani: nam lupi Sabinorum lingua uocantur hirpi. Sorani
uero a Dite: nam Ditis pater Soranus uocatur: quasi lupi Ditis patris” (Servius, in Verg.
A. 11.785);
“haut procul urbe Roma in Faliscorum agro familiae sunt perpaucae quae uocantur
Hirpi; hae sacrificio annuo quod fit apud montem Soractem Apollini super ambustam
ligni struem ambulantes non aduruntur, et ob id perpetuo senatus consulto militiae
omniumque aliorum munerum uacationem habent” (Pliny, NH 7.2.19).
Other sources are Vergil (A. 11.785-9), Silius (5.175-81), Servius (in Verg. A.
11.787), and Solinus (2.26, echoing Pliny). There are no epigraphic attestations of this
cult: the word sorex read in LtF 231-232 is often interpreted as the name of the
Soractean priesthood, but even if it is indeed a priestly title and not (as I think it is) a
ghostword, the sorex need not have been connected with the worship on Mount
Soracte (cf. Macurdy 1921 and Peruzzi 1963b). The name Hirpi Sorani is usually
regarded as the cultic epithet of a small group of gentes performing a hereditary ritual,
but it recalls such totemic tribal names as Hirpini, derived, like Hirpi, from hirpus,
and Picentes, derived from picus, and Servius’ account is not incompatible with an
explanation of the Hirpi Sorani as a small group of immigrants from the Sabellic-
speaking area on the other side of the Tiber: see §2.5.2 and §9.3.
With the exception of Servius (in Verg. A. 11.785), all sources (including Serv.
in Verg. A. 11.787!) link the Hirpi Sorani to the worship of Apollo. Taylor
(1923:83-91) explains this at some length through the assumption of a confusion of
Apollo Lycaeus and Mars; others have doubted the whole idea of the worship of
Apollo on Soracte (REA s.v. Hirpi Sorani). There is epigraphic evidence for a cult of
Apollo at the Tempio Maggiore on Colle di Vignale, however, the oldest cultic centre
of Falerii Veteres (see §14.1.2 and Sant pp.85-6), already for the Early Faliscan
period, in the dedication apolonos EF 10 (c.500-475, apparently the oldest mention of
the deity in an Italic language); from Falerii Veteres, too, is the inscription apolo LF
65 (300-250). Falerii Novi has yielded a dedication [. u]mpricius c f | aburcus
q | [a]polinei dat Lat 219 (c.120-50).
20
The confusion with the Hirpini (if it is indeed a confusion) may be due not only to the
similarity in name, but also to the fact that the poisonous fumes described here were also a
recognized feature of the Hirpinian temple of Mefitis at Ampsanctus (Cic. Div. 1.36.79, Plin.
NH 2.95.208).
33
CHAPTER 2
The sixth- or seventh-century ‘Ceres-inscription’ (EF 1) contains the earliest

epigraphic attestation of Ceres in the phrase ceres far *[0-2]e[1-3]tom
*[3-5]uf[1-4]ui[..]m *[3-4]*ad and, many centuries later, a place ad Cereris near
Falerii Novi is mentioned in CIL XI.3083 (cf. Taylor 1923:76-7).
The Ceres-inscription was thought also to contain references to Liber, but the
readings l[o]ụfir, lọuf[ir, louf[i]ṛ and the interpretation of euios as EÜioj that gave rise
to this idea are now largely abandoned (see §12.2). There is certainly no evidence for
seventh-century “important orientalising inscriptions that document the spread of the
cult of Dionysus among the town’s aristocrats” (ArchFal p.15). Faliscans are men-
tioned in connection with the Bacchanalia-upheaval (Liv. 39.17.6), but there are no
indications that Bacchic worship was present much earlier at Falerii Veteres, as
Peruzzi (1964a:158-9, 1964b) suggested.
A deity attested thus far only at Falerii Veteres is Mercus or Titus Mercus
known from the dedications titoi | mercui | efiles MF 113-117, titoi : mercui MF
118-122, mercui MF 123-126 (texts reconstructed from multiple examples), found in
the temple ‘ai Sassi Caduti’ (see §14.1.3 and Sant p.113). This Titus Mercus may
have had similarities with Roman Mercury, and possibly with the Oscan deity Mercus
known from the dedication mirikui Cm 24: this is discussed extensively by Combet
Farnoux (1980:113-69).
Attested only for Falerii Novi are the worship of Mars, implied by the occur-
rence of a mensis Martius in the Faliscan fasti (Ovid. Fast. 3.87-90) and by the
Faliscan sors inscribed Mauors telum suum concutit (Liv. 22.1.11, Plut. Fab. Max.
2.3; cf. Taylor 1923:78-9). Ovid (Fast. 3.89) furthermore informs us that the mensis
Martius was the fifth month among the Faliscans. The worship of the Capitoline
Triad, is attested for Falerii Novi by the second-century Latin dedication [di]ouei
iunonei mineruai Lat 213, although there is no ground to assume that the triad itself
originated at Falerii (thus Girard 1989). The Roman cults of Minerva Capta (Ovid
Fast. 3.843-4; cf. Girard 1989) and of Ianus Quadrifrons (Serv. in Verg. A. 7.607,
Macrob. Sat. 1.9.13) are mentioned in the sources as having been brought to Rome
from Falerii, apparently after the war of 241 (“perdomitis ... Faliscis” Ovid, “captis
Faleriis” Servius). Worship of Minerva at Falerii Novi is attested in the public
dedication to Minerva de | zenatuo sententiad LF/Lat 214.
Beside these cults, there are epigraphic attestations of the haruspex (LtF 231-
232): as said, the rex (MF 90-91, LF 249, and LtF 231), too, may have had a sacral
function (§2.3.3). The sources also ascribe the priesthood of the fetiales (Serv. in
Verg. A. 7.695, see §2.3.3) to a Faliscan origin and mention two Faliscan festivals, the
Struppearia (“in quo coronati ambulant,” Fest. 410.12-5L, see §6.6.6) and the
decimatrus (Fest. 306.4-6L, see §6.6.2): the latter has parallels in Latin festivals like
the quinquatrus, sexatrus, and septimatrus.
34
The names of most of the Faliscan gods of cult (Ceres, Titus Mercus, Ianus,
Mars, and probably Minerva), priesthoods, and festivals are therefore Italic: only Juno
seems to have been of Etruscan origin (see REA s.v.). Mythological scenes on gems
and mirrors found in the ager Faliscus bear Etruscan legends (cf. e.g. the Etruscan
inscriptions Etr XXV, XXVII, XXVIII, XXXI, XXXIII, XLI, and L), but most of
these may have been imports from other areas (with the probable exception of Etr L):
the sole instance of a Faliscan legend is the Middle Faliscan inscription canumede
[die]s pater cupi‹d›o menerua MF 62.
2.4. The early history of the ager Faliscus
2.4.1. Mythical origins. With the exception of Justin (20.1.13), who referred to it as a
Chalcidian colony, all ancient sources regarded Falerii as an Argive settlement (Cato
apud Plin. NH 3.5.51, Dion. 1.21.1, Steph. Ethn. 626.23-4 Meineke) founded by
Halaesus (Ovid. Fast. 4.73-4, Am. 3.13.32, Serv. in Verg. A. 7.695, Solin. 2.7): a
discussion of these sources may be found in Camporeale 1991. The attribution of
Italian towns to Greek or Trojan founders is of course commonplace, especially at the
periphery of Etruscan influence (see Alföldi 1963:228-38): the connection with Argos
will have been made because of the identification of the Faliscan cult of Juno with
that of Hera at Argos (Dion. 1.21.2, Ovid. Am. 3.13; cf. §2.3.4).
2.4.2. The history of the ager Faliscus until the fifth century BCE. The earliest
history of the ager Faliscus can only be inferred from the archaeological evidence.
The main surveys of this are Barnabei 1894a (but see §1.4.5) and Holland 1925, and,
based on new evidence, Potter 1976, 1979:52-92, Baglione 1987, and Petitti 1990.
Nowhere is there any indication or recollection of Faliscan having been brought
to the area in some way in the centuries immediately preceding its first attestations:
whether Faliscan is viewed as a separate language or as a Latin dialect, it was the
indigenous language in the area in the sense that it was present there before Etruscan.
If Faliscan is a dialect of Latin, therefore, the ager Faliscus and Capenas in all
probability originally have formed a part of a Latin-speaking area to the north of the
Tiber that disappeared (or became Etruscanized in a linguistic sense) with the spread
of Etruscan civilization and language from the coastal centres. Etruscan largely, but
perhaps not wholly, replaced Latin throughout South Etruria – except for the ager
Faliscus and the ager Capenas. I think that the assumption of such a ‘north bank’
Latin-speaking area (and thus a continuous Latin-speaking area that ranged from
Latium adiectum in the south to the ager Faliscus in the north) is not a very difficult
one: note e.g. the early Latin ‘Vendia-inscription’ 479†, dating from the late seventh
(or early sixth) century (cf. Cristofani 1993:25-7).
35
CHAPTER 2
The Etruscan language and culture certainly entered the ager Faliscus, and
Narce, its southernmost major site at the time, became a wholly Etruscan town. Yet
they did not spread rapidly enough or in a sufficient degree to replace the existing
language and culture entirely, perhaps because of the lack of communication with
major Etruscan centres such as Tarquinii and Caere. Significantly, the ager Faliscus
was at this time culturally independent enough from both Etruria and Latium to have
developed an alphabet of its own (§11.2.1-2). The contrast between Etruscan Narce
and Faliscan Falerii appears already in the earliest epigraphic material: of the 19 or 20
early Etruscan inscriptions from the ager Faliscus, 18 are from Narce and the south-
western ager Faliscus (Etr I-XV and XVIII-XX: Etr XXIX is from Corchiano, while
EF/Etr 5, from Falerii, is either Etruscan or Early Faliscan), while all certainly Early
Faliscan inscriptions (EF 1-4 and 6-10) are from Falerii.
An early attestation of the Faliscan ethnicon has been read by Poccetti (1997) in
the seventh-century inscription from Vetulonia, [mi a]ụṿileś feluskeś tuśnutaḷ[a
pa]|panalaś Vn 1.1. He convincingly interprets feluskeś as an Etruscan rendering of
Faliscus, pointing also to fourth-century veluske (probably a name) in an inscription
published by Colonna (1995). (To these could perhaps be added early fifth century
[m]i larisa fẹḷẹṣkẹnas am**[?---] AS 1.40.) If feluskeś is indeed ‘Faliscan’, this has
some very interesting consequences. The form is clearly an Italic, probably Latin,
ethnonym, and its use here by a Faliscan living ‘abroad’ would imply that the
ethnonym was already used by the Faliscans themselves, and that they were therefore
already identifiable as a distinct group. But if Narce was the original main site of the
area and Falerii succeeded it as such only during the fifth century, as is usually
assumed, why was the ethnonym of the area derived from the name of Falerii already
in the seventh? This must mean that Falerii was for some reason regarded as the most
important site even before or during the floruit of Narce. I do not find it hard to
envisage the older autochthonous centre that gave its name to the area as being on its
way to being eclipsed by an emerging competitor whose success was due to its better
connections with the Etruscanized area to the south, and possibly the support of Veii.
A more dubious early source for Faliscan history has been seen in one of the
frescoes of the late fourth-century François Tomb at Vulci. The scene depicted may
refer to events of c.500, as some of its figures also appear in the story of the expulsion
of the Etruscan kings from Rome. It shows local heroes killing foes designated with
ethnonyms like rumaχ ‘Roman’, sveamaχ ‘Sovanian’, and velznaχ ‘Volsinian’: one is
labelled venθicaụ[...]*[.]plsaχs Vc 7.30, and Heurgon (1961:66) suggested that this
plsaχs should be interpreted as ‘Faliscan’. Some scholars hesitatingly adopted this
interpretation (e.g. Alföldi 1963:66, Torelli 1966:1212, Scullard 1967:122-3), but
there is no evidence for it, as Di Stefano Manzella (1977:162) has shown. 21
21
The word has also been taken as the ethnicon of Salpinum (e.g. Pareti 1952:310, Accame in
Alföldi 1963:222 n.1, Baffioni 1967:157 n.144).
36
The early authors on Faliscan stressed the importance of alleged early Sabine
invasions (§1.5; still Pallottino 1987:105), and although there seems to be no evidence
for the large-scale invasions which they presupposed, there may well have been small-
scale immigrations from the Sabellic-speaking areas. Perhaps the Hirpi Sorani (cf.
§2.3.4) were one such group (thus already Taylor 1923:90): although usually regarded
as hereditary priests, the sources, especially Servius (in Verg. A. 11.785), seem to
indicate that they may have been a tribal group (populi in Servius’ account) of
Sabellic origin that migrated to the area around Mount Soracte guided by their totemic
animal22 and preserved an identity, including an ethnonym, until Roman times. In the
Faliscan onomasticon, there are several gentilicia that are of Sabellic origin (§7.8.2),
and there are linguistic traces, too, of the presence of speakers of Sabellic languages
in the area (§9.3), especially in inscriptions from the ager Capenas.
During the late sixth and early fifth century the most important development is
the growth and the expansion of Veii as the major Etruscan centre of the Lower Tiber
basin. To the east, Capena, probably already a cultural or economic dependency, now
became a political dependency as well. During the fifth century, however, Faliscan
Falerii eclipsed Etruscan Narce as the central site of the ager Faliscus, perhaps
because of its more central location with regard to the trade routes through the South
Etruria (cf. §2.2.3), whose importance steadily increased with the emergence of Veii
as the dominant centre to the south (cf. Baglione 1987).
The vacuum arising from the decline of Narce was filled by the extension of
Veientan power into the area that may well have been the south-western ager Faliscus,
beyond the range between the Monti Sabatini and Mount Soracte (§2.1.1), where Veii
either reinforced or founded Sutrium and Nepete, thus controlling the western access
route to the ager Faliscus through the ‘Sutri Gap’. The effects of this ‘second Etruscan
wave’ into the agri Faliscus and Capenas, however, were to be short-lived, and
quickly to be superseded by the expansion of Rome into South Etruria, which would
change the situation completely.
2.5. Falerii, Veii, and Rome in the fifth and fourth centuries BCE
2.5.1. The Fidenate wars and the siege of Veii. The history of the ager Faliscus from
the late fifth to the middle of the third century is documented by the much later
historical sources, especially Livy. Diverse as they are, these present a consistent
picture of a Falerii doggedly resisting the gradual extension of Rome’s influence in
South Etruria. It should of course be stressed that such sources can only be used with
22
Such treks are reported for the Picentes (Str. 5.4.1, Paul. Fest. 235.16-7L) and the Hirpini
(Str. 5.4.12, Paul. Fest. 93.25-6L): to the latter Servius (in Verg. A. 11.785, 11.787) assigned
both Mount Soracte and the ritual performed there by the Hirpi Sorani.
37
CHAPTER 2
an adequate amount of critical sense, especially where the earlier periods are con-
cerned, but this does not mean that they are altogether useless: although individual
events may be questioned, the main trends and events are clear, and often supported
by archaeological evidence. A useful survey of the historical material is provided by
Shotter 1976: cf. also Cornell 1986, 1991, and Gnade 2002:136-9 for discussions on
the relative merits of historical and archaeological evidence. In the discussion in this
and the following sections, I follow Livy’s account as the primary source.
The earliest events in Roman-Faliscan relationships to be recorded are the late
fifth-century wars over Fidenae. Due probably to the uncertainty about the date of the
duel between A. Cornelius Cossus and Lars Tolumnius (cf. Ogilvie 1965:563-4),
famed as the second occasion on which a Roman commander brought home the spolia
opima, the sources are confused: Livy’s account (4.17-34) probably contains repeti-
tions, Diodorus (12.80.6-8) mentions only the second of the two wars described by
Livy, and Florus (Epit. 1.6/12.9) telescopes the wars into the siege of Veii.
In Livy’s account, Fidenae defects to Veii in 438 and their joint armies cross
into Roman territory in 437 (4.17.1-11). There they are joined by the Faliscans, whose
eagerness to go home makes the king of Veii, Lars Tolumnius, decide for the battle in
which he is killed (4.17.11-18.8). The Romans then conduct punitive expeditions into
the agri Veientanus and Faliscus in 436 (4.21.1-2). When the Romans threaten
Fidenae in 435 (4.21.6-22.6), the Faliscans are unwilling to participate in a new war
(4.21.8), although it later appears that they took part in the battle at Nomentum
(4.32.3). After the fall of Fidenae, Veii and Falerii in vain seek help from the Etruscan
League at Fanum Voltumnae in 434 and 432 (4.23.4-24.2, 25.7-8).
In 427 war erupts again, with the Veientes and Fidenates making forays into
Roman land (4.30.5-6) until Fidenae is defeated in 426 (4.30.5-34.7). It is unclear
whether the Faliscans participated in this second war. Frontinus’ (Str. 2.8.3) reference
to a campaign against the Faliscans by T. Quinctius Capitolinus during his consulate
in 446 is usually regarded as an erroneous reference to his famous campaign against
the Volscians in that year. However, it could be a recollection of an otherwise
unknown campaign against the Faliscans by T. Quinctius Cincinnatus Poenus when
he was tribunus militum consulari potestate in 426: like T. Quinctius Capitolinus, he
inflicted a defeat upon the Volscians during his consulate, in 431 (4.31.1-34.7).23
Falerii and Capena next appear as the only allies of Veii during the siege of
402-395, being credited by Livy with the not unjustified fear that “quia proximi
regione erant, deuictis Veiis bello quoque Romano se proximos fore” (5.8.5). Sending
relief forces in 402 and 399 (5.8.4-12 and 5.13.9-13, perhaps a repetition), and asking
23
The campaign by the dictator Q. Servilius Priscus (in 418, cf. REA s.v. Servilius 75) men-
tioned by Frontinus (Str. 2.8.8) on the other hand may in fact have been the campaign against
the Aequians described by Livy (4.47.1-7).
38
for help from the Etruscan League at Fanum Voltumnae in 397 (5.17.6-10), they
become involved in the war, which is carried to their own lands from 401 onwards
(5.10.2, 5.12.5, 5.14.7, 5.16.2). M. Furius Camillus, before taking command of the
siege of Veii in 396, considers the threat posed by Falerii and Capena serious enough
to deal with these towns first, breaking the resistance of Capena in 395 and granting
the city a foedus (5.19.7-8, 5.24.3; cf. W.V. Harris 1971:89).
After the fall of Veii, he tries the same tactics against the Faliscans in 394,
which results in an unsuccessful siege (5.26.3-10). The deadlock is broken by a
Faliscan schoolmaster delivering the sons of the Faliscan nobility as hostages to the
Romans, an offer from which Camillus indignantly refuses to profit, whereupon the
Faliscans, moved by his noble gesture of returning the boys unharmed, immediately
offer unconditional surrender (5.27): a story-with-a-moral that apparently enjoyed
great popularity, if this can be judged from the number of sources in which it appears
(Dion. 13.1-2, Plut. Cam. 9-11, V. Max. 6.5.1, Fron. Str. 4.4.1, Polyaen. Strat. 8.1.7,
Flor. Epit. 1.6/12.5-6, Avit. fr.2 apud Prisc. CGL 2.427.1-6, Eutrop. 1.20, Hier. Ep.
57.3, Oros. 3.3.4, and Zonar. 7.22, from Cassius Dio).
It is possible that Falerii did not escape capture, however. According to
Diodorus, the Romans took the town by storm in 392 (14.96.5) but concluded a peace
in the next year (14.98.5); Livy twice (5.43.7, 6.7.4) makes Camillus refer to the
capture of Falerii, and the main cause of his soldiers’ later discontent is his refusal to
let them sack the town, which hardly understandable if Falerii was surrenderd and not
taken. The euocatio of Minerva, Janus Quadrifrons, and perhaps of Juno Curritis,
usually placed at the conclusion of the war of 241, could then have taken place at this
date, repeating the euocatio of Juno Regina from Veii in 396 (Liv. 5.21).
2.5.2. The wars of the fourth and early third century BCE. While Rome is trying
to regain its control over the Latin league after the Gallic siege, the cities of Etruria
revolt in 389 (Liv. 6.2.2). The main bones of contention are Sutrium and Nepete,
controlling the route through South Etruria (§2.2.3): “loca opposita Etruriae et uelut
claustra inde portaeque” (Liv. 6.9.4). After some Etruscan successes, the Romans
regain control (Liv. 6.3.1-10) and establish themselves by grants of citizenship to
‘faithful’ Veientes, Capenates, and Faliscans in 388 (Liv. 6.4.4; cf. Harris
1971:192-9). War is continued in 387 under the leadership of Tarquinii (Liv.
6.4.8-11): in the confused fighting, again centred on Sutrium and Nepete (Liv. 6.9.3-4,
6.9.7-10.6), the Romans come off best. They secure the area by founding colonies at
Sutrium, Nepete, and Capena shortly afterwards (the exact dates are debated).
Tarquinii, together with Falerii, again takes advantage of the situation by
overrunning South Etruria in 358 (Liv. 7.12.5-6) or 354-353 BC (Diod. 16.31.7) as far
as the salinae near the mouth of the Tiber in 356 (Liv. 7.17.6-9) or 353-352 (Diod.
16.36.4). For Falerii this may have meant regaining control over Sutrium and Nepete,
39
CHAPTER 2
although the former is the Roman base camp again in 357 (Liv. 7.16.7-8). The next
years see a number of indecisive skirmishes and one pitched battle, decided in favour
of the Romans (Liv. 7.17.2-5, Fron. Str. 2.4.18). In 351 hostilities are concluded with
forty-year indutiae (Liv. 7.22.5-6), which in 342 are replaced by a foedus (Liv.
7.38.1), perhaps, as Shotter (1972:32) suggested, one similar to the one granted to
Caere in 353 (Liv. 7.20.8). An epigraphic record of this war is preserved in the first
century CE (!) Latin eulogium of the Tarquinian praetor, A. Spurinna (see Torelli
1975:67-92), which mentions the Faliscans in an unfortunately fragmentary passage
(falis[c---] is legible, but nothing of the context remains).
For the years around 300, there is again only Livy’s account. The Faliscans,
although not mentioned, will certainly have been present when Sutrium was besieged
by “omnes Etruriae populi praeter Arretinos” (Liv. 9.32.1-12) in 311, and the story
(Liv. 9.36) of Fabius’ journey through the silua Cimina (apparently to cross the Tiber
near Horta or the Grotta Porciosa site) to cement an alliance with the Umbrians,
shows that normal travel through the ager Faliscus was temporarily too dangerous.
After this, the Faliscans kept quiet for some time, so that on the campaign against
Volaterrae in 298, the Roman baggage train could be left at Falerii “modico praesi-
dio” (Liv. 10.12.7). The Faliscans are absent, too, from Livy’s account of the war of
295: writing of the events of 293, he states that the Faliscans “per multos annos in
amicitia fuerant” (10.45.6).24 The Romans are sufficiently wary of them, however, to
post a guard over the ager Faliscus in 295 (Liv. 10.26.15), probably to prevent a
capture of Sutrium that would have cut their supply lines. Not surprisingly, at the
outbreak of the war of 293, it is the report of Faliscan participation that spurs the
Romans into action (Liv. 10.45.6): the Faliscans are quickly cowed into submission
by Sp. Carvilius Ruga’s capture of an unidentified Troilum and five castella, and are
granted indutiae annuae (Liv. 10.46.10-15). Archaeological sources show that around
this date the route through the ‘Sutri Gap’ began to be opened and enlarged, and the
area surrounding Sutrium and Nepete to be brought under culture (Potter 1979:96-7),
indicating that the Romans were firmly establishing themselves in the area.
The Faliscans appear in the sources “not as great instigators of action ... but as
ready to support those in the front line” (Shotter 1974:29), allying themselves first
with Veii and later with Tarquinii, and taking advantage of every opportunity to
subvert the growing Roman influence in South Etruria. In view of their consistent
record of ‘bad behaviour’, the relative clemency with which the Romans treated this
dubious ally, located in such a strategic position on the routes to both Etruria and
Umbria, is remarkable: they are in fact often treated far better than the situation seems
24
I can see no way of reconciling this with Frontinus’ mention (Str. 2.5.9) of a stratagem by
Cn. Fulvius (in 295?), “cum in finibus nostris exercitus Faliscorum longe nostro maior castra
posuisset”.
40
to justify. The clement treatment of the Faliscans after their surrender to Camillus as
described by Livy almost seems to symbolize their relationship with the Romans.25
2.6. The war of 241 BCE and its consequences
2.6.1. The war of 241 BCE. In the consulate of A. Manlius Torquatus Atticus and
Q. Lutatius Cerco, i.e., in the consular year 241-240 BCE, the Faliscans became
involved in a new war against Rome that would mark the end of the ager Faliscus as
an independent political unit. Two interesting discussions of this war and its conse-
quences are Loreto 1989 and Di Stefano Manzella 1990.
The date of 241-240 appears in Polybius (1.65.2), Livy (Per. 20), Valerius
Maximus (6.5.1), Eutropius (2.28), and Zonaras (8.18.1, from Cassius Dio), and is
epigraphically confirmed by the Fasti Triumphales (AUC ĐXII), which mention the
triumphs de falisceis granted to both consuls on the first and the fourth of March
240, and by the inscription on a South Etrurian cuirass published by J.-L. Zimmermann
(1986), q lutatio c f a manlio c f | consolibus faleries capto . Only Orosius
(4.11.10) places the war slightly later, in 238.
The surprising fact that the Faliscans seemingly waited until the end of the
Punic War instead of taking advantage of it is often ascribed to the expiration of the
hundred-year foedus concluded in 342 or of the indutiae granted in 293, curiously
making the date of a revolt dependent on a treaty imposed by the oppressor. Salmon
(1969:65) suggested that the Faliscans felt threatened by the colonization of
Spoletium in 241, but there appears to be no reason why they should, and the only
source to mention both events together, [Liv.] Per. 20, presents them in the reverse
order: “Falisci cum rebellassent, sexto die perdomiti in deditionem uenerunt.
Spoletium colonia deducta est.” It is much more plausible that 241 was not the year
when the Faliscans started their revolt, but the year when the Romans were free to
take action against them (Loreto 1989:720-1, Di Stefano Manzella 1990:342). The
conflict may have started much earlier: its occasion and cause are unknown, although
both a refusal to provide troops for the Punic War and problems in the renewal of the
foedus of 342 have been suggested (Loreto 1989:726-7).
The sources describe the war as a six-day campaign ([Liv.] Per. 20, Eutrop.
2.28; ™n Ñl…gaij ¹mšraij, Polyb. 1.65.2) with two pitched battles (Zonar. 8.18.1, from
Cassius Dio) that cost the lives of 15,000 Faliscans (Eutrop. 2.28, Oros. 4.11.10),
25
AntFal (p.18) describes these wars as “bitter struggles against expansionism of Rome,
whose fury against Falerii, which vied with Rome’s ceramic producers for ever more distant
markets, was no doubt largely due to serious trade rivalry.” There appear to be few signs of
‘Roman fury’, however, and I doubt whether Rome would go to war over pottery markets at a
time when it was land and possessions that marked the status of a town or individual.
41
CHAPTER 2
whereupon Falerii was surrendered ([Liv.] Per. 20, V. Max. 6.5.1). Both consuls were
granted a triumph (Fast. Triumph. AUC ĐXII), which lends some credibility to the
astonishing number of casualties for an army of such a small area, because of the
requirement of 5,000 enemy casualties to apply for a triumph (Di Stefano Manzella
1990:342, cf. V. Max. 2.8.1: see, however §2.7.1e). The Faliscans had to hand over t£
te Ópla aÙtîn kaˆ tÕn †ppon kaˆ t¦ œpipla kaˆ tÕ douleàon kaˆ tÕ ¼misu tÁj cèraj
(Zonar. 8.18.1, from Cassius Dio), which sounds as if Rome was making a nasty
example of Falerii (“aduersus quam saeuire cupiens populus Romanus,” V. Max.
6.5.1), perhaps as a warning to other unreliable allies.
Loreto (1989:730-3) rightly concludes that the war must have had a symbolic
value to the Romans that far exceeded its actual importance (witness the double
triumph for such a minor campaign): after the Punic War, they needed a quick and
decisive victory to re-establish themselves at home and abroad.
2.6.2. The division of the territory and the founding of Falerii Novi. The sources
report that as a consequence of the war of 241, half of the Faliscan territory was
forcibly ceded to the Romans: “[M£llioj Torkou©toj] tÕ ¼misu tÁj cèraj ¢fe…leto”
(Zonar. 8.18.1, from Cassius Dio), “agro ... ex medietate sublato” (Eutrop. 2.28).
According to Loreto (1989:723), the ceding of half the territory was not in itself
extremely severe, since the usual sanction was one-third: the same sanction had in fact
been imposed on Caere in 273.
The ceded area would obviously have been centred on the new Falerii and
extended westward to Sutrium and Nepete, so that the territories of Veii, Capena,
Sutrium, Nepete, and Falerii Novi now constituted one continuous stretch of land
under Roman control. Di Stefano Manzella (1990:345) suggests that the ceded
territory extended eastward to the Tiber to include the fertile tablelands north of the
Rio Maggiore and the Treia, and cut what remained of the ager Faliscus in half. This
is indeed very plausible: it seems in fact extremely likely that the Roman area
extended north-eastward towards Gallese and the Grotta Porciosa site, placing the
Romans in direct control of the strategically located Tiber crossing that gave access to
the Sabine interior and Umbria (note the colonization of Spoletium in 241, which in
[Liv.] Per. 20 (quoted in §2.6.1) directly follows the Faliscan war), and which in 220
would be used for the Via Flaminia.
The town of Falerii itself was destroyed and a new Roman Falerii was built
c.4.5 km to the west, in the middle of a section of flat tablelands. This was done
probably somewhere between 240 and 220, when the Via Amerina was extended
northward from Nepete; the same is implied by Zonaras, who explicitly places the
demolition of the old town and the founding of the new town as ‘later’ than the war
itself: Ûsteron d' ¹ mèn ¢rca…a pÒlij e„j Ôroj ™rumnÕn ƒdrumšnh katesk£fh, ˜tšra d'
òkodom»qh eÙšfodoj (8.18.1, from Cassius Dio).
42
The usual interpretation of Zonaras’ words is that Falerii Veteres was destroyed
and its inhabitants resettled en bloc to a less defensible site, like the inhabitants of
Volsinii in 265 (Zonar. 8.7.4-8), but this is an over-simplification. The continued use
of some of the temples of Falerii Veteres attested by Dionysius (1.21.1-2) and Ovid
(Am.3.13) has long been confirmed by archaeological findings (see Potter 1979:
99-101, Andrén 1940:88). Continued habitation at the site of the old town is implied
by the toponomastical data (cf. §2.1.2), although the archaeological evidence for it is
slight, consisting mainly of a few Republican burials (see Moscati 1985b:70-1).26
Di Stefano Manzella (1990:349-50) also rightly questions the idea that the
population of Falerii Novi consisted simply of transferred inhabitants of Falerii
Veteres. It seems indeed unlikely that the Romans would deport their enemies to a
less defensible site only to provide this with walls that are still among the best
preserved works of Republican military architecture, and much more plausible that
many or most of its inhabitants were Roman or Latin immigrants (like the craftsman
t fourios *[ ]f ̣ in Lat 216?). As Falerii Novi was the administrative and military
centre of a newly-occupied territory, its administration must have been pro-Roman,
and will have included few members of the ruling class of Falerii Veteres: Di Stefano
Manzella (1990:349-50) suggests that it may have comprised Faliscan families whose
loyalty to Rome in 389 had been rewarded with the citizenship (§2.5.2).
A related point is the legal status of Falerii Novi. It is usually assumed that the
town was at first either a ciuitas foederata or a municipium sine suffragio, as it was
during the first centuries CE (cf. CIL XI.3083, 3103, 3112, 3116, 3121, 3125, 3127,
3147, and 3155a,1). The honores in the cursus honorum from Falerii Novi (§2.3.3)
also point to municipal rather than to colonial status, as M. Mancini (2002:38-40)
points out. At some time during its history, however, the town must have been a
colonia, since in CIL XI.3089 and 3094 the emperor Gallienus is honoured as
redintegrator coloniae Faliscorum.27 It is apparently this earlier colony that is
referred to by Pliny (NH 3.5.51) and the Liber Coloniarum (217.5). Since these
statements are difficult to reconcile with the epigraphic evidence for a municipium,
there is a tendency to disregard them altogether.
Di Stefano Manzella (1990) suggested that Falerii Novi was in fact a (Latin)
colony already from its foundation in or shortly after 241, becoming a municipium
after the Social War. Colonial status is indeed more in accordance with the strategic
importance of the area, as well as with the amount of trouble taken by the Romans to
26
The few sepulchral inscriptions from Falerii Veteres in the Latin alphabet (LtF 140, 171-174)
may belong to such post-241 burials.
27
The restoration of colonial status was a piece of propaganda by which Gallienus, whose
maternal ancestors were Egnatii from Falerii, attempted to draw attention to his patrician fore-
bears (De Blois 1976:134). This second colonia Faliscorum is mentioned in CIL XI.3089-94.
43
CHAPTER 2
restructure its settlement and road-structure (§2.1.3). A complicating factor is that

Falerii Novi belonged to the tribus Horatia (cf. CIL XI.3100, 3112, 3123, 3125, 3136,
3930 (?), 7494, and the inscription quoted in note 268), not to one of the South Etrurian
tribus created in the fourth century for Veii, Capena, Sutrium, and Nepete (the Arnen-
sis, Sabatina, Stellatina, and Tromentina).28 The colonia mentioned by Pliny and the
Liber Coloniarum is explained by Di Stefano Manzella (1990:366-7) as being not a
colonia in the legal sense, but a land-allotment to veterans of the triumvirs at Falerii
Veteres, administered from Falerii Novi.
2.6.3. The ager Faliscus after 241. The loss of a very large number of men of
fighting age, the surrender of tÕ douleàon, the ceding of half the territory, the replace-
ment of its main site by a new settlement probably populated at least partly by
immigrants, and the subsequent construction of the Roman roads, must greatly have
changed life in the ager Faliscus. This change is very visible in the abrupt change in
the area’s settlement pattern described by Potter (1979:98-101). Falerii Veteres was at
best reduced to an insignificant township and replaced as the main centre by the new
Roman Falerii.
Of the other centres, Narce and Corchiano ceased to exist shortly after 241;
Vignanello, perhaps located in the part of the ager Faliscus that was not ceded to the
Romans, disappeared in the second century, while the site at Grotta Porciosa may
have survived as a Roman settlement. The main type of settlement now became
clusters of farmsteads, many of which new foundations, their Faliscan predecessors
having been abandoned in the mid-third century. All this appears to reflect a (perhaps
deliberate) attempt at fragmenting and ruralizing the area. The major routes from the
area, although following the lines of pre-existing ones, differ from these in the kind of
traffic they serve, opening up the area to long-distance traffic while at the same time
depriving it of its function as a crossroad (§2.1.3).
All this cannot have been accomplished without a major uprooting of Faliscan
society, whereby traditional loyalties were disrupted and pro-Roman families, and
even homines noui like Marcius Acarcelinius, the ‘fatherless’ son of a freedwoman
(LF 221-223) came to the forefront (cf. Terrenato 2007 for a discussion of such social
changes). Such an uprooting, both economical and social, cannot have been without
far-reaching consequences for the (linguistic) identity of the inhabitants of the area.
The new Falerii showed no signs of independence, and does not seem to have differed
from the average Roman provincial town. The sources only refer to it as the location
of one of the many omina portending Flaminius’ defeat in 216 (Liv. 22.1.11, Plut.
Fab. Max. 2, Oros. 4.15.1), as the birthplace of one of the ringleaders in the Baccha-
28
Di Stefano Manzella (1990:349) points to the fact that after the Social War the Horatia was
also the tribus of Spoletium, which was colonized in 241.
44
nalia-upheaval in 187-186 (Liv. 39.17.6), and as the birthplace of two brothers who
served under Varro in Spain (R. 3.16.10-11). Cicero still names Falerii as a possible
place for land grants (Leg. agr. 2.25.66), although the town is later described as
populous by Strabo (5.2.9). Its main economical resource now seems to have been the
rearing of sheep and cattle, the latter the source of the uentres Falisci (Var. L
5.22.111, Mart. 4.46.8, Stat. Silv. 4.9.35). Already Cato (Agr. 4.1, 14.1) had referred
to the praesepe Faliscum; Ovid mentions the Faliscan sheep (Epist. 4.8.41) and bulls
(Am. 3.13.14=Fast. 1.84=Epist. 4.4.32), whose famed whiteness was ascribed by
Pliny (NH 2.106.230) to the properties of the local water. Other sources mention the
growing of flax and the linen-industry (Grat. 40, Sil. 4.223). All in all, the picture
painted by the sources is that of a rather somnolent rural tranquillity.
2.7. Sociolinguistic factors influencing language preservation or loss
In §§2.3-4 I described a number of factors that may have played a role in Faliscan
ethnic identity. The feelings of identity of a group speaking a specific language or
dialect can be a powerful or even decisive influence in the preservation of that
language or dialect, but other sociolinguistic factors must be taken into account. The
difficulties in applying these to the situation in the ager Faliscus are great, and the
influence of each individual factor can only be estimated in a very general way. I
present here several that can to some extent be assessed.
(a) Economic status. Although it is impossible to establish how wealthy the average
Faliscan was at any given period, something may be said about the area as a whole.
During much of the period of its independence, the ager Faliscus may well have been
considered wealthy (“ciuitas Italiae opulenta quondam fuit,” Eutrop. 2.28): the area
was well situated with regard to trade routes, possessed fertile land, especially for
pasture, and was traversed by transhumance routes (§2.1.1-3). In fact, I think it very
likely that the area’s economy was a major factor in keeping it independent, especially
during the earlier periods. Falerii started building monumental temples like the one at
Contrada Celle in the fifth century (Sant pp.110-3), fourth- and third-century exports
of Faliscan pottery material have been found throughout the adjacent areas, and there
may have been exports of a less durable nature such as cattle and linen (cf. §2.6.3).
After the war of 241, this must have changed, as the Romans now controlled the main
routes to and from the area and created new roads for long-distance traffic and trade
(§2.1.3). During the division of the area into a Roman and a Faliscan part, from
shortly after the war of 241 until the implementation of the lex Iulia at the end of the
Social War (90-89), trade with Roman or Latin citizens may have been restricted by
the ius commercii, depending on the status of both areas (§2.6.2-3).
45
CHAPTER 2
(b) Social status. This is a point on which the sources are silent, and perhaps their
silence is eloquent enough: other towns or peoples were not deemed worth the effort
of serious comment at a time when Roman historiography was still finding its own
ground. Indirect data may be glimpsed from Roman literature. Dench (2005:300-1)
points to the caricatures and stereotypes of Italic and Latin rustics that may be found
in Roman literature from the Roman comedy of the second century onward: the
emerging importance of the City was apparently strengthened in its identity by
facetious references to those dwelling outside it or not conversant with its mores.
However well the Faliscans may have thought of themselves, and whatever status they
may have had among the city-states of Etruria (cf. e.g. their repeated appeals to the
councils of the Etruscan councils at Fanum Voltumnae mentioned in §2.5.1), there are
no indications that they were held in any kind of regard at Rome.
(c) Language status. What status the language or dialect has can be approached from
two directions: the status it had within the group itself and the status it had among the
‘outsiders’. Unfortunately, there are simply no sources on the status of Faliscan either
among the Faliscans or among the outsiders: the latter are silent to the extent that
there are not even contemporary Roman sources to remark on the fact that the
Faliscans spoke a different language or dialect in any way. 29
Dench’s (2005:298-361) excellent treatment of language as a factor in Roman
ethnic identity understandably starts from the first century BCE, when authors such as
Caesar and Cicero were establishing what was ‘good’ Latin and by and large decided
that this was the Latin of the Roman upper class. In this period, the mention of ‘rustic’
Latin begins to make its appearance, rustici and antiqui often being treated side by
side by authors on language: well might the rustic still speak as of old, but in the City,
men moved with the times. Earlier authors had made jokes and puns based on the
other Italic languages, e.g. Lucilius’ primum Pacilius thesorophylax pater abzet 581M
(where abzet ‘has died’ = Paelignian afđed Pg 9) but there are no more examples of
this from the first century BCE onward: Roman Latin now became de rigueur, as
appears from the remarks on the apparently well-remembered rural speech of
Augustus (Suet. Aug. 87-8).
(d) Functional distribution of the language involved. A language or dialect can stay
alive much longer if it has specific functions within the community that speaks it,
especially if these functions are connected to other possible rallying points for ethnic
identification such as religion (cf. e.g. the survival of Hebrew in the Diaspora).
Unfortunately, I can find no trace of an institution where Faliscan played a part, apart,
perhaps, from ritual: from the customs in Rome (especially the instauratio) and the
way in which the Umbrian rituals of the Tabulae Iguuinae are presented, it could be
29
The earliest author who remarks upon the Latin of Falerii is Varro (L 5.162, see §6.6.1).
46
expected that the precise wording of rituals was important in the ager Faliscus as well,
and its autochthonous cults may have retained prayers recited in Faliscan while the
written and spoken language of the area became successively more like Roman Latin.
Although there is no evidence for this, it is worth noting that the last ‘Faliscan’
inscription, or, perhaps rather ‘inscription trying to appear Faliscan’ (LF/Lat 214) is a
public dedication and may perhaps reflect an older wording of the same text. Other
institutions appear to be lacking: there are no indications that Faliscan played any role
in the carmina Fescennina, for instance.
(e) Population size. The number of speakers of Faliscan is the one factor where some
quantification can be done, if only to illustrate the difficulties and the overly large
number of a priori assumptions involved. The population of the ager Faliscus can
never have been very great, and may be estimated to some extent from calculations
based on land-use such as the ones applied by Ámpolo (1980) to early Latium. The
overall size of the area, depending on where the borders were situated, was no more
than 400-500 km² at the very most. Even today a sizeable part of the terrain (e.g. the
river-gorges) is unusable for agriculture: in antiquity, this part must have been larger,
given the amount of woodland on the ranges to the north and west (§2.1.1). Part of the
usable land was used for the rearing of cattle and sheep (cf., for a later period, Ov.
Epist. 4.8.41, Am. 3.13.14=Fast. 1.84=Epist. 4.4.32, Plin. NH 2.106.230) rather than
for food-crops: in all, the arable land cannot have exceeded 30,000 ha, which would
have fed 45,000 people at the very best. For the inhabitants to be considered wealthy,
however, the true number of inhabitants must have been much lower, probably in the
order of 20,000-22,500, at any given period, if not lower than that.30
A very much higher number, however, is implied by the size of the Faliscan
army in the war of 241, where the sources (Eutrop. 2.28, Oros. 4.11.10) give the
number of Faliscan casualties as 15,000. This number may of course be overstated,
but in view of the quite extraordinary double triumph awarded for this victory, it
cannot have been much lower than 10,000, unless there had been a gross relaxation of
the legal requirement of 5,000 enemy casualties for a triumph (V. Max. 2.8.1). Even if
the number of casualties was c.10,000, the army itself must have been 15,000-20,000
strong or more, and as it would have included most freeborn men in the age range of
18-45, this would point to a very much higher number of inhabitants than the one
arrived at above, perhaps even 75,000-80,000. I find this very hard to accept: either
the Faliscan army had been bolstered up with troops from other areas, or both consuls
were given a triumph on just one total of 5,000 enemy casualties.
30
Ámpolo (1980:28-9) reached much lower estimates for Latin towns of comparable size:
(1) Rome (seventh and sixth centuries): area 435 km², est. 18,000 inhabitants; (2) Tibur: area
351 km², est. 15,000 inhabitants; (3) Praeneste: area 262.5 km², est. 10,000-12,000 inhabi-
tants.
47
CHAPTER 2
A related point is the number of non-Faliscans that moved into the area after the
war and the division of the territory. Falerii Novi may have housed a few thousand
inhabitants, perhaps,31 but it can only be guessed at how many people were given
allotments in the rural area that was now Roman. Apparently there was still land to
spare at the time of Cicero (Leg. agr. 2.25.66), and Di Stefano Manzella (1990:366-7)
assumed that it was used for a land-grant to veterans of the triumvirs: this, however,
was several (civil) wars after the time Faliscan had ceased to exist.
(f) Distribution of the speakers. As said above (§2.1.1-2), the ager Faliscus was a
well-defined area with clear natural boundaries, with a major sites at Falerii and
originally at Narce as well, and a number of smaller towns e.g. in the north at
Corchiano, Vignanello, Grotta Porciosa, and in the south-east, e.g. at Rignano
Flaminio. This changes abruptly after the war of 241: Falerii Veteres is reduced to
insignificance and at worst depopulated, several of the smaller sites disappear, and the
area becomes ‘ruralized’, perhaps on purpose: Roman Falerii Novi now emerges as
the one dominant centre, and it may be presumed that this town had a largely or
predominantly Latin-speaking population (§2.6.2). Half of the land becomes Roman
property, with speakers of Latin settling among the speakers of Faliscan: the Roman
area may even have cut the Faliscan area in two. This change in habitation pattern
must have had severe consequences for the distribution of the speakers of Faliscan.
(g) Family and intermarriage. If what we know about the role of the family in Rome
is applicable to the ager Faliscus, the family or clan must have been one of the most
important social networks.32 In the ager Faliscus the coherence of the family structure
is in a sense underlined even further by the use of the family tombs (§2.3.1). Although
it is hard to tell if family traditions included keeping to the old ways in matters of
language too, there are signs of family mannerisms like the spelling p in the name
Umpricius in a Latin inscription from Falerii Novi (Lat 219). Related to the subject of
family is that of intermarriage. The heterogeneous origins of the gentilicia found in
the Faliscan texts show that Faliscan-speaking families probably intermarried with
Etruscan-, Sabellic- or Latin-speaking ones, but these data cannot be taken at face
value, let alone quantified (cf. §7.10.3). During the period when the ager Faliscus was
divided into a Roman and a Faliscan part, from shortly after the war of 241 until the
implementation of the lex Iulia at the end of the Social War, marriage with Roman or
Latin citizens may have been restricted by the ius conubii.
31
Cristofani’s (1984:30-1) estimates of the populations of a number of Etruscan cities are
based on the size of these towns’ urban area: according to these calculations, the inhabitants
of Falerii Novi (c.38 ha) may have numbered slightly over 6,000.
32
Terrenato (2007:13) in fact claims that “clan mentality in many cases came before civic
loyalty and ethnic identity.” This may well be correct, but in the absence of data on the
subject from the ager Faliscus I cannot pursue this further.
48
2.8. Linguistic evaluation: Faliscan and its neighbours
2.8.1. Faliscan and Etruscan. As said in §2.4.2, I regard Faliscan and Capenate as
the remains of a Latin once spoken throughout South Etruria, but which was replaced
by Etruscan, except in the ager Faliscus and Capenas. It is surprising that a relatively
small area like the ager Faliscus was able to preserve its linguistic independence, and
as this autonomy was clearly not based on great military might or superior resources,
it can only be explained by geographical position. Until the opening of the ‘Sutri Gap’
in the late fourth century, the silua Cimina must have formed a considerable barrier
and may have significantly slowed the eastward spread of Etruscan influence from the
great Etruscan centres of coastal South Etruria to the area. 33 From the south, the ager
Faliscus was easily accessible, but here there were no major Etruscan centres until the
emergence of Veii. Yet the physical barrier of the silua Cimina cannot by itself have
proved insurmountable, and can in any case have had no real influence on the
preservation of Latin in the ager Capenas. At least as important, therefore, must have
been the strategic position with regard to the trade routes along the Tiber and to the
interior. The economic importance of these routes may have been responsible not only
for the area’s independence, but may also have been a factor in the eventual ascen-
dancy of the centrally placed Falerii.
In view of the rise and expansion of Veii in the sixth and fifth centuries, it
remains to be seen whether the ager Faliscus could eventually have preserved its
independent identity. Veii, located on the doorstep of the ager Faliscus and Capenas,
became a dominant neighbour, not only economically and culturally, but also
politically, and by the late fifth century had swallowed Capena and was encroaching
on the ager Faliscus, founding or taking over Sutrium and Nepete. Yet even the
cultural or political usurpation by Veii of the major southern site of the area, at Narce,
did not lead to a Faliscan submission to Veii. Cultural identity may have played a part
here, for the epigraphic material shows that Narce was Etruscan and Falerii was
Faliscan at least as far as the language was concerned. I find it hard to imagine that
the rise of Veii, partly at the cost of the ager Faliscus, would not have been accompa-
nied by some form of friction or conflict in the sixth and fifth centuries, a conflict that
took place beyond the reach or the interest of the Roman and Greek historians. These
depict Falerii as the Romans encountered it, an ally of Veii in the wars of the late fifth
and early fourth centuries. At that time both towns were forced to cooperate against
Rome to protect their areas and trade-routes through the Tiber valley, and self-interest
may have overcome any lingering disagreements.
33
By then, Etruscan influence had long since reached the Tiber valley to the north of the ager
Faliscus. It is an interesting fact that Sutrium, although located on the ‘Sutri Gap’, appears to
have been founded and controlled from the south rather than from the west.
49
CHAPTER 2
Veii was then removed from the board by Rome, with the effect of making the
ager Faliscus into something of an ‘independent agent’ within South Etruria. For some
reason the Romans chose to preserve the political independence of the ager Faliscus
instead of settling it like the lands of Veii and its dependencies, and they continued to
do so in spite of frequent Faliscan revolts, perhaps because subduing the area would
have cost more trouble than occasionally defeating its probably not overly large
military might. The ager Faliscus thus found itself in a more or less precariously
independent position on the border between the Roman and the Etruscan spheres.
Although politically its alliance with the Etruscan states continued, economically the
area probably still depended largely on the now Roman south; culturally, it had been
Etruscan to a large extent, but linguistically, the ager Faliscus had been brought back
into the orbit of the Latin-speaking community.
2.8.2. Faliscan and the Sabellic languages. There most certainly were contacts with
the Sabellic area: the Tiber crossings near Lucus Feroniae at the south end of the ager
Capenas and near the Grotta Porciosa site at the north-eastern end of the ager Faliscus,
which connected South Etruria with the Sabine- and Umbrian-speaking interior must
have been important to traffic. Although reference to incursions has often been made,
these appear to have been on a fairly small scale and appear to concern the ager
Capenas more than the ager Faliscus. Most of the contacts were probably in the form
of trade and transhumance farming. Although Sabellic epigraphic and linguistic
features occur in several inscriptions (mostly from the ager Capenas, but MLF/Cap
474* is purportedly from Falerii Novi), these are probably phenomena of interference
rather than of borrowing: extensive linguistic influence from Sabellic languages is
certainly absent, however intensive the contacts between the speakers of Faliscan and
Sabellic languages may have been. For a further discussion, see §9.3.
2.8.3. Faliscan and Latin. The inhabitants of the ager Faliscus must have been in
fairly frequent contact with Latium from the early days on, as is shown not only by
the cultural peculiarities that the areas shared, but also by what is known of the trade
routes along the Tiber and the areas they served. How these contacts were affected by
the emergence of Narce and Veii is difficult to assess, especially where the linguistic
side is concerned. It is noteworthy that a major morphological change like the
replacement of the second-declension genitive singular ending -osio by -i, which must
have taken place during the floruit of Veii, affected both Latium and the ager Faliscus
(cf. §4.4).
Then, in the early fourth century, Veii was destroyed, and its dependencies
Capena, Sutrium, and Nepete passed into Roman hands, the latter two now linked to
Rome by the first stage of the Via Amerina. Although this must again have brought
the inhabitants of the ager Faliscus into more frequent contact with speakers of Latin
50
as spoken in Latium (cf. §9.4.1), it did not result in a spread of this variant of Latin to
the ager Faliscus, which together with its lively political autonomy preserved its
cultural and linguistic independence while the ager Capenas became Latinized
(§9.4.3). Yet in spite of the ager Faliscus adhering to its own form of Latin, important
morphological changes that took place during this period again seem to have affected
both Latium and the ager Faliscus, namely the replacement of the first-declension
genitive singular ending -as by -ai (§4.2.2) and the replacement of the third singular
perfect ending -ed by -et (§5.2.2e).
The outcome of the war of 241 must have changed this situation drastically. The
historical and archaeological evidence points to a decline in population, and perhaps
to a dispersion of the remaining speakers of Faliscan over farmsteads and townships
that were insignificant in comparison to the new Roman centre of the area. Immi-
grants from Rome and Latium, and probably also from the Roman colonies at Capena,
Sutrium, and Nepete, came to this new Falerii and to other parts of the area, taking out
allotments and settling among the native population. New roads opened up what had
been a fairly closed-off territory to traffic on a larger scale and over much greater
distances than before. The ager Faliscus, which until the end of the First Punic War
had been an autonomous force of local significance within the network of the city-
states of South Etruria, now became an insignificant part of a rapidly expanding world
of which Rome was at the centre.
All these factors would have been conducive to a speedy adaptation of Faliscan
to mainstream Latin, while there seem to have been few factors that would have been
an inspiration to the preservation of Faliscan. If Falerii Novi was a colony, contacts
with the inhabitants of the remainder of the area may to some extent have been
restricted by the regulations of the ius commercii and the ius conubii, but these can
hardly have exercised any crucial influence. Traditions of family may have played
some role in the preservation of Faliscan, but these must have been offset by the
social uprooting following the war. Traditions of religion, too, may have played a
role, but even though the cult of Juno may have been important to the Faliscan ethnic
identity, religion was apparently not powerful enough to preserve its dialect.
With the loss of its political independence, the area lost an important possible
rallying-point for the preservation of its cultural or linguistic identity. The Roman
roads, which bypassed the native centres, took away its strategic position and
commercial role, and even the name of the town after which the whole area was called
had been usurped by the new Roman centre. Under these circumstances, the local
dialect, too, was bound to disappear, especially as it does not appear to have had any
specific status or function that could have ensured its survival: Faliscan does not
appear to have been associated with the carmina Fescennina, or to have played a role
in the cults that continued at Falerii Veteres, or to have been used for long in the
administration of the new town (§9.4.1).
51
CHAPTER 2
Faliscan seems to have disappeared from the written record during the first half
of the second century, both in the Roman-controlled and in the ‘independent’ part,
fated to dwindle into an obscurity of twenty centuries. Whatever non-standard
linguistic features the area preserved after that date were now just part of a local
variant of what now became ‘rustic Latin’, ‘Falerian’ rather than ‘Faliscan’ – sic rure
loquuntur.
52
Chapter 3
Phonology
Faliscan phonology presents some very interesting features. First of all, the Faliscan material is
relatively old in comparison to the inscriptions of the Latin and most of the Sabellic languages:
most of it dates from before the middle of the third century. Second, the Faliscan material shows
some very interesting developments, especially in the voiced aspirates (§3.2.8, §3.3.3, §3.5.2)
and the diphthongs (§3.7).
This chapter opens with a few methodological considerations on the limitations of the
Faliscan material and the status of Proto-Italic (§3.1). The arrangement of the remainder is
more or less chronological. It starts with the developments of the Proto-Italic period, which
Faliscan shares with the other Latin dialects and the Sabellic languages (§3.2); continues with
the Proto-Latin developments, where Faliscan should show the same developments as Latin, but
different ones from the Sabellic languages (§3.3), and ends with the developments of the Early,
Middle and Late Faliscan periods (§3.4-7). A short conclusion is drawn in §3.8.
3.1. Methodological issues
3.1.1. Method and material. The phonology of Faliscan stands, in a sense, on a dif-
ferent level from the parts of speech that I discuss in chapters 4-7.
On the one hand, phonological developments can often be observed and defined
much more clearly than e.g. the developments in morphology, where analogy may
play a much larger role: phonemes, after all, although by definition meaningful
sounds, do not themselves convey a ‘meaningful concept’ in the sense morphemes
and lexemes do. For the same reason, the influence from other languages as a factor in
a phonological development is less and different than, e.g., in the lexicon and espe-
cially the onomasticon: see, however, §3.8.
A problem, on the other hand, is that in the case of fragmentarily preserved lan-
guages there is often too little material for a comprehensive interpretation, especially
in the case of diachronic developments. The Faliscan material can therefore often only
be interpreted clearly against the background of a picture of the larger developments
as they are known from Latin and the Sabellic languages. Since this study starts from
the assumption that Faliscan is a Latin dialect, I have chosen in several cases to pre-
sent the Faliscan material against the larger background of the developments as they
are known to have taken place in Latin. The aim is then to show that the few Faliscan
data tie in with what is known about the much better documented Latin, and are no-
where at odds with it. Wherever this is possible, the relevant data from the Sabellic
languages are then reviewed for comparison: these usually show a different picture.
53
CHAPTER 3
Another problem which applies especially to the Middle and Late Faliscan de-
velopments is the question of which inscriptions represent what can be called Faliscan
and which do not, e.g., because they represent the Latin of immigrants from Rome or
Latium. I have excluded on principle all inscriptions that I regard as Latin (Lat 217,
218, 219, 237 and 238, 240, 250, 251, 268, 291, 296, 377, 393, 456 and the Late Falis-
can or Latin 214), as well as those that I regard as Latino-Faliscan or Capenate, as
sources for data on Faliscan phonology, and used these only as additional material or
to highlight specific points, after the main developments of Faliscan have been made
clear. The risk of a circular argument is great, however, for it is easy to exclude spe-
cific inscriptions as being Latin rather than Faliscan, and thus ending up using only
those inscriptions that exactly confirm the preconceived idea about Faliscan that lead
to the exclusion of the ‘non-Faliscan’ inscriptions in the first place: see §3.6.6.1 for an
illustration of this point.
3.1.2. The concept of Proto-Italic. As has been said in §1.4.1, ‘Proto-Italic’ is de-
fined here as a chronological stage during which a set of phonological and morpho-
logical developments took place that together set off the Italic languages from the
other IE families, all of which developments are assumed to have preceded any de-
velopments that were limited to either the Latin or the Sabellic branch of the Italic
languages. As soon as either branch of the Italic languages shows independent devel-
opments, the Proto-Italic period must be regarded as closed and the Proto-Latin and
Proto-Sabellic stage must be assumed to have begun. The initial stage of Proto-Italic
must therefore coincide with (late) PIE, whereas the final stage of Proto-Italic consti-
tutes the initial stage of both Proto-Latin and of Proto-Sabellic.
Since Proto-Italic is defined as a chronological stage in the development of the
Italic languages, it must have had an existence in time and in place. However, as the
Proto-Italic period is a linguistically defined period but is also prehistoric, ascribing
dates to such a remote period is extremely difficult, if not impossible. Meiser’s
(1998:54) approximative dates of c.4000 to c.1800 BCE (apparently defined as start-
ing with the breaking up of (Western) PIE and ending approximately 1100 years be-
fore the appearance of the first written documents) is as good a guess as anyone’s.
To the Proto-Italic stage can (or rather, must) be ascribed all developments that
occur in both the Latin and the Sabellic branches of the Italic language family, unless
these must be ascribed to separate parallel development, either because they can be
dated absolutely (by written evidence) or because they can be dated later relatively as
having taken place after a development that is demonstrably post-Proto-Italic. The
explanation of any feature that is found in only one Italic language or in only one
branch of the Italic languages by a Proto-Italic development therefore implies that
either the explanation is unsound or that the language in question cannot be regarded
as Italic.
54
PHONOLOGY
3.1.3. The PIE phonemic inventory. I end this section with a few short notes on the
phonemic inventory of PIE and Proto-Italic as presented in fig.3.2. Firstly, it will be
clear that I approach PIE and the developments of the IE languages from a laryngeal-
ist perspective. As a consequence, I have been reluctant to include */a ā/ (cf. Lubotsky
1989), and have not included */ī ū/ (cf. Beekes 1990:173-5). It goes beyond the scope
of this study to review the whole laryngeal discussion: as far as the question of
whether Faliscan is a Latin dialect or an independent Italic language is concerned, the
laryngeals are of relatively minor importance, since they are assumed to have disap-
peared during the Proto-Italic period. Secondly, whether Proto-Italic had an inherited
PIE phoneme */θ/ or the sound *[θ] was only a positional variant of */t dh/ (cf. Meiser
1998:29) is in my view unclear, but for the scope of this study, this is irrelevant, since
it disappeared during the Proto-Italic period, and no reflexes of words where it pre-
sumably occurred are found in the Faliscan material.
The PIE occlusive series can and perhaps should be reconstructed in an entirely
different way than is done here, namely according to the glottalic theory. I mention
this especially in view of Baldi & Johnson-Staver’s (1989) glottalic reconstruction of
the development of the voiced aspirates in the Italic languages:
PIE Proto-Latin Latin

h h h
p t k p t k p t k
I
p t k
#bh #dh #gh ph th kh θ X f f h

II
-b- -d- -g- b d g b d g(u)
III p’ t’ k’ (?)
Fig.3.1. Glottalic reconstruction of the development of the voiced aspirates.

(After Baldi & Johnson-Staver 1989:96.)
I have not adopted this perspective, not because I think the glottalic reconstruction is
wrong in itself, but because I wonder whether the distinction between a voice-
less/voiceless aspirated, voiced/voiced aspirated, and glottalized series was indeed
preserved until the Proto-Italic, let alone the Proto-Latin period, and had not devel-
oped and/or merged into different series before that date, within PIE itself. Stuart-
Smith (2004:17-8) in fact notes that in the course of the development it would in any
case be necessary to assume a stage where the voiced occlusives would have become
voiced aspirated occlusives.
55
CHAPTER 3
vowels /a? e o/ /ā? ē ō/
semivowels laryngeals resonants
vocalic /i u/ /ħ1 ħ2 ħ3/ / / / /

positional
variation
consonantal /i u/ /ħ1 ħ2 ħ3/ /m n/ /l r/
sibilants /s/
labial dental (palatovelar labio-
velar?) velar
voiceless /p/ /t/ (/k/?) /k/ /k /

occlusives
voiced /b/ /d/ (/g/?) /g/ /g /

voiced aspirated /bh/ /dh/ (/gh/?) /gh/ /g h/
Reconstructed phonemic inventory of PIE (final stage) =

Reconstructed phonemic inventory of Proto-Italic, initial stage
↓
vowels /a/ /e/ /i/ /o/ /u/ /ā/ /ē/ /ī/ /ō/ /ū/
‘semivowels’ resonants
vocalic / // /
positional
variation
consonantal // / / /m/ /n/ /l/ /r/
sibilant /s/
labial dental velar labio-
velar
spirants voiceless / / (/θ/?) /x/ (/x /?)

voiced /β/ /ð/ /γ/ /γ /
occlusives voiceless /p/ /t/ /k/ /k /

voiced /b/ /d/ /g/ /g /
Reconstructed phonemic inventory of Proto-Italic, final stage =

Reconstructed phonemic inventory of Proto-Latin and Proto-Sabellic, initial stage
Fig.3.2. Development of the phonemic inventory of Proto-Italic.
56
PHONOLOGY
3.2. Proto-Italic developments
3.2.1. From Proto-Italic to Proto-Latin. The phonological developments that can

with reasonable certainty be ascribed to the Proto-Italic period have been summarized
by Meiser (1986:37-8 from the perspective of the history of the Sabellic languages,
1998:54 from the perspective of the history of Latin). Of these, I discuss only those
that are in some way relevant to the evaluation of the Faliscan material. 34 Since there
is no doubt among scholars that Faliscan is an Italic language, Faliscan should be ex-
pected to show the same Proto-Italic developments and the same outcomes of these
developments as Latin and the Sabellic languages. The Proto-Italic developments are
therefore of relatively minor relevance to the question of whether Faliscan is a Latin
dialect or a separate Italic language.
3.2.2. Presumed merger of the palatal and the velar series. It is a debated and de-
batable point if, beside a labiovelar series /k g g h/, PIE had both a palatal or pala-
tovelar series */k g gh/ and a ‘true velar’ series */k g gh/.35 Apart from the fact that
such a triple-series system is debatable from a typological point of view, it is pre-
served in none of the IE languages: the distribution of palatals and velars throughout
the IE languages appears to be complementary, the centum-languages (among which
the Italic languages) having velars and labiovelars, the satem-languages, palatovelars
and labiovelars. For discussions of this point, see Steensland 1973 (extensive) and
Baldi & Johnston-Staver 1989:88-97 (from a glottalist perspective).
The rather slight Latin evidence for a Proto-Italic distinction between a palatovelar
and a velar series is evaluated by Schrijver (1991:425-34). It consists of the fact that
after a reconstructed velar */k/, but not after a reconstructed palatovelar */k/, PIE */e/
usually appears in Latin as /a/:
PIE */ke/ → Latin /ke/ PIE */ke/ → Latin /ka/
The instances where Schrijver (1991:434) deems this probable are calidus ← PIE
*/kel-/, candeo ← PIE */(s)kend-/, carpo ← PIE */(s)kerp-/, carro ← PIE */kers-/,
scabo ← PIE */skebh-/, and scando ← PIE */skend/, to which Rix (1996:160 n.9)
adds castrum ← PIE */kegh-/. The counterexamples are cena ← PIE */kert-/, and sce-
lus ← PIE */skel-/, to which Meiser (1998:83) adds celsus ← PIE */kel-/, cingo ←
34
Those discussed by Meiser, but not by me, are: (1) */ū/ → /ī/ (e.g., in pius), (2) */g/ → //
(e.g., in maius), (3) */m/ → /n/ (e.g., in uenio), and (4) the developments of */θ/, namely
*/kθ/ → */ks/ (→ /s/) and */#kθ/ → /#s/. For the development of the accent, see §3.6.6.
35
Meillet (1894:294-9) suggested that the velars (in effect, /k/) developed from a positional
variant of the palatovelars that subsequently acquired phonemic status: PIE */#sk/ → */#sk/
(de-palatalization) → */#k/ (loss of movable /#s/).
57
CHAPTER 3
PIE */keng-/, and, with velar */g/ and */gh/, gemo ← PIE */gem-/ and hedera ← PIE
*/ghed-/. The Italic evidence that can be added to this consists of the corresponding
Sabellic forms, Oscan castrous TB 13 etc., Umbrian kastruvuf TI Va.13 etc., Oscan
kersnu Cm 14 etc., Umbrian śesna TI Vb.9 etc., and Umbrian śihitu anśihitu TI
VIb.59. If these differences can indeed be attributed to an opposition of a palatovelar
and a velar series, the development */ke/ → */ka/ must have taken place before a
merger of both series in Proto-Italic, and such a merger must have taken place at a
very early Proto-Italic date, since a palatovelar-velar opposition is not reflected in any
other Italic development.
Although therefore apparently irrelevant to the study of Faliscan, this point is in fact
relevant to the discussion on the development of the voiced aspirates. If (Western)
PIE did not have a separate palatovelar series, or if the PIE palatovelar series disap-
peared due to a PIE or early Proto-Italic merger of the palatovelar and the velar series,
the (presumed) PIE distinction between */gh/ and */gh/ is irrelevant to any further de-
velopments of the voiced aspirates. It is therefore impossible to ascribe the unex-
pected Faliscan reflexes fifiked EF 9, f[.f]ịqod EF 1, and lecet MF 88 (§3.3.3.2) to the
fact that these forms contained an original PIE */gh/ (PIE roots */dheigh-/ and */legh-/)
as opposed to an original PIE */gh/.
3.2.3. Developments of the laryngeals. A major development that can with certainty
be ascribed to the Proto-Italic period is the disappearance of the laryngeals. For the
Italic languages, the most detailed description of the way these phonemes disappeared
and the traces they left behind is Schrijver 1991. Brief summaries of the major trends
may be found in Meiser 1998:105-10, 1986:36. The list below is not a complete over-
view, but merely places the Faliscan lexemes and morphemes where a laryngeal may
be reconstructed beside the reconstructed Proto-Italic developments:
PIE Proto-Italic (final) Faliscan reflexes
*/#ħ1e/ → */#e/
*/ħ1eg-/ */eg-/ eqo EF 1, 467*, eco EF 3, eko EF
6, 7; eco LF 378, 383
*/#ħ2e/ → */#a/
*/ħ2eidh-/ */að-/ efiles MF 113, 115, efile MF 114,
efi[les] MF 116
*/#ħ3e/ → */#o/
*/ħ3ekt-/ */okt-/ ? oct- MLF 353 (very dubious)
*/#ĦRC/ → */#aRC/
*/ħ2rgntom/ */arg tom/ arcentelom EF 1
58
PHONOLOGY
*/#ĦRV/ → */#RV/
*/ħ1me/ */mē/ (cf. §4.7.2) med EF 1, 9; met MF 470*
*/ħ1leudhero-/ */lo ðero-/ l]oifirtạ MF 41, loferta LF 221;
loifiṛtato MF 31, loifirtato MF 32
*/h1reudho-/ */ro ðo-/ ? rufia EF 4 (see §3.7.2)
*/ħ3rēg-/ */rēg-/ rex MF 90, perhaps also ṛẹx MF
91 (very uncertain), rẹ[x] LF 249
[and rex LtF 231]
*/CĦC/ → */CaC/
*/dhħ1-k-/ */ ak-/ faced MF 471*, facet MF 470*
*/pħ2tēr/ */patēr/ pater MF 62
*/sħ2kro-/ */sakro-/ sacra MF 127
*/Ceħ2C/ → */CāC/
*/keħ2ro-/ */kāro-/ karai EF 1
? */meħ2no-/ */māno-/ cognomen man[o]mo MF 80,
[m]ano[m]o MF 149; gentilicium
mania LF 225, m{e}ania LF 224.
The word may have been a non-IE borrowing in PIE (cf. Schrijver 1991:143).
? */meħ2tēr/ */mātēr/ mate LF 222
The /ā/ is often regarded as due to a PIE lengthened vowel (PIE */mātēr/ rather than
*/meħ2tēr/), but can also be ascribed to a laryngeal (cf. Schrijver 1991:341).
*/Ceħ3C/ → */CōC/
*/deħ3u-/ or */doħ3u-/ */dō -/ douiad EF 1 (see §5.3.1.4)
*/CiĦC/ → */CīC/
*/uiħ1nom/ */ īnom/ uino MF 59-60
*/uiĦros/ */ īros/ → */ iros/ duum]|uiru LF 243, duum[uir LF
247, duu[muiru LF 248, [duu]ṃ-
uiru LF 249 (Latin borrowing?)
*/Cħ2eC/ → */CaC/
*/lħ2eiuo-/ */la o- names: leiuelio MF 79, [leu]elio
MF 90, [leu]elio MF 159; leuieis
Lat 251, leuia LtF 327, leueli MF
14, le[ueli]o MF 146; leue[lia]
MF 147
The PIE form may have been either */lħ2eiuo-/ or */leħ2iuo-/ (EDL s.v. laevus).
*/skħ2eiuo-/ */ska o-/ names: sceụa MLF 312, sceiuai
LF 379.
The PIE form may have been either */skħ2eiuo-/ or */skeħ2iuo-/ (Schrijver 1991:270).
59
CHAPTER 3
? */CĦiC/ → */CīC/
h
*/d ħ1ilios/ */ īlos/ fileo MF 467*, hileo MF 146, etc.
*/dhħ1ilieħ2/ */ īlā/ filea MF 14 etc.
Schrijver (1991:242) and EDL (s.v. filius) suggest either */dhħ1ilios/ or */dheħ1ilios/ as
the PIE form.
*/CRĦC/ → */CRāC/
*/prħ3-mo-/ */prāmo-/ pramom EF 2 (a debated word,
but probably related to Latin
prāndium, §6.2.59)
*/CRĦV/ → */CaRV/
*/ghrħ2-u-/ */haru-/ harịṣp ̣[ex LtF 231, harisp[ex LtF
232
*/CĦRC/ → */CaRC/
? */bhĦrs/ */ ars/ far EF 1
The /a/ may be of non-IE origin and not due to a laryngeal (Schrijver 1991:113-4).
Unclear are */sok (ħ2)io-/ → soc ̣[iai] EF 1, sociai EF 4, where the presence of the
laryngeal is doubtful (cf. Schrijver 1991:249, De Vaan EDL s.v. socius), and
praụ[i]os EF 1 (if connected to Latin prāuus): EDL (s.v. prāvus) gives no etymology
for this word, but it seems likely that the /ā/ is due to a laryngeal: perhaps */preħ2-uo-/
→ */prā o-/?
The verbal roots reconstructed as ending in a laryngeal have not all been included in
the list above, since their reflexes are sometimes different due to analogies or restruc-
turations within the paradigms of the verb. They are: (1) */dħ3-/ in porded EF 1, see
§5.3.1.14 (and (2) in the extended form of this root */deħ3 -/ or */doħ3 -/ in douiad
EF 1, see §5.3.1.4); (3) */kubħ2-/ (or */kubħ2-/?) in cupat MF 40 etc. and cupa]nt MF
80 etc., see §5.3.1.2-3; (4) */pi-pħ3-/ in pipafo MF 59, ‹pi›pafo MF 60, see §5.3.1.13;
(5) */prħ3-/ in pe para[i EF 1, see §5.3.1.12.
Laryngeals in verbal suffixes and endings that are attested for Faliscan are re-
constructed for: (1) seite EF 4 (either s{e}ite or s‹ie›te), reflecting the old PIE optative
*/ħ1s-iħ1-té/, used as subjunctive (cf. §5.3.1.18); (2) the primary ending of the first per-
son singular PIE */-oĦ/ → Proto-Italic */-ō/, attested for Faliscan in carefo MF 59,
carẹ[f]o MF 60, pipafo MF 59, ‹pi›pafo MF 60 (see §5.2.4a); (3) the first person sin-
gular perfect ending PIE */-ħ2e/, which developed either into Proto-Italic */-a/ →
*/-a/ (innovation) or into Proto-Italic */-ħ2e-/ (innovation) → */-a/ (see §5.3.1e):
note that the stage -ai is reflected only by Faliscan pe para[i EF 1 (cf. Untermann
1968a:165-9).
60
PHONOLOGY
Laryngeals in nominal endings that are attested for Faliscan are reconstructed
for the endings of the first declension: nominative singular */-eħ2/ → /-ā/ (or */-ħ2/ →
*/-a/, cf. §4.2.1), genitive singular */-eħ2s/ or */-eħ2es/ → */-ās/ (§4.2.2), dative singu-
lar */-eħ2ei/ or */-ħ2ei/ → /-ā/ (§4.2.3), accusative singular */-eħ2m/ → /-ām/ (§4.2.2).
3.2.4. Voicing of /t#/ and subsequent drop of /i#/. The primary and secondary end-
ings of the third person (see §5.2.4.1-2) show that these must have been subject to two
successive developments, namely (1) voicing of */t#/, and (2) loss of */i#/ (cf. Meiser
1998:98, 73-4).
voicing of /t#/ loss of /i#/
sec. 3rd sg. */-t/ → */-d/ → */-d/
sec. 3rd pl. */-nt/ → */-nd/ → */-nd/
prim. 3rd sg. */-ti/ → */-ti/ → */-t/
prim. 3rd pl. */-nti → */-nti/ → */-nt/
These endings are reflected in Early Faliscan fifiked EF 9, f[.f]iqod = fifiqo(n)d EF 1
vs. Middle Faliscan cupat MF 40 etc., cupa]nt MF 80 etc. (For further attestations, see
§5.2.41-b,e.)
This development must belong to a period where the PIE accent had not yet
been replaced by the Italic initial accent (§3.6.6), as */i#/ was preserved where it car-
ried the accent in PIE or where it occurred in alternation with an accented */i#/ within
the same paradigm (Rix 1996:158 n.7, Meiser 1998:74). It may therefore belong to an
early phase of Proto-Italic.
3.2.5. Merger of */eu/ with */ou/. According to the generally accepted view, the in-
herited diphthong */e / merged with */o / already during the Proto-Italic period (cf.
Pfister 1977:55, 69-70, Meiser 1998:59; Von Planta 1892:157-8). There are, however,
several instances of eu in Latin and Faliscan that have led a number of authors (e.g.
Pisani 1943:259, 1964:346, LHS pp.70-1, Wachter 1987:99, 374) to doubt this view
and assume that */e / was preserved until the third century BCE. The instances are as
follows (mainly from Blümel 1972:29-30):
Latin:
x neuen  deiuo CIL I2.455 (Ardea, third century: sometimes, but in my view errone-
ously, regarded as Faliscan, as is explained in §18.3.2);
x neuna dono CIL I2.2845 (Lavinium, early third century);
x neuna fata CIL I2.2846 (Lavinium, early third century);
x [---]euam (beside iouxmen|ta) CIL I2.1 (Rome, sixth century?);
x cozeulodorieso (also given as cozeui oborieso) quoted by Varro (L 7.26) and
leucesiae (also given as leucesie) quoted by Terentius Scaurus (CGL 7.28.11),
both purportedly from the Carmen Saliare.
61
CHAPTER 3
Faliscan:
x euios EF 1: There is general agreement among the editors that euios is onomastic:
it is therefore quite possible that euios is from another language altogether (cf.
§7.1.1), as G. Giacomelli’s (1963:41-2) interpretation of this form as EÜioj shows,
even if this interpretation may no longer be tenable (see §12.2).
x euotenosio EF 3 in ecoquto*euotenosio. In my view, this is to be read as eco quto
*e (?) uotenosio, where uotenosio is the genitive of the name that appears in the
vocative uoltene at the end of the text, and therefore to be read as uo(l)tenosio or
uo‹l›tenosio: see §12.3.
x Safarewicz (1954:101-3) also pointed to leueli MF 14 and heva EF 9, but in the
former (as in le[ueli]o MF 146 and leue[lia] MF 147) eu represents /ę̄ / ← /a /,
cf. leiuelio MF 79 (see §3.7.6), while the latter is now read as tele*[1-2].
As far as I know, there have been no suggestions that there are similar problematic
cases of eu in the Sabellic languages: Umbrian sevakne TI IIa.21 etc. represents /sē -
akni/), Oscan helleviis Cp 36 etc. and serevkid Po 1, sereukid Lu 62A.3 represent
/hel° is/ and /ser° kid/, Volscian deue VM 2 and Umbrian deueia TI VIa.9, 10 repre-
sent a monopthongized reflex of */de -/, Samnite evklúí Sa 1A.3, 25, B.4 is a bor-
rowing from Greek Eâkloj, in Paelignian peumpuni Pg 26 the eu is a rather curious
spelling for /ŏ ̣/, and Praesamnitic eṿies Ps 5 is now usually read as ef ̣ies. The instances
of ev and eu in Oscan neṿpus Po 68 (sometimes regarded as Etruscan, and not in-
cluded in WOU), the name meuies Fr 16, and the abbreviation ev tPo 30, 32, 34, and
perhaps also the Umbrian toponym Meuania, remain unexplained.
The only clear cases of a problematic eu are therefore neuen and neuna, equated with
Latin nouem ← PIE */ħ1ne n/ and nōnus ← PIE */ħ1ne nos/ ever since Pisani
(1943:259) and Vetter (1953:332-3, 1956) interpreted neuen  deiuo CIL I2.455 as a
dedication to the di nouensides. (For other interpretations, see §18.3.2.) There appears
to be no way in which these forms can be explained as ‘lawful exceptions’ to the rule
*/e / → */o /, due e.g. to a specific phonological context that prevented the operation
of this merger.36 There are therefore four possible solutions: (1) PIE */e / survived
until the third century in (some dialects of) Latin (Pisani 1943:259; also Wachter
1987:99, 374); (2) eu is due to an analogical preservation (Safarewicz (1954:103),
who suggested that neuen may have been preserved after decem); (3) eu is to be as-
cribed to an origin different from Proto-Italic */e / (Lipp (in Meiser 1998:59), who
assumes that eu represents /ē /); (4) eu is due to a secondary development (Steinbauer
(in Meiser 1986:37), who ascribed the eu to a dissimilation of /o /).
36
The fact that in many of the forms quoted here eu may reflect /e. / rather than /e / (Blümel
1972:30, R. Giacomelli 1978:26-7) cannot constitute a counter-argument, as Proto-Italic */e /
merged with */o / also in antevocalic position, cf. e.g. nouus, and nouem and nōnus them-
selves.
62
PHONOLOGY
A different solution has been proposed by Schrijver (1991:452), who suggests

that */e / and */o / merged into a neutral diphthong that could be realized as [e ],
[ə ], or [o ], and rendered both as ou and as eu.37 I find this a rather sweeping as-
sumption about the vowel system of the Italic languages just to explain three com-
paratively late forms of one word, but two points might be considered as evidence in
its favour: (1) the earliest Etruscan rendering of this diphthong in onomastic borrow-
ings from the Italic languages was av (cf. lauv|cies Etr XXIX), pointing to a central
vowel as its first member: the spelling becomes uv only from c.500 onwards (Stein-
bauer 1999:37-9); (2) podloucei CIL I2.2833 (sixth century) and polouces CIL I2.548
(between 350 and 250?, cf. Wachter 1987:106-8), both from Greek PoludeÚkhj, may
show that Greek eu and Latin ou were still seen as equivalents: cf. the Greek tendency
to render Lucius as LeÚkioj, although this could also be a ‘learned reconstruction’.38
There is therefore in my view no compelling evidence that PIE */eu/ was preserved
beyond the Proto-Italic period, let alone until the third century. Neuen and neuna,
even taken together with the unclear [---]euam (beside iouxmen|ta), the quotations
from the Carmen Saliare, and the unexplained Oscan forms, constitute too unreliable
a basis for such an assumption in view of the overwhelming evidence in favour of a
Proto-Italic merger */e / → */o /.
3.2.6. Drop of intervocalic // and subsequent contractions. The loss of intervocalic
// and the subsequent contractions of the surrounding vowels lie at the base of the
eventual form of the Italic conjugational system and must have been of Proto-Italic
date. The developments insofar as they have reflexes in the Faliscan material are:
1st conj. */-eħ2-e/o-/ → */-ā-e/o-/ → */-ā-e/o-/ → */-ā-/
nd e e e
2 conj. */-eħ1- /o-/ →*/-ē- /o-/ → */-ē- /o-/ → */-ē-/
Faliscan forms that reflect these developments are cupat MF 40 etc. and cupa]nt MF
80 etc. (first conjugation, but from */kubħ2-e-e/o-/ → */kubħ2-a-e/o-/, see §5.3.1.2-3),
and second-conjugation saluete EF 4, saluetod EF 3, salueto EF 4, teneṭ MLF 361 (if
indeed attested), and probably lecet MF 88 (see §5.3.1.11). (In carefo MF 59, carẹ[f]o
MF 60, /-ē-/ probably reflects */-eħ1-/ rather than */-eħ1-e/o-/, see §5.3.1.1.)
Loss of intervocalic // and subsequent contraction also affected the ending of
the third declension nominative plural */-ees/ → */-e.es/ → /-ēs/, attested for Faliscan
in efiles MF 113, 115, efil]es MF 117, efile MF 114, and perhaps salues EF 3.
37
Thurneysen (1887), whence also Von Planta (1892:115-6), assumed that in some contexts
*/o / developed to */a / at a Proto-Italic date, but this is rather an attempt to explain instances
of /a/ that are now mostly ascribed to a laryngeal origin. Cf. also Rix 1996:158.
38
Pfister (1977:69), pointing to polouces, assumed that the merger of */e / with /o / took
place “sehr frühe”, but so late that it could include early Greek borrowings containing /e /.
63
CHAPTER 3
3.2.7. The developments of */ /. The reflexes of the sonantic liquids */ / are the
same in Latin and in the Sabellic languages, and therefore the result of a Proto-Italic
development. (The sonantic nasals */ / developed slightly differently in both
branches, and their development must therefore be post-Proto-Italic: see §3.3.2.) The
sonantic liquids developed as follows:
/ / → /or/ (and /ur/) / __C / / → /ar/ / __V
/ / → /ol/ / __C / / → /al/ /__ V
(Cf. Pfister 1977:44, 46-7, Meiser 1998:63-4 for Latin, and Von Planta 1892:31-5,
Meiser 1986:37 for the Sabellic languages). Since these developments are Proto-Italic,
the same developments can be assumed for Faliscan. In the Faliscan forms where so-
nantic liquids can be reconstructed, however, these occur side by side with a laryn-
geal, which influenced the outcome: PIE */ħ2rgntom/ → */arg tom/, whence arcente-
lom EF 1 (cf. Latin argentum, Oscan aragetud Cm 7 etc.) and */prħ3-mo-/ →
*/prāmo-/ pramom etc. EF 2 (cf. Latin prandium).
3.2.8. Developments of the voiced aspirates (I): Proto-Italic. The developments of

the PIE voiced aspirates, */bh dh gh g h/,39 are probably the most debated and complex
subject of Italic phonology. The discussion concerns not so much the general trends
and outcomes of these developments, as these are sufficiently clear, but the way in
which these developments took place. This discussion is of special relevance for this
study, as Faliscan seems to show some unexpected reflexes: an important argument
for those who assign to Faliscan a more independent status, either within the Italic
languages or within the Latin branch. Recent overviews of the developments can be
found in the following publications: for the Italic languages in general, Stuart-Smith
2004; for Latin, Pfister 1977:145-50, Meiser 1998:101-5, and Eichner 1992:73-6; for
the Sabellic languages, Meiser 1986:73-8; for Faliscan, Wallace & Joseph 1993. Note
that in this section I discuss only the Proto-Italic stages of the development, or rather,
I try to establish which stages of the development belonged to the Proto-Italic period,
as this is relevant to the question at what stage Faliscan started to develop independ-
ently. For the Proto-Latin and Faliscan developments, see §3.3.3 and §3.5.2.
Broadly speaking, the developments of */bh dh gh g h/ are as follows:
(1) After a nasal, */bh dh gh g h/ lost their aspiration and merged with the correspond-
ing voiced occlusive (Nbh Ndh Ngh Ng h/ → */Nb Nd Ng Ng /): the disappearance of
the voiced aspirates in this position may have preceded all other developments, al-
though this is not certain (cf. Stuart-Smith 2004:211, Meiser 1998:104).
39
As discussed in §3.2.2, if a separate aspirated palatovelar */gh/ is assumed to have existed in
PIE at all, it must have merged with the aspirated velar */gh/ during the (early) Proto-Italic
period at the latest, as there is no evidence that */gh/ developed in a way different from */gh/.
64
PHONOLOGY
(2) Word-initially, the reflex of */bh dh g h/ is represented by f- in Latin, Faliscan, and

in the Sabellic languages: this f- most likely represents a voiceless labiodental spirant
/f/ (or perhaps a voiceless labial spirant / /: see §3.3.3).
(3) Word-initially, the reflex of */gh/ (as well as of */gh/ → */gh/) is represented by h-
in Latin, in Faliscan, and in the Sabellic languages: this h- most likely represents a
voiceless glottal spirant /h/ (or perhaps voiceless velar spirant /x/: see §3.3.3).
(4) Word-internally, the reflex of */bh dh g h/ is represented by -f- in the Sabellic lan-
guages and in Faliscan: in the Sabellic languages, this -f- most likely represents a
voiced labial spirant /β/ (or a voiceless spirant realized as [β], see below). In Latin, on
the other hand, the word-internal reflexes of */bh dh g h/ are b d u /b d /, with /b/ in
some environments also appearing as the reflex of */dh g h/ (see Meiser 1998:104,
Stuart-Smith 2004:41-5).
(5) Word-internally, the reflex of */gh/ is represented both in Latin and in the Sabellic
languages by h-: Faliscan apparently had a different development (§3.3.3).
(6) In Latin at least, (*/#ghR/ →) */#ghR/ developed to /#gR/ (e.g. glaber ← PIE
*/ghlħ2-dh-ro-/, gramen ← PIE */ghrħ1-(s)-mn/): see Stuart-Smith (2004:152-3). This
development is only of indirect relevance to the subject of this study.
These on the whole very similar reflexes point to a development that was at least
partly Proto-Italic: on the other hand, the different word-internal reflexes of */bh dh
g h/ in Latin show that the process cannot have been completed during the Proto-Italic
period.
The development, or rather, the series of developments, that has to be assumed
is a complex one and requires at least two major stages (cf. Picard 1994, Stuart-Smith
2004:195-224), since from occlusives, these sounds became spirants, and from
voiced, they became voiceless at least in word-initial position. The discussion centres
on the order in which these developments took place, and on which stages of these
developments can be ascribed to the Proto-Italic period and which to the post-Proto-
Italic period. Further points that have to be taken into account are how far back the
distinction between the initial and the internal developments goes, and when the re-
flexes of */bh dh g h/ merged into one sound represented by f, a merger that in word-
initial position took place in Latin, in Faliscan, and in the Sabellic languages, and in
Faliscan and the Sabellic languages also in word-internal position.
The first stage in the development is usually assumed to be the change from
voiced aspirated occlusives to spirants, and I see no reason to doubt this, since the
Latin word-internal reflexes /b d / can be (and usually are) explained by successive
Proto-Latin developments. In that case, it seems likely that they changed into voiced
spirants */β ð γ γ /, if only as an intermediate stage: I greatly doubt whether they
could have developed directly into voiceless spirants */ θ x x / (as was first proposed
by Ascoli (1868).
65
CHAPTER 3
The Latin word-internal reflexes /b d / show that the distinction between word-
internal */β δ γ / and their reflexes was preserved until the post-Proto-Italic period. It
seems likely that the voiced quality of these Latin reflexes is a continuation of the
voiced quality inherited from the Proto-Italic reflex, especially as there does not ap-
pear to have been a Proto-Latin revoicing to which it can be ascribed.
This is all the more likely as Rix (1957) has shown that there is evidence that
the Sabellic word-internal reflex of */bh dh g h/, represented by f, was also voiced.
This evidence consists in the first place of the relation between, on the one hand,
Greek SaÚnion and Saun‹tai, and Latin Samnium, Sabini, and Sabelli, and on the other
hand Paelignian and Samnitic safinim nPg 2, Sa 4, which together point to an original
*/saβen-/. Beside this, there is the variation of Samnitic alliba nSa 1b (alliba ST) vs.
allifa nSa 1c, of Oscan alabaternum nCm 3a vs. alafaternúm nCm 3e and alafaternum
nCm 3b-d,f-g, and of Oscan [#]ena#rum nSi 2a (or [v]enavrum ST) vs. Latin Ve-
nafrum.
To this, Meiser (1986:73-4) added the variation of Oscan meḅithi Lu 34 vs. me-
fithi aLu 15, 33, [m]efithi Lu 32, mefitei Lu 16, mefiteí Hi 3, 4, and of stabalano Lu 5
vs. staflatas Cp 24, as well as the fact that in the Pyrgi tablets the name θefa|riei Cr
4.4,4-5 is transcribed in Punic as tbry’. Slightly less convincing, in my view,
are his comparison of Meuania with Latin Mefanus and Etruscan mefanatei Co 1.28,
mefanateś ̣ Co 1.3, and mefanatial Cl 1.1428-1429, and of Scaeua to Etruscan scefi Pe
1.630, 1.1211, scefia Pe 1.201 (cf. §7.7.57). The evidence is reviewed by Stuart-Smith
(2004:84-91), who likewise decides in favour of -f- representing a voiced phoneme,
namely a labial spirant /f/ with a realization [β] in voiced environments (i.e., in inter-
vocalic position and adjacent to a liquid).
Taken by themselves, these instances do not prove that this voiced quality of
word-internal -f- in the Sabellic languages was inherited from Proto-Italic. However,
taken together with the reflexes of Latin, they are strong evidence that the reflexes of
*/bh dh g h/ and, by extension, of (*/#ghR/ →) */gh/, as they were inherited by Proto-
Latin and Proto-Sabellic from Proto-Italic were voiced. On this basis, it can safely be
assumed that these internal reflexes were voiced spirants */β ð γ γ / by the end of the
Proto-Italic period.
What happened in word-initial position, however, is more of a question. In Latin, Fal-
iscan and the Sabellic languages the reflex of */#bh #dh #g h/ is represented by f-, and
that of */#gh/ by h-. There is no evidence that f- ever represented a voiced phoneme in
this position: also, there is no distinction between the reflexes of */#bh #dh #g h/ even
in Latin. Therefore, at some point during the development of the voiced aspirates, a
split between the developments in word-initial and in word-internal position must
have occurred. The question is whether this split occurred already during the Proto-
Italic stage or independently: since the reflexes in Latin, Faliscan, and the Sabellic
languages are the same, the former possibility seems the more likely one.
66
PHONOLOGY
If the reflexes of */#bh #dh #gh #g h/ were still */#β #ð #γ #γ / by the end of the
Proto-Italic period (as they were in word-internal position), this would mean that the
distinction between the word-initial and the word-internal developments, namely the
devoicing and the merging of the reflexes of */#β #ð #γ / into one sound that could be
represented by f, must be ascribed to independent parallel developments in Proto-
Latin and Proto-Sabellic. Alternatively, there may already have been a merger of */#β
#ð #γ / by the end of the Proto-Italic period (either only */#γ / → */#β/, but perhaps
also a merger of */#β/ and */#ð/), so that only the subsequent devoicing has to be as-
cribed to independent parallel developments in Proto-Latin and Proto-Sabellic. Al-
though perhaps possible, I find this an uneconomic solution, and do therefore not
adopt it. In my view, it is more economic to assume that the */#β #ð #γ #γ / were de-
voiced to */# #θ #x #x / by the end of the Proto-Italic period.
The next question is whether the merger of */# #θ #x / was also Proto-Italic.
This merger probably had two stages, namely */#x / → */# / (a merger made possible
by the labial element of labiovelar */#x /), followed by */#θ/ → */# / (as in the word-
internal merger in Proto-Sabellic). There seems to be no arguments either for or
against this merger being of Proto-Italic date, although it affects the development of
(*/#ghR/ →) */#ghR/ → /#gR/ observable in Latin, which must have predated it. (This
development probably (but not necessarily) predated the Proto-Italic devoicing of
*/#β #ð #γ #γ / to */# #θ #x #x /: see Stuart-Smith 2004:152-3.)
Whether or not */# #x/ developed further to */f h/ already during the Proto-
Italic period cannot be ascertained. As the development produced similar reflexes in
Latin and in the Sabellic languages, it is more economic to ascribe the entire devel-
opment to the Proto-Italic period: however, this does require a repetition of the Proto-
Italic merger of */# #θ #x / by the Proto-Sabellic word-internal merger of */β ð γ /.
For this reason as well as for reasons of word-initial and word-internal parallellism, I
prefer to assume that by the end of the Proto-Italic period the reflexes of */bh dh gh
g h/ were word-initial */# #x/ beside word-internal */β ð γ γ /.
The split between the word-initial and word-internal developments of the voiced aspi-
rates caused a morphophonological anomaly in the reduplicative perfect of verbs with
a root starting in an original voiced aspirate. Thus, e.g., by the end of the Proto-Italic
period the present and the perfect of */dhħ2k-/ must have been */ ak-/ : */ eðak-/,
regular from a phonological, but irregular from a morphological point of view. Gener-
alization of the word-initial reflex would have given */ ak-/ : */ e ak-/, preserving
the perspicacity of the paradigm, but producing a (then) abnormal word-internal */ /,
while generalization of the word-internal reflex would have given an even less appeal-
ing */ ak-/ : */ðeðak-/. Meiser (2003:160-1, 178-80) sees this anomaly as the cause of
the disappearance of the reduplicative perfect of roots starting in a voiced aspirate, at
least in Latin: see §5.3.1.6-7.
67
CHAPTER 3
In view of the voiceless quality of the reflexes in word-initial position, the question
may be asked whether the voiced quality of word-internal */β ð γ γ / is a direct con-
tinuation of that of PIE */bh dh gh g h/ or the result of a Proto-Italic revoicing of an
intermediate voiceless reflex (an idea going back to Ascoli 1868, cf. Eichner 1992:73
with n.54). The latter solution would mean that */bh dh gh g h/ or their reflexes */β ð γ
γ / were devoiced in every position, the former, that they were devoiced in word-
initial position only. Stuart-Smith (2004:195-224) is in favour of the former solution,
noting that for phonetic reasons a devoicing is more likely to have taken place in
word-initial position only. This would mean that the Proto-Italic developments of the
voiced aspirates can be envisaged as represented in fig.3.3,A.
The latter solution, however, a devoicing in all positions followed by a voiceless
stage and a later revoicing in word-internal positions, is invoked to provide an expla-
nation for the development of the cluster */sdh/ e.g. in Latin uastus (← PIE
*/ueħ2sdho-/), for if the reflex of */dh/ had remained voiced, the expected outcome
would be PIE */ueħ2sdho-/ → Proto-Italic */ āsdho-/ → */ āsðo-/ *[ āzdo-] →
†uadus †/ ādo-/). (For a different solution (*/sdh/ → */sth/ → */st/), see Stuart-Smith
2004:42-3.) Also, such a revoicing can be connected with the voicing of intervocalic
/s/, which according to some authors (e.g. Meiser 1998:95-6) can be ascribed to the
Proto-Italic period (cf. §3.2.9). If this solution is maintained, the Proto-Italic devel-
opment of the voiced aspirates can be envisaged as represented in fig.3.3,B.
I should stress that for the purposes of this study the exact outcome of the develop-
ment at the end of the Proto-Italic period is of minor importance from a practical point
of view, but of some importance from a theoretical point of view. Assuming that Fal-
iscan is a Latin dialect, the differences between Faliscan and Roman Latin should be
traceable exclusively to post-Proto-Italic and preferably to post-Proto-Latin develop-
ments. It is therefore important to reconstruct the various stages of the development as
exactly as possible.
3.2.9. Proto-Italic voicing of intervocalic /s/? The voicing of intervocalic /s/, i.e., the
creation of a positionally conditioned realization [z] for intervocalic /s/, is sometimes
assumed to have taken place during the Proto-Italic period. Other than the possibility
to link this voicing to a revoicing of (internal or intervocalic) voiceless spirants (see
§3.2.8), there appears to no be reason to assume that this voicing was Proto-Italic,
however: the fact that both Latin-Faliscan and Umbrian must have had [z] as the first
stage of intervocalic rhotacism and that there are Oscan inscriptions where z or z is
used to represent intervocalic /s/ (§3.5.3) do not require that the realization [z] existed
already in the Proto-Italic period. Intervocalic voicing is a widespread and ‘natural’
phenomenon, and may very well have occurred independently in both branches of the
Italic languages.
68
PHONOLOGY
A. */bh dh gh g h/
↓
voiced aspirates become voiced spirants
*/β ð γ γ /
↓
initial: internal:
voiced spirants are devoiced voiced spirants are preserved
*/ θ x x / */β ð γ γ /
↓
merger of */φ θ x /
*/ x/
B. */bh dh gh g h/
↓
voiced aspirates become voiced spirants
*/β ð γ γ /
↓
voiced spirants are devoiced
*/ θ x x /
↓ ↓
initial: internal:
voiceless spirants are preserved voiceless spirants are revoiced
*/ θ x x / */β ð γ γ /
↓
merger of */ θ x /
*/ x/
Fig. 3.3. Two scenarios for the Proto-Italic development of the voiced aspirates.
3.2.10. */p...k/ → */k ...k /. Another development that can be ascribed to the Proto-
Italic period is the assimilation */p...k / → */k ...k /. In Faliscan, this development is
attested in cuicto MLF 310 (and perhaps in cuitenet MLF 361), reflecting PIE */pnk -
to-/ → */k k -to-/. This development is observable also in Latin (quinque, coquus
etc., Meiser 1998:97-8), and must be ascribed to the Proto-Italic period as it is the
source of the o-vocalism of Sabellic */pompe/ (e.g. Oscan pomtis TB 14, púmperia(s)
Cp 28 etc., Umbrian pumpeřias TI IIb.2, and perhaps puntis TI III.4, puntes TI III.9,
10, and, in the onomasticon, in Paelignian ponties Pg 5, Oscan púntiis Po 1,
puntieis Cm 28, pomptiej Me 1, pomptoiej Me 3). In view of Oscan pettiur Sa 17,
Umbrian petur- TI VIb.11 (← PIE */k et ōr/), this vocalism can have developed only
between labiovelars (Meiser 1986:41, cf. Untermann 1957a:187).
69
CHAPTER 3
3.3. Proto-Latin developments
3.3.1. Proto-Latin. Like the concept of Proto-Italic discussed in §3.1.2, the concept of
Proto-Latin assumes that there was a chronological stage in the development of the
Italic languages during which a set of phonological and morphological developments
took place that together set off Latin (in its broadest sense) from the Sabellic lan-
guages. Meiser (1998:54) suggests that this stage can be set, broadly, between c.1800
and c.700 BCE, i.e., during the last 1100 years before the earliest texts.
If Faliscan is a Latin dialect, as I assume, it should be expected that Faliscan
shared all Proto-Latin developments, as well as the possible outcomes of these devel-
opments: where there are differences between Latin and the Sabellic languages, Falis-
can should be expected to side with Latin. As said in §1.5, many scholars regard Fal-
iscan as a language close to Latin. In the context of the successive stages of the Italic
languages, I can interpret this only as meaning (1) that Faliscan is assumed to have
shared some but not all developments of Proto-Latin, and (2) that the developments
shared by Latin and Faliscan must be assumed to precede any developments that oc-
cur only in Latin or only in Faliscan.
3.3.2. The development of */ /. As was said in §3.2.7, the developments of the

sonantic liquids are the same in Latin and in the Sabellic languages and can therefore
be ascribed to Proto-Italic, while the developments of the sonantic nasals are different
in both branches and must therefore be ascribed to the post-Proto-Italic period. In
Proto-Latin, the development was */ / → */em/ and */ / → */en/. In the Sabellic
languages on the other hand the development was */ / → /an/ in initial syllables and
*/ / → /en/ elsewhere (Untermann 1957a:188): there are no certain instances of the
Sabellic reflexes of */ /.
The Faliscan material shows the Proto-Latin development in initial syllables (1)
in Proto-Italic */k k tos/ (cf. §3.2.10.1) → Proto-Latin /k enk tos/, attested in cuicto
MLF 310 (and perhaps in cuiteneṭ MLF 361?), (2) in Proto-Italic */ ð(e)ro-/ → Proto-
Latin */enf(e)rā(d)/, attested in ifra MF 40, (3) in Proto-Italic */t -ē-/ → Proto-Latin
*/tenē-/ → teneṭ LF 361, and (4) in Proto-Italic */k sōr/ → Proto-Latin */kensōr/ →
cen]|so LtF 231, censo LtF 232, [---]sor LtF 232, ce(n)s]or (?) LtF 233 (in Faliscan, the
word is perhaps a Latin loan, cf. §6.3).
Attestations of the same development in medial and final syllables are (1) Proto-
Italic */arg tom/ → Proto-Latin */argentom/, whence arcentelom EF 1, and (2) the
accusative of the consonantal stems Proto-Italic */- / → Proto-Latin */-em/, probably
attested in arute MF 269 and larise MF 270, MF 371, 372, which I regard as accusa-
tives in -e(m) used as nominatives in Etruscan-Faliscan languages contacts (see §4.5.3
and §9.2.2.4).
70
PHONOLOGY
3.3.3. Developments of the voiced aspirates (II): Proto-Latin and Faliscan. As

discussed in §3.2.8, I assume that by the end of the Proto-Italic period the reflexes of
*/bh dh g h/ were */ / in word-initial position and */β ð γ / in word-internal position;
the reflex of (*/gh/ →) */gh/ was /x/ in word-initial position and /γ/ in word-internal
position. These reflexes therefore form the basis of the Proto-Latin and the Proto-
Sabellic development.
(1) Development in word-initial position. Both in Proto-Latin and in Proto-Sabellic
the labial spirant */# / is assumed to have developed into a labiodental spirant /#f/,
and the velar spirant */#x/, to have developed into a glottal spirant /#h/, but at what
stage this should have happened is unclear. Stuart-Smith (2004:45-8, 59-61) leaves
the possibility open that in Latin and Faliscan f- and h- may still have represented /# /
and /#x/ during the earlier stages at least, and there appears to be no evidence either
for or against this assumption.
Since the development in Proto-Latin and Proto-Sabellic shows the same out-
come, Faliscan should be assumed to show the same development and have f- and h-
in initial position. This is indeed the case, but the picture is obscured by the fact that
in Faliscan f- sometimes occurs where */#gh/ (or */#gh/) is reconstructed, and h-,
where */#dh/ is reconstructed. This has lead to a number of sometimes very implausi-
ble suggestions with regard to the development of the voiced aspirates in Faliscan. In
my view, the forms where h- occurs for an expected f- are due to a Middle and Late
Faliscan development /#fV/ → /#hV/, whereas the forms where f- occurs for an ex-
pected h- are hypercorrect forms due to the same development. This is discussed in
§3.5.2.
The regular Faliscan development of the original voiced aspirates in word-initial
position is as follows:
PIE Proto-Latin Faliscan
*/#bh/ /# / /#f/
*/bhĦrs/ or */bhars/ */ ars/ far EF 1
*/bher-/ */ er-/ possibly fertrio Cap 391, and
f MF 58, MLF 352, Cap 392 (if
= Fertor)
? */ a(r)βar-/ fafarṇ MF 136, f ̣af ̣ MF 139
If this name is connected to the potamonym Farfarus, the Latin form of this name,
Fabaris, shows that the f reflects an earlier */bh/ (cf. §6.5.1, §7.8.1.57).
PIE Proto-Latin Faliscan reflex
*/#dh/ */# / /#f/
*/dheigh-/ */ eγ-/ f[.f]ịqod EF 1, fifiked EF 9
*/dhħ2k-/ */ ak-/ faced MF 471* and facet MF
470*
71
CHAPTER 3
*/dhħ1ilios/ */ īlos/ fileo MF 471*, hileo MF 146

h
*/d ħ1ilieħ2/ */ īlā/ filea MF 14 etc.
? */dhermo-/ */ ermo-/ firmio MF 54, fir mia MLF 302
It is not certain that the name Firmius is connected with firmus (cf. §6.3.27).
*/#g h/ */#f/ /#f/
? */g hredhsnom/ */ resno-/ frenaio (=frena‹r›io?) MF 471*
It is not certain that the name Frenaeus/Frenarius is connected with frenum
(§7.8.1.65, §6.2.29).
The following forms in f- have not been included: (1) the unclear fitaidupes EF 1 and
(b) f- in fasies MF 41, felicinate MF 42, fulonia MLF 313, which are of Etruscan ori-
gin. Further instances of word-initial f- (and h-) are discussed in §3.5.2.
Slightly more complex is the development of */gh/ (and */gh/). Word-initially, the
expected reflex of these phonemes is h- /#h/, and this is found in:
PIE Proto-Latin Faliscan reflexes
*/#ghV/ */#xV/ /#h/
*/ghe/o-/ (or */ghe/o-/) */xe-ke/ (loc.) hec ̣ MF 88 etc. (see §6.2.33)
*/gh ħ2-u-/ */xaru-/ harịṣp ̣[ex LtF 231, harisp[ex LtF
232
It is not completely clear whether haruspex is in fact a Faliscan word, since it occurs
in a context where Latin influence might already be assumed.
A case where the development is in accordance with the previous examples, but not in
accordance with the Latin development, is
(? */#gh/ →) */#gh/ */#x/ /#h/
? */gheu-/ ? */xu-/ hutị[.]ilom EF 1
Since Herbig (CIE 8079), hutị[.]ilom has usually been derived from (the zero-grade
of) the root */ghe / ‘to pour’. If this derivation is correct, Faliscan shows a different
reflex from Latin, which has /#f/ in fundo ← */ghund-/ and all other derivations from
this root. This /#f/ is usually ascribed to a separate development (*/#ghu/ →) */#ghu/
→ */#fu/,40 which assumes that the development of */#ghu/ was similar to that of
*/#g h/. Since there are no other words that go back to */#ghu-/ → */#ghu/, it cannot
be checked in how far this development was a ‘regular’ one. I doubt whether, as Stu-
art-Smith (2004:206) suggests, the Faliscan form can be interpreted as having had the
same development as Latin, with a subsequent development /#fV/ → /#hV/: it would
then be the earliest attestation of this development, for which see (see §3.5.2).
40
This idea is ascribed to Ernout, but I have not been able to trace it further than the short
note in DÉ p.208 where he discussed the origins of Latin f-.
72
PHONOLOGY
In word-initial position therefore Faliscan shows the same reflexes as Latin and the
Sabellic languages, with the sole exception of hutị[.]ilom, if this is indeed connected
with */gheu-/, where the development differs from that of Latin.
(1) Development in word-internal position. In word-internal position, the develop-
ments of Faliscan and Latin were different. First, the reflexes of */β ð γ /. In Latin, the
reflexes of */β ð γ / normally appear as b d u /b d / with a ‘fortition’: on this subject,
see Stuart-Smith (2004:203-6). In Faliscan, however, the reflex of */β ð γ / is always
/f/, as it is in the Sabellic languages, and the Faliscan development will have been
similar to the one in Proto-Sabellic (see Meiser 1986:73-8). There is good evidence,
however, that the Sabellic reflex was voiced (see §3.2.8), while for Faliscan such evi-
dence would consist entirely of the name ruvries in Etr XIX = Fa 3.2 as a rendering of
Italic */ruβr-/ ← */ruðr-/ (cf. Stuart-Smith 2004:60). In view of the fact that in Latin,
word-internal -f- probably represented a voiceless sound (Stuart-Smith 2004:46), I do
not think that the voiced or voiceless quality of the Faliscan reflex can be established.
*/bh/ */β/ /f/
*/-bh-/ */-β-/ carefo MF 59, carẹ[f]o MF 60,
pipafo MF 59, ‹pi›pafo MF 60
The future suffix -f- is an analogical Latin-Faliscan formation after the imperfect suf-
fix */-bh-/ → */-β-/, cf. §5.2.1.2. The different reflexes in Latin and Faliscan show that
its creation must precede the separate developments of */-β-/ in Latin and Faliscan.
*/tibhero-/ */tiβeros/ tif MLF 460
This assumes that tif reflects a name like Latin Tiberius (Faliscan *Tiferios). The
Latin counterpart is attested in a Faliscan inscription as tiperilia LF 229.
*/dh/ */ð/ /f/
*/ħ2eidh-/ */að-/ efiles MF 113, 115, efile MF
114, efi[les] MF 116
Efiles may be a calque on Latin aediles, but is in that case still illustrative of the de-
velopment, as it represents an adaptation of the Latin form.
*/ħ1 le dhero-/ */louðero-/ loifiṛtato MF 31, loifirtato MF
32, l]oifirtạ MF 41, loferta MF
222
*/h1reudho-/ */ro ðo-/ rụfia EF 4 (dubious)
In both words, Latin has a reflex b due to the Latin development of */ð/ → /b/ after /u/
(Meiser 1998:104). The connection of rụfia with */ro ðo-/ is not without problems:
see §3.7.2.
*/ dh(e)ro-/ */ ð(e)rā(d)/ (abl.) ifra MF 40
73
CHAPTER 3
In Latin, infra is problematic, as the expected form would be †indra, not infra. Latin
infra has been explained both by assuming that the word was reanalyzed as a com-
pound */en+ð(e)rā(d)/, so that /ð/ would develop as if it were in initial position,41 and
by assuming that the word reflects dialect Latin: see §6.2.37.
In a number of other, mainly onomastic forms, the origin of the f is not clear: (a)
fragmentary text: *[3-5]uf[1-4]ui[..]m EF 1 (often read after Vetter as loufir, loufom, or
roufom), [---]c ̣fli[---|---]ạfc[---] MF 107, abbreviated ef LtF 205 (ef(iles)? or effodio?);
(b) of unknown (probably Italic) origin: ofetios EF 4, ofete EF 4, oufilio MF 48,
aufilio MF 49, aufilio MF 50, aufi[lio MF 53, oufilio MF 275, oufilio MF 276, aufilo
MLF 348, oufilo MF 470*; ofiti or fofiti MF 58; (c) of Etruscan origin: [---]fate MLF
285 (genitive of a toponymic name in ...fas), polafio (=pol‹fa›io?) MLF 353.
Faliscan therefore has a different reflex than the one usually found in Latin. There is
some evidence for the occurrence of -f- in Latin as well. The epigrapic evidence con-
sists of trifos (or triḅos, cf. Vine 1998:261) = tribus in the fifth-century Garigliano
inscription, to which can be added the gloss “rienes quos nunc uocamus antiqui ne-
frundines appellabant quia Graeci nefroÚj eos uocant” Fest. 341.35-343.1L = “renes
antiqui vocabant nefrundines, quia Graeci eos nefroÚj dicunt” Paul. Fest. 342.13-14L
and the few words in the Latin vocabulary with -f- such as af, bufo, furfur (?), rufus,
scrofa, and uafer. These words are often ascribed to ‘dialect influence’, although to
which dialect in particular is hardly ever made clear: it is very well possible that they
were borrowed from a Sabellic language (thus e.g. Coleman 1990).
On the basis of this evidence, the development of word-internal */β ð γ / ap-
pears to be the most crucial phonological difference between Latin and Faliscan. The
Faliscan development appears to have been very similar to the Proto-Sabellic one (see
above): the voiced labiovelar spirant */γ / and the voiced dental spirant */ð/ merged
with the bilabial spirant */β/; this bilabial spirant */β/ then may have developed into a
labiodental spirant that was probably voiceless, /f/. I can see no evidence that there
were any other dialects within the Latin group that shared this development: in Latin,
the spirants underwent a ‘fortition’, */β/ and */ð/ merging with the corresponding oc-
clusives /b/ and /d/, and */γ / with / /.
There appears to be no way to date these different developments apart from the
fact that they must have preceded the earliest epigraphic texts, where the reflexes of
both developments are already present. If this difference in development is compara-
tively recent, it might point to a period where Faliscan was isolated from the rest of
the Latin-speaking area, perhaps by the spread of Etruscan (cf. §2.4.2).
41
Note that such a re-analysis would appear to be possible only at a stage when the word-
internal and word-initial reflexes of the voiced aspirates where still similar, i.e., at the latest at
the Proto-Italic stage when word-initial */θ/ corresponded to word-internal */ð/.
74
PHONOLOGY
The Faliscan development of (*/gh/ →) */gh/ shows even more surprising reflexes.
Whereas in Latin and in the Sabellic languages (*/gh/ →) */gh/ developed to a spirant
that could be represented by h (in Latin probably /h/, in the Sabellic languages per-
haps voiced [γ]), the Faliscan reflex appears to be /g/:
PIE Proto-Latin Faliscan reflexes
h h
(? */g / →) */g / */γ/ ?
*/dheigh-/ */ eiγ-/ fifiked EF 9. f[.f]ịqod EF 1
The use of k and q points (at least at first sight) to an occlusive, /g/. This /g/ can be
explained through analogy with the present */dhingh-/ → Proto-Latin */fing-/, where
*/g/ was the regular development of (*/gh/ →) */gh/ after a nasal (§3.2.8). The use of k
and q is therefore not necessarily a ground to regard fifiked and f[ f]ịqod as perfects of
/dh h1k-/ instead (cf. §5.3.1.6). Latin, too, has several words derived from */dheigh-/
where the reflex is /g/, namely figulus, figura, and effigies. For figulus, the /g/ is ex-
plained by assuming not */dhighelos/ but */dhighlos/ as the original form, which would
have developed to */ iglos/ and thence, with anaptyxis, to */ ig°los/. Figura and effi-
gies have likewise been explained through analogy (see EDL s.v. fingo).
(? */gh/ →) */gh/ */γ/ ?
h
*/leg -/ */leγ-/ lecet MF 88
Here, too the evidence points to a reflex /g/. The verb is not attested elsewhere: Latin
only preserve the related nouns lectus and lectica, and South Picene veia|t MC.1
represents a different formation */leγ-ā-t/. Explaining /g/ by analogy seems to be im-
possible in this case: there does not appear to be any model for it. In Meiser’s view
(2003:68 n.118) lecet still represents /leγet/: note that if in Middle Faliscan c could
still represent /γ/, then it is logical to assume that Early Faliscan k in fifiked EF 9 and q
in f[.f]ịqod EF 1 likewise represented /γ/. Yet I very much doubt whether this is feasi-
ble for an inscription dating from the late fourth or early third century at the earliest.
On the basis of these forms, G. Giacomelli (1963:125) assumed that PIE intervocalic
(*/gh-/ →) */gh-/ developed in Faliscan into /g/. The same conclusion is drawn by Stu-
art-Smith (2004:58, 62, 63-4) with barely a comment apart from noting that Faliscan
is unique among the Italic languages in showing this reflex. I cannot convince myself
that this is the right solution. Within ‘regular’ Latin, a development of word-internal
*/γ/ → /g/ would be understandable: it would be a ‘fortition’ like the one in the word-
internal development of */β ð/ to /b d/. However, in Faliscan, word-internal */β ð/ did
not develop into /b d/ but into /f/ (there is no instance of the reflex of word-internal
*/γ /), which would make a development */γ/ → /g/ even more irregular. Since fifiked
and f[.f]ịqod can be explained by a very plausible analogy, this leaves only lecet as
evidence: I find this a very narrow basis on which to assume a development that not
only runs counter to the developments within Faliscan itself, but has no parallel in any
of the Italic languages either.
75
CHAPTER 3
An unexplained instance of word-internal h is the gentilicium marhio LF 336. This is

in all probability an onomastic borrowing (G. Giacomelli 1963:204, Stuart-Smith
2004:58), cf. Campano-Etruscan ma r hie s Cm 6.1 and Oscan marahis Cm 14.C6,
marahij Lu 2 etc.
3.3.4. Other possibly Proto-Latin developments. The following developments can-

not be dated with certainty but may have taken place before the earliest inscriptions:
(1) /e/ → /o/ before velar /l/. In Latin, /e/ regularly developed to /o/ before a velar /l/
(Pfister 1977:56, Meiser 1998:82). An example of this is the frequent Faliscan name
Volta and its derivations (see §7.7.1.84-85), from Etruscan *Velte/*Velθe. Early Falis-
can uo(l)tenosio (or uo‹l›tenosio) and uoltene EF 3 show that this process was already
under way by the time of the earliest inscriptions. On the other hand, ueltur MF 266,
MLF 339 has an e, suggesting that this was a recent onomastic borrowing, or that the
name was kept in its Etruscan form. In Latin, the /o/ developed further to /u/, but this
is not observable in Faliscan: Volta still occurs as uoltai in LF/Lat 214 (probably
c.150) and in the abbreviation uo Lat 251 (late second century). In felic ̣inate MF 42,
Faliscan reflects Etruscan Felicinat-, while Latin Fulginas and Fulginiae/Fulginium
reflect the development /el.C/ → /ol.C/ → /ul.C/.
(2) */ri/ → */ / → /er/ between dentals. In Latin, /ri/ between dentals (in effect, in
between /t/ and /t s n/) developed to a secondary */ / and thence to /er/ (Pfister
1977:57-8, Meiser 1998:80), cf. tertius as the reflex of PIE/Proto-Italic */tri-t(i)o-/.
The development may be attested for Faliscan in the name Tertineius (tertinei
MLF/Cap 474*, tertineo LF 213), if this is connected with the ordinal. The develop-
ment did not take place in the Sabellic languages, cf. Oscan trístaamentud Po 3 vs.
Latin testamentum (← */terstāmentom/ ← Proto-Italic */tristāmentom/), and trstus (=
tr‹í›stus?) Cm 14 vs. Latin testis (← */terstas/ ← Proto-Italic */tristas/).
(3) reduction of consonant groups. Note the developments of the following conso-
nant groups:
x */nts#/ → /ns#/ in aruz MF 257. Aruz in all probability represents /arruns/ rather
than /arrunts/, the name having been adapted to Faliscan morphology (cf.
§9.2.2.1). If so, the name shows the Latin development of /nts#/: in Umbrian, the
cluster developed into /f#/ via */nss#/ (Meiser 1986:98-100).
x */rs#/ → */rr#/ (→ /r#/) in far EF 1, as in Latin (Meiser 1998:116), and in the Sa-
bellic languages (Umbrian far TI Vb.10, 15, Oscan far Cp 37).
x */nkt/ → /nxt/? In Latin, the /k/ in this group may well have been reduced to /x/,
as is shown by the frequency of quintus/Quintus (although Quinctius regularly
keeps its c in the inscriptions in CIL I2). Faliscan, however, shows cuicto MLF
310 (and cuiteneṭ MLF 361?), with omission of n: see §3.5.7a.
76
PHONOLOGY
x ? */rkn/ → /rn/ in urnam EF 1, ụrneḷ[a EF 1, if these words are connected to Latin

urceus and go back to */urknā/ (cf. Meiser 1998:122).
x */dm/ → /mm/ in umom EF 2 if, as seems likely, this word reflects an earlier
*/udmom/ that occurs in Hernician udmom He 2 (cf. caementum from */kad-
mentom/, Meiser 1998:121).
In view of Latin cella and Oscan kellaked Sa 10, 11, Faliscan cela MF 12, 83, 84,
MLF 285 will have had /ll/, which is unexplained (cf. EDL s.v. for suggestions); for the
ll in putellio see also §3.5.5.3.
3.4. The Faliscan phonemic system
The developments described in the preceding sections, which all precede the Early
Faliscan inscriptions, resulted in a phonemic system as represented in fig.3.4.
vowels /a/ /e/ /i/ /o/ /u/ /ā/ /ē/ /ī/ /ō/ /ū/
‘semivowels’ resonants
nasal coronal
// / /
/m/ /n/ /l/ /r/
sibilants /s/
fricatives
spirants /f/ /h/
labial dental velar labio-

velar
occlusives voiceless /p/ /t/ /k/ /k /

voiced /b/ a /d/ /g/ /g / b
a
) Very rare, see §3.6.1. – b) Not attested.
Fig. 3.4. The phonemic inventory of Early Faliscan.
This system as a whole does not differ greatly from that of Latin (cf. Meiser 1998:52):
the main difference from that of the Sabellic languages is the preservation of the
labiovelar series, which in the Sabellic languages had merged with the labial series at
the Proto-Sabellic stage (see Meiser 1986:79-92), and the structure of the vowel sys-
tem, which in the Sabellic languages had undergone some major changes at the Proto-
Sabellic stage (the ‘Proto-Sabellic vowel shift’, see Meiser 1986:39-54).
77
CHAPTER 3
The phonemic differences between Faliscan and Latin lie mainly in the phono-
tactics, notably the occurrence of word-internal /f/ (§3.3.3). There may also have been
phonetic differences in the realizations of some of the phonemes (see §3.5.1 and
§3.6.1).
The system presented here underwent changes during the Middle Faliscan pe-
riod in the vowels due to the monophthongization of the diphthongs (§3.6.1, §3.7).
3.5. The Faliscan consonants
3.5.1. The consonant system of Faliscan. After the changes brought about by the
disappearance of the voiced aspirates and their subsequent changes to a labiodental
spirant /f/ and a glottal spirant /h/, the consonant system of Faliscan must have been as
represented above in fig.3.4. A few remarks with regard to individual consonants:
(a) Most consonants directly continue the corresponding consonants of PIE and Proto-
Italic. The exceptions are, of course, /f/ from PIE */bh dh g h/ and /h/ from PIE */gh/
(and */gh/), cf. §3.3.3.
(b) The Faliscan alphabet has no b (§11.2.3-4) and /b/ must in fact have been one of
the rarest phonemes in Faliscan. As PIE */bh dh/, which in Latin constituted the main
source for word-internal /b/, became /f/ in Faliscan (§3.3.2), Faliscan /b/ can only re-
flect PIE */b/ (which was itself very rare), as in cupat MF 40 etc. (← PIE */kubħ2-/ or
*/kubħ2-/) and probably in pipafo MF 59, ‹pi›pafo MF 60 (← PIE */pi-bħ3-/ ← */pi-
pħ3-/), or, in the onomasticon, a /b/ in (onomastic) borrowing, as in batio MLF/LtF
359 and blaisiís Sab 468*, or a /b/ that was due to secondary developments, as in pi(s)
LF 242 ← */d is/ (cf. 3.5.6.2).
(c) The only labiovelar that is attested for Faliscan is /k /, spelled as cu (§11.2.4.2),
occurring (1) in -cue MF 80, -c ̣ue MF 158, -cuẹ MF 170, -cue MLF 313 = Latin -que,
(2) in cuicto MLF 310 (and perhaps cuiteneṭ MLF 361) = Latin Quinctus, and (3) in
cụestod LF 242, cues[tor] LF 243, c]ues[tor LF 245, cue[stor LF 247 (perhaps a bor-
rowing from Latin) = Latin quaestor: another instance, (4) cua MF? 129 = Latin quar-
tus etc., is very doubtful (§6.3.63). Cu also occurs in the Etruscan name θanacuil MF
49, tanacu[il] MF 101, θan‹a›cuil MLF 347. There is no Faliscan reflex of */g /; the
aspirated labiovelar */g h/ may be reflected in frenaios MF 471* (see §3.3.3.1). In
spite of this, it is quite clear that Faliscan preserved the labiovelars, as opposed to the
Sabellic languages, where the labiovelars merged with the labials */k g / → */p b/.
(d) The realization of the consonantal phonemes can in many cases not be ascertained.
I have assumed here that /f/ was realized as voiceless labiodental spirant, as in Latin
(§3.5.8, §3.3.3). The /h/ may have been realized quite strongly, not as the weak
Hauchlaut of Latin, since there are no Faliscan instances of omission of h- (§3.5.2).
78
PHONOLOGY
Syllable-final nasals were often omitted in writing (§3.5.7a), indicating that they were
realized as a nasalization of the preceding vowel (/Vn/ [V]/); similarly, /k/ was some-
times omitted in syllable-final position before a dental (§3.5.7c), which might point to
a realization [x] in this position. The fairly frequent omission of /t/ and /d/ in word-
final position (§3.5.7c) might point to a realization as a weak dental or alveolar tap [ ]
n this position: the same may be true of /r/ in word-final position (§3.5.7b). Word-
final /s/ is omitted in writing virtually everywhere, and may have been realized as a
weak glottal sound, [h] or [ ] (§3.5.7d).
3.5.2. The Middle Faliscan development /#fV/ → /#hV/. As mentioned in §3.3.3,

Faliscan has several forms that do not show the regular reflexes of the voiced aspi-
rates, but instead show f- where h- would be expected, or, conversely, h- where f-
would be expected. This has been the subject of a number of discussions, especially as
the Latin grammarians quote similar forms (see below), which suggests that this varia-
tion was not limited to Faliscan. A good summary of the difficulties involved is given
by Wallace & Joseph (1991). Basically, four solutions have been proposed:
(1) separate development of Faliscan and rural Latin.42 Campanile (1961:3-9) rather
vaguely regarded the f-/h- variation as an illustration of the differences between urban
and rural Latin: his list shows that he equated the etymologically unexpected forms
with rural Latin. More explicit is R. Giacomelli (1978:9-22, later (2006:99) abandon-
ing the idea), who assumed a development whereby in rural Latin */#bh #dh/ devel-
oped into a sound represented by h-, while */#gh/ (and */#gh/, §3.2.1) developed into a
sound represented by f-, i.e., the opposite situation of the one in Roman Latin. The
confusion found in the Faliscan material (and presumably also in the forms quoted by
the Latin grammarians) could then be interpreted as switching between informal and
formal language. Although I agree that a geolinguistic distinction may be involved,
which from a Roman point of view may have been regarded as a distinction between
formal/informal language, this assumes that Faliscan was an ‘informal Latin’, which
from the point of the Faliscans it clearly was not: futhermore, there appears to be no
reason for a Faliscan to switch between an urban Roman and a local style before the
war of 241 (§9.4). Wallace & Joseph (1991:88-9) point out that, if this variation is
ascribed to an early different development of Faliscan, the Early Faliscan material
would be expected to show this development, which in fact it does not (see §3.3.3).
42
According to Wallace & Joseph (1993:88), this idea goes back to Meillet (1933:98-9), but
Meillet’s formulation is vague: “Ces correspondances [between f and h] s’expliquent mal à
l’epoque historique: elles doivent remonter au temps où le latin avait des spirantes φ (spirante
bilabiale), þ, x, dont aucune n’était stable et qui ont abouti à f (labiodentale) et à h, après di-
verses hésitations dont les parlers latins ont gardé trace de manières diverses. L’innovation
phonétique essentielle est que les spirantes f [sic] (bilabiale), þ et x, n’ont pas persisté; il n’est
resté que f (labio-dentale) et h, avec quelques flottements dans la répartition.”
79
CHAPTER 3
(2) phonetic (?) confusion. Hiersche (1965) suggested that the variation was in fact
due to a phonetic confusion between a voiceless bilabial spirant ([ ]) represented by f-
and a voiceless velar spirant ([x]) represented by h-. It is very questionable whether
these were in fact the sounds represented by f- and h- in Middle Faliscan (Stuart-
Smith 2004:60-1), and Hiersche’s arguments are not convincing. A similar suggestion
was made by Stolte (1926:42).
(3) merger. Untermann (1964:178) suggested that the sounds represented by f- and h-
merged, but did not elaborate this suggestion. It is apparently also the solution pro-
posed by Wachter (1987:504-6), who assumes that both developed into “einen gerun-
deten Hauchlaut” (1987:505). This, too, had been suggested by Stolte (1926:42). It
comes in fact close to the Hiersche’s idea, the difference being that the merger would
operate on the phonemic level, and Hiersche’s confusion on the phonetic level.
(4) influence from Etruscan. G. Giacomelli (1963:126, 1978:515), Pfiffig (1969:44),
Pisani (1978:45), and R. Giacomelli (1979:153) assumed that the Faliscan variation
was due to a development /#fV/ → /#hV/ that is attested for Etruscan. This seems to
be impossible, however: the attestations of the Etruscan variation f-/h- (including at
least one hypercorrect form, ferclite Cl 1.835 = `Hrakle‹toj) show that this variation
was limited to the north and north-east of the Etruscan-speaking area, and occurred
only from the third century onwards, later than the development started in Faliscan
(cf. Rix 1984:221, Steinbauer 1999:63). The Etruscan inscriptions from the agri Falis-
cus and Capenas in fact show no trace of this variation, which would be expected if
there were some connection between the Faliscan and the Etruscan developments.
Note that this solution assumes an Etruscan influence that is close to Pisani’s Italic
Sprachbund (cf. §1.3.2.1), for which I can see no evidence in Faliscan (§9.2.2).
By far the most satisfying solution is the one proposed by Wallace & Joseph (1991)
and defended by Joseph & Wallace (1993), namely a development /#fV/ → /#hV/ that
took place from the fourth century onward (the first instance that can be linked to this
development is late fourth-century foied MF 59-60). As they note, such a develop-
ment is ‘natural’, and quite common, occuring for instance in Spanish (ferrum →
hierro, filius → hijo), and, as said above, in Etruscan.
Such a development would lead to f- representing [h], an etymologically justi-
fied but phonetically anomalous spelling. The forms where h- occurs instead of an
etymologically justified f- (ultimately going back to PIE */bh dh g h/) would then be a
‘phonetic’ spelling. This ‘h- for f-’ is in effect attested with certainty only in hileo MF
146 beside fileo MF 470*, filea MF 14 (and the abbreviated instances, all of which
have f-: see §6.3.24-25). The forms with f- instead of an etymologically justified h-
(ultimately going back to */gh/ or */gh/), would then be hypercorrect. This ‘h- for f-’ is
attested in fe MF 56, fe 305 beside hec MF 88, 95, 146, 158, LF 223, he[ MF 150, he
LF 220, 221, 224. Another instance, where the form with the regular h- is not attested,
80
PHONOLOGY
but which is derived from the same root as hec, is foied MF 59-60, corresponding to
Latin hodie. Note that in both cases the etymologically correct form is better attested
than the ‘unexpected’ form, indicating that the etymologically correct spelling was in
fact ‘regular’ and the other ‘irregular’ (or, in the case of f- for h-, ‘wrong’ from an
etymological point of view).
Beside these lexical forms, there are a number of onomastic instances with f-/h-
variation where it cannot be established whether this variation represents ‘regular f-
and secondary h-’ or ‘regular h- and hypercorrect f-’. Thus, there is firmio MF 54 and
firmia 302 beside hirmia MF 18, ḥiṛṃeo MF 19, hirmio MF 213. This name has been
connected to the adjective firmus (Proto-Latin */ ermo-/ ← PIE */dhermo-/): if this is
the case, the forms with f- are regular and those with h- secondary. On the other hand,
the name has also been connected with Etruscan Hermena (cf. hermana MF/Etr 264)
in which case the h- may be regular and the f- hypercorrect.
Another instance is the name Folcosius/Holcosius, in the forms fulczeo LF 329,
folcozeo LF 330, folcuso LF 331 and folcosio LF 333 beside holc[osi] MF/LtF 140
and the unclear *olcuzeo LF 332. In this case, neither Folcosius nor Holcosius is at-
tested elsewhere: the closest parallel is Latin Holconius.
Other onomastic forms with h- that may be illustrative of this development are
hadenia MLF 360 (connected to Fadenius?), hap MF? 46 (an abbreviated form of
Fabius?), hạθ̣i MF 13 (connected to Fadius?), hescuna MLF 346 (connected to the
toponym Fescennium?), and perhaps also hac****na MF 89.
The phonological development as described above is limited to word-initial antevo-
calic position, but there are two hypercorrect instances where h is written for f in other
positions, namely hlauelea LtF 325 = Latin Flauilia, and perhaps also in ohi*[---] LtF
300, if this is the praenomen Aufilus/Oufilus. Both these instances are probably erro-
neous extensions of the custom of writing h for f to positions where this was not due
to a regular development: both instances are from Latino-Faliscan texts written in the
Latin alphabet, perhaps by people not well acquainted with the rules of Faliscan or-
thography.
The only point on which I do not entirely agree with Wallace & Joseph is their expla-
nation of how it was possible for the variation to occur: they assume that the variation
reflects dialectal variants within Faliscan. Although possible, I do not altogether relish
this idea. In my view, it may well be possible to ascribe the distribution of the two
variants to the influence of the written form, where the spelling with f- was main-
tained even though the realization had changed from [f] to [h]: I already mentioned
the fact that the etymologically correct forms outnumber the other forms. The spelling
might further be influenced by the fact that the words always occur within well-
established formulas (filius/filia in the formula of filiation, and hec in the formula hec
cupat/cupant), which may further have influenced the spelling.
81
CHAPTER 3
There is some evidence for the same variation in Latin. The Praenestine inscriptions
have f- for h- in foratia CIL I2.166, and, in Greek names, in felena CIL I2.566, feliod
CIL I2.1446 = Heliod(orus), and probably fercle CIL I2.564 (if to be read thus) = Her-
cules. The Latin grammarians, too, mention a variation between f- and h-:
“nos non tam propter illas causas quas supra proposuimus ¢ harenam dicimus² , quam
propter originem uocis, siquidem, ut testis est Varro, a Sabinis fasena dicitur, et sicut s
familiariter in r transit, ita f in uicinam adspirationem mutatur. similiter ergo et haedos
dicimus cum aspiratione, quoniam faedi dicebantur apud antiquos; item hircos,
quoniam eosdem aeque fircos uocabant. nam et e contrario quam antiqui habam dice-
bant nos fabam dicimus” (Velius Longus CGL 7.69.6-10)
“quam Falisci habam, nos fabam appellamus, et quem antiqui fariolum, nos hariolum”
(Ter. Sc. CGL 7.13.8).
“ircus quod Sabini fircus; quod illic fedus, in Latio rure hedus: qui in urbe, ut in multis
a addito haedus” (Varro L 5.79)
“Faliscos Halaesus condidit: hi autem immutato h in f Falisci dicti sunt, sicut febris dici-
tur quae ante hebris dicebatur, Formiae quae Hormiae fuerunt, ¢pÕ tÁj ÐrmÁj: nam poste-
ritas in multis nominibus f pro h posuit” (Serv. in Verg. A. 7.695), cf. “uenerat Atridae fa-
tis agitatus Halaesus, a quo se dictam terra Falisca putat” (Ov. Fast. 4.73-4); “oppidum
Formiae, Hormiae dictum, ut existimauere, antiqua Laestrygonum sedes” (Plin. NH
3.59), and “Formium oppidum appellatur ex Graeco, uelut Hormiae, quod circa id cre-
brae stationes tutaeque erant, unde proficiscebantur nauigaturi.” (Paul. Fest. 73.19-22L)
“faedum antiqui dicebant pro haedo, folus pro holere, fostim pro hoste, fostiam pro
hostia.” (Paul. Fest. 74.9-10L)
“horreum antiqui farreum dicebant a farre.” (Paul. Fest. 91.6L)
“horctum et forctum pro bono dicebant.” (Paul. Fest. 91.14L); “forctes, frugi et bonus,
siue ualidus” (Paul. Fest. 74.14L)
“Fordicidis boues fordae, id est grauidae, immolabantur, dictae a fetu.” (Paul. Fest. 74.1-
2L), vs. “horda praegnans, unde dies, quo grauidae hostiae immolabantur, Hordicidia.”
(Paul. Fest. 91.17-8L)
“hanula parua delubra, quasi fanula.” (Paul. Fest. 91.25L)
In my view, this material is of some interest as a possible illustration of the develop-
ment described above, but not of much practical use. The linguistic context in which
these forms were found by the authors is unknown, and the attributions to the antiqui,
the Sabines, or the Faliscans, appear to be quite arbitrary. At best, it looks as if the
grammarians and glossographers were aware of a (chronological and/or geographi-
cal?) variation between f- and h- within their Latin sources, and simply ascribed the
variant that was not the usual one in Roman Latin to an (unspecified) older stage of
Latin, or to dialects on its margin. Note, for instance, that where they describe this
variation as a chronological development, Velius Longus, Terentius Scaurus, and Fes-
tus regard f- as the older stage (although Terentianus also notes that the reverse is true
in the case of haba), while Servius clearly regards h- as older, even stating that “pos-
82
PHONOLOGY
teritas in multis nominibus f pro h posuit”: a difference that is clearly due to the fact
that Velius, Terentianus Scaurus, and Festus try to explain forms with a non-standard
f-, while Servius tries to explain a non-standard h-. It also explains why the Faliscans
can be credited on the one hand with haba instead of faba (Terentianus Scaurus) but
on the other hand are said to have derived Faliscus from Halaesus (Servius, Ovid).
Interestingly, in the cases where this can be established, the grammarians’
‘standard form’ appears to be the etymologically correct one: in fordus and for(c)tis,
the f- goes back to */#bh/, while in haedus, hariolus, holus, and hostis, and perhaps
also in hircus and hostia, h- goes back to */#gh/ or */#gh/; faba may be an old borrow-
ing (Proto-Italic */fafā/, EDL s.v.). If due to a (dialectal) development /#fV/→ /#hV/,
haba, hordus, horctus would directly reflect this development, while faedus, fariolus,
fircus, folus, fostis, and fostia would be hypercorrect. In none of these cases did the
form that resulted from the development become standardized in Roman Latin, and
they were therefore in effect ‘irregular’.43 The etymologies of Formiae and harena are
unknown or unclear: horreum and farreum are unrelated, but show that a variation
f-/h- was sufficiently well-known to allow Festus to use it as an ‘etymology’.
This material indicates that the development was not limited to Faliscan, but
occurred in other Latin dialects as well, since it would hardly be feasible that all these
forms were in some way derived from Faliscan sources. The fact that the Faliscans are
singled out at all as a dialect in which the grammarians detected this variation may
simply be due to the fact that in the case of the Faliscans the variation was better
known, as it affected the names of the people itself (note the connection made be-
tween Falisci and Halaesus) and of their two main towns, Falerii and Fescennium.
3.5.3. Rhotacism and allophones of /s/. Word-internal rhotacism took place in Latin
somewhere before the end of the fourth century (Pfister 1977:145-7, Meiser 1998:95-
6). In the Sabellic languages, word-internal rhotacism took place only in Umbrian,
probably at the same time as in Latin (Meiser 1986:240-1). In a few southern Oscan
inscriptions, intervocalic /s/ is rendered by z: the Tabula Bantina has angetuzet, cen-
sazet, egmazum, eizac, eizasc, eizazunc, eizeic, eizeis, eizẹ[i]s, eizoic, eizuc, ezum;
beside this, there is eizidom Lu 5, #enzei Lu 31 The first stage, voicing of intervocalic /s/
to [z], may go back to the Proto-Italic stage (§3.2.9).
If word-internal rhotacism took place in Faliscan at the same time as it did in
Latin, Early Faliscan should be expected to show an unrhotacized /s/, but there are
unfortunately no attestations of /s/ that do not remain unrhotacized later. (kaisiosio EF
7) or do not appear in post-rhotacism (the ending -osio in kaisiosio EF 7 and aịṃiosio
EF 467*). In Middle and Late Faliscan, rhotacism might be expected, but the only
43
An exception might be hīlum (cf. also ni-hil), if this is identical with fīlum (probably from
Proto-Italic */fī(s)lom/ ← PIE */g hiĦ(s)lom/ (EDL s.v. fīlum).
83
CHAPTER 3
form that can be connected to rhotacism is carefo MF 59, carẹ[f]o MF 60, probably
from Proto-Italic */kas-/ ← PIE */kħ2s-/. Since these attestations are from the late
fourth century, it can on this basis be assumed that intervocalic rhotacism took place
before that date in Faliscan, as it did in Latin.
This view has been attacked by Bonfante (1966), who assumed that Faliscan
shows voicing of intervocalic /s/ to [z], but not a fully rhotacized /r/, regarding carefo
as a calque on Latin carebo. This is largely based on the observation by Sittig (1932)
and Belardi (1964) that the form of the r in carefo MF 59 and carẹ[f]o MF 60, , is
not only different from the other r in these inscriptions ( ), but unique and designed
especially with the aim of representing [z]. Even apart from the general unlikelihood
that a special sign would be developed to denote a positional variant of /s/ where the
lettre morte z might have been used, this statement is erroneous: the shape of r occurs
also e.g. in MF 101, where it is in fact used for a, , just as the other type of r, ,
could be (see §11.2.4.2). Unfortunately, this erroneous statement has been repeated by
several authors, even R. Giacomelli (1978:44) and Stuart-Smith (2004:63 with n. 67).
The Middle and Late Faliscan inscriptions also show a number of forms with an un-
rhotacised s. In several other instances, s represents /ss/, as in keset LF 242 kese[t LF
243 = /gessē ̣t/, and probably in the names fasies MF 41 (cf. Latin Fassius), ruso MLF
318 (cf. Latin Russo), and perhaps in mesio MF 148 (cf. Latin Messius). Larise MF
270 represents the Etruscan name laris, either with an epenthetic [e] or used as the
basis for an accusative larise(m), see §9.2.2. In the verbal form esú(m) Cap 389, 404,
465, the /s/ is preserved by analogy with the rest of the paradigm.
In several onomastic instances, the spelling s may indicate that the name was a
recent onomastic borrowing, or it may have been preserved simply because it was the
traditional way of spelling the name: note that for most Faliscan instances there is
usually an unrhotacized parallel from the Latin onomasticon. The spelling s is in fact
standard in the name Caesius and its derivations (as it is in its Latin counterpart
Kaeso), caisioi MF 20 etc. and cesịe MF 257, cesies MF 265 (for all derivations, see
§7.7.1.18 and §7.8.1.33). Other instances are uisni MF 82 (Latin Visinius), mesio MF
148 (if reflecting Maesius), folcuso LF 331, folcosio LF 333 (see below), and
nomesina MF 272, MLF/Etr 289, numesio LtF 377, [.]osena MLF 206.
Already from the Early Faliscan period onward, there are instances where z is used
instead of s. This usage has been discussed by Bonfante (1966), R. Giacomelli
(1978:32-43), and Wachter (1987:43-50), while Van Heems (2002) discusses a very
similar variation in the Etruscan inscriptions from Volsinii. Basically, there are three
possibilities: (1) z is a lettre morte within the Faliscan alphabet that can be used indis-
criminately instead of s; (2) z denotes one specific realization of /s/ and therefore
represents the same phonetic value in every instance, and (3) z can denote various
realization of /s/ and may therefore represent various phonetic values, depending on
84
PHONOLOGY
the context. If z represents a realization of /s/ different from [s], the only ones that can
be considered are [ts] and [z], as appears from the values of z in the other languages of
ancient Italy. The instances are the following:
(a) Word-initial: lexical or semi-lexical: zextos EF 1 and zextoi LF 330; zot MLF 285;
zenatuo LF/Lat 214; onomastical: zaconiọ MF 153 and zaconiai MF 154; zuconia MF
271 and perhaps zu[con]|eo MF 56; zeruatronia MF 272; zertenea LF 221. The idea
that z- may have represented [ts] is difficult: there are no other signs of affricates in
Faliscan or Latin, and as a phonetic tendency such a ‘strengthening’ of /#sV/ would be
in contrast with the ‘weakening’ of /#fV/ to /#hV/ (§3.5.2). Bonfante (1966) suggested
that z represented [z], attributing the instances in word-initial position to a voicing of
/#sV/ in saṇdhi /V#sV/. This can apply only to the cases of mama zextos EF 1, poplia
| zuconia MF 271, lete zot MLF 285, uipia  zertenea LF 221, and de | zenatuo LF/Lat
214; in the case of folcozeo | zextoi LF 330, it is possible only if it is assumed that in
Faliscan /s#/ was reduced completely to zero (which I do not, see §3.5.7d). For the
cases where a saṇdhi is impossible, namely uel zu[con]|eo MF 56, folcozeo | zextoi
LF 330, and cauio  nomes|ina  maxomo | zeruatronia MF 272 (where there must
have been a ‘pause’ between the two names, ‘Gavius Nomesina Maxumus; Serva-
tronia’), an analogical spread of either the realization [z] or of the spelling z- has to be
assumed. 44 (Van Heems (2006) notes the same problem in the Etruscan inscriptions
from Volsinii.). It is possible that in some of the cases of z-, it can be ascribed to
Etruscan influence: note that in the cases of zuconia MF 271, zu[con]|eo MF 56, and
zertenea LF 221 the z- is also found in the Etruscan form of the name (Zuχ-, Zertn-).
(b) Word-internal: only onomastical: fulczeo LF 329, folcozeo LF 330, and *olcuzeo
LF 332 vs. folcuso LF 331 and folcosio LF 333. In these instances, it is not unreason-
able to assume that -z- represents [z], as Bonfante (1966) proposed. After rhotacism,
intervocalic /s/ in onomastic borrowings from a non-rhotacist language such as Etrus-
can may well have been realized as [z].
(c) Word-final: aruz MF 257, morenez MF 269, and perhaps ạcṛẹẓ MF/Etr 67. In aruz,
-z could potentially represent /-ts#/, with aruz representing /arrunts/, but in view of the
fact that the name is always adapted to Faliscan morphology (gen. ar]uto MF 169,
aruto MF 257, 266, aronto MF 265; acc. arute MF 269?), it is more likely that the
nominative was /arruns/. However, -z may have represented a realization of /s#/ that
was ‘stronger’ than the weakened /Vs#/ of Faliscan (§3.5.7d), i.e., [s] instead of [h]: in
the case of aruz, because /s/ followed a consonant rather than a vowel, in the cases of
morenez MF 269 and ạcṛẹẓ MF/Etr 67, because these words are ‘Etruscoid’ forms in
-(i)es (§9.2.2.2c), where -s was usually written rather than omitted (§3.5.7c).
44
In [---] zaconiọ MF 153 and [---] zaconiai MF 154, the preceding word is missing.
85
CHAPTER 3
(d) unclear cases of z are: e**azieputilepe EF 4, probably calin[---]|rezo[---] MF 57 ,

me[.]uelau[.]suae rfa zeuoc*na MLF/Etr 357.
All in all, I see no reason to question the idea that Faliscan had complete intervocalic
rhotacism, probably at the same time as Latin and Umbrian. The use of z instead of s
may have served to represent the realisation of intervocalic /s/ as [z]: in word-initial
and word-final position, its use seems to be due to Etruscan orthographic influence
and represents [s] rather than [z] or [ts].
3.5.4. /t/ realized as [d]? Just as s is sometimes replaced by z, so t is sometimes writ-

ten as θ. Since θ properly belongs to the Etruscan and not to the Faliscan alphabet
(§11.2.4), this spelling can be regarded as influenced by Etruscan in every case. This
is most clear in the cases of θania MF 81, θanacuil MF 49, and arθ[3-5]rẹ MF/Etr
267, where the use of θ is in all probability due to the fact that in Etruscan these
names were written with θ. In the cases of [---]nθia MLF 212, senθia MLF 362,
uolθeo MF 276, and salθan MF/Etr 77, however the /t/ could potentially represent a
specific realization of /t/ after a nasal or a liquid. In that case, a voiced realization [d]
seems the most likely possibility. The θ could then be regarded as a graphical variant
of d: in fact, in MF 276, and perhaps also in MF 49 (which is known only from apog-
raphs), the θ has the shape , and these instances have in fact been read as uoldeo and
danacuil by the earliest editors. Similarly, in my suggested reading hạθ̣i MF 42 the θ
could represent a [d] due to intervocalic voicing. On the other hand, a voiced realiza-
tion of /t/ represented by θ is hard to envisage in uesθi MF 83. I am inclined to regard
all instances of θ as due to Etruscan orthographic influence.
3.5.5. Palatalizations. In a number of instances, palatalization of liquids and nasals

has been proposed: the point is elaborated by R. Giacomelli (2006:91-3), who regards
these palatalizations as one of the ‘volgarismi’ observable in Faliscan.
The data, unfortunately, are unclear. I regard a palatalization certain in the case
of /d/, where palatalization is well-attested especially for Latin: the other instances of
palatalization are uncertain at best: note that they all involve cases of /CV/, where the
Faliscan spelling in such groups frequently employed e instead of i, implying that the
group was /CiV/ rather than /CV/.
(1) Palatalization of /d/ → // or //. The only instance is foied MF 59-60, which,
seeing that it goes back to */hō(d)+dē(d)/, may reflect either the palatalizaion of
/VdV/ or of /#dV/. The palatalization of /VdV/ must have preceded the Latin devel-
opment of /ð/ → /d/ in view of the reflex medius (← PIE */medho-/) instead of
†meius; the palatalization of /#dV/ can be placed in the late fourth century (Meiser
1998:111). This is the only palatalization observable in the Faliscan material that may
be considered phonemic rather than phonetic.
86
PHONOLOGY
(2) Palatalization of /s/ → //. The palatalization of /s/ → // has been proposed by
Pisani in the case of his development of /-oso/, and is apparently adopted by Meiser
(1998:117). If occurring at all, which I doubt very much, it must have occurred later
than the fourth-century word-internal rhotacism in Latin, which changed many of the
names in /-Vso-/ to /-Vro-/ (cf. Papisius, Valesios, and Fousios as the older forms of
Papirius, Valerius and Fūrius) but apparently spared the morpheme /-oso/ (or its re-
flex */-ese/). For more detailed discussion of this issue, see §4.10.
(3) Palatalization of /l/ (→ // or //)? The idea of a Faliscan palatalization of /l/
goes back to Herbig (1914b:251), who interpreted tito polafio MLF 354 as tito p̣ola
fio (‘Titus Pola jr.’), with fio “wohl sicher über *fiius aus *fil’ius”: it was adopted by
Pisani (1962:59), G. Giacomelli (1963:246) and Orioles (1972:78-9). In view of the
fact that when the word is written in full, it is always spelled with e, fileo MF 470*,
hileo MF 146, file (abbr.) MLF 308, as well as other instances of antevocalic e for i
(§3.6.2) it is unlikely that the word was pronounced [fīljo] at all: it was in all probability
realized as [fīl o]. Hirata (1967:68) plausibly connected polafio to gentilicia in
Polf-/Pulf-, in which case it could be an error for pol‹fa›io with a suffix as in latinaio
MLF 210, uoltaia MF 196, and ac ̣iuaiom Cap 465. In any case, there is no evidence to
assume a palatalization here.
Prosdocimi (in G. Giacomelli 1978:508) pointed to seruio Lat 250, which repre-
sents the name of the consul of 106 BCE, Q. Seruilius Caepio, as a confirmation of
Herbig’s idea, but in my view this is simply an error for serui‹li›o, not an attestation
of a palatalization, as M. Mancini (2002:27 with n.25) rightly notes. Even if it is re-
garded as a palatalization, it occurs in an inscription that is almost 50 years later than
any datable inscription that is possibly Faliscan, and it cannot be assumed that it
represents Faliscan, or even a Faliscan feature that entered local Latin. It can certainly
not be used as an argument to assume a Faliscan feature that is not attested otherwise.
R. Giacomelli (2006:91-3), however, accepts the cases of polafio MF 354 and
seruio Lat 250 as evidence. He also adduces lullio MF 207 and putellio MF 152 (cf.
also uolḷia MF 47) and regards these rare geminated spellings as indicative of a pala-
talization: note that these forms would point to [ ] rather than to the [] of polafio and
seruio.45 He is certainly right in pointing out the rarity of such geminated spellings in
Faliscan (§11.2.5.5), and palatalization may explain them, but it goes too far, in my
view, to compare them to the Oscan spellings such as tti and nni representing palatal-
ized /t/ and /n/ respectively, and to regard the Faliscan spelling as a ‘sabinismo-
oschismo’.
45
Why he also adduces “malio [MF 39] ~ lat. Mal(l)ius”, “tali [MF 84, uncertain] ~ lat. Tal-
lius”, and “tulo passim ~ lat. Tull(i)us” (2006:92) is unclear: Mallius and Tallius do not ap-
pear to be evidence for a palatalization, and Tullius is a patronymic derivation from Tullus
(§7.5, §7.7,1.76, §7.8.1.156), not a different spelling that could represent a palatalized /l/.
87
CHAPTER 3
(4) Palatalization of /r/ → // or //? The evidence for this consists of the gentilicium
clipeaio MF 470* beside clipịaṛ[io] LF 230 (where the earliest editors read clipiai[)
clipeaṛ[io] LtF 231 and cḷ[i]peario LtF 233, and frenaio MF 471*, which would be
expected to be formed with the same suffix /-āro-/ (Clipearius ‘Shieldmaker’,
Frenarius ‘Bridler’). The forms in -aio could then point to a palatalization of /r/. On
the other hand, either form could of course just be an error for clipea‹r›io and
frena‹r›io: it is also possible that frenaio was formed not with /-āro-/ at all, but with a
suffix as in latinaio MLF 210, uoltaia MF 196, and ac ̣iuaiom Cap 465.
(5) palatalization of /n/ → [ ]? The evidence for this consists of two gentilicia of
Etruscan origin ending in -no where Latin has -nius: ueicọno MF 88 (and perhaps also
in tuconu MF 85, if read as t u(e)conu) beside Latin Veconius and Viconius (§7.8.1.174)
and precono[---] MLF 361 beside Latin Praeconius (§7.8.1.126). To this could be
added aino MFL 352 (§7.8.1.8) if this is indeed an adaptation of an Etruscan name in
-na (cf. eina MF 57), and uelmi|no MLF 316 beside uelmineo MLF 305, 307, 309, 310,
312, 313, and uelminẹo MLF 308, uel|mineo MLF 315. In these forms, -no could poten-
tially represent [- o ] with a palatalized /n/, as is suggested by R. Giacomelli (2006:93).
In my view it is more likely that aino, ueicọno, and precono were derived in another
way than in Latin, i.e., not adapted as -na → -nius but simply thematized as -na → -nus
(§7.8.2.2) and that uelmi|no is an error for uelmi|n‹e›o: the inscriptions of this group
contain several errors.
3.5.6. Minor developments. Two other developments that appear to have taken place
during the historic period are the following:
(1) Closing (and lengthening?) of /e/ before /n/. In Latin, /e/ was closed to /i/ before
/n.s n.f n.x/, probably at least partly during the historical period (Pfister 1997:100,
Meiser 1998:78-9, 81). The same development can also be observed in Faliscan (1) in
ifra MF 40, representing */infrā/ (or possibly /īnfra/) ← Proto-Latin */enð(e)rā(d)/ ←
/ ð(e)rā(d)/ and (2) in cuicto MLF 318 (and perhaps in cuiteneṭ MLF 361) represent-
ing */k inkto-/ (or */k īnkto-/) ← Proto-Latin */k enk to-/. In Latin, the vowel pre-
ceding the /n/ was lengthened as well as a compensatory lengthening due to drop of
the nasal, followed by a restoration of the nasal /VN.C/ → /V.C/ → /Vn.C/ (Pfister
1997:100, Meiser 1998:78-9). As explained in §3.6.1, it cannot be established whether
the vowel was lengthened in Faliscan as it was in Latin, although this does not seem
unlikely if Latin lengthening was a compensatory lengthening due to the omission of
the nasal (see §3.5.7a).
(2) /#d V/ → /#bV/. In Latin, /#d V/ became /#bV/ by the middle of the third cen-
tury: du- still appears as an archaism in duonoro CIL I2.9 and duelonai CIL I2.581,2
and in Plautus, e.g. arte d ellica Epid. 450. In Early Faliscan, there are likewise in-
stances of du- in duenom duenas EF 3. Although the reflex of /d eno-/ itself does not
88
PHONOLOGY
appear in Middle or Late Faliscan, there are now two instances of pi(s) in LF 242 that
correspond to Latin bis /bis/ ← duis /d is/ (“duis ... pro d…j ponebatur”, Paul. Fest.
58.17L). Assuming that these words reflect Faliscan and do not represent Roman
Latin (and I see no reason why they should, cf. §1.3.2.2), they show that the same
development took place in Faliscan.
3.5.7. Omission of syllable- and word-final consonants. The general tendency to

drop of syllable- and word-final consonants reflects a tendency to form ‘perfect sylla-
bles’ of the type CV by a reduction of the elements of the coda (cf. Venneman
1988:21-27). The results of such a reduction can be phonemic, or be limited to the
phonetic realization of the consonants involved (e.g., by weakening them or, in the
case of nasals, by substituting a nasalized lengthened vowel for the consonant). In
Faliscan, this reduction affected mainly nasals, and, to a degree, liquids, although re-
duction of other syllable- and word-final consonants, especially /s#/, is also well-
attested. Note that I prefer to use the term ‘omission’ (which refers to the representa-
tion of a sound in writing) rather than ‘drop’ (which refers to the reduction of a pho-
neme to zero). In many cases, it may be assumed that the phoneme was not reduced to
zero altogether, but still maintained some form of phonetic realization, and that there
was a synchronic variation in its realization. In these cases it would be unjustified to
speak of ‘drop’ of the consonant, only of its omission in writing.
(a) Nasals. The omission of a syllable-final nasal is especially frequent in Faliscan
(for attestations, see fig.3.5): in fact, its omission is about as frequent as its being writ-
ten out. The omission represents a reduction of the nasal consonant with a transfer of
the nasal quality to the preceding vowel, which was in all probability lengthened. The
frequency of the occurrence may also be illustrated by the fact that it may be found
already in Early Faliscan, if salueto EF 4 is read as salue(n)to or salue(n)to(d) (cf.,
however, §5.2.4b), or if ofetios EF 4 is regared as related to the potamonym Ufens
(cf., however, §3.7.2 and §7.2.2).46
Omission of /m#/ may be related to the length of the preceding vowel. In the
Early Faliscan period -m is written throughout, both where the ending is /Vm#/, as in
the first-declension accusative singular urnam EF 1, and probably also eitam EF 5
(and in arcentelom hutị[.]ilom EF 1 if this is interpreted with Peruzzi (1964a:163-4) as a
second-declension genitive plural), and where the ending is /Vm#/, as in the second-
declension accusatives and neuter nominatives, *[0-2]e[1-3]tom EF 1, arcentelom
hutị[.]ilom EF 1, duenom EF 4, propramom EF 2, [u]mom, umom, umọ[m] EF 2, and
46
A prehistoric phonemic drop of /n#/ occurred in the nominative singular of the ōn-stems, as
in apolo MF 65, cupi‹d›o MF 62 and perhaps also in quto EF 3, if representing /g�tō/ or
/g�ttō/ (note, however, the uariae lectiones qutoṇ and qutoṇe). This development was of
Proto-Latin date: in the Sabellic languages, these nominatives were recharacterized by the
addition of /-s#/ with subsequent assimilations of the resulting cluster /-ns#/: see §4.5.1.3.
89
CHAPTER 3
possibly ui[no]m EF 1 (if to be restored thus). However, in the Middle and Late Falis-
can inscriptions -m is always omitted after a short vowel, in the accusatives
uino MF 59-60, mạcistratu LF 242, and also in arute MF 269 and larise MF 270, MF
371, 372, which I regard as accusatives in -e(m) used as nominatives (§9.2.2). In the
genitive plural, it is written in tulom MF 72 and ac ̣iuaiom Cap 465*, but omitted in
[fel]ịcinatiu LF 384.
The /n/ can also be omitted in the word-final clusters /nd#/ and /nt#/, occurring
in the secondary and primary endings of the third plural. This is already found in the
Early Faliscan inscriptions, f[.f]ịqod EF 1 = f[.f]ịqo(n)d and possibly also in *[3-4]*ad
EF 1, if this is a plural *[3-4]*a(n)d. For Middle and Late Faliscan there are the in-
stances zot MLF 285, and cupat MF 146, cupạṭ MF 158, cupat LF 223, and perhaps
cupa[?t] MF 95 (if this is indeed a plural), vs. cupa]nt MF 80. Latin parallels for such
forms are dedrot CIL I2.378, emerut CIL I2.1513, and sut CIL I2.1824. Omission of /n/
in the cluster /ns#/ occurs in aruz MF 257, cf. Latin aros CIL I2.2081.
(b) Liquids. The omission of syllable- and word-final liquids is rarer than the omis-
sion of syllable-and word-final nasals. However, as in the case of syllable-final nasals,
omission of syllable-final l is found already in Early Faliscan, namely in uotenosio EF
3, if, as seems likely, this is the same name as uoltene in the same inscription. In Mid-
dle Faliscan, the omission is limited to a few cases of omission of syllable-final r, in
ma(r)cena MF 269, aca(r)celini LF 222, ma(r)ci LF 222, and probably also se(r)torio
MLF/Cap 476*, and of one instance of omission of syllable-final l, in uo(l)tiliạ LF 223,
vs. an overwhelming number of instances where syllable-final r and l are written out.
Further instances may be the names fafarṇ MF 136 and f ̣af ̣[---] MF 139, if they are re-
lated to the potamonym Farfarus. In view of the Latin form of the potamonym, Faba-
ris, this need not necessarily be an instance of the omission of a syllable-final liquid.
Latin, too, shows instances of omission of syllable-final r: mamor CIL I2.2, dosuo CIL
I2.270, asom CIL I2.560 (and perhaps prosepnai CIL I2.558, cf. Wachter 1987:115-6),
and, from the second century BCE, controuosias CIL I2.584,2, suso CIL I2.584,7, 8, 15.
In word-final position, too, omission of a liquid is rare: uxo MF 17, mate LF
221, uxo LF 222, uxo LF 242, mino LtF 173, cen]|so LtF 231, and censo LtF 232 vs. ux
or MF 41, uxor MF 42, pater MF 62, uxor MF 101, uxor MF 265, ueltur MF 266,
[u]xor MLF 301, ueltur MLF 341, p[reto]|r MLF/LtF 241, [pre]tor LF 247, pre]tor LF
248, cen]|sor LtF 231, and cen]|sor LtF 232. It may already occur in Early Faliscan, if
quto EF 3 represents /g�tor/ or /g�ttor/ (§6.2.31). In Latin, too, the omission of word-
final -r is rare, occurring only in marma CIL I2.2 (vs. twice marmar in the same in-
scription), uxo CIL I2.1829 (vs. 31 instances with -r), and in censento CIL I2.583,77
and rogato CIL I2.583,78: on the other hand, it appears to have been almost regular in
the names Maior (maio CIL I2.76, 161, 233, 329, 344, 347, 2471, 2482 vs. maior CIL
I2.1340) and Minor (mino CIL I2.126, 139, 194, 197, 198, 330, m]ino 77 vs. minor CIL
I2.271): cf. Latino-Faliscan mino LtF 173.
90
PHONOLOGY
written out omitted

syllable- arcentelom EF 1 ? salueto EF 4
final ? ofetios EF 4
aronto MF 265 aruto MF 257
ceso MF 232 aruto MF 266
cincia MF 135 [---ar]uto MF 169
ian[ta MF 146 arutlo MF 195
uentarc[i... MF 80 arutielia MF 96
[---]nθia MF 212 arute MF 269
cicio MF 40
cicoi MF 40
iata MF 147
iata MF 158
ifra MF 40
? puponio MF 54
? malio MF 39
senθia MLF 362 iata MLF 362
ortecese MLF 339
cuicto MLF 310
cuiteneṭ MLF 361
upreciano MLF 363
upreciano MLF 364
senti Cap 399
sen.ti Cap 430
word-final urnam EF 1
eitam EF 5
*[0-2]e[1-3]tom EF 1
arcentelom hutị[.]ilom EF 1
duenom EF 4
propramom EF 2
[u]mom, umom, umọ[m] EF 2
ui[no]m EF 1
tulom MF 72 uino MF 59-60
arute MF 269
larise MF 270
? tuconu MF 85
larise MF 371
larise MF 372
[fel]ịcinatiu LF 384
mạcistratu LF 242
ac ̣iuaiom Cap 465
word-final f[.f]ịqod EF 1
cluster *[3-4]*ad EF 1
cupat MF 146 cupa]nt MF 80
cupạṭ MF 158
aruz MF 257
zot MLF 285
cupat LF 223
Fig.3.5. Omission of syllable- and word-final nasals in Faliscan texts.
91
CHAPTER 3
A special case are the instances where /r#/ is represented by -d. This occurs
three times in Late Faliscan, twice in [.]a protacio m f mạcistratu | keset cụestod
pi pretod pis LF 242 and once in la cotena la f pretod de | zenatuo senten-
tiad uootum | dedet LF/Lat 214, while Latin has a parallel in opeinod deuincam ted
CIL I2.547. In the cases of LF/Lat 214 and CIL I2.547, this was often ascribed to a
saṇdhi development of /r#d/, but LF 242 (published in 1990) now shows that this ex-
planation is difficult, if not impossible. Neither does it seem likely, as has sometimes
been assumed, that it is simply an error.
Combined with the omission of /r#/ and /r.C/, I think it likely that /r/ was real-
ized at least in word-final position, and probably also in syllable-final position as a
rather weak sound, probably an alveolar flap [ ], as Peruzzi (1997:64-5) suggests (al-
though I do not adopt his suggestion that this is due to Umbrian influence). In the
written form of the word, this flap could then be represented by the etymologically
justified -r, by the phonetically close -d, or be omitted entirely.
Another interesting solution has been proposed by M. Mancini (2002:40), who
rejects Peruzzi’s idea of an alveolar flap, and suggests instead that the spelling with -d
was meant to represent the Latin realization of /r#/, which presumably was ‘stronger’
than the Faliscan realization (although I cannot see why this spelling should be an
instance of code-switching between Faliscan and Latin). This would tie in with the
possibility that there were likewise attempts to render the realization of Etruscan /s#/,
which was apparently stronger than the Faliscan realization (§3.5.7d): see §9.2.2.1.
(c) Occlusives. The omission of occlusives can be divided into two cases: omission of
syllable-final /k/ before a dental and word-final /t/ and /d/.
Omission of syllable-final /k/ before a following dental can be observed in lete
MLF 285, let MLF 361, and in sesto LF 329 (beside zextoi LF 330 from the same
tomb), perhaps also in fita EF 1 (cf. §5.3.2.1) and possibly in cuiteneṭ MLF 361, if
this should be connected with the name Quinctus (which, however, appears in cuicto
MLF 310. Omissions as in lete have only one parallel in Latin, uitoria CIL I2.550
from Praeneste, but on the other hand were common in Umbrian, where they were due
to a development */kt/ → /xt/ (cf. Meiser 1986:92-3, 179). It seems not unlikely that
a similar development was responsible for the Faliscan and Praenestine forms. In that
case, cavies  uhtav[---] Etr XLV from Lucus Feroniae, which has been regarded as
showing Sabellic influence in uhtav[---], may represent a local pronunciation.
Word-final occlusives were something of a rarity in the Italic languages: the
only ones regularly encountered are /d#/ and /t#/: of these, /d#/ disappeared in Latin
during the first half of the third century (after long vowels), and this disappearance is
observable also in Faliscan. In the case of the ablative in -d and the adverbs derived
from such ablatives, /-d#/ is written out in the Early Faliscan ablative pramọḍ, pra-
mod, propramod EF 2 and in Middle Faliscan, foied MF 59-60, but ifra MF 40 ←
*/enf(e)rā(d)/. The forms sententiad and rected in LF/Lat 214 are archaisms, compa-
92
PHONOLOGY
rable to Latin sententiad, oquoltod, poplicod CIL I2.581,8, 10, 15 (186). The secon-
dary ending of the third singular was originally also /-d#/, but this was later replaced
by /-t#/ (§5.2.4e). In Faliscan, this is attested in the Early Faliscan forms porded EF 1
and fifiked EF 9 and Middle Faliscan faced MF 470*, and on the other hand Middle
Faliscan facet MF 471* and Late Faliscan keset LF 242. This replacement had been
ascribed to the disappearance of /d#/, but this seems hardly likely, since its replace-
ment, /t#/, could also be omitted: at best, the disappearance of /d#/ may have been a
factor that contributed to the replacement. Latin forms that show omission of /-d#/ are
kapia CIL I2.1, dede CIL I247b, 377, 380, 477, 2438 (although in some of these forms
it may be /-t#/ rather than /-d#/ that is omitted). A special case is Middle Faliscan met
MF 470* (cf. Early Faliscan med EF 1, 9). This form may point to a loss of voice in
/d#/, but is in my view perhaps rather a (‘hypercorrect’) form influenced by the im-
mediately following facet, which is the earliest Faliscan form that shows the replace-
ment of the secondary ending /-d#/ in the third singular perfect, indicating that /d#/ in
/mēd/ had already been weakened or dropped.
In the ending of the third singular, /-t#/, and of the third plural, /-nt#/, the /t#/
could regularly be dropped: (1) cupat MF 40, cupa MLF 305, cupa LF 221, and proba-
bly cup[a] MF 161 vs. cupat MF 220, cupat LF 224, and probably c]up[at] MF 159;
this has also been suggested in the case of i*ice LF 309, 315, but I very much doubt
that this is a verb. The forms cu[pa] LF 226 and cuba LtF 326 show that in the ending
of the third plural, both /n/ and /t#/ could be dropped. Such forms are also known from
Latin: probauero CIL I2.22, dedero CIL I261, dedro CIL I2379, probaru CIL I21635,
courauro CIL I22542.47
(b) Sibilant. The omission of -s was a stock feature of Middle and Late FaliscanFalis-
can, and can in fact be regarded as one of the better known features of Faliscan. As said
in §1.4.4, however, it is also a feature of Latin in general, and cannot in itself be used as
a certain sign that the inscription in which is occurs is Faliscan.
In the Early Faliscan inscriptions, -s is always written out and never omitted: zex-
tos EF 1, praụ[i]os EF 1, lartos EF 6 and apolonos EF 10, ceres EF 1, fitaidupes EF 1,
and (after a long vowel) titias ... duenas EF 3. However, from the Middle Faliscan pe-
riod onward, there is an overwhelming tendency to omit -s. The number of instances is
so large, in fact, that it is not practical to present all instances in tabular form, as I have
done in the other cases of omission that are discussed in this section. Omission of -s
after a short vowel occurs in 107 inscriptions: MF 12, 19, 39, 40, 48, 49, 50, 51, 52, 54, 56,
79, 80, 88, 89, 90, 94, 95, 97, 100, 105, 137, 141, 142, 143, 146, 147, 148, 149, 151, 152, 153,
159, 162, 164, 175, 178, 181, 193, 195, 197, 259-260, 265, 272, 275, 276, 376, 470*, 473*;
MLF 206, 210, 211, 305, 307, 308, 309, 310, 312, 315, 316, 318, 319, 324, 346, 348, 349,
47
Latin also has forms of /nt#/ where the /t/ is dropped, but not the /n/: coraueron CIL I2.59,
curarun CIL I2.1616, dan CIL I2.1618, dedron CIL I2.30, tabificanque CIL I2.2540c.
93
CHAPTER 3
350, 351, 352, 354, 355, 358, 359, 360, 363, 364, 464, MLF/Cap 476*; LF 213, 220, 223,
224, 225, 226, 228, 231, 242, 244, 329, 330, 331, 332, 333, 335, 336, 378; LtF 171, 215, 223,
290, 341, 377; Cap 388, 391, 420, 433, 435, 462. These yield a total of 164 cases of omis-
sion in the second-declension nominative singular ending -o(s), 8 instances in the third-
declension genitive singular ending -o(s), 2 instances in the third-declension genitive
singular -e(s) (which could potentially represent /-ē ̣s/ ← */-es/, see §4.5.2), and 1 in-
stance in the third-declension nominative singular ending -e(s) (§4.5.1.5) or 175 in-
stances in all. Against this, there are only a 3-6 instances from 2-5 inscriptions of -s
being written out: cauios frenaios MF 471* and cauios LF 382, and perhaps also
pạrtis MF 79 (this may be an abbreviated or syncopated form, and occurs side by side
with leiuelio), and in two instances from the ager Capenas, írpios Cap 389 and pacios
Cap 392, where Sabellic influence is clearly present (§9.3.2). This means omission of
-s after a short vowel outnumbers the writing out of -s in the same position by 29 to 1:
in other words, the cases where -s is written out after a short vowel are c.3% of the
total. After a long vowel, however, the instances of omission are noticeably fewer: cra
MF 59-60, and efile MF 116 (if this is not simply an error for efile‹s›), and probably
canumede MF 62, vs. efiles MF 113, efiles MF 115, efil]es MF 117, and aiedies Cap
390. A distinct category are the Etruscoid forms in -(i)es (§9.2.2.2c), where the -s is
usually written out: ḷ[oc]ịes MF 12, fasies MF 41, ulties MF/Etr 64, calitenes and ce-
sies MF 265, uenelịes MF 258, petrunes LF 226, plenes LtF 231 and morenez MF 269
and ạcṛẹẓ MF/Etr 67, vs. satelie MF 42, cesịe MF 257, and tulie LF 383, and perhaps
acre MF/Etr 279, ame MF/Etr 280, ame MF/Etr 282. This may imply that in Etruscan
forms and names the /s#/ was realized ‘more strongly’ than in Faliscan (§9.2.2.1).
Omission of word-final -s occurs with quite some frequency also in Latin, as the
indices to the CIL I2 or to Wachter 1987 show: note also the weakness of /s#/ in Repub-
lican Latin quantitive poetry. There is no single area within Latin in its broadest sense
where the frequency of the omission is as high as in Faliscan: even in the ager Capenas
the frequency is much lower (omission in Cap 388, 391, 420, 433, 435, 462 vs. írpios Cap
389 and pacios Cap 392). In spite of this, I very much doubt whether the omission of -s
was phonemic, that is, that e.g. the second-declension nominative singular ending had
become /-o#/ in Middle Faliscan: even during the Middle Faliscan period it was possi-
ble to write -os, as in cauios frenaios MF 471*. I assume that the omission of -s repre-
sents a realization as a weak glottal sound, [h] or [ ], much like syllable- and word-final
/s/ is realized in Andalusian Spanish (e.g. estamos /estamos/ [e tamo ]). The great num-
ber of instances of omission of /s#/ may in fact have been due to an orthographic con-
vention: it may have been the case that in the ager Faliscus, the convention was not to
write this weak sound, while in Latium the rule appears to have been to write it as -s in
spite of its weak realization. This is borne out by the fact that Latin inscriptions from the
ager Faliscus generally have -us as the ending of the second declension nominative sin-
gular, with -u appearing only in Lat 251: see §3.6.6.1.
94
PHONOLOGY
3.6. The Faliscan vowels
3.6.1. The vowel system of Faliscan. The vowels that Proto-Latin inherited from the
final stage of Proto-Italic are /a e i o u/ and /ā ē ī ō ū/ (cf. Meiser 1998:55-7, 60-2). Of
these, the short vowels /e i o u/, and in some cases /a/, were normally inherited from
PIE: the Proto-Italic developments of the laryngeals added new /a/’s (§3.2.3), and the
Proto-Italic developments of */ / added new /e/’s, /o/’s and /u/’s (§3.2.7), and the
Proto-Latin development of */ / added new /e/’s (§3.3.2). The long vowels are
generally due to Proto-Italic developments of */VĦ/ and */RĦ/ (§3.2.3) and early
contractions after the loss of intervocalic // (§3.2.6), although in a few cases the long
vowel goes back to a lengthened PIE vowel, as in */pħ2tēr/, */mātēr/, */ħ3rē -s/.
Insofar as this can be ascertained, the vowel system described here was also the
one found in Early Faliscan. It should be noted that in Faliscan and Capenate inscrip-
tions the quantity of the vowel is not indicated anywhere: this is only done in uootum
LF/Lat 214, and uiitam and aastutieis Lat 217 in two Latin inscriptions from Falerii
Novi. The quantities of Faliscan vowels can therefore only be inferred (a) from ety-
mologies, (b) from developments that affected only long or only short vowels, (b)
from corresponding words or names in other Italic languages where the quantity of the
vowel is known, either because it is expressed in writing or because it can be estab-
lished in other ways (e.g., in the case of Latin, through quantitative verse).
During the Middle Faliscan period several changes took place that not only changed
individual vowels, but also affected the Faliscan vowel system itself.
As a result of the Middle Faliscan monophthongization of the diphthongs (§3.7),
three new long vowels were added: (1) /ē ̣/, both from /e/ and from /e/ ← /o/
(§3.7.5), (2) /ę̄/ ← /a/ (§3.7.6), and (3) /ō ̣/, both from /o / (including /o / ← Proto-
Italic */e /, §3.2.5) and from /a / (§3.7.2, §3.7.4). These new vowels must have had
phonemic status, cf. such (unattested) pairs as */fīliē/̣ ‘sons’ and */fīlię̄/ ‘daughters’.
The emergence of these long vowels also changed the vowel system as a whole by
introducing more degrees in openness, especially in the front vowels.
Among the short vowels, a new sound /ẹ/ appears (§3.6.2). This is most clearly
observed in the tendency to spell antevocalic /i/ as e, and /e/ before /r.C/ as i. Al-
though there are therefore two different spellings, these in all probability represent the
same sound, antevocalic e expressing ‘an /i/ that is more open than a normal /i/’ and i
before /r/, ‘an /e/ that is more closed than a normal /e/’. That this sound is not merely
a positional variant, but must have had phonemic status is shown by the fact that there
are instances of e for i and i for e in other positions as well, showing that in Faliscan
/e/ and /i/ were merging into /ẹ/. There may have been similar merger of /o/ and /u/,
but the indications for this are few (§3.6.3).
95
CHAPTER 3
i ū
i
u
u
ə? ?
ē e
o ō
e o
ă
ā
a
Fig. 3.6. The Middle and Late Faliscan vowel system.
The dotted circles accompanied by a vowel sign in bold case represent the letter by which the encircled
sounds could be represented. For digrams to denote /ē ̣/, /ę̄/ and /ọ̄/, see §§3.7.2-6. For the merger of /i/
and /e/ into /ẹ/, see §3.6.2; for the possible merger of /o/ and /u/ into /ọ/, see §3.6.3. For ə, see §3.6.6.
The Faliscan vowels are represented by the signs a e i o u, as follows (cf. fig.3.6):
x a represents (1) /a/ as in far EF 1 (← PIE */a/?), pater MF 62 (← PIE */ħ2/), ar-
centelom EF 1 (← PIE */ħ2r/); (2) /ā/ as in mater LF 221 (← PIE */ā/?), karai (←
PIE */eħ2/), pramo- EF 2 (← PIE */rħ/).
x i represents (1) /i/, as in pipafo MF 59; (2) /ẹ/ ← /e/ as in loifirtato MF 32 (see
§3.6.2); (3) /ī/ as in uino MF 59-60 (← PIE */iH/);
x e represents (1) /e/ as in eqo EF 1, arcentelom EF 1 (← PIE */ /); (2) /ẹ/ ← /i/ as
in fileo MF 471* (§3.6.2); (3) /ē/ as in rex MF 90 (← PIE */ē/); (4) /ē ̣/ ← /e o/
(§3.7.5); (5) /ę̄/ ← /a/ (§3.7.6);
x o represents (1) /o/ as in euios EF 1 (← PIE */o/); (2) /ō/ as in carefo MF 59 (←
PIE */oĦ/); and (3) /ō ̣/ ← /a o / (§3.7.2, §3.7.4);
x u represents (1) /u/ as in cupat MF 40 (← PIE */u/); and (2) presumably /ū/ (←
PIE */uĦ/ (no attestations); (3) u is also used where o would be expected (§3.6.3)
The signs i and u are also used to denote the second element of the diphthongs ai ei
(oi is not attested) and au ou respectively.
Beside these vowel notations, there are a few Capenate instances of the Sabellic signs
í and ú, representing (Sabellic) /ẹ/ and /ọ/ respectively: see §9.3.2.
96
PHONOLOGY
3.6.2. Middle and Late Faliscan variation of i and e. As said in §3.6.1, there are a
number of instances where there is variation in the spelling between e and i. The
clearest instances are those where antevocalic i is written as e. (For the cases where
antevocalic e represents a monophthongized diphthong, see §3.7.7).
x fileo MF 471*, hileo MF 146, filea MF 14
x fulczeo LF 329, folcozeo LF 330, *olcuzeo LF 332 vs. folcosio LF 333
x ḥiṛṃeo MF 19 vs. hirmia MF 18, firmio MF 54, fir mia MLF 302, hirmio LF 213
x zu[con]|eo MF 56 vs. zuconia MF 271,
x *(*)coṇẹo LtF 290 and peṭṛọṇẹọ MF 473* (cf. the regular derivation of Etruscan
gentilicia in -u : Latin -onius)
x zertenea LF 221 (cf. Latin Sertinius)
x uecineo LF 220, 224, 225, uecinea LF 223, uecin[e]a LF 221 (cf. Latin Vicinius)
x anileo or manileo MLF 355 (cf. Latin Annilius or Manilius) and hlau|elea LtF 325
(cf. Latin Flauilia)
x ? pauiceo MF 12
x unclear are [---]leo MF 153, [---]l[e]a MF 155, [---]oxie[.]eai MLF 310, [....]nea
LtF 301
x in the patronymic adjectives iu]nẹo MF 141, ịuneo MF 151, iuneo LF 220 and
uolteo MF 275, uolθeo MF 276, the e probably stands for [ẹ] not for [ę̄] (§7.5.2)
x the gloss Struppearia (§6.6.6) may represent the same tendency.
The, reverse, i for antevocalic e, is found in
x clipịaṛ[io LF 230 (vs. clipeaio (clipea‹r›io?) MF 470*, clipeaṛ[io LtF 231, and
cḷ[i]peario LtF 233, cf. Latin clipeus/clupeus).
The phenomenon thus occurs both during the Middle and Late Faliscan periods, and
also occurs in Latino-Faliscan (hlau|elea LtF 325, and probably *(*)coṇẹo LtF 290).
Interestingly, there are no instances from the Capenate or Latin inscriptions, with one
clear exception, namely the spelling Feronea for Feronia in all instances of this name
in the dedications from the shrine at Lucus Feroniae: feroneai Cap 434, fe]ronea Cap
431, feronea Cap 433, and fero]ṇea Cap 437. This may have been a local custom,
especially since antevocalic i is never written as e in other Latin inscriptions from the
area.
Although these instances can be explained as /i/ having a more open realization
[ ] before a vowel, comparable to e.g. Latin fileod CIL I2.2658 and fileai CIL I2.52,
they may be part of a larger tendency, since there are instances where e occurs instead
of an expected i in other positions than antevocalic, in zertenea LF 221 (instead of
zertinia) and upreciano MLF 363, upreciano MLF 364 vs. u]mpricius Lat 219, and
f||e LF 332 = fe(leo). Wachter (1987:126, 488-9) regards such instances of e for i as
indicative of a generally more open pronunciation of short /i/ as opposed to /ī/.
97
CHAPTER 3
The ‘reverse’, i for original /e/, is found, apart from the already named clipịaṛ[io
LF 230, only in loifiṛtato MF 31, loifirtato MF 32, and loifirtạ MF 41. Here, however,
the i represents a more closed /e/ before /r.C/, as in Latin stircus CIL I2.401 and
mirqurios CIL I2.553 (cf. Meiser 1998:81): in these cases, therefore, the tendency to
realize /e/ as [ ] is a conditioned variant and not free as in the cases described earlier
in this section.
Another instance may be the name Firmius in ḥirmia MF 18, ḥiṛṃeo MF 19,
firmio MF 54, hirmio LF 213, fir mia MLF 302. Although this name is consistently
spelled with i, it may be related either to (a) the Etruscan gentilicium Hermana, oc-
curring in the area in hermana MF 264 (cf. her Etr VI-VII) or (b) to the adjective
firmus, where the i reflects an older /e/ (*/fermo-/ ← Proto-Latin */ ermo-/ ← PIE
*/dhermo-/) and Latin apparently standardized the variant with the closed form of the
/e/.48 Counterexamples are loferta LF 222, and mercui MF 113-126, zertenea LF 221,
tertineo LF 213, tertinei MF 473*, fertrio Cap 391, and se(r)torio MLF/Cap 476*.
Other instances are hard to distinguish, in view of the fact that the material is
onomastic rather than lexical: it is therefore hard to tell whether a name like pupelio
MF 149 represents Pupilius (in which case it would be an instance of variation be-
tween e/i, but could also be interpreted as a case of a ‘non-reduced’ vowel, cf.
§3.6.6.1) or Pupellius (in which case it is not an illustration of a variation between
e/i).
Probably Sabellic rather than Latin-Faliscan is írpios Cap 389 and the name of
the Hirpi Sorani known from literary sources (cf. §2.3.4, §6.6.5), probably derived
from a Proto-Sabellic */hẹrpo-/ ← Proto-Italic /xerk o-/ (← PIE */gher(s)k o-/?). The
instances where the variation e/i may be due to vowel-reduction in unstressed medial
syllables are discussed in §3.6.6.
3.6.3. Middle and Late Faliscan variation of u and o. Beside the variation between
e and i described in the preceding sections, there are cases of variation between o and
u. This tendency, however, is much less clear: in a number of cases the variation is
due to influence from Etruscan orthography, and it does not only involve /o/ but /ō/ as
well. The instances where the variation o/u may be due to vowel-reduction in un-
stressed medial and final syllables are discussed in §3.6.6.
In some cases, the o is simply a preservation of the /o/ while in Latin this (later)
changed into /u/. Examples are zot MLF 285 and perhaps sot LtF 172 vs. Latin sunt
(still sont CIL I2.1529), and the name Publius/Publia, which in Faliscan is always
written with o (13 attestations, see §7.7.1.50), as in older Latin, e.g. popliosio CIL
I2.2832a (cf. §18.3.3.1).
48
Cf. perhaps Umbrian ferim-e TI III.16 and Oscan fí]r[í]mens Po 2 (cf. WOU s.vv.).
98
PHONOLOGY
The spelling u for /o/ mostly appears in contexts where orthographic influence
from the o-less Etruscan alphabet can be assumed. Thus e.g. cutri MF 200 vs. Latin
Cotrius, puponio MF 54 vs. Pomponius (but cf. §7.7.1.51) and on the other hand orte-
cese MLF 339 vs. Etruscan urtcsnas Etr XXXV, polafio (=pol‹fa›io) MLF 354 vs.
Etruscan Pulfna. Etruscan gentilicia in -u are usually adapted to -ōnius, as in Latin
(§7.8.2.4), e.g. zuconia MF 271 and zu[con]|eo MF 56 (and perhaps also su|[con ---]
MF 191) vs. larisa zuχus Etr XXXII): for further instances, see §7.8.2.4. G. Giacomelli
(1963:84) in fact interpreted cicoi MF 40 as a rendering of an Etruscan *Cicui, the
feminine form of Cicu.
The anaptyctic vowel in the Etruscan name Arnθ is once written as o in aronto
MF 265: elsewhere it is always written as u in aruz MF 257, aruto MF 257, 267, arute
MF 269, and the derivations arutielia MF 96 and arutlo MF 195. The spelling else-
where is also u, except in Oscan arrontiej tLu 1. This may be due to words like zo(n)t,
where /ont/ could realized as [õt] and perhaps even already as [ t], but the correct
(historical) spelling was o(n)t: this spelling may have been transferred (as a hypercor-
rect spelling) to the /u/ in /arrunt-/, where u was the historically correct spelling.
The use of u for /ō/ in petrunes LF 226 vs. peṭṛọṇẹọ MF 473* is probably best
explained by an Etruscan intermediary (cf. also the Etruscoid ending -es, see
§9.2.2.2c). However, Wachter (1987:412-3) points to the fact that in Latin inscriptions
from the area of the Lacus Fucinus, /ō/ is represented by u several times: dunom VM
3, semunu Pg 9, ptruna Pg 52, apunies Pg 53, peumpuni Pg 26, fadatruni MV 2, uic-
turie CIL I2.2486. The origin of the u in these names may therefore go back further, to
the original Sabellic form. Another such case is the exceptional Latin terebuni CIL
I2.312 vs. Trebōnius (cf. Wachter 1987:187).
A difficult case is fulczeo LF 329, folcuso LF 331, and *olcuzeo LF 332 vs. fol-
cozeo LF 330, folcosio LF 333, and holc[osi] LtF 140, where there is variation both in
the first and in the second syllable, and there is no equivalent name in Etruscan or
Latin: the only closely related names are Etruscan Hulχena and Hulχnies, and Latin
Holconius (§7.8.1.64). Both u’s are probably best ascribed to Etruscan influence in
spelling. The u in the first syllable could conceivably be ascribed to a Late Faliscan
closing of /olC/ → /ulC/, but this closing is not attested for Faliscan (§3.3.4.1). The u
in the second syllable was in all probability long (cf. Latin names in -ōsius/-ōrius and
-ōnius), and can therefore not be ascribed to reduction of a medial vowel: its omission
in fulczeo LF 329 must therefore be due to an error (fulc‹u›zeo or fulc‹o›zeo), not to
syncopation.
Unclear is the relation of the Faliscan gentilicium pupelio MF 149, pu]peḷ[i---]
MF 150, pup[elio MF 151 to Latin popil[i] Lat 295, popili Lat 296, popili Lat 478* (all
imports): the u could be due to Etruscan influence, but at least in Latin there appears to
have been a quantitative difference between the gentilicia in Pŏp- and those in Pūp-
(Schulze 1904:213), and the two gentilicia may therefore be unrelated. The same is
99
CHAPTER 3
true in the case of pupiias MLF 304 (praenomen or gentilicium?) and the abbreviated
gentilicium popi Cap 420.
Finally, posticnu MLF/Cap 474* may be a rendering of a Sabellic first-
declension nominative singular *posticnō (see §9.3.2).
3.6.4. Variation between i and u. In Latin, there is a tendency to variation between i

and u in pairs such as libet/lubet, clipeus/clupeus (in later centuries also spelled cly-
peus), probably pointing to a realization [ü] (Meiser 1998:80). There is no sign of this
variation in Faliscan, where the gentilicium Clipearius (which apparently does not
occur elsewhere, §7.8.1.42) is always spelled with i (clipeaio (=clipea‹r›io?) MF
470*, clipịaṛ[io LF 230, clipeaṛ[io LtF 231, cḷ[i]peario LtF 233). Faliscan does show
a related development, however, in the development of the diphtong /o / to /o/ in
loifiṛtato MF 31, loifirtato MF 32, and l]oifirtạ MF 41 (see §3.7.3).
3.6.5. Anaptyxis. As in Latin, there are very few instances of anaptyxis: a difference
from Sabellic languages like South Picene, Paelignian and Oscan, where its occur-
rence was regular. One instance is the Etruscan name Arnθ. In Etruscan, this name is
occasionally spelled with an anaptyctic u: arunθ Ar 1.7, arunθia Vt 1.73, Ar 1.53,
arunθial AS 1.236, arunθia[ Ar 0.3, arutia Pe 1.846 (Latin alphabet), aruzinale OA
0.1, arunzina Cr 6.2. In the Italic languages, the name only occurs in its anaptyctic
form, due to the phonotactic impossibility of a cluster /rnt#/ or /rnts#/ → /rns#/: in
Latin, as Arruns, in Oscan in the gentilicium arruntiis Po 58, arrontiej tLu 1. In Falis-
can, the name occurs as aruz MF 257, aruto MF 257, 267, arute MF 269, and the
derivations arutielia MF 96 and arutlo MF 195, and once as aronto MF 265 (§3.6.3).
Another instance may be Vetter’s interpretation of putellio MF 152 if this is derived
from */pu-tlom/: see §63.62. In the gloss cenaculum (§6.6.1) the spelling with an an-
aptyctic vowel probably reflects the regular Latin spelling of this word.
3.6.6. Developments of vowels due to stress. The change from the PIE tonal accent
to a (heavy) initial stress in all likelihood occurred during the Proto-Italic period (cf.
Meiser 1998:66). The developments of the vowels due to this initial stress are of later
date, however: both the weakening of vowels and the syncopation in medial and final
syllables do not seem to have taken place before the sixth century and to have been
completed by the fifth (Rix 1966, Meiser 1998:66-7).
There are hardly any indications of vowel-weakening or syncopation in Falis-
can, and Faliscan is indeed notorious for not having vowel-weakening (a point that is
repeated from author to author). This idea, however, is mainly based on the form cun-
captum in LF/Lat 214: this inscription was discovered very early in the history of Fal-
iscan studies and attracted considerable attention at the time, but it is unclear in how
far it actually represents contemporary Faliscan. That there has never been a serious
100
PHONOLOGY
attempt to contradict this view is due to the scarcity of material that could shed some
light on the subject. Early Faliscan inscriptions pre-date the developments that can be
ascribed to stress, and Early Faliscan forms like pe para[i EF 1 corresponding to Latin
peperi are therefore in no sense an argument against (or for) Faliscan having ever had
weakening. In view of the fact that both Latin and the Sabellic languages (and, for
those willing to accept an Italic Sprachbund, Etruscan) show extensive evidence of
various forms of weakening and syncopation in medial and final syllables, it would a
priori be very surprising if Faliscan did not show features of such developments.
(1) Weakening of vowels. In the Middle Faliscan material, there are simply very little
forms where weakening could be expected at all. Only a small part of the material is
lexical, and here there are simply no the instances are:
(a) maximo MF 89 beside max]om[o] MF 90, ṃaxoṃ[o] 98 and [m]ano[m]o MF 149
(man[o]mo MF 80 and [---]a*ome MF 156 (if interpreted as m]ax̣ome) cannot count
as counterexamples). Unfortunately, MF 89 is known only from an apograph: if it is
indeed attested, it shows a reduction of the original medial vowel, as in Latin.
(b) harịṣp̣[ex LtF 231 and harisp[ex LtF 232: in this case, the form may be due to
weakening /u/ → /i/ that is also found in open syllables in Latin (Meiser 1998:68).
The form can also be explained as formed after compounds, e.g. the priestly title pon-
tifex, especially as Latin haruspex shows no reduction of the vowel.
(c) pipafo MF 59, ‹pi›pafo MF 60, an athematic laryngeal stem, would be expected to
have /a/ as a result of the laryngeal developments: in Latin, this /a/ was this was re-
duced in medial syllables: cf. bibĕre, dăre and dăbo beside reddĕre. If in the Faliscan
forms the -a- represents /a/, it shows no reduction, having perhaps been preserved due
to a paradigmatic analogy: there are other explanations of this form where the a is
regard as /ā/ (§5.3.1.13).
The remainder of the material is onomastic, where it is difficult to establish whether
the forms with medial i show a reduction of another vowel or not (cf., for instance,
pupelio MF 149, which represents either Pupellius or an ‘unreduced’ Pupilius). Ex-
cluding these instances, as well as the instances where the name is patronymic and
formed with a suffix -ilius or an Etruscan name adapted to a Latin-Faliscan -inius, this
leaves only the possibility that ueicono MF 88 is connected with uecineo LF 220 etc.
and with [u]eculia MF 88. It seems unlikely that such a connection existed, however.
In the Latin inscriptions, a lack of reduction appears to be attested in cuncaptum
LF/Lat 214 (mid-second century) but this form may represent a learned reconstruc-
tion. The same can be true of falesce Lat 218 (late second century) which may be re-
constructed after Falerii, and of the e in the gloss decimatrus (§6.6), where some MSS
have decematrus, which may have been reconstructed after decem. The only clear
forms that can be interpreted as instances of vowel reduction in the area are Latin
aciptum, quolunda, and saipisume in Lat 217 (late second century).
101
CHAPTER 3
Weakening or reduction of vowels in final syllables, however, is well-documented,

since it affected several nominal endings.
The best documented instances are those of /o/ in the endings of the second-
declension nominative singular /-os#/ and the second declension accusative singular
and neuter nominative singular /-om#/. Here, the Middle and Late Faliscan inscrip-
tions generally have an unreduced -o, or, rarely -os. The exceptions are the genitive
plural [fel]ịcinatiu LF 384 (beside ac ̣iuaiom Cap 465 and probably tulom MF 72),
which shows that the /-ōm/ had been shortened to /-om/, and the unexplained tuconu
MF 85. Beside the instances of /o/, Faliscan also has a genitive in -e(s) in [---]fate
MLF 285, which may represent a non-reduced /-es/ (§4.5.2).
The Latino-Faliscan and Capenate inscriptions likewise generally have nomina-
tives in -o: the instances are [---]ilio LtF 215, clipeaṛ[io LtF 231, cḷ[i]peario LtF
233,*(*)coṇẹo LtF 290, ṃunio LtF 377, genucilio Cap 434 and donom Cap 430. The
exceptions are the verbal form esú Cap 389, 404, 465, where ú probably represents the
Sabellic vowel /ọ/, the unexplained cl · anu or clanu Cap 397, which may be an ab-
breviation, and the unclear murụ[?---] LtF 173.
The oldest Latin inscriptions from the area also still have -os in loucilios Lat
268 (fourth century) and fourios Lat 216 (c.220?). From the late third century on-
wards, however, they generally have -us: spurilius Lat 237 (after c.240), spurilius Lat
238 (after c.240), calpurnius Cap 432 (before 211), [---]rcius Cap 435 (before 211),
egnatius Lat 291 (second century), lectu(s) (twice) and amplius Lat 251 (late second
century), latrius Lat 218 (c.125), u]mpricius and [?]aburcus Lat 219 (c.120-50), ful-
uius Lat 250 (106), didius and uettius Lat 455 (c.100-50). Beside this there are the
forms in -um in sacru(m) and cuncaptum LF/Lat 214 (mid-second century),
gonlegium, aciptum, and opiparum Lat 217 and donum Lat 218. (Dono Cap 431 and
dono[?] Cap 433 may be a datives in /-ō/). The only exception is zenatuo LF/Lat 214,
which may be illustrative of the tendency found in Latin to preserve /o/ after /u/, / /,
or /k / longer than elsewhere (Meiser 1998:84).
This picture seems straightforward, but the problem is that this is a case of the prob-
lem that was touched upon in §3.1.1. The distinction between Latino-Faliscan and
Capenate on the one hand and Latin on the other depends upon whether or not the
inscriptions in question show dialect features that are compatible with those attested
for Faliscan. In several cases, the ending of the second-declension nominative is in
fact the only feature in the inscription on which this distinction is based.
(2) Syncopation. The only clear instance of syncopation in a Faliscan word is:
x maximo MF 89, max]om[o] MF 90, ṃaxoṃ[o] 98, where the original form is
probably */mag-isVmo-/ (cf. Cowgill 1970:125).
The other instances all appear to be irregular:
102
PHONOLOGY
x arutlo MF 195 must represent a syncopated form aru(n)t(e)lo of a name like Ar-
runtulus: it may also represent an orthographic shortening.
x reiclio MF 99 (cf also reic ̣[lio] MF 98 and reị[cli.] MF 100) is most likely a syn-
copated form for reic(i)lio = Raecilius (§7.8.1.132).
x uisni MF 82 is a syncopated form for uis(i)ni or uis(e)ni if this name is connected
to Visinius (§7.8.2.177).
x neln LtF 300 (and nel[n---] LtF 299) with the surprising cluster /ln/ is probably
due to syncopation. What the non-syncopated form would have been is unclear.
x [---]lnia MF 146, although fragmentary, may be an instance of syncopation.
x [---o]stro LF 244 and [---]ostr[o LF 245 may contain a syncopation and represent
...ost(e)ro: see also §6.2.94.
There are two cases where a vowel is omitted that may have been long, namely fertrio
Cap 391 for Fertōrius and fulczeo LF 329 (beside folcosio LF 333, folcozeo LF 330,
folcuso LF 331, *olcuzeo LF 332) for Folcōsius. In these cases the omission of the
vowel may present a graphical contraction or an error rather than syncopation.
All in all, the instances of syncopation in internal syllables appear to show that the
phenomenon did occur in Middle and Late Faliscan, but irregularly. 49
Syncopation of short vowels in final syllables occurred as a regular development in
Proto-Sabellic (Meiser 1986:59-60) but not in Latin or Faliscan. The one Faliscan
instance is partis in leiuelio pạrtis | uolti MF 79, where the ending appears to show
not only a syncopation, but also combines this with a very rare preservation of -s
(§3.5.7d). The gentilicium is without parallels, and it may be that the form is incom-
plete. This syncopation (if it is one) can be compared to the incidental Latin instances
uibis CIL I2.552, mercuris CIL I2.563 and caecilis CIL I2.1028.50
Resuming the point that was made at the beginning of this section, I believe that there
are indications that Faliscan did have stress-related treatment of vowels: if weakening
is attested only dubiously, it is clear from the syncopation that stress did affect the
vowels. Material for this in internal syllables is rare. The fact that the clearest in-
stances of stress-related vowel-weakening occur in Latin inscriptions from the area is
of course not a ground to assume that the phenomenon was due to Latin influence in
the area, but rather to the fact that the Latin inscriptions contain more lexemes than
the Faliscan inscriptions, which mainly consist of onomastic data.
49
In felic ̣inate MF 41 and [fel]ịcinatiu LF 384, Faliscan has Feliginas, while Latin has Fulgi-
nas and the toponym Fulginiae/Fulginium, where the u shows that the syncopated form had
existed early enough to have developed as /elC/ → /olC/ → /ulC/ (cf. §3.3.4.1).
50
Note that far EF 1 is not an instance of syncopation (which in a pre-fifth century text would
have been surprising in any case): Umbrian farsio TI VIb.2 etc. points to a Proto-Italic */βars/
(Meiser 1986:154, 174), since */βarVs-/ would have given *farfio (Schrijver 1991:113-4).
103
CHAPTER 3
3.7. The Faliscan diphthongs
3.7.1. General. At the end of the Proto-Italic period, the following diphthongs were
present in Proto-Latin: */a o a e o/, and at least two long diphthongs, */ā ō/,
which may already have been shortened everywhere except in the endings of the first-
and second-declension dative singular (cf. Meiser 1998:60): for the possibility of a
Faliscan diphthong /�/, see §3.7.8. In Faliscan, the short diphthongs were all mo-
nophthongized during the Middle Faliscan period, earlier than in (Roman) Latin,
where this happened during the late third and the first half of the second centuries.
A problem in the description of the Latin diphthongs is that there seems to have
been a considerable difference between rural and urban Latin, or rather, a difference
between rural and urbane Latin that was exaggerated by the Latin authors on lan-
guage. These regarded a diphthongal pronunciation as ‘correct’ and belonging to the
better layers of urban society, and associated the monophthongal pronunciation with
rural or lower-class Latin (cf. R. Giacomelli 1978:25-31, 2006:75-7). The degree to
which a monophthong was acceptable, however, appears to be very much connected
to when that particular diphthong was monophthongized: authors do not comment on
the monophthongization of /o / → /ū/, which was completed by the end of the third
century, and hardly on the monophthongizations of /o/ → /oe/ → /ū/ and /e/ → /ī/,
which belong to second century, but they do comment on /a/ → /ae/ → /ę̄/ and /a /
→ /ō ̣/ (see Blümel 1972:10-4, 28-9, Pfister 1977:63-4, 68-9, also Rocher 1928): what
they describe or prescribe is in fact the status quo of the early first century BCE.
3.7.2. The development of /o /. The diphthong /o / had its origins both in PIE */ou/
as well as in PIE */eu/, which merged with */ou/ during the Proto-Italic period: as I
said, I do not adopt Schrijver’s suggestion (1991:452) that Proto-Italic */e / and */o /
merged into a neutral diphthong that could be realized as [e ], [ə ] or [o ] (§3.2.5).
In the Middle Faliscan inscriptions, /o / is spelled both as ou and as o (for attes-
tations, see fig.3.7), which points to a monophthongization /o / → /ō ̣/ during the
Middle Faliscan period. The same monophthongization took place in Latin, where /ō ̣/
had developed further to /ū/ at least in Roman Latin probably by the end of the third
century: the earliest instance of the spelling u seems to be luciom CIL I2.9 (cf. Blümel
1972:32-3, Meiser 1998:59-60). There are no indications that /ō ̣/ developed to /ū/ in
Faliscan as well: Herbig’s (CIE 8225) and Vetter’s (1953:302) suggestion that Middle
Faliscan nut*[---] MF 103 was a form of nutrix was rightly rejected by G. Giacomelli
(1963:82-4). Peruzzi (1964d:312) assumed that nutṛ[---] was a Luxuslehnwort from
Roman Latin, but even in that case it would still be the earliest instance of u for origi-
nal /o /. Latino-Faliscan rụfi in LtF 292 is uncertain, and if to be read thus, reflects
second-century Latin.
104
PHONOLOGY
Early Faliscan has no attestations of /o /: Vetter’s l[o]ụf[ir (1925:27-8), lọuf[ir

(1939:156), and louf[i]ṛ (1952:280) in EF 1 are untenable (§12.2, see also §3.7.3).
There appear to be two problematic instances of this diphthong in EF 4, namely the
woman’s name rufia (read by Vetter (1939:151, 1953:258) as rọfia) and the man’s
name ofetios (cf. also ofete in the same inscription?), connected with the potamonym
Ūfens by Vetter (1953:286-7). These forms presuppose a monophthongization of /o /
already during the Early Faliscan period, far earlier than the other Faliscan and Latin
monophthongizations: note that this leaves unexplained why the diphthong could still
be written as ou in Middle Faliscan. Reading rufia also requires a development /ō ̣/ →
/ū/ that is not attested for Faliscan and even in Latin is not attested before c.200. If
rufia (or rọfia) is connected to rūfus ← Proto-Italic */ro ðo-/ ← PIE */ħ1reudho-/), it
can only be a misspelling for r‹o›ufia (or rọ‹u›fia). The alternative is to connect rufia
to the Proto-Italic */ruðro-/ ← PIE */ħ1rudhro-/) that underlies Latin rŭber, but then
the name would be *rufra or *rufria (or is rufia is an error for ruf‹r›ia?).51
In neighbouring languages, ou was monophthongized even earlier than in Faliscan
and Latin: South Picene has one instance of tútas TE.5 (beside toúta TE.7 and toútaih
RI.1), which dates from the sixth or the first half of the fifth century. In Umbrian and
Volscian, */o / had been monophthongized to o /ō/ before the time of the earliest in-
scriptions (Meiser 1986:122-4). In Etruscan, where the diphthong occurs almost ex-
clusively in (onomastic) borrowings from the Italic languages, it is first rendered as au
or av (cf. lauv|cies Etr XXIX from Civita Castellana) and from the early fifth century
onwards as uv (Rix 1984:205-6, Steinbauer 1999:39).
3.7.3. The development */lo β-/ → /loβ-/. A different development of /o / is found

in the word corresponding to Latin liber- (← PIE */ħ1le dhero-/). There are three Fal-
iscan attestations of the spelling oi, namely loifiṛtato MF 31, loifirtato MF 32, and
l]oifirtạ MF 41. These correspond to the Latin forms “a loebeso, Liberum” (Varro L
6.1) and “loebesum et loebertatem antiqui dicebant liberum et libertatem” (Paul. Fest.
108.5-6L): although these may be learned reconstructions (thus Blümel 1972:25), an
intermediate stage /o/ is in any case likely to explain */o / → /ī/ in Latin līber, līber-
tas, etc. The development is assumed to have been */o / → /o/ between /l/ and labial
(Pfister 1977:70, Meiser 1998:87),52 in all probability related to the variation, in Latin,
between u and i in pairs such as lubet/libet and clupeus/clipeus (cf. §3.6.4).
51
Cf. Umbrian rofu TI VIIa.3, rofa TI VIIa.6 with */rōf-/ ← Proto-Italic */ro ð-/, but rufrer
TI VIa.14 with */rufr-/ ← Proto-Italic */ruðr-/, and the Etruscan onomastic borrowings with
ruvf- reflecting */ro f-/ in Ta 1.216, 6.12, AT 1.4, 12, 49, 107, 108, 111, Vc 6.6, AV 6.2, and
with rufr- reflecting */rufr-/ AT 1.7, 8, 9, Pe 1.150, OB 3.2 (and ruvr- Etr XIX = Fa 3.2).
52
There appear to be no other words that show this development, cf. glūbo ← Proto-Italic
*/glo β-/ ← PIE */gleubh-/ and lūbricus ← Proto-Italic */slouβriko-/ ← PIE */sleubh-/: ap-
parently the development did not take place when the /l/ was preceded by another consonant.
105
CHAPTER 3
Interestingly, this development of */o / appears to have been restricted to Latin

and Faliscan: Paelignian has loufir Pg 11, and Oscan, lúvfreís Fr 5, and perhaps Sam-
nitic l?]úvfríkúnúss Sa 4, all with /o /.53 Paelignian lifar Pg 9 is unclear (cf. WOU
s.v.): it has been interpreted as a Latin borrowing ‘Liber’ (Vetter 1953:149), but more
plausible is the interpretation as a Paelignian passive verb equivalent to Latin
libet/lubet (thus e.g. Jiménez-Zamudio 1986:43).
In view of this difference between Latin and Faliscan on the one hand and the
Sabellic languages on the other hand, the spelling o in Faliscan loferta LF 221, and
possibly also in lo[---] MF 155 ? and lo MF 165 is surprising. R. Giacomelli’s hope
(1978:29) that loferta could be shown to be earlier than l]oifirtạ seems unlikely to be
fulfilled, as loifiṛtato, loifirtato, and l]oifirtạ are from tombs at Falerii Veteres and
loferta from a tomb near Falerii Novi that also yielded several Latino-Faliscan in-
scriptions. Pisani (1964:337-8, 342) suggested that the two spellings distinguished a
noun loferta from a woman’s name l]oifirtạ = l]oifirta(s), gen. loifiṛtato, loifirtato,
but this explains little, as both would still be derived from the same Proto-Italic
*/lo ðero-/. He seems to suggest that loferta is in fact a hypercorrect form due to con-
fusion between the (exceptional) /o / → /o/ and the (normal) development /o/ → /ō ̣/
→ /ū/: unfortunately, there are no Faliscan attestations of the latter development, al-
though it is very likely to have existed (cf. Blümel 1972:19).
3.7.4. The development of /a /. The diphthong /a / occurs in Faliscan only in the

Italic name Paula, the Etruscan name Aulena (from the praenomen Aulus) and in
names whose origin is unclear (see fig. 3.7). The Middle Faliscan co-occurrence of
Aufilus/Aufilius and Oufilus/Oufilius shows that /a / could be confused with /o /, im-
plying that, like /o /, /a / was monophthongized to /ō ̣/. The only clear instance of the
spelling o for /a / is Late Faliscan pola LF 227, however: the spelling of this name
with o is so frequent throughout Central Italy (see Kajava 1987:50-9) that the name
Pola/Polla may perhaps have led a life of its own, independent from the adjective
paulus/paullus. (Note also polae in Lat 251 beside claudia in Lat 393.)
In Latin, too, /a / was monopthongized to /ō ̣/, a change associated by both an-
cient and modern authors with rural or lower-class Latin. The date of this merger is
unclear: the earliest instance seems to be pola (!) in CIL I2.379 from Pesaro,54 from
the first half of the second century, and there appear to be indications for its presence
in Plautus as well (see Pfister 1977:68-9, Meiser 1998:61-2). On the other hand, the
Roman upper class still favoured the pronunciation [a ] well into the first century CE,
as the story of Vespasian’s plostrum (Suet. Vesp. 22) shows.
53
Umbrian vufiune TI Ia.20, uofione TI VIb.19 is sometimes also derived from this root, but
the derivation is in my view too unclear to count as an example: see WOU s.v.
54
Beside pola, the cippi Pisaurenses also have pisaurese CIL I2.378 and lo|ucina CIL I2.371.
106
PHONOLOGY
words and names with /ou/

ou o
louci MF 41 locia MLF 293
lo MF? 33
ḷ[oc]ịes MF 12
louria MF 41 loriea MLF 314
? loferta LF 221
? lo MF 165
? lo[---] MF 155
fourios Lat 216
loucilios Lat 268
soueis (hypercorr.) Lat 217
words and names with /au/
au ou o
aufilo MLF 348 oufilo MF 470* ? ohi*[..] LtF 300?
aufilio MF 49 oufilio MF 48
aufilio MF 50 oụ*[..]o MF 52
aufi[lio MF 53 oufilio MF 275
oufilio MF 276
pola LF 227
?o MF 13
? ofiti (=Aufiti ?) MF 58
? olna (=Aulena?) MF 82
claudia Lat 393
polae LtF 251
Fig.3.7. The spelling of /a /, /o /, and /ọ̄/ in initial syllables.
3.7.5. The developments of /e/, /o/, and /ō/. There are comparatively few data on
the developments of /e/, /o/, and /ō/ in Faliscan, as opposed to Latin (cf. Blümel
1972:15-28, Pfister 1977:64-8, Meiser 1998:58-9, 70-3). In Latin, monopthongization
of */e/ to /ē/̣ took place during the third century: the resulting /ē ̣/ was then closed to
/ī/ by the middle of the second century (Blümel 1972:15-6, Meiser 1998:58).
In Middle Faliscan, */e/ was monophthongized to /ē ̣/, as is shown by hec /hēk/ ̣
← */he-ke/, which is consistently spelled with e (for attestations, see fig.3.8). The
other instances are from the onomasticon, and depend on the dubious connections
between ueicọno MF 88, [u]eculia MF 81, and uecineo LF 220 etc. (cf.
§7.8.1.162,172-174, and on the interpretation of forms that may show monophthongi-
zation of antevocalic /a/ and /e/ (see §3.7.6). There are no Faliscan texts that show a
further closing of /ē ̣/ to /ī/: as in Latin this merger took place around the middle of the
second century, even the Late Faliscan texts are simply too early for this.
107
CHAPTER 3
oi and oe ei e i
with original /e/
ḥec MF 88
hec MF 146
hec MF 158
fe MF 56
hec LF 223
he LF 220
he LF 221
he LF 224
[h]e LF 226
pleina MF 80
pleina MF/Etr 199
ueicọno MF 88 ? [u]eculia MF 80
? u[eculi]a MF 81
? uecineo LF 220
? uecineo LF 224
? uecineo LF 225
? uecin[e]a LF 222
? uecinea LF 223
? ueiuatia MLF 462
? ueiueto MLF 463
? uei LtF 205
from Latino-Faliscan and Latin inscriptions:
with original /e/
heic LtF 231
eidus Lat 393
plenes LtF 231
plenese LtF 251
with original /o/
loidos Lat 217
coiraueront Lat 218
coer Lat 456 idem Lat 456
Fig.3.8. The spelling of /e/, /o/, /ē ̣/, and /ī/ in initial syllables.
The development of /o/ in Faliscan is likewise badly documented. In Latin, /o/ gen-
erally merged with /e/ in (closed) internal and final syllables, and subsequently
shared the development /e/ → /ē/̣ → /ī/ (Meiser 1998:70, 72). This implies that it had
reached the stage /e/ probably somewhere during the first half of the third century
(see above). In initial syllables, however, /o/ developed, through an intermediate
stage /oe/ that was preserved in some words (cf. perhaps the unclear poe[ MF? 130?),
to /ū/ (for the material, see Blümel 1972:16-28).
108
PHONOLOGY
The only Faliscan instances of oi in initial syllables are secondary: loifiṛtato MF 31,
loifirtato MF 32, l]oifirtạ MF 41 with /o/ ← /o / (§3.7.3) and foied MF 59-60 with
/oē/ (/o ē/?) ← */o.dē/ (§3.5.5.1). I do not adopt M. Mancini’s idea (2002:28-33)
that lete MLF 285 represents /lē ̣tę̄/, the monophthongized form of a locative */lotā/.
Beside these instances, oi and oe appear only as archaisms in Latin inscriptions from
Falerii Novi and Lucus Feroniae (see fig.3.9).
The other instances are all from case-endings. The second-declension nomina-
tive plural ending /-o/ (§4.3.6) is found in monophthongized form /-ē ̣/ in lete MLF
285: this can be compared to socie CIL I2.5 (early third century) from the area of the
Lacus Fucinus and ploirume CIL I2.9 (c.200) from Rome.
The instances of a non-monophthongized -oi are usually interpreted as datives,
but sometimes as genitives (§4.4.4). If these forms are datives, as seems likely in the
case of titoi MF 113, 116, 118, 122, [t]ịtoi MF 119, [ti]ṭoi MF 120, [ti]toi MF 121,
[t]ịto[i] MF 115, and tito‹i› MF 114, -oi represents the long diphthong /-ō/ (§4.3.2).
Where these forms can be interpreted as genitives, as seems possible or probable to me
in the cases of cicoi MF 40, titọi MLF 305, zextoi LF 330, caisioi MF 20, and tiroi
colanioi MF 69-71, and possibly also *****oi LF 333 and eiṃoi MLF 293, -oi repre-
sents either the long diphthong /-ō/ or disyllabic /-o.ī/ or /-ō.ī/, as is explained in §4.4.4.
Note that these forms show that, unlike /-o/, the long diphthong /-ō/ (like the long
diphthong /-ā/ in the first-declension dative, cf. §3.7.6), was not monophtongized.
Of the neighbouring languages, South Picene shows preservation of -eí and -uí (/-ō/) in
endings. Umbrian shows monophthongization of */e/ to /ē ̣/ and of */o/ to /ō ̣/: in final
syllables, however, both */o/ and */ō/ were monophthongized to /ē ̣/ (cf. Meiser
1986:122-3, and 66-68 on Proto-Sabellic */ō/). Monophthongization of */o/ in closed
final syllables can also be observed in Volscian (uesclis VM 1) and in Marsian (i]ouies ·
puclẹ[s] VM 4). All these monophthongizations took place before the time of the first
inscriptions. For Etruscan ei and ui, see Steinbauer 1998:35-7.
3.7.6. The developments of /a/ and /ā/. The Italic diphthong /a/ developed during
the Proto-Italic period from inherited */h2e/ (§3.2.3); under specific, morphologically
conditioned circumstances, the long diphthong /-ā/ developed from */-eh2/ (cf.
§4.2.3.). Proto-Latin and Proto-Sabellic therefore inherited both */a/ and */-ā/. The
diphthong /a/ also occurs in the onomasticon, where it can of course be of non-Italic
(e.g., Etruscan or Greek) origin. In the Faliscan inscriptions, as in Latin, there are
clear signs of monophthongization of /a/ in initial syllables. As in the treatment of /e/
and /o/, I shall follow the relevant developments in Latin first, since these are better
documented than the ones in Faliscan. (An overview of the relevant material may be
found in Blümel 1972:9-13).
109
CHAPTER 3
In initial syllables (which is virtually the only position where the diphthong is attested
for Faliscan), Latin originally shows a spelling ai, with a new spelling ae appearing
from the early second century onwards, first in aedem CIL I2.581,1 (186), and, from
the first half of the second century onwards, a new spelling e, in cedito and cedre CIL
I2.366 from Spoleto, cesula CIL I2.376 from Pesaro, grecia CIL I2.350 from Praeneste,
pretod CIL I2.365=LF/Lat 214 from Falerii Novi, and fedra CIL I2.1413 from Rome.
The problem lies in what phonemic or phonetic reality these various spellings
represent. According to the usual interpretation (Pfister 1977:62-3, Meiser 1998:58,
61-2), ai represents the diphthong /a/, ae, a more open diphthong /ae/, and e, a mo-
nophthong /ę̄/. This monophthong is often regarded as dialectal and/or due to substra-
tum influence, and not part of Roman Latin or upper-class Roman Latin, since Quin-
tilian (Inst. 1.7.18), Terentianus Scaurus (CGL 7.16.5-6), and Marius Victorinus (CGL
6.32.4) apparently describe the sound as (still?) a diphthong.
Blümel (1972:9-15) argued at some length that already ae represents the mo-
nophthong /ę̄/. There seems to be no compelling evidence for this, however: also, in a
development /a/ → /ę̄/, an intermediate stage is thoroughly plausible, and it seems
likely that the ae represents this stage at least in the earlier instances. On the other
hand, the monophthongization to /ę̄/ may have taken place quite early, depending also
on the interpretation of the datives in -a as representing /-ę̄/ (see below), of the cases
where the form with e became the standard form (cf. Pfister 1977:62, Meiser
1998:62), and of the rendering of Latin words in Greek and Greek words in Latin (cf.
Blümel 1972:11-2).55 The testimony of the ancient authors on the persistence of /ae/
cannot be decisive, except for what they regarded as standard Latin, and even then
they may be describing a wished-for state based on the status quo of the second or
first century BCE rather than the actual situation.
In Faliscan, there are likewise three ways of spelling this diphthong, namely ai, ei and
e (for the instances, see fig.3.9.). The spelling ai clearly represents the diphthong /a/,
but already during the Middle Faliscan period the predominant spellings are ei and e,
so that already in Middle Faliscan ai must have become a ‘historical’ spelling: there
are in fact no instances of ai from inscriptions that are with certainty Late Faliscan.
The Middle Faliscan spelling ei may represent a different diphthong, and the spelling
e, a monophthong: this means that already during the Middle Faliscan period ei must
also have become a ‘historical spelling’: it is in fact the spelling that is least used. As
in the case of Latin, it is not clear which diphthong and which monophthong are repre-
sented by ei and e. In my view, there are two possibilities:
55
It is remarkable how soon the introductions of ae and of e follow each other: this either
points to a quick development of /ae/ → /ę̄/, or to the introduction of the monophthongized
vowel from other languages or dialect, or, in case ae already represents /ę̄/, of the spelling e
from other orthographical traditions.
110
PHONOLOGY
ai ei e
aịṃiosio EF 467* eiṃoi MLF 293
ạ[i?]ṃ MF 89
efiles MF 113
efiles MF 115
efi[les] MF 116
efile MF 114
aino MLF 352 eina MF 80
cailio MF 90 c ̣elio MF 94
cailio MLF 358 celio MF 95
cailio MF 376 ce[lio MF 96
cail[ia MF 92 c]elio MF 97
cai[lia MF 93 celio MF 105
celio LF 331
kaisiosio EF 7 ceisio MF 276 cesio LF 331
ḳai[s]i[o MF 51 cẹịṣị[.] MF 140 cesi MF 94
cais[io] MF 153 ceịsịa LF 235 cesi MF 263
caisio MF 351 cesie MF 257
caisioi MF 20 cesies MF 265
c[ai]sia MF 41 cẹsiḷị[a] MF 99
cesula LF 229
kreco MF 147
crạ[i---] MF 141 cre[---] MF 142
leiuelio MF 79 leueli MF 14
[leu]elio MF 90
[leu]elio MF 159
le[ueli]o MF 146
le[---] MF 148
mecio MLF 211
mesio MF 148
precono MLF 361
pretod LF 242
pret[or LF 247
cụestod LF 242
cues[tor LF 243
c]ues[tor LF 245
cue[stor LF 247
reic ̣[lio MF 98
reiclio MF 99
reị[cli.] MF 100
sceiuai LF 379 sceua LF 312
pretod LF/Lat 214
leuia LtF 327
? ef LtF 211
Fig.3.9. The spelling of /a/ and its reflexes in initial syllables.
111
CHAPTER 3
(1) The spelling ei represents essentially the same sound as Latin ae. Faliscan would
then have indicated the changed quality of the diphthong by changing the first letter of
the digram (ai → ei), and Latin, by changing the second (ai → ae): alternatively, the
Latin diphthong may be thought of as /ae/ and the Faliscan diphthong, as / /. The
Faliscan spelling may have been influenced by the spelling of Etruscan, where ai was
replaced by ei during the fourth century (Steinbauer 1999:34-5), but the use of ei to
denote the reflex of /a/ is also found in Latin ceisia CIL I2.559 from Praeneste and
queistores (beside uicturie) CIL I2.388 from Trasacco. If the Faliscan diphthong rep-
resented by ei was (approximately) the same as the one in Latin, the monophthong
represented by e may well have been /ę̄/.
(2) The spelling ei represents a more closed diphthong than Latin ae. Since the spell-
ing ei is also used for the diphthong /e/ and e for its monophthongized reflex /ē ̣/, it is
possible on the basis of the spelling alone that /a/ in fact merged with /e/ and, like it,
was monophthongized to /ē/.̣ 56 This seems to be the interpretation of Blümel
(1972:13, 34-5), who represents the Latin monophthong as /ę̄/ and the Faliscan and
Praenestine monophthong as /ē/, and perhaps also of R. Giacomelli (1978:25).57
Which of these two interpretations is the correct one cannot be established: there is no
way to show whether the ei and e that represented the reflex of /a/ were phonemically
or phonetically different from the ei and e that represented the reflex of /e/. Since for
the purpose of this study I assume that Faliscan is a Latin dialect, I chose the first in-
terpretation, being unwilling to assume different developments for the various Latin
dialects unless there are positive arguments for doing so. Whether it represents /ę̄/ or
/ē ̣/, the spelling e for the reflex of /a/ is already the predominant one during the Mid-
dle Faliscan period, indicating that the process of monophthongization was concluded
by the middle of the third century. This is substantially earlier than Latin, where the
first clear attestations of monophthongization are from the second century, and then
from areas where substratum influence can be assumed.
In antevocalic position, the spelling ai occurs both in initial syllables, as in kaios EF
4, aiedies Cap 390 (and perhaps aieạ MF 110, if to be read thus), and in medial sylla-
bles, as in uoltaia MF 196, latinaio MF 210, ac ̣iuaiom Cap 465 and perhaps also cli-
peaio MF 470* and frenaios MF 471* (if these are not errors for clipea‹r›io and
frena‹r›ios, or due to palatalization of /r/, §3.5.5.). On the other hand, there appear to
be several cases of gentilicia ending in the nominative in -ieo/-iea or, in the genitive,
in -ei, which points to a name ending in -aeus/-aea or -eius/-eia:
56
This is in fact the Latin development of /a/ in closed medial and final syllables, where /a/
merged with /ē ̣/, and thence with /ī/ by the middle of the second century.
57
R. Giacomelli does not clearly differentiate between graphemes and phonemes at this point,
which rather obscures his arguments.
112
PHONOLOGY
x aieạ MF 110 (if to be read thus, which is very doubtful), can represent Aieia or
possibly Aiaea (§7.8.1.9);
x loriea MLF 314 (beside loụṛia in MF 41) is clearly either Luriaea or Lurieia
(§7.8.1.91);
x letei MF 470* can be derived from Etruscan leθaie Etr XLVIII and represent the
genitive of an unattested Letaeus, or from Etruscan Leθe as the genitive of a like-
wise unattested Leteius (§7.8.1.85);
x catinei MF 469* is the genitive, either of Catineius, which occurs in Latin, or of
an unattested *Catinaeus (cf. §7.8.1.37);
x tertinei MLF 471* and tertineo LF 213 are forms of either *Tertineius or
*Tertinaeus (Latin only has Tertinius, cf. §7.8.1.151);
x uelcei LF 332 is probably a form of Velceius, but could conceivably be a form of
Velcaeus (§7.7.1.81).
x uelmineo MLF 305, 307, 309, 310, 312, 315 and uelminẹo MLF 308: although
consistently spelled with -e-, it is assumed that the name represents an unattested
Velminaeus, due to possibility of reading v]ẹḷṃẹṇaie in Ve 3.19.
These forms are discussed by G. Giacomelli (1962:363), and point to a mo-
nophthongization of antevocalic /aV/ → /-ę̄V/ and /ē ̣V/. I do not adopt this interpreta-
tion in the cases of (1) uecineo LF 220, 224, 225, uecinea LF 223, uecin[e]a LF 221,
which is consistently spelled with -e- and can be compared to Latin Vicinius; (2)
zertenea LF 221, spelled with -e- but comparable with the Latin Sertinius; (3) the pat-
ronymic adjectives derived from Iuna and Volta, which are ịuneo MF 151, iuneo LF
220, perhaps also iu?]nẹo MF 141, and uolteo MF 275, uolθeo MF 276, since I read
uoltio in LF 224, and when the patronymic adjectives of Iuna and Volta are used as a
praenomen or a gentilicium, they are always spelled with i (cf. §7.5.2).
In final syllables, -ai only occurs in endings. The Early Faliscan inscription have at-
testations both of the short diphthong /-a/, in the first-declension nominative plural
sociai EF 4 (§4.2.6) and in the first singular perfect ending pe para[i EF 1 (§5.3.1.12)
and of the long diphthong /-ā/ in the first-declension dative singular soc ̣[iai] ... karai
EF 1 (§4.2.3). For the Early Faliscan period, monophthongization is not expected.
In Middle and Late Faliscan, -ai occurs only in forms that can represent either
the first-declension dative ending /-ā/ or the genitive ending /-āī/ → /-ā/: the latter
interpretation is not adopted by everyone (see §4.2.3). Probably genitives are uoltai
MF 367-370, iunai LF 112, iunai MLF/Cap 475*, and sceiuai LF 379; probably da-
tives are citiai MF 270, zaconiai MF 154, popliai MLF 308, and possibly [..........]naị[
MF 17 and [---?]uoxie[.]eai MLF 310; either genitives or datives are uoltiai MF 165
and [---]ạltai MF 109. None of these forms are ever monophthongized: note especial-
ly sceiuai MLF 379, where the initial syllable has -ei- but the final syllable has -ai.
113
CHAPTER 3
There are two instances of -e that have been explained as monopthongized end-
ings Colonna’s (1972:446-7) interpretation of [---]ronio : uol[t---|---]a*ome MF 156
as a sepulchral inscription of the type ‘X [made this grave] for Y’ (§8.10.2), and M.
Mancini’s (2002:28-33) interpretation of lete MLF 285 as a locative /lē ̣tę̄/ ← */lotā/.
This is very doubtful on several grounds: note that the interpretation requires a pro-
ductive locative for Faliscan (§8.2.1). To these two instances can be added eco tulie
MLF 383 if tulie is a genitive /tullię̄/: it is, however, rather an Etruscoid nominative in
-e(s), for which cf. §9.2.2.2c.
Another group of instances, however, are the datives in -a in the Latin inscriptions
from Lucus Feroniae, fe]ronea in Cap 431 and feronea Cap 433 (both before 211), and
in menerua sacru LF/Lat 214 (probably mid-second century). Such datives in -a ap-
pear to have spread from Latium and the Latin-speaking colonies, and are found in
Southern Etruria from the fourth and third century onwards: the first instances in Etruria
are mursina CIL I2.580 from Cortona (?), and menerua CIL I2.2909 from Veii. (Interest-
ingly, the datives in -a from the agri Faliscus and Capenas are occur in Latin inscrip-
tions.) Wachter (1987:483-4) suggests that in these forms -a represents /-ę̄/: in other
words, the monophthong reflex of the sound spelled -ae could be spelled both as -e and
as -a. In that case, the spread of the dative in -a would be no more than the introduction
of an alternative way of spelling /-ę̄/, which in my view is a more attractive solution
than the morphophonological explanations that have until now been proposed for these
forms (for which see Villar 1986).
With regard to the surrounding languages, the Faliscan monopthongization of /a/ was
therefore slightly earlier than the Latin one, but more or less contemporary with the
Etruscan one, where the spelling ai was replaced by ei during the fourth century (Rix
1984:205-6, Steinbauer 1999:34-5). As for the Sabellic languages, Umbrian, Volscian
and Marsian show complete monophthongization of /a/ (and /ā/) by the time of the
oldest inscriptions, and it cannot be established at what time it occurred.
3.7.7. A diphthong /�/? As said in the discussion on the forms (titoi) mercui (§14.1.3,
§4.6.2), there is a reasonable possibility that a dative in -ui occurs in titoi | mercui MF
113, [t]ito[i] | ṃercu[i] MF 115, titoi | mercụ[i] MF 116, tito | mercui MF 114, titoi 
mercụ[i] MF 118, [t]itoi  mercu[i] MF 119, [ti]toi  mercu[i] MF 120, [ti]toi  mercui
MF 121, titoi  ṃ[e]rcui MF 122, [m]ẹrcui MF 124, and ṃercui MF 125 (the same end-
ing is restored in MF 117 and 123). Although a diphthong /u/ or /ū/ is admittedly alien
to the Italic languages, it is hard to imagine that in this form, and probably also in its
Oscan parallel mirikui Cm 12 (which, as it predates the introduction of ú and í, may be
an o-stem dative in /-ō/), -ui represents anything other than /-u/ or /-ū/, perhaps an
analogical creation after the first-declension dative /-ā/ and the second-declension da-
tive /-ō/. Note that -ui cannot represent /-uī/ m /-uē/̣ m */-ue/, as in Faliscan /ē ̣/ had
not merged with /ī/ (§3.7.5).
114
PHONOLOGY
3.8. Summary of §§3.2-7
As was said at the beginning of this chapter, the phonological material in general allows
to take a more structural look at the position of Faliscan, especially from a diachronic
perspective (§1.2). If we working as strictly as possible from a theoretical development
Proto-Italic > Proto-Latin > ..., the developments observable in Faliscan can be placed
within this framework.
As was to be expected, Faliscan shows no reflexes of any individual develop-
ments at the Proto-Italic stage: from a theoretical perspective, this would have been
mutually exclusive with the position of Faliscan as an Italic language. I point especially
to the Proto-Italic merger of */e/ with */o/, where it has been suggested that Faliscan
shows signs of the preservation of */e/ (§3.2.1): in my view, there is no evidence for
this, and certainly not from Faliscan.
At the post-Proto-Italic stage, however, we find the first, and probably earliest,
development where Faliscan develops differently, namely the Proto-Latin development
of the word-internal reflexes of the original voiced aspirates, (§3.3.3). (Note that if Fal-
iscan is regarded as a Latin dialect and shows a different development from the other
Latin dialects, this development can of course no longer be called ‘Proto-Latin’.) How-
ever much I would see that things were otherwise, it cannot with reason be maintained
that Faliscan in this respect resembles other Latin dialects where the development was
the same as in Faliscan: there simply is no evidence for such dialects within Latin. The
development in Faliscan appears to be similar to that in the Sabellic languages, and all
the material from within Latin that reflects a similar development can without problems
be ascribed to interference from a Sabellic language, presumably Sabine.
For Faliscan, this must imply a period of independent development, which can
unfortunately not be dated either to a recent or to an early period. It should be noted,
however, that the Faliscan development is completely in line with the general trend of
the development of the voiced aspirates in the Italic languages, namely a development
resulting in spirants both word-initially and word-internally, while it is the Latin devel-
opment that is in fact the unique one, resulting in occlusives word-internally. This could
potentially be taken as indicating that the traditional position, namely that this was a
local Roman development that spread through the rest of the Latin-speaking area, is in
fact correct, even though evidence is lacking.
This development is even stranger in view of the fact that Faliscan shows no other
independent developments during the post-Proto-Italic period: in fact, all other Proto-
Latin developments that can be traced in the Faliscan material show that Faliscan sided
with Latin wherever there was a difference between the Proto-Latin and the Proto-
Sabellic developments, such as e.g. the preservation of the labiovelars (§3.4) or the de-
velopment of /er/ between dentals (§3.3.4.2). This remained so during the later periods,
as is shown e.g. by the development of /o/ between labials (§3.7.3), and by fairly re-
115
CHAPTER 3
cent developments such as rhotacism (§3.5.3), which in my view is attested for Falis-
can, and the development /#fV/ → /#hV/, which probably operated in other Latin dia-
lects as well (§3.5.2).
Faliscan is slightly earlier than most other Latin dialects, and certainly earlier than
that of Rome, in the monophthongization of the diphthongs (§3.7), but the general ten-
dency of these monophthongizations is the same as in Latin, and where the final result
is different, this is because in Latin the monophthongization process reached its final
stages only during the second century, when Faliscan texts were no longer produced. A
similar picture can be drawn in the case of the second-declension nominative singular
ending -os, both with regard to the omission of word-final -s, where Faliscan is so con-
sistent in its omission that I have ventured to suggest that this might have been an
orthographical rule (§3.5.7d), and with regard to the weakening of the vowel (§3.6.6.1),
where our view of the situation is obscured by the fact that the second-century material
from the area is Latin rather than Faliscan.
All in all, the only phonological feature that separates Faliscan from Latin are, as I said,
the word-internal reflexes of the voiced aspirates, a feature that points to a significantly
different development that was perhaps of reasonably early date. Yet this is the only
phonological feature that separates Faliscan from Latin, while every other phonological
development is either in agreement with the Latin development or is of very recent date.
I shall return to this point in §10.1.2.
116
Chapter 4
The nominal and pronominal inflections
The study of Faliscan declensional morphology provides a number of interesting cases where
Faliscan can be compared with both Latin and the Sabellic languages. Apart from this, the
Faliscan data have played an important role in the discussions on the history of the second-
declension genitive singular. After a short preliminary remark (§4.1), this chapter deals with
the attested case endings of the first declension (§4.2), the second declension (§4.3), the second-
declension genitive singular (§4.4), the third declension (§4.4), the fourth declension (§4.5),
and the fifth declension (§4.6) respectively. The second part of the chapter is devoted to the
personal pronouns (§4.7), the demonstrative pronoun (§4.8), and the relative pronoun (§4.9).
The chapter concludes with a short evaluation of the position of Faliscan from a morphological
perspective (§4.10).
4.1. The nominal and pronominal inflections: methodological issues
Studies on specific cases or endings apart, Faliscan nominal morphology has been
treated comprehensively by Deecke (1888:262-6), Stolte (1926:48-58), and G. Gia-
comelli (1963:131-50). The studies on specific endings mostly concentrate on the
history of the ending of the second-declension genitive singular (see §4.4), for in this
case especially, the relative antiquity of the Faliscan material comes into its own.
As in the case of phonological data (§3.1.1), only the inscriptions classed as
Faliscan have been used as a source for the primary data. The Latino-Faliscan, Cap-
enate, and Latin inscriptions have only been used to provide additional material.
It may be asked to what extent names of Etruscan origin were adapted to the Italic
paradigms of Faliscan, as they were in Latin. Where this can be ascertained, it appears
that Etruscan names in -a were declined according to the Faliscan first declension, that
Etruscan names like Larθ and Arnθ were declined according to the consonant-stems,
and that Etruscan toponymic gentilicia formed with the suffix -te/-ti were declined
according to the i-stems. These forms have therefore not been treated separately. An
exceptional case are the ‘Etruscoid’ forms in -(i)es: these are discussed in §9.2.2.2d.
The main problem in the description of the Faliscan nominal declension is that a
number of forms can be interpreted in several ways, especially those without context.
The most notable examples of this are the first-declension forms in -ai and the second-
declension forms in -oi, which can be interpreted both as genitives and as datives. The
interpretation depends on syntactic and typological comparison with similar texts in the
Italic languages and in Etruscan. This is discussed in §8.2.3-4 and §8.8.1.
117
CHAPTER 4
4.2. The first nominal declension
4.2.1. The first-declension nominative singular. The clearest attestations of the

nominative are those where it occurs as the subject of a verb phrase. The instances of
the feminine nominative subjects are
eqo ụrneḷ[a ti?]tela fitaidupes  arcentelom hutị[c?]ilom  pepara[i? EF 1
[ma]rco  pleina  marcio  man[o]mo  cauiacue | [u]eculia  uoltilia  uentarc[i
.....  hec  cupa]nt MF 80
uol[ta ]**[---] | iatac ̣ue  ḷ[---] | hec  cupạṭ MF 158
[po]plia[  ---| hec]  cup[a] MF 161
uipia  zertenea  loferta | marci  acarcelini | mate  he  cupat LF 221
marcio  acarcelinio | cauia  uecinea | hẹc ̣ cupat LF 223
In these and in all other instances, which are too numerous to enumerate here (for
attestations see §7.4.1 with fig.7.2), the ending is -a. Faliscan therefore, like Latin,
preserved the unrounded vowel /-a/ or /-ā/ (see below), whereas in the Sabellic
languages the vowel was rounded to /-ō/ (Umbrian -a, -u, -o, Oscan -u, -ú, -o, -o) before
the fifth century (Meiser 1986:44). The instances of a nominative with a rounded vowel
in Faliscan, caui  tertinei  || posticnu MLF/Cap 474*, and perhaps sta sediu Cap 466,
may be due to interference from a Sabellic language: see §9.3.2.
The masculine first-declension stem nominative subjects are
mama z[e]xtos med f[if]iqod EF 1
The other instances are uolta MF 15, [u]olṭa MF 149, iuna MF 39, 48, 96, i[un]a MF
148, [---]a MF 53, olna MF 82, eina MF 87, [4-5]ạ and hac****a MF 89, hermana
MF/Etr 264, tetena MF 266, macena MF 269, mar||cna MF 270, nomes|ina MF 272;
[.]osena MLF 206, hescuna MLF 346, [iu]na MLF 363, uicina MF 371-372; iuna
LF 307, and, in isolation, iuna MF 73, 198, MLF 297, 298, pleina MF/Etr 199.
In all these cases the ending is -a, and Faliscan therefore had asigmatic first-
declension masculine nominatives, as did the other Italic languages:60 note that Greek
names in -aj and -hj were given a nominative in -a both in Latin (e.g. Anchisa Naev.
Poen. 25.1S, Aeacida Enn. Ann. 273V; see LHS p.454) and in Oscan (arkiia Po 65,
santia Cm 40, but arímmas Po 52).
60
The exceptions are Latin Graecisms like cottas CIL I2.2877 from Sicily, hosticapas Paul.
Fest. 91.15L and paricidas Paul. Fest. 247.24L, and Oscan praenomina like maras Cm 14,
markas Po 66 (from autopsy: margas ST), and tanas Sa 27 (originally indigenous second-
declension names with -as /-as/ m /-os/, as in Messapic?).
118
THE NOMINAL AND PRONOMINAL INFLECTIONS
Although the quantity of the first-declension nominative -a in Faliscan cannot be

established (cf. §3.6.1), it is worthwile to devote a few words to the problem, as I shall
have to return to it later (§4.4.11). The ending of the first-declension nominative
singular (and of the neuter plural nominative-accusative, which was originally identical
with it), is usually reconstructed as PIE */-eħ2/ → Proto-Italic */-aħ2/ → */-ā/. In Latin,
however, the ending is /-a/ already at the earliest time the quantity of the vowel can be
ascertained, namely the mid-third century, when the first quantitative verse appears.61 It
is therefore assumed that in Latin */-ā/ was shortened before the third century, and, as
this shortening cannot be described as a regular phonological process, various ana-
logical models have been proposed for it, but none of these is really convincing: for a
discussion, see Beekes 1985:21-5. In the Sabellic languages, on the other hand, the
vowel of the nominative ending was rounded, while the short vowel of the vocative, /-a/
m Proto-Italic */-a/ m PIE */-ħ2/, remained unrounded at least in Umbrian: the
attestations of the vocative, all from the texts in the Latin alphabet, are always spelled
-a, while in the Latin alphabet the nominative is always spelled -o. This implies that
there was a quantitative difference between the Sabellic vocative and nominative
endings at least at the time the vowel was rounded.
Lejeune (1949) suggested that in Oscan and Umbrian, the vowel of the nomina-
tive ending it was shortened after it was rounded, as it was also in Latin, and that in all
three languages, this was due to a tendency to shorten the most open long vowel in
word-final position.62 If this were correct, there would seem to be no reason why this
shortening should not have taken place also in Faliscan, but as things stand, it is
doubtful whether Lejeune’s theory can be accepted.
A radically different solution has been proposed by Beekes (1985:20-37).
According to him the PIE ending was not */-eħ2/, but */-ħ2/, which would be reflected
by (among others) Latin /-a/. In some languages, among which Oscan and Umbrian,
*/-ħ2/ would then have been replaced by */-eħ2/ or */-aħ2/ after the accusative
*/-eħ2m/ o */-aħ2m/. However, the assumption that Latin and the Sabellic languages
had different morphophonological developments already before the Proto-Italic
disappearance of the laryngeals is inadmissible (cf. §3.1.2). It is more likely that the
remodelling of the nominative took place later, and that the PIE nominative */-ħ2/
developed regularly to Proto-Italic */-a/. This is also implied by the fact that several of
the inflectional types of the stems in */-iĦ/ (e.g. the devī-type stems in */-iħ2/) had a
Proto-Italic nominative in */-ī/ that could only have developed from */-iĦ/ (see
61
Et densis aquilā pennis obnixa uolabat Enn. Ann. 147V is a lengthening metri causa (Skutsch
1985:58), not a reminiscence of the older /-ā/, as Steuart (1925:131) suggested.
62
Cf. the Oscan neuter accusative plural segonw aizniw ... ‹w›psanw ... stabalano Lu 5. Lejeune
later (1970:300) suggested that the shortening preceded the rounding, but this would leave the
difference between the Umbrian nominative and vocative unexplained.
119
CHAPTER 4
Schrijver 1991:363-90). It would then be possible to envisage a Proto-Italic remodelling

of */-ī/ to */-ia -a/ after the rest of the paradigm (and possibly after the second declen-
sion with */-ios -os/), resulting in a Proto-Italic */-ia -a/ that was preserved in Latin but
remodelled in Proto-Sabellic to */-iā -ā/ after the /ā/ in the endings of the oblique cases.
4.2.2. The first-declension genitive singular. The first-declension genitive singular

ending provides one of the presumed morphological differences between Faliscan and
Latin.
The communis opinio on the development of the ending of this case in the Italic
languages is that the Proto-Italic ending was */-ās/, which either reflected the PIE
proterodynamic ending */-eħ2s/ (or */-eħ2es/) or was a replacement of PIE hysterody-
namic ending */-ħ2os/. This /-ās/ was preserved in the Sabellic languages (South Picene
íitas  estas  amge|nas AP.3, saf|inas  tútas TE.5, safina[s TE.7, ḥ[1-2]lpas CH.1,
selah AQ.3?, fitias RI.1, Umbrian -as o -ar, Oscan -as, -aj). In Latin, /-ās/, epigraphi-
cally attested in late-seventh- or mid-sixth-century eco urna tita uendias Lat 479†, and
in fourth- or third-century manias CIL I2.2917b and pias CIL I2 sub 479,63 was replaced
by an /-āī/ created after the second-declension ending /-ī/, probably first in the mascu-
line forms. The first attestations of -ai are from the early third-century pocula deorum
(aecetiai CIL I2.439, belolai CIL I2.441, fortunai CIL I2.443, lauernai CIL I2.446,
meneruai CIL I2.447, uestai CIL I2.452). In literary Latin, Andronicus and Naevius still
have -as: Andronicus’ topper citi ad aedes uenimus Circae 27L is a modernization by a
copyist, either of an original Circai or of an original Circas. In Ennius (who only once
has -as in dux ipse uias Ann. 41V) and Plautus, the normal genitive is /-a/ or /-ā/ (cf.
Blümel 1972:39, Pfister 1977:118-9), /-āī/ being employed for metrical convenience at
or near the line-end (Skutsch 1975). After this time the genitive in -as is found only in
formulas like pater familias (first Cato Agr. 2.1).
This model has been challenged by Schrijver (1991:360-3), who raises the
traditional objections (1) that it is hard to credit the relatively small number of mascu-
line forms with being the origin of /-āī/, and (2) that /-āī/ is not the expected result if the
model were /-ī/. He proposes that during the Italo-Celtic period the PIE hysterodynamic
genitive ending */-ħ2os/ was replaced by the new second-declension genitive */-iĦ/. In
Latin, the resulting */-ī/ was then remodelled to /-āī/ after the /ā/ of the other endings,
and this form was eventually generalized in favour of /-ās/, the reflex of the PIE
proterodynamic genitive */-eħ2s/. In the Sabellic languages, on the other hand, the
ending of the proterodynamic flexion was retained and the reflex of the hysterodynamic
ending discarded. I find it very hard to accept this, not in the least because in the earliest
Latin and Faliscan epigraphic material we find only -as, the forms in -ai appearing only
63
The forms in -as in deuas | corniscas | sacrum CIL I2.975 and anabestas CIL I2.969 may be
plural datives with -as representing /-ę̄s/ or /-ē ̣s/: cf. §4.2.3.
120
from the early third century onwards.64 Schrijver in fact completely disregards the
epigraphic evidence, stating that “the normal Gsg. of the Latin ā-stems from the earliest
documents onwards is -āī (> -ae)” (1991:360). Neither do I find the assumption of a
centuries-long coexistence of two apparently productive morphemes /-ās/ and /-ī/ for
the same category attractive. I therefore maintain the communis opinio as set out above.
The situation in Faliscan is less clear than in Latin. Faliscan certainly had a genitive in
-as, which appears in
titias duenom duenas EF 3
titias MF 201
pupiias MLF 304
Whether Faliscan, like the Sabellic languages, preserved the genitive in -as (thus e.g. G.
Giacomelli 1963:139-40, 1978:517), or, like Latin, shifted to a genitive in -ai, is
unclear. The evidence for a Faliscan genitive in -ai is inconclusive and consists of:
(a) Contextless forms in -ai in the pottery inscriptions:
iunai MF 74, 107, Cap 475*
uoltai MLF 367-370
sceiuai LF 379
These could be Besitzerinschriften consisting of genitives in -ai. They can also be
interpreted, however, as Geschenkinschriften consisting of the name of the recipient in
the dative, as is shown by vultasi Etr XLII and perhaps by the second-declension forms
in -oi, which are generally interpreted as datives: see §8.2.3-4 and §8.8.1.
(b) Forms in -ai with a context, of which the following can be interpreted as genitives:
[---]*i  u[o]ltiai lo MF 165
tito  uel|mineo  iun|aị ị*ice MLF 315.
The former could be translated as ‘..., a freedman/freedwoman of ... Voltia’ and the
latter as ‘Titus Velminaeus ..., son of Iuna’ (rather than ‘Titus Velminaeus [made this
grave?] for Iuna’). Possibly also genitives, but perhaps rather datives, are the forms in
-ai (‘X [made this grave] for Y’) are larise  mar||cna  citiai MF 270 and cauio
uelminẹo | popliai file MLF 308: see §8.10.2. Unclear are [--- ] zaconiai MF 154 and
cuicto uelmineo | [---]uoxie[.]eai MLF 310.
(c) Eco tulie LF 383, where tulie could be a genitive with a monophthongized ending,
/tullię̄/, but could also be a Etruscoid nominative tulie(s) (see §9.2.2.2d). I tend towards
the latter interpretation, as there is no other instance of a monophthongized ending in
either the genitive or the dative of the first declension (§3.7.6).
64
The same is true for the second declension, which in this theory would have had a genitive in
*/-iĦ/ o /-ī/ from the Italo-Celtic stage onwards: the early Latin and Faliscan epigraphic
material shows only -osio, while the genitive in -i does not appear before the late fourth century.
121
CHAPTER 4
(d) Some of the Faliscan forms in -oi that can be interpreted as genitives: the easiest
explanation of these forms is that they are modelled on genitives in -ai: see §4.4.4.
This material is so ambiguous that whether or not Faliscan is assumed to have had a
genitive in -ai is based mainly on a priori arguments. However, the reluctance of
authors to accept a Faliscan genitive in -ai is probably partly due to the fact that this
ending is supposed to be modelled on the second-declension genitive singular ending -i
/-ī/, which was long thought to be non-existent in Faliscan (§4.4.1). I myself am
inclined to assume the existence of a Middle and Late Faliscan first-declension genitive
in -ai, regarding the instances under (a) and at least the first two instances under (b) as
attestations, with those referred to under (d) as supporting this assumption.
A few remarks with regard to the Faliscan genitive in -ai. (1) It seems likely that
a Faliscan genitive in -ai is to be ascribed to the same shift as in Latin, taking place
simultaneously also in the ager Faliscus and Capenas: there seems to be no reason to
assume that it reached the area due to influence from Latin, as e.g. Devine (1970:20-1)
assumed. (2) Devine’s (1970:20) suggestion that the Faliscan genitive in -ai is attested
only from masculine forms and might not yet have penetrated the feminine forms, is
now contradicted by sceiuai LF 379. (3) It cannot be ascertained whether, in the
Faliscan genitives in -ai, this -ai represents /-āī/, or whether at some point this became a
diphthong /-ā/ or /-a/ as in Latin. If tulie in LF 383 is a monophthongized genitive
(which I doubt), this would imply the latter.
4.2.3. The first-declension dative singular. The only incontrovertible attestation of the
dative comes from Early Faliscan, in
praụ[i]os urnam  soc ̣[iai] p ̣orded karai EF 1
Most of the forms in -ai from the later periods appear in such a context that they can
also be interpreted as genitives (see §4.2.2 and §8.8.1). Of these forms, those that can
perhaps most plausibly be interpreted as datives are the sepulchral inscriptions of the
type ‘X [made this grave] for Y’ (cf. §8.10.2):
larise  mar||cna  citiai MF 270
cauio uelminẹo | popliai file LF 308
Perhaps datives too, but unclear, are iii l[.......]|[..........]naị[?---] MF 17 (‘the third
bed (?) ... for ...na’) and [---]ạltai  MF 109 (a dedication?). Unclear, too, are [--- ]
zaconiai MF 154 and cuicto uelmineo | [---]uoxie[.]eai LF 310. Of the remaining forms
in -ai, I interpret the contextless forms in -ai in Besitzerinschriften, iunai MF 74, 107,
Cap 475*, uoltai MLF 367-370, and sceiuai LF 379, as genitives rather than as datives.
A dative interpretation is possible, however, in view of vultasi Etr XLII and perhaps
also of the second-declension forms caisioi MF 20 and tiroi colanioi MF 66-68: see
§8.8.1. I likewise regard the forms in -ai in [---]*i u[o]ltiai lo MF 165 and tito 
uel|mineo  iun|aị ị*ice MLF 315 as genitives rather than as datives.
122
Although ambiguous, the material shows that Faliscan in all probability had a dative in
-ai reflecting Proto-Italic */-ā/, whether the reflex of the PIE proterodynamic ending
*/-eħ2e/ or a remodelling of the reflex of the PIE hysterodynamic ending */-ħ2e/. The
same ending is found in Latin (-ai, later -ae) and in the Sabellic languages (Umbrian -e
/-ę̄/ m */-āi/, Oscan -aí, -ai). In both branches the long diphthong of the Proto-Italic
ending may have been shortened: in Latin before c.250 (Blümel 1972:38-9), perhaps
even at a prehistoric date (Villar 1987a:49, 1987b); for the shortening in the Sabellic
languages, see Meiser 1986:66-8.
Monophthongization of the dative ending is regular in Umbrian and Volscian, and
is found sporadically also in Latin inscriptions, often from areas where substratum
influence can be assumed (e.g. uictorie CIL I2.388 from the Lacus Fucinus, supunne
CIL I2.2111 from Foligno, uictorie CIL I2.2631 from Veii). Although in Middle
Faliscan // was monophthongized word-internally, there are no certain instances of
monophthongization in the dative ending (§3.7.7): I therefore find it hard to adopt
Colonna’s (1972c:446-7) interpretation of MF 156 as [---]ronio  uol[t---|--- m]axome
as ‘...ronius son of Volt..., for ... Maxuma’.
Another group of instances, however, is constituted by the datives in -a in
menerua sacru LF/Lat 214 and in the Latin inscriptions from Lucus Feroniae, fe]ronea
Cap 431, and feronea Cap 433. These datives in -a seem to have spread from Latium
and the Latin-speaking colonies, and are found in Southern Etruria from the fourth and
third century onwards: mursina CIL I2.580 from Cortona (?), and menerua CIL I2.2909
from Veii (the instances from Lucus Feroniae, too, are from the third century). Wachter
(1987:483-4) suggests that in these forms -a represents /-ę̄/. In that case the spread of
the dative in -a would be no more than the introduction of an alternative way of spelling
/ę̄/ (in other words, the sound spelled -ae could now be spelled both as -e and as -a).
4.2.4. The first-declension accusative singular. The accusative singular is attested

only for Early Faliscan, in
eitam EF 5
The Proto-Italic form was */-ām/, reflecting the PIE hysterodynamic ending */-eħ2m/.
In the Sabellic languages, where the regular spelling of the ending was -am (attested
e.g. in South Picene viam, tokam TE.2, deiktam, ok[r]ikam, qora CH.1, postiknam
CH.2, koram AQ.2, in Umbrian -am, Oscan -am), the long vowel appears to have been
preserved, as is implied both by Oscan paam Po 3 and p]aam Sa 4, and by the fact that
the ending of the ia-stems was not affected by the Sabellic Endsilbensynkope, as was its
io-stem counterpart */-om/. In Latin, */-ām/ was shortened to /-am/, but the date of this
shortening is debated (cf. Blümel 1972:45). It is impossible to establish the quantity of
the vowel in Faliscan: in Early Faliscan, it was in all probability still /-ām/, but it may
later have been shortened to /-am/.
123
CHAPTER 4
4.2.5. The first-declension ablative singular. The Proto-Italic ending was */-ād/,
formed after the inherited second-declension ablative */-ōd/, reflexes of which are
found both in Latin (-ad /-ād/ o -a /-ā/) and the Sabellic languages (Umbrian -a m
*/-ād/, Oscan -ad). The Faliscan evidence for this ending is uncertain or indirect, but
points to the same form -ad /-ād/ o -a /-ā/ (cf. §3.5.7c):
(a) From the Early Faliscan period there is only a very faint possibility that *[3-4]*ad
in EF 1 may be an ablative.
(b) The Middle Faliscan adverb ifra MF 40 is of ablatival origin, reflecting Proto-Latin
*/enð(e)rād/, with loss of /d#/.
(c) Late Faliscan de | zenatuo sententiad LF/Lat 214 is the only direct attestation of the
ablative singular, but it is very questionable in how far this inscription (still) represents
Late Faliscan: the ablative in -ad will by this time (mid-second century) have been an
archaism in any case.
4.2.6. The first-declension nominative plural. The nominative plural is attested only
for Early Faliscan in
ues saluete sociai EF 4
This shows that in Faliscan the Proto-Italic nominal ending */-ās/ (m PIE */-eħ2es/) had
been replaced by the pronominal ending */-a/ during the Proto-Latin period. The
replacement of the inherited second-declension nominative plural ending */-ōs/ by the
pronominal ending */-o/ (§4.3.6) will have taken place at the same time. This shift
constitutes one of the stronger morphological links between Faliscan and Latin, since in
Sabellic the Proto-Italic endings */-ās/ and */-ōs/ were not only preserved (reflexes of
the first-declension ending are Umbrian -as o -ar and Oscan -as), but extended to the
pronouns by a Proto-Sabellic shift that took the opposite direction as the one in Faliscan
and Latin (cf. Von Planta 1897:227).
4.3. The second nominal declension
4.3.1. The second-declension masculine nominative singular. In Early Faliscan, the

second-declension nominative singular ending is -os, attested in
mama z[e]xtos med f[.f]ịqod EF 1
praụ[i?]os urnam  soc ̣[iai] p ̣orded karai EF 1
perhaps also euios EF 1 (context unclear)
probably ofetios kaios uelos amanos EF 4 (context unclear)
In the later periods, the nominative ending is generally written as -o with omission of -s,
due to the weak realization of /Vs#/ to [h] or [ʔ] (§3.5.7d). I quote here only the
instances of the nominative subject:
124
[---]o cicio cicoi  cupat  ifra MF 40

uel zu[con]|eo  fe [cupa] MF 56
..... : hec : cupa]nt MF 80
[uo]ltio[  ]ueicọno  lecet  ḥec MF 88
[---] celio [---]|[---]* hec  cupa[t] MF 95
cạ[u]io  leụ[eli]o  cau[i] | hileo  ian[ta  ..]lni[a] | hec  cupat MF 146
[leiu]elio [ ---|---]io ca[--- | he c]up[at] MF 159
oufilio  clipeaio  letei  fileo  met  facet MF 470*
precono[?] | cui teneṭ [?] | let MLF 361
uoltio uecineo | maxomo | iuneo he  cupat LF 221
marcio  acarcelinio | cauia  uecinea | hec cupat LF 223
ca uecineo[ ]uoltio | he cupat LF 224
tito[  ]acarcelinio  | ma  fi pop petrunes ce f | [h]e cu[pa] LF 225
tito  uelmineo | titọị  fe cupa MLF 305
perhaps tito | uelmineo | nu i*ice MLF 309 (if i*ice is a verb)
perhaps tito  uel|mineo  iun|aị ị*ice MLF 317 (if ị*ice is a verb)
The remaining instances of -o can be found in §3.5.7d: it is also the normal form in the
Latino-Faliscan and Capenate inscriptions: [---]ọ LtF 173, [---]ilio LtF 215, cḷ[i]peario
LtF 233, *(*)coṇẹo LtF 290, [---]nio LtF 341, ṃunio LtF 377, uomanio Cap 388, fertrio
Cap 391, clanidio Cap 394, anio Cap 420, [---]no Cap 433, genucilio Cap 435.
The only instances of -os from the later periods in Faliscan inscriptions are
cauios frenaios faced MF 471*
cauios MLF 382
(These instances show that, even in Faliscan, /s#/ after a short vowel was only weak-
ened, probably to [h] or [ʔ], and not reduced to zero: see §3.5.7d.) The phonemic form of
the morpheme was therefore still /-os#/.) Beside these two instances, there are the Latin
inscriptions med loucilios feced Lat 268 (probably an import) and t fourios *[ ]f
Lat 216 (probably written by a Latin immigrant), and the Capenate inscriptions a
írpios esú Cap 389 and f pacios Cap 392. These attestations show that in the Middle
and Late Faliscan periods the ending -os was associated with Latin and Capenate rather
than with Faliscan. The nominative in -us is found only in purely Latin inscriptions:
latrius Lat 218, [. u]mpricius ... aburcus Lat 219, spurilius Lat 237, 238, fuluius Lat
250, lectu (twice) and datus Lat 251, all from near S. Maria di Falleri or its surround-
ings, egnatius Lat 291 from near Corchiano, and calpurnius Lat 432, [---]rcius Lat 436,
and didius and uettius Lat 456, all from Lucus Feroniae.
Faliscan therefore, like Latin, preserved the Proto-Italic */-os/ which in the
Sabellic languages was syncopated to /-s/ by the Proto-Sabellic Endsilbensynkope
(resulting, e.g, in Umbrian -s and -s o -r in io-stems, and Oscan -s, -j).
125
CHAPTER 4
There is at least one instance, but possibly even three, of syncopated io-stem
nominatives from the ager Capenas that can be ascribed to interference from Sabellic
languages (see §9.3.2): pa‹qu›is blaisiís Sab 468*, in an inscription that is Sabellic
rather than Faliscan or Latin, and possibly also k pa aiedies Cap 390, if interpreted as
‘K. Aiedius son of Pa.’, and perhaps k sares esú Cap 404 (but cf. §8.8.2). An isolated
Middle Faliscan instance of a syncopated io-stem nominative may be partis in leiuelio
partis | uolti MF 79. There is no reason to ascribe this instance, too, to interference
from Sabellic languages: the form may be compared with the occasional instances of
such syncopation in Latin inscriptions, e.g. uibis pilipus CIL I2.552, mirquris CIL
I2.563, and caecilis CIL I2.1028.
In Latin, second-declension stem nominatives in */-Cros/ were syncopated to
*/-Crs/ o */-Cer/ (type ager) and those in */-Vros/ to */-Vrs/ o */-Vr/ (type puer). The
Faliscan inscriptions provide no attestations of this syncopation (Vetter’s readings
l[o]ụfir, lọuf[ir, or louf[i]ṛ in EF 1 are impossible): cf., however, ucro[---] MF 138,
[---]ro  MF 175 and [---]ro  MF 178, all of which are unclear.
4.3.2. The second-declension dative singular. As explained in §4.4.4 and §8.8.1, many
of the Faliscan forms in -oi can be interpreted both as datives and as genitives. In all
probability datives are
titoi | mercui | efiles MF 113, titoi | mercu[i] | efi[les] MF 116, titoi  mercu[i]
MF 118, [t]ịtoi  mercu[i] MF 119, [ti]ṭoi  mercu[i] MF 120, [ti]toi  mercui
MF 121, titoi  ṃ[e]rcui MF 122; with a damaged ending [t]ịto[i] | ṃercu[i] |
efiles MF 115; also tito (probably = tito‹i›) | mercui | efile MF 114
possibly also locia eiṃoi MLF 293
The contextless forms
caisioi MF 20
tiroi  colanioi MF 69-71
may well be datives too (cf. vultasi Etr XLII and §8.8.1), although they could also be
interpreted as genitives. I regard rather as genitives the forms in -oi in tito  uelmineo |
titọi  fe cupa MLF 305, uoltio | folcozeo | zextoi | fi LF 330 and [. ]folcosio | *****oi
LF 333, The only form in -oi that is quite clearly not an second-declension dative is
cicoi in [---]o cicio cicoi  cupat  ifra MF 40: see §4.4.4. Faliscan therefore had a
dative in -oi that reflected Proto-Italic */-ō/ m PIE */-ō/. In the Sabellic languages,
*/-ō/ was preserved, and possibly (but not necessarily, cf. Meiser 1986:66-8) shortened
to /-o/: South Picene titúi AP.1, Umbrian -e /-ē ̣/, Oscan -ui, -úí). In Latin, the dative in
-oi, attested in numasioi CIL I2.1, duenoi CIL I2.4 and in Marius Victorinus’ remark
“populoi Romanoi pro populo Romano scito priores scribere” (CGL 6.17.20), was
replaced by /-ō/ at an uncertain date, probably not too long after the archaic period,
either by a phonological or a by morphological development (see Villar 1987b). Tito in
126
tito | mercui | efile MF 114 has been regarded as an indication that the same replace-
ment took place also in Middle Faliscan (Thulin 1907:303), but as long as there is no
other evidence for this, I prefer to regard it as an error for tito‹i›. It cannot be ascer-
tained whether Faliscan -oi represents /-ō/ or the shortened /-o/.
4.3.3. The second-declension accusative and neuter nominative singular. The only
form that may be interpreted as a second-declension masculine accusative is tulom MF
72, if interpreted as ‘Tullum’, but this is rather a genitive plural (§4.3.7). The remaining
second-declension accusatives are all neuter, and occur in the following objects:
ceres  far *[0-2]e[1-3]tom ... douiad EF 1
eqo ụrneḷ[a ti?]tela ... arcentelom hutị[.]ilom  pepara[i EF 1
possibly ui[no]m EF 1 (context unclear)
(with omission of -m:)
foied uino pipafo MF 59, foied uino ‹pi›pafo MF 60
The accusative therefore continues PIt. */-om/ m PIE */-om/, which was preserved both
in Latin (-om o -um) and in the Sabellic languages, e.g. Umbrian -u(m), -o(m), Oscan
-úm, -om, -om.
The only certain attestation of the neuter nominative is from Early Faliscan:
eco quto *e uotenosio titias duenom duenas EF 3
Either nominative or accusative are the forms in -om in
propramom  pramed [u]mom pramod pramed umom  pramod propramọḍ
(or pro pramọḍ)  pramod umọ[m] EF 2
Faliscan therefore preserved Proto-Italic */-om/, as did Latin (-om o -um) and the
Sabellic languages, e.g. Umbrian -um, -o(m), Oscan -um, -úm, -om, -om.
4.3.4. The second-declension ablative singular. The ablative singular is attested only
for Early Faliscan, in
propramom  pramed [u]mom pramod pramed umom  pramod propramọḍ
(or pro pramọḍ)  pramod umọ[m] EF 2
This Early Faliscan ablative in -od /-ōd/ reflects Proto-Italic */-ōd/ m PIE */-ōd/. The
same ending is found in Latin (-od o -o), and in the Sabellic languages, e.g. Umbrian
-u m */-ōd/, Oscan -úd, -ud. On this ablative and its syntactic functions, cf. the
interesting notes by Vine (1993:191-213).
There are no attestations from the later periods, where from Middle Faliscan
onwards onwards the expected forms would be -od /-ōd/ o -o /-ō/ (§3.5.7c): Renzetti
Marra (1990:338) suggests that [--- os]tro LF 244 and [---]ostr[o LF 245 could be
ablatives, but I hesitate to adopt this interpretation. For the ablative singular of the first
declension, see §4.2.5.
127
CHAPTER 4
4.3.5. The second-declension vocative singular. The vocative singular is attested once
in Early Faliscan, as
saluẹ[to]d uoltene EF 2
The ending reflects Proto-Italic */-e/ m PIE /-e/. The same ending is found in Latin (-e)
and Umbrian (-e).
4.3.6. The second-declension nominative plural. The only attestation of the nomina-
tive plural is the subject-nominative lete in
lete zot xxiiii MLF 285
where -e renders /-ē ̣/, the result of the monophthongization of */-o/ → /-e/ (see
§3.7.5).65 Faliscan therefore, like Latin, had a nominative reflecting /-o/, originally the
ending of the pronouns, which had replaced the nominal ending */-ōs/ during the Proto-
Latin period: the first-declension nominative plural sociai EF 4 shows that this re-
placement had also taken place in the first declension (cf. §4.2.6). In the Sabellic
languages, the shift operated in the oppositie direction, with the nominal endings /-ās/
and /-ōs/ replacing the pronominal endings /-a/ and /-o/.
Survivals of Proto-Italic */-ōs/ have been read by Norden (1939:206-7) and Vetter
(1953:280) in euios  mama z[e]xtos EF 1 (‘the Evii, Mama and Sextus’), and by Vetter
(1925:26, 1953:284) in ofetios kaios uelos amanos EF 4 (‘the Ufentii, Gaius and Velus,
Amanus’), but, if nominatives at all, these forms in -os are rather a nominative singular
/-os/, not nominative plurals in /-ōs/: see §8.2.1.
4.3.7. The second-declension genitive plural. The genitive plural is attested with
certainty only in a Capenate inscription:
ac ̣iuaiom esú Cap 465
probably also tulo MF 72
Peruzzi (1964a:166) interpreted also arcentelom huti[.]ilom EF 1 as a genitive plural.
The attestations duum]|uiru LF 243 and duu]muiru LF 249 (and duum[uiru LF 247,
duu[muiru LF 248, and duum]uir[um Lat 240) are fossilized rather than paradigmatic
forms, and were in all probability borrowed as such from Latin (§9.4.2).
Faliscan thus shows a form /-om/ that reflects Proto-Italic */-om/ m PIE /-ōm/. In
Latin, /-om/ was gradually replaced by */-ōsom/ o /-ōrom/ (first duonoro CIL I2.9) o
/-ōrum/ modelled on the first-declension genitive plural */-āsom/, which was in its turn
modelled on the pronominal ending */-som/. The Sabellic languages only show the
innovation /-āsom/: in the second declension, the old ending */-om/ was preserved (e.g.
Umbrian -u(m), -o(m); Oscan -úm, -om).
65
I cannot agree with M. Mancini’s (2002:28-33) interpretation of these forms as reflecting a
locative /lē ̣tę̄/ ← */lotā/.
128
4.4. The second-declension genitive singular
4.4.1. The problem. The second-declension genitive singular endings -osio and -i
constitute without doubt the most debated problem of Faliscan and Latin historical
morphology. Basically, the problem is that the Latin morpheme -i /-ī/ was identified
with Celtic -i (an identification that constituted one of the core arguments for the
assumpton of an Italo-Celtic stage) and therefore had to be of PIE origin. The Faliscan
ending -osio on the other hand was clearly identical with, e.g. Sanskrit -asya and
Homeric Greek -oio and therefore had to be of PIE origin as well. The obvious explana-
tion for this was to assume that Faliscan and Latin were separate, though closely related
languages, each preserving a different inherited variant. The occurrence of -i in Middle
and Late Faliscan was ascribed to influence from Latium (thus e.g. Lejeune 1952b:125,
Bonfante 1966:9, and Devine 1970:24).
This explanation was not without its critics, especially when scepticism with
regard to an Italo-Celtic stage increased. Thus, G. Giacomelli (1963:142-4) suggested
that Faliscan might have had both -osio and -i, the former as the nominal and the latter
as the pronominal ending, with -i later being extended to the nouns and generalized. I
might add that it is a priori very unattractive, as it would either assume that Italo-Celtic
had two different morphemes or that Faliscan was not an Italic language, and that it is
quite difficult to assume a spread of a morpheme. The publication, in 1978, of the
archaic Latin inscription from Satricum in 1978 (now CIL 12.2832a) with its attestation
of -osio, showed that the assumption that Latin had -i and Faliscan -osio could no
longer be maintained without at least some modification. The ‘pre-Satricum’ scholar-
ship has been extensively reviewed by Devine (1970), and I shall therefore limit myself
to a discussion of the evidence and of the implications of the Satricum inscription for
the relationship between Faliscan and Latin.
4.4.2. The Early Faliscan genitive in -osio. In the Early Faliscan inscriptions, the
ending of the second-declension genitive singular is -osio throughout. The attestations
are
eco quto *e uotenosio EF 3
eko kaisiosio EF 6
aịṃiosio eqo EF 467*
The first of these instances was published already in 1887, but it was not until 1912 that
Herbig (CIE 8163) proposed to interpret this form as a genitive. Although at first not
generally accepted, this interpretation was eventually vindicated by the publication of
the other forms in 1933 and 1952 respectively. Even after the publication of the third
instance, Knobloch (1954) interpreted the Faliscan forms in -osio as adjectives in
-osio(s) that expressed a specific possessive relationship (i.e., a relationship of posses-
129
CHAPTER 4
sion with one specific person rather than with a group like the gens). This has been
rejected by Devine (1970:25-32), who argues that the existence of such adjectives in
Italic cannot be demonstrated (a point with which I disagree, see §4.4.11), and that it is
in any case unlikely that they would have been derived with the suffix /-oso-/. Apart
from these objections, interpreting -osio as -osio(s) presupposes that omission of -s
occurred already in Early Faliscan, which is not attested (§3.5.7d). Knobloch first
(1954:38-9) ascribed this to dissimilation of the two /s/’s in the ending /-osos/, but later
(1966:48) discarded this explanation by reading quton euotenosio[m] with a similarly
anachronistic omission of -m. The Satricum inscription makes his theory even less
tenable: it could now be maintained only if popliosio ualesiosio | suodales were to be
read as popliosio(s) ualesiosio(s) | suodales, with a nominative plural ending /-ōs/
which in Latin had long been replaced by /-o/ (§4.3.6) and with omission of -s after a
long vowel (§3.5.7d).66
4.4.3. The Middle and Late Faliscan genitive in -i. The forms in -osio are all from the
Early Faliscan inscriptions: in Middle and Late Faliscan the second-declension genitive
ending is -i. The Middle and Late Faliscan instances are the following:
uipi  leueli | filea MF 14
louci  teti  uxor MF 41
caui  felic ̣inate | uxor MF 42
*[.]pi  uesθi  cela MF 83
caui[  ]ṭ**(*)[i]  cela MF 84
cesi fi MF 94
marci  acarcelini | mate LF 221
maci  acacelini  uxo LF 222
uelcẹị || fe LF 332
letei  fileo MF 470*
caui  tertinei  || posticnu MLF/Cap 474*
probably also leiuelio pạrtis | uolti[ MF 79
Ambiguous are the contextless forms in -i in the Besitzerinschriften, serui MF 34-36,
ani MF 45, f ofiti MF 58, anni LtF 63, uli MF? 261-262, caui  turi MF 273, marci 
anel[i] MLF 472*, uolti  catinei MLF 469*, and in the sepulchral inscriptions uolti |
teti MF 11 and teti atron or teti atronị MF 13. These forms can be genitives or abbrevi-
ated io-stem nominatives (§8.8.1). Similar forms are the potter’s signature c ̣ cutri
MF 200, and the Capenate Besitzerinschriften c pscni Cap 387, c aci Cap 395, sex |
senti Cap 399, and sex  senti Cap 430. Uncertain with regard to reading or interpreta-
tion are the forms in -i are [---]*i  u[o]ltiai lo MF 165, cẹsi t  fere (?) MF 263,
66
Note that this instance can hardly be called ‘possessive’, unless this term is used in a very
wide sense (“référent d’une appartenance familiale ou sociale”, Lejeune 1990a:77).
130
oct*i[....]uoltili MLF 354, and c ạu[---]ịsi LF 236. Some of these forms (e.g. uipi
MF 14, caui  tertinei MLF/Cap 474*, serui MF 34-36) have at some time or other been
interpreted as Etruscan feminines, but even if this is possible for some of the forms,
which I greatly doubt (§9.2.2.2a), it is beyond doubt that most of the the forms in -i are
genitives. As is described in §4.2.2, it is probable that in Faliscan, as in Latin, the
genitive in /-ī/ had spread to the first declension as well.
4.4.4. The Middle and Late Faliscan genitive in -oi. As is argued in §8.8.1, a number
of forms in -oi can be interpreted as genitives at least from a syntactical or contextual
perspective. The most interesting case is
This is the only instance where it is impossible to interpret the form in -oi as a dative,
and I regard it as a genitive. (G. Giacomelli (1963:84) interpreted cicoi as a Faliscan
rendering of Etruscan cicui, cf. §9.2.2.2a) A genitive interpretation is also possible in
tito  uelmineo | titọi  fe cupa MLF 305
uoltio | folcozeo | zextoi | fi LF 330
[. ]folcosio | *****oi LF 333
perhaps locia eiṃoi MLF 293
although this would be the only instance of a filiation in a Faliscan Besitzerinschrift
(§7.5.1). Ambiguous are the contextless forms in -oi:
caisioi MF 20
tiroi colanioi MF 69-71
These two forms can be interpreted as genitives in Besitzerinschriften, but can as well
be datives (cf. vultasi Etr XLII): see §8.8.1. In all probability datives are titoi
MF 113, 116, 118, 122, [t]ịtoi MF 119, [ti]ṭoi MF 120, 121, [t]ịto[i] MF 115.
Zextoi, the earliest Faliscan form in -oi to be published, was in fact interpreted as a
genitive by Jordan (1881:511) and Deecke (1888:263). This was rejected by J. Schmidt
(1905:31), however, and since Herbig’s discussion (1914) of the forms in -oi, they have
generally been regarded as datives. The exception to this is Pisani (first 1933b:624 n.1),
who regarded titoi | mercui | efiles MF 113 etc., caisioi MF 20, and cicoi MF 40 as
genitives in /-oī/, in his view a further development of the older /-oso/. I cannot agree
with this (see §4.4.10): if the forms in -oi are genitives, they are analogical formations
after the first-declension genitive -ai, and can be explained in three ways:
(a) -oi = /-oī/ modelled on the first-declension genitive singular in /-āī/, but with
retention of the short vowel (thus Devine 1970:20-1).
(b) -oi = /-ōī/ modelled on the first-declension genitive singular in /-āī/ even in the
length of the vowel preceding the ending, like the Latin second-declension genitive
plural */-ōsom/ was modelled on the */-āsom/ of the first declension.
131
CHAPTER 4
(c) -oi = /-ō/ modelled on the dative /-ōi/ after the example of the first declension
where, in the course of the process of the shortening of the original dative ending */-ā/
and the genitive ending /-āī/, both endings at a one point were both /-ā/.
Note that all three analogies assume the existence of a Faliscan first-declension genitive
singular in /-āī/, which is not accepted by most authors: see §4.2.2. If the replacement of
-osio by -i was due, as I think, to a morphological replacement in both the second and
the first declension at the same time, the resulting dissimilarity between, in the second
declension, old genitive -osio /-oso/ : new genitive -i /-ī/ : dative -oi /-ō/ (: genitive
plural */-ōsom/?) and, in the first declension, old genitive -as /-ās/ : new genitive -ai
/-āī/: old dative -ai /-ā/ (: genitive plural */-āsom/?) may have led to the emergence of
such an analogical genitive in -oi.
4.4.5. The alleged Middle and Late Faliscan genitives in -io and -oio. A number of
forms in -io and -oio were explained by Ribezzo (1930:98-9, 1931b:79, 1933:80,
1936:158) as second-declension genitives, due to a (PIE) confusion between the
genitive /-ī/ and possessive adjectives in /-io/ [sic]. Attestations of these genitives he
found in Faliscan, in Ardeatine (but supposedly Faliscan) titoio 483†, and in Praenes-
tine taseio CIL I2.555. Not only is this derivation impossible, but this theory assumes a
long coexistence of several productive morphemes for one category, in the case of the
io-stems even of a genitive that was homomorphemic with the nominative. In spite of
these objections, the interpretation of (some) forms in -io as genitives was adopted e.g.
by Bolelli (1943:56), Campanile (1961:20 n.19), and Meiser (1998:117, 133). Of the
instances listed by Ribezzo (1930:98-9), the only case where a genitive interpretation of
a form in -io would be attractive is poplia  calitenes | aronto  cesies | lartio  uxor
MF 265, but in Ribezzo’s translation, ‘Publia Calideni Arruntis filia, Caesii Lartis
uxor’, lartio is the genitive of a consonant-stem. In the other cases the forms in -io are
simply masculine praenomina and gentilicia, or are based on misreadings.
4.4.6. The second-declension genitive singular in Latin. The non-archaic second-

declension genitive ending is -i /-ī/. The oldest epigraphic attestations are early third-
century aesclapi CIL I2.440, keri CIL I2.445, saeturni CIL I2.449, and uolcani CIL
I2.453. In literary Latin, -i is the normal ending already in Andronicus (Saturni 2L, 14L,
uerbi 3L, Liberi 30R; Taenari 34R; Nerei 5L, Ulixi 16L). The /-ī/ spread also to the first
declension, where /-āī/ had replaced /-ās/ probably c.300, so that by the end of the third
century the normal form was /-ā/, /-āī/ being used only as metrical convenience
(§4.2.2). The ending was extended to the fifth declension, where the oldest attestation of
the genitive, from the second half of the third century, is rei dinai cau[s]a CIL I2.366.
The attestations from Ennius, magnam quom lassus diei | partem Ann. 236-7V and ille
uir, haud magna cum re sed plenus fidei Ann. 338V, show the disyllabic form /-ēī/ at
the line end; monosyllabic /-ē/ is attested for Terence and Plautus (cf. LHS I pp.445-6).
132
It had long been assumed that Latin preserved traces of the ending /-oso/ (see
below). The direct evidence for the existence of this ending -osio is the inscription
found in 1977 at Satricum, [...]*ei steterai popliosio ualesiosio | suodales mamartei
CIL I2.2832a, dated to c.510-490 by Stibbe (1980a:36-8; it is impossible to agree with
Ferenczy’s (1987) date of c.385). Although there is virtual consensus that the language
of the inscription is Latin, it remains debated what form of Latin (Satrican? Roman?
Latian?) it represents, and consequently how representative the attestation -osio is for
Latin as a whole. However, even in the unlikely case that the Satricum inscription is
Faliscan (thus Lucchesi 2005, see §18.3.3), there is no doubt that the forms in -osio
represent an inherited form: unless it is assumed that Faliscan does not belong to the
Latin branch of the Italic languages, the ending */-oso/ must therefore have existed in
Proto-Latin even if by c.500 it did no longer exist in every Latin dialect.67
Apart from this direct attestation of -osio, Latin has long been regarded as originally
having had a genitive ending */-oso/. The evidence is as follows:
(a) Two epigraphic forms are assumed to reflect later developments of /-oso/. Third-
century Ardeatine titoio 483† has been regarded as a genitive in */-oo/ m */-oso/ e.g.
by Pisani (first 1933b:624 n.1; see §4.4.10), which was one of the grounds on which the
inscription has been regarded as Faliscan. There seems to be nothing against taking
titoio as a nominative, however. Late fourth-century Praenestine taseio CIL I2.555 has
been regarded as a genitive by Ribezzo (first 1930:98) and Dirichs (1934:55, from
*/-eso/), but this difficult form can be interpreted more attractively as e.g. a patronymic
adjective or as rendering of Greek QaseÚj (see Devine 1970:118-28).
(b) Cuius is usually derived from a PIE */koso/ reconstructed on the basis of Sanskrit
kasya. The assumption is that, in Latin, */koso/ became */koo/ by palatalization of
/VsV/, and that this */koo/ was then remodelled by addition of /-s/ after the analogy
of the nominal genitives in /-s/. The resulting /koos/ then developed regulary to
quoius CIL I2.7, and thence to cuius. This derivation of cuius is not without problems,
even apart from the possibility that Sanskrit kasya may not reflect a PIE */koso/ but
instead reflects a transfer of a nominal ending to the pronouns, and the debated issue of
how cuius relates to huius, eius, etc. A Latin development */VsV/ o /VV/ is unlikely
(cf. §4.4.10), and, since in the Latin (and Faliscan) -osio is attested only in nominal
forms, it is unclear why /-oso/ was replaced by /-ī/ in the nouns while pronominal
*/koso/ developed to cuius.
67
Coleman’s (1986:120-2) suggestion that the Satricum inscription is Volscian is untenable.
Suodales (with /d/ ← */dh/) is Latin (but cf. Lucchesi 2005:166-7), and so is steterai: the
Volscian form is sistiatiens VM 2 (=sist{i}atiens or sistiat‹i›ens, cf. Wallace 1985). Alphabet
and ductus are typical of the late archaic Latian inscriptions (Wachter 1987:76-7): contemporary
Sabellic inscriptions are written in the Etruscan or South Picene alphabet (e.g. the fifth-century
‘Satricum hatchet’, see Colonna 1984, Rix 1992b:251).
133
CHAPTER 4
It may be preferable to re-examine other derivations of cuius, especially the

identification with Greek po‹oj, since cuius was used as an adjective e.g. by Plautus (e.g.
nam haec litteratast, eapse cantat cuia sit Rud. 478) and Terence (e.g. uirgo quoiast?
Eun. 321). The same adjective use is found in the case of the Oscan form púiiu Sa 31,
púiieh Cp 41, showing that it may have been original (“antiqui dicebant sicut meus mea
meum sic cuius cuia cuium”, Serv. in Verg. E. 3.1) and have survived as a colloquial-
ism, as appears from its use in comedy and its reflexes in the Romance languages (e.g.
Spanish cuyo).68 If the adjectival use of cuius had been a later development, it is hard to
explain why eius, with its adjective counterpart suus, did not share this development.
(c) Continuously quoted (and rejected) as evidence for the existence of a Latin genitive
ending */-oo/ is Ennius’ Mettoeo¢que² Fufetioeo Ann. 126V, quoted by Quintilian
(Inst. 1.5.10-2). Pace Meiser (1998:133), this is a Homerism, comparable to Lucilius’
Ixionies alochoeo 25M (based on X 317), the uinoeo bonoeo ascribed by Quintilian
(Inst. 8.6.33) to Ovid, and Ausonius’ oÙ…noio bÒnoio Ep. 6.42 Prete. As such, this form
does not require an existing or remembered Latin genitive ending */-oo/.69
4.4.7. The second-declension genitive singular in the Sabellic languages. The

Sabellic languages all show second-declension genitive endings that reflect */-es/ (e.g.
Umbrian -es o -er; Oscan -eis, -eís, -hij), the genitive ending of the i-stems that spread
to the second declension and the consonant-stems (§4.5.2) probably already during the
Proto-Sabellic period. There are no traces of the inherited second-declension genitive
ending, but as the Faliscan and Latin attestations of -osio show that this ending must
have been present in Proto-Italic, */-oso/ may have been the form that was replaced by
*/-es/. The relation of (Proto?-)Sabellic */-es/ to the -es found in Praesamnitic (sixth-
century || bruties (or f ̣rạties) || esum || Ps 4, | ief ̣ies (or ieṿies) || esum  p[.]les  adaries Ps
5, fifth-century luvcies cnaiviies sum Ps 13, cnaives flaviies p Ps 14, (Etruscan?)
mamerces huσinies Ps 11), in South Picene (sixth century a*pies esum TE.4, possibly
also in postin  viam  videtas  | tetis (for teti‹e›s?)  tokam  alies  e|smen  | vepses :
vepe‹t›en TE.2) is debated. I prefer to regard -es as /-ēs/, ̣ a monophthongized form of
*/-es/ (thus also Meiser 1986:20), rather than as due to interference from Etruscan (see
Devine 1970:38-40, Agostiniani 1982:253-8, Joseph & Wallace 1987:683 n.23).70
68
Numitorius’ “dic mihi, Damoeta: ‘cuium pecus’ anne Latinum? non, uerum Aegonis nostri;
sic rure loquuntur” (Don. Vita 17), a quip on Vergil’s “dic mihi, Damoeta: cuium pecus? an
Meliboei?” (E. 3.1), shows that the adjectival use of cuius was not regarded as comme il faut.
69
Ennius may have chosen -oeo for metric reasons, as Mēttī(quĕ) Fūfĕtī would not have fitted
the metre. Note that the text of this passage is quite corrupt: see Devine 1970:12-4.
70
Colonna (1975:165 n.3bis) interpreted the Praesamnitic forms as nominatives where “il
personale dovrebbe fungere grammaticalmente da aggettivo”. A genitive in -es occurs also in
Oscan (herettates  súm Fr 4, kanuties sim Cm 42).
134
A different explanation of Sabellic /-es/ was proposed by Arena (1974), Prosdo-

cimi (1974, 1979:142-4), and Agostiniani (1982:259-60), who regarded /-es/ as a reflex
of /-oso/. In their view, the inherited */-oso/ was first frontalized to */-ese/, an ending
read in Ps 4 and Ps 5, which they read as -esiie sum rather than as -es || esum. This
frontalization of */-oso/ would have taken place under the influence of the //: for this,
Prosdocimi adduced the Praesamnitic forms in -ies quoted above, which he regards as
nominatives in /-es/ m */os/. Even assuming that these forms are nominatives, I am
not convinced that a // could cause such frontalization in Italic, and it remains unex-
plained how it could affect the /o/ in the preceding syllable. From this /-oso/, the */-es/
of the remaining Sabellic languages is derived, either through */-esi/ (loss of /-e/
followed by metathesis)71 or through */-ese/ (metathesis followed by loss of /-e/). Like
the frontalization, the metathesis of /s/ or /si/ is without parallel in the Italic languages,
however,72 and the loss of /-e/ is difficult: note that it apparently did not affect the
Umbrian second-declension vocative ending /-e/. The same authors also read South
Picene TE.4 as a*piese sum, with an -ese whose place in this development is unclear
(*/-ēse/
̣ m */-ese/?). All in all, I do not think that this theory constitutes a convincing
alternative to the communis opinio that the Sabellic second-declension genitive ending
*/-es/ is in fact due to an extension of the i-stem genitive ending.
4.4.8. The second-declension genitive in other languages of ancient Italy. Although

the data provided by the non-Italic languages are irrelevant unless one assumes the
existence of a Sprachbund (cf. §1.3.2.1) that operated also on the morphological level, I
briefly discuss them here because of the evidence that Venetic and Lepontic also shifted
from a reflex of */-oso/ to /-ī/. The genitive in -i has long been recognized in Venetic,
in ego urkli Le 60, [ego o]n[]ti [vh]remais|ti Le 65, entollouki Le 148,
Veneto-Latin ostinobos friui Le 110bis, ceutini | keutini Le 150A-B, and enoni ...
ecupetaris Le 236 (cf. Lejeune 1974:84-95, 1989:72), and is also often read in the
Lepontic inscriptions alkouinos | aśkoneti PID 274, atekua | aśouṇị PID 302, and
[---?]raṇeni | [---?]ualaunal PID 255 (see Devine 1970:54-63). Evidence for an older
genitive in */-oso/ is provided by four archaic forms in -oiso from the Venetic and
Lepontic areas (see Lejeune 1989). Possibly Venetic is the early- to mid-fifth-century
gravestone inscribed padros pompeteguaios and kaialoiso, the latter probably a genitive
(“nom du curateur”, Lejeune 1989:71). Lepontic are an early-sixth-century Besitzerin-
schrift χosioiso, and a fifth-century potter’s stamp and a Besitzerinschrift that both read
71
A similar explanation (/-es/ derived by metathesis from */-esi/ m */-esa/) had in fact been
proposed long ago by Bopp (1857:386).
72
Prosdocimi and Arena adduced the (questionable) metathesis of */-Vsi/ to /-Vs(i)/ proposed
by Kiparsky (1967) for a number of Greek endings, but this is irrelevant as long as it is not
shown that such a palatalization could also occur in Italic.
135
CHAPTER 4
plioiso. Venetic and Lepontic therefore show a replacement of an archaic -oiso, clearly
reflecting an older */-oso/, by a genitive in -i /-ī/. At least in Venetic this replacement
can be dated to between the (mid)-fifth century, the date of the kaialoiso-inscription,
and the third century, the date of the oldest Venetic attestations of -i.
The evidence from the other languages is unclear. The Messapic forms in -aihi
have long been recognized as second-declension genitives, but it is unclear whether this
ending represents /-aī/ m */-oī/ or something else, e.g. Pisani’s /-aihi/ m */-oso/ (see
Devine 1970:42-53). The evidence from the Sicilian languages consists of sixth-century
adiomis | raroio and adaioi, both of which were interpreted by Paino (1958:168) as
genitives in -oio in accordance with Pisani’s theory (for which see §4.4.10), and of
Lejeune’s (1989:73) reading of the late-sixth- or early-fifth-century Ve 186 as touti
kemai poterem ‘vase de Toutios pour Kema’.73
4.4.9. The relation between /-oso/ and /-ī/. Resuming the sections §4.4.2-8, I
conclude that Early Faliscan and archaic Latin show a genitive in -osio /-osio/ reflecting
a PIE */-oso/, but from the late fourth century onwards in Faliscan and from the third
century onwards in Latin the ending of the genitive singular is -i /-ī/, and that this
ending therefore replaced the ending /-oso/ somewhere in the fifth or fourth centuries.
In the Sabellic languages, the unattested inherited second-declension ending (perhaps
*/-oso/?) was replaced by the i-stem ending */-es/ probably already during the Proto-
Sabellic period. The significance of this chronological sequence -osio ... -i in Faliscan
and Latin was called into question by Untermann (1964:178-9). He assumed that
already in Early Faliscan the genitive was /-ī/, and that /-oso/ was no longer a case-
ending but an isolated and eventually disappearing form used exclusively in Besitzerin-
schriften to express possessive relationship. As /-oso/ was even in his view originally
undoubtedly a genitive ending, however, this amounts to saying that the original ending
/-oso/ was replaced by /-ī/, not between c.450 and c.350, but at some earlier (prehis-
toric) date. Neither is this theory supported by the popliosio ualesiosio | suodales of the
Satricum inscription, which can hardly be called possessive in the sense this word has
when applied to a Besitzerinschrift. De Simone (1980:82-3) therefore suggested a
syncretism of a possessive in -osio and a genitive in -i (the latter ending, having the
wider scope, eventually prevailing), which requires a period during which both endings
could be used for both categories. This, however, assumes the existence of a separate
possessive form in early Latin, something I at least am unwilling to accept.
The replacement of -osio by -i has been interpreted both as a phonological
development and as a morphological replacement: this is discussed in §4.4.10.
73
Note that Ve 186 is from the area traditionally referred to as Elymian, whereas the other two
are from the area traditionally ascribed to the Sicels. It is therefore uncertain whether the two
inscriptions represent the same or two different languages or dialects.
136
4.4.10. The ending /-ī/ as a phonological development of /-oso/. The idea that /-ī/
directly reflects /-osio/ goes back to the early nineteenth century. In recent times it has
been maintained by Pisani (1933b:620-4, 1934, 1935:167, 1937:235-6, 1955,
1964:344-5, 1981), whose theory is adopted e.g. by Safarewicz (1955:103-5) and
Wachter (1987), and independently by Must (1953:303-4, from */-eso/). Pisani’s
theory has been formulated in several ways. The earliest formulation runs as follows:
“Come *quoso-s ha dato quoiios, *luposo doveva dare anzitutto *lupoo [...], questo
*lupee (cfr. ›peo - sequere, Ó - iste ipse), onde *lupiī lupī: e mediano seguìto da conso-
nante doppia resta di norma e, ma qui la doppia era un suono i, cui seguiva un i, e
l’assimilazione, che ha luogo anche per vocale accentata (cfr. *cenis > cinis, *ne mīs >
nimīs, *vegil > vigil [...]) doveva a maggior parte subentrare nel presente caso; che Pom-
peiius faccia al voc. e gen. Pompeī e non *Pompī, è dovuto al nom. Pompeiius dat.
Pompeiiō ecc. Per la finale e > ī cfr. il voc. fīlī da *fīle [...] e sopratutto il voc. Pompeī.
[...] Quanto ad iī > ī cfr. flagitī gen. di flagitium ecc.” (Pisani 1933b:623).
For Faliscan, Pisani (first 1933b:624 n.1) assumed a slightly different development in
which the thematic vowel was retained, and this version of the theory was later applied
also to Latin. In this variant, the stages of the development are as follows (after Pisani
1964:344-5): */-oso/, attested in Faliscan euotenosio EF 3, kaisiosio EF 7, and aịṃiosio
EF 467* becomes */-oo/, attested in Ardeatine titoio 483† and required by the
derivation of cuius from */koso/ (see §4.4.6), and /-oo/ in its turn becomes /-oī/,
attested e.g. in Faliscan caisioi MF 20, tiroi colanioi MF 69-71, and titoi | mercui |
efiles MF 113 etc., which is then contracted to -i /-ī/.
This idea has been rejected so often that I shall only briefly touch on the most difficult
points: an extensive discussion may be found in Devine 1970:93-105.
(a) A development /-oso/ o /-oo/ is not in line with the attested Italic or Latin
palatalizations, which involve mainly dental and velar occlusives, e.g. peius m
*/pedos/ and maius m */magos/. Meiser (1998:117) likewise does not quote any
examples of this palatalization other than cuius and titoio. The usual Latin development
of /VsV/ (and the comparable /VsV/), however, appears to be rhotacism, not palatali-
zation, cf. Venerius m */eneso-/ (Devine 1970:98) and haurio m */has-/ (Meiser
1998:117, explaining it as due to analogy with hausi haustum).
(b) Titoio as an attestation of /-oo/ is questionable, since the word occurs in isolation,
and although isolated dialects may of course preserve older stages of a development, it
would be surprising to find -oio in third-century Ardeatine if the replacement of -osio
by -i is assumed to have taken place (as Pisani (1981:139) suggested) in a Sprachbund
embracing not only Latin and Faliscan, but also Messapic, South Picene and Venetic.
(c) Cuius cannot constitute an additional argument for a development /s/ o //, since
in Pisani’s theory it involves the same morpheme: it rather requires an explanation why
the phonological development of /-oso/ had different outcomes in the nouns and in the
pronouns.
137
CHAPTER 4
(d) The steps */-oo/ o */-ee/ o */-ie/ o */-iī/ are questionable, and the contrac-
tion of */-ee/ or */-ie/ cannot be parallelled by filie /fīle/ → fili /fīlī/; the alternative
*/-oo/ o */-oī/ is even more unlikely. Meiser (1998:72) in fact suggests that /-oso/
wold have developed to */-ese/ due to vowel weakening (quoting ›peo : sequere).74
(e) Although some of the Faliscan forms in -oi can in my view be interpreted as
genitives (§4.4.4, §8.8.1, §8.10.2), this is difficult in the case of titoi | mercui in MF 113
etc.: all the more as in Pisani’s interpretation this is a genitive of Titus Mercuuius,
which requires the assumption of different endings for the second-declensions and the
io-stems (whereas in tiroi colanioi MF 69-71 both have the ending -oi).
(f) It is unclear whether */-eī/, */-iī/, or */-oī/, could in fact be contracted to /-ī/
without an intermediate diphthongal stage, whereas the genitive ending /-ī/ cannot go
back to a diphthong: see §4.4.11.
4.4.11. The ending /-ī/ as a reflex of PIE */-iĦ/ (*/-iħ2/?). It has long been realized
that the genitive ending /-ī/ cannot go back to a diphthong. Broadly speaking, */o/ and
/e/ were monophthongized in Latin as /o/ o /e/ o /ē/̣ o /ī/, a process that seems to
have started at least in some dialects already in the fourth century (e.g. socie CIL I2.5
from the Lacus Fucinus). The last stage, the merger of /ē ̣/ with /ī/, took place only
c.150, and neither the spelling i for original */e/ or */o/ nor the hypercorrect spelling ei
for original /ī/ are found before this date. Whereas e.g. the second-declension nomina-
tive plural (originally */-o/) or the consonant/i-stem dative singular (originally */-e/)
are normally written as -ei and -e, there is not one instance from before c.150 of the
genitive ending being written as -ei or -e. Also, in the io-stems the endings containing
an original diphthong, e.g. the locative singular and the nominative plural, are virtually
always found uncontracted as -iei, -ie, or -ii, whereas before the first century the
genitive of these words is always found as -i (for an analysis of the material, see Devine
1970:5-9). Note also that in Faliscan the merger of /ē ̣/ with /ī/ never took place at all
(§3.7.5), but that the genitive ending is always spelled as -i.
If /-ī/ cannot go back to a diphthong, or to a form that was contracted during the
historic period, it may well reflect */-iĦ/. As the development */VĦ/ o *// is to be
dated to the Proto-Italic period (§3.2.3), this /-ī/ must have been present already in
Proto-Italic in some form or other. It can hardly already have been a second-declension
genitive ending, as is assumed e.g. by Schrijver (1991:361-2), for this would require
that Proto-Italic and Proto-Latin preserved two apparently productive morphemes
within one and the same category and declension for a very long time (millennia,
perhaps). There are no indications that the ending /-ī/ belonged to a different inflection
and/or that it was originally the ending of another case than the genitive. The possibility
74
Note the leucesie in Terentius Scaurus’ (CGL 7.28.11) unfortunately garbled quotation
from the Carmen Saliare.
138
that /-oso/ was the Latin ending and /-ī/ an ending that belonged to another language
and was borrowed into Latin (as well as Faliscan) by the end of the sixth century can
also be excluded, both because of the a priori unlikelihood of such a borrowing due to
the borrowing hierarchy constraint (§1.3.2.2) and because there are no parallels for
Latin borrowing a declensional morpheme from another language. Apart from that,
there appears to be no language that could have been the source for this borrowing.
The two main theories with regard to the original morphological category of /-ī/
are those proposed long ago by Wackernagel and by Sommer. Wackernagel (1908)
proposed that the origin of /-ī/ was to be found in the Indo-Iranian adverbial forms in -i
(cvi-forms) occurring in compounds with the roots kr ‘to make’, bhu ‘to become’, and
as ‘to be’ with the meaning ‘to make/become/be what is expressed in the first member
of the compound’ (abhūtatadbhāve kbhvastiyoge saṃpadyakartari cviḥ, Pāṇini
5.4.50). According to Wackernagel, the original use of /-ī/ in Latin was in expressions
like lucri facere, which he assumed to have originally meant ‘zum Gewinn machen’.
The form was then at a certain point regarded as a genitive and /-ī/ as a genitive ending.
Although popular for a long time, this theory succumbed under A. Bloch’s (1960)
critical review, although a variant was still defended by Blümel (1970:109). Not only
were the functions and the syntactic contexts in which the cvi-forms operated such that
they could never have developed into expressions like lucri facere (for which a
meaning ‘zum Gewinn machen’ is in any case doubtful), but it is likely that the Indo-
Iranian forms in -i had emerged within in Indo-Iranian, if not within Vedic itself.
Today, the most popular theory still seems to be the one proposed originally by
Sommer (1902:371 n.3), who equated the genitive in /-ī/ with the Indo-Iranian feminine
nominative singular in -ī in the declensional type devī. Here, too, the main problem is
how to come from a nominative to a genitive. It is not necessary to go back to “an early
IE where there was no opposition between a derivational and an inflexional suffix”
(Devine 1970:109): both devī and the Latin genitive in /-ī/ can be derived from an
feminine adjective in */-iħ2/. If, as Beekes suggested, the PIE nominative of this type
was */-iħ2/ and this developed to */-ī/ in Proto-Italic (see §4.2.1), this provides the only
possible instance of an Italic morpheme */-ī/ that could be in some way responsible for
the genitive in -i. The two main problems of the transfer of this /-ī/ to the second-
declension genitive are that it is necessary to assume the existence (1) of ‘specific’
possessive adjectives and (2) of a category were */-ī/ was preserved and developed into
a genitive, while in the nominative it was reformed to /-ia a/.
The problem of the existence of ‘specific possessive adjectives’, that is, adjectives
used to denote a possessive relationship with a specific individual rather than with a
group or class, has been discussed by Devine (1970:24-34), who concluded that such
adjectives are not attested at all in Italic. Most or even all Italic languages, however,
appear to have had patronymic adjectives, either still productive, as in Faliscan, or
fossilized as gentilicia. I find it difficult to agree with Devine’s assumption that these
139
CHAPTER 4
patronymic adjectives were not possessive: in a society where children were in the
manus of the pater familias to the extent that they could legally be sold or executed, the
relationship of a father to his son or daughter can hardly be given any other semantic
label than ‘possessive’. Neither can it be maintained that in e.g. Seruios Tulios the
adjective Tulios expressed generic possession (‘a Tullan Servius’): the essence of a
patronym is that it denotes someone as the child of a specific father (‘Servius son of
Tullus’, that Tullus we all know). The problem is a different one, namely that there are
no indications that one could say equos Tulios as easily as Seruios Tulios: in all Italic
instances of ‘possessive relationships’ dating from before the fourth century, this
relationship is expressed by the genitive, e.g. Early Faliscan 2, 6, and 467*, archaic
Latin 479†, CIL I2.2832a, Praesamnitic Ps 4 and Ps 5, South Picene TE.2 and perhaps
TE.4, and possibly also Palaeovolscian VM 1. This, however, is partly a problem of
interpretation, for a sixth-century Besitzerinschrift Tulios or Tulia would a priori be
more likely to be interpreted as a nominative ‘Tullius’ or ‘Tullia’ than as an adjective
‘Tullian = belonging to Tullus’: note that kaṇaios in eqo kaṇaios 482†=CIL I2.474 has
been interpreted as a possessive adjective by Wachter (1987:92-3).
The second problem, the assumption that in certain contexts what was in fact an
ia-stem nominative in */-ī/ was preserved instead of being remodelled to /-a -ia/ is even
graver. With great hesitation, I venture to suggest that a likely candidate would be the
neuter plural nominative-accusative collective, which was homomorphemic with the
feminine nominative singular described above, in such phrases as
*/tullī esti/ (m */tulliħ2 ħ1esti/) ‘it’s Tullus’s things’ (lit. ‘it’s Tullian stuff’)
*/tulloso esti/ ‘it’s Tullus’s’
It is then necessary to assume that these forms in */-ī/ were for some reason no longer
regarded as paradigmatic before the Proto-Italic remodelling of the feminine nominative
singular and the neuter nominative-accusative plural */-ī/ to /-ia -a/ and survived in
some very specific niche of the morphological system until they replaced the genitive
ending in the fifth and fourth centuries.
In every theory (including my own suggestion) it remains in any case unclear why
/-ī/ replaced /-oso/, for in both theories the morpheme /-ī/ originally only played a very
minor role in the morphological system. The only one to address this problem recently
is Peruzzi (1978b:346-7), who suggested that when /-oso/ had become /-orio/ by
rhotacism (cf. what was said in §4.4.10 on the development of /VsV/!), it became less
distinguishable from the adjective ending /-ōrios/ which by this time was realized as
[-ōrio]. I do not think that this can be upheld: even if /s#/ was completely reduced to
zero, which I doubt very much (the realization was rather [-ōrioh] or [-ōrioʔ], see
§3.7.5d), /-orio/ and [-ōrio] or /-ōrio/ would (pace Peruzzi) still be distinguished by the
difference in vocalic quantity, which is a phonemic difference, not ‘just’ a difference in
phonetic realization.
140
4.5. The third nominal declension
4.5.1. The nominative singular of the consonant- and i-stems. There are few
attestations of the third-declension nominative singular, but enough to show that
Faliscan in this form corresponded to Latin.
(1) Stems in occlusives. The attestations of occlusive-stem nominatives are limited.
First, there is the dental-stem Arruns:
aruz MF 257; arute MF 269
The form aruz probably represents [arrũs] = /arruns/, although it could also represent
[arrũts] = /arrunts/ (cf. §3.3.4, §9.2.2.1). Arute MF 269 is either an Etruscan nominative
with an epenthetic [-e] or [-ə] or an accusative arute(m) used as a nominative, probably
the latter (§9.2.2.4). Apart from Arruns, there is the velar stem rex:
rex MF 90, ṛẹx MF 91, rex LtF 231
As in all Italic languages, these occlusive stems had a sigmatic nominative. From a
morphophonological perspective, the Faliscan forms, with /-ns/ m */-nts/ and /-ks/,
correspond to those of Latin: in the Sabellic languages, these nominatives were
assimilated differently (Umbrian -f and -s, Oscan -s and -ss).75
(2) Stems in liquids. The only attestations of stems in /l/ are the Etruscan names
Tanaquil and Vel:
θanacuil MF 49, tancuil MLF 347,
uel MF 56, 82, and perhaps ue]l MF 191
Stems in /r/ are more frequently attested, and apart from the Etruscan name Veltur
consist of words of IE origin:
ueltur MF 266, MLF 339
pater MF 62
uxor MF 41, 42, 101, 265, [u]xor MLF 301
p[reto]|r LF 243, [pre]tor LF 248
cen]|sor LtF 232
[---]or LtF 233
(with omission of -r:)
mate MF 220
uxo MF 17, LF 221, 242, LtF 300
ce]so LtF 230, censo LtF 232
perhaps also quto EF 3, if this represents /gtor/ or /gttor/ (see §6.2.31)
75
Paelignian pristafalacirix and sacaracirix Pg 9, and Marrucinian lixs MV 1 may represent
developments where these languages differed from Oscan and Umbrian.
141
CHAPTER 4
(with -d instead of -r:)

cuestod LF 242
pretod LF 242, LF/Lat 214
Both the stems in /l/ and the stems in /r/ appear to have had an asigmatic nominative in
Faliscan, as in the other Italic languages. The omission of -r and its spelling as -d are
phonological, not morphological, variants: see §3.5.7b.
(3) Stems in nasals. More interesting is the nominative of the nasal stems, which shows
a clear contrast between Latin and the Sabellic languages. The only certain attestations
are ōn-stems:
apolo MF 65
cupi‹d›o MF 62
perhaps also quto EF 3, if this represents /gtō/ or /gttō/ (see §6.2.31)
Apolo (adapted either from Etruscan Apulu or from Greek 'ApÒllwn), and cupi‹d›o
MF 62 both show a nominative in -o /-ō/. Although MF 62 has been regarded as Latin
inscription (in which case there would be no purely Faliscan attestations of the ōn-stem
nominative), in my view the inscription can be counted as Faliscan: see also Wachter
1987:367-9. Faliscan therefore had an asigmatic nominative singular in the ōn-stems, as
did Latin. This constitutes a marked contrast with the Sabellic languages, where these
stems had a sigmatic nominative */-ōns/ o /-ōf/ (Umbrian -u(f), Oscan -úf).
(4) Stems in /s/. Of the attestations of the nominative singular of the stems in /s/, neither
appears to be completely representative of this group. The first attestation are the
theonym Ceres in
ceres  far *[0-2]e[1-3]tom  *[3-5]uf[1-4]ui[..]m  *[3-4]*ad EF 1
The nominative ceres was originally */keres/, but in Latin this appears to have had an
irregular lengthening of the vowel in the last syllable /kerēs/ (§6.4.2), probably as it was
a personification. Whether this lengthening took place in Faliscan as well cannot be
established.
Another attestation of an s-stem is
mino LtF 173
which can represent either /mins/, with omission of the original /s#/ (§3.5.7d), or
/minor/ with omission of an /r#/ (§3.5.7b) that had replaced the /s#/ after the oblique
cases where the intervocalic /s/ had been rhotacized, as in Latin honos → honor after
honoris etc. Note that, in Latin, in the case of the name Minor the spelling Mino is the
rule rather than the exception: see §3.5.7b.
(5) Stems in /i/. The i-stem nominative is attested in
ortecese MF 339
larise MF 270, 371, 372
142
Although ortecese may well be connected to Etruscan urtcsnas Etr XXXV, it appar-
ently represents /ortikensis/, with an ending -e [-ẹh] or [-ẹʔ] /-is/ (§3.6.2), a reflex of
Proto-Italic */-i-s/. Latin had the same -is, occasionally also spelled as -e(s) (e.g.
milita]re CIL I2.48, militare CIL I2.49, aidiles CIL I2.8); in the Sabellic languages, */-is/
was syncopated to /-s/ by the Endsilbensynkope, with various assimilations of the
resulting /-Cs/. Larise is either an Etruscan nominative with an epenthetic [-e] or [-ə]
after an /s#/ that was realized more strongly than in Faliscan, or an accusative larise(m)
used as a nominative (§9.2.2.4). If the latter, it shows an Etruscan name in -is apparently
being declined according to the consonant-stems, not according to the i-stems.
genitives in -os or -us genitives in -es or -is
early third century [---]erus CIL I2.2885 iunone{ne}s CIL I2.444

salutes CIL I2.450
ueneres CIL I2.451
[---]es CIL I2.2884b
third century nationu CIL I2.60
diouo CIL I2.60
third/second century salutus CIL I2.62 cereres CIL I2.973
diouos CIL I2.360 diouis CIL I2.361
? artoro CIL I2.126
? usoro CIL I2.346
second century nominus CIL I2.581,7
kastorus CIL I2.589,1
Fig.4.1. Third-declension genitive endings in Latin before c.150.
4.5.2. The third-declension genitive singular. In Faliscan, the consonant-stem genitive

ending is -os throughout:
lartos EF 6
apolonos EF 10
loifiṛtato MF 31, loifirtato MF 32,
aruto MF 257, 266, [--- a]ruto MF 169, aronto MF 265
perhaps also [....]o MF 17, and [---]ono MF 102
G. Giacomelli (1963:148) recognized only loifiṛtato MF 31 and loifirtato MF 32 as
genuine attestations, regarding the other forms as transcriptions of Greek and Etruscan
forms. She first (1963:147) regarded the Faliscan genitive in -o(s) as one of the main
morphological differences between Faliscan and Latin, but later (1978:517-8) compared
the Faliscan forms with the Latin genitives in -os/-us. These are not as sporadic as they
143
CHAPTER 4
are sometimes made out to be: before c.150, they occur as frequently in the epigraphic
material as the forms in -es/-is: see fig.4.1. Both -os and -is are also found in the fourth
declension, -uos in early second-century senatuos in CIL I2.581,8, 17, 21, 23 (cf. also
zenatuo LF/Lat 214), and -uis in Terence (Haut. 287), and, according to Gellius (4.16),
in Nigidius and Varro.
It is usually assumed that Latin /-os/ and /-es/ reflected different PIE forms */-os/
and */-es/. If this was the case, the attestations would imply that in Faliscan /-os/ was
standardized, while in Latin both forms were preserved side by side for several
centuries longer, /-es/ eventually becoming the standard form. The assumption that
Proto-Latin had two productive morphemes for the same category, which were pre-
served for at least several centuries in Faliscan and even longer in Latin, is unattractive,
however. Neither are there any indications that the two were (originally) dialectal, let
alone diglossic, variants, as R. Giacomelli (1978:57-9) suggested.
Various solutions have been proposed to this problem: for a discussion, see
Wachter 1987:492-3. In my view, the easiest solution is to assume that /-os/ reflects
Proto-Italic */-os/ and that /-es/ was a Latin innovation. This was first suggested by
Szemerényi (1969:977-8, later (1989:173) abandoning the idea), and independently by
Beekes (1986:176-80, altogether rejecting a PIE */-es/) and Wachter (1987:492-5). The
last two assume that /-es/ was an analogical creation of the early third century: “neben
den mutmasslichen Paradigmata -ād/-am/-ai/-ās (neben -āī), -ūd/-um/-uē ̣/-ọ̄s, -īd/-im/-ē ̣
/-is (bzw. *-ē ̣s?) nahm sich -e/-em/-ē ̣/-os m.E. so merkwürdig aus, dass ein Ersatz des
-os durch -es regelrecht nahelag!” (Wachter 1987:495).76 Assuming that until the third
century Latin had only /-os/ does away with the coexistence of /-os/ and /-es/, and also
removes a morphological difference between Faliscan and Latin. In any case, Faliscan
aligns with Latin here, for in the Sabellic languages the inherited consonant-stem
ending was replaced by the i-stem ending */-es/, probably already in Proto-Sabellic. 77
There are also two attestations of the i-stem genitive singular:
felic ̣inate MF 42
[---]fate MF 285
It could be argued that felic ̣inate and [---]fate are not representative of Faliscan, as both
forms end in an Etruscan suffix, but in view of the ending of [fel]ịcinatiu LF 384
(§4.5.3.3) and the way such forms were declined in Latin, I think it is safe to regard
them as Faliscan. The ending -e can of course stand for [-ēḥ ] /-ēs/,
̣ the expected reflex of
Proto-Italic */-es/ found in Latin (-eis, -es /-ēs/̣ o -is /-īs/) and the Sabellic languages
76
The frequent use of the ending -os/-us in official documents and in the names of gods implies
that speakers of Latin regarded it as more traditional and associated it with an earlier period.
77
If South Picene ---]nips TE.7 is a syncopated consonant-stem (Marinetti 1985:133), the
transfer of */-es/ to the consonant-stems could be dated to the sixth or fifth century.
144
(Umbrian -es /-ē ̣s/ o -er /-ēr/, ̣ Oscan -eís, -eis, -hij), but as omission of -s after a long
vowel is rare in Faliscan (§3.5.7d), it is more likely that -e represents either [-eh] or [-eʔ]
/-es/ or [-ẹh] or [-ẹʔ] /-is/ (§3.6.2). In that case, /-es/ or /-is/ may originally have been an
i-stem ending which at some point was transferred to the consonant-stems (perhaps first
at Rome, see Wachter 1987:495).
As far as can be ascertained from these few instances, the distinction between the
consonant- and the i-stem genitives was still quite clear in Middle Faliscan.
4.5.3. Other consonant- and i-stem endings. Of the other third-declension endings,
there are few and sometimes dubious attestations:
(1) The consonant-stem accusative singular. There only uncertain attestations of the
consonant-stem accusative, namely
arute MF 269
larise MF 270, 371, 372
These forms have been interpreted as nominatives with an epenthetic [-e] or [-ə], but
may be accusatives in -e(m) used as nominatives: see §9.2.2.1,4. In that case, Faliscan
would align with Latin in having an ending /-em/ m */-/, for in the Sabellic languages
the consonant-stem accusative ending had been replaced by the second-declension
ending /-om/ (e.g. Umbrian -um, -om, Oscan -om).
The only Faliscan attestation of a neuter consonant-stem accusative is far EF 1,
reflecting Proto-Italic */bhars/ (← PIE */bhĦrs/ or */bhars/, see §6.2.23) with an
assimilation */rs#/ → */rr#/ → /r#/ (§3.3.4).
(2) The i-stem nominative plural. The i-stem nominative plural is attested in
efiles MF 113, 115, efil]es MF 117
perhaps salues EF 3, if from an i-stem adjective */sali-/ (§6.2.71)
(with omission of -s:)
efile MF 114
Since omission of -s after a long vowel is fairly rare in Faliscan (§3.5.7d), efile MF 114
is probably an error rather than an omission on phonological grounds: the inscription
contains another irregularity in the ending second-declension dative singular (§4.2.3).
The ending -es /-ēs/ reflects Proto-Italic */-ēs/ m PIE */-ees/ (§3.2.6), found also in
Latin (-es) and the Sabellic languages (Umbrian -es o -er; Oscan -ís).
(3) The i-stem genitive plural. The i-stem genitive plural is attested in
The form reflects the PIE ending /-i-om/ with the closing of /o/ →/u/ in closed final
syllable (§3.6.6.1), as attested also for Latin (-iom o -ium). The same ending appears in
the Sabellic languages (Umbrian -io(m), Oscan -íúm).
145
CHAPTER 4
4.6. The fourth and fifth nominal declensions
4.6.1. The fourth-declension genitive singular. The only instance of a fourth-

declension genitive singular is de | zenatuo sententiad LF/Lat 214. The ending -uo(s)
[-uoh] or [-uoʔ] /-uos/, formed after the consonant-stem genitive ending /-os/ (see §4.5.2)
is also found in early second-century Latin senatuos CIL I2.581,8, 17, 21, 23. It is an
analogical innovation existing alongside the usual Italic genitive */-os/ (PIt. */-os/ m
PIE */-os/) that is reflected both in Latin (senatous CIL I2.2197, -us) and in the
Sabellic languages (Umbrian trifor TI VIb.54 etc., Oscan castrous TB 13). In the case of
zenatuo, it cannot be ascertained whether the ending -uo(s) was the standard Faliscan
ending or an alternative existing beside an unattested */-ō ̣s/ m */-os/. It is not even
clear whether -uo(s) is a Faliscan form at all: the language of LF/Lat 214 can equally
well be regarded as Latin, In view of the consonant-stem /-os/ it is not unlikely that the
ending /-uos/ occurred in Faliscan, but the ending occurs here in a formula that may be
of Latin origin, in which case it may have been taken over as part of the formula (note
that the Latin instances of the genitive senatuos also occur exclusively in this formula)
and need not have occurred in Faliscan outside this formula.
4.6.2. The fourth-declension dative singular. The development of the Italic dative
singular endings is debated: I follow here mainly Lejeune’s (1944:99-102) account (for
other theories, see LHS I pp.442-3). The Proto-Italic form was probably */-o/, an
analogical creation beside the genitive */-os/ after the i-stems, which had a dative
*/-e/ beside a genitive */-es/ (Lejeune 1944:92-101):
i-stems u-stems
──────────────
genitive */-es/ */-os/
↓
dative */-e/ → */-o/
Reflexes of this */-o/ are Latin -u /-ū/ and Umbrian trifo TI VIIa.11, manuve TI IIb.23,
and perhaps ahtu TI IIa.10, 11. The Latin ending -ui /-ūī/ goes back through /-uē/̣ to an
/-ue/ (cf. senatuei CIL I2.586,11) which was an analogical formation after the conso-
nant-stem dative /-e/ (Lejeune 1944:100).
The only Faliscan forms that have plausibly been interpreted as fourth-declension
datives are the forms in -ui in
mercui MF 113, 114, 121, ṃ[e]rcui MF 122, [m]ẹrcui MF 124, ṃercui MF 125
damaged: ṃercu[i] MF 115, mercụ[i] MF 116, 118, mercu[i] MF 119, 120
Although a diphthong // is admittedly alien to Italic, it is hard to imagine that in this
form, and probably also in its Oscan parallel mirikui Cm 12 (which, as it predates the
146
introduction of ú and í, may represent an second-declension dative in /-ō/), -ui repre-

sents anything other than /-u/ or /-ū/, an ending created after the first-declension dative
/-ā/ and the second-declension dative /-ō/. The ending does not reflect Proto-Italic
*/-o/, nor can it be equated with Latin -ui /-uī/ m /-uē ̣/ m */-ue/, as in Faliscan /ē ̣/ had
not (yet) merged with /ī/ (§3.7.5).
4.6.3. The fourth-declension accusative singular. The fourth-declension accusative

singular is attested only in
mạcistratu LF 242
The form occurs as part of the formulaic expression mạcistratu | keset, which is directly
equivalent to the Latin formula magistratum gessit (TLL 62.1939,1-1940,56). On the
one hand, this makes it quite certain that mạcistratu is indeed an accusative singular
mạcistratu(m) and not an accusative plural mạcistratu(s) [-ūh] or [-ūʔ] /-ūs/ (which
would have been difficult in any case, as omission of -s after a long vowel is fairly rare
in Faliscan, §3.5.7d): on the other hand, it opens up the possibility that the entire
formula was adopted from Latin and that the word and its ending may therefore not
reflect Faliscan. The same ending is found in Latin and in the Sabellic languages (e.g.
Umbrian trifo TI VIb.58, VIIa.47).
4.6.4. The fifth declension. The only Faliscan forms that can be associated with the
fifth declension do not reflect the original *eħ1-stems, but the paradigm of dies that was
formed analogically after that of these *eħ1-stems (Schrijver 1991:366).
The nominative is attestated in [3-4]s pater 62. The text can be restored as [die]s
pater, [iouo]s pater, or [ioui]s pater: if restored as [die]s, the form would correspond to
Latin dies /diēs/, a Proto-Italic formation with an /ē/ that was either due to the accusa-
tive */dēm/ m Proto-Italic */dem/, or to preservation of a PIE */dēs/ (cf. Wachter
1987:151-2, Waanders 1988:57). No Sabellic counterpart is attested, although Oscan
zicolom TB 14 etc. appears to have been derived from */dē-kelom/, which likewise had
/ē/. Note that MF 62 has been regarded as Latin, although it is in my view Faliscan: see
also Wachter 1987:367-9.
The ablative singular is attested in foied MF 59-60. This adverb goes back either
to a fossilized ablative */hōd+dē(d)/ or */ho-dēd/ (cf. §6.2.34). This */dēd/ is also
found in third-century Latin e]od died CIL I2.2872 (vs. die in CIL I2.366, the other
version of this text) and Umbrian ri TI Va.5, re TI VIIb.2 m */rēd/. It reflects a (Proto-
Italic?) */-ēd/ formed after the second-declension ablative */-ōd/ and/or the first-
declension ablative /-ād/. In view of the other attestations of this ablative in /-ēd/, it is
not necessary to assume that the /-d/ was not part of the ending of the ablative on
which the adverb was based, but was added only when the ablative */-ē/ had been
fossilized into an adverb, by analogy with the adverbs in /-ōd/ and /-ād/ that derived
their -d from a second- or first-declension ablative (cf. G. Giacomelli 1963:150).
147
CHAPTER 4
4.7. The personal pronouns
4.7.1. The nominative of the first person singular. The nominative of the pronoun of
the first person singular is attested in the Faliscan ‘iscrizioni parlanti’ as eqo/eko/eco:78
eqo ụrneḷ[a ..]tela fitaidupes  arcentelom hutị[.]ilom  pe  para[i? EF 1
eko lartos EF 6
eko kaisiosio EF 7
m adicio eco MLF 378
eco tulie MLF 383
The Faliscan form therefore corresponds to Latin ego (first seventh- or sixth-century
eqo 482† and eco 479†, then sixth century CIL I2.479 and 2917c), which represents an
/egō/ that later became /ego/. If this is due to iambic shortening, which is attested for
Latin only from the late third century onwards (Pfister 1977:104, Meiser 1998:76-7), it
seems unlikely that such a shortening can be assumed for Middle or Late Faliscan. The
same */egō/ occurs also in the Sabellic languages. The evidence for this is as follows:
(a) South Picene ekúsim CH.1. As ekúsim corresponds to rufrasim in the same text, it is
clearly an enclitic -sim ‘I am’ preceded by an ekú that may represent */eg/, but could
conceivably be a form of the demonstrative pronoun */eko-/ (cf. WOU s.v. ekúsim).
(b) Samnitic íív Sa 31. The unique íív occurs in pis  tiú  íív  kúru Sa 31, usually
interpreted as ‘Who [are] you? – I [am] a kora’, but íív representing /ē/ is very difficult
to explain on the basis of an original */egō/ (see WOU s.v. íív).
(c) Umbrian ef TI VIa.4. This has unconvincingly been interpreted as equal to íív
(above) by Vetter (1935:188, 1942:60-1, 1953:231). Untermann (WOU s.v. eite) rightly
rejects this explanation.
(d) The nominative of the pronoun of the second person singular attested as /tēom/ in
Samnitic tiú Sa 31 and Oscan tiium Cp 37. On the assumption that /-om/ in continues a
PIE suffix */-om/, W. Petersen (1930:168 n.13) and Bonfante (1935:183) suggested that
the nominative of the first person may have been */egm/ (cf. WOU s.v. tiium).
4.7.2. The accusative of the first person singular. The accusative singular is attested
in the Early and Middle Faliscan ‘iscrizioni parlanti’ as med/met:
med EF 1, EF 9
met MF 470*
78
I do not adopt Peruzzi’s (1964d:310-1) interpretation of the ...uatụ..eco... read by Gamurrini
(1883:166) and Deecke (1888:145-6) in MF 91 as containing an eco.
148
(Latin rather than Faliscan is med  loucilios  feced Lat 268.) The form med is also
found in Latin (first probably seventh- or sixth century m]ẹḍ in 479†, then sixth- or
fifth-century med in CIL I2.4, 2658). In Latin, it coexisted with the similar accusatives
of the second person singular ted /tēd/ (first sixth-century ted in CIL I2.4) and the third
person reflexive sed /sēd/, and with the homomorphemic ablatives med ted sed /mēd tēd
sēd/. The unattested Faliscan accusatives of the second person singular and the third
person reflexive may therefore likewise have been */tēd sēd/, and, depending on the
view of how these forms arose, the Faliscan ablatives may have been */mēd tēd sēd/
(which would fit in with other Faliscan ablatives in -d, for which see §4.2.5, §4.3.3).
In Latin, both the accusatives and the ablatives lost the /-d/: from the third century
onward, me te se */mē tē sē/ were the normal forms both in the epigraphic (first me CIL
I2.500, 501, te CIL I2.412a,c, 547) and in the literary attestations (from Andronicus
onwards), although med, ted, and sed occasionally occur in second-century inscriptions
(e.g. sed CIL I2.581,13, 14, CIL I2.582,21) and for metrical convenience in the works of
Naevius, Ennius, and Plautus. A similar loss of /-d/ may also be assumed to have taken
place in the Middle Faliscan period: the occurrence of met MF 470* may well be a sign
of this weakening (see §3.5.7c), the spelling with -t perhaps influenced by the facet that
immediately follows it, where -t was a recent replacement of -d.
As the corresponding PIE accusatives are usually reconstructed as */ħ1me te se/, the
Latin and Faliscan forms require some explanation with regard to the origin of the /-d/
and of the long vowel. Various explanations have been proposed:
In the explanation proposed by Meillet (1922:50, whence e.g. DÉ s.v. mē), the
/-d/ was an inherited PIE suffix with parallels in Vedic mád and tvád: this presupposes a
Proto-Italic */med ted sed/ that was lengthened to /mēd tēd sēd/, although it is not made
clear when and why this lengthening would have taken place. The explanation is
difficult to maintain if the suffix /-om/ in the corresponding Sabellic accusatives (see
below) is also assumed to be of PIE origin, for this presupposes that Proto-Italic
preserved two different morphemes for the accusative of the personal pronouns.
The Latin forms have also been explained as analogical creations after the
ablatives /mēd tēd sēd/ (that probably were analogical creations themselves, after the
ablatives of the first and second nominal declensions). In the traditional version of this
explanation (e.g. Osthoff 1884:127, W. Petersen 1930:185), the accusatives are
explained as hypercorrective forms of */mē tē sē/ that emerged when the ablatives /mēd
tēd sēd/ were losing their /-d/: this, however, is impossible in view of the Early Faliscan
and early Latin accusatives with -d, which precede the disappearance of /-d/ by several
centuries (cf. §3.5.7c). If this explanation is to be maintained, the accusatives can only
be analogical creations after /mēd tēd sēd/ themselves: this would also explain the /ē/.
Szemerényi (1973:58, 1989:226) derived the accusatives /mēd tēd sēd/ in an
altogether different way, by assuming that /mēd/ and /sēd/ were analogical creations
after /tēd/, which, through */tēt/, was the regular phonological outcome of a redupli-
149
CHAPTER 4
cated */tēte/ (described awkwardly (1973:58) as “the sequence of emphatic tē and

unemphatic te”). There are several reasons why I find this solution unappealing, the
most important being (1) that it does not explain the origin of the /ē/; (2) that the later
attestations of tete imply that the word had always been analyzed as a reduplicated
form, which would have prevented it from becoming */tēt/; (3) that */tēt/ could have
become /tēd/ by a regular process only if this happened at the same time that the
secondary ending of the 3rd person singular */-t/ became /-d/: this process, however,
must have preceded the drop of the word-final short vowel by which */tēte/ became
*/tēt/, or the primary ending of the 3rd person singular, */-ti/, would also have become /-
d/, which it did not; (4) that there is no reason why the pronoun of the second person, let
alone the reduplicated form, should be the model for an analogical creation.
Whatever their origin, the Faliscan and Latin accusatives */mēd tēd sēd/ clearly differ
from the corresponding forms in the Sabellic languages. The accusative of the first
person singular is attested in Palaeoumbrian setums  míom | face Um 4=480†,79 of the
second person singular in South Picene tíom TE.5, Umbrian tiom TI VIa.43 etc. (more
than 40 attestations), and of the third person reflexive in Oscan siom TB 5, 6, 9. These
accusatives /mēom tēom sēom/ are usually regarded as accusatives /mē tē sē/ (← PIE
*/h1me tue se/), with an /-om/ that is either an inherited PIE suffix */-(Ħ)om/ (cf. e.g.
Vedic ahám) or taken over from the second-declension accusative singular ending/-om/
(cf. WOU s.vv. miom, siom, tiium).
4.7.3. The nominative of the second person plural. The second person nominative/
vocative plural is attested once from Early Faliscan as
ues EF 4
This form represents either /es/, with a short vowel as in the possessive pronoun, or
/ēs/, with a long vowel as in Latin uos /ōs/. It presents a considerable problem, for
although it corresponds to later Latin uester and Umbrian uestra TI VIb.61, it clearly
differs from the earlier Latin uoster and from Paelignian uus Pg 9 (twice), which seem
to point to a Proto-Italic */s/ and a possessive */ostero-/. Vetter (1939a:153,
1953:287) explained ues as an ‘incidental’ form, formed with the nominative plural
ending of the i-stems /-ēs/ ← */-ees/. The use, even incidental, of i-stem endings in the
personal pronouns is unparallelled, however.80
79
This renders untenable La Regina’s interpretation, hesitatingly adopted by Marinetti (1985:
104), of South Picene ma in ma kuprí koram opsútạ ninis rakinevíi p̣omp̣[4-5]í AQ.2 as a first
person accusative singular.
80
Latin ques CIL I2.581,3, 24 adduced by Vetter is not a valid parallel as qui actually is at least
partly an i-stem. The (second- and first-century) nominatives heis, his etc. are regular o-stem
nominatives with an added -s, maybe a graphical convention only (Bakkum 1994).
150
Others have pointed to non-Italic parallels for ues. Campanile (1968:90) adduced
Old Irish sí, Welsh chwi, derived from */swēs/; R. Giacomelli (1978:65-6) compared
Pisani’s derivation of Gothic izwis from */eswes/. These parallels are debatable,
however, and, more importantly, they are irrelevant unless either Faliscan is treated as a
non-Italic language or the relation of these forms to the forms in the other Italic
languages is made clear.
A solution from within the Italic languages has been proposed by Ribezzo
(1936:166-7) and Peruzzi (1967b:118-9), who pointed to Umbrian uestra TI VIb.61.
Pace R. Giacomelli (1978:65), this cannot reflect an earlier */ostrād/, as there is no
evidence that the development /o-/ o /e-/ that was responsible for Latin uoster o
uester occurred also in Umbrian: cf. e.g. early ku-vurtus TI Ib.11 and late co-uortus TI
VIIa.39 etc. The possibility that uestra is a Latin loan (LHS I p.466) can be excluded,
for a borrowing at the level of the pronominal system would imply language contact on
a scale unattested by the Umbrian texts (cf. §1.3.2.2). Umbrian uestra, therefore, like
Faliscan ues, would seem to point to an earlier */s estro-/.
The problem, however, is greater than Faliscan ues: the evidence points to the
existence of both an early */s estro-/ in Faliscan and Umbrian and to an early */s
ostro-/ in Latin and Paelignian, i.e., both vocalisms are attested for both branches of
Italic: since neither can convincingly be explained as a borrowing from the other, both
would therefore presumably going back to Proto-Italic, as neither form can convinc-
ingly be connected to corresponding forms in other IE languages. A way out of this
dilemma would be if it could be demonstrated that the one of the two vocalisms
(probably the /o/-vocalism of the Latin and Paelignian forms, both of which are later
than Faliscan ues) is due to a later development. Perhaps one vocalism originally
belonged to the personal pronouns and the other to the possessive pronouns, and this
anomaly was equalled out in various ways in the various languages; or was an original
/s estro-/ changed dut to influence from /ns nostro-/?
4.8. The demonstrative pronoun
There are no direct attestations of demonstrative pronouns. However, there are two
important indirect attestations in the adverbs:
(a) hec ‘here’: hec MF 94, 146, 158, ḥec MF 88, hẹc ̣ LF 223, [and heic LtF 231]; he LF
220, 221, 224, [h]e LF 226; he[ MF 149; (with hypercorrect f- for h-) fe[(c?) MLF 56,
fe MF 305 derived from a pronominal locative */he-ke/;
(b) *hoied ‘today’: (with hypercorrect f- for h-) foied MF 59-60, probably derived from
a fossilized ablative phrase */hō(d)+dē(d)/, or, alternatively from a compound */ho-
dē(d)/ (cf. §6.2.34).
151
CHAPTER 4
These attestations show that Faliscan had a demonstrative pronoun corresponding to

Latin hic instead of to its Sabellic counterparts */eko-/ and */ekso-/. With regard to the
declension, hec and hoied point to an old locative */he-ke/ and perhaps to an old
ablative */hō(d)/, while the pronominal origin of the nominative plural ending /-a/ of
the first nominal declension (cf. §4.2.6) makes it likely that the feminine and masculine
nominative plurals of this demonstrative pronoun were */ha/ */ho/ in Early Faliscan,
and probably */hę̄/ and */hē/̣ in the later periods.
4.9. The relative pronoun
A relative pronoun cui is read by Renzetti Marra (1990:337) in [---?] precono[

?--|---?] cuiteneṭ[ ?---|---?] let MLF 361. I very much doubt whether cui is indeed a
relative pronoun. If it is, the following tenet seems to suggests that cui is a nominative
nominative */kī/ (ultimately from */ko/): I cannot accept Renzetti Marra’s suggestion
that the form can be a locative /kī/ (ultimately from */ke/) or a dative /kuī/ (ultimately
from */koe/). Note, however, that all these interpretations presuppose a merger of
(/o/ →) /e/ → /ē/̣ with /ī/ that is not attested in Faliscan (§3.7.5) and would be very
early even in Latin, where this merger operated c.150.
4.10. Summary of §§4.2-9
If the preceding sections are resumed with regard to the question of whether Faliscan
should be classed with Latin or with the Sabellic languages, the result is the following.
(a) Faliscan participated in the common Proto-Italic innovations of the first- and the
fifth-declension ablatives /-ād/ (§4.2.5) and /-ēd/ (§4.6.4), modelled on the second-
declension ablative /-ōd/.
(b) Of the phonological changes that affected the endings, typical Sabellic develop-
ments like the rounding of word-final /-ā/ (§4.2.1) or the Endsilbensynkope (§4.3.1) are
not found. Faliscan also aligns with Latin with regard to the results of the assimilations
of */-Cs/ in the consonant-stem nominatives (§4.5.1). As in Latin, monophthongization
of // does not seem to have affected the first-declension dative and genitive singular
endings, unlike Umbrian and Volscian (§4.2.2-3). Loss of /-d/ affected the ending of the
ablative singular from the mid-third century onwards, as in Latin, but unlike the
Sabellic languages that either preserved it or lost it much earlier (§4.2.5, §4.3.4, §4.6.4).
152
It is more interesting to look at the morphological innovations. The following innova-

tions are prehistoric, and, as they are shared with Latin but not with the Sabellic
languages, probably Proto-Latin:
(c) As in Latin, the first- and second-declension nominative plural endings */-ās/ and
*/-ōs/ were replaced by the pronominal endings /-a/ and */-o/, whereas in the Sabellic
languages the opposite transfer took place, the endings /-ās/ and /-ōs/ being extended to
the pronouns (§4.2.6, §4.3.6).
(d) As in Latin, the nominative singular of the ōn-stems was asigmatic and lost its /n#/,
whereas in the Sabellic languages the nominative of these stems was sigmatic */-ōns/
(§4.5.1).
The other significant differences can be ascribed to the historical period, and therefore
definitely to the period where Latin and 3the Sabellic languages were separated.
(e) As in Latin, the second-declension genitive singular /-oso/ was replaced by an /-ī/
that reflected PIE */-iĦ/ (perhaps */-iħ2/) between the first half of the fifth century and
the second half of the fourth century, whereas in the Sabellic languages the unattested
inherited form (*/-oso/?) was replaced by the i-stem genitive ending */-es/ probably
already during the Proto-Sabellic period, and at the latest in the sixth century (§4.4.2-7).
(f) As in Latin, the first-declension genitive singular /-ās/ was probably replaced by /-āī/
during the late fourth and early third century, whereas in the Sabellic languages /-ās/
was preserved (§4.2.2).
(g) The consonant-stem genitive singular was /-os/, which was probably the original
form also in Latin, whereas in the Sabellic languages the unattested inherited form was
replaced at an early date by the i-stem genitive in */-es/ (§4.5.2).
(h) The i-stem genitive singular reflected either Proto-Italic */-es/, as in Latin and the
Sabellic languages, or was /-es/ or /-is/, an innovation that is found also in Latin
(§4.5.2).
(i) The fourth-declension genitive singular shows an innovative form /-uos/ that is
known also from Latin: it is unclear whether this form was the standard Faliscan ending
(§4.6.1).
The only differences between Latin and Faliscan are in the second- and fourth-
declension dative singular:
(j) The inherited diphthongal second-declension dative singular was preserved in
Faliscan, either as /-ō/ or as /-o/, as in the Sabellic languages, whereas in Latin it was
replaced by /-ō/ after the archaic period (§4.3.2).
(k) The fourth-declension dative singular appears as /-/, which may have a parallel in
Oscan; there are no attestations of */-o/ or */-ue/ (§4.6.2).
153
CHAPTER 4
The only significant difference between Latin and Faliscan in the nominal declensions
is therefore the preservation of the diphthongal second-declension dative /-ōi/ or /-o/,
which is of historic date and which, contrary to the Latin innovation /-ō/, left the
‘horizontal’ perspicuity of the first- and second-declension paradigms intact. The
different formations in the fourth-decelension dative cannot, I think, count very heavily,
although this is admittedly partly due to the lack of data. In my view, the conclusion can
only be that as far as can be established from the lacunary material, Faliscan aligns
closely with Latin on virtually every significant morphological opposition between
Latin and the Sabellic languages.
The only regular point of comparison provided by the personal pronouns is the first
person singular accusative med, which clearly links Faliscan to Latin med ted sed /mēd
tēd sēd/, as opposed to the very different Sabellic /mēom tēom sēom/ (§4.7.2). The
Latin-Faliscan forms either preserved a PIE suffix */-d/ (common retention) or may
have been new formations after the ablative (common innovation). The indirect data on
the demonstrative pronouns shows that Faliscan had a pronoun with the same stem as
Latin hic, where the Sabellic languages used */eko-/, */ekso-/, etc. (§4.8). The relation
of Faliscan ues to Latin uos and uoster/uester, Paelignian uus, and Umbrian uestra and
its implications for the position of Faliscan remains unexplained: the vocalism of these
forms poses a problem that comprises both branches of the Italic languages and cannot
be solved satisfactorily due to a lack of data (§4.9).
154
Chapter 5
The verb
The verbal forms attested in the Faliscan inscriptions are few and limited in range, but they
provide material of great interest for the study of the Italic verbs, since most Faliscan forms are
of a relatively early date. They therefore figured prominently in the studies by Herbig (1913b)
and Lejeune (1955). Recently, interest has been rekindled by the publication of the Middle
Faliscan perfects faced MF 470* and facet MF 471*. The chapter opens with a short remark on
the problems encountered in the study of the Faliscan verb (§5.1). The Faliscan verb is then
discussed, first its general structure (§5.2), then the individual forms (§5.3). The chapter
concludes with a short comparison of Faliscan with Latin and the Sabellic languages (§5.4) on
the subject of the verbs.
5.1. The verb: methodological issues
Surprisingly perhaps, the main problem in the evaluation of the Faliscan verb is not the
lack of material, nor the necessity to exclude the Latin material (cf. §4.1, §6.1, §8.1): in
spite of the not overabundant data, the forms that are attested give a rather good general
view of the Faliscan verb, even providing material for a comparison with Latin and the
Sabellic languages. The problem is rather that much of the material is Early Faliscan,
and that Early Faliscan forms are difficult to compare both to the later material and to
the contemporary Latin or Sabellic material. This has led to comparisons between Early
Faliscan and e.g. third-century Latin that sometimes resulted in an erroneous picture of
how the Faliscan verb relates to the Latin verb. This is especially true in the cases of the
forms fifiked EF 9 and f[.f]ịqod EF 1.
5.2. The Faliscan verb
5.2.1. Conjugational system. The overall structure of the Faliscan verb follows the
four-conjugational system, as in Latin and in the Sabellic languages.
(a) first conjugation: cupat MF 40 etc. (for the attestations, see §5.3.1.2), perhaps also
subj. prameḍ, pramed EF 3, and imp. (?) urate Etr/EF 385.
(b) athematic laryngeal verbs (probably included in the first conjugation): present not
attested, but pipafo MF 59, pafo = ‹pi›pafo MF 60 (stem in */-ħ2/) and porded EF 1
(stem in */-ħ3/) belong in this category.
155
CHAPTER 5
(c) second conjugation: saluete EF 4, saluetod EF 3, salueto EF 4, teneṭ LF 361, and

probably lecet MF 88 (if not a third-conjugation form), also carefo MF 59, carẹ[f]o
MF 60.
(d) third conjugation: present not attested, except possibly for lecet MF 88 (which is
probably a second-conjugation form); fifiked EF 9, f[.f]ịqod EF 1, keset LF 242, and
kese[t LF 243 also belong in this category.
(e) mixed conjugation: probably douiad EF 1 (§5.3.1.4); the presents of faced MF
471*, facet MF 470*, and of pepara[i EF 1 are not attested, but will have belonged
in this group as well.
(f) fourth conjugation: not attested: douiad EF 1 could be a fourth-declension form,
but is probably rather of the mixed conjugation (§5.3.1.4).
Outside the conjugational system there is the verb ‘to be’, attested in esú Cap 389,
404, 465, zot MF 285 (and perhaps sot LtF 172?), and seite (=s{e}ite or s‹ie›te) EF 4.
5.2.2. Formation of the tenses. The only tenses attested in the Faliscan material are
present, perfect, and future: as might be expected from the nature of the material,
imperfect, pluperfect, and future perfect are not attested.
(1) Present tense. See §5.2.1.
(2) Future tense. The future is attested only in pipafo MF 59, pafo = ‹pi›pafo MF ̣,
and carefo MF 59, carẹ[f]o MF 60. These forms show a suffix -f- /-f-/ ← */-β-/: the
same suffix occurs in Latin as -b- /-b-/, the productive suffix of the future of the first
and second conjugations. Both suffixes reflect a Proto-Latin innovation */-β-/,
modelled on the imperfect suffix */-β-/ ← Proto-Italic */-bh-/ (which also occurs once
in the Oscan pluperfect fufans Cm 1A,10).
Beside this, Latin also had an ā/ē-future in the third and fourth conjugations,
going back to the original subjunctive. Although not attested for Faliscan, it seems not
unlikely that Faliscan likewise had this future (see §5.3.1.13). In the Sabellic
languages, the future was formed with an s-suffix (probably a continuation of the PIE
aorist subjunctive or desiderative): there is no trace of either the ā/ē-future or the b-
future.
(3) Perfect tense. There are no attestations of Faliscan perfect forms with productive
perfect suffixes such as Latin /-- -u-/ or Sabellic /-f-/ and /-tt-/. The attested
formations are:
(a) reduplicative perfect: fifiked EF 9, f[.f]ịqod EF 1 (/fifig-/, see §5.3.1.7-8) and
probably pepara[i EF 1 (/pepar-/ ← Proto-Italic */peprħ3-/).
(b) either reduplicative perfects that lost their reduplicative syllable or old aorists:
porded EF 1 (see §5.3.1.14) and faced MF 471*, facet MF 470* (see §5.3.1.6).
156
THE VERB
(c) sigmatic perfect: keset LF 242, kese[t LF 243. The sigmatic perfect, which
continues the old sigmatic aorist, occurs in Latin but not in the Sabellic languages. 81
Keset and kese[t occur in the formula magistratum gero that may have been borrowed
from Latin, but it seems unlikely that this means that keset was Latin rather than
Faliscan (cf. §9.4.2).
5.2.3. Formation of the subjunctive. Forms representing moods other than the
indicative are attested only for Early Faliscan. The ā-subjunctive appears in douiad
EF 1 and probably in *[3-4]*ad EF 1, tulas EF 385, and tulate EF 385 (urate in the
same inscription is perhaps rather an imperative, see §5.3.1.23). The ē-subjunctive
may be attested in prameḍ, pramed EF 2, if this is a verbal form. Both these old
subjunctives are found in Latin as well as in the Sabellic languages. As in Latin and
the Sabellic languages, the subjunctive of the verb ‘to be’, appearing in seite (= s{e}ite
or s‹ie›te, §5.3.1.18) EF 3 is a continuation of the old optative.
5.2.4. The endings. Note that the primary endings are all attested only for Middle and
Late Faliscan, and the secondary and imperative endings only for Early Faliscan.
Passive endings are not attested at all.
(1) Primary endings (attested for Middle and Late Faliscan only):
1st sg. : -o /-ō/ (← Proto-Italic/PIE them. */-oħ2/), in fut. ind. pipafo MF
59, pafo=‹pi›pafo MF 60, carefo MF 59, carẹ[f]o MF 60;
-m /-m/ (← Proto-Italic/PIE athem. */-mi/), in ind. pr. of ‘to be’,
esú = esú(m) Cap 389, 404, 465
2nd sg. —
3rd sg. : -t /-t/ (← Proto-Italic/PIE */-ti/), in pr. ind. cupat MF 40, cupat
MF 220, cupat LF 224, lecet MF 88; (with the -t omitted:)
cupa MLF 305, cupa LF 221; (either singular or plural:)
[cuba LtF/Lat 326]; (restored:) c]up[at] MF 159, cup[a]
MF 161, cupa[?t] MF 95, cu[pa] LF 226
1st pl. —
2nd pl. —
3rd pl. : -nt /-nt/ (← Proto-Italic/PIE */-nti/), in pr. ind. cupa]nt MF 80,
cupat MF 146, cupạṭ MF 158, cupat LF 223; (either
singular or plural:) cupa[?t] MF 95, cu[pa] LF 226 [and
cuba LtF 326].
These endings do not differ from the corresponding endings in Latin and the Sabellic
81
It has been suggested that the Umbrian future perfect sesust TI VIa.5 is a sigmatic perfect,
but this seems unlikely (cf. WOU s.v. sistu).
157
CHAPTER 5
languages. The endings of the third person reflect the Proto-Italic drop of */-i/
following the voicing of word-final occlusives (§3.2.4) that caused the contrast
between the primary endings of the third person /-t -nt/ and the secondary endings
(see below).
(2) Secondary endings (attestated for Early Faliscan only):
1st sg. —
2nd sg. : -s /-s/ (←Proto-Italic/PIE */-s/), in pr. subj. tulas EF 385
3rd sg. : -d /-d/ (←Proto-Italic/PIE */-t/), in pr. subj. douiad EF 1,
probably also in *[3-4]*ad EF 1 (can also be plural),
perhaps also in prameḍ, pramed EF 2
1st pl. —
2nd pl : -te /-te/ (←Proto-Italic/PIE */-te/), in pr. subj. tulate Etr/EF 385,
pr. subj. (continuing an older optative) seite EF 4
rd
3 pl. : -(n)d /-nd/ (←Proto-Italic/PIE */-nt/), perhaps in pr. subj. *[3-4]*ad
EF 1 (if this form is plural and not singular)
Early Faliscan still shows a clear distinction between the primary and the secondary
endings. So, in all probability, did early Latin, but there the primary endings were
later generalized throughout (except in the first singular), and the distinction between
primary and secondary endings thus largely disappeared (cf. Meiser 1998:216-7). The
date of this replacement is unclear: judging by the few epigraphic attestations of Latin
forms where secondary endings can be expected (e.g., 3rd sg. fut. esed CIL I2.1, 3rd sg.
pr. subj. sied CIL I2.4, 3rd sg. pf. vhevhaked CIL I2.3, feced CIL I2.4), the replacement
seems to have started after the fifth century, and to have been completed by the third.
In Faliscan, it was certainly under way by the late fourth century (see below under
(e)). In the Sabellic languages, on the other hand, the distinction between the primary
and secondary endings was preserved. It is therefore all the more unfortunate that the
Faliscan secondary ending of the third plural can be regarded as dubiously attested at
best, for this ending provides a clear contrast between Latin, where it was replaced by
the primary ending /-nt/, and the Sabellic languages, where it was replaced at a Proto-
Sabellic date by a new secondary ending /-ns/ (cf. Shields 1980).
(3) Imperative endings (attested for Early Faliscan only):
2nd sg. —
3rd sg. —
2nd pl. : -te /-te/ (←Proto-Italic/PIE */-te/), in saluete EF 4, probably also
in urate Etr/EF 385 (which could also be a subjunctive)
3rd pl. —
This ending is the same as the corresponding ending in Latin; the corresponding
Sabellic ending is not attested.
158
THE VERB
(4) Future imperative endings (attested for Early Faliscan only):

2nd sg. : -tod /-tōd/ (←Proto-Italic/PIE */-tōd/), in salue[to]d EF 3
3rd sg. —
nd
2 pl. : -tod /-tōd/ (←Proto-Italic/PIE */-tōd/), in salueto EF 4 (see below)
3rd pl. —
The same ending is found in Latin (still archaising licetod and datod beside exuehito
and exferto in CIL I2.366) as well as in the Sabellic languages (Oscan líkítud Cm
1B,10, Umbrian -tu). In EF 4 the ending -to = -to(d) or -to‹d› is used for the plural
(see also §5.3.1.17). This may reflect an inherited paradigm where the future
imperative either had only one ending /-tōd/ (or /-tō/, Prosdocimi 1990:304-5), or had
a singular and second plural ending /-tōd/ beside a third plural ending */-ntōd/
(Szemerényi 1953:946): the endings of the second plural, Latin -tote /-tōte/ (formed
after the imperative) and Umbrian -tuta, -tutu, -tuto /-tōtō/ ← */-tōtā/, are post-Proto-
Italic formations. The ending of the third plural, Latin -nto /-ntō/ ← */-ntōd/ (the
corresponding Sabellic ending is not attested),82 may be inherited or a new formation.
(5) Perfect endings. The endings of the perfect are attested for the Early, Middle and
Late Faliscan periods, and thus offer some insight into the development of the perfect
endings in the Italic languages.
1st sg. : -ai /-a/ (innovation on Proto-Italic */-a/ ← */-ħ2a-/ ← PIE
*/-ħ2e/), in pepara[i EF 1
2nd sg. —
3rd sg. : -e-d /-d/ (= them. secondary ending), in porded EF 1, fifiked EF 9;
faced MF 471*;
-et /-ē ̣t/ (innovation containing the primary ending /-t/), in facet
MF 470*; keset LF 242
(with missing or omitted ending:) kese[ LF 243 and
possibly i*ice LF 309, 315 (see §5.3.1.9)
1st pl. —
2nd pl. —
3rd pl. : -o-nd /-nd/ (= them. secondary ending), in f[.f]iqod EF 1
The first singular ending is -a[i /-a/, a reformation of the original inherited perfect
ending, Proto-Italic /-ħ2a/ ← PIE /-ħ2e/, with an /-i/ that may have been derived from
the primary endings (Untermann 1968a:165-9): the same ending is found in Latin.
The date of this reformation was either Proto-Italic (*/-ħ2a-/ or */-a-/) or Latin-
Faliscan (*/-a-/): as the Sabellic languages appear to show no sign of the old singular
82
It has been suggested that Oscan eítuns Po 27, 29, 31 etc. shows a third plural future
imperative ending -tuns /-tōns/ (cf. WOU s.v. eítuns): this ending would then be modelled on
the secondary ending /-ns/ (a Proto-Sabellic innovation) and also be a recent formation.
159
CHAPTER 5
perfective endings (cf. below, note 83), this cannot be established with certainty. The
Faliscan form is the only attestation of the stage /-a/: the Latin attestations show only
the later stages of its phonological development, -ei /-e/ → -e /ē/̣ → -i /-ī/ (first ueixei
CIL I2.14 and petiei beside accumulaui, genui, optinui CIL I2.15, in the elogia
Scipionum). In the Sabellic languages, there is only one attestation of the first singular
perfect, in Oscan [man]afum Cp 37,1 manafum Cp 37,3, where the ending is /-o-m/,
the (secondary) ending of the thematic aorist. Herbig (1914:238 n.1) interpreted tulom
MF 68 as having this ending as well, but this is unlikely (see §5.3.1.22).
In the third singular perfect, there was a shift in the endings. Early and Middle
Faliscan show forms that continue the (secondary) ending of the thematic aorist
ending -e-d (← Proto-Italic/PIE */-t/) in Early Faliscan porded EF 1, fifiked EF 9, and
Middle Faliscan faced MF 471*, but the Middle and Late Faliscan forms facet MF
470* and keset LF 242 show a different ending. A similar shift occurs in Latin. The
early Latin attestations of the third singular perfect, vhevhaked CIL I2.3 and feced CIL
I2.4, also show -e-d, and this ending also shows up in a contaminated form -id in the
much later fecid (beside dedit) in CIL I2.561. The older -d, however, was replaced by
-eit /-et/ → -et /-ē ̣t/ → -it /-īt/ → it /-t/.83 This ‘new’ ending represents the older
perfect ending */-e/, i.e., the inherited PIE */-e/ with a prehistoric addition of an /-i/,84
that was recharacterized with the primary ending /-t/ to /-e-t/: the Latin forms dede
CIL I2.47, 2438 and fece CIL I2.416 may in fact reflect the ending /-e/ (Untermann
1968a:169-70). The earliest Latin attestations of the new ending -et appear to be from
the middle and the second half of the third century (cepet CIL I2.25, dedet CIL I2.48,
CIL I2.49; for other attestations of the variants, see the indices to CIL I2).
This situation differs from the one in the Sabellic languages. There, as in Early
and Middle Faliscan and early Latin, the ending of the third singular perfect was the
old aorist ending -e-d (in some languages, such as Umbrian, reduced to zero), but
contrary to what happened in Latin and Faliscan, this ending was maintained and not
replaced by a new ending. An exception to this is formed by a handful of southern
Oscan forms, afamatet Lu 18 (beside afaamated Lu 6), dedet Lu 19, and anafaket Lu
18. Since this -et cannot represent /-e-t/ or /-ē-t/, ̣ these forms must represent a
replacement of the secondary ending -d by the primary ending -t, perhaps connected
to a weakening or desonorization of word-final /-d/.85
83
Like Wachter (1987:270), I do not think that in Latin this replacement was due to a
weakening of word-final occlusives, although this may have contributed to the process.
84
The date of this addition is unknown: if Palaeoumbrian face Um 4=480† represents the
inherited perfect ending /-e/ rather than /-e-d/ or /-ē ̣/ ← */-e/ and is representative of the
Sabellic languages in general, then */-e/ must have been a post-Proto-Italic innovation. All
other Sabellic attestations of the third singular perfect point to /-e-d/, however.
85
Note, however, the forms aflkeit (= a‹na›f‹a›keit?) Lu 13 and the unexplained liokakeit Lu
39, which apparently have an ending /-et/.
160
THE VERB
The Middle Faliscan forms faced and facet are both from the late fourth century,
showing that the replacement of -ed by -et was in progress by that time. The nature of
the new ending is not clear: -et can represent either /-ē ̣t/, in which case the new ending
is the same one as in Latin, or /-e-t/, with a replacement of the /-d/ by the primary
ending /-t/ similar to the one in southern Oscan. In my view, the Faliscan replacement
is best regarded as connected with the replacement of the ending in Latin, and -et is
therefore probably /-ēt/.
̣ In Faliscan, the weakening of word-final occlusives (§3.5.7c)
may have made this replacement easier, although it is unlikely to have caused it.
The Faliscan ending of the third plural perfect also shows an old thematic aorist
ending, -o-(n)d /-o-nd/: in this respect Faliscan differs from both Latin and the
Sabellic languages. In Latin, the ending was the inherited perfect ending */-ēri/ (←
PIE */-eħ1-ri/), first attested in steterai CIL I2.2832a (with an -ai probably influenced
by the endings of the first singular /-a/ and the second singular, which was then either
the inherited */-ta/ or the new formation */-is-ta/), and later as -ere /-ēre/. Beside
this, Latin later developed the endings /-eront/ and /-ēront/, the latter apparently a
contamination of /-ēre/ × /-eront/, but there is no indication that in Latin the aorist
ending was ever used for the third plural perfect. The Sabellic languages had a Proto-
Sabellic innovation /-ns/: the original ending, whether */-nd/ or */-ēri/, is unattested.
In view of this difference, it may be questioned whether Faliscan f[.f]ịqod is
representative of the standard Early Faliscan paradigm. Faliscan, early Latin, and the
Sabellic languages all show the old aorist ending as the standard ending of the third
singular of the perfect: Faliscan f[.f]ịqod therefore represents a very obvious
analogical extension of the aorist endings to the third plural of the perfect. This use of
the aorist ending for the third plural of the perfect may have been incidental: the
Faliscan material does not show whether -o-nd ever was or became the standard
ending, or coexisted alongside another ending (presumably */-ēri/). Conversely, the
Latin and Sabellic material only shows that there the aorist ending for the third plural
perfect was never standardized, not that it never occurred there as well.
5.3. The attested Faliscan verbs.
5.3.1. Verb forms. The following list contains all the verb forms attested in the
Faliscan material.
1. carefo MF 59, carẹ[f]o MF 60, 1st sg. fut. ind. The /-ē-/ continues the PIE stative
suffix */-eħ1-/ rather than */-ee-/ (cf. Meiser 2003:90-1).
2. cupat MF 40, c]up[at] MF 159, cup[a] MF 161, cupat MF 220, cupa MLF 305,
cupa LF 221, cupat LF 224, [also cubat LtF 231], 3rd sg. pr. ind.;
3. cupa]nt MF 80, cupat MF 146, cupạṭ MF 158, cupat LF 223, 3rd pl. pr. ind.;
161
CHAPTER 5
(either singular or plural:) cupa[?t] MF 95, cu[pa] LF 226, [cuba LtF 326].
In all Italic languages where the verb is attested (Latin cubo, South Picene qupat
MC.1, AP.3, Marrucinian cibat MV 6, Paelignian incubat Pg 10) the verb belongs
to the first declension: as it is hardly a causative or a denominative, this may be
due to the largyngeal of the root, */kebħ2-/, affecting the suffix (*/kubħ2-ee-/ →
*/kubħ2-ae-/, cf. Meiser 2003:66 n.112).
4. douiad EF 1 (also in Herbig’s (1913:75) restoration [dou]ịad in EF 1), 3rd sg. pr.
act. subj. Faliscan douiad has rightly been compared to the Latin subjunctive
forms duam (cf. DÉ, LEW s.vv. dō, duam). The forms continue a root /*deħ3-/ or
*/doħ3-/,86 an extension of */deħ3-/ (see WOU s.v. pordouitu, Meiser 2003:
182-3). The Faliscan form probably represents a third-conjugation /dō--ā-d/
based on an old aorist stem (thus Meiser 1987:189), cf. also the Latin subjunctive
duim etc. from the same root; interpreting douiad as a fourth-conjugation form
/dōī-ā-d/ seems more difficult.
5. esú Cap 389, 404, 465 = esú(m) /esom/, 1st sg. pr. ind. The form esum is attested
for several Sabellic languages: South Picene esom TE.4, Praesamnitic esum Ps 4,
5, Hernician esu He 3, and Umbrian esu Um 18. For Latin, it was already known
from Varro (“sum quod nunc dicitur olim dicebatur esum, et in omnibus personis
constabat, quod dicitur esum es est, eram eras erat, ero eris erit” L 9.57), and is
now attested in the inscription ṃorai eṣo[m] from the ager Signinus (Colonna
1994). Esom also appears in the Garigliano inscription (early fifth century), pari
med esom kom meois sokiois trivoiai deom duo[...]nei (see M. Mancini 1997).
Both branches of the Italic languages therefore show an /esom/ alongside /som/,
although the epigraphically attested Latin and Capenate forms are all from areas
where Sabellic influence is a distinct possibility. Since due to the constraints
imposed by the borrowing hierarchy (§1.3.2.2) it is very unlikely that either
branch borrowed the form from the other (cf. and also Joseph & Wallace
1987:680-1), /esom/ is in all probability a Proto-Italic form: it may, however, have
been borrowed as part of a formula: see §9.3.2.
The usual explanation is that /som/, which cannot be derived regularly from
PIE */ħ1ésmi/, was a Proto-Italic innovation, and that /esom/ was an analogical
creation after /som/ (in short, /som es est/ → /esom es est/). Joseph & Wallace
(1987) argue against this, assuming that /esom/ was the original form, and that
/som/ was a later development from /esom/. They propose that the development
was in fact PIE */ħ1ésmi/ → Proto-Italic */ħ1esmi/ → */esmi/ → */es/ →
*/esəm/ → /esom/. This derivation follows regular phonological processes in all
but the final two steps.
86
In earlier publications, Meiser gave the root as */do(Ħ)-/ (1986:186-91): cf. also (on Latin
duim etc.) “Aoriststamm *do < *deh ... das Präs. bietet falisk. douiad ‘det’ ” (1998:184).
162
THE VERB
M. Mancini’s (1997:32-3) objection to */es/ → */esəm/, that Proto-Italic

(word-internal) /Vsm/ was retained (and later developed to /Vm/ in Latin), is
irrelevant: in */es/, the cluster is word-final and, moreover, contains a
morpheme boundary. The development of */es-/ can therefore to be compared
to that of the accusative singular of the consonant-stems, and the problem is rather
why */es-/ should develop to /esom/ while the accusative */-C-/ developed to
/-C-em/. Joseph & Wallace assume that the */-/ in */es-/ was secondary, and
therefore did not develop like the accusative */-/. For the final step, */esəm/ →
/esom/, they assume either a labialization of /ə/ → /o/ due to the following /m/, or
an analogy after the thematic secondary ending /-om/. It is unclear, however, why
in that case the accusative in /-em/ did not also change to /-om/ at a Proto-Italic
date under the influence of the second-declension accusative, especially since
/-om/ later replaced /-em/ in the Sabellic languages.
More problematic, in my view, is the way in which Joseph & Wallace
(1987:687, 689-90) derive /som/ from /esom/ by assuming that /esom/ lost its
initial /e-/ in enclitic position (/-esom/ → /-som/) separately in Latin and in
the Sabellic languages. They are probably right in assuming that such a loss
cannot have been a Proto-Italic process and therefore has to be ascribed to
separate processes in Latin and Sabellic, but their explanation is slightly awkward
in that it ascribes this loss to two processes that operated in dissimilar contexts: in
Latin, the syncopation or reduction of medial syllables (including those in
compounds like necdum ← */neke-dum/), and in Sabellic, the syncopation
processes that mainly affected final syllables. If correct, Joseph & Wallace’s
explanation would place the emergence of /som/ somewhere in the fifth century
for Latin (cf. §3.6.6) and perhaps also for the Sabellic languages, late enough for
/esom/ to survive in the epigraphic record. I wonder, however, whether an
accented /esom/ and an enclitic /som/ might not have existed side by side from a
Proto-Italic date onwards (cf. also Meiser 1998:221-2).
The Latin and Capenate forms all preserve an unrhotacised /-s-/: except in
the case of esom in the Garigliano inscription, which predates rhotacism, this
point, too, needs to be addressed. In the esum quoted by Varro (assuming that he
used a post-rhotacism source), this is probably due to analogy with the /s/ in the
remainder of the paradigm: it is not necesssary to assume that /esom/ was in fact
rhotacised to */erom/ and that this hastened its disappearance, as Joseph &
Wallace (1987:691-2) suggest. The Capenate forms can therefore represent both
Latin and a Sabellic language: if they are Sabellic, they are most likely to be
Sabine. However, the only evidence for the presence or absence of rhotacism in
Sabine seems to be the forms quoted by Varro (L 5.74, Velius Longus CGL
7.69.7-9) and Festus (Paul. Fest. 8.14L), which seem to suggest that rhotacism did
not occur there.
163
CHAPTER 5
6. faced MF 471*, facet MF 470*, 3rd sg. pf. act. ind. The different endings are due
to a replacement of the old aorist ending -d by a new perfect ending, see §5.2.4e.
The publication of these two Faliscan forms faced/facet has re-opened the
discussion on the Italic perfects (and aorists) from the root */dheħ1-/. Starting with
the non-reduplicative (aorist) forms from the root */dheħ1-/ in the Italic languages,
these are as follows. Latin had fec- /fēk-/ at least from the fifth century onwards
(first attested in duenos med feced CIL I2.4 and nouos  plautios  med romai 
feced CIL I2.561). The Sabellic languages, on the other hand, have fak-/fac- /fak-/
in Paleoumbrian face Um 4=480† (c.560), and in Umbrian fut. pf. fakust TI
IV.31, fakurent TI Ib.32, facurent TI VIIa.43: Oscan anafaket Lu 18, may be
another instance (see below). Perhaps a similar from is ]face in an inscription
from the Vestinian area (Mattiocco 1986:92, 95).
The easiest solution to explain this co-occurrence of /fēk-/ and /fak-/ is to
assume that both are derived from an Proto-Italic aorist that still showed
paradigmatic Ablaut, with a singular */dheħ1k-/ → */ɸēk-/ and a plural */dhħ1k-/
→ */ɸak-/: in Faliscan and the Sabellic languages, the plural form of the root was
generalized, in Latin, the singular form (thus Wallace 2004:179, Poccetti 2005:
27-8). Such a paradigmatic Ablaut existed in the roots */deħ1- dħ1-/ ‘to give’ and
*/ħ1es- ħ1s-/ ‘to be’, but it must have been a very rare phenomenon the Proto-
Italic.
A similar co-occurrence of /fak-/ and /fē(k)-/ is apparently found within the
Umbrian present: inf. façiu TI IIa.16, façu TI IIb.22; 3rd sg. subj. façia TI IIa.17
but also feia TI Va.23, Vb.3; fut. imp. fetu TI Ia.3 etc. (47 attestations), feitu TI
VIa.22 (20 attestations), fetu TI VIa.22 etc. (53 attestations), feitu TI VIb.3 etc. (5
attestations). This alternation may reflect two different formations, */fak--/ and
*/fē--/ (cf. WOU s.v. fakiiad), or fei-/fe- reflect a form that developed directly
from */fak--/ (cf. Meiser 1986:124 (/fetu/ ← */fakitōd/) and Berenguer & Luján
2005:198-202).
The alternative explanation is that /fak-/ and /fēk-/ are reduplicative perfects that
had lost their reduplicative syllable (thus e.g. Berenguer & Luján 2005:209-10,
De Simone 2006:162-3, 172). In this explanation, two new problems present
themselves: (a) why these reduplicative perfects should have lost their
reduplication (and thus become identical with the old aorist), and (b) why there
should have been two reduplicative perfects /fefak-/ and /fefēk-/ in the first place,
especially since /fefēk-/ has a full-grade root, whereas the reduplicative perfect
normally has a zero-grade root. Also, the Latin data seem to argue against this, for
in Latin the attested reduplicative perfect is vhevhaked CIL I2.3 /fefak-e-d/, but
the later form is not /fak-/ but fec- /fēk-/. Before attempting to answer these two
questions, however, a brief look at the material on the reduplicative perfects of the
root */dheħ1k-/ is needed.
164
THE VERB
The attestations of /fefak-/ are the following. The two Latin attestations of
this reduplicative perfect are the vhevhaked CIL I2.3 on the fibula Praenestina,
whose authenticity has been a point of debate, although most authors now accept
its authenticity,87 and the ṿḥe[ in the Vendia-inscription 479†, which can be
restored both as ṿḥe[ked and as ṿḥe[vhaked (see §18.3.1). Oscan shows
attestations of */fefak-/ in the perfect fe‹f›acid TB 10 and the future perfect
fefacust TB 11, 17, fef[acust] TB 33. Another instance is perhaps Oscan anafaket
Lu 18, which is either a reduplicative perfect with drop of the reduplicative
syllable after a prefix (cf. LHS I p.587) or a form of the perfect /fak-/.
The evidence for /fefēk-/ consists in effect only of Praesamnitic fefiked Ps
20 (with an early closing of /ē/ → /ī/). The other possible attestations,
Palaeoumbrian hehike read by Firmani (1977) in Um 2, and South Picene fefeh
read by Marinetti (1999; still tefeí 1985:233) in CH.2 are now read differently as
heṛuseḥ or heṛuseṣ and as tefeh respectively. Some authors, e.g. Poccetti
(2005:28), also include Oscan fifikus Cp 37 and Faliscan fifiked EF 9 and f[.f]ịqod
EF 1 among the attestations of /fefēk-/, but I still prefer to regard these forms as
perfects from the root */dheiǵh-/ (cf. below on fifiked/f[.f]ịqod).
In my view, the explanation of these forms must start from the assumption
that at least /fēk-/, /fak-/, and /fefak-/ reflect Proto-Italic forms: (a) /fak-/ and
/fefak-/ are both attested for both branches of the Italic languages (unless
vhevhaked is discarded as evidence); (b) /fēk-/ can only be a aorist full-grade root
(hardly a reduplicative perfect that lost its reduplicative syllable, see below); (c)
/fefak-/ is a reduplicative perfect, which ceased to be a productive category during
the Proto-Italic period (although new reduplicative perfects may still have been
formed later by creative analogy, as may be the case with /fefēk-/).
The only way in which /fēk-/ and /fak-/ could be of a later date is by
assuming that they were originally reduplicative perfects /fefēk-/ and /fefak-/ that
somehow lost their reduplicative syllable. This is a difficult assumption especially
in the case of /fefēk-/, as the Italic reduplicative perfect had a zero-grade root, but
there are also more general objections against this idea. Berenguer & Luján
(2005:209-10) suggest that the reduplicative syllable could be dropped because
it was regarded as more or less equivalent to a prefix. I find this idea
unappealing, for three reasons: (a) unlike prefixes, reduplication only occurred
in well-defined morphological categories; (b) if true, a far more widespread
drop of reduplicative syllables would be expected; (c) this explanation reduces
the reduplicative perfect to a non-category, since it assumes that the category’s
distinctive feature could be more or less freely omitted.
87
Among the recent authors on the perfect of facio, Meiser (2003:178-80), Poccetti (2005:
30-2), and De Simone (2006) all assume the authenticity of vhevhaked.
165
CHAPTER 5
A drop of reduplicative syllables certainly occurred in Latin (and perhaps

also in the Sabellic languages, cf. e.g. Oscan anafaket Lu 18), but only in verbs
with a prefix (cf. LHS I p.587): this was probably because forms in which the
root was preceded by both a prefix and a reduplicative syllable did not conform
to the desired bisyllabic structure (see Meiser 1998:210, 2003:160). De Simone
(2006:162-3) explains the emergence of /fēk-/ and /fak-/ by the attested drop of
the reduplicative syllable in compounds (cf. LHS I p.587) spread to non-
compounded verbs, but the arguments against this idea are mutatis mutandis
much the same as the ones named above.
I would suggest that the various forms can more easily be explained by the
assumption of a Proto-Italic confusion between an aorist and a reduplicative
perfect. If this aorist still had functional vowel gradation (i.e., sg. */dheħ1k-/ vs. pl.
*/dhħ1k-/) in Proto-Italic, this must have been quite exceptional, and the expected
development would be that either the root of the singular or the root of the plural
would be generalized throughout the paradigm. The evidence seems to point to
the existence of both these paradigms, one with */dheħ1k-/ → */ɸēk-/ and one
with */dhħ1k-/ → */ɸak-/. The co-existence of two variant roots in the aorist
beside a reduplicative perfect */dhedhħ1k-/ → */ɸeðak-/ may well have led to the
analogical creation of a new reduplicative perfect */d hedheħ1k-/ → */ɸeðēk-/.
The creation of */ɸeðēk-/ may have taken place after the Proto-Italic
period: this limits the number of forms that has to be assumed for Proto-Italic, and
the material allows it. In that case, however, it must be assumed that the
languages where */ɸeðēk-/ was formed preserved */ɸak-/, */ɸēk-/, and */ɸeðak-/.
The Sabellic languages then either preserved Proto-Italic */ɸak-/, */ɸeðak-/ and
*/ɸeðēk-/ (all three of which are reflected in the epigraphic material) or preserved
Proto-Italic */ɸak-/, */ɸēk-/, and */ɸeðak-/ (like Latin-Faliscan, see below) and
formed a new perfect */ɸeðēk-/ → /feβēk/, with */ɸēk-/ subsequently vanishing,
apparently without trace.88 The reduplicative perfects */ɸeðak-/→ /ɸeβak-/ and
*/ɸeðēk-/ → /feβēk/ then survived in the south, and /fak-/ in the north. These
varying outcomes can be ascribed to different preferences: e.g. in the south, a
desire to preserve a clearly marked perfect (pr. /fak--/ : pf. /feβak-/) and in the
north, a desire to preserve the perspicacity of the paradigm by generalization of
just one form of the root (pr. /fak--/ : pf. /fak-us-).
Latin-Faliscan must have preserved Proto-Italic */ɸak-/, */ɸēk-/, and
*/ɸeðak-/, but the distribution of these forms is suprising, since Latin uniquely
shows both an (early) reduplicative perfect /fefak-/ (← */feβak-/ ← */ɸeðak-/)
and a (later) non-reduplicative perfect /fēk-/. Meiser (2003:160-1, 178-80)
88
Marrucinian (or Paelignian, cf. Jiménez Zamudio 1986:43) fec = fec(ed)? MV 3 may
represent a survival of */fēk-/. Fecront NDI 223=ILLRP 303 is Latin, not Marsian.
166
THE VERB
explains this on the basis of the distributional patterns of the perfect formations:
“Das [reduplizierende Perfekt] findet sich nur bei Verben mit anlautender
Okklusiva, Gruppe aus “s + Okklusiva” und Nasal” (2003:160). In his view, the
discrepancy between the word-initial and the word-internal reflexes of the
voiced aspirates caused an anomaly in the reduplicative perfect of the verbs
with a root starting in an original voiced aspirate (see §3.3.3), and it was this
anomaly which in Latin (and Faliscan?) 89 eventually lead to the standardization
of the non-reduplicative perfect in these verbs.
Taking into account that Proto-Latin must have had */ɸak-/ → */fak-/,
*/ɸēk-/ → */fēk-/, and */ɸeðak-/ → */feβak-/ → */fefak-/, this means that if
Meiser’s explanation is correct, it is in fact Faliscan that shows the more
expected development, dropping /fefak-/ in favour of /fak-/, while in Latin
/fefak-/ was for some reason replaced by /fēk-/. Meiser (2003:179-80) suggests
that this may be due to dialect variation; De Simone (2006:173-4) in fact
ascribes /fēk-/ to Roman Latin.
7. fifiked EF 9, 3rd sg. pf. act. ind.
8. f[.f]ịqod EF 1, 3rd pl. pf. act. ind. (-od = -o(n)d). Herbig’s (1913b:74-80) and
Buonamici’s (1913:40) restoration f[if]iqod (cf. fifiked) is probably correct.
Assuming that k in fifiked and q in f[if]ịqod represent /g/ (cf. §11.2.3), both
forms can represent a reduplicative perfect /fifig-/. At least in fifiked, and probably
also in f[.f]ịqod, the vowel of the reduplicative syllable was either assimilated to
that of the root, or preserved a (Proto-Italic?) reduplication where the vowel of the
reduplicative syllable followed that of the root, especially where this was /i/ or /u/
(cf. Meiser 2003:159-60). Since the PIE root was */dheiǵh-/, the presence of an
intervocalic /g/ must be due analogy with the /g/ in the present stem */fing-/ (as in
Latin fingo) ← */dhingh-/, where it was the regular development of */gh/ (← */ǵh/)
after the nasal infix (§3.2.8), unless it is assumed that in Faliscan intervocalic
*/gh/ developed to /g/ (G. Giacomelli 1963:125, Stuart-Smith 2004:62) which
would explain its occurrence here and in lecet MF 88: see §3.3.3.
The formation of the perfect was radically different from the one found in
Latin, Early Faliscan having a reduplicative perfect with assimilation of the vowel
in the reduplicative syllable, and Latin, an s-perfect finxi that reflects an aorist,
with analogical extension of the nasal infix. In view of what was said under
faced/facet (above) on the distribution of the reduplicative perfects with roots in
/f-/, it is possible that the Faliscan reduplicative perfect /fifig-/ later disappeared
and/or that Latin originally also had a reduplicative perfect which was later
dropped in favour of finxi.
89
Middle Faliscan faced MF 471* and facet MF 470*, published in 2003-2004, were of
course unknown to Meiser when he formulated this theory.
167
CHAPTER 5
The idea that fifiked and f[.f]ịqod (as well as Oscan fifikus Cp 37, and perhaps
Samnitic fif[ (= fíf[íked]?) in Sa 4) are in fact forms of facio was rejected already
by Lejeune (1955), but has been revived in the recent discussion on faced/facet
(see above), e.g. by Poccetti (2005:32-4). To me, this idea remains unacceptable.
The vowel of the root is spelled as i both in fifiked and in f[.f]ịqod (where
readings like f[.f]ẹqod or f[.f]ạqod can certainly be excluded), and this i must
represent //, as in this interpretation the i in the reduplicative syllable of fifiked
(and presumably in that of f[.f]ịqod) can be explained only by assimilation to the
vowel of the root. However, a Faliscan /-fk-/ cannot be derived by any regular
process from either */-dhħ1k-/ or */-dheħ1k-/, and fifiked and f[.f]ịqod are far to
early to ascribe it to a reduction of /a/ (← */ħ1/), which cannot reasonably be
placed before the fifth century (§3.6.6), let alone of /ē/ (← */eħ1/).90
9. ? i*ice LF 309, ị*ice LF 315 (probably to be read either as ip ̣ice or as iḍice). The
word is usually explained as a verb because of the structure of the inscription tito 
uel|mineo  iun|aị ị*ice LF 315, where editors have interpreted iunai as a dative,
being reluctant to assume a genitive in -ai for Faliscan (§4.2.2).
Herbig (1923:231-2) suggested that ip ̣ice = impigit ← */en-pepag-/ (root
*/peħ2ǵ-/), a reduplicative perfect of an *impingo ‘to fasten upon’, with a
reduction of the medial syllable as was then assumed for porded EF 1. Although
the possibility of such a reduction is now rightly doubted in the case of porded, by
the time of Late Faliscan i*ice it may not have been impossible (cf. §3.6.6); in
any case, the reduction might be due to the omission, found mainly in Latin, of
reduplicative syllables when the verb carries a prefix (cf. below under porded).
Since reduction of word-internal syllables was assumed to be entirely absent in
Faliscan (cf. §3.6.6), this explanation was rejected by Stolte (1926:61), who
revived the alternative rejected by Herbig, namely ip ̣ice = impīgit ← */en-pīg-/,
perfect of an *impingo ‘to paint upon’. Of such a perfect */pīg-/ there appears to
be no trace, however, nor does its existence seem likely (cf. Meiser 2003:152-8
on the distribution of the various perfect formations), except as an analogical
creation. In my view, there can be reasonable doubt whether i*ice is a verb at all:
I would rather regard it as a cognomen (see §7.9.1.3).91
90
In Latin the assimilation of the vowel of the reduplicative syllable to that of the root did not
take place if the root vowel was a secondary /i/ or /u/ due to weakening of /a/, /e/, or /o/ (LHS
I p.586).
91
Because of the constraints of the borrowing hierarchy (§1.3.2.2) and the rarity of Etruscan
lexical borrowings in Faliscan (§6.3, §9.2.2), I think it unlikely that i*ice is an Etruscan verb in
-ce, as was suggested by Ribezzo (1931b:192) from a Faliscan *ipi = Latin ibi, which is
impossible as the Faliscan form would have been *ifi) and Pisani (1964:341, from Etruscan ipi
‘olla sepolcrale’, derived in the same way as turuce Cr 3.17 etc. was derived from dîron).
168
THE VERB
10. keset LF 242, kese[t LF 243, 3rd sing. pf. ind. The verb shows an s-perfect as in
Latin. Both Faliscan attestations occur in the formulaic phrase magistratum gessit,
which is identical with the phrase used in Latin (TLL 62.1939,1-1940,56, cf.
§6.2.30, §9.4.2) and may be either a copy or a calque (cf. Renzettti Marra
1990:339-40).
11. lecet MF 88, 3rd sing. pr. ind. As the PIE root is */leǵh-/, the c is difficult: either
the Faliscan development was (*/ǵh-/ →) */gh/ → */γ/ → /g/ (G. Giacomelli
1963:125, Stuart-Smith 2004:62), or c (still) represents /γ/ (Meiser 2003:68
n.118): cf. §3.3.3. Unlike the case of fifiked EF 9, f[.f]ịqod EF 1, it is not possible
to ascribe the /g/ to an analogy. Since e can represent both /ē/ and /ẹ/, -et may
represent either third-conjugation [-ẹ-t] = /-e-t/ or /-i-t/ (as appears to be implied
by its IE cognates, cf. DÉ, LEW, EDL s.v. lectus) or second-conjugation /-ē-t/
(thus Vetter 1953:301). The latter possibility may be the more likely one: Meiser
(2003:68 n.118) points to the productivity of the stative suffix */-eħ1-/ in verbs
like Sabellic */stāē-/ underlying forms like e.g. Oscan stahínt Cp 25, staíet Cm
1B.31, Cp 24, Samnitic staít Sa 1B.22, and Latin sedeo beside sido.
12. ? pepara[i EF 1, 1st sg. pf. act. ind. The use of punctation to separate the
reduplicative syllable from the root is parallelled only by the vhevhaked of the
fibula Praenestina CIL I2.3.92 In spite of this curious division, pe  para[ has been
read and interpreted as pepara[i = Latin peperi since Herbig (1913b:84-7), a
reading that in my view is still not entirely certain. Other proposals have been
made (e.g., Peruzzi’s (1964:160-4) pe  par a[dke]douiad ‘per par accedat’ and pe
 para[te ke]douiad ‘’per parate accedat’), but none of these have ever affected
the accepted reading.
If pepara[i is read, this is a reduplicative perfect /pepar-/ ← */peprħ3-/, the
expected early form of the Latin perfect peperi. The ending -a[i /-a/ represents an
innovation by the addition of /-i/ to the Proto-Italic ending */-ħ2a/ ← PIE */-ħ2e/,
although it cannot be said whether this addition was Proto-Italic or Latin-Faliscan
only (see §5.2.3e). The form is the only instance of the ending of the first singular
perfect that still shows the stage /-a/.
13. pipafo MF 59, pafo=‹pi›pafo MF 60, 1st sg. fut. act. ind. Editors have questioned
whether pipafo and pafo represent two different formations or if one is simply an
error for the other. Since the two kylikes on which these forms occur were
obviously meant as a pair (or even as part of a series), I assume that pafo is a
graphical error for ‹pi›pafo.
92
Campanile (1986), in the discussion on the authenticity of the fibula Praenestina, has
shown convincingly that pepara[i cannot have been the alleged faker’s model for vhevhaked,
as the Ceres-inscription was still unknown when the fibula Praenestina first appeared.
169
CHAPTER 5
Whether Faliscan pip- represents /pib-/ (← PIE */pi-pħ3-/ ← */pi-bħ3-/), or

the later /bib-/ found in Latin (cf. §5.2.2.13, §6.2.3) cannot be established. The -a-
represents either /ā/ or /a/, according to how the form is analysed:
(a) pipafo represents /pipa-f-ō/, /piba-f-ō/ or /biba-f-ō/ from an original athematic
*/pi-pħ3-/ or */pi-bħ3-/, the /-a-/ continuing the laryngeal: Latin bibo would then
show a regular reduction of the medial vowel (Schrijver 1991:413). If pafo is a
different formation, /ba-f-ō/ ← */bħ3-/ would then have a direct parallel in Latin
dabo /da-b-ō/ ← */dħ3-/ (thus, e.g., LEW s.v. bibō, Schrijver 1991:412-3).
(b) pipafo represents /pipā-f-ō/, /pibā-f-ō/ or /bibā-f-ō/, assuming that the verb in
Faliscan belonged in the first conjugation (thus G. Giacomelli 1963:152, 238-9,
pointing to pairs like cumbere : cubare). In this case, it remains unexplained why
the verb belonged in the first conjugation, however.
(c) pipafo represents /pip-ā-fō/, /pib-ā-fō/, or /bib-ā-fō/, assuming that an
(unattested) ā/ē-future *pipam /pip-ā-m/, /pib-ā-m/, or /bib-ā-m/ etc. was
reformed to an f-future with retention of the /ā/, after other f-futures (cf. carefo),
especially the f-future of the first conjugation, which would have had /-ā-f-ō/.
14. porded EF 1, 3rd sg. pf. act. ind. The form is often regarded as an originally
reduplicative perfect */por-ded-e-d/ with loss of the medial syllable: a
reduplicative perfect from the root */deħ3-/ is found both in Latin (dedi) and in the
Sabellic languages (Umbrian teřust TI Ib.34, dirsust TI VIIa.43, and Oscan
dedens Cm 2, 4, 9, ded[ens Sa 24, dedenj Lu 2). As the reduction or loss of a
medial syllable is difficult to maintain at this date, Meiser (2003:106 n.31) and
Wallace (2004:179 n.24) argue that the form is rather a continuation of an old
aorist. Alternatively, the form could in my view still be explained as a
reduplicative perfect, with the loss of the reduplicative syllable due to the
tendency, observable at least in Latin, to drop the reduplicative syllable if the verb
carries a prefix (cf. above under faced/facet).
15. ? prameḍ, pramed EF 2, possibly 3rd sg. pr. act. subj. /prām-ē-d/. Although the
word is often interpreted as an adverb derived from a */prĦ-mo-/ that also
underlies Latin prandium (see §6.2.59), Pisani (first 1937:233) observed that
pramed could well be a subjunctive, deriving it from a *prāmere or a *prāmāre
that would correspond to Latin promo. The form would be the only instance of an
ē-subjunctive in Faliscan. (Note that interpreting pramed in this way appears to be
the only feasible possibility of finding a verb in this inscription.)
16. saluete EF 4, 2nd pl. act. imp.
17. salue[to]d EF 3, salueto EF 4, 3rd sg. fut. imp. Salueto shows an omission of -d
that may represent a very early weakening of /d#/ (§3.5.7c), or may be a simple
graphical error (salueto‹d›).
170
THE VERB
The form salueto is apparently used as a plural: ofetios kaios uelos amanos
[all probably men’s names, cf. §7.2.1] salueto salues seite EF 4. The possibility
that the verb takes the number of the last (singular) name is excluded by the
immediately following plural salues seite (=s{e}ite or s‹ie›te). This may reflect an
inherited paradigm where the future imperative still had only one (singular)
ending */-tōd/. The attested Italic future imperative plural endings were all post-
Proto-Italic formations: see §5.2.4d. Prosdocimi (1990:304-5) explains this
plural use of salueto (with -to, not -tod) as due to a different paradigm, comparing
Umbrian, where -to is used indiscriminately for the singular and the plural, in
spite of the existence of a plural forms -tuta, -tuto but the phonological context in
the much later Umbrian texts is quite different, and his comparison of the future
imperative endings -tod and -to with the co-occurrence of an ablative in -od and
an instrumental in -o (Prosdocimi 1990:320-1) goes rather far.
Interestingly from a syntactic (rather than a morphological) perspective is
the side-by-side occurrence, in EF 4, of the imperative, the future mperative, and
the subjunctive, as ues saluete sociai ofetios kaios uelos amanos salueto salues
seite (=s{e}ite or s‹ie›te).
18. seite (=s{e}ite or s‹ie›te?) EF 4, 2nd pl. pr. subj. Since in Early Faliscan ei cannot
represent /ī/, the form must be an error or an alternative spelling for either site
(s{e}ite) or siete (s‹ie›te). To confuse matters further, the word could also be read
as seitei.
If seite is to be read as s{e}ite /s-ī-te/, it is the direct continuation of the old
optative, PIE */ħ1s-iħ1-té/, used as subjunctive; if it is to be read as s‹ie›te (/s-ī-ē-
te/ or /s-iē-te/?), it is apparently influenced by the singular forms */s-iē-m s-iē-s
s-iē-d/ ← PIE */ħ1s-iéħ1-m ħ1s-iéħ1-s ħ1s-iéħ1-t/ (cf. Latin siem sies sied) or by the
/-ē-/ of the ē-subjunctive. The ending is still the original secondary ending: the
corresponding Latin form sitis is attested only from the time after the
generalization of the primary endings to the subjunctive. Comparable forms are
found in Umbrian: 2nd sg. sir TI VIb.7, 26, sei TI VIa.23, si TI VIb.26, 3rd sg. si
TI Va.6, 24, 27, Vb.3, 7, 3rd pl. sis TI Va.6, sins TI VIIb.4.
19. ? teneṭ in [---] cuiteneṭ[---|---] let MLF 361 = /tenē-t/, 3rd sg. pr. act. ind. I regard
the attestation as uncertain, and even if the word teneṭ is regarded as attested, the
(syntactic) context as well as the specific meaning of the word here is unclear (cf.
Renzetti Marra 1990:336-7).
20. tulas Etr/EF 385 = /tul-ā-s/, 2nd sg. pr. act. subj.
21. tulate Etr/EF 385 = /tul-ā-te/ 2nd pl. pr. act. subj.
Both forms appear side by side in the Etruscan or Early Faliscan inscription
tulate tulas urate Etr/EF 385. Pisani (1943:262-3, 1964:347) quite plausibly
connected tulate and tulas with Latin tollo (“sopportate - sopporta!”) as a
171
CHAPTER 5
‘congiuntivo radicale’, i.e., a zero-grade root */tħ2-/ + subjunctive suffix /-ā-/,

corresponding to Latin forms like attolat Pac. 42R, 228R etc. In tulate, the ending
used for this subjunctive is the secondary ending, as might be expected in an early
subjunctive (see §5.2.4b).
22. ?? tulom MF 68, 1st sg. pf. act. ind.? Herbig (1914:238 n.1) tentatively interpreted
this form as an (originally reduplicative) perfect */tetul-/ with the old thematic
aorist ending */-o-m/, which clashes with Herbig’s own interpretation of Early
Faliscan pe  para[ EF 1 as a first singular perfect pepara[i with the ending -a[i
(cf. §5.2.4e), although admittedly it would agree with the aorist endings in the
third singular porded EF 1, fifiked EF 9 and the third plural f[.f]ịqo(n)d EF 1. For
a Middle Faliscan inscription, however, assuming an aorist ending in the first
singular perfect is extremely doubtful, unless it is assumed that the Faliscan
paradigm was radically different from that of contemporary Latin: the only
parallel would be Oscan man]afum Cp 37,1, manafum Cp 37,3, itself unique in
being the only attestation of a first singular perfect from the Sabellic languages.
For tulom, an interpretation from the onomasticon (‘Tullorum’) is more probable.
23. urate EF/Etr 385 /ōrā-te/, 2nd pl. pr. act. imp.? (The alphabet of the inscription is
Etruscan rather than Faliscan, and u- may therefore represent /-/ as well as /-/.)
Urate occurs in the Etruscan or Early Faliscan inscription tulate tulas urate
Etr/EF 385. These three forms were all explained as ‘congiuntivi radicali’ by
Pisani (1943:262-3, 1964:347, translating urate as “chiacchierate!”), and although
this is plausible in the case of tulate and tulas (see above), it is difficult in the case
of urate. Pisani connected this form with Oscan urust TB 14, 16, which is derived
from */ħ2er-/, probably through a reduplicated */ħ2e-ħ2or-/ (cf. WOU s.v. urust).
The ‘congiuntivo radicale’, however, requires a zero-grade root (cf. Meiser
2003:41-2), and this */ħ2r-/ would be expected to develop into */ar-/ rather than
into */or-/ or */ur-/ (cf. Schrijver 1991:56-73, and §3.2.3). Pisani’s explanation
can therefore only be maintained if the vocalism of urate is explained, e.g., by an
analogical extension of the vocalism of the o-grade. Furthermore, explaining
urate as a subjunctive */r-ā-te/ requires that the Faliscan (and Oscan) verb was
(thematic) */r-/, not /ōrā-/ as in Latin.93 It is in my view easier to assume that
urate is an imperative: co-occurrence of the imperative, the future imperative, and
the subjunctive is also found in EF 4, saluete ... salueto salues seitei (=s{e}ite or
s‹ie›te?): see §8.3.
24. zot MLF 285 [and sot LtF 172?], 3rd pl. pr. ind. (For z- = /s/, see §3.5.3). The form
corresponds to Latin sunt, older sont CIL I2.1529, which is assumed to be a Latin
innovation on Proto-Italic */sent/ ← PIE */ħ1s-énti/, with the vocalism influenced
93
This in turn implies that Latin oro was not a Proto-Italic formation with /-ae-/ but a Latin
denominative verb derived from os, not connected with Oscan and Faliscan */r-/.
172
THE VERB
by the thematic endings. In the Sabellic languages on the other hand, the original
vocalism was preserved (Oscan sent Po 32, Si 4, 5, 6, set Cm 1A.16, Po 32, Cp
24, 29, 30, Samnitic Sa 1A.1; Umbrian sent TI VIa 15, 27, 36, 46, VIb 29).94
25. *[3-4]*ad EF 1, either 3rd sg. pr. act. subj. in /-ā-d/ or 3rd pl. pr. act. subj. in -a(n)d
/-ā-nd/, the latter possibility suggested by Peruzzi (1964a:157). The only
restoration that fits both the size of the lacuna and the traces surrounding it
appears to be p ̣[ore]ḳad ‘porrigat’ (Peruzzi 1964:157). Other proposals are
ạ[dtul]ạd (Ribezzo 1927:151-2), f ̣[in]ḳad (Olzscha 1965:123), [fe]ṛad (Vetter
1953:280), and [pa]ṛad (Joseph & Klein 1981:294), and ạ[dkap]ịad (Herbig CIE
8079), ạ[ddou]ịad (Ribezzo 1936:46), and [dou]ịad (Herbig 1913:75).
5.3.2. Nominal forms. Two Early Faliscan forms have with some likelihood been
interpreted as past participles formed with the common Italic participial suffix /-to-/:
26. ? fita or fitai EF 1. The letters fitaidupes have often been divided either as fita
idupes or fitai dupes, with fita or fitai explained in one of the following ways:
(a) connected with Latin fingo and its derivations in fict- (e.g. Thulin 1908:259,
Herbig CIE 8079, G. Giacomelli 1963:41-3, Peruzzi 1964a:161-2): This
presupposes a very early omission of -c- (fita = [fiˀta]?), or even a development
*/kt/ → /xt/ (fita = [fixta]?): see §3.5.7c. Fita = fi(c)ta would have a parallel in
Middle or Late Faliscan lete MLF 285, corresponding to Latin lecti.
(b) connected to Latin fitum est Andr. 29L (Stolte 1929:107, Vetter 1953:280),
Umbrian fito TI VIb.11 (Herbig CIE 8079): This is not impossible, although the
parallels are not equivalent: the /ī/ in Latin fītum can only be due to influence
from the present fī-o (← */bh--/), whereas the /ī/ in Umbrian fito can also be due
to the Umbrian development */ū/ → /ī/. A Faliscan fita could therefore reflect
both a common (Proto-Italic) innovation and a Latin innovation.95
(c) connected to Latin fetus (Pisani 1946:53): As there appears to be no way in
which Faliscan i can be derived from /ē/, this would appear to be impossible.
27. *e[..]tom EF 1. This has been restored in various ways: ḍe[lec]ṭom ‘delectum’
(Olzscha in Radke 1965a:136-7, Radke 1965a:134), ṃe[re]ṭom ‘meritum’ (Vetter
1953:280), me[lc]tom or me[lq]tom ‘mulctum’ (Vetter in Knobloch 1958:138),
me[le]tom or me[la]tom ‘molitum’ (Joseph & Klein 1981:294).
94
M. Mancini (1997:36-8) assumes that the Latin vocalism was original and the Sabellic
vocalism an innovation. Note that Rix (1993:338) read an isolated Samnitic instance of the o-
vocalism súṇṭ in Sa 17, although he did not maintain this reading in ST, which has súḷ [.
95
Martzloff (2006:67-75) in fact divides fitaidupes as fit aidupes, but as I find his inter-
pretation of aidupes is not convincing, I have not included his form fit here.
173
CHAPTER 5
5.4. Summary of §§5.2-3
The discussion of the attested Faliscan forms in §5.3 illustrates the major difficulty in
the evaluation of these forms: not only is the Faliscan material itself lacunary, but so is
the contemporary material in the other Italic languages, especially as many of the
Faliscan verb forms are from comparatively early inscriptions. Insofar as a consistent
picture of the Faliscan verb can be drawn on the basis of this material, it is quite clear
that Faliscan participated in all the changes of the Proto-Italic period, such as the
formation of the four-conjugational system, the merger of the aorist into the perfect, and
the development of the endings (most notably the development of primary */-ti/ and
secondary */-t/ to primary /-t/ and secondary /-d/). Interestingly, in the morphology of
the verb (as in the morphology of the noun, see §4.10), there appear to be no instances
where Faliscan sides with Sabellic as opposed to Latin (but see below on the ending of
f[.f]ịqod EF 1). Where morphological differences can be established between Latin and
the Sabellic languages, Faliscan sides with Latin in the following cases:
(1) the f-future (§5.2.2.2): The Faliscan suffix /-f-/ in Middle Faliscan pipafo MF 59,
‹pi›pafo MF 60 carefo MF 59, carẹ[f]o MF 60 corresponds to the Latin suffix /-b-/ in
the formation of future of the first and the second conjugation (common innovation
based on the suffix of the imperfect), while the Sabellic languages had an s-future
throughout. It is quite possible that Faliscan, like Latin, had an ā/ē-future in the third
and fourth conjugation.
(2) the s-perfect (§5.2.2.3c, §5.3.1.10): The Late Faliscan s-perfect in keset LF 242,
kese[t LF 243 occurs also in Latin (common retention of the suffix of the old sigmatic
aorist), but not in the Sabellic languages.
(3) the perfect endings (§5.2.4e): The Early Faliscan perfect pepara[i EF 1 and the
Middle and Late Faliscan perfect ending -et (probably representing /-ē ̣t/ ← */-e-t/) both
point to perfect endings /-a/ in the first singular and /-e/ in the third singular. These
forms represent innovations of the inherited endings Proto-Italic */-a/ ← */-ħa/ ← PIE
*/-ħ2e/ and Proto-Italic/PIE /-e/ by the addition of /-i/. This addition may have been
Proto-Italic (*/-a/ → /a-/ or even */-ħa/ →*/-ħa-/) or Latin-Faliscan (*/-a/ → /a-/). In
the Sabellic languages, there appear to be no attestations of these endings (cf. note 83
on face Um 4=480†). In Early and Middle Faliscan, the ending of the third singular
perfect was the old thematic aorist ending -e-d /-e-d/. During the Middle Faliscan
period, this was replaced by a new ending -et, which probably represents /-ēt/̣ ← */-e-
t/, an innovation based on an older perfect ending */-e/. The same shift occurs in Latin,
apparently at the same time as in Faliscan, that is, between the fifth and the third
centuries. There are indications of a shift of /-e-d/ to /-e-t/ in southern Oscan, but this
shift is different in nature and unrelated to the one in Latin and Faliscan.
174
THE VERB
(4) the vocalism of third plural present of ‘to be’ (§5.3.1.24): The vocalism of Middle
Faliscan zot MLF 285 (and perhaps also sot LtF 172) corresponds to that of Latin sont
CIL I2.1529, later sunt, representing a common innovation after the /-o-nt/ of the
thematic verbs, whereas Sabellic preserved the old e-vocalism.
There are, however, several points on which Faliscan (apparently) differed from Latin.
In some cases, these differences are apparent only and due to the limitations of the
available material (the comparison of Early Faliscan to later Latin forms as described
above). Where differences can be established, these are of a later date.
(5) the perfect of ‘to make’ (§5.3.1.6): The Middle Faliscan perfect forms faced MF
471*, facet MF 470* differ from the attested Latin perfects vhevhaked CIL I2.3, later
feci. Even this difference, however, to some degree shows a common factor in Faliscan
and Latin, for both agree in substituting an aorist for an older reduplicative perfect (a
common innovation, shared by Umbrian and perhaps Vestinian, but not by Oscan); if in
Faliscan and in Latin this replacement can be ascribed to the fact that reduplicative
perfects from roots in a voiced aspirate had become an anomaly, Faliscan and Latin also
agree in the application of this morphophonological rule, which did not apply in the
Sabellic languages (Meiser 2003:173-4). On the other hand, Faliscan and Latin differ in
that Faliscan then generalized the root /fak-/ (originally the root of the aorist plural)
while Latin generalized the root /fēk-/ (originally the root of the aorist singular), with
Faliscan perhaps showing the more expected replacement of /fefak-/ by /fak-/: Latin is
in fact the only Italic language to generalize /fēk-/. This difference may be due to dialect
differences within Latin-Faliscan: Faliscan may not have been the only Latin dialect to
have had /fak-/.
(6) the perfect of ‘to knead’ (§5.3.1.7-8): The Early Faliscan reduplicative perfect
fifiked EF 9, f[.f]ịqod EF 1 corresponds to Oscan fifikus Cp 37 rather than to the later
Latin finxi, which continued an old sigmatic aorist. In view of what was said on
faced/facet and the anomaly in the reduplicative perfect of verbs with a root in an
original voiced aspirate (§5.3.1.6), Latin finxi may have replaced an earlier Latin
reduplicative perfect, and in Faliscan the reduplicative perfect may have disappeared in
favour of a non-reduplicative type. (Cf. Meiser 2003:171-2 on such replacements.)
(7) the perfect of ‘to give’ (§5.3.1.14): Early Faliscan porded EF 1 can be explained
either as a reduplicative perfect that lost its reduplication (not due to syncopation, but
due to loss of the (Latin?) reduplication in compounds, see §5.3.1.6) or as an old aorist
form (Meiser 2003:106 n.31). Such an aorist may also have existed in Latin: Meiser
(2003:105-6) suggests that the non-reduplicative Latin present of dare may be due to
influence from the old aorist.
(8) the root of ‘to drink’ (§5.3.1.13): The Middle Faliscan forms pipafo MF 59,
‹pi›pafo MF 60 can be explained in several ways: (a) if the -a- represents /-a-/ ←
175
CHAPTER 5
*/-ħ3-/, Faliscan differs from Latin in that the medial vowel in athematic laryngeal verbs
was apparently not reduced, as it was in Latin (separate retention of a phonological
feature); (b) if the -a- represents an /-ā-/ belonging to the stem, Middle Faliscan pipa-
somehow belongs to the first conjugation (separate innovation of a morphological
feature); (c) if the -a- represents an /-ā-/ belonging to the suffix, Faliscan differs from
Latin in that the future of these verbs was apparently formed in a different way, perhaps
as a recent analogical formation (separate innovation of a morphological feature). In
each of these explanations, Faliscan differs from Latin: in the first and last explanation,
this difference would appear to be of fairly recent date.
(9) the ending of the third plural perfect: One of the more vexing differences
between Faliscan and the other Italic languages is the ending of f[.f]ịqod EF 1. This
form shows an old thematic aorist ending -o(n)d /-o-nd/ that is incompatible with both
the Latin ending */-ē-ri/ → /-ēre/ (inherited from Proto-Italic/PIE */-eħ1-ri/) and the
Sabellic ending /-ns/. The Sabellic ending is a (Sabellic) innovation on an older */-nd/,
and Faliscan would therefore in this respect stand closer to the Sabellic languages
than to Latin. However, in view of the fact that Faliscan, Latin and the Sabellic
languages all used the corresponding third singular aorist ending -ed /-e-d/ in the
perfect, Faliscan -o(n)d may represent an analogical extension of the aorist endings to
the third plural of the perfect. In that case, -o(n)d coexisted with an inherited perfect
ending (probably */-ēri/) and may never even have been the standard ending. (Cf. also
Meiser 2003:89 n.41 on the merging of the aorist endings with the perfect.)
(10) /esom/ ‘I am’ (§5.3.1.5): Capenate esú Cap 389, 404, 464 has parallels both in
Latin and in the Sabellic languages (common retention rather than common
innovation): the lack of rhotacism can be explained as due to Sabellic influence, but can
also be due to an analogical preservation of the intervocalic /s/ within Capenate (or
within Capenate and other Latin dialects).
In several of these cases (certainly 10, probably also 5-7, perhaps also 9) therefore, the
differences between Faliscan and Latin can be explained as apparent rather than real,
and as partly due to the lacunary state of the extant material: such differences as there
are can be ascribed to recent developments (usually dating from between the Early
Faliscan period and the Latin inscriptions of the third century and later). The real
differences are in the formation of pip- or pipa- (8) and in the ending -ond (9). The
former may never be decisively explained, as it depends on the phonological
interpretations of the written form. The latter may be regarded as an apparent difference
only, due to a lack of contemporary material for comparison, and although this is not a
satisfying explanation, it may at least be tested as more material (both Faliscan and
Latin) becomes available in the future. What can be said with certainty is that there are
no cases of differences between Faliscan and Latin where the solution has to be sought
outside the framework of Latin-Faliscan.
176
Chapter 6
The lexicon
In this chapter, the lexical elements attested in the Faliscan inscriptions are compared with the
corresponding elements in Latin, the Sabellic languages, and, where necessary, Etruscan. The
chapter opens with some remarks on methodological issues (§6.1). Following this, the lexical
elements attested in the Early, Middle, and Late Faliscan inscriptions are discussed (§6.2) and
compared with the lexica of Latin, the Sabellic languages, and Etruscan (§6.3). Appended to
this discussion are separate lists of the theonyms (§6.4) and of the geographical names and
ethnonyms (§6.5) that occur in the inscriptions from the area, and a brief discussion of the Fal-
iscan glosses (§6.6).
6.1. The lexicon: methodological issues
For the purposes of this study, the most important aim in looking at the lexicon is to
establish a list of more or less securely attested Faliscan words and other lexical ele-
ments and compare this ‘Faliscan lexicon’ with the equivalents, cognates, and corre-
sponding words in Latin, the Sabellic languages, and Etruscan. This is necessary for two
reasons: first, because in this way the similarities and differences between, on the one
hand, Faliscan and Latin, and, on the other hand, Faliscan and the Sabellic languages
can be established; second, because such a comparison of the lexicon is necessary if the
issue of lexical borrowing is to be assessed in any systematic way. A largely synchronic
comparison of the Faliscan lexicon is therefore in my view more important and more
revealing about the status of Faliscan than is a diachronic derivation of the individual
lexemes from their Indo-European roots, although etymological arguments must still
play a part where it is necessary to explain connections with Latin or Sabellic cognates.
Since Faliscan is not related to Etruscan in the sense that it is related to Latin or the Sa-
bellic languages, comparison of the Faliscan lexicon with corresponding Etruscan
words will serve mainly to establish their possible Etruscan origin.
There is of course always a risk of regarding only those inscriptions as Faliscan
that fit one’s pre-conceived mental image of what Faliscan is, and exclude those that do
not fit this image as being Etruscan, Sabellic, or Latin, which of course leads to a dan-
gerous circular argument. This risk is, I think, especially great in establishing the lexi-
con, for the (consciously or subconsciously) conceived mental image of a fragmentarily
preserved language such as Faliscan is of course likely to be based primarily on the few
items of speech that can be ‘understood’ because they carry a meaning of their own, in
other words, by what is known, or perceived as known, of its lexicon.
177
CHAPTER 6
I have therefore used only those lexical elements that occur in Early, Middle and
Late Faliscan inscriptions, that is, the inscriptions that show morphological, phonologi-
cal, or onomastic features that are consistent with Faliscan: in other words, the lexicon
is based on the inscriptions that can be considered Faliscan on the basis of other criteria
than the lexical elements they contain.
The lexical elements in the Etruscan inscriptions (Etr I-LI) have been excluded:
the inscriptions that are Etruscan in their morphology, phonology, onomasticon, and
alphabet also show a markedly different lexicon, and can safely be said to represent a
different language. Possible interferential forms and borrowings from Etruscan and Sa-
bellic in the Faliscan lexicon are, of course, remarked upon in the list in §6.2, and dis-
cussed in §9.2-3.
As it is assumed here that Faliscan is very closely related to Latin, the question
arises whether and how Faliscan and Latin lexical elements can in effect be distin-
guished. I have used the material from the Latino-Faliscan inscriptions (since these
show at least some features of having been written with Faliscan rather than Latin in
mind. With the exception of esú Cap 389, 404, 465, which is clearly a dialect form (see
§5.3.1.5), I have added the data from these inscriptions between [ ]. I have, however,
excluded all Latin inscriptions, that is, the inscriptions in the Latin alphabet that show
few or no Faliscan features. These are mostly late, and reflect a more general ‘rural
Latin’: 217 (c.125), 218 (c.125), 219 (c.120-50), 237 and 238, 240, 250 (106), 251 (late
2nd century), 268 (4th century, but probably imported), 291 (2nd century), 296 (an im-
port), 377, 386, 393, 420 (c.150), 431-438 (all from before 211), and 456 (c.100-50).
With some hesitation, I have treated Late Faliscan or Latin 214 (c.150?) as an interme-
diate case: although the language of this inscription does not differ from contemporary
Latin, it was obviously meant to give the impression of being a Faliscan inscription. I
have therefore included f., pretod, and sacru, which also occur in Middle and Late Fal-
iscan inscriptions, and excluded words attested from this inscription alone.
I have been reluctant to include lexical elements that are attested only in the ono-
masticon, as it cannot be established whether these elements were also (still) in use as
part of the lexicon. For instance, while the gentilicium Firmius may well be derived
from the adjective firmus, this provides no information on whether or not the adjective
was (still) used in Faliscan: names, being primarily referential elements of speech, may
move between (language) communities with far greater ease than lexical elements, even
if these names also have a lexical meaning. I have therefore included only those cases
where it is probable that the onomasticon reflects words in current use, namely (a) nu-
meric praenomina; (b) cognomina, since during the Middle Faliscan period these were
still a new feature of the onomasticon (§7.9) and may therefore be expected to consist
mainly of lexical elements, and (c) the gentilicia Clipearius and Frenaeus/Frenarius,
which are probably new formations based on the nouns clipeus and frenum respectively.
178
THE LEXICON
6.2. The epigraphically attested lexicon
The following list contains all the lexical elements from the Early, Middle, and Late
Faliscan inscriptions, with additions from the Latino-Faliscan or Latin inscriptions (cf.
§6.1). In the cases of the obscure passages of the Early Faliscan inscriptions EF 1-4, I
have on the whole not included words or interpretations that have been proposed only
once or only by one author, but not adopted by other authors. For ease of reference only,
the words have been placed under the nearest Latin equivalent (to ‘iron out’ the differ-
ences in spelling), according to the alphabetical order of the modern alphabet.
Symbols preceding the lemmata: † = rejected by me, but discussed here because of fre-
quent mention in the literature on Faliscan; ? = attestation doubtful; ° = attested only
in the onomasticon; * = reconstructed.
1. aedilis n. ‘edile’. Nom. pl. efiles MF 113, efiles MF 115, efil]es MF 117, efile MF
114, efi[les] MF 116 (dedications) [cf. also ef in LtF 205?]. The word was first in-
terpreted as ‘aediles’ by Erman (1917). – Direct Latin equivalent aedilis: the Sa-
bellic equivalents, Vestinian aidiles MV 2, Oscan aídíl Po 11, Po 16, aídilis Po 1,
Po 15, aídili[s Po 2, and Samnitic aídịḷí ̣[s Sa 14, are borrowed from Latin (see
WOU s.v. aídíl, Camporeale 1956:44-50 and La Regina 1968:436-46). Etruscan
origin of the word is rightly rejected by Bréyer (1993:137).
Faliscan efiles has been regarded both as an adaptation of Latin aedilis and as a
calque on a Middle Faliscan *efis96 after Latin aedis : aedilis (G. Giacomelli
1963:243-4, Rix 1994:96 n.36; cf. Campanile 1961:7). I doubt, however, whether
Latin was already important enough in the ager Faliscus before the war of 241-
240 to be used for the name of a Faliscan magistracy, and if it were, there seems
to be no reason why Latin aedilis should be adapted and not just borrowed as it
was, as happened in Vestinian and Oscan. Reasons for assuming a Latin origin for
the word are apparently (1) that the aediles derived their name from the Roman
temple of Vesta or of Ceres, and (2) that quaestor, praetor, duouir, and censor are
probably Latin borrowings. The former is a spurious argument, for the fact that
Roman tradition connected the origin of their aediles to a local temple does not
exclude the possibility that the institution was more widespread and did not
(everywhere) go back to the Roman institution; the latter is a false comparison,
for quaestor, praetor, duouir, and censor are all attested from public inscriptions
and cursus honorum from Roman Falerii Novi, whereas the efiles are named in
connection with pre-Roman dedications at Falerii Veteres.
96
Indications for the existence of a Sabellic aif... → ef... have been seen in (1) the Aequian
toponym Aefula, (2) the Samnitic gentilicium aífineís Sa 31 (also read as aṿfineís), and (3)
Palaeovolscian (?) efieị or efieṣ in iúkúh  ko  efieṣ VM 1. Cf. WOU s.v. aídíl.
179
CHAPTER 6
2. argentum n. ‘silver’. Acc. sg. (n.) arcentelom EF 1, which is usually regarded as a

diminutive either of the noun itself, or of the corresponding adjective (see §12.2).
[I do not adopt Martzloff’s (2006:68-9) derivation of arcentelom from arceo (cf.
adulesc-ent-ulus), similar in sense to Greek ¢lexif£rmakon.] – Direct Latin
equivalent argentum: Oscan has a formal equivalent aragetud Cm 7, araget[ud
Cm 6, arage|[?nteis TB A.5, but the meaning here is ‘money’ rather than ‘silver’
(cf. WOU s.v. aragetud), which may be due to Greek influence. In the unlikely
case that Pisani (1964:71) was right in connecting the n-less Oscan forms to
Greek ¢rg»j, there is no direct Sabellic equivalent at all.
3. bibo vb. ‘to drink’. 1st sg. fut. act. pipafo MF 59, ‹pi›pafo MF 60. Editors have
questioned whether pipafo and pafo represent two different formations or if one is
simply an error for the other. Since the two kylikes on which these forms occur
were obviously meant as a pair, I assume that pafo is an error for ‹pi›pafo. – Since
Latin bibo is due to an assimilation of */pib-/ → /bib-/ after the reduplicative pre-
sents at an unknown date, Faliscan pip- may represent either /pib-/ or /bib-/. The
-a- can be explained in various ways (see3.2.1.13): the verb may have been an
athematic laryngeal verb /piba-/ or /biba-/ (← */pi-bħ3-/ ← */pi-pħ3-/), but also
/pibā-/ or /bibā-/, assuming that it was included in the a-conjugation (thus e.g. G.
Giacomelli 1963:152, 238-9). – Direct Latin equivalent or close cognate bibo. In
Sabellic, there is only a very distant cognate in Umbrian puni TI Ia.4 etc., which
has been explained differently (WOU s.v. poni.)
4. bis adv. ‘twice’: pis LF 242, pi LF 242. – Direct Latin equivalent bis ← duis (at-
tested in Cic. Or. 153 and Paul. Fest. 58.17L). No direct Sabellic equivalent, al-
though Umbrian duti TI VIb.63 corresponds in sense; Umbrian dupursus TI
VIb.10, the equivalent of Latin bipes, shows a compound formed with du- rather
than with dui- (WOU s.vv. dupursus, dur). See also 18. duo.
5. bonus adj. ‘good’. Gen. sg. f. duenas EF 2; nom. sg. n. duenom EF 2. The relation
of this word to manus (below) is unclear. – Direct Lat equivalent duenos (already
in CIL I2.4) o bonus, of unclear etymology. Sabellic on the other hand has
*cupros (see WOU s.v. cubrar) in South Picene kuprí AQ.2 (adv.) and qupíríh
AP.2 (adv.), and in Umbrian cubrar Um 7, cupras Um 17, cupr[as Um 20; cf.
also “ciprum Sabine bonum”, Var. L. 5.159.
6. careo vb. ‘to lack’: 1st sg. fut. act. carefo MF 59, carẹ[f]o MF 60. – Direct Latin
equivalent careo. No Sabellic equivalents, but perhaps remote cognates in
Umbrian kastruvuf TI Va.13 etc., castruo TI VIa.30 etc., and Oscan castrid TB 8,
castrous TB 13 (cf. WOU s.v. castrous), and perhaps Oscan kasit Cp 33, Cp 34,
kas[it Cp 33 (cf. WOU s.v. kasit), if these are derived from the same */kas-/ ←
*/ḱħ2s-/ that underlies Latin careo, castus, and castra.
180
THE LEXICON
7. carus adj. ‘dear, beloved’. Dat. sg. f. karai EF 1. – Direct Latin equivalent carus.
Probably derived from an Italic */kāro-/ ← PIE */keħ2ro-/. No known Sabellic
equivalent or cognate.
8. cella n. ‘burial chamber, tomb’. Nom. sg. cela MF 12, cela MF 83, cela MF 84,
cela MLF 285. (The uncertain [---]cẹla[---] in MF 166 occurs in an inscription on
a tile and, if an attestation of cela at all, would refer to the loculus instead of to the
burial chamber.) – Direct Latin equivalent cella, although this is never used for a
tomb (TLL 3.759,19-761,80), perhaps because chamber tombs did not play a role
in the burial ritual of Latium. Samnitic kellaked Sa 14, 15, referring to the con-
struction of a cistern (?), may be influenced by or borrowed from Latin (WOU s.v.
kellaked). The etymology of cella is unclear (cf. EDL s.v.).
Bréyer (1993:341-2) assumes that Latin cella is a borrowing from Etruscan cela
‘burial chamber, tomb’, occurring in vel  aties  velθurus  lemniσa  celati  cẹsu
Ta 1.66 and cela  sal  θn Vc 0.40. This is based largely on the assumption that
Latin cella did not have the meaning ‘burial chamber, tomb’: she does not men-
tion the Faliscan attestations, in spite of the fact that these are the only contempo-
rary instances of cella, clearly have the meaning ‘burial chamber, tomb’, and oc-
cur in precisely the same context as the Etruscan attestations. According to
Bréyer, the original meaning was ‘Grabkammer’, which was then extended in
Latin (apparently within Latium) to ‘enge Kammer, enger Raum’ and thence to
‘Vorratskammer’, whether underground or above ground.
In my view, the material can be interpreted equally well, if not better, by assuming
an original Latin-Faliscan word cella with the meaning ‘covered or enclosed
room, chamber’ (whether underground or above ground), which could easily be
applied to underground burial chambers in areas where these were used. (Faliscan
lectus and cubo are also everyday Latin words with a secondary funerary mean-
ing.) Its incidental occurrence in Etruscan texts would then have to be ascribed to
interference from Faliscan or South Etrurian Latin. Bréyer’s point that Etruscan
would have had no reason to borrow a word for a type of tomb that was (exclu-
sively?) Etruscan is spurious, since cela is clearly not a borrowing, but an inter-
ferential form occurring in only two inscription in stead of the normal Etruscan
word for ‘tomb’, namely σuθi/σ́uθi (ET lists at least 60 instances, not counting the
numerous instances of the derivations in σ́uθin-).
9. [censor n. ‘censor’. Latino-Faliscan attestations only. Nom. sg. cen]|so LtF 231,
censo LtF 232, [---]sor LtF 232; probably also [--- ce(n)s]or LtF 233 (or is this [-
-- ux]or?). The word may well be a borrowed Latin term for a Roman magistracy
introduced after the war of 241. – Direct Latin equivalent censor: Oscan, censtur
TB 27, 28 kenzsur Fr 1, censtur TB 8, 20 and Samnitic keenzstur Sa 4 are borrow-
ings from Latin (WOU s.v. kenzsur) that formed the basis for kensortathi Lu 5.]
181
CHAPTER 6
10. °clipeus/clipeum n. ‘(round) shield, buckler’. Attested in the gentilicium Cli-

pearius (‘Shieldmaker’) in clipeaio (clipea‹r›io?) MF 470*, clipịaṛ[io] LF 230
[and clipeaṛ[io] LtF 231, cḷ[i]peario LtF 233]. – Direct Latin equivalent cli-
peus/clupeus or clipeum/clupeum, of unknown etymology. The variation clip-/
clup- may point to a borrowing, but can also be explained within the phonology
of Faliscan, see §3.6.4). The suffix -eus has been compared with borrowings from
Etruscan (DÉ, LEW s.v. clipeus), and Etruscan origin of the word is assumed also
by Bréyer (1993:291-2, adducing Clipearius). No known Sabellic equivalent or
cognate.
11. [? coniunx n. ‘wife’. Latino-Faliscan and Latin attestations only. The attestations,
con[---] LtF 174, and cọ* LtF/Lat 341, are very uncertain. Note that in Latin
sepulchral inscriptions the word used is always uxor, never coniunx (cf. §7.4.2).
See also 87. uxor.]
12. cras adv. ‘tomorrow’. cra MF 59-60. – Direct Latin equivalent cras, of unclear
etymology. No known Sabellic equivalent or cognate.
13. cubo vb ‘to lie’, spec. ‘to lie in a tomb or grave’. (1) 3rd sg. pr. act. ind. cupat MF
40, c]up[at] MF 159, cup[a] MF 161, cupat MF 220, cupa MLF 305, cupa LF
221, cupat LF 224, [also cubat LtF 231]; (2) 3rd pl. pr. ind. cupa]nt MF 80, cupat
MF 146, cupạṭ MF 158, cupat LF 223; (3) either sg. or pl. cupa[?t] MF 95,
cu[pa] LF 226 [and cuba LtF 326]. The attestations are all from the sepulchral
formula hec cupat/cupant, ‘lie(s) here’, where the verb has a secondary meaning
within the lexical subset related to burial (§6.3.2.4). For the formula, which is
nowhere attested with the frequency it has in the Faliscan sepulchral inscriptions,
see §8.10.1. – Direct Latin equivalent cubo (used de mortuis in CIL I2.1259, 1638
and 2135 (quoted in §8.10.1) and Lucr. 3.892,97 cf. TLL 4.1278,82-1279,9); direct
Sabellic equivalents in South Picene qupat MC.1, qupat AP.3, Vestinian cibat MV
7, and, slightly differently (a transitive compound), Marrucinian encubat MV 8,
Paelignian incubat Pg 10 (all from sepulchral inscriptions); cf. also “cumbam
Sabini uocant eam quam militares lecticam” Paul. Fest. 56.26L. See also 40.
*lego or *legeo.
14. cupido n. ‘(sexual) desire’. Attested as as theonym Cupido ‘Desire’ in cupi‹d›o
MF 62. See also §6.4.3. – Direct Latin equivalent cupido (first attestation also in
the theonym cupido ILLRP 1204). No known Sabellic equivalents; a remote
cognate may be *cupros ‘good’ (cf. WOU s.v. cubrar). The regular Sabellic root
for ‘desire’ would appear to have been /her-/ ← PIE */ǵher-/ (cf. WOU s.vv.
herentas, heriiad).
97
With ossa as subject, cubo occurs in CIL I2.1312 from Rome and in Ovid. Am. 1.8.108 =
Ep. 7.162 = Tr. 3.3.76 (a mock-epitaph).
182
THE LEXICON
15. *dies n. ‘day’. Attested indirectly in the adv. foied MF 59-60 (either from an
ablative */hō+dē(d)/ or from a compound */ho-dē(d)/, see 34. hodie); [related is
the theonym [die]s pater MF 62, see §6.4.4.]. – Direct Latin equivalent dies: the
Oscan semantic equivalent zicolom TB 14 etc. shows a different formation */dē-
k°lom/ (cf. WOU s.v. zicolom).
16. do vb. ‘to give’. (1) 3rd sg. pf. act. of a compound *por-do in porded EF 1 [and 3rd
sg. pf. act. ind. dedet LF/Lat 214]; [(2) nom. sg. n. pf. ptc. datu LF/Lat 214].
Porded has often been explained as a reduplicative perfect that lost its medial
syllable, but it is probably rather an old aorist: see §5.3.1.14. – The simple verb
from the root /*deħ3-/ is well-attested both for Latin (do, reduplicative perfect
dedi) and for the Sabellic languages (reduplicative present *did- in Umbrian,
Vestinian, and Paelignian; reduplicative perfect *ded- in Umbrian, Marsian,
Paelignian, and Oscan: for attestations see WOU s.v. didet). The compound with
por- has a cognate only in Umbrian purtuvitu TI IIa.24 etc., which has a different
stem. See also 17. *duo or *duio.
17. *duo or *duio vb. ‘to give’. 3rd sg. pr. act. subj. douiad EF 1, also restored by
Herbig (1913:75) in the damaged first part of the same text as [dou]ịad (the most
widely adopted reading of this word, see also 93. *[3-4]*ad). Faliscan douiad has
rightly been compared to the Latin subjunctive form duam (cf. DÉ, LEW s.vv. dō,
duam, EDL s.v. do). It is derived from a verbal root /*deħ3-/ or */doħ3-/
(/*do()ħ3-/ Meiser 1986:186-91) that occurs also in Umbrian purtuvitu TI IIa.24
etc., (cf. WOU s.v. pordouitu, Meiser 2003:182-3), related to the root of Latin do
(/*deħ3-/). See also §5.3.1.4. – Direct Latin equivalent in duam etc., and in
Umbrian purtuvitu TI IIa.24 etc. (for attestations, see WOU s.v. purdouitu). See
also 16. do.
18. ? duo card. ‘two’. Several editors have divided fitaidupes in EF 1 as fitai dupes,
interpreting dupes either as an equivalent of Latin bipes (Thulin 1908:259), or as
an equivalent of Latin dupondius (Vetter 1925:29-30). Note that in Latin, duis
used as the compound form only in the old formations ducenti, duplex, and
dupondius: later formations have bi- ← *dui- (cf. above s.v. bis) or duo- (see 19.
duouir). Direct Umbrian equivalent dupursus TI VIb.10 (the equivalent of Latin
bipes). See also 4. bis.
19. duouir n. ‘member of the board of two’. Late Faliscan attestations only. Gen. pl.?
duum]|uiru LF 243, duum[uir LF 247, duu[muiru LF 248, [duu]ṃuiru LF 249
(all from cursus honorum, cf. §2.3.3). (Duumuirum is an analogical formation af-
ter the genitive plural, cf. DÉ s.vv. duouir, uir, LEW s.vv. duomvir, vir; IEW s.v.
uiro-s, EDL s.v. vir). The word is in all probability an imported Latin word du-
ouir. No Sabellic equivalent. See also 89. uir.
183
CHAPTER 6
20. ego pers. pron. 1st sg. ‘I’. Nom. sg. eqo EF 1, 467*, eco EF 3, eko EF 6, EF 7; eco
LF 378, 383. (The eco read in MF 91 by Peruzzi (1964d:310-1) is too uncertain.)
The Faliscan forms probably represent /egō/, but it is possible (although unlikely)
that this was shortened to /ego/ as in Latin. See §4.7.1. – Direct Latin equivalent
ego /egō/ → /ego/, first attested as eco in 479†: Sabellic equivalents are South
Picene ekú- in ekúsim CH.1, and perhaps Samnitic íív Sa 31: see §4.7.1 and WOU
s.vv. ekúsim, íív. See also 47. me.
21. ? eita- (Etr.) n./adj. (?). eitam EF 5. Either a noun or an adjective with an Italic
ending, or an (adaptation of) an Etruscan word (Bakkum 1991): Peruzzi (1964a:
169-70) compared Etruscan itan. R. Giacomelli’s (1978:78-82) comparison with
the much later Oscan eítiuvam Po 3 etc. is unconvincing, both because it entails a
different suffix and because it requires a concept of ‘money’ two centuries before
the first Etruscan money was coined.
22. facio vb ‘to make’. 3rd sg. pf. act. ind. facet MF 471*, faced MF 470*. – The verb
has direct parallels in Latin (facio) as well as in the Sabellic languages (present
*fac-/*faci- attested for Umbrian and Oscan, for attestations see WOU s.v.
fakiiad). For the use of this verb in signatures, see §8.9.2. The formation of the
perfect, however, was very different in the various languages: beside Faliscan
faced/facet stand Latin fēci, Umbrian fut. pf. fakust TI IV.31, fakurent TI Ib.32,
facurent TI VIIa.43, but also reduplicative perfects like Latin vhevhaked CIL I2.3,
Oscan fefacid TB 10, fut. pf. fefacust TB 11, 17 and fef[acust] TB 33, and
Praesamnitic fefakid Ps 20. For a discussion of these forms, see §5.3.1.6. – Some
authors have interpreted fifiked EF 9 and f[.f]ịqod EF 1 as forms of the same verb:
this was already rejected by Lejeune (1955): see §5.3.1.8.
23. far n. ‘emmer (Triticum dicoccum Schr)’. Acc. far EF 1. [I reject Pisani’s
farṃẹ[n]ṭom (1946:54) for phonological reasons.] – Direct equivalents in Latin
far, Umbrian far TI Vb.10, 15, farer TI Vb 9, 14 (and the derivations farariur Um
9 and farsio TI VIb.2, fasiu TI IIa.12, fasio TI VIb.44) and Oscan far Cp 37. There
are cognates in other IE languages (cf. Polomé 1992:69), but the word may
ultimately be of non-IE origin (cf. Schrijver 1991:113-4, WOU s.v. far).
24. filia n. ‘daughter’. filea MF 14; abbreviated to f in MF 155, LF 229, 234?, 242,
249 [and LtF 231, LtF/Lat 300, 305];
25. filius n. ‘son’. fileo MF 470*, hileo MF 146; fragmentary fi|[?leo MF 94; abbre-
viated to file MLF 308, abbreviated to fi MF 15, abbreviated to f||e LF 332, ab-
breviated to f LF 213, 234?, 242, 247, 249 [and LtF 171, 172, 174, 231, 327,
LF/Lat 214, 325, Lat 216]. [I reject Herbig’s (1914b:251) interpretation of tito po-
lafio MLF 354 as tito pola fio ‘Titus Pola filius’, with a fio as a palatalized [fīo],
see §3.5.5c.]
184
THE LEXICON
Direct Latin equivalents filia and filius. These words are Latin-Faliscan innova-
tions (cf. Lejeune 1967, Hamp 1972): The corresponding Sabellic words are
*puclom ‘son’ in South poqloh AQ.1, Marsian pucle[s VM 4, Paelignian puclois
Pg 5, Oscan puklum Cp 37,4 etc., and Oscan futír ‘daughter’ Si 8, 9, Samnitic
fuutreí Sa 1B.5, futreí Sa 1A.4, and futre[ís Sa 30, which continues PIE */putlom/
and */dhugħ2tēr/.98 Umbrian once has abbreviated fel ‘son’ Um 29, perhaps an
interferential form from Latin or Faliscan (cf. WOU s.v. fel: the inscription shows
both Latin and Etruscan features). *Puclom has distant cognate in Latin puer; a
Faliscan cognate *putellius ‘infant’ was read by Vetter (1953:303-4) in MF 152 as
‘infant’, but I reject this for phonological reasons (see 62. †putellius). Filius and
filia reflect PIE */dhħ1ilio-/ or */dheħ1ilio-/ (cf. Schrijver 1991:242), not forma-
tions based on Proto-Italic */fēlā/ ‘breast’ ← PIE */dheħ1leħ2/ that underlies the
Latin verb felo/fello, and its Umbrian cognate feliuf TI Ia.14, filiu TI VIb.3 (cf.
DÉ, LEW s.v. fīlius, IEW s.v. dhe(i)-, and WOU s.v. feliuf).
26. fingo vb. ‘to form, to knead’. (1) 3rd sg. pf. act. ind. fifiked EF 9; (2) 3rd pl. pf.
act. ind. f[.f]ịqod (=f[.f]ịqo(n)d) EF 1 (usually read as f[if]ịqod, although it is also
possible to read f[ef]ịqod). The attestations are both signatures of the iscrizione
parlante-type ‘... made me’, where the use of this word may have been formulaic.
[(3) very uncertain is nom. or dat. sg. f. pf. ptc. fita or fitai read in in fitaidupes EF
1, cf. below s.v. fita-/fitai-.]. – The attestation of fifiked is beyond doubt (cf. now
Gulinelli 1996), and on this basis f[if]ịqod, which goes back to Herbig (1913:74-
80) and Buonamici (1913:40), has become the accepted reading, probably rightly.
Both forms show a reduplicative perfect with (at least in fifiked) assimilation of
the vowel of the reduplicative syllable to that of the root (as happened e.g. in
Latin spopondi, tutudi etc.).
Since the root underlying the word was */dheiǵh-/ (DÉ, LEW s.v. fingō), the
presence of an intervocalic /g/ must be due to analogy after the present stem
*/dhingh-/ (DÉ, LEW s.v. fingō), where (*/ǵh/ →) */gh/ regularly developed to /g/
after the nasal infix. Alternatively, intervocalic */gh/ may have developed to /g/
instead of to /h/ in Faliscan: see §3.3.3. Although the verb is therefore the same as
in Latin, the formation of the perfect was different, Faliscan having an old
reduplicative perfect with assimilation of the vowel in the reduplicative syllable,
and Latin having an s-perfect with analogical extension of the nasal infix finxi,
perhaps, however, as a replacement of an earlier reduplicative perfect (§5.3.1.8) –
Latin equivalent fingo; distant cognates in Oscan feíhúss CA B.5, feíhúís CA B.19
(cf. WOU s.v. feíhúss).
98
DÉ s.v. fīlius erroneously stated: “L’italo-celtique a perdu les noms indo-européens du
«fils» (got. sunus, etc.) et de la «fille» (got. dauhtar, etc.). Ces noms ont été remplacés par des
noms nouveaux, familiers, ou fabriqués.”
185
CHAPTER 6
27. ? °firmus. The name Firmius/Hirmius in ḥiṛṃeo MF 19, firmio MF 54, hirmio LF
213; f. ḥirmia MF 18, firmia MLF 302 has been connected with the adjective
firmus (← PIE */dhermo-/) e.g. by Campanile (1961:5-6), although the
connection was rejected by G. Giacomelli (1963:193).
28. ? fit/fita/fitai. Editors have usually divided fitaidupes in EF 1 either as fita idupes
or as fitai dupes (fit aidupes Martzloff 2006:66-74). The resulting fita or fitai is
usually connected either (1) to Latin fingo and its derivations in fict- (e.g. Thulin
1908:259, Herbig CIE 8079, G. Giacomelli 1963:41-3, Peruzzi 1964a:161-2), (2)
to Latin fio (Stolte 1929:107, Vetter 1953:280, Martzloff 2006:66-74) or Umbrian
fitu TI VIb.11 (Herbig CIE 8079), and (3) to Latin fetus (Pisani 1946:53). For a
discussion of these suggestions, all highly conjectural, see §5.3.2.26 and §12.2.
29. ? °frenum n. ‘rein’: perhaps attested indirectly in the gentilicium Frenaius or
Frenarius, in frenaios or frenarios MF 471*. Latin equivalent frenum (probably
derived from frendo (DÉ, LEW, EDL s.v. frēnum). No known Sabellic equivalents
or cognates.
30. gero vb. ‘to wear’, spec. (with magistratum) ‘to function as a magistrate’. 3rd sg.
pf. act. ind. keset LF 242, kese[t LF 243. The Faliscan forms represent /ges-s-/
with an s-perfect as in Latin. Both Faliscan attestations occur in the formulaic
phrase magistratum gessit which is identical with the phrase used in Latin (TLL
62.1939,1-1940,56): cf. Renzettti Marra 1990:339-40. – Direct Latin equivalent
gero with pf. gessi.
31. ? *gutto/*guttum/guttur n. ‘pitcher’. Nom. sg. quto EF 3. (The word is often read
as qutoṇ, which I regard as impossible.) Quto is a rendering or adaptation of the
Etruscan word that appears as qutun (e.g. in mi qutun lemausnas Etr III from
Narce), qutum Cm 3.1, Cr 2.18, 19, 30, qutumuza Ve 2.1, and quṭus Vs 1.116,
q̣utus Vs 1.120. Whether in EF 3 it is a borrowing or an interferential form cannot
be said: neither does the occurrence, in Latin, of borrowings such as guttus and
guttur throw any light on this. The phonological form represented by quto is
unclear. The q- can represent either /k-/, as in the Greek kèqwn or Hesychius’
kîqon (k 4788 Latte) from which the word is ultimately derived (cf. Colonna
1974:140-2), or /g-/ as in Latin guttur, guttus etc. which are thought to have a
similar origin. The -u- may represent either /ū/, preserving the long vowel of the
original Greek word, or /u/ as in Latin guttur, guttus etc.; the -t- may likewise
represent /t/ or /tt/ (cf. Bréyer 1993:198-9 on cuturnium, 209-10 on gūtus/guttus).
Reading the ending -o as an ōn-stem nominative /-ō/ or as an o-stem masculine
nominative -o(s) /-os/ is impossible as the word appears to be qualified by the
neuter adjective duenom (eco quto *e uotenosio titias duenom duenas). The
ending might represent a second-declension neuter nominative -o(m) /-om/, but
186
THE LEXICON
this requires that an omission of -m that is without parallels in Early Faliscan

(§3.5.7a). Alternatively, the word might perhaps be derived, not from Greek
kèqwn or kîqon, but (through a hypothetical Etruscan intermediary *qutur?) from
a Greek *cÚtwr, in which case the ending might represent -o(r) (cf. Latin guttur).
This requires assuming an omission of -r that is unparallelled (although not
impossible) in Early Faliscan, but makes it more probable that the word was
neuter. – Possibly a direct Latin equivalent guttur and cognate in guttus. No
Sabellic equivalents or cognates.
32. [haruspex n. ‘haruspex’. Latino-Faliscan attestations only: harịṣp ̣[ex LtF 231,
harisp[ex LtF 232. No known Sabellic equivalents. Bréyer (1993:351-4) rightly
argues against an assumed Etruscan origin of the word, which was based on the
older false readings of the Faliscan attestations such as haracna.]
33. hic adv. ‘here’. ḥec MF 88, hec MF 94, hec MF 146, he[ MF 149, hec MF 158,
he LF 220, he LF 221, hẹc ̣ LF 223, he LF 224, [h]e LF 226; [and heic LtF 231];
hypercorrective spelling with f- for h- in fe[(c?) MLF 56, fe MF 305. [I do not
adopted the hac ̣ ‘hanc’ read by Vetter (1953:301) in MF 89.) All attestations are
from the sepulchral formula hec cupat/cupant ‘lie(s) here’. The adverb is a
fossilized locative */he-ke/ of a demonstrative pronoun /ho-/. The same pronoun
also underlies foied (see 34. hodie) – Direct Latin equivalent heic → hic. No
cognate in Sabellic languages, where the pronominal root /ho-/ is absent and
/eko-/ is used instead (cf. §4.8).
34. hodie adv. ‘today’. foied MF 59-60 (with hypercorrect f- for h-, see §3.5.2). It
cannot be ascertained whether the Faliscan form represents /hōēd/ from a
fossilized ablatival phrase /hō+dē(d)/, or /hoēd/ with a short /o/ as in Latin,
either reflecting a compound form */ho-dē(d)/, or the result of a shortening of an
earlier */hōdiē/, but hardly a locative (as Meiser (1998:78) suggests). The /-d/ in
foied may be due to its presence in the ablative underlying the word, but as in the
fifth declension the d-ablative was an analogical development after that of the o-
stems and the a-stems, it may also have been added to an already existing adverb
*hoie after analogy with those adverbs that had /-d/ because they where derived
from o-stem or a-stem d-ablatives: see §4.6.4 and G. Giacomelli 1963:150. –
Direct Latin equivalent hodie. No cognate in Sabellic languages: the existence of
a direct formal equivalent in Sabellic is unlikely, as in Sabellic the pronominal
root /ho-/ is not used, and the equivalent of dies, Oscan zicolom TB 14 etc., is
derived differently.
35. hutị[.]ilom EF 1. Most modern editors have adopted Vetter’s (1953:280)
restoration hutị[c]ilom: this has been interpreted in various ways (see §12.2),
most of which involve a derivation from PIE */ǵheu-/ ‘to pour’. If that derivation
187
CHAPTER 6
is correct, Faliscan would apparently differ from Latin, where the derivations
from this root (fundo etc.) all have /#fV/ (DÉ, LEW, EDL s.v. fundō): see §3.3.3.
The only author to have suggested a fundamentally different restoration is Radke
(1965:138), restoring huti[p]ilom, but his derivation of this word as Etruscan huθ
‘four’ + a Latin suffix apparently related to -plus/-plum ‘-fold’ is unconvincing:
there is no reason why Faliscan should derive such a word from an Etruscan
numeral rather than from its word for ‘four’ (cf. also §1.3.2.2).
36. † indu- praef. ‘in, within’. G. Giacomelli (first 1963:41, 43) and Peruzzi (1964a:
163-4) divided fitaidupes EF 1 as fita idupes, interpreting idupes as containing a
praefix i(n)du- equivalent to Latin indu- ← endo /endo/. It is is very unlikely,
however, that Early Faliscan already had the form indu- when Latin still had endo
in CIL 12.4: the closing of the vowels took place only at a later date (cf. §3.6.6.1).
37. infra adv. ‘below, underneath’. ifra MF 40, in cupat ifra ‘lies below’, a variation
on the usual sepulchral formula hic cuba(n)t ‘lie(s) here’. – Direct Latin
equivalent infra, with a lengthened /ī/: whether Faliscan had /i/ or /ī/ cannot be
ascertained (§3.5.6.1). The etymology of the Latin form is difficult, for in Roman
Latin an original Proto-Italic */dh(e)rā/ → Proto-Latin */ð(e)rād/ would have
become †end(e)rad /end(e)rād/ → †indrad /īndrād/ → †indra /īndrā/ (§3.3.3).
Infra is therefore explained either by assuming that it was originally a non-Roman
form (DÉ s.v. īnferus) or by assuming that, when the form was still */enð(e)rā/, it
was re-analysed as a compound */en+ð(e)rā/, so that the */ð/ developed to /f/, the
regular word-initial development (LEW, EDL s.v. īnferus). Both are ad hoc
solutions: the attestation of Faliscan ifra shows that the former is at least possible
(G. Giacomelli 1963b). No known Sabellic equivalent.
38. ? i*ice (vb., n., or adj.?). i*ice LF 309, ị*ice LF 315. The word is usually ex-
plained as a verb because of the structure of tito  uel|mineo  iun|aị ị*ice LF 315,
where editors have interpreted iunai as a dative, being reluctant to assume a geni-
tive in -ai for Faliscan (§4.2.2). Suggestions have been ip ̣ice = impĭgit pf. of
*impingo ‘to fasten upon’ (Herbig (1923:231-2), ip ̣ice = impīgit pf. of *impingo
‘to paint upon’ (Stolte 1926:61), an Etruscan verbal form in -ce (Ribezzo
1931b:192, Pisani 1964:341): see also §5.3.1.9. I regard it rather as a noun or an
adjective used as a cognomen, which fits the structure of both texts.
39. lectus n. ‘bed’, spec. ‘the burial place for the dead in a tomb or loculus’. Nom.
pl. lete MLF 285, abbreviated (acc. sg.?) let MLF 361, probably to be restored in 
iii  ḷ[.......|..........]naị[?---|....]o uxo MF 17 (either ‘the third bed …’ or ‘three
beds …’). [Also in lectu Lat 251 (twice), with the same meaning.] In Latin this
specific meaning of lectus (de mortuis, peculiariter de loco, ubi cadauer conditur,
TLL 72.1099.18-20) occurs only in CIL I2.1990=Lat 251. M. Mancini (2002:
188
THE LEXICON
28-33) therefore explains lete as */lētę̄ ̣ /, the monophthongized form of the loca-
tive of */lotā/ with the same meaning as lectus, unnecessarily introducing a hypo-
thetical word to avoid giving lete (and perhaps let) a meaning that is in fact at-
tested in the area, and assuming that Faliscan had a functional locative (cf. §8.2.1)
with an ending that could be monophthongized (§3.7.6). For the omission of the
syllable-final /k/ (lete = le(c)te), see §3.5.7c.
40. *lego or *legeo vb. ‘to lie’, spec. ‘to lie in a grave or tomb’: 3rd sg. pr. act. ind.
lecet MF 88. Once in place of the usual cubat in the sepulchral formula hec cupat
‘lie(s) here’, cf. §8.10.1. The Faliscan form is not without problems, since, as the
IE cognates point to an original */leǵh-/, the expected Faliscan form would be
*leh- */leh-/ rather than /leg-/. The /g/ would therefore have to be due to an anal-
ogy, unless it is assumed that in Faliscan intervocalic (*/ǵh/ →) */gh/ developed
to /g/ instead of to /h/: see §3.3.3. It is also unclear whether the Faliscan form
represents */leg-e-t/ (as would be implied by its IE cognates, cf. DÉ, LEW s.v. lec-
tus) or /legē-t/ (thus Vetter 1953:301): see §5.3.1.11. The verb was apparently re-
placed by cubo both in Faliscan (where cubo is the formulaic verb in sepulchral
inscriptions) and in Latin (where it is not attested at all, although its existence is
implied by the derivations lectus and lectica), in which case its occurrence in Fal-
iscan can be described as an archaism (R. Giacomelli 2006:42). Interestingly, the
unique attestation of South Picene veia|t /eāt/ MC.1 ← Proto-Sabellic */leγāt/
occurs side by side with qupat in apaes : qupat [: e]smín : púpúnis : n|ír : mefiín :
veia|t : vepetí MC.1. (See also DÉ, LEW s.v. lectus, WOU s.v. veiat).
41. liberta n. ‘freedwoman’ and libertus n. ‘freedman’: Nom. sg. f. l]oifirta MF 41,
loferta LF 221, perhaps abbreviated in ṭị [] ṭịria lo[?---]|l[e]a  cs  f MF 155.
Another attestation (masc. or fem?) perhaps in [---]*i  u[o]ltiai lo MF 165 [and
perhaps l LtF/Lat 292 (very uncertain)]. The word clearly designates the freedman
and freedwoman, although it is unclear what the status of the freedman was in
Faliscan society, and whether it differed from that of the Etruscan lautni and the
Latin libertus: see §2.3.2 and Rix 1994:94-6. – Direct Latin equivalents libertus
and liberta. No Sabellic equivalents or cognates. Liberta (and libertus!) has been
explained as a calque on the Etruscan feminine form lautniθa, but this has rightly
been rejected by Rix (1994:88-91).
42. libertas n. ‘freedom’. Gen. sg. loifiṛtato MF 31, loifirtato MF 32. – Direct Latin
equivalent libertas. No attested Sabellic equivalent, although there is a close cog-
nate in Paelignian loufir Pg 11 and abstract nouns derived by means of the suffix -
tas are attested from the Sabellic languages.
43. magistratus n. ‘magistracy’. Acc. (sg.?) macistratu LF 242 [and to be restored in
LF 243]. The attestation is from a formula appearing in LF 242 as macistratu |
189
CHAPTER 6
keset, which is directly equivalent to Latin magistratum gessit (TLL 62.1939,1-

1940,56) and may have been borrowed from Latin as part of the vocabulary
relating to public office (§6.3.6): this does not imply that the words themselves
did not exist in Faliscan. – Direct Latin equivalent magistratus. No attested
Sabellic equivalents or direct cognates.
44. °manus adj. ‘good’. Attested in the superlative used as a cognomen Manumus
‘Most Good’ (rather than ‘Best’) in man[o]mo MF 80 and [m]ano[m]o MF 149,
and either man]om[o or max]om[o MF 89; indirectly also in the gentilicium
Manius in mania LF 225, m{e}ania LF 224. In Latin, manus was ousted probably
already at an early date by bonus: since the latter occurs also in Early Faliscan, it
is interesting to find the superlative manumus as a cognomen in Middle Faliscan,
for if cognomina were a new element in the Middle Faliscan onomasticon (cf.
§7.9.2), then the adjective manus must have been current recently enough to be
remembered, especially as the form used for the cognomen is not the regular form
but the superlative. – In Latin, the attestations of manus are all from
glossographers (Var. L. 6.2.4, Fest. 112.24-5L, 132.3-7L, Paul. Fest. 109.4-7L,
133.10-2L, 151.6-7L, Macr. 1.3.13), where the word often has a sacral
connotation, e.g. “Matrem Matutam antiqui ob bonitatem appellabant, et maturum
idoneum usui, et mane principium diei, et inferi di Manes, et subpliciter boni
appellati essent, et in Carmine Saliari Cerus Manus appellatur creator bonus”
(Paul. Fest. 109.4-7). Apparently already by the time of the earliest attestations
that could still be found or remembered, the word was associated with divine
benevolence rather than human goodness, which may be why it disappeared from
common use: even its onomastic derivation in Latin, the praenomen Manius, was
not frequent. No Sabellic equivalent, although the stem ma- has been seen in
Samnitic maatúís Sa 1A.10, B.13 (WOU s.v. maatúís, DÉ, LEW s.v. mānis, -e).
45. mater n. ‘mother’. Nom. sg. mate LF 221. – Direct Latin equivalent mater; direct
Sabellic equivalents in South Picene matereíh AP.2, Umbrian matres Um 17, 19,
18, matrer Um 7, Samnitic maatreís Sa 30 (cultic epithet of a deity).
46. maxumus irregular superl. adj. ‘greatest’. Attested as a cognomen Maxumus/
Maximus in maximo MF 88, ṃaxoṃ[o] MF 98, ma]x̣ọmo MF 162, maxomo LF
220 and max]om[o or man]om[o MF 89, probably not [--- m]a*ome MF 156
(pace Colonna 1972c:446-7) – Direct Latin equivalent maxumus/maximus from
*/mag-isVmo-/ (see §3.6.6.1,2). The Sabellic languages had */mag-imo-/ →
*/maimo-/ → */mamo-/ attested in Oscan maimas TB 3, 7 (DÉ, LEW s.v. mag-
nus, WOU s.v. maimas).
47. me pers. pron. 1st sg. acc. ‘me’. Acc. sg. med EF 1, 9; met MF 470*. (The abl. sg.
[me]d has been proposed as a reading by G. Giacomelli (1963:44, 46, 1978:527)
190
THE LEXICON
in EF 3). For the various explanations of the Faliscan and Latin form med, see
§4.7.2. The -t in Middle Faliscan met can be explained either as an indirect
attestation of weakening of word-final consonants (§3.5.7c) and/or as a hyper-
correct form copying the replacement of /-d/ by /-t/ in ending of the third singular
perfect, cf. the facet that immediately follows met: see also §5.2.4e – Direct Latin
equivalent med /mēd/ o me /mē/. The only certain instance of a Sabellic semantic
equivalent is Palaeoumbrian míom Um 4=480†, with a suffix /-om/ that also
occurs in the personal pronouns of the second person and third person reflexive,
attested in Oscan tiium Cp 37 and siom TB 5, 6, 9 respectively (cf. WOU s.vv.
miom, siom, tiiom). See also §4.7.2. For the nominative ego, see 20. ego.
48. [minor irregular comp. adj. ‘smaller, younger’. Latino-Faliscan attestation only.
Probably in mino LtF 173, where its is apparently a woman’s name, although the
lexeme itself is likely to have existed in Faliscan. See §7.7.1.43.]
49. † nutrix n. ‘wet-nurse’. Herbig (CIE 8225) considered interpreting nut*[---] MF
103 as nutṛ[ix, which was adopted by Vetter (1953:302). G. Giacomelli (1963:82-
4) rightly rejected this, as the Middle Faliscan form would have been *notrix or
*noutrix (cf. Latin noutrix CIL I2.45): see §3.7.2. Peruzzi’s attempt (1964d:312)
to uphold Herbig’s interpretation by assuming that nutr[ix was a Luxuslehnwort
from Roman Latin is unconvincing: it would still be the oldest instance of the
spelling u for original /o/, only then in Roman Latin instead of in Faliscan. (Note
there are no attestations of ‘professional designations’ in the Faliscan sepulchral
inscriptions.)
50. ? °oct... ord. ‘eight’ or card. ‘eighth’. Perhaps attested in the praenomen oct*i[....]
MLF 353, which Herbig (CIE 8204) in fact read as oc ̣ṭọ. Most of the Latin names
in Oct- discussed by Schulze (1904) and Kajanto (1965) are derived from the
cardinal octo or the ordinal octauus, although oct*i[....] could conceivably
represent a name in /ō ̣kt-/ m /akt-/ (§3.7.4). If attested, there are direct
equivalents in Latin octo, octauus etc., as well as close cognates in the Oscan
names úhtavis Cp 36, ota#ij Lu 36.
51. † olla n. ‘urn, pot’. Olna in uel [] uisni  olna MF 82 was interpreted by Deecke
(1888:131-2) as olla, but this is based on an erroneous etymology. It is rather a
second gentilicium, as was first proposed by Schulze (1904:73 n.3).
52. ? ora- (vb.?) in tulate tulas urate EF/Etr 385. The text is regarded as Etruscan by
most editors and (therefore?) left unexplained. Pisani (1964:347) suggested a
connection with Oscan urust TB 14, 16 (and thus indirectly with Latin oro, cf.
WOU s.v. urust), interpreting it as ‘chiacchierate’. Although this is not without
difficulties (see §5.3.1.23), a connection between urate, urust, and oro is not
impossible.
191
CHAPTER 6
53. pario vb. ‘to bring forth, to give birth to’. 1st sg. pf. act. ind. pepara[i EF 1
(context unclear). Notwithstanding the punctation, this form has been read and
interpreted thus since Herbig (1913:84-7), and although other proposals have
been made (e.g. Peruzzi’s (1964a:160-4) pe  par a[dke]douiad ‘per par
accedat’and pe  para[te ke]douiad ‘’per parate accedat’), these are not
improvements, nor generally accepted. Assuming that the text is indeed to be read
thus, pepara[i (cf. Untermann 1968a:166 n.5) shows a reduplicative perfect
/pe-par-/ that would also have been the early form of Latin peperi (DÉ, LEW s.v.
pariō). Direct Latin equivalent peperi; no Sabellic equivalent, although the root
may have distant Sabellic cognates (cf. WOU s.v. amparitu, perstu).
54. [*pescum (Sab.) n. ‘votive offering’. Only in Torelli’s (1974:741-6) reading of
Cap 431, which starts with the word p ̣esco. This is a Sabellic word, cf. Marsian
pesco VM 5: it is usually derived from */perḱ-sḱ-o-m/, and related to Umbrian
persklum TI Ia.1 etc. (← */perḱ-sḱ-elo-m/): see WOU s.vv. pesco, persklum.99]
55. pater n. ‘father’. Only indirectly attested in the theonym [...]s pater MF 62,
restorated either as [ioui]s pater or [iouo]s pater, or as [die]s pater. – Direct
Latin equivalent pater, direct Sabellic equivalent Oscan patir Cp 2, dí{}pa{}tír
Po 22, etc.
56. por- praefix. See 16. do. Occurring in porded EF 1. In Latin, por- occurs in the
compounds polliceor, porgo, porricio, and portendo. (Cf. DE s.v. por- and porrō
in porod CIL I2.560.)
57. *postigna (Sab.) n. prob. ‘likeness, statue’?. Nom. sg. posticnu MLF 474*. The
form may be a Sabellic first-declension nominative singular (§4.2.1, §9.3.2), for
the only known equivalent or cognate is South Picene postiknam CH.2 (acc. sg.).
The meaning ‘statue’ assumed for both words is a conjecture based on the fact
that the Faliscan inscription, caui  tertinei  | posticnu 474*, occurs on a bronze
base which in the earliest drawings (Ritschl 1862 tab.XXXVI,B and Garrucci
1862 tav.IV.2) shows marks of the feet of a statuette.100 The etymology is
unknown (WOU s.v. postiknam): the word is perhaps a compound of Sabellic
*posti (cf. WOU s.v. pústin, also La Regina 1981:132). No known Latin cognate.
99
Untermann (WOU s.v. pesco) refers to Cap 431 but erroneously describes this inscription as
‘aus dem Marsergebiet (Luco AQ)’.
100
Untermann (WOU s.v. postiknam) all but rejects this, stating “der Gegenstand zeigt jedoch
keine Spur einer Statue (Lejeune).” Lejeune’s (1952:115) autopsy, however, is less decisive:
“Dans l’état actuel de l’objet, il n’y a pas de moindre trace de soudure de la statuette jadis
supportée par la base; à peine peut-on discerner l’emplacement du pied gauche à une légère
différence de coloration de la patine, et il est difficile d’en affirmer seulement autant pour la
pied droit; sans doute le bronze a-t-il subi un décapage?” (my italics throughout).
192
THE LEXICON
58. praetor n. ‘praetor’. Nom. sg. pretod LF 242, p[reto]|r LF 243, pret[or LF 247,
pre]tor LF 248 [and pretod LtF/Lat 214] (all from cursus honorum in sepulchral
inscriptions); nom. pl. pret[ores LF 213 (public work). The word may well be a
borrowed Latin term for a magistracy that was introduced in the ager Faliscus as
part of the Roman presence there after the war of 241. Although Latin praetor
was sometimes also used as a ‘translation’ or ‘equivalent’ of Etruscan zilaθ, it is
unlikely that this is the case in the Faliscan attestations, as these all occur in the
context of the Roman administration of Falerii Novi. – Direct Latin equivalent
praetor; the Oscan equivalent in the abbreviated pr TB 23, 27, 28 may be a
borrowing from Latin (cf. WOU s.v. pr). For the etymology, cf. also DÉ, LEW s.v.
praetor: Bréyer (1993:378) discusses the possible derivation from or connection
by popular etymology with Etruscan purθ (doubtful, as the Etruscan magistrate
equated with the praetor was not the purθ, but the zilaθ).
59. ? pramo- and propramo-. (1) Abl. sg. pramod, pramod, pramod EF 2; perhaps
also pro pramọḍ EF 2, although this may also be a form of the compound
propramo-; (2) adv. prameḍ, pramed EF 2; (3) compound nom. or acc. sg.
propramom, and abl. sg. propramọḍ EF 2, although this may also be read as also
pro pramọḍ. The repetition of these enigmatic words forms the core of EF 2. The
words are in all probability derived from a */prĦ-mo-/101 that also underlies Latin
prandium (← */prām(o)+edom/, cf. DÉ, LEW s.v. prāndium, IEW s.v. per 2.A,g).
Since prandium is usually interpreted as ‘early meal’ (like German Frühstück, or
Greek ¥riston), /prāmo-/ has been credited with a meaning ‘early’, even though
the reflexes of */prĦ-mo-/ in Germanic (Old Saxon formo, Old English furma)
and Baltic (Lithuanian pìrmas, Old Prussian pirmas) mean ‘first’ in the sense of
‘foremost’ rather than ‘earliest’. Pisani (first 1937:233) observed that pramed
could also be the subjunctive of a verb rather than an adverb, deriving it from a
*pramo (either *prāmāre or *prāmĕre) that would correspond to Latin promo.
60. ? pro… Unclear, occurring in pro[---] LF 244. Renzetti Marra (1990:338) inter-
preted pro[---] as “un pro[ elemento architettonico di cui è data la localizzazi-
one”. If I understand this correctly, it would refer to (a place in) the loculus,
which is elsewhere referred to as lectus (MF 17, MLF 285, perhaps MLF 361, and
Lat 251 (twice)); placing a body before another is described by anteponat in Lat
251. Pro[---] may be part of a (woman’s) name, perhaps pro[tacia], the gen-
tilicium attested from the same tomb in LF 242, 244.
61. † puia (Etr.) n. ‘wife’. Although the possibility of reading puia in [.?]ạ*ịạ | lepuia
| uoltilia MF 144 was rejected already by Herbig (CIE 8243a), this reading is still
101
The nature of the laryngeal is uncertain in view of West Greek and Boeotian pr©toj vs.
Attic-Ionic prîtoj (cf. Beekes 1969:214-6, who seems to favour /ħ3/).
193
CHAPTER 6
adopted by Vetter (1953:305). It would be the only sepulchral inscription where

[HUSBANDGEN WIFE] precedes [FILIATION] (attestations in §7.4.2). Note also that
[HUSBANDGEN WIFE] or [WIFE HUSBANDGEN] is part of an (onomastic) formula in
which the word used for WIFE is always uxor (§7.4.2), which would make it even
more unlikely that it would be replaced by puia. If in spite of these objections
puia is still adopted, it is likely to be an interferential form, not a borrowing.
62. † *putellius/*putellium n. ‘infant’; *putellus adj. ‘little’. Vetter (1953:303-4, 80)
interpreted putellio in uolti[o ] marc[---] | putellio MF 152 as ‘infant’ and on this
basis restored pu]tela ‘parvula’ in EF 1. The main problem is that the noun
*putellius/*putellium and the adjective *putellus are to be derived from PIE */pu-
tlom/, but that the suffix */-tlom/ was continued in Italic as /-klom/: */putlom/ in
fact occurs in the Sabellic languages as *puclum in South Picene poqloh AQ.1,
Marsian pucle[s VM 4, Paelignian puclois Pg 5, Oscan puklum Cp 37,4 etc. (In
spite of this, R. Giacomelli (2006:42) still considers putellio as a possible
‘sabinismo’.) A *putellius/*putellium and *putellus would therefore require a
separate Faliscan development at the Proto-Italic stage, which is inadmissible
(§3.1.2). *Putellius/ *putellium could be a later derivation from /pu-/ (which oc-
curs also in Latin puer/puella), but in that case it is unclear what the origin of the
suffix would be. It is easier to interpret putellio as an onomastic element. A better
case for a p... ‘child’ could in fact be made on the basis of cẹịṣ[i.] | holc[osi] | ar
 p[...] MF 140 (or is this another case of puia?).
63. ? °quartus ord. ‘fourth’: very dubiously attested[---]*[5-7] cua MF? 129, which
has been interpreted as an abbreviation of a name like Latin Quartus by Thulin
(1907:305), an interpretation that has been rejected by later editors.
64. quaestor n. ‘a magistracy, quaestor’. Nom. sg. cụestod LF 242, cues[tor] LF
243, c]ues[tor LF 245, cue[stor LF 247 [and possibly also [---]or LtF 233; q LtF
231 is probably not an attestation.] Probably (but not necessarily) a borrowing
from Latin. – Direct Latin equivalent quaestor: the Sabellic counterparts,
Umbrian kvestur TI Va.23, Vb.2 cvestur Um 9, and its derivative kvestretie TI
Ib.45, IIa.44, Oscan kvaísstur Po 3, Po 4, k]vaísstur Po 9, Po 10, kv]aísstur Po 14,
k#aistor Lu 6, Lu 7, k#]aistor Lu 10, k#ais. Lu 8, kvaístureí Cm 1A.2, kvaízstur
Po 8 are borrowings from Latin (WOU s.v. kvaísstur).
65. -que encl. conj. ‘and’. -cue MF 80, -c ̣ue MF 158, -cue MLF 313, probably also -
cuẹ MF 170, possibly also -c[ue MF 108. All attestations are from names joined
in sepulchral inscriptions. – Direct Latin equivalent -que. In the Sabellic lan-
guages, the formal equivalent -pe is often assumed to occur as a suffix (as e.g. in
Latin uterque) in Umbrian putrespe TI IV.14 and seipodruhpei TI VIa.11, but this
view has convincingly been challenged (cf. WOU s.v. pútereípid, seipodruhpei).
194
THE LEXICON
66. ? qui (rel.?) pron. ‘who’. (Nom. sg.?) cui LF 352 (not a locative or a dative, as
Renzetti Marra (1990:336-7) suggests: see §4.9). The context ([---?] precono[---|-
--] cuiteneṭ[---|---] let) is unclear: I doubt whether the text in fact contains a pro-
noun at all. – If attested, there is a direct Latin equivalent qui, and Sabellic
equivalents (with */k/ o /p/) in e.g. Umbrian pisi TI Va.3, 10 etc. (indef.) and
poi TI VIa.5, VIb.24 etc. (rel.), Oscan pis Cp 32 etc. (rel.) and paí Cm 1B.8 etc.
(rel.). (Cf. the indices to ST, and WOU s.vv. pis, poi for all Sabellic attestations).
67. °quinctus ord. ‘fifth’. Attested only in the praenomen cuicto MLF 310. The Latin
equivalent has /ī/ as a result of compensatory lengthening (cf. Pfister 1977:100,
Meiser 1998:78-9, 81): it is unclear whether this can also be assumed for Faliscan
(§3.5.6.1). – Direct Latin equivalent quin(c)tus, both in the lexicon and the ono-
masticon. The Sabellic equivalent, attested only in the onomasticon (Paelignian
ponties Pg 5, Oscan púntiịs Po 1, puntịẹis Cm 28, pomptiej Me 1, pomptoiej Me 3,
abbreviated pompt tSa 9, 11), shows an o-vocalism that probably originated in the
cardinal (cf. §3.2.10.1 and WOU s.vv. pomtis and púmperiais),
68. rex n. ‘king’ (perhaps a sacral title). Nom. sg. rex MF 90, perhaps also ṛẹx MF 91
(very uncertain), rẹ[x] LF 249 [and rex LtF 231]. The only title to occur both at
Falerii Veteres and at Falerii Novi: it may have been a sacral function, like that of
the Roman rex sacrorum (cf. §2.3.3). The title is found at the end of a cursus
honorum in LF 249 and LtF 231, either because the status of the office was very
high, or because the office fell outside the usual Latin cursus. – Direct equivalents
in Latin rex (already recei CIL I2.1, rex CIL I2.2830) and probably in Oscan rego
Lu 5 (but cf. WOU s.v. rego): a close cognate is Marrucinian recen[ai MV 1.
69. ? °russus adj. ‘reddish’. If the cognomen Ruso is read in cauio[---]|ruso[?---]
MLF 318 [and perhaps in ce  paui[ceo ru?]so LtF 290], this might be connected
to the adjective that appears in Latin as russus (note the unrhotacized s = /ss/).
70. sacer adj. ‘sacred’: nom. sg. f. sacra MF 127 (dedication) [and nom/acc. sg. n.
sacru LtF/Lat 214 (dedication)]; possibly abbreviated to sa MF? 76, 131, al-
though these are perhaps rather abbreviations of onomastic elements. – Direct
Latin equivalent sacer; direct Sabellic equivalents Umbrian sacru Um12, Um 13,
adv. sacre Um 6, sacr Um 19, and Oscan sakoro Me 1, Me 2, Me 3. An i-stem
*sacris occurs both in the Sabellic languages (cf. WOU s.v. sakrim) and Latin
(porci ... sacres Pl. Men. 289-90, sacrem Fest. 420.26ffL). Bréyer (1993:381-3)
discusses a possible Etruscan origin of the word.
71. salueo vb. ‘to be healthy, to fare well’ (imp. used as a greeting). 3rd sg. pr. imp.
salue[to]d EF 3, salueto EF 4; 2nd pl. pr. imp. saluete EF 4. – Direct Latin
equivalent salueo; in the Sabellic languages, the only cognate is only the adjective
*saluos. Cf. DÉ s.v. saluus, LEW s.v. salvus, WOU s.v. salavs.
195
CHAPTER 6
72. saluis adj. ‘healthy, well’. Nom. pl. salues EF 4. The Faliscan form is surprising
in that it apparently shows a *saluis where Latin and the Sabellic languages have
*saluos. Perhaps this may be compared to the coexistence of *sacros and *sacris
in Latin and the Sabellic languages (cf. 70. sacer). – Close cognate in Latin saluus
and Sabellic *saluos in Umbrian saluom TI VIa.51 etc., Marrucinian salaus MV
7, salas MV 6, Oscan salavs Cm 18, 38, 39, sala#j Lu 40.
73. °scaeuus adj. ‘(coming from the) left’ → ‘propitious, well-omened’ (“scaeua, id
est sinistra, quod quae sinistra sunt, bona auspicia existimantur” Var. L. 7.97). At-
tested only in the woman’s name Scaeua in sceụa MLF 312, sceiuai LF 379. –
Direct Latin equivalent scaeuus, which also seems to have been used in the ono-
masticon rather than in the lexicon. No known Sabellic equivalent except for the
Paelignian gentilicium scaifia Pg 14 (see also §7.7.1.57). Schulze (1904:369-419)
assumed an Etruscan origin for the word, which in view of Greek skai(#)Òj is un-
necessary. In Latin the word was also more common in the onomasticon (where it
was perhaps, like manus, because of its sacral association), and that this is re-
flected in its use in Faliscan and Paelignian.
74. °sextus ord. ‘sixth’. Attested indirectly in the praenomina z[e]xtos EF 1, sesto LF
329, zextoi LF 330, and sextia LF 311. (For the abbreviated attestations of this
praenomen, see §7.7.1.61.) In view of the uncertain quantity of the e of the corre-
sponding Latin form (cf. Pfister 1977:190), Faliscan *sexto- may represent either
/seksto-/ or /sēksto-/, perhaps the latter (cf. §3.5.7c). – Direct Latin equivalent
sextus; for the Sabellic languages only derivations of the cardinal are attested in
Umbrian sestentasiaru TI III.2 (WOU s.v. sestentasiaru), and Oscan sehsík[ Po 19
(WOU s.v sehsík[) and indirectly in the gentilicium sehsímbrịí ̣ṣ Po 36.
75. socia n. ‘(female) companion, girlfriend’. Early Faliscan attest-ations only: dat.
sg. soc ̣[iai] EF 1; nom./voc. pl sociai EF 2. – Direct Latin equivalent socia, of
uncertain etymology (DÉ, LEW s.v. socius, Schrijver 1991:249, Meiser 1998:98).
No known Sabellic equivalent or cognate.
76. † [*sorex n. ‘a type of priest, sorex’. Many editors have read sorex in LtF 231 (in
my view, this should be read as cen]|so rex) and restored it as sor[ex] in LtF 232.
First interpreted as a cognomen Sorex (Garrucci 1860:277-9), it soon became en-
tangled with the Hirpi Sorani and their cult of Apollo on Mount Soracte (§2.3.4),
which was also known as Sorax (Porph. in Hor. carm. 1.9). In spite of it being
based on an assumption only, this interpretation began to lead an independent life
to such an extent that Peruzzi (1963b:435-40) in fact re-analyzed the word as de-
rived from a */sor-ag-s/ ‘lot-shaker, cleromantis’.]
77. † statuo vb. ‘to place’, spec. ‘to place as a sacred object’. Thulin (1907:307)
interpreted statuo MF? 29 as the active counterpart to the sta(t) read by him in
196
THE LEXICON
MF? 128 and MF? 28, presumably similar in meaning to Greek ¢nat…qhmi. It is
rather a Besitzerinschrift ‘Stat. Vo.’. See also 78. †sto.
78. † sto vb. ‘to stand’, spec. ‘to stand as a sacred object’. Thulin (1907:304, 307)
interpreted [---]sta[?---] MF? 128, from the temple ‘ai Sassi Caduti’ and sta MF
28, as verbal forms sta(t) meaning ‘to stand (as a sacred object)’. It is rather a Be-
sitzerinschrift ‘Sta.’. See also 77. †statuo.
79. sum vb. ‘to be’. The verb is attested in three forms:
(a) 1st sg. pr. ind. esú Cap 389, 404, 465, a form occurring both in the Sabellic
languages (South Picene esom TE.4, Praesamnitic esum PS 4, 5, Hernician esu He
3, and Umbrian esu Um 18) and in the border of the Latin-speaking area (ṃorai
eṣo[m] from the ager Signinus (Colonna 1994) and pari med esom kom meois
sokiois trivoiai in the Garigliano inscription): for a discussion, see §5.3.1.5;
(b) 3rd pl. pr. ind. zot MLF 285 [and sot LtF 172?], which corresponds to Latin
sunt, older sont CIL I2.1529, while the Sabellic languages preserved the vocalism
of Proto-Italic */sent/ in Oscan sent Po 32 etc., set Cm 1A.16 etc. and Umbrian
sent TI VIa.15 etc.: see §5.3.1.24; and
(c) 2nd pl. pr. subj. seite = s{e}ite or s‹ie›te EF 4, which has comparable forms in
Umbrian (2nd sg. sir TI VIb.7, 26, sei TI VIa.23, si TI VIb.26, 3rd sg. si TI Va.6,
24, 27, Vb.3, 7, 3rd pl. sis TI Va.6, sins TI VIIb.4): see §5.3.1.18.
The verb itself of course has direct equivalents in Latin sum etc. and in the
Sabellic languages (for attestations, see WOU s.v. ezum).
80. ? sus... (n. or adj.?). The sus[---] in LF 227 is unclear: it may be part of a name, as
many editors have suggested (but in that case it would appear to be the only in-
stance of a woman’s cognomen); perhaps it is rather to be read as s us[or ?---] or
s ux̣[or ?---].
81. teneo vb. ‘to hold, to occupy’. 3rd sg. pr. act. ind. teneṭ in [---] cuiteneṭ[---|---] let
MLF 361. Reading, context (‘qui tenet ...’?) and the specific meaning of the word
here are unclear (cf. Renzetti Marra 1990:336-7) – Direct Latin equivalent teneo;
direct Umbrian equivalent tenitu TI VIb.25.
82. ? °titus (adj. ‘prosperous, propitious’?): dat. sg. m. titoi MF 113, tito MF 114, titoi
MF 116, [t]ito[i] MF 115, titoi MF 118, [t]itoi MF 119, [ti]toi MF 120, [ti]toi
MF 121, titoi MF 122. All the attestations are from the theonym Titus Mercus,
where Titus is probably a cultic epithet, since Mercus occurs by itself in MF 124-
126. It is unclear from these attestations whether the word may still have had a
lexical function or was already restricted to the onomasticon: the praenomen Titus
is not very frequent in the ager Faliscus (§7.7.1.74). – Latin equivalents in the
praenomen Titus and also titius (in the name of the Tities and the aues titiae ‘birds
197
CHAPTER 6
of good omen’, Var. L 5.81). The meaning of titus is debated: it is often explained
as ‘genius’ or ‘phallus’ (cf. e.g. Bréyer 1993:398-400), but at least for Latin and
Faliscan an adjective ‘propitious, well-omened’ is not unlikely (Combet Farnoux
1980:113-69). The interpretation of the South Picene equivalents titúí TE.5, titum
AP.1 and titiúh CH.2, titienom TE.3 is unclear (cf. WOU s.v. titienum, titúí, ti-
tiúh).
83. ? tol-/tul- vb. tulate and tulas in tulate tulas urate Etr/EC 385 have been regarded
by most editors as Etruscan and (therefore?) left unexplained: Pisani (1964:347)
connected tulate and tulas with Latin tollo (“sopportate - sopporta!”), in which
case it is a subjunctive based on a zero-grade root */tħ2-/: see §5.3.1.20-21. The
verb has direct parallels in Latin tollo from the zero-grade root */tnħ2-/ and in
Umbrian antentu TI Ia.20 etc., ententu TI Ib.12 etc., pertentu TI IIa.31, sutentu TI
IIa.23 (for all attestations, see WOU s.v. -tentu) from the full-grade root */telnħ2-/.
I do not adopt Herbig’s (1914:238 n.1) interpretation of tulom MF 68 as a per-
fect: see §5.3.1.22.
84. † ulna n. Several editors have interpreted olna in uel [] uisni  olna MF 82 as a
noun related to the burial (ulna ‘loculus’ Garrucci, ulna ‘pulvinus’ Vetter
1953:299). It is rather a second gentilicium, as was first proposed by Schulze
(1904:73 n.3).
85. ? *ummum n. ‘type of vase’. Early Faliscan attestation only. [u]mom, umom,
umọ[m] EF 2. Apart from a connection with Latin umidus and umor that was
rightly rejected already by Braun (in Giglioli 1935:241), there were no explana-
tions of this umom. M. Mancini (2003:239-41, 2004:205-7) rightly compares ud-
mom in the Hernician (?) inscription [---]matas udmom ni hvidas mi kait[sis ---]
He 2, which is probably the name of a type of vase. Rix (1998a:250-1) explains
the word as /ud-mom/ ← PIE */ed-/, comparing Latin unda ← */ud-nā/. The
Faliscan word probably represents */ummom/ ← */udmom/ (cf. §3.3.4.3).
86. urna n. ‘urn, vase’. Early Faliscan attestations only. Acc. sg. urnam EF 1; nom.
sg. of the diminutive ụrneḷ[a EF 1. – Direct Latin equivalent urna, of unclear
etymology, perhaps an adaptation of a borrowed word (see DÉ, LEW s.vv. urceus,
urna); Umbrian cognates perhaps in urnasier TI Va.2, 15 and urnasiaru TI III.3,
although other etymologies for this word have been proposed (see WOU s.v.
urnasier).
87. uxor n. ‘wife’. Nom. sg. uxo MF 17, uxor MF 41, uxor MF 42, ux[o(r) MF 43,
uxor MF 101, ux[o(r)] MF 102, uxo LF 222, uxo LF 242, uxor MF 265 [and uxo
LtF/Lat 300, [u]xor LtF/Lat 301], perhaps us[o(r) or ux̣[o(r) in LF 227 (see also
sus...). The usual word for ‘wife’, always in the female onomastic formula in se-
pulchral inscriptions (cf. §7.4.2). Direct Latin equivalent uxor, of unclear etymol-
198
THE LEXICON
ogy (DÉ, LEW s.v. uxor); possible direct Sabellic equivalents in Paelignian usur
Pg 9 (uncertain) and Oscan usurs Cp 37 (uncertain), usurum TB 6 (the interpreta-
tion of these forms is debated, see WOU s.v usur).
88. uinum n. ‘wine’. Acc. sg. uino MF 59-60; very doubtful is Herbig’s (CIE 8079)
restoration acc. sg. u[in]o[m] EF 1 (revived by Radke 1994). – Direct Latin
equivalent uinum; direct Sabellic equivalents in Umbrian vinu TI Ia.4, 22, Ib.6 etc.
and Volscian uinu VM 2. The i in the Early Faliscan, Umbrian, and Volscian
forms points to an original */iħ1nom/: for this form and its relation to its IE cog-
nates such as Greek (#)o‹noj, cf. Beekes 1987:22-3.
89. uir n. ‘man’. Late Faliscan attestations only. Attested indirectly in (gen.pl?)
duum]|uiru LF 243, duum[uir LF 247, duu[muiru LF 248, and [duu]ṃuiru LF
249, which is probably an imported Latin word. Although therefore it cannot be
used as an argument that uir also occurred in Faliscan, this seems likely in view
of its occurrence both in Latin uir and in Umbrian uiro TI VIa.42, 50 etc., al-
though the meaning of this word might differ slightly from that of its Latin
equivalent (cf. WOU s.v. uiro).
90. uos pers. pron. ‘you (pl.)’. Early Faliscan attestation only. Nom/voc. pl. ues EF 4.
The attestation is clear, but the vocalism is difficult to explain in view of that of
the direct equivalents Latin uos and Paelignian uus Pg 9 (twice) vs. that of the
possessive pronoun, Latin uoster/uester and Umbrian uestra TI VIb.61 (cf. DÉ
s.v. uōs, LEW s.v. vōs, WOU s.vv. uestra, uus): see §4.7.3.
91. *e[..]tom in EF 1 has been restored in various ways: ḍe[lec]ṭom ‘delectum’
Olzscha in Radke 1965:136-7, Radke 1965:134), ṃe[re]ṭom ‘meritum’ (Vetter
1953:280), me[lc]tom or me[lq]tom ‘mulctum’ (Vetter in Knobloch 1958:138),
me[le]tom or me[la]tom ‘molitum’ (Joseph & Klein 1981:294). [Pisani (1946:54),
however, connected it with the preceding far as farṃẹ[n]ṭom, a reading which I
reject, mainly on phonological grounds.] For an extensive discussion of all
restorations for this part of EF 1, see §12.2.
92. [2-3]tela in EF 1 has been restored in various ways: preferable, in my view, is
ti]tela (Watkins 1995a:129, based on eco urna tita uendias 479†); possible, too,
are lu]tela ‘lute(ol)a’ (G. Giacomelli 1963:41-3, 1978:525-6) and pa]tela ‘patula’
= ‘larga, ampia’ (Pisani 1946:53, 1964:347-8). For pu]tela ‘parvula’ (Vetter
1953:280), see 62. †*putellius. For an extensive discussion of all restorations for
this part of EF 1, see §12.2.
93. *[3-4]*ad in EF 1 has been restored in various ways, but always as a 3rd sg. pr.
act. subj. (although it may also be a 3rd pl. pr. act. subj., cf. Peruzzi 1964:157).
The only restoration that fits both the size of the lacuna and the traces surrounding
it appears to be p ̣[ore]ḳad ‘porrigat’ (Peruzzi 1964:157). Other proposals have
199
CHAPTER 6
been ạ[dkap]ịad (Herbig CIE 8079), ạ[dtul]ạd (Ribezzo 1927:151-2), ạ[ddou]ịad

(Ribezzo 1936:46), [dou]ịad (Herbig 1913:75), f ̣[in]ḳad (Olzscha 1965:123),
[fe]ṛad (Vetter 1953:280), and [pa]ṛad (Joseph & Klein 1981:294).
94. [---]ostro, reconstructed from [---o]stro LF 244 and [---]ostr[o---] LF 245.
Although the reading is certain, restoration and meaning are unclear. Renzetti
Marra (1990:338) interprets LF 244 as referring to a double burial in one loculus,
with po]stro, but the syncopation required for this is irregular in Faliscan
(§3.6.6.2).102 For the Sabellic languages, *post(e)ros is attested for Umbrian
(postra TI VIb.13) and Oscan (pústreí Cp 33, abbreviated pústr Cp 34).
6.3. The Faliscan lexicon and its Latin and Sabellic equivalents
6.3.1. The Faliscan lexicon in comparison. Comparison of the Faliscan lexical

elements with the corresponding elements in Latin and the Sabellic languages show that
the extant Faliscan lexicon is essentially Latin. The extant Faliscan lexicon is extremely
limited, however: it is not even possible to construct a more or less complete Faliscan
core vocabulary with words like ‘water’/‘fire’, ‘day’/‘night’ (although ‘father’/ ‘mother’
and ‘son’/‘daughter’ are attested). In the comparisons with Latin and the Sabellic
languages, the picture may be influenced by the fact that far less is known of the
Sabellic than of the Latin lexicon, as well as by the different nature of the Sabellic
epigraphic material. Even with these restrictions, however, it is abundantly clear that the
extant Faliscan lexicon is Italic, and that where Latin differs from the Sabellic
languages, Faliscan sides with Latin (e.g. in the use of filius/filia for ‘son’/‘daughter’,
and in the use of the indicative pronoun /ho-/ rather than /eko-/).
As far as lexical borrowing or interference is concerned, there is two instances of
words that have only Sabellic parallels, namely p ̣esco Cap 431 and posticnu MLF/Cap
474*. Since p ̣esco apparently shows a Sabellic development of the internal cluster */rk-
sk/ and posticnu appears to have a Sabellic ending, these words can be regarded as
interferential forms from a Sabellic language: see §9.3.2. Umom EF 2 on the other hand
also has only Sabellic parallels, but shows an assimilation */udmom/ → /ummom/ that
is Latin rather than Sabellic (§3.3.4.3).
Borrowings from Etruscan are apparently limited to clipeus, *gutto/*guttum/
guttur, and possibly of cella, all of which also occur in Latin. This is all the more
remarkable as the Faliscan area must have contained a sizeable number of native
speakers of Etruscan and was presumably in far more frequent contact with Etruscan
than Latium (§9.2.1). It might be objected that this absence of Etruscan influence in the
102
I admit, however, that I can think of no Latin word that fits the text apart from nostro or
uostro, ostro = austro, clostro = claustro, plostro = plaustro, or rostro.
200
THE LEXICON
Faliscan lexicon is a false picture, biased by the limited amount of attested Faliscan
lexical elements, but when the same Faliscan lexicon is compared to the attested
Sabellic lexicon, there are many direct equivalents or close cognates, even though there
are far fewer Sabellic than Etruscan texts. Neither can this conclusion be regarded as
biased by the different nature of the texts, for the typology of the Faliscan texts has
more in common with that of the Etruscan texts than with that of the Sabellic texts.
6.3.2. Lexical subsets. Looking at several subsets of the lexicon reveals a picture that is
in some respects more detailed:
(1) Numerals. Of the numerals from 1-10 the following are attested in some way: ‘two’
in the adverb bis, and possibly in du-; ‘four’ dubiously in Qua...; ‘five’ in the
praenomen Quinctus; ‘six’ in the praenomen Sextus; ‘eight’ dubiously in Oct…; ‘ten’ in
the gloss decematrus (§6.6.2). This entire subset appears to be common Italic.
(2) Family ties. As in all Italic languages, ‘father’ and ‘mother’ are pater and mater:
‘son’ and ‘daughter’, on the other hand, are filius and filia, new words that Faliscan
shared with Latin, while the Sabellic languages continued PIE words, *puclom and
*fu(h)tēr. ‘Wife’ is uxor, which may be attested also from Sabellic, but has no Indo-
European cognates: coniunx, if attested at all, may be due to Latin interference.
(3) Social groups. The only words that belong under this heading are liberta and
libertas. These correspond to Latin words, but it cannot be established whether the
connotation of the words was identical. The word is not a calque on Etruscan lautniθa.
(4) Burial ritual. The word for ‘burial chamber, tomb’ is cella, which occurs also in
Latin, but not in this meaning, and in Etruscan: I assume that cella is Latin rather than
Etruscan. The word for ‘loculus’ or ‘place in the loculus’ is lectus, which also occurs in
Latin.103 The formulaic verb for ‘to lie in a tomb’ is cubo, which is used in this sense
also in Latin and in the Sabellic languages (cf. also §8.10.1). These all appear to be
common words where the specialized sepulchral meaning is secondary. Faliscan also
provides an instance of *lego or *legeo, which has an (indirect) parallel in South
Picene, but must have occurred in Latin as well.
(5) Dedications and ritual. Here the only word that can be considered certain is sacer,
which occurs both in Latin and the Sabellic languages. (The only attestations of longer
texts of this type are the Latin dedications that have been excluded from the
comparison, namely Latin 217-218, 219, 377, and Capenate 421 and 431-438.) P̣esco
Cap 431 is an interferential form.
103
Steinbauer (1999:472, 473) translates σuθi as “Grab(stelle)” and tamera as
“(Grab)kammer”. The Etruscan word corresponding to ‘loculus’ may have been tunu- or tusu-
(cf. Steinbauer 1999:483, 484).
201
CHAPTER 6
(6) Public office. The Middle Faliscan inscriptions only yield aedilis (perhaps a calque
based on Latin), and rex, a ‘general Italic’ word. All other words for public offices or
magistracies, quaestor, praetor, duumvir, and censor, are attested only from Late
Faliscan, Latino-Faliscan, and Latin inscriptions, and reflect the structure of the Roman
administration of Falerii Novi: it may be questioned whether these words were ever
really part of the Faliscan lexicon (§9.4). The same applies to the Latin formula for
functioning as a magistrate, magistratum gero.
6.4. Theonyms
I have included the theonyms in the chapter on the lexicon rather than in the chapter on
the onomasticon, for although they are onomastic elements in the strict sense of the
word and are therefore primarily referential, theonyms refer to individual entities that
are thought of as having an everlasting lifespan, and can therefore not be bestowed upon
newborn members of the group while others die. They are names that refer to specific
individuals that are an enduring part of a group’s cultural heritage, and as such may be
taken over, together with the deity to which they refer, by other groups, even those with
altogether different languages (cf. below on Apollo and Ganymedes), or they may even
be translated (cf. below on Cupido). Note that the list below contains only those deities
that are in some way attested epigraphically: deities mentioned only in the literary
sources are discussed in §2.3.4.
Symbols preceding the lemma: † = rejected by me, but discussed here because of fre-
1. Apollo. Nom. apolo LtF 63 (dedication, although also explained as an abbreviated
human name); gen. apolonos EF 10 (dedication of c.500-475, the first attestation
of the name in an Italic language); [dat. [a]polinei Lat 219 (dedication)]. An early
loan from Greek, the deity is attested all through Central and Southern Italy. The
Faliscan forms, like Etruscan Apulu (first Vc S.1 and OI S.5, both from the first
half of the fifth century, cf. apulu Etr XXXI from Falerii Veteres, if not a falsum),
Latin Apollo (first in the fourth- and third-century inscriptions from Praeneste,
CIL I2.563 and ILLRP 54?), and Marsian apols VM 7 (cf. also aplone VM 6),
show the o-vocalism of Attic-Ionic 'ApÒllwn, while Vestinian apellune MV 10
and Oscan appelluneís Po 14, appelluneí Si 20, [a]ppellounhi Me 1, appell[ou]nhi
Me 2, appellounhi Me 3, and a]p̣p̣ellounhij Me 5 show the e-vocalism of West
Greek 'Apšllwn.104 The difference is due to two separate borrowing processes.
For the worship of Apollo in the ager Faliscus, see §2.3.4.
104
For the distribution of 'ApÒllwn and 'Apšllwn in the Greek dialects, see Buck 1955:46.
202
THE LEXICON
2. Ceres. Nom. ceres EF 1 (context unclear, (sixth (or seventh?) century) is the
earliest attestation of the deity’s name: cf. Le Bonniec 1958:303-4). The
equivalents in the Italic languages show slightly different formations: Latin Ceres
(first in third- or second-century (gen.) cereres CIL I2.973?) was an s-stem
*/keres-/ → /kerer-/ (nom. /kerēs/), while in Oscan keri Cp 37 and Samnitic kerrí
Sa 1.A 3, B 7, it became an e-stem /ker(e)sē-/ (cf. WOU s.v. kerrí). The Sabellic
languages also have derivations such as *Cerrio in Umbrian, Paelignian,
Marrucinian, and Oscan (see WOU s.vv. kerríiúí and cerfum).
3. Cupido. Nom. cupi‹d›o MF 62 (label in mythological scene, the first attestation
(c.380-370) of both the word and the theonym). – Direct Latin equivalent Cupido,
epigraphically attested in cupido ILLRP 1204). The Latin and Faliscan forms are a
Lehnübersetzung of Greek ”Erwj or PÒqoj. In Latin, the noun is feminine, but the
personification is masculine, probably due to its representation in art. (The figure
on the Faliscan vase is also male.) No known Sabellic equivalents or cognates.
4. Diespiter. See 7. Iupiter or Diespiter.
5. Ganymedes. Nom. canumede MF 62 (label in mythological scene, the first
attestation (c.380-370) of the theonym in an Italic language). It is surprising to
find Ganumede(s) = Greek Ganum»dhj here, as in Latin the name encountered is
usually Catamitus (first Pl. Men. 144; for other attestations, see TLL Onomasticon
2.255,20-60), a borrowing from Etruscan Catmite (catmite Ta S.12 and c ̣aθmite
OI S.46). Catmite is often derived from Ganum»dhj by the assumption of several
ad hoc phonetic adaptations (cf. the critical discussion in Bréyer 1993:155-6): it is
easier to derive it from a Greek *katatmhtÒj ‘gelding, catamite’.105
6. † Euius. Based on the reading l[o]ụfir in EF 1 (see 8. †Liber), G. Giacomelli
(first 1963:41-2) and Peruzzi (1964a:166) interpreted euios EF 1 as EÜioj = Liber.
If l[o]ụfir is rejected, there is no ground for this interpretation (which formed the
basis for G. Giacomelli’s (1963:46) euotenosio EF 3 = ‘colui che tiene da Evio’).
7. Iupiter or Diespiter. Nom. [...]s pater MF 62 (label in mythological scene),
variously restored as [die]s pater, [ioui]s pater, or [iouo]s pater. The various
forms of are discussed by Wachter (1987:150-3), who notes that dies pater was
the usual form of the nominative in the fourth century. [Die]s pater has a direct
Latin equivalent in Latin diesptr CIL I2.564, d]iespater CIL I2.568, while a [ioui]s
or [iouo]s (or [iou]s?) would be have equivalents in Latin Iupiter and in the
Sabellic languages in Umbrian iupater TI IIb.24, iuvepatre TI IIa.5, IIb.17, IIb.22,
IIb.26, III.22, and iuve patre TI IIb.7, and Oscan dí{}pa{}tír Po 22.
105
I wish to point out that this derivation, which in my view is the correct one, is not my own,
but that I cannot recollect in which publication I first read it.
203
CHAPTER 6
8. † Liber. The name of Liber has been read (1) by Wissowa (1902:23), who in
interpreted loifiṛtato MF 31 and loifirtato MF 32 as loifir tato ‘Liber pater’, which
was rejected by Herbig (CIE 8011); (2) by Vetter (first 1925:27-8), who read
l[o]ụfir in the damaged first part of EF 1, which both Radke (1965:134-5, 1994)
and my own autopsy have shown to be epigraphically impossible; (3) by AntFal
(p.43), who read leper ‘Liber’ in the unclear first part of EF 4, which is
epigraphically difficult (the text probably reads lepeḳ) and linguistically
impossible.
9. Mercus or Titus Mercus. Dat. titoi | mercui MF 113, tito | mercui MF 114, [t]i
to[i] | ṃercu[i] MF 115, titoi | mercụ[i] MF 116, titoi  mercụ[i] MF 118, [t]itoi 
mercu[i] MF 119, [ti]toi  mercu[i] MF 120, [ti]toi  mercui MF 121, titoi 
ṃ[e]rcui MF 122, [---?]merc[ui] MF 123, [m]ẹrcui MF 124, ṃercui MF 125,
meṛ[cui] MF 126 (all dedications). Mercus (a u-stem /merku-/, connected with the
/merk-/ underlying Latin merx etc.) is undoubtedly connected with Latin
Mercurius (derived from the same */merku-/ as the Faliscan name, cf. DÉ s.v.
merx) and Oscan mirikui Cm 12 (either */merku-/ or */merko-/: the attestation
predates the introduction of the sign ú). Titus is clearly a (cultic) epithet, probably
meaning ‘propitious, well-omened’ or ‘prosperous’ (see also §6.2.82, where
Sabellic cognates are given). For an more extensive discussion of the name, see
§14.1.2 and especially Combet Farnoux 1980:113-69.
10. Menerua/Minerua. Nom. menerua MF 62 (label in a mythological scene, the
first attestation (c.380-370) a non-Etruscan text) [and menrva Etr XXVIII,
menerua LF/Lat 209 (dedication)]. The deity has been thought to have been of
Etruscan origin (see REA s.v. Minerva, DÉ s.v. Minerua, LEW s.v. Minerva), but
may well have been Italic (*/menes-ā/, cf. Rix 1991:117-120, 1998b:209). Apart
from labels in mythological scenes on mirrors (for which see ET), the name
appears in Etruscan only in a few inscriptions from South Etruria (menervas Ve
3.10, meṇ[er]ạvas Ve 3.29, men]ervạ[s Ve 3.33, and men]ervas Cr 4.1). On the
basis of Ovid. Fast. 3.383-4 it has been assumed that worship of Minerva reached
Rome through euocatio of the Faliscan cult of Minerva (cf. §2.3.4), although it is
not necessary to assume that this was the way in which the cult spread (thus
Girard 1989). – Direct Latin equivalents Menerua and Minerua: Wachter
(1987:448) notes that Menerua is both older and more frequent (10 instances
including Lat 209) than Minerua (5 instances including Lat 218, three of which
from the first century).106 Direct Sabellic equivalents in Paelignian minerua Pg 4,
meneruai Pg 8, and Oscan menerẹ(vas) Po 38.
106
Interesting but hardly significant is the fact that both forms are first attested from the ager
Faliscus, Menerua first in CIL I2.454=MF 62, Minerua first in CIL I2.364=Lat 218.
204
THE LEXICON
Of the theonyms thus attested, Ceres, Iupiter or Diespiter, and Mercus have equivalents
or close cognates both in Latin and in the Sabellic languages, while Menerua/Minerua
is probably likewise of Italic origin as well. Greek loans are Apollo and Ganumedes,
while Cupido is a Lehnübersetzung of Greek ”Erwj or PÒqoj: of these, Apollo also oc-
curs in Etruscan and Latin, while the Sabellic languages have *Apello; Ganumedes oc-
curs also in Latin, although contemporary Latin and Etruscan rather used Catamitus and
Catmite respectively. 107 For other deities associated with the ager Faliscus see §2.3.4.
6.5. Toponyms, potamonyms, and ethnonyms
I have treated the category of geographical names and ethnonyms separately, as they
differ in several respects from normal lexical items. Like anthroponyms, their function
is primarily referential, although there is a greater possibility that a geographical name
may consist of or contain lexical elements. Unlike anthroponyms, however, geograph-
ical names are connected to a (unique) geographic feature that exists within a specific
language area and cannot normally be relocated elsewhere. Toponyms therefore often
retain features of the language in which they originated even when the language of that
area is replaced by a different one, or when they are borrowed into another language.
Ethnonyms are another special case: these may originate either within the group they
designate or among ‘outsiders’ that come in contact with this group (§2.2.2), so that
several different ethnonyms may be used for the same group (thus e.g. Tusci and Etrusci
beside Rasenna, if that is indeed an ethnonym). Both geographical names and
ethnonyms may therefore reflect contact with other areas, including language contact
(see §9.1).
In collecting the attestations of geographical name, I have therefore included the
data from Latin and Etruscan inscriptions from the ager Faliscus and Capenas as a
possible source-material. Note also that with the possible exception of the ethnonym
Faliscus the data on geographical names are all derived from occurrences or derivations
in the onomasticon: the same limitations on the use of such forms therefore apply as in
the case of lexemes (cf.§6.1).
Symbols preceding the lemma: † = rejected by me, but discussed here because of fre-
107
The Etruscan labels in mythological scenes on mirrors, gems and vases also name the
Etruscan acaviser Etr XLI, aχavisur Etr X, σ́etlans Etr XLI, tinia Etr XXXI, turan Etr XLI,
turṃṣ Etr XXXI, and uslaṇes or usle*es Etr XLI (cf. Ambrosini 1995, Maras 2002), and the
Greek figures aχle Etr XXXII, atmite Etr XXVII, aivas Etr XXXII, alcestei Etr XXVII,
ariaθa Etr XXVIII, kukne Etr XXV, herkle Etr XXV, hercle Etr XXVIII, vile Etr XXVIII,
mine Etr XXVIII, θevrumines Etr XXVIII, and φerse Etr XL.
205
CHAPTER 6
POTAMONYMS
1. ? °Fa(r)farus. Perhaps attested indirectly in the gentilicium Fafarn- occurring in

abbreviated form as fafarṇ MF 136 and perhaps also in f ̣af ̣[---] MF 139. The gen-
tilicium might be connected to the name of the river Farfarus (G. Giacomelli
1963:191) mentioned by Ovid (Met. 14.328-30). Farfarus is apparently a local
form of the name, since Vergil (A. 7.716) mentions the same river under the Latin
name Fabaris, with the Roman word-internal /b/ corresponding to a word internal
/f/ in non-Roman Latin, Faliscan, and the Sabellic languages (cf. §3.3.3). The
Latin form also has no r in the first syllable,108 so that the Faliscan name Fafarn-,
if connected to the name of the river at all, may represent /fafarn-/ as well as [far-
farn-] with omission of /r.C/ (§3.5.7b): perhaps rather /fafarn-/, as in MF 136 /r.C/
is written out in the second syllable. The Farfarus/Fabaris, originating in the Sa-
bine area near Reate (modern Rieti), flows into the Tiber close to Monte Soratte,
near the border between the agri Faliscus and Capenas.
2. ? °Nar. Perhaps attested indirectly in the gentilicium Narionius in narionio MLF
206 (Schulze 1904:80).
3. °Tiberis. Attested indirectly in the gentilicium Tiberilius in tiperilia LF 229, and
perhaps also in the abbreviated name tif MLF 460 from the ager Capenas. Tiper-
ilia is clearly derived from the Latin form of the name, Tiberius, while tif may be
an abbreviation of its Faliscan equivalent, which would have been *Tiferios.
4. ? °Vomanus. Attested indirectly in the gentilicium Vomanius in uomanio Cap
388, if this is indeed derived from a potamonym Vomanus, as Schulze (1904:481)
suggested: the nearest river of that name known from the sources was part of the
Po estuary. Perhaps Vomanius was derived from another river of the same name
closer to the agri Faliscus and Capenas.
TOPONYMS
5. [°Abella. (Latin attestation only.) Indirectly attested in the gentilicium abelese Lat
251. It cannot be established whether this Abellensis refers to the Campanian
town of Abella (modern Avellino). The adjective derived from this Abella was
Abellanus in Latin and in Oscan (abellanúí Cm 1.A3 etc.).]
6. ? °*Acarcelum (=Ocriculum?). Perhaps attested indirectly in the gentilicium
acarcelini LF 221, acarcelinio LF 223, acarcelinio LF 226. Peruzzi (1963b:441-
6) suggested that this name was derived from a toponym; A. Mancini (1981)
plausibly derived it from an */akarkelom/ that would be equivalent to (but not
108
The scansion of Vergil’s words, qui Tiberim Făbarimque bibunt, shows that Fabarim can-
not be a copist’s error for Farbarim.
206
THE LEXICON
necessarily identical with) Latin Ocriculum and Umbrian *Ukriçlum */okriʃlom/

m */okrikelom/ implied by the Etruscan gentilicium ucrislane Cl 1.2609,
2611-2613 etc. The Ocriculum that is known from the ancient sources is located
on the east side of the Tiber close to the Tiber crossing near the Grotta Porciosa
site (to which it was connected, from c.220 onwards, by the Via Flaminia), and
thus the nearest town beyond the borders of the ager Faliscus in this direction.
7. ? °*Cali- or °*Cale-. The name calitenes MF 265 may be derived from an Etrus-
can toponymic adjective *Calite, in turn derived from the name of an otherwise
unknown town Cali- or Cale-. Cale has in fact been proposed as the original
name of modern Gallese (Cifani 2002:33) in the northern ager Faliscus.
8. ? °Capena. Perhaps attested indirectly in the names ḳapena EF 4 (reading and
context uncertain), and kape Cap 403 (or is this to be read as k ape?).
9. ? °*Feliginum (=Fulginium/Fulginiae). Indirectly attested in the gentilicium
felic ̣inate MF 42 (gen. sg.) and [fel]ịcinatiu MLF 384 (gen. pl.). *Feliginas, oc-
curing also as an Etruscan gentilicium (felcinatial Pe 1.485, 1.1235, and felcinat-
nal Cl 1.2673) is derived from a toponym *Feliginum (*Felginum Rix 1965:233
n.133), an older form of the Fulginium that occurs as an alternative name of the
Umbrian town Fulginiae (modern Foligno) on the Via Flaminia.
10. ? °Fescennium. Perhaps attested indirectly in the gentilicium hescuna MLF 346:
Colonna (1990) suggests that this name may be derived from the Faliscan
toponym Fescennium (with the Faliscan change of /#fV/ o /#hV/, §3.5.2). For
the sources on Fescennium, perhaps the name of the site at Narce, see §2.1.2.
11. ? °*Ortica or °*Orticum. The gentilicium ortecese MLF 339 (probably connected
with the Etruscan gentilicium urtcsnas Etr XXXV) may be read as Orticensis, a
name that appears to be derived from an otherwise unknown toponym *Ortica or
*Orticum.109 This (rather than Colonna’s (1990:118) Χurcle) may be related to
the ancient name of modern Corchiano (older Orchiano), the earliest recorded
form of which is Orclanum,110 perhaps a contraction of an older *Orticulanum.
12. ? °?Veii. Very uncertain. The name of the town has been read in furc  t  p  c  ef
 i  uei  LtF 205. Even if this uei is connected to the name of the town, it is per-
haps rather an abbreviation of a (related) gentilicium like Veianius, attested in CIL
XI.3805 from Veii and in Varro (“fratres Veianii ex agro Falisco” R 3.16.10).
109
Or, perhaps, *Hortica or *Horticum, but neither Faliscan ortecese nor Etruscan urtcsnas
has h-, and there are no certain attestations of omission of h- in Faliscan (§3.5.2).
110
“ORCHIANUM “forte Orchia Castellum, quod tamen in antiquis donationibus Monasterii
Farfensis Orclanum dicebatur” ” (De Italiae medii aevi dissertatio chorographica, c.CCXX,
in Muratorius 1726 t. X): see also Gamurrini (1894b:146-7) on Corchiano and Vitorchiano.
207
CHAPTER 6
ETHNONYMS
13. Faliscus (see §2.2.2). The ethnonym is attested in falesce quei in sardinia
sunt Lat 218, a rare example of an Italic ethnonym being used by the members of
the group themselves. Feluskeś in [mi a]ụṿileś feluskeś tuśnutaḷ[a pa]|panalaś
Vn 1.1 may be an Etruscan rendering of *Falescos (Poccetti 1997). The same
may be true of fẹḷẹṣkẹnas in [m]i larisa fẹḷẹṣkẹnas aṃ**[?---] AS 1.40 (perhaps
a patronymic gentilicium ‘Faliscanson’, cf. creicnal Ar 1.4 ‘Greekson’ from
creice = Graecus): see §2.2.2 and §2.4.2. (The corresponding Etruscan ethnonym
may have been *Falsaχ or *Felsaχ.)
14. °Graecus. Attested in the praenomen kreco MF 150 (with k representing /g/, cf.
§11.2.4) and perhaps also in the gentilicium Grae... (or Crae...?) in cra[---] MF
MF 144, cre[---] MF 145, cṛ[---] MF 146 and in the abbreviated gentilicium cr
MF? 33. The ethnonym is also found in Etruscan names, mainly from Clusium
(creice Cl 1.1280, 1510, 1511, 2466, 2467, crei[ce] Cl 1.1512, cr[ei]c[e Cl
1.1514, creices Cl 1.1669, creicia Cl 1.352, 1513, cr]eicia Cl 1.794, c]reicia Cl
1.1515, creiceσ́a Cl 1.1281, 1282, 1302, 1686, 1744) and Tarquinii (creice Ta
6.15, creices Ta 1.17, creic[ia]l Ta 1.217), but also elsewhere (creici Cr 1.149, Vc
1.5, craica Vc 2.34, creice Pe 1.889, kraikaluś Fe 2.7): cf. Rix 1965:231. The
ethnonym occurs in Sabellic texts only in graex Pg 40, where it is a cognomen.
15. ? °Hirpi and Hirpini. Indirectly attested in the gentilicium írpios Cap 389, which
is probably connected to the Hirpi Sorani mentioned in the Latin sources (most
notably Serv. in Verg. A. 11.785 and Plin. NH 7.2.19). These were sometimes con-
fused with the Hirpini (e.g. by Servius): see §2.3.4 for further sources. The name
may well go back to the Sabellic word hirpus ‘wolf’, see §6.6.5.
16. °Latinus. Attested indirectly in the gentilicium latinaio MF 135. The ethnonym is
attested also from Etruscan inscriptions, most notably in mi tites̀ latines̀ Ve 2.4
(c.600) and mi latines Cm 2.57 (fifth century?), and has numerous derivations in
the onomasticon (the indices of ET name 75 attestations of latin-, 69 of which are
from Clusium). There appears to be no attestation of the ethnonym in Sabellic
texts.
17. ? °Sabinus. Probably indirectly attested in the abbreviated gentilicium sab Cap
400. Note that this is the Latin form with b, not the Faliscan form, which would
have been *Safin- (§3.3.3) or possibly *Safen- (from */saβeno-/, see Rix 1957).
The Sabellic inscriptions yield both the ethnonym, in South Picene safinús TE.5,
safinúm TE.6, safinas TE.5 and safina[ TE.7, and the name of the region in Pael-
ignian and Samnitic safinim nPg 2, Sa 4. (Cf. Dench 1997:48-9). There are no
Etruscan attestations of this ethnonym.
208
THE LEXICON
18. °Umber and °Umbricus.111 Attested indirectly in the gentilicia Umbrius in umrie
Etr XLIII, and its derivations Umbricius in u]mpricius Lat 219, and Umbricianus
in upreciano MF 363 and MF 364. (Note that Etr XLIII, MF 363, and MF 364
are all from the same tomb.) The ethnonym is attested from the Etruscan onomas-
ticon in several forms, including Umr- (umres AH 1.74, umria Cl 1.2620, 1.2621,
umriaś Cl 1.1294, 1.1913) and Umrc- (umrceś AS 1.129, umrcial AS 1.395), cor-
responding to the Umbrius and Umbricius/Umbricianus of the ager Faliscus (cf.
also Rix 1965:321). The ethnonym is attested from the Sabellic languages only in
South Picene ombriíen CH.2: there is no mention from the Umbrians themselves
(cf. Bradley 1997:56).
Although with the exception of Faliscus all the potamonyms, toponyms, and ethnonyms
are attested at best indirectly in the onomasticon, and often very dubiously, too, these
draw a picture that encompasses all the surrounding areas and peoples. The Faliscans,
Latins, Sabines, and Umbrians all make their appearance, as do the Greeks. Among the
towns that may be attested are both those of the area itself (Fescennium, and perhaps
Gallese and Capena), and those on the route of the Via Flaminia (Ocriculum and
Fulginium), and there is reference to the rivers Tiber and perhaps Farfarus/Fabaris.
6.6. Faliscan glosses
Only a few glosses are ascribed by the ancient authors explicitly to the Faliscans or
pertain to matters Faliscan, probably because the glossographers referred not so much to
Faliscan as to the local Latin of the second century. The Faliscan glosses were first
listed by Mommsen (1850:364): in later literature, they were increasingly ignored, apart
from their occurrence in the lists by Deecke (1888:230-42), Conway (1897:324, 384),
and Vetter (1953:362-78).
Glosses present several specific problems. Consisting of second-hand data whose
primary source is unknown, their reliability is questionable, and their attribution to a
specific language or dialect is at best unverifiable. This is particularly so in the case of
Latin dialect glosses, not only because these are more difficult to recognize than glosses
from other languages, but also because of the tendency to ascribe the differences from
‘standard’ urban Latin to unspecified antiqui or rustici (see also §3.5.2). A further
problem is that it is usually impossible to establish to what stage of a language or dialect
the author is referring: in the case of the Faliscan glosses, this is probably the ‘rustic
Latin’ of Roman Falerii.
111
Although in Latin the gentilicium is Umber, the Etruscan, Faliscan and Latin onomasticon
also shows an Umbricius that may be derived from an alternative ethnonym *Umbricus (cf.
the Greek ”Ombroj and 'OmbrikÒj).
209
CHAPTER 6
1. cenaculum. “ubi cenabant, cenaculum uocitabant, ut etiam nunc Lanuui apud

aedem Iunonis et in cetero Latio ac Faleris et Cordubae dicuntur” (Var. L 5.162). –
Cenaculum is dialectal only in its meaning: the word existed in Roman Latin, but
there meant ‘top-storey, garret, attic’ (OLD), the word for ‘dining room’ being
cenatio. The form of the word, probably from */kert(e)snā-/ (cf. WOU s.v.
kerssnaís) rather than from */kersinā-/ (Schrijver 1991:432), corresponds to that
of Latin rather than that of the Sabellic languages, which appear in Oscan kersnu
Cm 14, kerssnaís Cp 31 and Umbrian śesna TI Vb.9 etc. The suffix -culum shows
an anaptyxis that is not attested in the Faliscan inscriptions: its presence here is
probably due to the fact that in written Latin the anaptyctic form was the normal
one.
2. decimatrus. “quinquatrus appellari quidam putant a numero dierum, qui feriis his
[Scaliger: fere his MSS, Lindsay] celebrantur. quod scilicet errant tam hercule
quam qui triduo Saturnalia et totidem diebus Competalia. nam omnibus his
singulis diebus fiunt sacra. forma autem uocabuli eius exemplo multorum
populorum Italicorum enuntiata est, quod post diem quintum iduum est is dies
festus, ut apud Tusculanos triatrus et sexatrus et septematrus et Faliscos
decimatrus.” (Fest. 304.33-306.2L); “quinquatrus festiuus dies dictus, quod post
diem quintum iduum celebraretur, ut triatrus et sexatrus et septematrus et
decimatrus” (Paul. Fest. 305.10-2L).112 – MS E of Paulus Diaconus has decema-
trus. If this is correct, the form could be regarded as an instance of the presumed
Faliscan lack of weakening of short vowels in medial syllables (§3.6.6), but
decematrus may just as well have been remodelled on septematrus, or either or
both may be due to reanalysis after decem and septem (by anyone from the
original source of the word to the copist of E).
3. haba. “quem antiqui fircum, nos hircum, et quam Falisci habam, nos fabam
appellamus, et quem antiqui fariolum, nos hariolum” (Ter. Sc. CGL 7.13.8-9). –
The attribution of haba to Faliscan is not implausible in view of the Middle
Faliscan development /#fV/ o /#hV/, but cf. §3.5.2. Note that haba for faba is
also ascribed to the antiqui, both by Terentius Scaurus himself (CGL 7.13.8) and
by Velius Longus (CGL 7.69.10), echoing the passage quoted here.
4. Halaesus. “uenerat Atridae fatis agitatus Halaesus, a quo se dictam terra Falisca
putat” (Ov. Fast. 4.73-4); “Faliscos Halaesus condidit: hi autem immutato h in f
Falisci dicti sunt, sicut febris dicitur quae ante hebris dicebatur, Formiae quae
112
These words are often regarded as a compounds with ater ‘so-and-so-many days of the
dark = waning moon’ (DÉ s.vv. ater, Quinquatrus, LEW s.v. quinquatrus); the element -atrus
was in fact considered meaningless by Gellius (2.21.7). A discussion of the etymologies and
their possible relation to Etruscan is given by Bréyer (1993:465-6).
210
THE LEXICON
Hormiae fuerunt, ¢pÕ tÁj ÐrmÁj: nam posteritas in multis nominibus f pro h
posuit” (Serv. in Verg. A. 7.695). (For other sources on Halaesus, see §2.4.1; for
other grammarians on the variation of f- and h-, see §3.5.2.) – The derivation of
Faliscus from Halaesus presupposes that the Middle Faliscan change /#fV/ o
/#hV/ (§3.5.2) was well under way; Servius’ remark apparently shows that the
hypercorrect use of f for an /#hV/ that did not reflect an original /#fV/ was also
known. The oldest source for an Argive origin of Falerii is Cato (Plin. NH 3.51),
but it is unknown whether he already made the link with Halaesus; the first author
who certainly makes this connection is Ovid (both l.c. and Am. 3.13.31-5).
5. Hirpi. “Soractis mons est Hirpinorum in Flaminia conlocatus. in hoc autem monte
cum aliquando Diti patri sacrum persolueretur (nam diis manibus consecratus est)
subito uenientes lupi exta de igni rapuerunt. quos cum diu pastores sequerentur,
delati sunt ad quandam speluncam, halitum ex se pestiferum emittentem, adeo ut
iuxta stantes necaret: et exinde est orta pestilentia, quia fuerant lupos secuti. de
qua responsum est, posse eam sedari, si lupos imitarentur, id est rapto uiuerent.
quod postquam factum est, dicti sunt ipsi populi Hirpi Sorani: nam lupi
Sabinorum lingua uocantur hirpi. Sorani uero a Dite: nam Ditis pater Soranus
uocatur: quasi lupi Ditis patris” (Serv. in Verg. A. 11.785). (For other sources on
the Hirpi Sorani, see §2.3.4). – Hirpus is referred to as Sabine (Servius) as well as
Samnitic (Paul. Fest. 93.25-6L, Str. 5.4.12). The word is usually connected to
Latin horridus, hirsutus etc., and originally meant ‘shaggy animal’ (apparently
from a Proto-Italic */ǵher-ko-/: horridus etc. are usually derived from */ǵhers-/,
however): see EDL s.v. hircus. In the passage from Paulus Diaconus the word is
given as irpus, with a dialectal omission of h-. This h-less form is attested in the
gentilicium írpios Cap 389, although the omission of h- seems to have been non-
Faliscan (§3.5.2).
6. Struppearia. “stroppus est, ut Ateius Philologus existimat, quod Graece strÒfion
uocatur et quod sacerdotes pro insigni habent in capite. quidam coronam esse
dicunt, aut quod pro corona insigne in caput inponatur; itaque apud Faliscos diem
[Augustinus: idem MSS, Lindsay] festum esse qui uocetur Struppearia, quia
coronati ambulent; et a Tusculanis quod in puluinari inponatur Castoris struppum
uocari” (Fest. 410.6-9L). – Ateius’ derivation of the word from Greek strofe‹on or
strÒfion ‘headband worn by priests etc.’ (LSJ) seems to be correct. The Latin
spelling with u and pp may be due either to an Etruscan intermediary (Bréyer
1993:227-8, cf. §6.2.31 on *gutto/*guttum/guttur) or to Latin developments (cf.
Pfister 1977:155-6, Allen 1978:49 n.2). It seems likely that the word was quoted
simply because the festival was unknown in Rome or in Latium, but there may
have been a different reason, as appears from a comparison of the few instances
of struppus and stroppus in Latin:
211
CHAPTER 6
“struppi uocantur in puluinaribus ¢fasciculi de uerbenis facti, qui pro de²orum

capitibus ponuntur” (Fest. 472.15-6L, restored from Paul. Fest. 473.4-5L)
“tumque remos iussit religare struppis” (Andr. 9L)
“ubi id audiuit, lecticam iussit deponi, struppis, quibus lectica deligata erat, usque
adeo uerberari iussit, dum animam efflauit” (Gracch. ORF 49 = Gel. 10.3.5)
“tenuioribus [coronis] utebantur antiqui, stroppos appellantes: unde nata
strophiola” (Plin. NH 21.2.3)
In the last instance, several MSS have struppos, a reading that cannot be rejected
out of hand, since struppos may easily have been corrupted into stroppos under
the influence of strophiola, while it is difficult to see how or why stroppos could
have been corrupted into struppos. If stroppos is correct, it appears that there is a
difference in meaning between struppus ‘a (plaited) leather strap’113 (Andronicus,
Gellius) or ‘a bundle of twigs or herbs used in lectisternia’ (Festus) on the one
hand and stroppus ‘a headband or ribbon worn on festive occasions’ (Pliny,
Festus) on the other. If the two forms were distributed in this way, the original
reason for quoting Faliscan Struppearia may have been that it implied a use of
struppus as ‘headband’, which in Roman Latin would have been stroppus.
To sum up, it can be said that decimatrus, Hirpi, and Struppearia are quoted as words
for local Faliscan institutions, cenaculum, and perhaps Struppearia, to illustrate a
difference of meaning between the Faliscan and the Roman word, and haba and
Halaesus to illustrate the Faliscan realisation of /#fV/ as [h], which was ascribed also to
have existed also in ‘Old Latin’ and ‘Sabine’ (used in the sense of ‘marginal Latin’): for
such attributions, see §3.5.2. Interestingly, in several instances forms labelled as
Faliscan are quoted side by side with forms from other Latin dialects: with Lanuvian
and Corduban (and indeed ‘Latian’) in the case of cenaculum, with Tusculan in the case
of decimatrus and Struppearia, and with ‘Old Latin’ in the case of haba.114 The
exception is Sabellic Hirpi, which is probably a Transtiberine import.
113
OLD translates ‘a twisted cord’, but ‘a (plaited) leather strap’ is more appropriate: Andro-
nicus’ struppis translates tropo‹j ™nˆ dermat…noisi δ 782 = θ 53 (Erasmi 1975:82-3), and in the
passage from Gellius, the struppi are the shoulder-straps attached to the carrying-poles.
114
Similarly manus, attested in Faliscan in the cognomen Manumus (see §6.2.44) is ascribed
both to the antiqui (Var. L 6.2.4) and to the Lanuvians (Macr. 1.3.13).
212
Chapter 7
The onomasticon
As a large part of the Faliscan material is onomastic, no study of Faliscan can be complete
without an assessment of the Faliscan onomasticon: it has in fact been the subject of a sepa-
rate study by Hirata (1967). This chapter opens with a short discussion of the problems in-
herent in the use of the onomasticon as a subject of linguistic study (§7.1). The next section
treats the names in the Early Faliscan inscriptions (§7.2). This is followed by sections on the
onomastic formulas of men (§7.3) and of women (§7.4), the formulas of filiation (§7.5), and
the onomastic formulas of freedmen and -women (§7.6). The next sections treat the attested
praenomina (§7.7), gentilicia (§7.8), and cognomina (§7.9). The chapter concludes with a
discussion of the onomasticon from the perspective of ethnic identity (§7.10).
7.1. The onomasticon: methodological issues
7.1.1. Names and language. In using onomastic data as the basis of linguistic study,
several specific problems present themselves, and these are all the more important in a
study of material like the Faliscan inscriptions, where the onomastic data far exceed
the lexical ones. Onomastic elements are associated with a certain language rather
than a part of it, for although they adhere to the morphology and phonology of the
language in which they originate, they are not a part of its lexicon, in the sense that
they do not necessarily have a meaning apart from their reference to a specific person.
This is true even if a name consists of a lexeme, as in the case of a nickname such as
Plautus or a name of good omen such as Scaeua. Using these words as names changes
their point of reference: they no longer refer to ‘flat feet’ or ‘good luck’ themselves,
but to a specific person presumably possessing these. Such ‘telling names’ may have
an added value because they can be ‘understood’, but they are still names, not words.
For this reason, onomastic elements can move between language communities
with far greater ease than lexical or morphological elements (cf. §1.3.2.2). Names are
the constant companions of the people they refer to, even if a person moves between
language communities, both in the sense that that person migrates between areas
where different languages are spoken and in the sense that that person is bi- or multi-
lingual and changes between the frameworks of the different languages at his or her
disposal. The Faliscan material is very illustrative in this respect: whereas the main
conclusion of the preceding chapter was that the extant Faliscan lexicon contains few
or no Etruscan elements, in the onomasticon Etruscan names abound (§7.7.2, §7.8.2).
213
CHAPTER 7
A name is, in a sense, the ultimate vehicle of the whole range of a person’s iden-
tities. Apart from ‘just’ referring to a person, names, in countless unspoken ways, ex-
press a person’s ties to his or her gender, family, ethnic group, and religion, and are
therefore a kind of condensation of all that person’s identities. As such, a name is a
definition not only of who the person is, but also of what he or she is, and stays with
that person even, or perhaps even more, among strangers, whether these speak the
same language or a different one. Names are carriers, not of meaning, but of associa-
tions of personal and cultural significance: a great difference from lexical elements.
This of course does in no way preclude that a person’s name may be adapted in
various ways if it is used within the framework of a language different from the one it
originated in. For instance, the phonological form may be adapted in order not to
sound too ‘foreign’, especially if the name contains phonemes or phonotactics that are
alien to the language to which it is adapted. An example of this is the way in which
the Sabine chieftain Attus Clausus adapted his gentilicium to Claudius (Liv. 2.16).
The morphological form of a name may have to be adapted, if it is to be declined ac-
cording to the morphology of a different language: such adaptations resulted in the
Latin versions Arruns and Lars of the Etruscan names Arnθ and Larθ. Names may
have their derivational suffixes altered so that they resemble names of a different lan-
guage: thus, the Faliscan onomasticon has Succonius beside Zuχu (§7.8.2). Names that
have recognisable lexical elements may even be translated, so that Etruscan Ziχu be-
came Scribonius (Cl 1.318-320), or they may be replaced by similar-sounding ones, or
ones regarded (rightly or not) as etymological equivalents, so that in the ager Faliscus
the very frequent Gauius was ousted by Gaius (§7.7.1.24-25).
Such adaptations always show a desire to adapt and fit in, whether the choice to
do so is made willingly, hoping perhaps for a better acceptance or better chances
within a different community, or under some form of pressure, where people bearing
names from a specific ethnic or social background are discriminated against, or where
a new administration sets new rules as to the use of names. Similarly, being able to
preserve the old name unadapted in a new environment may also speak volumes both
about the person able to do so and about the environment in which this can occur.
How do these preliminary remarks apply to the study of the Faliscan onomasti-
con? First, the onomasticon is not so much of linguistic as of socio-linguistic interest.
This has already been pointed out in the preceding chapter: lexical elements can only
be derived from the onomasticon in exceptional circumstances. Even when a name
contains lexical elements, it can never be assumed as a matter of fact that these ele-
ments where also present in the lexicon, for the name may have originated in a differ-
ent area and contained lexical elements that in the ager Faliscus may have had differ-
ent associations or meanings, had become obsolete, or simply did not exist. On the
other hand, the way in which people choose, use, adapt or preserve their names may
be of considerable sociolinguistic interest.
214
THE ONOMASTICON
Second, gentilicia are often used to pinpoint the ‘roots’ of a family, as is done
e.g. by Peruzzi (1990:283-9) for the Etruscan gentilicia from Corchiano. This can of
course be done to some extent, and it can be of great use in socio-historical studies if
there is abundant additional historical material to substruct such links. Yet a simple
similarity of a name means close to nothing in the case of a socio-linguistic study. The
fact that a person at Falerii Veteres had an Etruscan gentilicium that occurs also or
even exclusively at Perusia is in itself not very significant, not only because the fami-
lies may still be entirely unrelated, but, more importantly, because it is impossible to
make any valid inferences about that person’s personal ethnic or linguistic back-
ground that are relevant to the interpretation of the text they occur in.
This is connected with a third and more serious fallacy, namely the assumption
that a person with an Etruscan or Sabellic or Latin name ipso facto had that specific
cultural background, identified with that specific cultural background, or was a
speaker of the language associated with that background. It is a dangerous kind of
simplification to equal names (even though they demonstrably originated within a
specific culture or language) with specific cultures, peoples, or languages, and treat
these as if they were in a one-to-one relation. Recent studies on ethnic identity have
shown that distinctions were certainly not so clear-cut (cf. e.g. Cornell 1997 on ethnic
identity in early Rome). I shall return to this question in chapter 9.
7.1.2. The problem of abbreviations. Some very simple problems in the material are
caused by abbreviations. In many inscriptions, names are abbreviated, and many Be-
sitzerinschriften consist entirely of abbreviated names. Although editors have gener-
ally ignored this (not small) number of inscriptions, they do contain data that might be
used for onomastic research. I have made use of this material as follows:
x Inscriptions consisting entirely of abbreviations are assumed to contain one name
even if they consist of consonant clusters such as mr or cs (which can be abbreviations
of Marcus or Caesius), unless the letters are separated by an interpunct.
x Inscriptions consisting of one abbreviated name have not been used, as it cannot be
established (a) whether the name is a praenomen or a gentilicium; (b) whether the
name is male or female; (c) what name is represented by the abbreviation.
x Inscriptions consisting of two abbreviations are assumed to consist of a praenomen
and a gentilicium: these have been used in the lists of praenomina and gentilicia
(§7.7.1, §7.8.1), but not in the discussion of onomastic formulas (§7.3-6), since it can-
not be established whether the name is male or female.
x Abbreviations in longer inscriptions are usually praenomina and have been used
both in §7.3-4 and §7.6. The abbreviations in FILIATION present a problem of their
own, for, unless these are followed by SON/DAUGHTER, abbreviations of patronymic
adjectives and of the father’s praenomen cannot be distinguised: see §7.5.
215
CHAPTER 7
7.2. Names in the Early Faliscan inscriptions
7.2.1. The Early Faliscan onomastic formulas. Considering the date of the Early
Faliscan inscriptions, the names occurring in these texts may be expected to be
praenomina and Individualnamen, and early instances of gentilicia.
Gentilicia are first found in the area in the Etruscan inscriptions: this may be due
to cultural factors, or it may simply be due to the fact that there are more Etruscan than
Faliscan inscriptions from the sixth and fifth centuries. The earliest instances may be
lar*s ruvries Etr XIX from Mazzano Romano (c.650-625), and leθaie Etr XLVIII from
Mazzano Romano (c.570-560). The first certain instances of gentilicia in Etruscan in-
scriptions from the ager Faliscus are from the second half of the sixth century, velθarus
velanas Etr XVI from Narce (c.550-500) and larisa zuχus Etr XXXII from Corchiano
(c.525-500). Gentilicia then appear regularly from the fifth century onwards in Etruscan
inscriptions: the first instance in a non-Etruscan inscription from the area occurs in the
Sabellic inscription pa‹qu›is blaisiís 468* of unknown provenance.
The single names that occur in the Early Faliscan inscriptions are the following:
eco quto *e uotenosio (= uo(l)tenosio or uo‹l›tenosio) titias duenom duenas
salue[to]d uoltene  EF 3
praụ[i]os urnam  soc ̣[iai] p ̣orded karai  EF 1
eko lartos EF 6
eko kaisiosio EF 7
tele*[1-2?] med fifiked EF 9
These show several names that occur also in the later inscriptions, namely the
praenomina Aemius, Caesius, Lars, and Titia. Voltenus appears to be connected to the
later Faliscan praenomen Volta (see §7.7.1.85), and might be considered a gentilicium if
the inscription is not too early for this. Tele... is perhaps a Greek name in Thle- or
Tele-. Praụ[i]os may be connected to Latin prauus, and be a nickname or a play on the
‘good’ or ‘beautiful’ theme of many of the early Faliscan, Latin, and Etruscan inscrip-
tions. (All these names, as well as those given below, are discussed in §7.2.2.)
More difficult to asses are the strings of names that occur in EF 1 and EF 4. Thus,
apart from the praụ[i]os already mentioned, EF 1 has:
ceres  far *[0-2]e[1-3]tom  *[3-5]uf[1-4]ui[..]m  *[3-4]*ad euios  mama
z[e]xtos med f[.f]ịqod  EF 1
Here, euios  mama z[e]xtos has been variously interpreted. The early interpretation as
PRAENOMEN GENTILICIUM COGNOMEN (Herbig CIE 8079 (‘Sextus Mama Euius’),
Ribezzo 1918:56 etc., Vetter 1939a:155) can be ruled out: even apart from the difficul-
ties encountered in interpreting the verb f[.f]ịqod as third person singular, it would be
216
THE ONOMASTICON
quite early for a gentilicium, and far too early for a cognomen (see §7.9). Most authors
have interpreted mama z[e]xtos as praenomina instead. Mama is probably of Etruscan
origin, comparable to the Latin praenomen Mamus (G. Giacomelli 1963:202, Salomies
1987:75), while z[e]xtos is quite clearly the numeral praenomen Sextus, which occurs
also in later inscriptions, albeit rarely (§7.7.1.62). If and how euios is connected to these
names is rather more difficult. It has been taken as a gentilicium euios to go with the
praenomina mama and z[e]xtos, which might not be impossible if euios is regarded as
singular (Meister 1916:101): interpreting it as a plural (Norden 1939:206-7, Vetter
1953:280) is difficult or impossible (cf. §4.3.6). Another possibility is that it is a third
praenomen (cf. perhaps Praesamnitic eṿies Ps 4), although in that case it is unclear why
it should be separated from mama z[e]xtos by an interpunct. G. Giacomelli (1963:41-2)
interpreted euios as a theonym to go with the fragmentary words that precede it, and
although this interpretation is based on Vetter’s untenable reading l[o]ụfir and can only
be maintained with great difficulty, it is possible that euios should be taken with the
preceding words rather than with mama z[e]xtos.
Even more difficult is the reading and interpretation of EF 4, where there appear
to be two groups of names, one of women’s names and one of men’s:
e**azieputilepeḳapena (e**azieputilepe ḳapena?) rufia ḳalẹptia ues saluete sociai
ofetios kaios uelos amanos salueto salues seiteiofeteqemeneseseie EF 4
Of the women’s names, rufia can hardly be interpreted as anything else than as a
woman’s name (cf. ruvries in Etr XIX), which leaves no other possibility than the
same interpretation for ḳalẹptia (even if individual letters of this part of this name have
been read differently). A third name ḳapena has often been read in the unclear first
part of the text, e**azieputilepeḳapena, and this would appear to be a gentilicium rather
than a praenomen (cf. Prosdocimi 1990:302-5). Of these names, rufia ḳalẹptia looks
like a praenomen followed by a gentilicium, but I wonder whether it is not far too
early to assume this formula: in view of the following ues saluete sociai, these may be
separate names belonging to two different women. The men’s names are likewise dif-
ficult. Ofetios sounds very much like Italic names such as Opetius or Ufentius, but
cannot be directly connected to either: see §7.2.2.9. Kaios is clearly the Latin
praenomen Gaius, but this name does not appear elsewhere in the Faliscan onomasti-
con: see §7.7.1.24); uelos is in all probability the Etruscan praenomen Vel (§7.7.1.80),
but whether it is a genitive (Pisani 1937:238-9, cf. §4.5.2) or a thematized nominative
*Velus is unclear. Amanos had tentatively been linked by G. Giacomelli (1963:173-4)
to the Etruscan gentilicium Amana (as a thematized nominative?); Salomies (1987:99)
also pointed to Amanus and Amanius in Latin. How these names are to be strung to-
gether is unclear. Vetter (1925:26, 1953:284) took ofetios kaios uelos amanos as a gen-
tilicium ofetios followed by three praenomina (‘the Ufentii, Gaius, Vel, [and] Amanus’),
just like he had interpreted euios  mama z[e]xtos in EF 1 (see above). If uelos is a geni-
tive, this could of course be a genitive of the father’s name (kaios uelos = ‘Gaius [son]
217
CHAPTER 7
of Vel’ or ofetios kaios uelos = ‘Ofetius [and] Gaius [sons] of Vel’). This could be
another argument against taking amanos as the gentilicium, for placing the filiation
after the praenomen and before the gentilicium is in accordance with the later
Umbrian and Volscian custom, but not with that of the Middle and Late Faliscan
texts: see §7.5. (The place of the filiation may not yet have been fixed, especially at a
time when gentilicia were just making their first appearance.) The only alternative
seems to be to leave amanos as an isolated name without filiation.
7.2.2. The names attested in the Early Faliscan inscriptions. The names that occur in
the Early Faliscan inscriptions are the following. For ease of discussion I have classed
the names under their closest Latin equivalent or approximation, and have arranged the
lemmata according to the order of our own alphabet.
1. Aemius. praen. m. aịṃiosio EF 467* (gen.). – A patronymic derivation of Aemus,
attested for Middle Faliscan in eiṃoi MF 293 and possibly in the abbreviation
ạ[i?]ṃ MF 89 (either a praenomen or a patronymic adjective), that can represent
either Aemus or Aemius. This name does not occur elsewhere in Italy except in
Venetic ·‹a›·i·mo·i· Le 26: see also §7.7.1.3.
2. Amanus. m. amanos EF 3. – Either a gentilicium (G. Giacomelli 1963:173) or a
praenomen (Hirata 1967:33-4): see §7.2.1. Salomies (1987:99) points to the
Amanus in Sil. 17.44 and the Latin gentilicium Amanius. The name could be an
adaptation of the Etruscan gentilicium Amana (amanas Vs 1.92, also ] amanas̽[
Ve 3.4?): see §7.8.2.
3. Caesius. praen. m. kaisiosio EF 7 (gen.). – The name occurs also in the later peri-
ods: see §7.7.1.18.
4. ? Caleptius. gent. (?) f. ḳalẹptia EF 4. – However these letters are read (kalketia
G. Giacomelli 1963:180, kalẹstia Hirata 1967:40-1, kalịptia Prosdocimi
1990:295, etc.), it can hardly be anything other than a woman’s name (cf. §7.2.1),
albeit one that is without parallels in the Etruscan, Latin or Sabellic onomasticon.
The same is true of the other readings that have been proposed.
5. ? Capena or Capenus. f. ḳapena EF 4. – The name looks like an Etruscan gen-
tilicium, or it might be the feminine of a gentilicium Capenus adapted from an
Etruscan *Capena (cf. §7.8.2). Etruscan had Capn- (capna AT 1.200, Cl 1.454,
Pe 1.869, 1.973, capnal Cl 1.202, 1.578, capna[l] Cl 1.203, 1.633, capnas Pe
1.975, capnaś Cl 1.2214; capnei Cl 1.201; capni Pe 1.436), and Capan- (capanei
Pe 1.213, 214). The Capenus Sequanus in [Liv.] Per. 120 appears to be a Gaul. If
and how the name is connected to the toponym Capena is unclear.
6. ? Euius. m. euios EF 1. – G. Giacomelli’s (1963:41-2) interpretation of euios as a
theonym EÜioj = ‘Liber’ was largely based on Vetter’s untenable reading (first
218
THE ONOMASTICON
1925:27-8) l[o]ụf[ir in the same text. If euios is an anthroponym (see §7.2.1),

possible parallels are Praesamnitic eṿies Ps 5 (also read as ef ̣ies), and perhaps in
the Latin gentilicium Euius (Hirata 1967:49). Salomies (1987:83) compares Euius
to the Oscan praenomen Ovis (úvis Cm 35, elsewhere abbreviated, see ST), but
this is difficult, as it requires that the rounding */e/ → */o/ did not take place
during the Proto-Italic period, but at a later stage (see §3.2.5).
7. Lars. praen. m. lartos EF 6 (gen.). – This is the only direct attestation of this
praenomen in a Faliscan inscription, although it is indirectly attested for Middle
Faliscan in the patronym lartio MF 265. It does recur in the Etruscan inscriptions
from the area (larθ Etr XXXIV, XXXV, XXXIX, lazi Etr XI-XV, lazia Etr
XVII): see §7.7.2.33.
8. Mama/Mamma. praen. m. mama EF 1. – This praenomen is of Etruscan origin
(thus G. Giacomelli 1963:202, cf. mama OA 2.58?), rather than a shortened form
of a name such as Mamarcus, as Stolte (1928:295) suggested. Salomies (1987:75)
also discusses the apparently Oscan praenomen Mamus. Note also the (patro-
nymic?) gentilicia Māmius and Mammius in Latin and in Oscan (maamiis Cm 47,
maamieịs Po 55).
9. ? Ofe(n)tius. m. ofetios EF 4. – Vetter (1953:286-7) rendered this name as Ufen-
tius, a name derived from the potamonym Ūfens (either the modern Ofento in
Southern Latium, or another, unknown, river of the same name): for such pota-
monymic names, see §7.8.2. Although attractive at first sight, this derivation pre-
supposes an impossibly early monophthongization of /o/ → /ō ̣/ (see §3.7.2), as
G. Giacomelli (1963:208) pointed out. There appear to be no other names that can
easily be connected to ofetios, however.
10. Prauius. m. praụ[i]os EF 1. – G. Giacomelli (1963:213) classed it as a gen-
tilicium, but I doubt whether it is not rather some kind of nickname or pun on the
Latin adjective prauus, referring to the ‘good’ or ‘beautiful’ theme of several of
the Early Faliscan inscriptions and their Latin and Etruscan counterparts: see
§7.2.1.
11. ? Rufia f. rufia EF 4. – Although editors generally read rufia (rọfia Vetter
1953:285-7), this presupposes an impossibly early monophthongization of /o/ →
/ō ̣/ → /ū/ (see §3.7.2), unless it is assumed that the name is entirely unrelated to
the Latin adjective rufus. For the feminine in -ia beside a masculine in -us, cf. Ti-
tia : Titus and Tullia : Tullus (§7.7.2).
12. Sextus praen. m. z[e]xtos EF 1. – The name is attested also for the later periods:
see §7.7.1.62.
13. Tele… m. tele*[1-2?] EF 9. – Unclear: perhaps a Greek name in Thle- or Tele-?
219
CHAPTER 7
14. Titia praen. f. titias EF 3 (gen.). – The name is attested for the later periods in
titias MF 201 (gen.), as well as in a number of attestations of its male equivalent
Titus: see §7.7.1.75.
15. Vel praen. m. uelos EF 4 (either a genitive or a thematized form, see §7.2.1). –
The praenomen is also attested for Middle Faliscan: see §7.7.1.80.
16. Voltenus m. uotenosio (= uo(l)tenosio or uo‹l›tenosio) EF 3 (gen.), uoltene EF 3
(voc.). – A problem with this is that the derivation is not clear, unless it is a the-
matized adaptation of an Etruscan gentilicium *Voltena, which itself would have
to be based on the Faliscan praenomen Volta (for which see §7.7.1.85): for such
thematizations, see §7.8.2.
Of these names, Lars and Vel are clearly Etruscan; perhaps Etruscan too are Mama,
Amanus, and Voltenus. Others are Latin-Faliscan: Rufia (?), Sextus, Titia, and probably
Prauius. Ofe(n)tius seems to be of Italic origin, and the same could be true of Euius, if it
is indeed a name. Of unclear origin are Aemius, Caesius, Caleptia (?), Capena, and
Tele...
7.3. Middle and Late Faliscan male onomastic formulas
As mentioned in the preceding section, the gentilicium had become a normal onomas-
tic element from the fifth century onwards at least in the Etruscan inscriptions from
the area, and by the Middle Faliscan period, the full formula for a man’s name in Fal-
iscan inscriptions had become PRAENOMEN GENTILICIUM. It was of course still possi-
ble to use PRAENOMEN or GENTILICIUM, but these formulas are massively outnum-
bered by the full formula PRAENOMEN GENTILICIUM. In many cases, especially in se-
pulchral or public inscriptions, FILIATION was added to PRAENOMEN GENTILICIUM: see
§7.5. During the Middle Faliscan period, cognomina begin to make their appearance,
so that the full formula now became PRAENOMEN GENTILICIUM [FILIATION] COGNO-
MEN: see §7.9.
In the enumeration of the instances below, I have only included the instances
where the man’s name is the primary subject of the text, as owner, maker, deceased,
official, or dedicant. I have excluded the instances where the man’s name occurs in
FILIATION (normally as PRAENOMEN) and HUSBANDGEN WIFE (normally as PRAENOMEN
GENTILICIUM), since these formulas ‘require’ a specific form of the man’s name.
These instances are discussed in §7.9 and §7.4.2 respectively.
(1) PRAENOMEN (14-16 instances). When using a single name, there is a preference
for the use of PRAENOMEN (in contrast to the women’s names, where GENTILICIUM
seems to have been preferred: see §7.4.1). The use of PRAENOMEN is virtually limited
to Besitzerinschriften, where, within the context of the household, this would have been
220
THE ONOMASTICON
enough. In sepulchral inscriptions it is understandably rare (2 instances, against 105

instances of PRAENOMEN GENTILICIUM): although in a family tomb the gentilicium
would not require specification, it would still be useful to know which family member
was buried where. (PRAENOMEN is also the normal formula in FILIATION, as the gen-
tilicium had already been named in the name of the son or daughter: see §7.6.)
(a) Besitzerinschriften (14-17 instances): caisioi MF 20, serui MF 34-36, tulom MF 68
(if indeed genitive plural of a praenomen); iuna MF 73, iuna MF 74, iunai LF 112, iuna
MF 198, uli MF? 261-262 (perhaps a gentilicium); cauios MLF 382, uoltai MLF 367-
370; iunai MLF/Cap 475*.
(b) sepulchral inscriptions (2 instances): iuna MLF 297, iuna MLF 298.
(c) others (0 instances): -
(2) GENTILICIUM (9-12 instances). This formula also occurs mainly in Besitzerin-
schriften, where of course an item might be regarded as the property of the family rather
than of one individual. There is only one uncertain instance of its use from the sepul-
chral inscriptions, where it would not be expected to be very frequent: given that most
burials were in tombs that belonged to one or two families, inscribing a loculus with
GENTILICIUM only would have been of little use.
(a) Besitzerinschriften (8-10 instances): ani MF 45, licinio MF 259-260, ulties MF/Etr
64 (or a woman’s name?), hermana MF/Etr 265; tulie MLF 383 (or a woman’s name?),
[fel]ịcinatiu LF 384; ac ̣iuaiom Cap 465, setorio MLF/Cap 476*; anni LtF 63.
(b) sepulchral inscriptions (1 instance?): ? manileo MF 355 (I would rather read this as
m anileo, i.e., PRAENOMEN + GENTILICIUM).
(c) others (1 instance): pleina MF/Etr 199 (signature?)
(3) PRAENOMEN GENTILICIUM (156-158 instances, 43-68 with FILIATION). This is by
far the most frequent formula in all the categories into which men’s names can be
divided, except, perhaps, the Besitzerinschriften. It is also the only formula where
FILIATION and sometimes COGNOMEN are added, showing that this was the ‘official’
formula. FILIATION only occurs in the sepulchral and the (mainly Latino-Faliscan and
Latin) public inscriptions and dedications, and once in a potter’s signature (oufilo 
clipeaio  letei  fileo MF 470*).
(a) Besitzerinschriften (22-24 instances): ? f ofiti MF 58 (doubtful), tiroi  colanioi MF
69-71, caui  turi MF 273, marci  anel[i] MF 472*, cauio  peṭṛọṇẹọ MF 473*; larise
uicina MF 371, larise | uicina MF 372, statio cailio MLF 376, m adicio LF 378 (or ma
dicio?), uolti  catinei MLF 469*; c  pscni Cap 387, ḳ  uomanio Cap 388, a  írpios Cap
389, k  pa  aiedies Cap 390, at  fertrio Cap 391, f  pacios Cap 392, sex | senti Cap
399, k  sares Cap 404, m  anio Cap 420, sex  senti Cap 430, p  iunio Cap 462,
ueiụeto Cap 464 (if read as uei ụeto, and not a falsum).
221
CHAPTER 7
Besitzer sepulchral other total

PRAENOMEN MF 10-11 MF 10-11
MF 10-11
MLF 3-5 MLF 2 MLF 5-7
MLF 5-7
all
16-19
Cap 1 Cap 1
Cap 1
+ FILIATION
GENTILICIUM MF 4-6 MF 0-1 MF 1 MF 5-8
MF 5-8
MLF 1 MLF 1
MLF 1
all
9-12
Cap 2 Cap 2
Cap 2
Lat 1 Lat 1
Lat 1
+ FILIATION
SINGLE NAME MF 14-17 MF 0-1 MF 1 MF 15-19
MF 15-19
TOTAL MLF 4-6 MLF 2 MLF 6-8
MLF 6-8
all
25-31
Cap 3 Cap 3
Cap 3
Lat 1 Lat 1
Lat 1
+ FILIATION
all 22-27 all 2-3 all 1 all 25-31
PRAENOMEN + MF 6-7 MF 20 MF 4 MF 30-31
GENTILICIUM MLF 5 MLF 11 MLF 4 MLF 20
LF 9 LF 1 LF 10
LtF 1 LtF 1
Cap 11-12 Cap 11-12
Lat 5 Lat 5
+ FILIATION? MF 14 MF 14 MF 57-58
MLF 3 MLF 3 MLF 29
LF 3 LF 3 LF 28 all
LtF 5 LtF 2 LtF 7 LtF 11 155-157
Cap 11-12
Lat 19
+ FILIATION MF 12 MF 1 MF 13
MLF 6 MLF 6
LF 13 LF 2 LF 15
LtF 2 LtF 1 LtF 3
Lat 6 Lat 8 Lat 14

all 22-24 all 105 all 28 all 155-157
TOTAL 44-51 107-108 29 180-188
Fig. 7.1. The onomastic formulas of men’s names.
222
THE ONOMASTICON
Besitzer sepulchral other total

PRAENOMEN MF 1 MF 4 MF 5
MLF 1 MLF 1 MLF 2
MF 6
MLF 3
all
+ FILIATION MF 1 MF 1 LF 2
11
MLF 1 MLF 1
LF 2 LF 2
GENTILICIUM MF 7 MF 9 MF 1 MF 17
MLF 4 MLF 2 MLF 6
LF 1 LF 1
MF 17
MLF 6
Lat 1 Lat 1 all
LF 1
+ FILIATION 28
LtF 2
Lat 2
LtF 2 LtF 2
Lat 1 Lat 1
SINGLE NAME MF 8 MF 13 MF 1 MF 22
TOTAL MLF 4 MLF 3 MLF 1 MLF 8
LF 1 LF 1
MF 23
MLF 9
Lat 1 Lat 1 all
LF 3
+ FILIATION MF 1 MF 1 39
LtF 2
MLF 1 MLF 1
Lat 2
LF 2 LF 2
LtF 2 LtF 2
Lat 1 Lat 1
all 13 all 24 all 2 all 39
PRAENOMEN + MF 19 MF 19
GENTILICIUM MLF 0-1 MLF 0-1
LF 6 LF 6
MF 23
Cap 0-2 Cap 0-2
MLF 0-1
Lat 2 Lat 2
LF 10 all
+ FILIATION?
35-38
+ FILIATION MF 4 MF 4 Cap 0-2
Lat 2
LF 4 LF 4
all 0-3 all 35 all 0 all 35-38

TOTAL 13-16 59 2 74-77
Fig. 7.2. The onomastic formulas of women’s names.
223
CHAPTER 7
(b) sepulchral inscriptions (105 instances, 39 with FILIATION, and a further 25 where
FILIATION is probable or possible): uolti ᛂ teti MF 11, cau‹i›o  pauiceo  | ḷ[oc]ịes MF
12 with FILIATION, o hạθ̣i MF 13, teti atron or teti atronị MF 13, uolta | neroni | ca fi
MF 15 with FILIATION, iuna | malio MF 39, iuna  oufilio MF 48, cauio  aufilio MF 49,
caui[o]  auf ̣ilio MF 50, ḳai[s]i[o  auf]ilio  iun[?eo] MF 51 with FILIATION; [---]a 
aufi[lio ?---] MF 53 (where FILIATION can be restored), puponio  firmio MF 54, uel
zu[con]|eo MF 56, ca lin[---] MF 57 (where FILIATION can be restored), leiuelio pạrtis |
uolti MF 79 with FILIATION, [ma]rco  pleina  marcio  man[o]mo MF 80 with FILIA-
TION, uel [] uisni  olna MF 82 (double gentilicium), [.]pi  uesθi MF 83, caui []
ṭ**(*)[i] MF 84, mar  eina MF 87, [uo]ltio [] ueicọno MF 88, [4-5]ạ hac****a 
[?]ạ[?]ṃ  maximo MF 89 with FILIATION, [leu]elio  cailio [ ...| max]om[o ] MF 90
(where FILIATION can be restored), la  ie[---] MF 93 (where FILIATION can be restored),
iuna  ce[lio---] MF 96 (where FILIATION can be restored), cẹịṣ[i.] | holc[osi] | ar  p[...]
MF 140 with FILIATION?, [---]io  cre[---] MF 142 (where FILIATION can be restored),
[---]o  cṛ[---|---]leo  c[---] MF 143 (where FILIATION can be restored), ca[u]io 
le[ueli]o  cau[i] | hileo MF 146 with FILIATION, i[un]a lẹ[---] MF 148, [u]olṭa  pupe-
lio | [m]ano[m]o MF 149, tulo  pup[elio ?---] | ịuneo MF 151 with FILIATION, uolti[o ]
marc[---] | putellio MF 152 (with FILIATION?), cais[io] |  zaconiọ MF 153, iu  uiụị[---]
MF 157 (where FILIATION can be restored), [---]l su|[---] MF 191 (where FILIATION can
be restored), cauio  arutlo MF 195, cauio ṣ*[---] MF 197 (where FILIATION can be re-
stored), aruz  cesịe  aruto MF 257 with FILIATION, ueltur  tetena | aruto MF 266 with
FILIATION, arute macena MF 269, larise  mar||cna  citiai MF 270 (with FILIATION?),
cauio  nomes|ina  maxomo MF 272, cauio  oufilio | uolteo MF 275 with FILIATION,
ceisio  oufilio | uolθeo MF 276 with FILIATION; c mecio  a[---] MLF 211 (where FILIA-
TION can be restored); tito  uelmineo | titọi MLF 305 with FILIATION, iuna uelmineo |
titio MLF 307 with FILIATION, cauio uelminẹo | popliai file MLF 308 (with FILIATION?),
tito ᛂ uelmineo | nu i*ice MLF 309 with FILIATION, cuicto uelmineo |[---?] MLF 310
(where FILIATION can be restored), uoltio [] uelmineo | titio MLF 312 with FILIATION,
uolta  uelmineo MLF 313, tito  uel|mineo  iun|aị ị*ice MLF 315 with FILIATION,
popli[o] | uelmi|no MLF 316, cauio  latrio MLF 324, m  tito  tulio  uoltilio  hescuna
MLF 346 (double gentilicium) with FILIATION, aufilo  aratio MLF 348, cauio  aratio
MLF 349, tito  ar‹a›tio MLF 350, caisio  tirio MLF 351, f aino MLF 352 (if not to be
read as faino?), oct*i[....] uoltili MLF 353 (if read as oct* i[....] uoltili or oct*i [...]
uoltili) with FILIATION, tito polafio MLF 354, m anileo MLF 355 (if not to be read as
manileo), [iu]na  upreciano MLF 363; uoltio  uecineo | maxomo | iuneo LF 220 with
FILIATION, marcio  acarcelinio LF 223, ca  uecineo [] uoltiọ LF 224 with FILIATION,
ca  uecineo LF 225, tito [] acarcelinio  | ma  fi LF 226 with FILIATION, l  clipịaṛ[io]
LF 230, c  clipeaṛ[io] | m  f LF 231 with FILIATION, c  ạu[---]ịsi LF 236 with FILIA-
TION, [.]a  protacio  m  f LF 242 with FILIATION, [---]o [] spuṛ[ilio LF 248 (where
FILIATION can be restored), sesto ᛍ | fulczeo LF 329, uoltio | folcozeo | zextoi | fi LF 330
224
THE ONOMASTICON
with FILIATION, cesio folcuso LF 331, celio *olcuzeo | ***io | poplia | uelcei f||e LF 332
(with FILIATION?), [..] folcosio | *****oi LF 333 with FILIATION, cauio | uetulio LF 335,
tito  marhio | uoltilio LF 336 with FILIATION, ueltur | ortecese LF 339; cailio  tirio
MLF/LtF 358, tito  batio MLF/LtF 359; m  cḷ[i]peario  m [ f LtF 233 with FILIATION,
m  pani[---] LtF 239, uo  nel[n---] LtF 299, m  neroni | a  f LtF 325 with FILIATION, st
 aco[---] LtF 327 (where FILIATION can be restored), [se]x  ne[?]ro[---] LtF 328
(where FILIATION can be restored), c  neroni LtF 340, m  aco[---]| rutil  ce[---] LtF
341 (with FILIATION?); m  spurilius  c  f Lat 237 with FILIATION, c  spurilius | m  f Lat
238 with FILIATION, Pu(blius) Fuluius C(aii) f(ilius) | C(aii) n(epos) Suto(r) Lat 250
with FILIATION, l  uecilio  uo  f Lat 251 with FILIATION, [. ]uecilio  l  f Lat 251 with
FILIATION, l  c  leuieis  l  f Lat 251 with FILIATION.
(c) others (28 instances, 14 with FILIATION): (i) signatures (4 instances, 1 with FILIA-
TION): c ̣ cutri MF 200, arθ[3-5]rẹ MF/Etr 267, oufilo  clipeaio  letei  fileo MF 470*
with FILIATION, cauios frenaios MF 471*; (ii) inscriptions on public works, public dedi-
cations (16 instances, 10-12 with FILIATION): a [.]osena MLF 206, uẹ narionio MLF
206, cauio lullio MLF 207, cauio latinaio MLF 210; [..] hirmio  m[ f  ] LF 213 with
FILIATION, ce  tertineo  c  f LF 213 with FILIATION; c* *(*)coṇẹo  l***(*) LtF 290
with FILIATION (or a cognomen?), ce  paui[ceo 1-2]so LtF 290 with FILIATION (or a
cognomen?), [---]ilio  c[  f ?---] LtF 215 with FILIATION; la  cotena  la  f LF/Lat 214
with FILIATION, l  latrius  k  f Lat 218 with FILIATION, c  salu[e]na  uoltai  f Lat 218
with FILIATION, [.  u]mpricius  c  f | [?]aburcus Lat 219 with FILIATION, c  egnatius 
s[ex ] f Lat 291 with FILIATION, c  didius  t  f Lat 456 with FILIATION, m  uettius  m 
f Lat 456 with FILIATION; (iii) private dedications (5 instances, 1 with FILIATION): [.]
ṃunio regena* Lat 377, mar  popi st  f Cap 421 with FILIATION, l  calpurnius Cap 432,
m  t  u  genucilio Cap 435 (three liberti), [---]rcius Cap 436 (a libertus); (iv) reversed
filiation (1 instance): marci  acarcelini LF 221; (v) statue (1 instance): caui  tertinei
MLF/Cap 474*; (vi) unknown (1 instance): st  clanidio Cap 394.
To these instances should be added the following 33 damaged sepulchral inscriptions:
(a) with only the praenomen preserved (7 instances, 1 with FILIATION): ueṇe[?---] MF
43, [u]ọlt[---] MF 145, kreco  [---] MF 147, uol[ta ]**[---] MF 158, [leu]elio  [---
|---]io  ca[---] MF 159, [u]olṭ[---| u]oltịḷ[i---] MF 163 with FILIATION; cauio [---]|
ruso[?---] MLF 318 (ruso may be a cognomen); (b) with only the gentilicium preserved
(19 instances, 8 with FILIATION): [.......]| neṛ[oni.] | ị[......] MF 16 with FILIATION, [---]
ḥiṛṃeo iu  MF 19 with FILIATION, [---] oụ*[..]o *ạ**[---] MF 52 with FILIATION, [---
c]ẹlio  cesi  fi|[---] MF 94 with FILIATION, [---] celio [---] MF 95 possibly with FILIA-
TION, [--- c]elio [---] MF 97 possibly with FILIATION, [---] reic ̣[lio] | [---] ṃaxoṃ[o]
MF 98 probably with FILIATION, [---] reiclio [?---] MF 99 possibly with FILIATION,
[......]| reị[cli.] | m*[....] MF 100, [---]iena  ụ[---] MF 102, f ̣af ̣[---] MF 139? possibly
with FILIATION, [---] crạ[i---| iu?]nẹo MF 141 with FILIATION [---]ronio  uol[t---] MF
156 with FILIATION, [---|---] ṃarcio LF 228 (or a patronym?) [---p]rotacio[---] LF 244
225
CHAPTER 7
possibly with FILIATION [---]fate MLF 285, [---?] precono[---] MLF 361, [---] upre-
ciano MLF 364, [---]nio  ia  * LtF 341 with FILIATION; [---]**  uei|[---] t  f LtF 327
with FILIATION; (c) with only the cognomen preserved (1 instance, with FILIATION): [---
ma]x ̣ọmo ᛂᛂ uoltilio MF 162 with FILIATION; (d) with only the FILIATION clearly pre-
served (6 instances, all with FILIATION): [---]cẹla[---|---] iun[---] MF 166, [---]f LF 247,
[---]*o  c  f LF 249; [---]ọ c ̣e ᛍ F LtF 171, [---] c  f  mo[---] LtF 172, [---?] decon[--
-|---]a  f LtF 174
I have not included the following instances: (a) unclear reading or interpretation: apolo
MF 65 (a theonym), namureṭua (?) MF/Etr 66, ạcṛẹẓ cat (?) MF 67, tuconu (?) MF 85
(perhaps PRAENOMEN GENTILICIUM, if read as t u(e?)conu), [--- pu]peḷ[i---] MF 150,
[---u]oltio  | [---]o  MF 164, [---]io  uolti[---] MF 167, [---ar]uto r[---] MF 169, cẹsit
 fere MF 263, puiatu MLF 208, laris  ṃ  ṛφ̣c ̣χ̣a | uạịẹṣịṿsị st (?) MLF/Etr 290;
[--- ?]anco ma LF/LtF 232, [---] cuba |[---]nte LtF 326; (b) either a man’s name or a
woman’s: uentarc[i ..... MF 80, uoll[---]MF 86 (GENTILICIUM?), [---] uenelịes MF 258
(probably a gentilicium), popl[---]| [u]elṃi[ne---] MLF 317, [---?]*a*ḳit*ue* a  f LF
234 (PRAENOMEN GENTILICIUM FILIATION?), [---]a  neln f|[---]uxo  ohi*[..] LtF 300;
(c) isolated names in -e(s): acre MF/Etr 279, ame MF/Etr 280, aṃẹ MF/Etr 282.
The material is presented in tabular form in fig.7.1. From this table the following ten-
dencies may be read:
x The use of the single name (whether PRAENOMEN or GENTILICIUM) is normal only
in the Besitzerinschriften (22-27 instances, out of a total of 44-51) and very rare in
sepulchral and other inscriptions (3-4 instances altogether, out of a total of 136-137).
x If the single name is used, it is PRAENOMEN (16-19 instances) rather than GEN-
TILICIUM (9-12 instances). The use of GENTILICIUM is very rare in all categories. This
is a contrast with the formulas of women’s names: see §7.4.1.
x PRAENOMEN GENTILICIUM is the normal formula in the sepulchral inscriptions (105
instances, out of a total of 107-108). It is also quite frequent in Besitzerinschriften
(22-24 instances, out of a total of 44-51).
x In the categories other than Besitzerinschriften and sepulchral inscriptions,
PRAENOMEN GENTILICIUM appears to have been the normal formula in public inscrip-
tions, including public dedications (20 instances altogether, out of a total of 29).
x Note also that in FILIATION, the normal formula for the father’s name is
PRAENOMEN, with the exception of ca uipi  leueli | filea MF 14.
Resuming, it may be said that PRAENOMEN GENTILICIUM FILIATION is the official for-
mula from the first Middle Faliscan inscriptions onward: in that respect, the Faliscan
usage does not differ from that of Etruscan, Latin, or the Sabellic languages. The use
and the frequency of FILIATION are discussed in §7.9.
226
THE ONOMASTICON
7.4. Female onomastic formulas
7.4.1. The formula of women’s names. In Faliscan usage, as in Etruscan, contempo-

rary Latin, and apparently also in the Sabellic languages (although the material is lim-
ited in this case) 115, it was normal for a woman to have a praenomen as well as a gen-
tilicium: pace G. Giacomelli (1963:160), there is no great difference between the ager
Faliscus and the surrounding areas in this respect. Views on this subject can be and
have been obscured by the later Roman usage, where women’s praenomina became
increasingly rarer (see Kajava 1995:114-8). This decrease in the use of praenomina
does not appear to have occurred in the ager Faliscus, however: the use of the double
name remains regular throughout all periods (see below). The full formula for
women’s names in Faliscan is therefore PRAENOMEN GENTILICIUM, which may be
extended with FILIATION (§7.5) and the marital formula HUSBANDGEN WIFE (§7.4.2).
Yet even in the ager Faliscus and Capenas women were often designated by one name
only, and this is usually GENTILICIUM, as in Latium, whereas in the case of men there
was a preference for PRAENOMEN (§7.3). There are no instances of Faliscan women
having a cognomen, as is to be expected: in Etruscan, woman’s cognomina are ex-
tremely rare in South Etruria, and occur with any frequency only at Clusium (cf. Rix
1965:40-2), while in Latin the earliest examples are probably from the second half of
the second century (Kajanto 1977a:64-7, Kajava 1995:30-1): see §7.9.
(1) PRAENOMEN (11 instances, 4 with FILIATION). This formula was used both in Be-
sitzerinschriften and in sepulchral inscriptions. The use in Besitzerinschriften can be
compared to the quite frequent use of PRAENOMEN in Besitzerinschriften where the
owner is male (see §7.3). In sepulchral inscriptions, the use of P RAENOMEN for women
(9 instances) is more frequent than that of PRAENOMEN for men (2 instances), even
though the number of recognizable women’s names in sepulchral inscriptions (59) is far
smaller than that of men’s (107-8). This is probably due to the fact that women were
buried with their husbands in the tomb of the husband’s family: what mattered in these
inscriptions was not the woman’s own gentilicium, which differed from that of her hus-
band’s, but the fact that she was related by marriage to one of the family owning the
tomb. In 6 out of the 9 instances of PRAENOMEN in women’s sepulchral inscriptions, the
name of the woman follows that of a man whose gentilicium is given (MF 48, 49, 50,
MLF 312, LF 242, LF 332) and who was presumably her husband.
(a) Besitzerinschriften (3 instances, 1 perhaps with FILIATION): titias MF 201; locia
eiṃoi MLF 293 with FILIATION (?), sceiuai LF 379.
115
Only the Paelignian sepulchral inscriptions (Pg 12-17, 28-33, 51-54) give some insight in
the formulas of women’s names; all other Sabellic languages yield only a few instances of
women’s names each (Um 3, 30, 38; MV 6; MV 7; Hi 4, 7; Po 51, 66?, Cm 25, Lu 46).
227
CHAPTER 7
(b) sepulchral inscriptions (9 instances, 3 with FILIATION): ca uipi : leueli | filea MF 14

with FILIATION, poplia MF 48, θanacuil MF 49, poplia MF 50, [p]ọpliạ MF 160; sextia
MLF 311, sceụa MLF 312; cauia  uxo  a  f LF 242 with FILIATION, poplia | uelcei f||e
LF 332 with FILIATION.
(c) others (1 instance): popliai MLF 308.
(2) GENTILICIUM (25-26 instances, 3 of which with FILIATION). If a single name is
used, GENTILICIUM appears to have been the preferred formula in the case of women,
whereas in the case of men there was a preference for PRAENOMEN. This formula, too,
appears to be used both in Besitzerinschriften and insepulchral inscriptions.
(a) Besitzerinschriften (9-10 instances) turia MF 22, 23, 24, 25, 26, 27, ? ulties MF/Etr
64 (unclear); pupiias MLF 304, seralia LF 380, tulie MLF 383.
(b) sepulchral inscriptions (15 instances, 3 with FILIATION): loụṛia MF 41, uolḷia MF
47, latria MF 75, θania MF 81 (where two other women are designated as ca 
u[eculi]a and ca  e[c]ṇata), [---]  cẹsiḷị[a] MF 99, cincia MF 135, uoltaia MF 196,
morenez MF 269, zeruatronia MF 272; fulonia MLF 313; m{e?}ania LF 224 (in LF 225
she is described as ca mania); hlau|elea  m  f  LtF 325 with FILIATION, plenes  q  f
LtF 231 with FILIATION; plenese Lat 251, claudia  c  f Lat 393 with FILIATION.
(c) others (1 instance): citiai MF 270 (in FILIATION?).
(3) PRAENOMEN GENTILICIUM (35-38 instances, 8 of which with FILIATION). As in the
men’s names, this appears to have been the official formula, as it is the regular formula
in sepulchral inscriptions (35 instances, out of a total of 59 recognizable women’s
names in sepulchral inscriptions) and is moreover virtually restricted to sepulchral in-
scriptions (35 instances out of a total of 35-38 instances of PRAENOMEN GENTILICIUM).
Yet even in the sepulchral inscriptions there is a large number of instances where a sin-
gle name is used (24 instances out of a total of 59 recognizable women’s names), and
the number of attestations of single names is slightly larger (39 instances) than that of
PRAENOMEN GENTILICIUM (35-38 instances).
(a) Besitzerinschriften (0-3 instances): ? uei uatia MLF 463 (unclear); ? ca  e**sa Cap
458 (very unclear), ? sta sediu Cap 466.
(b) sepulchral inscriptions (34 instances, 8 with FILIATION): poplia | ḥirmia MF 18;
fasies  c[ai]sia MF 41; cauia  satelie MF 42; cauiacue | [u]eculia  uoltilia MF 80 with
FILIATION; ca  u[eculi]a MF 81; ca  e[c]ṇata MF 81; tan[---] | cail[ia ?---] MF 92;
tanacu[il] | anelia MF 101; poplia | fafarṇ MF 136; [.?]ạ*ịạ | lepuia | uoltilia MF 144
with FILIATION; ian[ta  ..]lni[a] MF 146; iata  leue[lia] MF 147; ṭị [] ṭịria
lo[?---]|l[e]a  cs  f MF 155 with FILIATION; poplia  calitenes MF 265; poplia | zuco-
nia MF 271; [---]nθia MLF 212 (or a patronym?); ịạ  firmia  titia MLF 302 with
FILIATION; poplia  cocelia MLF 303; cauia loriea MLF 314; tana | lartia MLF 338 (or
is lartia a patronym?); tan‹a›cuil  aratia MLF 347; cauia | hadenia MLF 360; iata 
228
THE ONOMASTICON
senθia MLF 362; ca ạconia LF 220; uipia  zertenea LF 221; cauia  uecinea LF 223;
ca  mania LF 225; pola marcia  sus[?---] LF 227; cau[ia ] uecin[e]a  uotiliạ LF 222
with FILIATION; pop  petrunes  ce  f LF 226 with FILIATION; cesula  tiperilia  te f LF
229 with FILIATION; ceịsịa *e[0-4?]i*ia  ce LF 235 with FILIATION; cauia | uetulia LF
334; po[l]ae  abelese Lat 251; ? a[rria] | plaria LtF 340
(c) other (0 instances): -
To this should be added 17 damaged sepulchral inscriptions: (a) with only a praenomen
preserved (5 instances): cauia * *|[---]a MF 94, iatac ̣ue  ḷ[---] MF 158, [po]plia [ ---]
MF 161, cauia | [---] MLF 306, mino  s[---] LtF 173 (b) with only a gentilicium pre-
served (5 instances): iuna  ce[lio---] | arutielia [?---] MF 96 [---pu]peḷ[i---] MF 150,
[---]  zaconiai MF 154, [---] uenelịes  sapnonia MF 258, st  aco[---]| leuia[---] LtF
327; (c) with fragmentary women’s names: ueṇe[?---]na  | ux[o(r) ?---] MF 43, [---] |
cai[lia ---] MF 93, [---]iena  ụ[---|---]ono  ux[o(r)] MF 102, [---?]uoxie[.]eai MLF
310, [---]*lia  c  f LF 249 with FILIATION, [....]nea  *a | [u]xor ia* | ma  oṣcin* LtF
301 with FILIATION, [---]nio  ia  * [---]ilia  cọ* LtF 341.
I have not included: (a) unclear reading or interpretation: ṛica MF/LtF 21, namureṭua
(?) MF/Etr 66, ạcṛẹẓ cat (?) MF? 67 (praenomen or gentilicium?), ipa MF? 78,
[---]ạltai  MF 109, aie* MF 110, [---]a*ome MF 156, apa Cap 457; (b) either a man’s
name or a woman’s: uentarc[i ..... MF 80, uoll[---]MF 86 (GENTILICIUM?), [---] ue-
nelịes MF 258, popl[---]| [u]elṃi[ne---] MLF 317, [---?]*a*ḳit*ue* a  f LF 234, [---]a
 neln f|[---]uxo  ohi*[..] LtF 300; and (c) isolated names in -e(s): acre MF/Etr 279,
ame MF/Etr 280, aṃẹ MF/Etr 282.
The material is presented in tabular form in fig.7.2. From this table the following ten-
dencies may be seen:
x In single names, GENTILICIUM (28 cases) is preferred to PRAENOMEN (11 cases), in
spite of the fact that in the ager Faliscus women did have praenomina.
x Not only is GENTILICIUM more popular than PRAENOMEN, it appears to become
more popular as time progresses, perhaps due to Latin influence:
praenomina MF: 6 MLF: 3 LF: 2 LtF: - Cap: - Lat: -
gentilicia MF: 17 MLF: 6 LF: 1 LtF: 2 Cap: - Lat: 2
x The use of the single and the double name appears to have been equally popular:
single name MF: 23 MLF: 9 LF: 3 LtF: 2 Cap: - Lat: 2
double name MF: 23 MLF: 0-1 LF: 10 LtF: - Cap: 0-2 Lat: -
x The use of PRAENOMEN GENTILICIUM is virtually limited to sepulchral inscriptions
(35 out of 35-38 cases of the use of the double name). The single name can be used in
Besitzerinschriften (13 out of 39 cases of the single name) and in sepulchral inscriptions
(24 out of 39 cases of the single name).
229
CHAPTER 7
7.4.2. Adding the husband’s name: the marital formula.116 In several instances, the
woman’s onomastic formula is further expanded by adding HUSBANDGEN WIFE. Leav-
ing aside cultural and personal motives to add the husbands name, the primary aim in
doing so in the Faliscan family-tombs will have been to clarify the relationship be-
tween the various deceased buried in the same tomb, as a wife would of course have
had a gentilicium that differred from that of her husband’s family. With the exception
of LF 242 (below) the formula is only used in sepulchral inscriptions of women
whose husband had not or not yet been buried in the same loculus. In the case of
cau[ia ] uecin[e]a  uotiliạ | maci  acacelini  uxo LF 222, the text was in fact re-
placed after the husband’s burial in the same loculus by marcio  acarcelinio | cauia 
uecinea | hẹc ̣ cupat LF 223, which shows the usual Faliscan custom of simply stating
the names (sometimes joined by -cue) when husband and wife were buried together.
Adding this marital formula is mostly an Etruscan custom (cf. the numerous in-
stances of puia and puiac in ET). The Latin sepulchral inscriptions yield only 24 in-
stances where the name of the woman (either as the deceased or as the ‘procurateur’,
cf. §8.9.2) is accompanied by uxor (CIL I2.171, 184, 288, 300, 1220, 1289, 1294, 1328,
1349, 1352, 1424, 1432, 1490, 1536, 1595, 1824, 1829, 1830, 1843, 1886, 1907, 2284,
2460, 2636). The only Sabellic instance of this custom appears to be Paelignian [4-5]
p ̣racom p ̣[20-30] | ụsur pristafalacirix Pg 9. The relative frequency of the formula in
the ager Faliscus is probably due, not to direct Etruscan influence, but rather to the fact
that the Etruscan and Faliscan areas shared the same mode of burial, and the function of
the sepulchral inscription was therefore the same in both areas.
The instances of the addition of HUSBANDGEN WIFE occur in Middle Faliscan,
Late Faliscan, and Latino-Faliscan inscriptions:
(a) HUSBANDGEN WIFE (4-8 instances). This appear to have been the regular formula,
with the components in the same order as in the filiation formula FATHERGEN SON/
DAUGHTER (cf. §7.5). It occurs in cauia  satelie | caui  felic ̣inate | uxor MF 42, fasies 
c[ai]sia | louci  teti  uxor MF 41, poplia  calitenes | aronto  cesies | lartio  uxor MF
265; cau[ia ] uecin[e]a  uotiliạ | maci  acacelini  uxo LF 222, probably also in  iii 
ḷ[.......|..........]naị[?---|....]o uxo MF 17, [---]iena  ụ[---|---]ono  ux[o(r)] MF 102, pola
marcia  sus[?---] LF 227 (if sus[?---] = s us[o(r)), and perhaps in [---]a  neln
f|[---]uxo  ohi*[..] LtF 300. Note that LF 222, cau[ia ] uecin[e]a  uotiliạ | maci  aca-
celini  uxo, has both FILIATION and [HUSBANDGEN WIFE], in this order. This is to be ex-
pected, not because of a greater role of the father or of the gens the woman was born
into, but rather because FILIATION was at once a more common (everyone has a father,
but not everyone has a husband) and a more general (men and women both have fa-
thers, but only women have husbands) part of the onomastic formula.
116
I refer to this formula as marital rather than as gamonymic, restricting the latter term to the
Venetic use of the gamonymic adjective (cf. Lejeune 1974:60-3).
230
THE ONOMASTICON
(b) WIFE HUSBANDGEN (1-4 instances). This appears to have been a (rare) variant where
the members of the formula appear in reversed order, occurring in tanacu[il] | anelia  
| uxor  ia MF 101, perhaps also [....]nea  *a | [u]xor ia  * | ma  oṣcin* LtF 301, per-
haps also cauio [---]|ruso[?---] MLF 318 (if uso[?---] = uso[(r) … = ‘uxor …’), per-
haps also ueṇe[?---]na  | ux[o(r) ?---] MF 43. The fact that this reversed formula oc-
curs at all may indicate that the marital formula was not as fixed as FILIATION.
(c) An exceptional case is [.]a  protacio  m  f  mạcistratu | keset  cụestod  pi  pretod
 pis | cauia  uxo  a  f LF 242: (a) it is the only instance where the marital formula is
added to PRAENOMEN, while in elsewhere it is added to PRAENOMEN GENTILICIUM; (b)
it is the only certain instance where the marital formula consists of WIFE only; and (c) it
is the only certain instance where FILIATION follows the marital formula instead of pre-
ceding it, as in LF 222. All these exceptional features arise from the simple fact that the
normal usage in Faliscan inscriptions was to use the marital formula HUSBANDGEN WIFE
only in cases were husband and wife were not buried together: it almost looks as if the
composer of the text of the inscription, which focuses entirely on the husband and his
impressive cursus honorum, had misunderstood the marital formula.
(d) Unclear is m  cḷ[i]peario  m [ f | --- |---]or LtF 233 (if ---]or = … ux]or).
In a few inscriptions, a word other than uxor appears to have been used. In two Latino-
Faliscan or Latin inscriptions the word coniunx may have been used, but these texts are
too damaged to be sure: m  aco[ ]nio  ia  * | rutil  ce[ ]ilia  cọ* LtF 341 and [---?]
decon[---|---]a  f LtF 174, if to be read as de con[---]): note that, as said above, in Latin
inscriptions the word used is always uxor, never coniunx. I find it hard to agree with
Vetter’s (1953:305) suggestion to read lepuia in [.?]ạ ịạ | lepuia | uoltilia MF 144 as an
abbreviated husband’s name le followed by Etruscan puia ‘wife’: not only would this
be an instance of interference from Etruscan, which is very rare (§9.2.2), but it would be
an instance of interference within a formula, but it would also be an instance of HUS-
BANDGEN WIFE preceding FILIATION.
7.5. The formula of filiation
7.5.1. FILIATION. The onomastic formula can be extended with FILIATION. As may be
seen from the tables presented earlier in this chapter (figs. 7.1-2), its use appears to have
been ‘formal’. First, FILIATION is found only in sepulcral and official inscriptions (in-
cluding public dedications), with the exceptions of the Besitzerinschrift locia eiṃoi
MLF 293, the signature oufilo  clipeaio  letei  fileo  met  facet MF 470*, and the pri-
vate dedication mar  popi  st  f  n  mart  d  d  me Cap 421. Second, FILIATION is
normally added only to the full onomastic formula PRAENOMEN GENTILICIUM, although
there are rather more exceptions in the case of women: not only the Besitzerinschrift
231
CHAPTER 7
locia eiṃoi MLF 293, but also within the sepulchral inscriptions, ca uipi  leueli | filea
MF 14, cauia  uxo  a  f LF 242, poplia | uelcei f||e LF 332, hlau|elea  m  f  LtF 325,
plenes  q  f LtF 231, and claudia  c  f Lat 393. These exceptions are probably due to
the fact that in women’s names the use of the single name was more common than in
men’s names. With regard to the use of FILIATION over time, I can find no discernable
tendency in the tables other than a slight increase in its use in the case of men in the
Late Faliscan and Latin inscriptions, which is at least partly due to its use in public in-
scriptions, which are more frequent in these categories.
7.5.2. FATHERGEN [SON/DAUGHTER] and the patronymic adjective. In the Faliscan

inscriptions, FILIATION was always added after the gentilicium, as in Etruscan, Latin,
and the majority of the Sabellic languages, not after the praenomen, as was the custom
in Umbrian and Volscian.117 However, FILIATION could be expressed in two completely
different ways, namely (1) by the formula FATHERGEN [SON/DAUGHTER], and (2) by
means of a patronymic adjective (see also fig.7.3):
(1) The formula FATHERGEN SON/DAUGHTER and FATHERGEN. In Faliscan, one of the
ways to express filiation was the formula FATHERGEN filius/filia, as in Latin. Whether
FATHERGEN, which occurs in several inscriptions (see fig.7.3), was an independent for-
mula or just a shortened form of FATHERGEN filius/filia cannot be established. In Etrus-
can, the frequency of FATHERGEN clearly shows that it was a separate formula that could
be used as an alternative to FATHERGEN clan/seχ. In the Sabellic languages, FATHERGEN
was in fact the normal formula, whether placed after the praenomen, as in Umbrian and
Volscian, or after the gentilicium, as in Oscan and the other Sabellic languages. It may
be significant that in the only clear Middle Faliscan examples of F ATHERGEN, aruz 
cesịe  aruto MF 257 and ueltur  tetena | aruto MF 266, the names are of the persons
involved are all Etruscan.
(2) Patronymic adjectives. In Faliscan, filiation could also be expressed by patronym-
ic adjectives, derived from the father’s praenomen by means of the suffix /-io-/, e.g.
Marcus → Marcius ‘Marcusson’ or Titus → Titius ‘Titusson’. When the father’s name
itself was already derived with this suffix, the suffix -ilio- (possibly originally a diminu-
tive suffix) was used instead, e.g. Voltius → Voltilius. The same suffixes appear in the
Latin and Sabellic patronymic gentilicia, although in the Sabellic languages -idio- was
used rather than -ilio-. When the father’s name belonged to the first declension, as in the
case of the Faliscan praenomina Iuna and Volta, the resulting adjectives were Iunius
and Voltius rather than Iunaeus and Voltaeus. In view of the spellings ịuneo MF 151,
iuneo LF 220, perhaps also iu?]nẹo MF 141, and uolteo MF 275, uolθeo MF 276, this
117
Two inscriptions from the area that have been interpreted as showing the Umbrian order
are k  pa  aiedies  Cap 390 and the older reading of Cap 388, ṭ c ̣  uomanio.
232
THE ONOMASTICON
may be doubted, since the -eo may stand for /-ę̄ŏs/ ← /-ŏs (§3.7.6). On the other
hand, I read uoltio in LF 224, and when the patronymic adjectives of Iuna and Volta are
used as a praenomen or a gentilicium, they are always spelled with i (gentilicium iunio
Cap 462; for the attestations of the praenomen Voltius, see §7.7.1.86).
Patronymic adjectives are of course well known from other Indo-European lan-
guages: they were used e.g. in various Greek dialects (especially Lesbian, Boeotian, and
Thessalian, see Buck 1955:134-5) and in Venetic (cf. Lejeune 1974:52-7), and must
have been used in both Latin and the Sabellic languages as well, as many Latin and
Sabellic gentilicia are of patronymic origin. There are in fact several instances of Sa-
bellic forms in -is occurring in the usual position of the filiation that cannot be geni-
tives and could be patronymic gentilicia. This interpretation is debatable, however,
and some of the forms may simply be graphical errors. The instances are: Umbrian
titis in vu‹v›çis titis teteies TI Ib.45 (probably an error for tit‹e›s or titi‹e›s: this line and
the preceding one contain several errors), South Picene taruis in [---]s  taruis 
petrúnis AP.5 (unclear) and pet{i}eronis in noúínis  pet{i}eronis  efidans AP.4 (per-
haps a gentilicium), and Praesamnitic veneliis in vinuχs veneliis peracis estam tetet
venelei viniciiu Ps 3 (a patronymic adjective if peracis is a gentilicium).
It is therefore not the existence of patronymic adjectives in Faliscan that is re-
markable, but the fact that they continued to be used long after they had become fossi-
lized as gentilicia throughout the rest of Central and Southern Italy: the Faliscan materi-
al shows instances of patronymic adjectives even from the Late Faliscan inscriptions
(see fig.7.3). Although the remarkable use of the patronymic adjectives in Faliscan has
often been pointed out, it should be noted that this was always an option and not the
only possible variant: the patronymic adjective and the formula FATHERGEN [filius/filia]
occur side by side from the Middle Faliscan period onwards. There is no indication that,
originally, the Faliscan way of expressing filiation was exclusively by means of the pa-
tronymic adjective and that FATHERGEN [filius/filia] was due to influence from other tra-
ditions: FATHERGEN [filius/filia] appears already at times when it is hardly possible to
ascribe its use to Latin influence, and although it could conceivably be modelled on the
Etruscan formula FATHERGEN [clan/seχ], there is certainly no need to assume this.
On the other hand, the disappearance of the use of the patronymic adjective may
well be ascribed to Latin influence, as can be seen from fig.7.3, where the instances of
the patronymic adjective and of FATHERGEN [filius/filia] are presented in tabular form.
This clearly shows that the patronymic adjective is the more frequently used option in
the Middle Faliscan inscriptions, but in the Late Faliscan inscriptions it is less frequent
than FATHERGEN [filius/filia], while the Latino-Faliscan and Latin inscriptions have only
FATHERGEN [filius/filia].118
118
Even if my interpretation of the forms in -oi in MF 40, MLF 293, 305, and LF 330, 333 as
genitives is not adopted, this picture remains the same: in fact, it becomes more pronounced.
233
CHAPTER 7
patronymic adjective FATHERGEN [filius/filia]
MF Falerii Veteres marcio 80 uipi  leueli | filea ♀ 14

uoltilia ♀ 80 ca fi|[---] 15
iu]ṇeo 141 cesi  f 94
[---]leo 142 cau[i] | hileo 146
uoltilia ♀ 144 cs  f ♀ 155
mesio 148 ? ar  f ̣ a 140
ịuneo 151 cicoi 40
uoltilio 162 uolti[ 79
[u]oltil[i---] 163 ? ḷ[oc]ịes b 12
? ḷ[oc]ịes b 12
Corchiano lartio b 265 aruto 257
uolteo 275 aruto 266
uolθeo 276
originis ignotae letei  fileo 470*
MLF Corchiano eiṃoi ♀ d 293
Vignanello titia ♀ 302 titọi d 305
titio 309 iun|ai d 315
titio 314
Grotta Porciosa ? lartia ♀ c 339
LF Falerii Novi iuneo 220 m [ f] 213
uotilia ♀ 222 cf 213
? marcia ♀ c 227 la  f 214
? ṃarcio c 228 ma  fi 226
? uoltiọ e 224 ce  f ♀ 226
te f ♀ 229
af 234
mf 242
af♀ 242
. ] f 247
cf 249
cf♀ 249
ce]isi f 236
? uolti e 224
Carbognano- ***io 332 zextoi | fi d 330
Vallerano uoltilio 336 ? uelcei f||e ♀ 332
*****oi d 333
a
Perhaps to be read as ar p. b The interpretation of this form is unclear. c Perhaps a gentilicium rather
than a patronym c Possibly a dative. e I read a patronymic adjective uoltiọ, where previous editors have
read a genitive uolti. – Not included in this table are (1) damaged instances (MF 16, 158, 166; LF 211,
215); (2) abbreviated filiations (MF 19, 88, MLF 309, LF 235); (3) the problematic cases MF 11-12, 263;
(4) instances where previous editors have in my view erroneously presupposed a filiation: MF 152, MLF
354, Cap 388, 390.
Fig. 7.3. FILIATION in the Middle and Late Faliscan inscriptions.
234
THE ONOMASTICON
patronymic adjective FATHERGEN [filius/filia]
LtF Falerii Veteres ce ᛍ f 171

cf 172
af 174
Falerii Novi mf 231
qf♀ 231
Corchiano ... f a ♀? 300
ia f 301
Fabbrica di Roma af 325
mf♀ 325
? [s]t f ♀ 327
Grotta Porciosa ? ia f ̣ 342
Lat Falerii Novi * [ ] f 216
kf 218
uoltai f 218
cf 219
cf 237
mf 238
cfcnb 250
uo f 251
lf 251
lf 251
Corchiano s[ex ]f 291
Cap ager Capenas cf♀ 393
st f 421
tf 456
mf 456
a
Context unclear. – Not included in this table are (1) damaged instances (MF 16, 158, 166; LF 211, 215);
(2) abbreviated filiations (MF 19, 88, MLF 309, LF 235); (3) the problematic cases MF 11-12, 263; (4)
instances where previous editors have in my view erroneously presupposed a filiation: MF 152, MLF 354,
Cap 388, 390.
Fig. 7.4. FILIATION in the Latino-Faliscan and Latin inscriptions.
It would be interesting to know if this shift in usage in the formula of filiation was also
a shift in the expression of ethnic identity. The use of the patronymic adjective appears
to have been associated exclusively with Faliscan inscriptions, and may well have been
regarded as distinctive. In the sense that in the period after 240 the choice for FATHERGEN
[filius/filia] could be associated with adhering to a (Roman) Latin formula rather than to
a local standard, it can be regarded as a change in expressing identity at least in the use
of the onomastic formular. I doubt, however, whether this shift was very great or of
very great importance, as the use of FATHERGEN [filius/filia] was already well-established
within Faliscan itself.
235
CHAPTER 7
7.6. The names of freedmen and freedwomen
The texts from the ager Faliscus and Capenas contain several examples of freedmen
and freedwomen, some explicitly designated as such. As was discussed in §2.3.2,
there is no way of concluding whether the status of the Faliscan freedmen and
-women corresponded more to that of the Latin libertus or that of the Etruscan lautni.
The first four of the following instances have also been discussed by Rix (1994:94-6).
In the Middle Faliscan inscriptions, two women are explicitly designated as
freedwomen. The first occurs in MF 41, which consists of two inscriptions separated
by a triple interpunct, the first reading fasies  c[ai]sia | louci  teti  uxor ‘Caesia Fas-
sia, wife of Lucius Tettius’, the second, loụṛia | [l]oifirta ‘Luria, freedwoman’. The
freeborn woman is designated by PRAENOMEN GENTILICIUM [HUSBANDGEN WIFE], the
freedwoman by GENTILICIUM, but as it is not uncommon for a woman to be desig-
nated by GENTILICIUM in a sepulchral inscription (16 instances out of a total of 59), I
doubt whether this difference is in fact significant. It is useless to speculate whether
Luria was a freedwoman of Caesia Fassia, or, if not, what the relationship was between
the two.119
The second instance occurs in LF 221, which is a special case: uipia  zertenea 
loferta | marci  acarcelini | mate  he  cupa ‘Vibia Sertinia, freedwoman, mother of
Marcius Acarcelinius, lies here’. Here the freedwoman is interred in the tomb because
she is the mother of Marcius Acarcelinius, who lies buried in the same tomb, and who
apparently had become an important man in Falerii Novi. As has been suggested, the
gentilicium Acarcelinius may well be a new formation: the gens Sertinia may have
been the gens to which Vibia belonged when a slave rather than her original gen-
tilicium. Note that here the woman is designated by P RAENOMEN GENTILICIUM and a
‘reversed filiation’, as she does not owe her status and burial in the family tomb to her
father, but to her son.
Two other instances are less clear, and consist of fragmentary texts with the word
lo, which could be interpreted as an abbreviation of loferta. These are ṭị [] ṭịria
lo[?---]|l[e]a  cs  f MF 155 and [---]*i  u[o]ltiai lo MF 165. Of these, the first appears
to show another freedwoman with PRAENOMEN GENTILICIUM, although the text has also
been read as ṭị[n]ṭịria with GENTILICIUM only.
Further examples of freedmen and freedwomen designated as such can be found
in the Capenate dedications from Lucus Feroniae, where we find an a[rria] or sal |
plaria  t  l Cap 431 (see below), m  t  u  genucilio  sen  l Cap 435, and [---]rcius  l 
l Cap 436.
119
Peruzzi (1964b:140-2) implausibly connected these inscriptions to the role played by liber-
tae in Bacchanalia-upheaval, which, according to Livy (39.8-19), also affected Falerii.
236
THE ONOMASTICON
In Etruscan, freedmen and -women could also be designated by double gentilicia,

the second being the gentilicium of the former master, as has been discussed exten-
sively by Rix (1994:97-111). Of double gentilicia, too, there are at least two examples
from the ager Faliscus, uel [] uisni  olna MF 82 and m  tito  tulio  uoltilio  hescuna
MF 346. An isolated instance of a freedwoman apparently designated only by two gen-
tilicia is a[rria] | plaria  t  l in Lat 431 from Lucus Feroniae, but this depends on the
restoration, which is very dubious: the text has been read in a entirely different way by
Torelli (1941:741-6), and in his reading the name is sal | plaria  t  l, with a well-
attested praenomen Saluia.
7.7. The praenomina
7.7.1. The praenomina attested from the Middle Faliscan period onward. The
praenomina that are attested in the Middle Faliscan, Late Faliscan, Latino-Faliscan,
and Latin inscriptions from the ager Faliscus and Capenas are presented in the follow-
ing list (for the Early Faliscan inscriptions, see the list in §7.2.2).
I have included all abbreviations that can be considered praenomina, as ex-
plained in §7.1.2: contextless abbreviations are included between [ ], but only if the
abbreviations are attested elsewhere in a context where they clearly represent
praenomina, or if they can easily be matched with an existing praenomen. Names oc-
curring only in Latin inscriptions from the area are included as lemmata between [ ];
names occurring only in Etruscan inscriptions have been included only if the same
name or an obviously related name is attested in the Faliscan onomasticon. As the
data for the Sabellic onomasticon are relatively few, they have only been noted when
cognates or derivations are attested, not when they are not. Note that for ease of refer-
ence I have used the closest Latin equivalent as the header to the lemma (except in the
case of abbreviations) and ordered the lemmata according to the modern alphabet.
1. A. abbr., see Aulus.
2. ? Acr-. ạcṛẹẓ MF 67 (gen.?), acre MF/Etr 279. – G. Giacomelli (1963:172) hesi-
tatingly classed this name as a praenomen, in which case it is probably connected
with Etruscan Acri (acri Pe 1.871 (perhaps a gentilicium?), acriś Pe 1.86, Pe
1.928, acrial 1.1242): such a praenomen must have formed the base of the Latin
patronymic gentilicium Acrius. Hirata (1967:32) classed it as a gentilicium: see
§7.8.1.6.
3. Aemus. m. eiṃoi MLF 293 (gen. or dat.).
4. Aemius. m. aịṃiosio EF 467* (gen.). Pace G. Giacomelli (1963:173) and Hirata
(1967:33) this is a praenomen rather than a gentilicium.
237
CHAPTER 7
The abbreviation ạ[i?]ṃ MF 89 (either a praenomen or a patronymic adjective)

can represent either Aemus or Aemius.
The praenomen itself is not attested for any other language of ancient Italy except
Venetic ·‹a›·i·mo·i· Le 26, although Aemius must have formed the basis of the old
Latin patronymic gentilicium Aemilius, and an abbreviated gentilicium aí ̣m is at-
tested for Samnitic in m  t  g  aím  ḥ[n] tSa 15. The origin of the praenomen is
unclear: G. Giacomelli (1963:173) and Hirata (1967:33) suggested that it may
have been Etruscan.120
5. Aim. abbr., see Aemus.
6. ? Am- or Amm-. ame 280, 282 (without context). Hirata (1967:34) classed the
name as a praenomen, pointing to the Latin (patronymic) gentilicium Ammius. G.
Giacomelli (1963:173) regarded it as a gentilicium: see §7.8.1.12.
7. ? Ancus. m. very dubiously in [---?] anco ma LF 232. According to Salomies
(1987:20-1, in Latin the praenomen occurs only in the names of Ancus Martius
and of an Ancus Publicius from the time of Tullus Hostilius (Dion. 3.43.3). As
there are no attestations at all for Etruscan or the Sabellic languages, this makes it
even more unlikely that it should occur here. Its origin is either Etruscan or Sa-
bine: note also the Sabellic (?) form ”A(m)puj appearing (together with Tarp…nioj)
in the list of kings in Cod. Vat. 1307 (Conway 1897:48).
8. Ar. abbr., see Arruns.
9. Arruns. m. ar]uto MF 169 (gen.), aruz MF 257, aruto MF 257 (gen.), aronto MF
265 (gen.), aruto MF 266 (gen.), arθ[3-5]rẹ MF/Etr 267, arute MF 269 (acc. used
as nom.?). Cf. also Etruscan arnθial uṛ[4-5?] Etr XXVII. The abbreviation Ar.
that may perhaps be read in MF 140, 168 probably represents this praenomen.
Derived from this praenomen are the Faliscan gentilicia Arruntulus and Arrun-
tielius (see §7.8.1.20-21). The curious nominative arute MF 269 has been re-
garded as showing epenthetic [-e] added to an unusual word-final consonant or
cluster and as an accusative arute(m) used as a nominative: see §9.2.2.4. – An al-
most emblematically Etruscan praenomen attested from all over the Etruscan area
(numerous attestations in ET). In Latin literary sources it is always the name of
Etruscans: in Latin epigraphic sources, it appears outside Etruria only in ] ar
uesc[---] CIL III.10444 from Aquincum. The name is also attested for Oscan, as
arút Po 47. Apart from these attestations, it formed the base of the patronymic
gentilicium Arruntius that occurs both in Latin (e.g. [arr]untia[e] · c · l · erotidi
CIL XI.3189 from Vignanello) and in Oscan (arruntiis Po 58, arrontiej tLu 1).
120
The only Etruscan counterpart referred to by G. Giacomelli (1963:173) and Hirata
(1967:33), however, the gentilicium Eimi, does not occur in the indices to ET.
238
THE ONOMASTICON
10. At. abbr. m. at Cap 391 [without context also in LtF/Lt 294]. Perhaps Atta, Attus
or Attius (Stolte 1928:301), for which see Salomies 1987:21. This name is often
taken together with Appa, Appius: this name is read by G. Giacomelli (1963:176)
and Hirata (1967:36) in the contextless apa Cap 457.
11. Au. abbr., see Aufilus/Oufilus and Aulus.
12. Aufilus/Oufilus. m. oufilo MF 470*, aufilo MLF 348, perhaps also ohi*[..]
LtFLat 300? Derived from this is the gentilicium Aufilius/Oufilius (see §7.8.1.25).
Abbreviations: (1) Au., see under Aulus; (2) A., see under Aulus; (3) O., dubious
reading in MF 13. – A corresponding praenomen occurs in Etruscan as Aufle (4
attestations in ET) and Afle (6 attestations in ET) from Perusia: Rix (1965:66)
suspected that Aufle in fact rendered the Faliscan name. Latin had an (unrelated?)
name Ofillus or Ofellus (Salomies 1987:91), as well as patronymic gentilicia
Aufilius/Aufillius and Ofilius/Ofillius, see §7.8.1.25. There are no correspondents
or derivations in the Sabellic languages, except perhaps the Umbrian gentilicium
ufeŕie[r] Um 8, which would correspond to *Ofidius. Salomies (1987:91) also
points to Oscan ụ́pfals Cp 2, upfals Cp 3 etc., which occurs once as úff ̣[alleís]
Fr 1, but this name appears to be unrelated.
13. Aulus. m. The praenomen is found in the area in full only in Etruscan auvilesi Etr
VIII: note also a]ụṿileś feluskeś Vn 1.1 from Vetulonia, which according to Poc-
cetti (1997) may be ‘Aulus the Faliscan’. Possible abbreviations are (1) Au. Cap
459; (2) A. MLF 206, LF 234, 242, Lat 174?, 325, Cap 389: both abbreviations
could also stand for Aufilus. – This praenomen appears to be limited to the in-
scriptions from the period after c.240 (see §7.10.5), implying that its occurrence
in the area was due to Latin influence, even though it is of Etruscan origin. It was
not very frequent in Etruscan (ET gives slightly over 40 instances, and c.30 in-
stances of the abbreviation av) or in Latin: according to Salomies 1987:24-5, c.3-
4% of Roman men were called Aulus at any given time, with a maximum of c.6%
(c.10% in Etruria). It may be attested for Oscan in the abbreviation avl Sa 28.
14. C. abbr., see Gaius and Gauius.
15. Ca. abbr., see Gaius and Gauius.
16. Cau. abbr., see Gauius.
17. Caelius. m. celio LF 332; cailio 358; perhaps also unclear ce[---] LtF 231? Ab-
breviations of this name may be (1) Ce. m. LF 213, 226, 235, LtF 171 [and with-
out context in MF? 30, MF? 68, MLF 320, 321], although this could also be an
abbreviation of Caesius; (2) Cl. in cl  anu Cap 397. Caelius occurs also as a gen-
tilicium (see §7.8.1.31). – Both the praenomen and the gentilicium are (patro-
nymic) derivations of the Etruscan praenomen Caele (caile Vc 7.24, Vs S.4,
239
CHAPTER 7
kailes Cs 2.3). Latin had only the gentilicium: the praenomen occurs only (in its
Etruscan form) in the story of Caele Vibenna (Varro L 5.8). The praenomen is at-
tested for Oscan in the abbreviation kaíl Fr 1.121
18. Caesius and Caesula. m. kaisiosio EF 7 (gen.); caisioi MF 20 (gen.), ḳai[s]i[o
MF 51, cesi MF 94 (gen.), cais[io] MF 153, cẹsi MF 263? (gen.?), ceisio MF
276; caisio MLF 351; cesio LF 331; perhaps also fragm. ce[---] LtF 231?; f.
c[ai]sia MF 41. Either m. or f. is cẹịṣ[i.] MF 140. Indirectly attested in the patro-
nymic adjective or gentilicium cesilia MLF 211. Abbreviations of this name may
be (1) Ce. masc. LF 213, 226, 235, LtF 171 [and without context in MF 30, MF?
68, MLF 320, 321], although this may also be an abbreviation of Caelius; (2) [Cs.
without context in MF? 204, an abbreviation that may expressly have been chosen
to avoid confusion with Caelius]; (3) K. occurring in Lat 218, Cap 388, 390, 404.
This is of course the abbreviation of the Latin praenomen Kaeso, but it may well
have been used for the Faliscan name that came closest.122 – The name is proba-
bly originally a patronymic adjective of a *Caesus that also seems to underly the
(diminutive) praenomen cesula LF 229 (cf. also Kajava 1995:36). Etruscan, too,
had a praenomen Caisie (kaisie Cr 3.14, kais̀ ies̀ Vs 2.1; Latin had Kaeso, which
may have had a different origin (Salomies 1987:26-7), although Caesia occurs as
a women’s praenomen in ceisia · loucilia CIL I2.559 from Praeneste (but cf. Ka-
java 1995:36) and the existence of Caesius is indirectly attested by the Latin pat-
ronymic gentilicium Caesilius. Similar gentilicia are attested for Paelignian (cai-
sies Pg 27) and Oscan (kaísillieís Cp 25, caisidis Lu 51). The unrhotacised s is ei-
ther a case of a name preserving an archaic form or a simplification of /ss/.
19. Ce. abbr., see Caelius and Caesius.
20. ? Cincus. m. perhaps cicoi in [---]o cicio  cicoi  cupat  ifra MF 40. G. Gia-
comelli (1963:88, 184-5) regarded cicoi as an Etruscan feminine gentilicium =
Cicui (cf. Etruscan cencui Cr 1.724, 1.1491). I would prefer to regard the form as
a genitive in -oi (see §4.4.4) of a father’s praenomen that would apparently be
*Cincus, related to the gentilicium Cincius that is attested for Faliscan in cincia
MF 135. Not also the Etruscan gentilicium cẹncu Etr XXI: see §7.8.1.38.
21. Cl. abbr., cl  anu Cap 397. Cl. may stand for Caelius, like Cs. stood for Caesius.
G. Giacomelli (1963:185) and Hirata (1967:45) read the text as a gentilicium
clanu.
121
Vetter (1953:301-2) read [c]elio : cailio in MF 90, which forced him to render the
praenomen as ‘Gellius’ rather than as ‘Caelius’.
122
As Kaeso was already becoming rare in Latin (Salomies 1987:26-7), the relatively frequent
occurrence of K. in the area can be explained by assuming that either Kaeso was regarded (or
used) as a Latin counterpart of Caesius, or that the abbreviation K. was used for Caesius.
240
THE ONOMASTICON
22. Cs. abbr., see Caesius.

23. F. abbr. m. f MF 58?, MLF 352, Cap 392. Already Garrucci (SIL 813) interpreted
the f in Cap 392 as an abbreviation of Fertor, comparing the gentilicium fertrio in
Cap 391 (§7.8.1.60). Salomies (1987:71) agrees with this, quoting also f  grecia
CIL I2.350 from Praeneste, and rejects the F(austus) originally proposed by Hen-
zen (1864:146).123 Fertor is only attested in the name of Fertor Resius, king of the
Aequicoli: see Salomies 1987:102 and Ámpolo 1972.124
24. Gaius. m. kaios EF 4. (I do not adopt Bormann’s reading (CIL XI.3162b,5) caio
in LF 332.)
25. Gauius and Gauia. m. cau‹i›o MF 12, caui MF 42 (gen.), cauio MF 49, caui[o]
MF 50, caui MF 84 (gen.), ca[u]io MF 146, cau[i] MF 146 (gen.), cauio MF 197,
cauio MF 207, cauio MF 210, cauio MF 272, cauio MF 275, caui MF 273 (gen.),
cauios MF 471*, cauio MF 473*; cauio MLF 308, cauio MLF 318, cauio MLF
324, cauio MLF 349, cauios MLF 382, caui MLF/Cap 474* (gen.); cauio LF 336
(falsum?); f. cauia MF 42, cauia MF 80; cauia MLF 314, cauia MLF 360, cau[ia
LF 222, cauia LF 223, cauia LF 242, cauia LF 334. Cf. Etruscan cavies Etr XLV.
Abbreviations of these praenomina are: (1) Cau. m. or f. Cap 459 [and without
context in LtF 277, Cap 398]; (2) Ca. m. MF 15, LF 224, LF 225, LtF 231, fem.
MF 14, 81 (twice), LF 220, 225, Cap 458, m. or f. MF? 38, MF 57? [and without
context in MF? 133, MLF 323, Cap 405, 406, 407, 408.]; (3) C. m. MF 200, MLF
211, LF 213, c[ LtF 215, 249 (twice), 340, Lat 218, 219, 237, 238, 250, 251, 291,
Cap 387, 393, 395, 456, probably m. Cap 396, 400, 419, 424, 427, 429, m. or f.
LtF 236, LtF 172, 173.
For ease of discussion, I treat these praenomina together although I very much
doubt that they were related, let alone identical. Authors who regard them as iden-
tical (e.g. G. Giacomelli 1963:182-3) usually assume that the name was originally
/gāo-/ (perhaps related to gaudeo ← PIE */ǵeħ2uidh-/) and that this somehow
lost its //, although there appears to be no regular development to which this loss
can be ascribed. Salomies (1987:29) follows LHS I p.138) in assuming that
/gāo-/ or /gāio-/ was a derivation of a */gāo-/ (which in itself is perfectly fea-
sible): this would then have become */gāo-/ by the same process by which
Gnaeuos became Gnaeus, and this */gāo-/ was then reformed to /gāo-/ → /gāo-/
Gaius (apparently not to the rather more expected */gāo-/, since this would have
given *Gaeus).
123
Salomies erroneously ascribes the interpretation Fertor to Vetter (1953:328) and the inter-
pretation Faustus to Degrassi (ILLRP 1233).
124
Just like the Faliscans (see §2.2.3), Fertor Resius is named as the source of the ius fetiale in
CIL VI.1302, Lib. Praen. 1 (where this attribution is ascribed to Varro), and Vir. Ill. 5.4.
241
CHAPTER 7
The evidence, however, does not support this theory, for whereas Gnaeuos was
still used as an archaic form (attestations in Salomies 1987:29-30), there is no
trace at all of either *Gauos or *Gaos, in spite of Gaius being by far the more fre-
quently used praenomen;125 furthermore, the attestation of Early Faliscan kaios
EF 1 and the Etruscan derivations of Gaius (see below) would place this whole
process before the sixth century, and therefore well before the loss of // in
Gnaeuos. However, whether etymologically related or not, the distribution of
Gauius and Gaius in the agri Faliscus and Capenas indicates that they may well
have been regarded as equivalents, as is discussed below.
The praenomen Gaius is in all probability of Latin origin, and its attestations are
likewise almost exusively Latin, apart from one instance in Umbrian (cais Um
23) and one in Oscan (cạis Fr 10). Etruscan has a frequently attested gentilicium
Cae or Cai, which is probably derived from the Latin praenomen (cf. Rix
1965:217). Gauius on the other hand is usually associated with the areas where
Sabellic languages were spoken: it occurs in South Picene kaúieis AQ 3 and in
Oscan gavis Hi 10, ga[vis Cm 14,3, gaavị[eís] Fr 1, [g]a#ij Lu 45, [g]a#[ij] Lu
63, (g)a#im Lu 47, ga#in Lu 46, Samnitic [ga]avíeís Sa 14 (for the abbreviated in-
stances, see ST). There are also a number of instances from Etruscan, virtually all
from Southern Etruria (kạṿị[e Cr 2.56, kavie Cr 5.1, cavies Cr 2.74, kaviiesi AT
3.1 kaṿ[ies Vs 1.99, kavies Vs 1.159, cavies Fa 2.25=Etr XLV, kavias OA 2.11,
cavias OA 2.52; from Northern Etruria are cavial Vt 1.124 and caviaś Fe 2.15).
Yet the number of Faliscan instances of Gauius is in fact greater than that of the
Sabellic and Etruscan instances put together: perhaps Gauius should be regarded
as a common Central Italic, perhaps even Faliscan praenomen, rather than a Sa-
bellic one. Although it is certainly not unique to the ager Faliscus in the way Iuna
and Volta are, its frequency in the area allows it to be classed as a Faliscan
praenomen and may have played a role in establishing ethnic or cultural identity.
This is made even more feasible by the distribution of Gaius and Gauius. Gaius is
absent from the Middle and Late Faliscan onomasticon, while Gauius is the most
frequently attested Middle and Late Faliscan praenomen: on the other hand, there
is no trace of Gauius in the Latino-Faliscan and Latin inscriptions, where the ab-
breviation C. is the most frequently attested praenomen. This can be explained ei-
ther by assuming that the abbreviation C. was used for Gauius as well as for
Gaius (thus G. Giacomelli 1963:178), or, as I would prefer, by assuming that
Latin Gaius and Faliscan Gauius were regarded as equivalents (whether this was
etymologically justifiable or not is irrelevant). Someone called Gauius could then
125
Salomies (1987:29) calculates that c.20% of Roman men were called Gaius at any given
time, while describing Gnaeus as “In der republikanischen Zeit ebenso selten wie Aulus”
(1987:30), i.e., c.3-4%, with a maximum of 6%.
242
THE ONOMASTICON
just as easily give his name as Gaius and abbreviate this as C. when adapting to or
adopting the framework of Latin: see §7.10.5. With the spread of Roman influ-
ence, this may ultimately have lead to the disappearance of Gauius.
26. Graecus m. kreco  [---] MF 147. The name may have been a nickname, but the
occurrence of Graecus and Gr. elsewhere (Latin CIL I2.336, 3279) rather implies
that it was an older existing praenomen, perhaps Etruscan (Salomies 1987:71-2).
The Etruscan attestations of Craic-/Creic- (ET gives 29 attestations) are all of
gentilicia, which in itself could be in an indication that the praenomen existed if
these gentilicia were patronymic (note the patronymic gentilicium creicnal Ar
1.4). The name also occurs in Venetic (gra·i·ko·i· Le 77).
H- see also F-
27. ? Her-. A very dubious attestation in cẹsit  fere MF 263 (with hypercorrect f). G.
Giacomelli (1963:192) and Hirata (1967:51) compared the Latin gentilicium
Herius and various Etruscan gentilicia in Fer-; Salomies (1987:73) includes it in
her discussion of the praenomina Herius and Herennius. The interpretation of the
inscription is debated, note that fere does not occupy the position of a praenomen.
28. Iantus and Ianta. f. ian[ta MF 146, iata MF 147, iata MLF 362. Abbr. Ia. m. MF
101, LtF 301, LtF 341; f. MLF 302. – The Faliscan instances are almost the only
attestations of a very infrequently attested Etruscan praenomen: the only other at-
testations are Etruscan iantia Vn 2.7 and perhaps ianzu Vt 4.6, and possibly also
the Umbrian abbreviation ia Um 8; it must also have formed the basis for the
Latin patronymic gentilicium Iantius. Perhaps related, too, is Venetic ia·n·t··s· Le
124. G. Giacomelli (1963:195) noted that the form Iantus/Ianta is entirely based
on Herbig’s restoration (under CIE 8586) ian[ta in MF 146, while the other attes-
tations are without n: the omission of syllable-final n is so common in Faliscan,
however, that this can hardly be an argument to doubt Iant- (§3.5.7a).
29. Iuna. m. iuna MF 39, iuna MF 48, iuna MF 73, iuna MF 74, iuna MF 96, iunai
LF 112 (gen.), i[un]a MF 148, iun[---] MF 166, iuna MF 198; iuna MLF 297,
iuna MLF 298, iuna MLF 307, iun|aị MLF 315 (gen.), [iu]na MLF 363; also
iunai MLF/Cap 475* (gen.). Indirectly attested in the patronymic adjective: ịuneo
MF 151, iuneo LF 220, perhaps also iu?]nẹo MF 141. Abbr. Iu. MF 19, 157; Iun.
MF? 203, LF? 381. Derived from this is the patronymic gentilicium Iunius that
occurs once in the ager Capenas (see §7.8.1.76). – The name is exclusively Falis-
can, and has no counterparts in Latin, Etruscan or Sabellic, although the Latin
gentilicium Iunius could point to an existence in Latin as well (unless it is presup-
posed that the gens Iunia originated from the ager Faliscus.126 Iunius was derived
126
CIL I2.559 from Praeneste is sometimes quoted as an instance of Iunius used as a praenomen,
but the reading is t · iunio · setio rather than iunio · setio: see Wachter 1987:117.
243
CHAPTER 7
by H. Petersen (1962:352) from the name of the month, originally meaning ‘born
in June’, and Salomies (1987:114) assumed the same for Iuna, but in that case the
derivation is awkward. It remains in any case unexplained why Iuna and the other
uniquely Faliscan praenomen, Volta, were masculine names of the first declen-
sion, a category absent from the Latin or Sabellic onomasticon.
Since the name is uniquely Faliscan and occurs with relatively great frequency
(15 instances, which makes it the second most frequently attested praenomen), it
may well have been a carrier of Faliscan ethnic identity (see §7.10.5), which
would render its disappearance after the Middle Faliscan period all the more sig-
nificant: in the Late Faliscan inscriptions, there is only one instance of the patro-
nym iuneo in LF 220, and the name is entirely absent from the Latino-Faliscan
and Latin inscriptions. If the name did play a role in the ethnic identity of the ager
Faliscus, it may well have been regarded as connected with the name of Juno, the
central deity of the area (see §2.3.4), even though it is unlikely to have been de-
rived from it.
30. K. abbr., see under Caesius.
31. La. abbr. m. la MF 93, LtF 214 (twice) [and without context in MF 373-375,
MLF 286, MF/LtF 252, LtF 278]. It may be an abbreviation of either Laris or
Lars (see below): in MF 373-375 it is perhaps rather Laris, as these inscriptions
were found together with larise uicina MF 371 and larise | uicina MF 372, al-
though Lars was far more frequently used praenomen at least in Etruscan.
32. [? Laeuius. leuia LtF 327 (probably rather a gentilicium, see §7.8.1.81). – The
name also occurs as a gentilicium, see §7.8.1.81.]
33. Laeuilius. m. leiuelio MF 79, [leu]elio MF 90, [leu]elio MF 159. The name oc-
curs also as a gentilicium (see §7.8.1.82). – Laeuilius is a patronymic derivation
from Laeuius. The spelling ei/e represents /ę̄/ ← /a/ (G. Giacomelli 1963:199,
Hirata 1967:57) rather than the /ē ̣/ required by the connection with Liuius sug-
gested by Deecke (1888:129): note the Etruscan praenomen laives AV 2.1 and the
patronymic gentilicium laivenạ[ Ru 3.1, laiven[as] Vs 1.58. A praenomen Laeuus
occurs once in Latin, a Laeuus Cispius from Anagni at the time of king Tullus
(Fest. 476.11-2L): Latin also had a gentilicium Laeuius. The name is derived
from the adjective laeuus, like Scaeua from scaeuus: both had the meaning ‘well-
omened’ in the official sacral language: “laeua prospera existimantur, quoniam
laeua parte mundi ortus est” (Plin. NH 2.142), “scaeua, id est sinistra, quod quae
sinistra sunt, bona auspicia existimantur” (Var. L. 7.97). There are no attestations
of related names from the Sabellic languages, except perhaps for the abbreviated
gentilicium laí Sa 5.
244
THE ONOMASTICON
34. Laris. m. larise MF 270, laris MLF/Etr 290, larise MF 371, larise MF 372;
Etruscan larisa Etr XXXII, perhaps also lar*s Etr XIX. – An Etruscan
praenomen that is frequently attested from various locations (see ET). Absent
from the Sabellic languages (except perhaps for the abbreviation la Um 27), and
attested in Latin only indirectly in the patronymic gentilicium Larisius. The -e in
several of the Faliscan forms (see also under Arruns), may be an epenthetic [-e]
added after an /s#/ that was realized in more strongly than the weak Faliscan /s#/
(cf. §3.5.7d), perhaps reflecting an Etruscan pronunciation (§9.2.2.1,4).
35. Lars. m. lartos EF 6 (gen.). Indirectly attested in the patronym lartio MF 265.
Derived from this is the patronymic gentilicium Lartius (see §7.8.1.78). [Also
larθ Etr XXXIV, XXXV, XXXIX, also lazi Etr XI-XV, and lazia Etr XVII.] –
An emblematically Etruscan praenomen (for the numerous attestations, see ET).
In view of its frequency in Etruscan and its occurrence in the Etruscan inscrip-
tions from the ager Faliscus and Capenas, the number of Faliscan attestations is
surprisingly low. The Latin attestations are limited to inscriptions from Etruria
and the name of the consul, Lars Herminius (Salomies 1987:32), although Latin
had the gentilicium Lartius. The praenomen is not attested for the Sabellic lan-
guages, except perhaps for the Umbrian abbreviation la Um 27.
36. Lucius and Lucia. m. ḷ[oc]ịes MF 12?, louci MF 41 (gen.); f. locia MLF 293.
Abbreviations of this praenomen are probably (1) Lo. m. MF? 33 (In ṭị [] ṭịria
lo[?---]|l[e]a  cs  f MF 155 and [---]*i  u[o]ltiai lo MF 165, lo in is interpreted
rather as loferta = liberta); (2) L. m. LF 230, Lat 218, 251 (four times), 477*, Cap
428, 432, f. Cap 436. – Apparently a Latin praenomen that was frequently used in
Latin127 and occurs also in the Sabelllic languages (Praesamnitic luvcies Ps 13,
Umbrian vuvçis TI Ib.45, IIa.44, and vuvçiia TI IIb.26, and Oscan lúvkis Cp 36
etc. (12 instances in ST); ST furthermore gives over 50 instances of the abbrevia-
tion l in Marrucinian, Vestinian, Paelignian and Oscan). Borrowed into South
Etruscan inscriptions as Luvce and Luvcie (luvc ̣e AT 5.2, luvce Vs 1.282, luvces
Ta 1.220, luvciies Ta 7.31, luvcies Cr 2.139, Vc 6.12; f. luvci AT 1.102, luṿci AH
2.3, luvcia Ta 1.149, luvcial Ta 1.75). In the ager Faliscus it is not very frequent,
and it appears to be associated predominantly with the Late Faliscan and Latin in-
scriptions (see §7.10.5).
37. M. abbr., see Marcus and Marcius.
38. Ma. abbr., see Marcus and Marcius.
39. ? Maesus or Maesius, see below under Messus or Messius.
127
According to the calculations by Salomies (1987:34), c.20% of Roman men were called
Lucius at any given time.
245
CHAPTER 7
40. Mar. abbr., see Marcus and Marcius.

41. Marcus. m. [ma]rco MF 80. Indirectly attested in the patronymic adjective mar-
cio MF 80; marcia LF 227 (perhaps a gentilicium?), ṃarcio LF 228.
42. Marcius. m. marci LF 221 (gen.), maci LF 222 (gen.), marcio LF 223. (These
instances all refer to the same person, the son of a freedwoman: the praenomen
may simply be the patronymic adjective derived from his father’s name.)
Either Marcus or Marcius (probably the former): marci MF 472* (gen.). Abbre-
viations of these praenomina are: (1) Mar. in MF 87, Cap 421; (2) Ma. LF 226
(probably Marcius), 242, LtF 232, 301; M. in MLF 346, 355?, LF 213, 378, LtF
231, 233 (twice), LtF 239, 325 (twice), Lat 237, 238, Cap 420, 435, 456 (twice).
The abbreviations might conceivably belong to other praenomina.
A Latin praenomen, probably connected with the theonym Mars (thus Lib. Praen.
5, cf. Mamarcus : Mamars). It is well-attested in Etruscan inscriptions: ET gives
over 30 instances, nearly 20 of which from southern locations, especially Caere. It
is not attested from the Sabellic languages, although it is unclear what praenomen
is abbreviated by the abbreviations M. and Ma. that occur in Umbrian, Volscian,
Paelignian and Oscan texts (see ST: Oscan Mar. probably stands for Maras, but
might also conceivably stand for Marcus). In the ager Faliscus, the instances of
Marcus are almost all from the Late Faliscan, Latino-Faliscan and Latin inscrip-
tions, which might imply that its use was due to Latin influence: see §7.10.5.
43. ? Messus or Messius m. mesio in iuna lẹ[---] | mesio MF 148. I think it very
unlikely that mesio is a cognomen, as G. Giacomelli (1963:205) and Hirata
(1967:61) suggested (see §7.9): it is probably a second name. The absence of rho-
tacism suggestes that the s represents /ss/, in which case Messius is an obvious
candidate: this could either be a praenomen Messius or a patronymic adjective
from a praenomen *Messus (cf. Salomies 1987:127), or a gentilicium Messius as
in CIL XI.3782 from Veii. Alternatively, mesio could be a patronymic adjective
derived from a praenomen *Maesus, cf. the (patronymic?) gentilicium Maesius,
occurring both in Latin and in Oscan (maisim Lu 47).
44. Minor. f. mino LtF 173 (uncertain). Kajava (1995:48) cites 8-9 instances of this
praenomen, all from Praeneste, critically discussing (1995:118-124), but not re-
jecting, the traditional view that praenomina like Minor and Paula were used to
distinguish between sisters. Such a usage would be necessary if women did not
have ‘real’ praenomina: the fact that the name occurs here in a Latino-Faliscan in-
scription may reflect a Latin custom rather than a Faliscan one, connected with
the disappearance of women’s praenomina in Latium, which appears to have been
less pronounced in Faliscan.
45. Nu. abbr. m. nu MF? 202, MLF 309. The instances are both doubtful: if they are
indeed abbreviated praenomina, Numerius or a related name is the obvious candi-
246
THE ONOMASTICON
date. This praenomen occurs in Latin as Numerius, and with some frequency in
the Sabellic languages, both as Num(e)sis (e.g. Umbrian numesier Um 38, Oscan
ṇịụṃsis Cm 14,2, [n]iumsis Cm 6, niumsieís Cm 6, numyim Lu 46) and as
Nom(e)sis (e.g. nomyij Lu 47, noyim Lu 46). Etruscan has both the praenomen Nu-
mesie (numes Cm 2.48 (abbreviated?), numisiies Cm 2.8, numesiesi Ta 3.1, nume-
sia Vs 1.268, numuσ́ieś Ar 1.13), and the gentilicia Numsie and Numsina: the lat-
ter is also attested for the ager Faliscus (nomesina MF 272): see §7.8.1.107. Cf.
also the theonym Mars Numesius in LtF 377 from Ponzano Romano and Cap 421
from Capena. The origin of the name is debated, and may be either Italic or Etrus-
can (see Salomies 1987:39), probably rather the former (De Simone 2006:170-3).
46. O. abbr. m. in o hạθ̣i MF 13? Doubtful. See Aufilus/Ofilus.
47. Oct-. m. A dubious attestation in oct*i[....] uoltilio MLF 353, where oct is proba-
bly a numeral praenomen like Octauus,128 in Latin one of the rarer numeral
praenomina. In Oscan, only Ohtavis is found as a praenomen (ota#ij Lu 63), a
(patronymic) derivation of an unattested *Ohtaus. The existence of this *Ohtaus
is implied also by Etruscan Uhtave (usually used as a gentilicium, but apparently
as a praenomen in uhtaves Pe 1.817, ụḥtaṿes Pe 1.1267), where the h clearly
points to a Sabellic origin.
48. Paquius m. pa‹qu›is in paụqis blaisiís Sab 468*. This praenomen is in all prob-
ability the one represented by the abbreviation Pa. in Cap 390. A Sabellic
praenomen attested for Oscan in pak#hij Lu 40 (and pakkuiis Si 19?), and in ab-
breviated form in Marrucinian paq MV 8, pa MV 9: the Sabellic languages also
had several closely related praenomina (Oscan pạakiu Po 87, pakthij Lu 23, paa-
kul Cm 7) and gentilicia (Paelignian pacia Pg 4, Oscan pakidiej Lu 63, pakthij Lu
23, pakulliis Cm 14,5). The praenomen occurs in Latin inscriptions virtually only
in Central and Southern Italy (attestations in Salomies 1987:84), and it appears to
be unattested for Etruscan. Both Faliscan attestations are from inscriptions that
show Sabellic epigraphic and onomastic features.
49. Paula. f. pola LF 227, po[l]ae Lat 251. Kajava (1995:50-59) quotes numerous
instances of Paula/Paulla and Pola/Polla as a woman’s praenomen, and critically
discusses (1995:118-124) the traditional view that praenomina like Minor (above)
and Paula were used to distinguish between sisters. As in the case of Minor, the
attestations of Paula are from the Late Faliscan and Latino-Faliscan inscriptions,
perhaps implying that these praenomina were due to Latin influence, where such
praenomina became increasingly necessary as the custom of giving women ‘real’
praenomina diminished. The name is spelled tith o so soften that the form
128
Herbig (CIE 8204) in fact read the Faliscan attestation as oc ̣ṭọ, referring to Schulze’s re-
marks (1904:21) on Octo, where the latter discussed Celtic origins of Latin gentilicia.
247
CHAPTER 7
Pola/Polla may well have led a live independent from the adjective paula/paulla
(cf. §3.7.4).
50. ? Petro. m. A very dubious attestation in petr]ono MF 102 (gen.). A Sabellic nu-
meral praenomen (‘nordoskisch’ according to Salomies 1987:85-6), attested in
South Picene petroh TE.1 and in a number of Latin inscriptions from Central Italy
(see Salomies 1987:86). It also formed the base of the gentilicium Petronius at-
tested for South Picene (pet{i}eronis AP.5), Marrucinian (petroni MV 3), Paelig-
nian (ptruna Pg 52), and Latin, which occurs also in the ager Faliscus (see
§7.8.1.119). Etruscan has well-attested gentilicia Petru (more than 120 attesta-
tions) and Petruna, Petruni (more than 40 attestations).
51. Publius and Publia. m. popli[o] MLF 316, perhaps popl[---] MLF 317; f. poplia
MF 18, poplia MF 48, poplia MF 50, poplia MF 136, [p]ọpliạ MF 160, poplia
MF 265, poplia MF 271, poplia MLF 303, popliai MLF 308 (gen.), poplia LF
332. Abbreviations (1) Pop. f. in LF 226; (2) P. LF 337?, Lat 250, Cap 409, 462
(although this might conceivably stand for a name other than Publius, this is
unlikely in the case of the Latin and Capenate attestations). Surprisingly, in the
Faliscan inscriptions this praenomen appears to have been used almost exclu-
sively for women, while its use for men appears to be associated with the Latin
and Capenate inscriptions. – The praenomen is either of Latin or of Etruscan ori-
gin, perhaps rather the latter, although the Etruscan instances of the praenomen
are few (puplies Vs 1.29, pupli Cl 1.2079, 1.2080, 1.2344, pupli{na} Cl 1.2109,
perhaps also pup[liś] Cl 1.2179). In Latin, the praenomen was fairly well at-
tested, although not particularly frequent.129 There are no attestations for the Sa-
bellic languages, although these may be hidden in abbreviations such as po and p
(see ST for instances of these abbreviations).
52. ? Pumponius or Puponius. m. puponio in puponio  firmio MF 54. G. Giacomelli
(1963:214) and Hirata (1967:70) regarded the name as a gentilicium, but it is ap-
parently used as a praenomen here. It could conceivably be a patronymic
praenomen derived from the Sabellic praenomen Pompo (South Picene
pomp[úne]í AQ.2, Oscan pu(m)puf Cp 42), but u seems to point rather to an
Etruscan origin, e.g. the gentilicia Pumpu/Pupu and Pumpuni/Pupuni (for attesta-
tions, see ET). Latin had the gentilicia Pomponius as well as Puponius.
53. Pupia. f. pupiias MLF 304 (gen.). The name occurs in isolation and can be either
a praenomen or a gentilicium: a related praenomen Pupus occurs in Latin inscrip-
tions from Northern Italy (see Salomies 1987:129, Kajava 1995:64), and there are
several related Etruscan and Latin gentilicia (see §7.8.1.123, 124, 128, 129).
129
Salomies (1987:46) calculates the total of Roman men called Publius at 10%.
248
THE ONOMASTICON
54. Q. abbr., see under Quinctus.

55. ? Qua. abbr. [---]*[5-7] cua MF 129? If this is a praenomen at all, it looks like a
numeral praenomen such as Quartus. This makes the interpretation even more
doubtful, as the numerals 1-4 do not seem to have been used as praenomina dur-
ing this period in Latin except in Northern Italy: see H. Petersen (1962:348-50)
and Salomies (1987:111-2, 118).
56. Quinctus. m. cuicto MLF 310. Abbr. Q. LtF 231. A Latin numeral praenomen,
not attested in Etruscan (except in the gentilicium Cvinte, rendering a Latin name)
or the Sabellic languages, although the corresponding Sabellic form Pompt- oc-
curs as a (patronymic?) gentilicium in Paelignian (ponties Pg 5) and in Oscan
(pomptiej Me 1 etc., púntiis Po 1 etc.). The fact that Quinctus occurs only in Latin
and Faliscan may be connected to the fact that these praenomina had a lexical
meaning: cf. the case of Sextus (see below).
57. Saluia. f. sal | plaria  t  l in Torelli’s reading (1974:741-6) of Cap 431. Saluius is
well-attested praenomen of Sabellic origin, occurring throughout Central Italy:
see Salomies 1987:88-90. The female Saluia was popular as a slave-name (Solin
1996:9-10, Kajava 1995:69 n.88), as it is here.
58. Scaeua. f. sceụa MLF 312, sceiuai LF 379. I regard this name as a female
praenomen (cf. Solin 1996:57), not as a cognomen, as has been suggested (e.g.
Torelli 1967:536-7): there are no attestations in the Faliscan inscriptions of COG-
NOMEN alone, nor of women having cognomina (cf. §7.9). Furthermore, the name
is a name of good omen, derived from the adjective scaeuus (“scaeua, id est sinis-
tra, quod quae sinistra sunt, bona auspicia existimantur”, Var. L. 7.97), just as
Laeuus/Laeuius/Laeuilius is derived from laeuus (see under Laeuilius). Latin had
gentilicia such as Scaeuius etc. (Schulze 1904:226-7), while Etruscan had a
probably borrowed name sceua Cl 1.1243, 2028, sceuaσa Cl 1.1045 (cf. also
the gentilicium Sceva/Scevia (9 attestations, mainly from Clusium).130 For the
forms with -f-, Etruscan scefi Pe 1.630, scẹf ̣[i Pe 1.1211, scefia Pe 1.201, and the
Paeliginian gentilicium scaifia Pg 14, see §3.2.8.
59. Sen. abbr. in m  t  u  genucilio  sen  l Cap 435. Moretti (1975:133-4) suggested
that it could be the abbreviation of a cognomen, but I doubt whether this is likely
at this date: it is perhaps an unidentified praenomen, cf. the abbreviated Samnitic
praenomen sn tSa 21, tSa 32.
60. Seruius. m. serui MF 34-36 (gen. or abbr. nom.). It is impossible to ascertain
whether serui is the praenomen or the gentilicium Seruius: both PRAENOMEN and
130
The Latin gentilicium and the occurrence of the name in Etruscan make it even less likely
that Scaeua was a cognomen.
249
CHAPTER 7
GENTILICIUM are possible in Faliscan Besitzerinschriften (§7.3). – The origin of

the praenomen is either Latin or Etruscan: Salomies (1987:47-9) decides for the
latter, in spite of the fact that the praenomen is not attested for Etruscan (but cf.
the gentilicium σ́ervei Pe 1.1191, σ́ervi Pe 1.1190). The praenomen was not very
frequent in Latin either. The name is probably not connected to the noun seruus
(which has been thought to be of Etruscan origin, cf. Bréyer 1993:383-5).
61. Sex. abbr., see Sextus.
62. Sextus. m. z[e]xtos EF 1; sesto LF 329, zextoi LF 330 (gen.). Abbreviation: Sex.
in se]x LtF 328, s[ex Lat 291, sex Cap. 399, 430. – A Latin numeral praenomen,
not attested for Etruscan (except for the indirect attestation in the gentilicium
σ́ekstaluś Sp 2.71) or the Sabellic languages, although, in view of the occurrence
of other Sabellic numeral praenomina, it may well have existed also in the Sabel-
lic languages (cf. the Oscan gentilicium sehsímbrịí ̣ṣ Po 36, derived from the name
of the sixth month, which according to H. Petersen (1962) is also the original
meaning of the name Sextus). Apart from the Early Faliscan instance of z[e]xtos
EF 1, this praenomen occurs only in the Late Faliscan, Latino-Faliscan, and Latin
texts, which could imply that its popularity was due to Latin influence: see
§7.10.5.
63. St. abbr., see Statius.
64. Sta. abbr., see Statius.
65. Stat. abbr., see Statius.
66. Statius. m. statio MLF 376. Abbreviations that probably represent this praenomen
are: (1) Stat. MF? 29? (the text is statuo, probably to be read as stat uo); (2) Sta. f.
(?) Cap 466 [without context in MF? 28, sta[ MF? 128]; (3) St. m. LtF 327, Cap
394. – An Italic praenomen that occurs with some frequency in Latin (attestations
in Salomies 1987:90-1), but is primarily known from Oscan (Samnitic statis Sa
36 etc.: ST lists 14 attestations, not counting abbreviations or the use of the name
as a gentilicium). There are no Etruscan attestations.
67. T. abbr., see under Titus.
68. Tana. f. tana MLF 338. – An emblematical Etruscan female praenomen, occur-
ring both as Θana (numerous instances, mainly from Tarquinii, Clusium, and Pe-
rusia: see ET) and Tana (only AH 1.67, Cl 1.725, Pe 1.71, 1.135). Not attested for
Latin or the Sabellic languages.
69. Tania. f. θania MF 81. – Like tana, a frequently attested Etruscan female
praenomen (ET lists 170 instances), attested for Latin in tania · p ̣ap̣ric[i] / c · f
CIL XI.2977 from Tuscania.
250
THE ONOMASTICON
70. Tanaquil. f. θanacuil MF 49, tanacu[il] MF 101, tan‹a›cuil MLF 347. – A typical
Etruscan female praenomen, derived from the theonym Θana, occurring as Θan-
(θancvil Vs 1.287, Po 4.4, θancvilus OA 2.63), Θana- (θanaχvil Ta 7.3, θan]aχvil
Vs 1.190, θana]cviḷ[us Ta 2.1, θanakvilus̽ Cr 2.42), and Θane- (θaneχvil Ta 7.31,
θanecvilus AV 2.11). Not attested from Latin or the Sabellic languages: note that
it is apparently a nominal compound Θana+cvil, a type of praenomen that is ap-
parently absent from the Italic onomasticon.
71. Te. abbr., see Tettius.
72. Tettius. m. teti MF 13. The abbreviation Te. in LF 229 probably belongs to this
name. The name occurs also as a gentilicium Tettius (see §7.8.1.153). – Salomies
(1987:93) points to the abbreviated praenomen tet in CIL X.6098 from Urbino:
there also appears to be an attestation in South Picene tetis TE.2. The name occurs
only as gentilicium in Latin, Paelignian, and Etruscan: see §7.8.1.152.
73. Ti. abbr. ṭị in ṭị [] ṭịria MF 155. Perhaps Titia? G. Giacomelli (1963:223) and
Hirata (1967:79) suggested Tiberius, based on the Roman convention T. = Titus
and Ti. = Tiberius, but there is no indication that the Faliscan abbreviations were
distributed in this way: furthermore, there are no cases of a Tiberia even in Latin.
74. Tirrus. m. tiroi MF 69-71 (gen.). Derived from this is the gentilicium Tirrius (see
§7.8.1.155). – The name is not attested in Etruscan or the Sabellic languages, and
occurs in Latin only in tirri · craisli · tir · f CIL XIV.3110 from Praeneste, where it
is unclear whether the name is Tirrus or Tirrius.
75. Titus and Titia. m. tito MLF 305, titoi MLF 305 (gen.), tito MLF 309, tito MLF
315, tito MLF 346, tito MLF 350, tito MLF 354; tito LF 226, tito LF 336, tito
359. (The instances do not include the theonym Titus Mercus, for which see §6.4)
Indirectly attested in the patronymic adjective titia MLF 302, titio MLF 307, titio
MLF 312. The corresponding female praenomen appears to have been Titia rather
than Tita, a patronymic praenomen derived from Titus: titias EF 3 (gen.), MF 201
(gen.). The abbrevation of this praenomen is in all probability T. masc. LtF 327,
Lat 216, Cap 415, 425, 431, 435, 456. – The origin of the name is debated: I fol-
low Combet-Farnoux (1980:113-69) in assuming that it was originally an Italic
adjective meaning ‘propitious, well-omened’, as in the Faliscan theonym Titus
Mercus (cf. §6.4) and the aues titiae mentioned by Varro (L. 5.81). The
praenomen is well-known from Latin, although not frequently used,131 and occurs
also in South Picene (titúm AP.1, titúí TE.5), in Umbrian (titis TI Ib.45), and in
various Sabellic derivations (South Picene titienom TE.3, Paelignian titis Pg 15,
131
Salomies (1987:57) calculates the percentage of Roman men called Titus at c.3-5%, far
less than e.g. Gaius or Lucius (each c.20%).
251
CHAPTER 7
Oscan titieis He 3 (and titti tPo 13?), titidiej Lu 15, 27). Etruscan has a fairly
well-attested praenomen Tite (the indices to ET list over 20 instances, beside a
greater number where this name is used as a gentilicium).
76. Tr. abbr. tr Cap 461. Probably Trebius.
77. Tullus and Tullia. m. tulo MF 151, perhaps also tulom MF 72 (gen. pl.?). The
corresponding female praenomen may have been Tullia rather than Tulla, a patro-
nymic adjective derived from Tullus, occurring in tulie MLF 383 (although this
may also be an instance of the gentilicium Tullius). Derived from this is the pat-
ronymic gentilicium Tullius (see §7.8.1.156). – Latin had both the praenomen and
the gentilicium, the former occurring only in tul · tullius · tul · f CIL I2.1493, 1497
from Tibur. There are no attestations from the Sabellic languages, with the possi-
ble exception of the abbreviated gentilicium tu Um 39. Etruscan has Tule, which
occurs only three times: mini tule[ Ve 3.32, fasti  kainiei  tuleσ́a  kn[ Ar 1.1, and
larθ  tule  kavinei | ṭuś(urθi) Ar 1.94. The name is either Latin or Etruscan.
78. V. abbr. u Cap 435.
79. Ve. abbr. ue MF 43?, MLF 206 [and without context MF? 284, MLF 322].
80. Vel. m. uelos EF 4 (gen. or a thematized form, see §7.2.1), uel MF 56, uel MF 82,
perhaps ue]l MF 191. Also Etruscan velusa Etr XXXIV. – An Etruscan
praenomen, attested in a very large number of attestations (mostly from Clusium,
Volsinii and Tarquinii, see the indices to ET). It is not attested in Latin or in the
Sabellic languages, although Latin has gentilicia such as Velius and Veleius, both
of which occur also in Oscan (veliieis Cm 22, veleí[is] Hi 3).
81. Velce(i)us or Velcaeus. m. uelcei LF 332 (gen.). An adaptation of Etruscan
praenomen Velχe (Cl 1.1327, 1328, Ar 1.9): like Veltur, this name preserves the
Etruscan /e/ (cf. the Faliscan name Volta).
82. Veltur m. ueltur MF 266, ueltur MLF 339. Also Etruscan velθarus Etr XVI,
velθurusi Etr XIX, [u]eltur Etr XXXVIII. – A well-attested Etruscan praenomen
(ET gives more than 80 instances from various locations, mostly from Tarquinii).
There are no attestations from the Sabellic languages, while Latin had only the
gentilicia Velthurius, Volturius, and Vulturius, which show the usual Latin devel-
opment /e/→ /o/ before velar /l/ (§3.3.4.1). Faliscan does not show this develop-
ment in Veltur, although it appears in Volta (see below), indicating that Veltur
was still an Etruscan name, in contrast to the entirely Faliscan Volta.
83. ? Venel. m. dubiously attested in ueṇe[?---]na  | ux[o(r) ?---] MF 43. Perhaps
abbreviated uen in uentar[c..... MF 80? A derivation Venelius, either a patronymic
adjective or a gentilicium, occurs in uenelịes MF 258. – A well-attested Etruscan
praenomen (the indices to ET give more than 50 instances from various loca-
252
THE ONOMASTICON
tions). Latin had only the patronymic gentilicia Venelius and Venilius, but there
are attestations of the praenomen in Praesamnitic (uenel{i}eis Ps 12, uenilei Ps 3,
patronymic adjective ueneliis Ps 3) and Oscan (uenileis Cm 30).
84. Vibius and Vibia. m. uipi MF 14 (gen.); f. uipia LF 221. – Probably a patronymic
praenomen derived from Etruscan Vipe, which is attested for the areas surround-
ing the ager Faliscus: Tarquinii (vipe Ta 1.39, 1.92, AT 1.28, vipe[s Ta 1.93, vipes
Ta 1.237, AT 1.74), Volsinii (vipe Vs 1.233, vipes Vs 1.133, 1.231), and Horta
(vipes AH 1.8). It is therefore not unthinkable that Vibius in fact originated in the
ager Faliscus. The praenomen is also attested fairly frequently for Latin (attesta-
tions in Salomies 1987:96, mostly from Central and Southern Italy and Etruria),
where Vibius also occurs as a gentilicium. In the Sabellic languages it is attested
for Umbrian (vipies Um 5, vibie Um 37), Paelignian (fem. uibia Pg 51, abbrevi-
ated uib Pg 33) and is most common in Oscan (viíbis Fr 1 etc., f. #ibian Lu 46: the
indices to ST list at least 15 instances, not counting abbreviations). Salomies
(1987:96) regards Vibius as an Oscan praenomen, but as most of the Oscan in-
stances are from Campania, the use of the name there may be due to Etruscan in-
fluence. An Etruscan origin might also explain the relative frequency of the
praenomen in Latin inscriptions when compared to other Sabellic praenomina,
and the frequency of Etruscan gentilicia such as Vipe and Vipena/Vipina/Vipiena.
85. Volta m. uolta MF 15,[u]olṭa MF 149, uol[ta MF 158; uolta MLF 313, uoltai
MLF 367-370 (gen.); uoltai Lat 218 (gen.). Also Etruscan vultasi Etr XLII. Indi-
rectly attested in the patronymic adjective uolteo MF 275, uolθeo MF 276; uoltiọ
LF 224; perhaps also in u]oltio MF 164 (this may also be an instance of the gen-
tilicium Voltius) and uolti[---] MF 167 (this may also be an instance of the gen-
tilicium Voltius, or of the patronymic adjective Voltilius).
86. Voltius. m. uolti MF 11 (gen.), uolti MF 79 (gen.), [uo]ltio MF 88, uolti[o MF
152; uolti MLF 469* (gen.), uoltio MLF 312; uoltio LF 220, uoltio LF 330;
probably also ulties MF/Etr 64, if this is to be read as u(o)lties. Indirectly attested
in the patronym adjective uoltilia MF 80, uoltilia MF 144, uoltilio MF 162,
u]oltịḷ[i---] MF 163; uoltilio MLF 346; uotiliạ LF 222, uoltilio LF 336; perhaps
also uolti[---] MF 167 (this may also be an instance of the patronymic adjective
Voltius, or of the gentilicium Voltius).
Either Volta or Voltius: [u]ọlt[---] MF 145, [u]olṭ[---] MF 163. A patronymic ad-
jective, either Voltius or Voltilius, in uol[t---] MF 156. The abbreviation of these
names is in all likelihood the Vo. in LtF 299, Lat 251.
Both Volta and Voltius occur frequently in the Faliscan inscriptions, but the name
is not attested for Etruscan (apart from vultasi Fa 3.4=Etr XLII, which renders the
Faliscan name), Latin (which had the gentilicium Voltius), or the Sabellic lan-
guages. These names can therefore be regarded, together with Iuna and perhaps
253
CHAPTER 7
also Ianta and Tirrus, as specifically Faliscan praenomina. Unlike Iuna, however,
Volta and Voltius continued to be used into the Late Faliscan period: Volta in fact
makes its final appearance in the Latin 218, from the late second century. The
name is usually derived from an Etruscan *velt- or *velθ-, with the regular devel-
opment of /e/→ /o/ before a velar /l/ (§3.3.4.1). This Etruscan base, however, is
apparently not attested, but cf. the rare gentilicium Velti (velti Pe 1.277, 1.565,
1.1031, ve|l|tia(l) Pe 1.564, veltia(l) Pe 1.1087). What remains surprising is that
the word was included in the first declension instead of to the second.
7.7.2. The origins of the Faliscan praenomina. As might be expected in an area that
lies on the crossroads of several different cultures and languages, the Faliscan onomas-
ticon is of mixed origin: the same, however, could be said of the Latin or Etruscan, and,
to a lesser extent, of the Sabellic onomasticon. As said in §7.1.1, ascribing names to
languages or peoples is difficult (although in the case of the praenomina it is easier than
in the case of the gentilicia, cf. §7.8.2): in many cases it is unclear in what language the
name originated. Even if the origin of the name is clear, the name may have reached the
area through another language than the one in which it originated (as appears to have
been the case with Aulus and Publius): an important point, since such a name may
therefore have been associated with a different group than the speakers of the language
in which it originated. That having been said, the origins of the Faliscan praenomina are
probably more or less as follows:
(1) Faliscan: Exclusively Faliscan are Iuna and Volta (with its derivation Voltius):
though Volta is ultimately derived from an Etruscan praenomen, it appears to be a very
old derivation and has no counterparts in Etruscan: even its Etruscan base *Velte is at-
tested only indirectly at best. Interestingly, both Iuna and Volta are masculine
praenomina of the first declension, a category absent from the Latin and Sabellic ono-
masticon. Praenomina that occur chiefly in Faliscan and only sporadically in other lan-
guages are Aufilus (of Italic origin?), Iantus/Ianta (perhaps of Etruscan origin?) and
Tirrus (of unknown origin). Laeuius and Laeuilius also appear to have been Faliscan,
and the same is perhaps true of the female praenomen Scaeua. The derivation of Vibius
from Etruscan may have been Faliscan in origin, and the great frequency of Gauius in
the area may similarly point to a Faliscan origin of this name: both names occur with
some frequency also in Etruscan and the Sabellic languages, however.
(2) Latin: Several praenomina that occur in the area are of Latin origin: these are Gaius,
Lucius/Lucia, Marcus (with its probably Faliscan derivation Marcius), and the numeral
praenomina Quinctus, Sextus, and Oct-. Of these, Gaius, Lucius and Marcus occur with
any frequency only in the Latino-Faliscan and Latin inscriptions. The same is true of
Aulus and Publius, which are of Etruscan origin, but occurred regularly in Latin, and
may have been regarded as Latin rather than as Etruscan.
254
THE ONOMASTICON
(3) Sabellic: Interestingly, there is hardly any praenomen that is exclusively Sabellic
except the very dubiously attested Maesius or Messius, and Petro. Several praenomina
that occur only in abbreviated form in the ager Capenas and are perhaps Sabellic are
At(tus), F(ertor), Sen( ), and Tr(ebius).
(4) Italic: Of Italic origin, but not ascribable to either Latin, Faliscan, or the Sabellic
languages in specific are Gauius/Gauia (which could perhaps be of Faliscan origin)
Statius, Titus, the very dubiously attested Ancus, and, occurring only in abbreviated
form, Nu(merius). Among these praenomina, the frequency of Gauius/Gauia is surpris-
ing: it is the most frequently attested praenomen in the area (see below). Titus and Nu-
merius also occur in the theonyms Titus Mercus and Mars Numesius (see §6.4)
(5) Etruscan: Clearly Etruscan or of Etruscan origin are Arruns, Aulus, Laris, Lars,
Tana, Tanaquil, Tania, Vel, Velceius, Veltur, and Venel. Probably Etruscan as well are
Caelius, Publius, and Vibius, and possibly Seruius and Tullus. Of these names, how-
ever, Aulus, Publius, and Seruius are also well-known from Latin. Probably also Etrus-
can, but less certainly attested are Acr-, Am-/Amm-, Cincus, and Her-.
Difficult to ascribe to any specific origin are Caesius (Etruscan or Italic?), Aemus and
its derivation Aemius (Etruscan or Italic?), Pupius and Puponius (Sabellic?), and Tettius
(Etruscan or Italic?).
Apart from the origin, the frequency of the names must be taken into account. In view
of the great uncertainty in some instances (especially the abbreviated names), it is not
very useful to push quantification too far, but several tendencies are clear.
By far the most frequently used name is Gauius/Gauia (together 30 instances, not
counting abbreviations), followed by the specifically Faliscan names Iuna (15 instances,
not counting abbreviations or patronymic adjectives), Volta (7 instances, not counting
abbreviations or patronymic adjectives) and Voltilius (8 instances, not counting abbre-
vations). Together, these four names therefore make up 60 instances of the c.230 in-
stances of praenomina occurring in the inscriptions: note that the 60 instances do not
include abbreviations and the total of c.230 does. Of the names of unknown origin, only
Caesius/Caesia is frequent (11-12 instances, not counting abbreviations): if this name is
included with the other four, these five names together make up 71-72 instances of the
c.230 instances of praenomina, or nearly one-third of all attestations of praenomina.
Of the names of Latin origin, several are associated with the Latino-Faliscan and
Latin inscriptions instead of with the Faliscan ones: this is the case with Gaius, Lucius,
Marcus, and the Latin names of Etruscan origin Aulus and Publius. This picture is
partly based on the abbreviated praenomina in the Latino-Faliscan and Latin inscrip-
tions, however, and could therefore be biased. The distribution of Publius/Publia in the
Middle and Late Faliscan inscriptions is curious: whereas Publia occurs 11 times, Pub-
lius occurs only 2-3 times.
255
CHAPTER 7
Etruscan names are many in number, but most are comparatively rare. The most
frequent one is Arruns (7 instances), followed by Laris (4 instances), Vel (3-4 in-
stances), Tanaquil (3 instances), and Veltur (2 instances). The others, including Lars,
which in Etruscan is among the most frequent praenomina, are all attested in one or two
instances at best. (Cf., however, §7.10.5 with note 142.) In spite of this rather meagre
frequency, these names are still more frequent than they are in Latin or Sabellic inscrip-
tions, as may be expected for an area where the Etruscan presence must have been
large. The specifically Sabellic names not only are few, but each is used in only one or
two instances.
Keeping in mind that praenomina were given, not received like gentilicia, this implies
that the inhabitants of the ager Faliscus had some clear preferences in the names they
chose for their children, and as several of the most frequent names did not occur else-
where, these names may well have been a part of ethnic identity. This is discussed fur-
ther in §7.10.5.
7.7.3. Types of Faliscan praenomina. Among the praenomina occurring in the Falis-
can inscriptions, several groups can be identified according to their derivation:
(1) Patronymic praenomina. Many praenomina are of patronymic origin, and in the
light of the lasting use, in Faliscan, of patronymic adjectives (§7.5.2), this is hardly sur-
prising. Examples are Aemius, Caelius, Caesius, Laeuius and Laeuilius, Marcius, Mes-
sius, Vibius, Voltius and Voltilius, perhaps also Paquius, Pu(m)ponius, Saluia, Seruius,
Statius, Tettius, and possibly also Letaeus and Velceius. It is noteworthy that the one
instance of Marcius (LF 221, 222, 223) may well be an ad hoc praenomen for a man
who is apparently a ‘fatherless’ son of a freedwoman, reflecting his parentage not in his
gentilicium, but in his praenomen: this indicates that the process of using patronymic
adjectives as praenomina was a still continuous process. Since many gentilicia, too,
were of patronymic origin, it is not surprising to find the same name used both as a
praenomen and as a gentilicium, as in the cases of Caelius, Caesius, Laeuius, Laeuilius,
and Marcius. See also §7.8.2. on the patronymic gentilicia.
This derivation may also be the origin of the female praenomina in -ia beside a
male equivalent in -us: far from being the female form of the male name, these female
praenomina reflect patronymic adjectives, so that beside Titus ‘propitious (m.)’ stands
not Tita ‘propitious (f.)’, but Titia ‘daughter of Titus’. Examples of this are Titia, Tullia,
perhaps also Pupia, and possibly Early Faliscan Rufia (§7.2.2)
(2) Numeric praenomina. A number of praenomina are numeric: Quinctus, Sextus,
Oct-, and possibly also Qua-, based on Latino-Faliscan numerals, and perhaps Petro,
and, indirectly, Pu(m)ponius, based on Sabellic numerals. They may originally have
indicated the month of birth (rather than the sequence of sons within the family), which
256
THE ONOMASTICON
would explain why originally only numeric praenomina derived from the ordinals for
‘four’ and higher are attested: see H. Petersen 1962 and Salomies 1987: 111-20.
(3) Praenomina of good omen. Several names have a ‘propitious’ meaning, bearing a
connotation of ‘good omen’. As such I regard Laeuius and its derivation Laeuilius, de-
rived from laeuus, the female praenomen Scaeua, derived from scaeuus, and Titus
(which also occurs in the theonym Titus Mercus) and its derivation Titia. Perhaps also
Saluia may be included in this group, if related to saluus.
7.8. The gentilicia
7.8.1. The gentilicia attested from the Middle Faliscan period onward. The fol-
lowing list contains all gentilicia attested in the Middle Faliscan, Late Faliscan, La-
tino-Faliscan, and Latin inscriptions from the ager Faliscus and Capenas (for the Early
Faliscan inscriptions, see the list in §7.2.2).
I have included all abbreviations that can be considered gentilicia, as explained
in §7.1.2: contextless abbreviations are included between [ ], but only if the abbrevia-
tion is attested elsewhere in a context where it clearly represents a gentilicium, or if it
can easily be matched to an existing gentilicium. Names occurring only in Latin in-
scriptions are included as lemmata between [ ]; names occurring only in Etruscan in-
scriptions have been included only if the same or a related name is attested in the Fal-
iscan onomasticon. As the data for the Sabellic onomasticon are relatively few, it has
only been noted when cognates or derivations are attested, not when they are not. For
ease of reference I have used the closest Latin equivalent as the lemma (except in the
case of abbreviations) and ordered the lemmata according to the modern alphabet.
1. [Abellensis. f. abelese Lat 251 (dat.) Apparently a gentilicium derived from a
toponym Abella. This place may be identical with Campanian Abella, although
the toponymic adjective derived from that name was Abellanus in Latin (also used
as a gentilicium) and in Oscan (abellanúí Cm 1.A3 etc.). Solin (1972:165 n.2) re-
garded it as an ethnicon rather than a gentilicium, but in view of the other
toponymic gentilicia in the area (see §7.8.2), it may well be a gentilicium.]
2. Acarcelinius. m. acarcelini LF 221 (gen.), acacelini LF 222 (gen.), acarcelinio
LF 223, acarcelinio LF 226. The name is not attested elsewhere:132 Schulze
(1904:111, 368) suggested that it was derived from a gentilicium like Accaeus
with a suffix parallell to the one in Rup-arcellius: G. Giacomelli (1963:171) and
Hirata (1967:31-2) assumed that this gentilicium might be Acus (see below under
132
Schulze (1904:111) hesitatingly referred to CIL VIII.15474 as a further attestation of the
name, but that text reads gemina l fili|a carcelinia.
257
CHAPTER 7
Aconius). Already Peruzzi (1963b:441-6), however, pointed out that the name
might be a new formation, perhaps derived from a toponym, and A. Mancini
(1981) in fact quite attractively derived it from an */akarkelom/ that would be
equivalent to (but not necessarily identical with) the */okrikelom/ reflected in
Latin Ocriculum and Umbrian */okriʃlom/ implied by Etruscan ucrislane ET
Cl 1.2609, 2611-2613 etc. Cf. also Calzecchi-Onesti (1981:184-8, 165-7) on
acr-/arc- in Italic toponyms and on Ocriculum.
3. Aci. abbr. aci Cap 395. Probably Accius or Acilius: the latter is attested in CIL
XI.7531 from Falerii Novi and CIL XI.7768 from Fiano Romano.
4. Acciuaeus. m. ac ̣iuaiom (or aḷiuaiom) Cap 465 (gen. pl.). The name appears to be
a derivation of Accius (cf. above under Aci.) with the same suffix as in karkavaios
CIL I2.2917a (Colonna 1990a).
5. Aconius. f. ạconia LF 220, perhaps also m. *(*)coṇẹo LtF 290. Other attestations
may be aco[---] LtF 341 and aco[---] LtF 327. (G. Giacomelli (1963:172) pre-
ferred to interpret these texts as instances of a gentilicium Acus.) The name is an
adaptation of the Etruscan gentilicium Aχu (21 attestations, mostly from Clusium
and Perusia; cf. also Aχuna), and occurs in CIL XI.3115-3119 from Falerii Novi.
6. ? Acr-. ạcṛẹẓ MF 67 (gen.?), acre MF/Etr 279. Hirata (1967:32) classed the name
as a gentilicium, in which case it would be connected with Etruscan acriś Pe
1.951 and acries Vs 1.138: cf. the Latin gentilicium Acrius. G. Giacomelli
(1963:172) regarded it as a praenomen: see §7.7.1.2.
7. Adicius. m. adicio MLF 378. A Latin gentilicium Adicius occurs in CIL V.4251
from Northern Italy.
8. ? Aenus. Perhaps m. aino in f aino MLF 352, if this is not to be read as faino (see
Faenus). Like Latin Aenius, Aenus could be an adaptation of the Etruscan gen-
tilicium Eina (eina MF 57): see also §7.8.2.
9. ? Aieius. f. aie* MF 110 (read as aieạ by Herbig CIE 8032). The reading aieạ, the
interpretation ‘Aieia’, and the derivations proposed (see e.g. Stolte 1928:289) are
all equally doubtful. G. Giacomelli (1963:172) connected this name with Aiedius.
10. Aiedius. m. aiedies Cap 390 (nom. sg. or pl.?). The name is apparently not at-
tested elsewhere. For names in Aie-, see Schulze 1904:116-7: cf. perhaps the ab-
breviated Oscan gentilicium aie Po 89. The suffix -idius is of Sabellic origin: if
the inscription, k  pa  aiedies, is interpreted as ‘K. Aiedius, son of Pa.’, it shows
further Sabellic features in the nominative in -ies and in the (Umbrian-Volscian)
placement of FILIATION between the praenomen and the gentilicium: see §9.3.2.
11. Alliuaeus. m. aḷiuaiom (or ac ̣iuaiom) Cap 465 (gen. pl.). The name appears to be
a derivation of the Sabellic gentilicium Alis or Allis (Latin Allius) that occurs in
258
THE ONOMASTICON
South Picene alies TE 2 and Marrucinian alies MV 4 (twice), formed with the
same suffix as in early Latin karkavaios CIL I2.2917a (Colonna 1990a).
12. ? Am- or Amm-. without context ame 280, aṃẹ 282. G. Giacomelli (1963:173)
classed this name as a gentilicium, but Hirata (1967:34) as a praenomen (cf.
§7.7.1.6), pointing to the Latin (patronymic?) gentilicium Ammius (which occurs
in CIL XI.3080 from Falerii 133). The name appears to be of Etruscan origin.
13. Annius. m. ani MF 45 (gen.?), anio Cap 420, anni LtF 63 (gen.?).
14. Annilius. m. anel[i] MF 469* (gen.); f. anelia MF 101. Another attestation would
be manileo MLF 355, which can be read as m anileo as well as manileo.
The Latin gentilicia Annaeus, Annius, Annilius, and several others in Ann- are ad-
aptations of the well-attested Etruscan gentilicia Anae/Ane/Ani (for the numerous
attestations, see ET: from Civita Castellana is anae lauvcies Etr XXIX). Cf. also
the Faliscan poet Annianus or Anianus mentioned by Ausonius (Cent. 11). Gen-
tilicia of this group are also attested for the Sabellic languages, e.g. South Picene
anaiúm AP.1, Paelignian anaes Pg 10, annies Pg 40, annia Pg 15, 33, Oscan
annịeí Cp 38, Paelignian anniaes Pg 39, aniaes Pg 38, Vestinian aninies MV 11.
15. Ap. abbr. ap Cap 419.
16. Anu. abbr. anu in cl  anu Cap 397. The attestation is doubtful: the text was read
as clanu by G. Giacomelli (1963:185) and Hirata (1967:59-60), the latter compar-
ing Etruscan names in Clan-. Since the names in the Capenate inscriptions are
mostly Latin rather than Etruscan, I woud rather read cl  anu and compare the
Latin names in Anu- (cf. Solin & Salomies 1994:17-8).
17. Aratius or Arantius. m. aratio MLF 348, aratio MLF 349, ar‹a›tio MLF 350; f.
aratia MLF 357. The gentilicium may be a patronymic derivation from the Etrus-
can praenomen Araθ/Aranθ, in which case it could be read either as Aratius or as
Arantius. The Latin onomasticon has no corresponding names: perhaps the name
may be compared to Aradius (cf. Faliscan calitenes MF 265, which corresponds
to Latin Calidenus), but this name appears to be of Middle-Eastern origin
(Schulze 1904:113). Cf. perhaps also the Etruscan gentilicium araθenas Vs 1.88.
18. [Arn. abbr. in MF/Etr 37. It is not clear if this is a gentilicium or a praenomen: if
it is a gentilicium, it could be Arnius (Schulze 1904:412). See also §7.7.1.9.]
19. [? Arrius. f. a[rria] in arria | plaria Cap 431. The restoration is extremely doubt-
ful: a very different reading, in which the name would be sal | plaria, with a
praenomen sal(uia), was proposed by Torelli (1974:741-6): see also Kajava
(1995:70).]
133
The Iulia Ammia of this inscription seems to be of Oriental extraction, however, as she is
called tigranis | regis f (perhaps Tigranes of Armenia, executed in 36 CE, cf. Tac. Ann. 6.40).
259
CHAPTER 7
20. Arruntulus. m. arutlo MF 195.

21. Aruntielius. f. arutielia [?---] MF 96.
Both names are derivations of the Etruscan praenomen Arnθ, for which see
§7.7.1.9. Arutlo appears to be a diminutive: G. Giacomelli (1963:175-6) com-
pared Etruscan aruntleσa AS 1.227. The derivation of arutielia is difficult: it ap-
pears to be derived from Arruntius, but neither as a diminutive (which probably
woud have been arutela) nor as patronymic derivation (which would have been
arutilia or arutelia). The formation is reminiscent of the Sabellic gentilicia in
-iēnus, but it may well be an error, e.g. aruti{e}lia or arut{i}elia.
22. Atronius. m. atron (or atronị?) MF 13 (abbr. or gen.). Like Latin Atronius, it is an
adaptation of Etruscan Atru (atru Cl. 1.1298, atrus Ru 2.5, atruś Sp 2.76) or
Atrune (atrunias Cl 1.1347).
23. Au... in au[---] LF 236.
24. Aue. abbr. aụe Cap 396. G. Giacomelli (1962:176) suggested a connection with
Etruscan Avei (avei Pe 1.896, aveiś Pe 1.897, 1.898, 1.1132, 1.1297, AS 1.160)
and Latin Aueius (thus also Hirata 1967:37), comparing also Auius and Auienus.
25. Aufilius/Oufilius. oufilio MF 48, aufilio MF 49, auf ̣ilio MF 50, auf]ilio MF 51,
oụ*[..]o MF 52, aufi[lio ?---] MF 53, oufilio MF 275, oufilio MF 276. A patro-
nymic gentilicium derived from the (Faliscan?) praenomen Aufilus/Oufilus, for
which see §7.7.1.12. Latin, too, had Aufilius/Ofilius, but also Aufillius/Ofillius and
Aufellius/Ofellius: note the intervocalic -f-, pointing to a non-Roman origin of the
name. Cf. perhaps also the Umbrian gentilicium ufeŕie[r] Um 8.
26. ? Aufitius. m. ofiti in f ofiti MF 58 (if not to be read as fofiti). A Latin gentilicium
Aufitius occurs in CIL VI.6945 from Rome.
27. ? Aulena. m. olna MF 82. Editors usually interpret this as a noun, but no satisfac-
tory interpretation has ever been given for it: I would rather read it as a second
gentilicium (describing a freedman, cf. §7.6), comparing Etruscan Ulena (ulenas
Ru 2.4), Aulna (aulnal AS 1.11, Cl 1.1241, Pe 1.943, aulnas Vs 1.244, aulnaś Fe
3.3), Aulne (aulnei Cl 1.1308, Pe 1.93) and Latin Olnius and Aulenus.
28. Battius. m. batio MLF/LtF 359. The use of b is surprising, and probably points to
a non-Faliscan origin of the name. Latin had a gentilicium Battius (Schulze
1904:423).
29. [Blaesius. m. blaisiís 468*. The name, like Latin Blaesius, appears to be of Sabel-
lic origin, cf. Oscan blaisiis Cm 14.C8. The text also has a Sabellic praenomen
pa‹qu›is and shows Sabellic epigraphic and linguistic features (such as the synco-
pation of the final syllable).]
30. Ca. abbr. ca Cap 427, 428.
260
THE ONOMASTICON
31. Caelius. m. cailio MF 90, c]ẹlio MF 94, celio MF 95, ce[lio MF 96, c]elio MF
97, celio MF 105; cail[ia MF 92; cailio MLF 376. A patronymic gentilicium de-
rived from the praenomen Caelius and ultimately from Etruscan Caile (see
§7.7.1.17). The gentilicium also occurs in Latin; Etruscan had a related gen-
tilicium Cailina (cailinal Vt 1.43).
32. Caesilius. f. cesilia MLF 211. A patronymic gentilicium derived from the
praenomen Caesius (for which see §7.7.1.18). The gentilicium also occurs in
Latin (Schulze 1904:135) and in Oscan (kaísillieís Cp 25).
33. Caesius. m. cesịe MF 257, cesies MF 265, cf. also ceises Etr XXXIV. (The name
Caesius also occurs as a praenomen: see §7.7.1.18.). The name occurs in Etruscan
as Caise/Ceise (caise AH 1.80; ceises AT 1.67, 1.145, ceisi Ta 1.116, AT 1.67)
and Ceiσ́e (ceiσ́i Pe 1.325, 1.326, ceiσ́is Pe 1.323, 1.327, ceiσ́ial Pe 1.505), as well
as Caisie (caisies Cm 2.49): Latin had Caesius, as well as several other names in
Caes-.
34. ? Calinius. The name was read by Herbig (1910:187) in calin[---]|rezo[---] MF
57. The gentilicium occurs in Latin and in Oscan (kalinij Me 1, kaleinij Me 3,
k]ali[nij Me 2). As a gentilicium at the beginning of the text is very unusual, I
would rather read ca lin[---].
35. Calitenus. f. calitenes MF 265. The name has been equated since Herbig CIE
8387 with Latin Calidenus, for which cf. Schulze 1904:138. It is in all probability
related to the Etruscan gentilicium caliti Pe 1.1441, which looks as though it
might be derived from a toponym Cales. Cifani (2002:33), without referring to
calitenes, suggested that Cale may have been the original name of modern
Gallese in the north-eastern ager Faliscus. For Calit- : Calitenus cf. perhaps Volta
: Voltenus (Lejeune 1952b:124 n.1).
36. [Calpurnius. m. calpurnius Cap 432.]
37. Catineius. m. catinei MLF 469* (gen.). Latin has Catineius beside Catinius. Cf.
Etruscan catni Ta 1.166.
38. Cincius. m. cicio MF 40; f. cincia MF 135; cf. also Etruscan cẹncu Etr XXI. Like
Latin Cincius, the gentilicium is derived from the Etruscan gentilicium Cincu
(cincus OA 2.60, cincual Pe 1.53, cinc[ual] Cl 1.102; cf. also cincunia Pe 1.54,
1.748) or Cencu (15 attestations, apart from Fa 2.4=Etr XXI all from Clusium).
The gentilicium also occurs in CIL XI.3327 from Forum Cassii.
39. Citius. f. citiai MF 270 (gen. or dat., but cf. §9.2.3c). G. Giacomelli (1983:185)
pointed to Etruscan citia in TLE 495, which appears to be the only Etruscan paral-
lel. Latin has a gentilicium Citius.
261
CHAPTER 7
40. Clanidius. m. clanidio Cap 394. The name apparently does not occur elsewhere,
but is formed with the Sabellic suffix -idius. The closest parallel is Clandius in
CIL XI.2004 from Perusia, but this name is connected rather with the Etruscan
gentilicium Clante (Schulze 1904:529 n.6). Cf. perhaps also cl  anu Cap 397,
read as clanu by G. Giacomelli (1963:185) and Hirata (1967:59-60).
41. [Claudius. m. claudia Cap 393.]
42. Clipearius (and Clipeaeus?). m. clipeaio (clipea‹r›io?) MF 470*; clipịaṛ[io] LF
230, clipeaṛ[io] LtF 231, cḷ[i]peario LtF 233. A Berufsgentiliz ‘Shieldmaker’ de-
rived from clipeus/clupeus (a word that is probably of Etruscan origin, see
§6.2.9). It is unclear whether clipeaio is an error for clipea‹r›io or if this is a dif-
ferent derivation from the same noun: cf. also frenaios MF 471* instead of the
expected frena‹r›ios. The gentilicium Clipearius apparently occurs only in the
ager Faliscus (if this can be concluded from Schulze 1904:416): unclear is clu-
piaria | origo | q · mudasidius | arists CIL VI.4925 from Rome.
43. Cocilius. f. cocelia MLF 303. G. Giacomelli (1963:186) compared Latin Caucil-
ius/ Cocilius, but also Coclius and Coculnius. This last name is derived from the
Etruscan gentilicium Cuclni from Tarquinii (culcnial Ta 1.9, 1.14, culcnies Ta
1.31, 1.95, 1.96, culcni[es] Ta 1.97).
44. Colanius. m. colanioi MF 69-70 (gen. or dat.). Herbig (1914a:239) connected the
name to the Latin gentilicia Colus and Colius. Hirata (1967:46) compared the
Etruscan gentilicium Culni (kulnei Vs 1.208 culni Cl 1.1524, cf. also culnaial Cr
2.54, 2.55, 2.57). Latin has a gentilicium Colianius, which might be related.
45. [Cotena. m. cotena Lat 214. Cotena appears to be attested only here, although
Schulze also points to Kott…niaj in an inscription from near Faenza. The name re-
flects Etruscan names like Cutna/Cutne (12 attestations, mostly from Clusium).]
46. Cutrius. m. cutri MF 200 (abbr.). Latin Cotrius. The spelling with u is Etruscan
rather than Faliscan, cf. Popius : Pupius (below).
47. [Didius. m. didius Lat 456.]
48. ? Decon... Perhaps attested in decon[ LtF 174. Thulin (1907:292-4) interpreted
this as an instance of an unattested gentilicium Deconius, an adaptation of the
Etruscan gentilicium Tecunas from Volsinii (tequnas Vs 1.48, 1.153, 1.154, te-
quna[s 1.151).
49. ? E...sus in e**sa Cap 457. Unclear.
50. Egnatius. m. egnatius Lat 291 ; f. e[c]ṇata MF 81; also Ekn = Egnat[ LF 246.
Identical with Latin Egnatius, epigraphically attested for the area in CIL XI.3083
(twice) from Falerii Novi, and CIL XI.3257 (twice) from Sutrium. The name oc-
262
THE ONOMASTICON
curs in Etruscan as Ecnate/Ecnati (eknate Vs 1.299, ekṇạṭ[e] Ta 7.40, ecnate Vs

1.170; ecnaṭị[ Ta 1.256, ecnatial Ta 1.95) and Ecnatna/Ecnatni (ecnatna AS
1.316, ecnatnas Vs 1.202, 1.307, ecnatnal Cl 1.1455, ecṇ[a]ṭnal Vt 1.110,
ecnatn[al] Cl 1.1682; ecn]atni Cl 1.388, ecnatnei AS 1.232, ecṇaṭṇei Cl 1.1568).
Cf. perhaps also the abbreviated Volscian praenomen ec VM 2. The name may be
an Etruscan toponymic adjective in -te/-ti.
51. Eina. m. eina MF 57. Identical with Etruscan eina Cl S.17, eini Cl 1.1574, einis
Cl 1.1575. The name may originally have been Aina (cf. aina Ru 0.13, without
context?), an adaptation of which may be aino MLF 352, if the text, faino, is to be
read as f aino: see Aenus.
F- see also H-
52. [? Fab. abbr. without context hap MF? 46. Probably to be interpreted as an ab-
breviation of the gentilicium Fabius, with Faliscan h- for an original /#fV/ (see
§3.5.2). Latin Fabius, a Latinization of the Etruscan gentilicium Fapi (fapi Cl
1.220, fapis Pe 1.904 (used as a praenomen).]
53. ? Fac... in hac****a MF 89. The only Etruscan parallel would appear to be
φacsneal Pe 1.1191, in which case the Faliscan instance would show the Faliscan
spelling h- for original /#fV/ (see §3.5.2).
54. ? Fadius. m. perhaps hạθ̣i MF 13 (gen. or abbr.). If indeed to be read thus, the
name identical with Latin Fadius (Schulze 1904:132, 516), perhaps an adaptation
of an Etruscan name gentilicium hatina Pe 1.686, cf. also on Fadenius (below).
55. Fadenius or Hadenius. f. hadenia MLF 360 (either with h for original /#fV/ or
with a hypercorrect f for original /#hV/, see §3.5.2). Probabily connected to
Etruscan hatina Pe 1.686, Latin Fadenus (Schulze 1904:132). Cf. also Fadius.
56. Faenus. m. faino MLF 352. G. Giacomelli (1963:189) compared Latin Faenius:
for Faliscan Faenus : Latin Faenius, see §7.8.2. The text may have to be read as f
aino, cf. Aenus and Eina (above).
57. Fa(r)farn... fafarṇ MF 136 (abbr.), f ̣af ̣[---] MF 139. Herbig (CIE 8237) suggested
a connection with the name of the nearby river Farfarus (Ovid Met. 12.328-30) or
Fabaris (Verg. A. 7.716): see §6.5.1. For other potamonymic gentilicia, see Nari-
onius and Vomanius.
58. Fassius. f. fasies MF 41. The only direct parallel appears to be Oscan fassii[s
ZO 1. Latin had only Fassidius, derived with the Sabellic suffix -idius.
59. Feliginas. m. felic ̣inate MF 42 (gen.), [fel]ịcinatiu LF 384 (gen. pl.). The name
has parallels in Etruscan Felcinate/Felcinatne (felcinatial Pe 1.485, 1.1235; felci-
natnal Cl 1.2673) as well as in Latin Fulginas (cf. Schulze 1904:528). The names
263
CHAPTER 7
are derived from a toponym *Feligin- (*Felginum Rix 1965:233 n.133), which
may well be identical with Fulginium/Fulginiae (modern Foligno in Umbria, on
the Via Flaminia).
60. Fertorius. m. fertrio Cap 391. A patronymic gentilicium derived from Fertor,
which may be attested from the area in the abbreviation f: see §7.7.1.23. Latin had
both Fertorius and Hertorius. (Fertrio has also been interpreted as Fer(e)trio(s),
but there seem to be no parallels for a gentilicium Feretrius.)
61. Fescuna or Hescuna. m. hescuna MLF 346. Derived by Colonna (1990b:123
n.52) from the toponym Fescennium in the ager Faliscus, with the spelling h- for
original /#fV/ (cf. §3.5.2). Latin gentilicia that are perhaps related are Fescenna
(Schulze 1904:80) and Fescennius (Schulze 1904:231). Cf. perhaps also Etruscan
Hescanas from Volsinii (hescanas Vs 7.34, 7.38, hescan[as] Vs 7.35, ḥes[canas]
Vs 7.36, h[esca]na[s] Vs 7.31, hescnas Vs 1.183. 0.23).
62. Firmius. m. ḥiṛṃeo MF 19, firmio MF 54, hirmio LF 213; f. ḥirmia MF 18,
firmia MLF 302. The name has been connected with Latin Firmius and the adjec-
tive firmus, in which case it shows the Faliscan development /#fV/ → /#hV/ in
ḥiṛṃeo MF 19, hirmio LF 213, and ḥirmia MF 18 (§3.5.2). G. Giacomelli
(1963:193) connected it with Etruscan names such as hermana MF/Etr 265, in
which case the forms with f- would be due to hypercorrect (§3.5.2).
63. Flauilius. f. hlau|elea LtF 325. The spelling with hl- is unique, reflecting a hyper-
correct extension of the spelling h- for original /#fV/: see §3.5.2. Although Latin
had several gentilicia derived from Flauus/Flauius, the gentilicium Flauilius is
not attested for Latin, although its originally Sabellic counterpart Flauidius is.
64. Folcosius. m. fulczeo LF 329, folcozeo LF 330, folcuso LF 331, *olcuzeo LF 332,
folcosio LF 333; also holc[osi] MF 140. Folcosius is only attested here, although
Latin has a closely related gentilicium Holconius (see Schulze 1904:169, who
also compared gentilicia like Fulcennius, Fulcinius etc.). These names would ap-
pear to be adaptations of an unattested Etruscan *Fulχu or *Hulχu: cf. Etruscan
Hulχena (hulχenas Vs 1.28, 1.99, hulχnas Vs 2.35) and Hulχnies (hulχnies AT
5.2, hulχniesi Ta 5.2, 5.5, h[ul]χ[n]iesi Ta 5.4).
65. Frenaeus or Frenarius. m. frenaios MF 471*. The name is not attested else-
where: if frenaios is an error for frena‹r›ios, it could be a newly-formed
Berufsgentiliz ‘Bridler’, (related to Latin frenum ‘bridle’), like Clipearius ‘Buck-
ler’. In view of other gentilicia in -aeus (latinaio MLF 210, uoltaia MF 196), the
possibility of a gentilicium Frenaeus cannot be excluded, cf. Berenguer & Luján
2004:219-20.
264
THE ONOMASTICON
66. Fullonius. f. fulonia MLF 313. Like Latin Fullonius and Umbrian fulonie Um 7,
this name may be derived from Etruscan Fulu (19 attestations, mostly from Clu-
sium and Volaterrae) or Ful(u)na/Ful(u)ne (27 attestations, mostly from the ager
Saenensis). Already Schulze (1904:168) rightly rejected taking this name as a
Berufsgentiliz derived from Latin fullo. G. Giacomelli (1963:194) identified Ful-
with Fol- in Folcosius.
67. [Fuluius. m. Fuluius Lat 250. Also in CIL XI.3156 from Falerii Novi.]
68. [Furius. m. fourios Lat 216. Also in CIL XI.3164 and 3170 from Falerii Novi.]
69. [Genucilius. m. genucilio Cap 435.]
70. Gr. abbr. cr MF 33. Probably to be interpreted as Graecius or Graecilius, see
Grae....
71. Grae... crạ[i--- MF 141, cre[---] MF 142, cṛ[---] MF 143; possibly also the ab-
breviated gentilicium cr MF? 33. The fragmentary name is in all probability to be
read as Graec..., probably Graecius (cf. Schulze 1904:522) or perhaps Graecilius.
Latin has several gentilicia in Graec-, and Etruscan had both Creice (e.g. creice
Cl 1.1280: ET gives c.20 instances of this gentilicium, from various locations) and
Creicna (creicnal Ar 1.4). The gentilicium may be derived directly from the eth-
nonym, but also from the praenomen Graecus (attested for Faliscan as kreco MF
147), especially if this was an existing Etruscan praenomen, as Salomies
(1987:71-2) suggests and seems to be implied by the (patronymic) gentilicium
Creicna: see §7.7.1.26. See also 132. Raec(i)lius.
H- see also F-
72. ? Her-. Very dubiously attested in cẹsit  fere MF 263. Salomies (1987:73) in-
cludes fere in her discussion of the praenomina Herius and Herennius (cf.
§7.7.1.27), but G. Giacomelli (1963:192) and Hirata (1967:51) regarded it as a
gentilicium (cf. perhaps Hirius in CIL XI.2980 from Tuscania?).
73. Hermana. hermana MF/Etr 265, cf. Etr her Etr VI-VII. An Etruscan gentilicium
occurring also in hermanaś Cl 2.11, cf. also Hermena (hermenas Pa 3.1
h]ermenas Vs 1.152, perhaps also h]eṛmenaie Ve 3.19). Latin has Herminius and
Hermenius. G. Giacomelli (1963:192) connected the gentilicium Firmius (see
above) with this name.
74. Hirpius. m. írpios Cap 389. Since the name has been read correctly by Briquel
(1972:833-7) as írpios rather than the śrpios of all previous editors, it has become
possible to connect it with the cult of the Hirpi Sorani on Mount Soracte (see
§2.3.4): in Pliny’s account (NH 7.2.19), these are described as “familiae sunt per-
paucae quae uocantur Hirpi”. A Latin gentilicium Hirpius was already known
(Schulze 1904:234): see §6.6.5.
265
CHAPTER 7
75. Ie... in ie[---] MF 93. Various possible names may be found in Hirata 1967:54
and Solin & Salomies 1994:95-6. Cf. perhaps iegia · ty|che CIL XI.3447 from
Tarquinii, or Paelignian ieíis nPg 8?
76. Iunius. m. iunio Cap 462. The gentilicium is derived from the specifically Falis-
can praenomen Iuna (for which see §7.7.1.29): in view of the frequency of this
praenomen, however, it is surprising to find only one early attestation of the gen-
tilicium (and that from the ager Capenas): the name further occurs in CIL XI.3174
from Falerii Novi, and in CIL XI.3934 from Capena. Note that the Latin gen-
tilicium Iunius was derived by H. Petersen (1962:352) from the name of the
month rather than from the praenomen.
77. L... in ḷ[---] MF 158.
78. Lartius. f. lartia in tana | lartia MLF 338. It is unclear whether lartia is a patro-
nymic gentilicium or a patronymic adjective: in either case, it is derived from the
Etruscan praenomen Larθ (see §7.7.1.35), like the corresponding Latin gen-
tilicium Lartius.
79. Latinaeus. m. latinaio MLF 210. Perhaps derived from the Latin ethnicon
Latinus rather than from Etruscan gentilicium Latini (c.70 attestations, nearly all
from Clusium) or Latine (Cm 2.57, latines Ve 2.4). For the derivation, see G. Gi-
acomelli 1962.
80. Latrius. m. latrio MLF 324, latrius Lat 218; f. latria MF 75. Latin Latrius (and
Laterius?), Etruscan latrnei Cl 1.501, perhaps also Latin Latronius and Etruscan
ḷaθruni Pe 1.1091.
81. Laeuius. m. leuieis Lat 251 (abl. pl.); f. leuia LtF 327 (possibly a praenomen, see
§7.7.1.32).
82. Laeuilius. m. leueli MF 14 (gen.), le[ueli]o MF 146; f. leue[lia] MF 147. The
name also occurs as a praenomen, see §7.7.1.33.
Laeuius is in all probability a patronymic derivation from the praenomen Laeuus,
while Laeuilius is in its turn a patronymic derivation from the name Laeuius. This
praenomen is derived from the adjective laeuus ‘well-omened’: see §7.7.1.32-33
on the praenomina Laeuius and Laeuilius. (Cf. also the praenomen Scaeua,
§7.7.1.58) G. Giacomelli (1963:199) rejected a possible connection with Liuius;
Hirata (1967:57) unnecessarily suggested a connection with Laelius.
83. Le... in lẹ[---] MF 148, which may be read as Lae-, Le-, or Li-.
84. Lepuius (Laepuius?). f. lepuia MF 144. Unclear: there are no Latin, Etruscan or
Sabellic gentilicia that appear to be related, in spite of the suggestions made by G.
Giacomelli (1963:199), Hirata (1967:56), and others (which all concentrate on
Lep- in stead of on Lepu-). Cf. perhaps Latin Laeponius, which appears to be an
266
THE ONOMASTICON
adaptation of an Etruscan *Laipu: this Laipu might have been adapted in a differ-
ent way in Faliscan to Laepuius.
85. Letaeus. m. letei MF 470* (gen.). This unique name clearly renders the equally
unique Etruscan leθaie Etr XLVIII, probably derived from lete Sp 2.109: on
these names, see Vetter 1948:67-72 (who regards these as names implying de-
scent from serfs or bondsmen).
86. Licinius. m. licinio MF 259-260. Latin Licinius, an adaptation of Etruscan Licine
(lik[in]e Cl 2.18, licines̀ i Cr 3.13 licines̽i Cr 3.18; licne[ Vs 2.38, licni Ta 1.1222,
Cl 1.2206, licnis Cl 1.2207).
87. Lin... perhaps in ca lin[---]|re zo[---] MF 57. Herbig (1910:187) read calin[---]
with the gentilicium Calinius, see above.
88. [Lucilius. m. loucilios Lat 268 (import). Also in CIL XI.3109 from Falerii Novi.]
89. Lullius. m. lullio MLF 207. Latin Lollius (CIL XI.7487 from Falerii and 3864,
3887 from the ager Capenas), and Lulleius, Lolleius, Etruscan Lule (lule Cl 1.394,
luleσ́a Cl 1.395, 1.1470, 1.2589; lulia Cl 1.1136, 1.1955). The spelling with ll is
surprising (cf. §11.2.5.5 and §3.5.5.3).
90. Lurius. f. loụṛia MF 41.
91. Luriaeus or Lurieius. f. loriea MLF 314.
The second gentilicium is in all probability is a derivation from the first, either
Lurieius (G. Giacomelli 1963:200-1) or Luriaeus.134 The gentilicia Lurius and
Lurianus occur also in CIL XI.3181 from Falerii Novi. G. Giacomelli (1963:200-
1) and Hirata (1967:58) also pointed to Loreius.
92. Ma. abbr. ma Lat 451-452. It is also possible to read na.
93. Maecius. m. mecio MLF 211. Latin Maecius.
94. Mallius (or Manlius?). m. malio MF 39. Latin Mallius. The name could also be
read as ma(n)lio = Latin Manlius, which occurs in CIL XI.3254 from Sutri.
95. Manius. f. m{e}ania LF 224, mania LF 225. M{e}ania is an error for mania (as
appears both from the inscription and from the fact that both inscriptions refer to
the same person), not a distinct gentilicium (thus e.g. G. Giacomelli (1963:204-5)
and Hirata (1967:61)). Like Latin Manius, it is either a gentilicium derived from
the praenomen Manius, or directly from the adjective manus (which is also found
in the Faliscan cognomina, see §7.9).
134
I do not understand what phonetic realisation of this name is intended in Rix’s remark
(1994:94) on loriea, “>iea< Schreibung für [iə̯a] wie in etr. Θaniea, Veliea”: the e is appar-
ently assumed to be non-syllabic.
267
CHAPTER 7
96. ? Manilius. m. manileo MLF 355. Like Latin Manilius, it is a patronymic gen-
tilicium derived from the praenomen Manius. (Cf. also the gentilicium Manius.)
The form manileo occurs without context: it is therefore also possible to read the
text as m anileo (cf. above under Annilius),
97. Marcena. m. macena MF 269, mar||cna MF 270. The Etruscan gentilicium Mar-
cna/Marχna, Marcne/Marχne, Marcni/Marχni is very well attested (c.155 attesta-
tions, mostly from Clusium). I doubt whether this name is the direct basis of the
gentilicium Marcius (below).
98. Marcius. m. marc[---] MF 152, ṃarcio LF 228 (perhaps rather a patronymic ad-
jective); f. marcia LF 227 (probably a gentilicium rather than a patronymic adjec-
tive). Like Latin Marcius, the name is probably a patronymic gentilicium derived
from the praenomen Marcus, rather than adaptation of Etruscan Marcena. (In that
case, the expected form would probably be Marcinius, a gentilicium that does in
fact occur in Latin: cf. Schulze 1904:188.)
99. Marhius. m. marhio LF 336. The name may be connected to Marcius (above),
but G. Giacomelli (1963:204) may well have been right in pointing rather to
Campano-Etruscan marhies Cm 6.1 and suggesting a connection with the
Oscan praenomen Marahis (e.g. marahis Cm 14.C6, marahij Lu 2, marahieis Cm
28, maraen Lu 46), which may go back to an older (Etrusco-Sabellic?) */marxio-/.
100. Morren-. f. morenez MF 269. The name is an Etruscoid form in -ez = -es (cf.
§3.5.3, §9.2.2). Morenez is an adaptation of an Etruscan gentilicium such as
Etruscan Murina/Murine/Murini (32 attestations, mainly from the ager Saenen-
sis). There are no direct Latin equivalents: G. Giacomelli (1963:206) and Hirata
(1967:62) pointed to Latin gentilicia in Murr-.
101. [Munius. ṃunio Lat 337. The name also occurs in CIL XI.3941 from Capena.]
102. Na. abbr. na Lat 451-452. It is also possible to read ma.
103. Narionius. m. narionio MLF 206. Probably related to Latin Naronius: G. Gia-
comelli (1963:207) pointed to other names where there are derivational variants
-onius and -ionius. Nar(i)onius may be connected to the name of the nearby river
Nar: cf. the Latin gentilicium Narius (Schulze 1904:80). Cf. also Etruscan Nari
(nar[i] Pe 1.1126, n[ari] Pe 1.1127, naria Cl 1.2008, 2009, narial Pe 1.813,
nari[al] Pe 1.1080, narieś Cl 1.2010).
104. Neln... nel[n---] LtF 299, f.? neln LtF 300 (abbreviated?). The name is entirely
unclear: Neln... could conceivably reflect Naelĕn-/Naelĭn-, Nelĕn-/Nelĭn- or
Nilĕn-/Nilĭn-, but none of these possibilities has any parallels in Latin, Etruscan or
the Sabellic languages.
268
THE ONOMASTICON
105. Neronius. m. neroni MF 15, neroni LtF 325, ne[?]ro[---] LtF 328; neṛ[oni.] MF
16, neroni LtF 340. Latin Neronius, probably originally from a Sabellic *ner. Cf.
the abreviated Umbrian praenomen ner Um 10, 21, also occurring in Latin in-
scriptions from Umbria (attestations in Salomies 1987:80).
106. No. abbr. no Cap 425.
107. Nomesina. m. nomes|ina MF 272. Nomesina is the Faliscan rendering of Etruscan
Numσina (numσ́inal Cl 1.969, 1.1102, 1.1596, 1.2026, numσ̣[́ i]nal Cl 1.1103;
numσ́ine Cl 1.2027). Etruscan also had Numσie (numσ́ie Cl 1.753, numσ́i Cl
1.2025, 1.2028, numσ́is Pe 1.197, 1.198): this appears in Latin as Nu-
misius/Nomisius, e.g. in CIL XI.3110 and 3176 from Falerii, and in CIL XI.2958
from Tuscania.135 These gentilicia are derived from the Sabellic praenomen Nu-
mesis/ Nomesis: see §7.7.1.45). Cf. the theonym numesio · m[art]ẹ in LtF 377.
108. Orticensis. m. ortecese MLF 339; cf. Etruscan urtcsnas Etr XXXV. Orteces- and
urtcs- are in all probability connected (Colonna 1990b:136), and appear to be de-
rived from the same toponym: this cannot be Horta, as G. Giacomelli (1963:209)
has convincingly shown, but rather an unknown *Ortica/*Orticum.136 This may
be related to the name of modern Corchiano (older Orchiano): see §6.5.11.
109. Oscin. oṣcin* LtF 301 (abbr.?). The text and the possible parallels for the name
are unclear: cf. perhaps Latin Hoscinius and Etruscan Huzcena (huzcnạ[s Ta
1.250, huzcnai Ta 1.50, 1.51, ḥụzc̣ neṣc̣ Ta 1.185, huzecena[s] Cr 2.74), although
this requires an omission of h- that is not attested for Faliscan (§3.5.2).
110. P. abbr. p ̣ 454.
111. Pa. abbr. pa Cap 457.
112. Pacius. m. pacios Cap 392. Like Latin Pacius, the name is derived from the well-
attested Oscan praenomen Pacis (e.g. pakis Cp 37,9: for the many attestations, see
ST). In the Sabellic languages, however, the gentilicium only occurs in Paelignian
pacia Pg 4 (cf. also the Samnitic abbreviation pk Sa 51).
113. Pani... in pani[---] LtF 239. Perhaps Latin Panicius? Cf. also Etruscan Pance
from Caere (pa]nces Cr 1.66, panci Cr 1.59, panc[i Cr 1.64).
114. Panur... in au cau | panur 459. It is doubtful if panur is indeed a gentilicium: it
may be an abbreviation of the (slave?) name Panoàrgoj (Deecke 1888:217), cf.
the N. Munitor Panurcus in CIL XI.3166 from Falerii Novi.
135
L. Numisius Viator in CIL XI.3110, however, gives his tribus as Pollia and L. Numisius Pro-
culus in CIL XI.2958 as Stellatina, whereas the tribus of Falerii was the Horatia (§2.6.2).
136
The name may also have been *Hortica/*Horticum, but neither Faliscan ortecese nor Etrus-
can urtcsnas has h-, and there are no certain attestations of omission of h- in Faliscan (§3.5.2).
269
CHAPTER 7
115. ? Partius. m. pạrtis MF 79. The curious ending is the result either of an abbrevia-
tion (cf. Latin Partiscius?) or of a rare and irregular syncopation /-ios/ → /-is/
(§3.6.6.2) accompanied by an equally rare retention of -s (§3.5.7d). There are no
corresponding names in Latin or Etruscan: G. Giacomelli (1963:210) and Hirata
(1967:66) compared Etruscan partunus Ta 1.9, 1.13, 1.15, and parθanaś Cl
1.2035, and Latin Partuleius.
116. Pauicius. f. pauiceo MF 12, paui[ceo LtF 290, and perhaps to be restored in
cauio [ pauiceo ] | ruso [?---] MF 318. (This restoration is based on the assump-
tion that the inscription contains a cognomen ruso and that this cognomen also
occurs in ce  paui[ceo ru]so LtF 290.) G. Giacomelli 1963:210) equated the
name with Latin Paucius, comparing also Latin Pauillius. I greatly doubt the con-
nection with Faucius suggested by Hirata1967:66-7.
117. Pe. abbr. pe Lat 406, perhaps also Cap 403, if the inscription, kape, is to be read
as ka pe. Perhaps Pescennius, see below.
118. Pescennius. m. pscni Cap 387. Latin Pescennius (and Pescenius). Schulze
(1904:80) connected this name with the Etruscan gentilicium Fescenna in CIL
XIV.1016 from Ostia, which in turn would be connected with the toponym Fes-
cennium, one of the major sites of the ager Faliscus. Cf. also Hescuna (above).
119. Petronius. m. peṭṛọṇẹọ MF 473*; f. petrunes LF 226. A patronymic gentilicium
from the Sabellic numeric praenomen Petro (see §7.7.1.50). The name occurs in
Etruscan as Petru/Petrui (more than 130 attestations) and as Petruna/Petruni/
Petrunie (together with Petrn- more than 50 attestations), in Latin as Petronius
(e.g. CIL XI.3207 from Nepi), in Marrucinian as petroni MV 3, and in Paelignian
as ptruna Pg 52. The u in petrunes LF 226 may be due to Etruscan influence (cf.
§3.6.3, §7.8.2).
120. [Plarius. f. plaria Cap 431.]
121. Pleina. m. pleina MF 80, pleina MF/Etr 199, plenes LtF 231, plenese Lat 251
(dat.). Intriguingly, this obviously Etruscan name does not appear to be attested
elsewhere: whether it is identical with or related to plinialc Ta 1.113 is unclear.
Schulze (1904:89) and Stolte (1928:296) hesitated to equate this name with Latin
Plinius (Plīnius) because of the spelling Pl…nioj (without diphthong) in SIG 558:
they preferred to connect the name with Pleius.
122. ? Poenus. perhaps poe[?---] MF? 130, and puiatu MLF? 208. [Cf. perhaps also
puiunal Etr XX.] The attestations are very doubtful. G. Giacomelli compared
Latin Poenus and Etruscan Puina (puina Vt 1.137, 4.1).
123. Polfaeus. m. polafio (=pol‹fa›io) MLF 354. Hirata suggested that the name was a
misspelling for pol‹fa›io (cf. also latinaio MLF 210, uoltaia MF 196), comparing
270
THE ONOMASTICON
names like Polfennius. This is a plausible solution: Schulze (1904:216) in fact

gave a number of names formed from a base Polf-/Pulf-, while Etruscan has a
well-attested gentilicium Pulfna (more than 45 attestations in ET, all from Clu-
sium). Cf. also Marrucinian polfenis MV 1, Paelignian polf Pg 13.
124. Popi. abbr. popi Cap 421. Probably Popius or Popilius. See also under Pupius.
125. [Popilius. m. popil[i] Lat 295, popili Lat 296, popili Lat 478* (all imports).]
126. Praeconus or Preconus. m. in [---?] precono[---|---] cuiteneṭ[---|---] let MLF
361. Although the inscription is fragmentary and difficult to interpret, the name
precono appears to be certain. Latin had a gentilicium Praeconius, which Schulze
(1904:87 n.4) hesitatingly connected to Etruscan Percenna, which he in turn de-
rived from the Oscan praenomen perkens Cm 6, perkedn[eís] Cm 6. If that is cor-
rect, the Latin and Faliscan name may (originally) have been Pre- rather than
Prae-. For a Faliscan Praeconus beside a Latin Praeconius, see §7.8.2.
127. Protacius. m. protacio LF 242, p]rotacio LF 244. For the name, see Schulze
(1904:97, 366). Latin Protacius is attested in CIL XI.3208 from Nepi, CIL
VI.25097 (twice) from Rome,137 and probably also in CIL XII.5728 from Antibes.
128. Ps. abbr. ps Cap 415. Perhaps P(e)s(cennius), attested for the ager Capenas.
129. Pupilius or Pupelius. m. pupelio MF 149, pu]peḷ[i---] MF 150, pup[elio MF 151.
In all probability a patronymic gentilicium derived from a praenomen Pupius,
which in its turn is derived from the praenomen Pupus: see §7.7.1.53. Pupilius
may be identical with Popilius, with the u due to Etruscan influence, but note the
quantitative difference between the Latin gentilicia in Pŏp- and those in Pūp-
(Schulze 1904:213). Latin had Pupilius as well as Pupelius.
130. ? Pupius. f. pupiias MLF 304 (gen.). The name occurs in isolation and may be a
praenomen or a gentilicium (both PRAENOMEN and GENTILICIUM appear in
women’s names in Besitzerinschriften, see §7.4.1). If it is a gentilicium, the name
is probably identical with Latin Pupius. Alternatively, it could be identical with
Popius, probably attested in popi Cap 421, in which case the u could be due to
Etruscan influence: see under Pupilius. For Pupia as a praenomen, see §7.7.1.53.
131. [Quintus or Quintius: quinti Lat 477* (import).]
132. Raec(i)lius. m. Reic ̣[lio] MF 98, reiclio MF 99; reị[cli.] MF 100. G. Giacomelli
(1963:215) pointed to a Latin Raecilius as a parallel, but this gentilicium is not
mentioned in Solin & Salomies 1994: the closest parallels are Latin Raecius (e.g.
in CIL XI.3206 from Nepi) and Etruscan Reicna (8 attestations, all from Clu-
137
Renzetti Marra (1990:331 n.17) points to the fact that the M. Protacius Regulus in this text
is from the Horatia, which was also the tribe of Roman Falerii: see §2.6.2.
271
CHAPTER 7
sium). Kretschmer (1943:158) pointed to Raec- : Graec- (cf. `RaikÒj: “Ellhn He-
sych. r 58 Latte/Hansen). Raec- appears to be limited to the Northern Adriatic and
Istrian coast. Torelli (1967:536) proposed to read MF 99 as p]reiclio, with a gen-
tilicium Praec(i)lius that occurs also in CIL XI.3181 from near Fabbrica di Roma,
but, although possible in MF 98 and 99, this is unlikely in the case of reị[cli.] MF
100, where the name stands at the beginning of the line.
133. S... in (1) ṣ*[---] MF 197; (2) s[---] LtF 173.
134. Sab. abbr. Cap 400. In all probability Sabinus.
135. Sacconius. zaconiọ MF 153, zaconiai MF 154 (gen. or dat.). Latin Sacconius, an
adaptation of Etruscan Σaχu (σ́aχu Cl 1.2499, 1.2500, Pe 1.1175, σaχus OA 2.40,
σ́aχus̀ Pe 1.423, σ́aχus Cr 2.40, σ́acuσ́a Cl 1.1911).
136. [? Saluena. m. salu[e]na (or salu[i]na?) Lat 218. The name apparently occurs
only here in Latin: Etruscan had Salvina (salvinal Cl 1.1643, śalvinei Cl 1.2346).]
137. Sapnonius. f. sapnonia MF 258. There are no direct parallels in either Latin or
Etruscan. The closest parallel is Etruscan Sapu (sapu Cl 1.2358, śapu Cl 3.3, sa-
puσ́a Cl 1.1139, śapuσ́a AS 1.461, Cl 1.2016), which was adapted in Latin to
Saponius, but the formation of this gentilicium is different. Perhaps Sapnonius is
an adaptation of an Etruscan gentilicium *Sap( )nu.
138. ? Sarius. m. sares Cap 404. Latin Sarius, Vestinian saries MV 11. It is not neces-
sary to try to connect the name to that of Mount Soracte, as did Hirata 1967:73.
139. Satellius or Satilius. f. satelie MF 42. Latin Satellius, Satilius, Etruscan Σatna
(σatna-/σ́atna-, 24 attestations, mainly from Clusium, cf. ET).
140. Sedius or Saedius. f. (?) sediu Cap 466. The closest parallel is Latin Sedius or
Saedius (Schulze 1904:93) but in view of the suffix the name appears to be of Sa-
bellic origin and be a monophthongized form of the Sabellic gentilicium Saidius,
which occurs in Oscan saidiieis Cp 9.
141. Sentius. m. senti Cap 399, senti Cap 430; f. senθia MLF 362, perhaps also
[---]nθia MLF 212. Latin Sentius, Etruscan Sente/Σente (21 attestations, mainly
from Clusium).
142. ?Seralius: seralia LF 380. There are no parallels for this gentilicium: could it be a
misspelling for e.g. serania or seratia?
143. Sertinius. f. zertenea LF 222. Like Latin Sertinius, it is an adaptation of Etruscan
Zertna (zertnai Ta 1.52, 53, zertnas Vs 1.205): the z- in Faliscan zertenea proba-
bly reflects the z- in the Etruscan form.
144. Sertorius. m. setorio MLF/Cap 476*. Latin Sertorius, occurring in CIL XI.3181
from Fabbrica di Roma. Probably a patronymic gentilicium from the praenomen
272
THE ONOMASTICON
Sertor (cf. Salomies 1987:46-7), which was apparently of Etruscan origin (ser-
tur/serθur/śertur/śerθur: for attestations, see ET).
145. Seruatronius. f. zeruatronia MF 272. Latin Seruatronius also occurs in CIL
X.8230 from Capua. The name is Etruscan: cf. Schulze 1904:342 on the deriva-
tions in -atronius, Etruscan -atru.
146. Seruius. m. serui MF 34-36 (gen. or abbr. nom.). It is impossible to ascertain
whether serui is a praenomen or a gentilicium: as both PRAENOMEN and GEN-
TILICIUM occur in Faliscan Besitzerinschriften (§7.3), both are possible. – The
gentilicium Seruius has parallels in Etruscan σ́ervei Pe 1.1191, σ́ervi Pe 1.1190,
and in Latin Seruius and Seruilius. It is probably unconnected with the noun se-
ruus (sometimes thought to be of Etruscan origin, cf. Bréyer 1993:383-5).
147. Spurilius. m. spuṛ[ilio LF 248, spurilius Lat 237, spurilius Lat 238, perhaps to be
restored in [---]ilio LtF 215. Latin Spurilius. A patronymic gentilicium derived
from the praenomen Spurius, which is ultimately of Etruscan origin (cf. Salomies
1987:50-6). This derivation in all probability originated within Latin, and al-
though Spurilius can therefore be equated with Etruscan patronymic gentilicia
like Spurinna, it is not an adaptation of an Etruscan gentilicium.
148. Succonius. f. zuconia MF 271, probably also m. zu[con]|eo MF 56, perhaps also
[--- ue?]l su|[con ---] MF 191. Cf. larisa zuχus Etr XXXII. Latin Succonius: the
name is an adaptation of Etruscan Sucu (from Caere: sucus Cr 1.152, 1.155,
1.172, 2.31, sucui Cr 1.100) or Zuχu (mainly from Clusium: zuχu Cl 1.1619,
1.1769, 1.1770, 1.2173, zuχuś Cl 1.1771; zuχus Vs 1.136, zuχ̣uś Pe 1.965). A Soc-
conia Voluptas occurs in CIL XI.3223 from Nepi.
149. T... m. ṭ**(*)[i] MF 84 (gen.). The name can be read as either Ta... or Tri..., but
the tạḷị read by Herbig (1910:101 etc.), which would tie in with the Latin gen-
tilicium Talius, is perhaps too short.
150. Tar... perhaps in uentarc[i..... MF 80, if this can be read as uen tar[...... ‘Ven(el)
Tar...’. Latin and Etruscan have several gentilicia in Tar- (perhaps Tarc- or
Tarqu-?).
151. Tertineius. m. tertinei MLF/Cap 474* (gen.), tertineo LF 213. Tertineius is ap-
parently attested only for Faliscan: Latin has Tertinius.
152. Tetena or Tettena. m. tetena MF 266.
153. Tettius. m. teti MF 11.
Latin Tettius (thrice in CIL XI.2990 from Tuscania, also in Paelignian tettia Pg
16. The name also occurs as a praenomen: see §7.7.1.72. In Etruscan, the gen-
tilicium is Tetina (70 attestations in ET, mainly from Clusium and the ager Sae-
nensis) or Tetna (19 attestations, mostly from Clusium). Cf. also South Picene ti-
273
CHAPTER 7
tienom TE.3? Tettius may be an adaptation of the Etruscan name (G. Giacomelli
1963:223 and Hirata 1967:78-9), but in view of the existence of a praenomen Tet-
tius (cf. Salomies 1987:93), attested from the ager Faliscus in teti MF 13, the
names might be entirely unconnected (cf. above one Marcena and Marcius).
Latin has a number of gentilicia in Tetti-, which need not all be derived from the
Etruscan gentilicium.
154. Tiberilius. f. tiperilia LF 229. A patronymic gentilicium derived from the
praenomen Tiberius, itself derived from the potamonym Tiberis. Note that tiper-
ilia is the Faliscan spelling of the Latin gentilicium Tiberilius: the Faliscan form
would have been *Tiferilios, cf. perhaps tif MLF 460 (either a praenomen or a
gentilicium). The gentilicium is not attested from Etruscan (but cf. the gentilicium
teperi Pe 1.865, teperial Pe 1.875, 1.880) or the Sabellic languages.
155. Tirrius. m. tirio MLF 351, tirio MLF 358; f. ṭịria MF 155. A patronymic gen-
tilicium derived from the Faliscan praenomen Tirrus (see §7.7.1.74). The gen-
tilicium is attested for Latin in CIL XI.3132 from near Civita Castellana - Falerii
Novi: for Etruscan cf. perhaps tiria TC 2, 16, 28, ti[ria TC 37, tiriiai TC 26?
156. Tullius. m. tulio MLF 346; tulie MLF 383. A patronymic gentilicium derived
from the praenomen Tullus (see §7.7.1.77). The gentilicium also appears in Latin
(e.g. CIL XI.3036 and 3037 from near Viterbo): in Etruscan, it occurs as tule Ar
1.94.
157. Turius. m. turi MF 273 (and the abbreviation t MLF 274 on the same vessel); f.
turia MF 22, 23, 24, 25, 26, 27; [and without context probably also tur MF? 44
and tu MF? 38]. Latin has a gentilicium Turius, which occurs e.g. in CIL XI.3038
from near Viterbo and CIL XI.3064 from Horta. There are no equivalents or cog-
nates in Etruscan or the Sabellic languages.
158. [Umbricius. u]mpricius Lat 219. Also in CIL XI.3254 from Sutri.]
159. Umbricianus. m. upreciano MLF 363, upreciano MLF 364.
The attestations of Umbrius and of Umbricianus were found in the same tomb as
Etruscan umrie Etr XLIII. Pace Schulze (1904:258), Stolte (1928:300), and G.
Giacomelli (1963:232), the basis of these names is clearly the ethnonym of the
Umbrians. The Etruscan names umres AS 1.174, umria Cl 1.2620, 1.2621, uṃṛịaś
Cl 1.1294, umriaś Cl 1.2621, and umriś Pe 1.1268 as well as Latin Umbrius (and
Umberius) reflect the ethnonym *Umbros which appears in Latin as Umber and
in Greek as ”Ombroj. Umbricius may have been derived as Umbr-icius within the
Latin onomasticon, but the Etruscan gentilicium Umrce (umrceś AS 1.129,
umrcial AS 1.395) shows that it is more likely to have been derived as Umbric-
ius from a different ethnonym also reflected by Greek ’OmbrikÒj. Neither the eth-
nonyms nor the names are attested for the Sabellic onomasticon.
274
THE ONOMASTICON
160. ? Vatius. f. uatia perhaps in uei uatia MLF 463 (but its companion inscription has
uei ụeto MLF 464, if this is not a falsum). Both attestations are doubtful. If at-
tested, the gentilicium has an equivalent in Latin Vatius.
Ve- see also Vi-
161. [Vecilius. m. uecilio Lat 251 (twice).]
162. Veculius. f. [u]eculia MF 80, u[eculi]a MF 81.
I wonder whether Vecilius and Veculius are not in fact the same name. Latin had
both Vecilius and Vicilius, and Veculius occurs in CIL XI.3843 from Veii. G. Gia-
comelli pointed to an Etruscan gentilicium Vecu, but the expected adaptation of
that name would be Veconius/Viconius (which is in fact attested for Latin, see be-
low under Vicon(i)us). Whether Vecilius/Veculius is identical with Vicin-
ius/Vecinius (thus Stolte 1928:299) is another matter.
163. ? Vei... (Veianius?). uei|[---] LtF 327, perhaps also uei LtF 205 (abbr.). Both at-
testations are very unclear: the name may be Veianius, attested in CIL XI.3197
from Nepi, CIL XI.3805 from Veii, lazi veianes Etr XI-XIV and lazi veiaṇes Etr
XV, and in Varro (“fratres Veianii ex agro Falisco” R. 3.16.10),
164. Vel. abbr. uel Cap 424.
165. Velminaeus (or possibly Velmineius or Velminius). m. uelmineo MLF 305,
uelmineo MLF 307, uelminẹo MLF 308, uelmineo MLF 309, uelmineo MLF 310,
uelmineo MLF 312, uelmineo MLF 313, uel|mineo MLF 315, perhaps misspelled
uelmi|no MLF 316, [u]elṃi[ne---] MLF 317. The name is in all probability an
adaptation of the Etruscan gentilicium Velimna (velimna Pe 1.142 etc.: 23 attesta-
tions, from Clusium and Perusia). For the curious metathesis in Etruscan velimn- :
Faliscan uelmin-, G. Giacomelli (1963:228-9) rightly compared Latin Volminius
CIL VI.21470 and the more common Volumnius, both likewise derived from
Velimna. She also pointed to an Etruscan inscription from Veii (now Ve 3.19),
]ẹḷṃẹṇaie muluvanic[, which can be restored as v]ẹḷṃẹṇaie (cf. Ribezzo
1931c:93-4).138 From the Faliscan instances it is unclear whether the ending -eo in
uelmineo represents /-ios/, /-ēos/
̣ ← /-os/, or /-ę̄os/ ← /-os/: see §3.7.6.
166. ? Venelius. uenelịes MF 258. A patronymic gentilicium derived from the Etrus-
can praenomen Venel, possibly attested also from the ager Faliscus (see
§7.7.1.83). Latin Venilius, and a direct parallel in Etruscan venelies Vc 2.10.
167. ? Ventarc... uentarc[...... MF 80. Unclear: there are no parallels from the Latin,
Etruscan or Sabellic onomasticon that can be restored, nor is it clear if the name is
really a (second?) gentilicium. Perhaps it can be divided as uen tarc[......: see Tar-.
138
ET gives the text as [-?- h]eṛmenaie muluvanic[e -?-], which is equally possible (see the
drawing in Giglioli 1930:307). Both Velmenaie and Hermenaie do not occur elsewhere.
275
CHAPTER 7
168. Vestius. m. uesθi MF 83 (abbr.). Latin Vestius, perhaps of Etruscan origin.

169. [Vettius. m. uettius Lat 456.]
170. Vetulius. f. uetulia LF 334 (and m. uetulio LF 336, if not a falsum). The name has
parallels in Latin Vetulius, Etruscan vetlnei Pe 1.336 and vetlnal Cl 1.1467.
Vi- see also Ve-
171. Vi. abbr. ui Cap 414.
172. Vicina. m. uicina MLF 371, uicina MLF 372. There are no attestations of this
name in Etruscan: the closest cognate is veicnas Vs 1.203, veceṇes Ta 7.29. In
spite of the different vocalism, this name may therefore be identical with Vicin-
ius/Vecinius.
173. Vicinius or Vecinius. m. uecineo LF 220, uecineo LF 224, uecineo LF 225; f.
uecin[e]a LF 222, uecinea LF 223. G. Giacomelli (1963:226) equated the name
with Latin Vecenius, an adaptation of Etruscan Vecena (veceṇes Ta 7.29), but the
name may be identical with Latin Vicinius and an adaptation of Etruscan Veicna
(veicnas Vs 1.203). If so, it could be identical with uicina MF 371, 372 (see
above under Vicina).
174. Vicon(i)us: ueicọno MF 88, perhaps also in tuconu MF 85, if this can be read as t
u(e)conu. Latin has Veconius, Vecconius, and Viconius. The names are adapta-
tions of the Etruscan gentilicium Vecu, attested for Clusium (vecu Cl 1.843,
1.844, 1.845, 1.846, 1.847, 1.851, OI S.52, vecui Cl 1.848, 1.849, vecuσ́a Cl
1.852, vecus Cl 1.850). For a Faliscan Veconus beside Latin Veconius, see §7.8.2.
175. ?Vinu... possibly in [---] uịnụ[---] MLF 365. Latin Vinucius, Etruscan vinucenas
Vs 1.126, Oscan viínikiís Po 3, all apparently derived from the praenomen at-
tested in Praesamnitic vinuχs Ps 3 and the patronymic adjective viniciiu Ps 3.
176. ? Viui... perhaps in uiụị[---] MF 157.
177. Vis(i)nius. m. uisni MF 82 (abbr. or gen.). The name has parallels in Latin Visin-
ius (e.g. CIL XI.3614 from Caere) and Vesnius, and in Etruscan Viσ(e)na/ Viσne
(viσenas Vs 1.9 and viσnai Vc 1.31, viσnalś Vc 1.92, viσnei Vc 1.53).
178. Vo. abbr. uo MF? 29, Cap 429.
179. Vollius. f. uolḷia MF 47; uoll[---] MF 86. Other attestations of this name are per-
haps uli MF? 261-262, if this can be read as u(o)li. Latin likewise had a gen-
tilicium Vollius or Volius.139 The spelling with ll is surprising (cf. §11.2.5.5 and
§3.5.3.3): cf. Lullius.
139
I fail to see why Hirata (1967) included this name under Volta: even if Vollius and Volta
are ultimately derived from the Etruscan praenomen Vel, they are derived in entirely different
ways, and Vollius can in no way be considered to be a regular derivation from Volta.
276
THE ONOMASTICON
180. Voltaeus. f. uoltaia MF 196. A gentilicium derived from the Faliscan praenomen
Volta (see §7.7.1.85) but not as a patronymic gentilicium, since the patronymic
adjective from Volta is Voltius (§7.5.2): see G. Giacomelli 1962.
181. ? Voltius. ulties MF/Etr 64. G. Giacomelli (1963:232) and Hirata (1967:89) com-
pared Etruscan Ulθe (ager Saeniensis and Perusia) and Velθe (only in the Liber
Linteus), remarking “ha affinità con volta”. In my view, the name may be read as
u(o)lties (with a graphical contraction, see §11.2.5.8) which, as the form occurs in
isolation, can be interpreted either as a patronymic gentilicium Voltius or as the
praenomen Voltius (for which see §7.7.1.86).
182. Vomanius. m. uomanio Cap 388. Vomanius, which also occurs in Latin (e.g. CIL
XI.3338 from Blera), is perhaps derived from a potamonym Vomanus (Schulze
1904:481): the nearest river of that name known to modern authors is part of the
Po estuary, however, a long way from the ager Faliscus and Capenas. See also
§6.5.1. Other potamonymic gentilicia may be Fa(r)farn- and Narionius.
183. ACEPHALOUS FRAGMENTS (consisting of more than the endings [---]o, [---]io,
[---]eo, [---]a, [---]ia, or [---]ea: *(*)coṇẹo LtF 290 (perhaps ạcoṇẹo, see under
Aconius), *e[0-4?]i*ia LF 235, *i[....] MLF 353, [.]osena MLF 206, [..]lni[a]
MF 146, [....]nea LtF 301, [....]ta MF 146, [---]fate MLF 285 (gen. of a probably
toponymic gentilicium in ...fas), [---]iena MF 102, [---]ilio LtF 215 (perhaps
spur]ilio, see under Spurilius), [---]lio MF 137, [---]nio LtF 341, [---]nθia MLF
212 (perhaps se]nθia, see under Sentius), [---]rcius Cap 435 (e.g. mar]cius?),
[---]ronio MF 156 (probably ne]ronio, see under Neronius).
7.8.2. The origins of the Faliscan gentilicia. Many of the gentilicia in the list in
§7.8.1 can be placed into one of several clearly recognizable categories (for an over-
view of the derivational suffixes, see G. Giacomelli 1963:132-49). Some of these
categories can be connected with an origin of the name in a specific language, either
Etruscan, Latin-Faliscan, or a Sabellic language. As has been said in §7.1.1, one of the
major motives of looking at the onomasticon in a linguistic study is that the onomasti-
con can provide (socio)linguistic data on the ethnic background of the population, and
its contacts with the areas around it. The first problem here is that it is often difficult
(to say the least) to ascribe the origin of a gentilicium to a specific language, as has
been explained in §7.1.1. Although I have divided the gentilicia into groups according
to their most likely origin, it should be stressed that in many cases attribution is very
uncertain. The inferences that may be drawn from these data are discussed in §7.10.3.
(1) patronymic gentilicia in -ius, -ilius and -idius (cf. §7.5.2). This is clearly the
category in which most of the gentilicia belong that can be placed in a specific cate-
gory. This category can be subdivided according to the praenomina from which they
are derived (see also §7.7.2): (a) from Faliscan praenomina: Aufilius/Oufilius, Iunius,
277
CHAPTER 7
Laeuius and Laeuilius, and Voltius; (b) from praenomina that occur both in Latin and
Faliscan: Lucilius (only on an import), Marcius, Quinctius (only on imports), (c) from
praenomina that are Latin rather than Faliscan: Spurilius (from an originally Etruscan
praenomen), Tiberilius (note the -b-); (d) from Sabellic praenomina: Fertorius,
Neronius, Petronius, Tettius (?), Vinu... (?), possibly also Pu(m)ponius, if this is in-
deed a gentilicium (§7.7.1.52), and (Etruscan?) Marhius; (d) from praenomina com-
mon to both Latin/Faliscan and the Sabellic languages: Genucilius (only on an im-
port), Manius and Manilius, and Tullius; (e) from Etruscan praenomina:
Aratius/Arantius, Caelius, Lartius, Sertorius, and Venelius; (f) from praenomina of
uncertain origin: Caesius, Graec... (Etruscan?), and Tirrius; (g) patronymic gentilicia
with the Sabellic suffix -idius: Aiedius, Clanidius, and perhaps Didius.
In view of the suffixes with which these gentilicia are derived, they are clearly
of Italic origin, but they are of a type that is so common in the Italic languages that it
is in many cases impossible to ascribe gentilicia of this type to any specific Italic lan-
guage. Although it is probable that many originated in the language to which the
praenomina belonged from which they were derived, this is by no means necessary: a
well-established name like Spurilius is derived from a praenomen of Etruscan origin,
but one that also occurred in the Latin onomasticon, and the derivational suffix is
clearly Latin-Faliscan. Patronymic gentilicia derived from Faliscan praenomina will
have originated within the ager Faliscus: these are Aufilius/Oufilius, Iunius, Laeuius
and Laeuilius, and Voltius. The gentilicia that are formed with the suffix -idius that is
the Sabellic equivalent of Latin -ilius may be assumed to be of Sabellic origin.
(2) Etruscan patronymic gentilicia in -na and their adaptations: Aulena, Cotena,
Eina, Hac...na, Fescuna (if this is not a toponymic gentilicium), Hermana, Marcena,
Nomesina, Pleina, Salu[e?]na, Tetena/Tettena, Vicina, perhaps also Early Faliscan
Capena (§7.2.2); and, adapted from such names, Licinius, Vicinius, Visinius, Aenus,
Faenus, Poenus, and perhaps the Early Faliscan names Amanus, Capena, and Volte-
nus (§7.2.2). These names constitute a clearly recognizable Etruscan type, and may
have been recognized as patronymic even by those who did not speak Etruscan (much
like Scottish names in Mac... or Irish names in O’... are recognizable as patronymic to
speakers of English that know no Gaelic), especially since they often contain recog-
nizable praenomina (e.g. Aulena, Marcena, Nomesina, Tettena).
Most of these names remain unadapted both in Latin and in Faliscan: in accor-
dance with their nominative in -a, such names could be declined according to the first
declension (§4.2.1). Note that Faliscan had two frequently occurring local male
praenomina in -a, Iuna and Volta, which may have made it even easier to adopt Etrus-
can gentilicia in -na without adaptation of the suffix. On the other hand, if required,
such names could be adapted. In Latin, this was usually done by means of -na →
-inius, and examples of such adaptations in the Faliscan onomasticon are Licinius and
Vicinius (occurring beside Vicina) and probably also Visinius. In Faliscan, there are
278
THE ONOMASTICON
also cases where such gentilicia were thematized without further adaptation of the
suffix, as -na → -nus: possible examples of this are Aenus, Faenus, Poenus, and per-
haps Early Faliscan Amanus, Capena, and Voltenus (see §7.2.2).
(3) other names derived from praenomina: Arruntulus, Arruntielius, Voltaeus. These
names are derived from praenomina, but not with the -ius and -ilius of the patronymic
derivations (cf. §7.5.2). The derivational suffixes of Arruntulus and Arruntielius are
Italic, although they are derived from Etruscan Arruns. Voltaeus on the other hand is
derived from Faliscan Volta, but with a suffix that recalls Etruscan names such as
Velminaeus, Letaeus from the ager Faliscus and other Etruscan names in -aie: G. Gia-
comelli (1962) regarded the suffix as Italic, however.
(4) adaptations of Etruscan gentilicia in -u. These names are a category that does
present morphological difficulties, and therefore had to be adapted in some way in
order to be declined. Apparently, the nominative in -u was comparable to the nomina-
tives in -o of the ōn-stems in (§4.5.1.3), and the usual way of Latinizing these gen-
tilicia was therefore by means of -u → -onius. Faliscan examples of this are Aconius,
Atronius, Fullonius, Sacconius, Sapnonius, Seruatronius, and Succonius, and possibly
Decon.... Yet Faliscan has two names where the Latinization was apparently by means
of -u → -onus, Viconus and perhaps also Praeconus. This adaptation is in a sense
comparable to the adaptation -na → -nus described above under (2). Related as well
appears to be Folcosius, with -u → -osius. A different, and simpler type of adapting
these names, occurring also in Latin, is -u → -ius, as in Cincius. Unclear is Laepuius:
it looks as if this was adapted by means of -u → -uius.
(5) geographical: (a) toponymic: Abellensis, Acarcelinius, Feliginas, Fescuna (?),
Orticensis, Veianius (?), perhaps also Calitenus and Egnatius, and the ...fas in
[---]fate MLF 285; (b) potamonymic: Fa(r)farn..., Narionius (?), Vomanius (and,
indirectly, Tiberilius); (c) ethnonymic: Grae... (perhaps rather patronymic?), Latin-
aeus, Sab(in-), Umbrius, Umbricius, and Umbricianus. Several of the toponymic and
potamonymic gentilicia are connected with local toponyms and potamonyms. Note
that Abellensis and Tiberilius have a -b- that can only point to a (Roman) Latin origin,
while in the case of Fa(r)farn- (if connected with the potamonym at all), the name is
derived from the apparently local name Farfarus, not from the Latin form Fabaris.
The ethnonymic names, too, mainly reflect the peoples inhabiting the areas surround-
ing the ager Faliscus and the ager Capenas: the exception is Graec..., but this may be
a patronymic gentilicium derived from an older praenomen Graecus. With regard to
the derivation, note the Etruscan toponymic adjective suffix -te/-ti in Feliginas and the
...fas in [---]fate MLF 285, and perhaps also in Calitenus and in Egnatius.140
140
Note also alsi*is Etr XL, which, if to be read as alsiṭis, is probably a toponymic name de-
rived from Alsium. The suffix -te/-ti is also used in the toponymic adjective Capenas.
279
CHAPTER 7
(6) Berufsgentilizen: Clipearius and Frenarius. This is a rare group in the Faliscan
onomasticon, and both names are not attested elsewhere. The nouns from which they
are derived, clipeus and frenum, occur also in Latin, although clipeus may be of
Etruscan origin (§6.2.9). The derivational suffix is Italic.
Whereas these gentilicia are more or less recognizable, there remains a large group
that cannot be ascribed so easily to one language or the other. I have divided these
according to the languages in which they have most parallels.
(7) gentilicia that appear to have only local parallels: Protacius, Turius.
(8) gentilicia that have parallels in both Etruscan and Latin: Annius and its deriva-
tion Annilius, Calitenus (toponymic?), Calpurnius, Catineius, Colanius, Egnatius
(toponymic?), Fabius, Fadius, Fadenius, Firmius, Lullius, Polfaeus, Satellius, Sen-
tius, Sertinius. Most, if not all, of these names appear to be of Etruscan origin, but
since they are attested for the Latin onomasticon as well, some may have reached the
area through the Latin onomasticon rather than in their Etruscan form: the adaptation
of these names may therefore have taken place elsewhere than in the ager Faliscus.
(9) gentilicia that have parallels only in Etruscan: Letaeus/Lete(i)us, Velminaeus/
Velmine(i)us. Both are adaptations of Etruscan gentilicia in -aie, and are not attested
outside the area. Probably also of Etruscan origin are Acr- (?), Am-/Amm-, Arn.
(10) gentilicia that have their main parallels in Latin: Acciuaeus, [Claudius], Cu-
trius (although the vocalism points to an Etruscan intermediary), Flauilius (patro-
nymic?), [Fuluius], [Furius], Maecius, Mallius, [Munius], Tertineius, Vestius (of
Etruscan origin?), Vettius (of Sabellic origin?), Vecilius, Veculius, Vetulius.
(11) gentilicia that are probably of Sabellic origin: Alliuaeus, Battius (?), Blaesius,
Fassius, Hirpius, Pacius, Pescennius, Sedius/Saedius/Saidius. Most of these occur in
inscriptions from the ager Capenas: see §9.3.2
(12) gentilicia that are probably Italic rather than Etruscan: Arrius, Calinius, Cocil-
ius, Plarius, Popi. and Popilius, Pupius and Pupilius, Sarius.
(13) gentilicia of indeterminate origin: Adicius, Aieius (?), Aufitius, Citius (Etrus-
can?), Latrius (Etruscan?), Lurius, Neln-, Oscin., Pani..., Panur..., Partius, Pauicius,
Raec(i)lius, Seralius, Vatius, Ventarc... (?), Viui... (?), Vollius. Several of these gen-
tilicia are of dubious attestation.
7.9. The cognomina
7.9.1. Attestations of cognomina in the Faliscan inscriptions. There are several

attestations of cognomina from the Faliscan inscriptions. All are from sepulchral in-
280
THE ONOMASTICON
scriptions and occur solely in combination with the onomastic formula PRAENOMEN
GENTILICIUM. The attestations are:
1. Maxumus (5-7 attestations):141 [4-5]ạ hac****a  [?]ạ[?]ṃ  maximo MF 89,
with FILIATION? COGNOMEN, [---] reic ̣[lio] | [---] ṃaxoṃ[o] MF 98, [---ma]x̣ọmo
ᛂᛂ uoltilio MF 162, with COGNOMEN FILIATION, cauio  nomes|ina  maxomo MF
272, uoltio  uecineo | maxomo | iuneo LF 220, with COGNOMEN FILIATION; per-
haps also (either Maxumus or Manumus) in [leu]elio  cailio [ ...| ma.]om[o ] rex
[ ..]** MF 90, probably with FILIATION COGNOMEN, and perhaps **xi[..] in the
very fragmentary MF 91 (the titulus posterior of MF 90) which can be little else
than ṃạxi[mo. Maximus ‘the Greatest’ is a cognomen indicating power or suc-
cess, but it is surprising to find this cognomen in at least four different families.
This may suggest that the adoption of this cognomen was not due to some ex-
traordinary achievement, but rather to a feat that was attained with some regular-
ity by deserving individuals, e.g. membership of a ruling body or functioning in a
specific high magistracy.
2. Manumus (2-3 attestations): [ma]rco  pleina  marcio  man[o]mo MF 80, with
FILIATION COGNOMEN, [u]olṭa  pupelio | [m]ano[m]o MF 149; perhaps also (ei-
ther Maxumus or Manumus) in [leu]elio  cailio [ ...| ma.]om[o ] rex [ ..]** MF
90, probably with FILIATION COGNOMEN. Manumus, ‘the Best’ or perhaps rather
‘the Most Good’, on the other hand, may have had a sacral connotation (cf. §6.2.1
s.v. manus).
3. ?i*ice (2 attestations?): tito ᛂ uelmineo | nu i*ice MLF 309 with FILIATION COG-
NOMEN?, tito  uel|mineo  iun|aị ị*ice MLF 315 with FILIATION COGNOMEN?. It is
unclear if i*ice is a cognomen: most editors have interpreted it as a verb (see
§5.2.1.9, §6.2.38), but this is partly due to the interpretation of iun|ai in MLF 315
as a dative rather than as a genitive. If it is a cognomen, it is unclear how i*ice is
to be read: neither iice nor iḍice seems to produce a promising cognomen
(I(m)ige(r)? I(n)ḍige(ns)?).
4. ? Ruso (2 attestations?): cauio[---] | ruso[?---] MLF 318, perhaps to be restored
in ce  paui[ceo ru?]so LtF 290. The cognomen Ruso is well-known from Latin
sources (cf. Solin & Salomies 1994:394) and occurs in CIL XI.3254 I.13 from Su-
tri. It is probably derived from russus/rūsus ‘reddish’, with the -o that is fre-
quently used in derivation of Latin cognomina: assuming an Etruscan connection
(G. Giacomelli 1963:216) is not necessary. Both attestations are doubtful, how-
ever: note that reading ru]so in LtF 290 would provide the only instance of a cog-
141
I cannot adopt Colonna’s (1972c:446-7) interpretation of [---]ronio  uol[t---|---]a*ome
MF 156 as containing a cognomen m]ax̣ome: this would be the only instance of mo-
nophthongization of /-ai/ to /-ę̄/.
281
CHAPTER 7
nomen in a public inscription from the ager Faliscus at a time when this was rare
even in Latin texts (cf. Kajanto 1977a:67).
5. Previous editors have read several other cognomina in the texts. Of these, the fol-
lowing can be rejected for reasons discussed under the individual inscriptions:
cela MF 12, MF 83, MF 84, cẹla MF 166 (all instances of the noun cela ‘cella’);
rezo MF 56 (fragmentary text); uentarc[…… MF 80 (rather a name of another in-
dividual); θania MF 81 (a woman’s name), tuconu MF 85 (unclear, but if it is a
cognomen, it would be a name consisting only of COGNOMEN occurring in isola-
tion), [---]ono MF 102 (perhaps rather a genitive in [---]ono(s)), kreco MF 147 (a
praenomen), sus[?---] LF 227 (unclear, perhaps s us[o(r)]?), sorex LtF 231 (a
ghostword), ***io LF 332 (badly legible, probably a patronymic adjective), esχ
Cap 389 and 404 (to be read as a verbal form esú), posticnu MLF/Cap 474* (a
Sabellic noun, cf. South Picene postiknam CH.2), and velusa Etr XXXIV (rather
an Etruscan genitive). Still worth considering perhaps are putellio MF 152 (rather
a gentilicium?) and abbreviated sen Cap 435 (rather a praenomen?). Note that G.
Giacomelli (1963) and Hirata (1967) sometimes appear to use the term cognomen
also for a second gentilicium: for which see §7.6.
7.9.2. The chronology of the Faliscan cognomen. Rix (1965 passim, cf. 1965:379-80)
has shown that in Etruscan cognomina started to appear in the fifth century and became
current for men from the third century onward: cognomina for women remained scarce
in most of Etruria, except at Clusium (Rix 1965:40-2). In Rome, cognomina may have
made their first appearance in the fifth century, but they remained a prerogative of the
patrician families until well into the second (Kajanto 1977a:64-7).
The scarcity of cognomina in the Middle and Late Faliscan inscriptions may be
due to the fact that they were a new feature in the ager Faliscus as well or that they were
limited to a specific group, as in Rome. Also, the occurrence of the filiation after the
cognomen in [---ma]x ̣ọmo ᛂᛂ uoltilio MF 162 and uoltio  uecineo | maxomo | iuneo LF
220 may indicate that it did not yet have a fixed place in the onomastic formula. Neither
are there secure indications that Faliscan cognomina were hereditary, apart from the
occurrence of i*ice in MLF 309 and MLF 315, both from the gens Velminaea. An-
other possible but even more dubious instance would be ruso, if MLF 318 is to be re-
stored as cauio [ pauiceo] | ruso[?---], and LtF 290 as ce  paui[ceo  ru?]so LtF 290.
Although there is no evidence of the existence of something resembling the Ro-
man patrician class in Faliscan society, there are a few indications that the Faliscan cog-
nomina may have been associated with high social status. Thus, [leu]elio  cailio [ ...|
max/man]om[o ] rex [ ..]** MF 90 shows not only a cognomen but also the (sacral?)
title rex, and high status has also been assumed in the case of uoltio  uecineo | maxomo
| iuneo LF 220, even apart from the fact that Maximus is in itself a name implying
282
THE ONOMASTICON
greatness.142 If Faliscan cognomina were not yet hereditary, the repeated occurrence of
Maxumus would then point to individual rather than to family status, perhaps referring
to some notable (political or military?) achievement or success. This might partly ex-
plain the occurrence of Maxumus as a cognomen of individuals of at least four differ-
ent families.
There are no clear instances of cognomina from the Late Faliscan period apart
from uoltio  uecineo | maxomo | iuneo LF 220. Harisp… in c  clipeaṛ[io] | m  f 
harịṣ[ex LtF 231 and [---]| harisp[---|---]sor LF 232 is a priestly title than a cogno-
men. Apart from this, there is only the doubtful instance of Rutilus in m  aco[---]| rutil
 ce[---] LtF 341. Even sepulchral inscriptions with quite extensive cursus honorum
from this period yield no cognomina, and neither do the public inscriptions from Falerii
Novi and along the roads of the ager Faliscus.143 This absence of cognomina after the
war of 241-240 could imply that as far as cognomina were concerned, the Faliscan
onomasticon was adapted to the Roman usage of the period, where cognomina were
still a patrician prerogative, and did not regularly appear in public inscriptions.
If that is true, the Latin inscriptions from the area from the second half of the sec-
ond century and later might be expected to contain more cognomina, for around this
date the cognomen also begins to appear in the names of plebeians and freedmen in
inscriptions from Rome and Latium. Cognomina indeed appear in the dedication [.]
ṃunio regena* | numesio  m[art]ẹ | d d l m Lat 377, where regena* may well be a
cognomen, in the sepulchral inscription of Pu(blius) Fuluius C(ai) f(ilius) | C(ai)
n(epos) Suto(r) Lat 250 (106 BCE), which also has a very formal double filiation,144
and in the dedication [.  u]mpricius  c  f | [?]aburcus  q  | [ap]olinei  dat Lat 219
(c.120-50 BCE).
7.10. The Faliscan onomasticon and the question of identity
7.10.1. Names as markers of identity. As was said in §7.1.1, names are markers of
identity, not only the identity of the person they refer to, but also of the ramifications
of that person’s identity with regard to gender, family, social group, ethnic group, etc.
142
On an epigraphic level, note that [ma]rco  pleina  marcio  man[o]mo MF 80 and [4-5]ạ
hac****a  [?]ạ[?]ṃ  maximo MF 89 are both decorated with a painted border, an exceed-
ingly rare feature in Faliscan sepulchral inscriptions (§11.1.4.1c). On the other hand, cauio 
nomes|ina  maxomo MF 272 was scratched on a tile in a very careless manner.
143
Kajanto (1977a:67) explained the absence of cognomina in Republican public inscriptions
as a relic from the time when cognomina were not yet a regular part of the name.
144
In accordance with what was said in the preceding note, the names of the consuls still ap-
pear in this inscription without cognomina as C(aio) Atilio (et) Q(uinto) Servi‹li›o.
283
CHAPTER 7
In societies where family names exist, such as those of ancient Italy, a distinc-
tion should be made between gentilicia and praenomina. Gentilicia are inherited: they
express the relationship to the family, and the social or ethnic group from which this
family comes. Praenomina are given, that is, chosen by a parent, and especially in an
area like the ager Faliscus, with its heterogenous ethnic make-up, they can represent a
choice for a certain identity. From this perspective it could be said that the gentilicium
represents the origins of the family, which remain unchangeable and may be of great
importance in a society that values tradition and the ancestors, while the praenomen,
which may change from generation to generation, can denote the direction in which
the family is heading, and of which group they want their children to be a part.
A third point by which identity, especially social identity, may be established is
by the onomastic formula: for the period and the area that is the subject of this study,
this means e.g. the use of double gentilicia, the use of cognomina, and the differences
between the onomastic formula of men and of women.
7.10.2. The onomastic formula and social status. The onomastic formula can denote
social groups and the importance of an individual or section of the group with regard
to the other members.
First, there are specific formula for freedmen and -women (§7.6), where Falis-
can appears to follow the Etruscan usage of giving a double gentilicium in some texts
(MF 82, 346), but the Latin usage of naming the former master with the word libertus
or liberta in others (MF 155, 165). Apparently, it was important that freedmen and
freedwomen were recognizable by a distint onomastic formula.
A second point involving both the onomastic formula and social status is the use
of the cognomen (§7.9). Whether or not it was restricted to a specific group, as it was
in Rome, where cognomina were long a prerogative of the patrician families, cannot
be established. There are indications, however, that it was associated with high status,
and that it may have depended on certain individual (political?) achievements.
Thirdly, there are the onomastic formulas of women (§7.4). It is noteworthy that
these are more liable to variation than those of men, and that it is more usual for a
woman than for a man to be designated by one name only, in which case there is a
preference for using the gentilicium – unlike in the case of men, where the praenomen
is preferred. This may foreshadow a tendency observable also in Rome, where the
importance of the praenomen as part of the (official) onomastic formula appears to
have been on the decrease during the late Republican period (cf. Kajava 1994:114-
24). The fact that there are no instances of Faliscan women having cognomina is also
significant, although it does not set the ager Faliscus apart: during the Middle and
Late Faliscan period, women’s cognomina were very rare in Etruria, except for the
area of Clusium, and probably still non-existent in Latium (§7.9.2).
284
THE ONOMASTICON
7.10.3. Gentilicia and the question of ethnic origin. Having established, albeit in a
very tentative way, the linguistic origins of the gentilicia that occur in the Faliscan
inscriptions (§7.8.2), it is time to look at what implications can be drawn from these
data. A number of provisos have been made earlier (§7.1.1), and these severely limit
the inferences that can be made. One thing that can safely be said, however, is that the
majority of the Faliscan gentilica appears to be of Etruscan origin, while only a very
small number of names can with some certainty be regarded as local: the long persis-
tence of the patronymic adjective in the ager Faliscus may imply that the development
of the gentilicia proceeded at a slower rate than elsewhere. However, in view of the
gentilicia, a large number of the inhabitants appear to be of Etruscan descent at least.
The point, however, is what this means from an ethnic and linguistic perspec-
tive. As said in §7.1.1, as long as there is nothing to show that people with Etruscan
gentilicia were (still) Etruscan in the sense that they were regarded as Etruscan or
thought of themselves as Etruscan (in whatever sense they defined this), this means
little. In what sense were families with Etruscan gentilicia Etruscan? Were there fami-
lies that were obviously Etruscan to their neighbours, or families whose connection
with what modern scholarship calls Etruscan was only very slight? Posing the ques-
tion from a linguistic perspective: is it safe to assume that such families spoke Etrus-
can as their first language, and, more importantly for the linguistic study of the area as
a whole, that they spoke Etruscan as their first language while the majority of people
with a non-Etruscan gentilicium did not?
The answer to most of these, and similar, questions can at best be a surmise, at
worst an assumption. If the Roman onomasticon is reviewed in a similar way, it can
likewise be said that a great number of the Roman gentilicia were of Etruscan origin.
Yet it is obvious that fourth- or third century Rome was not an Etruscan city at least in
the linguistic sense: as Cornell (1997) has shown in his article on ethnicity in early
Rome, from the earliest time onwards, Roman was something that you could become.
Whatever the original identity of a person or that person’s family, that identity could
be changed: not for nothing he quotes the story of Tarquinius Priscus, son of a Greek
potter, married to an Etruscan wife (Liv. 1.34), bearing an Etruscan name, who even-
tually became a Roman king. In other words, gentilicia may be markers of the origi-
nal identity of the family, but the individual members of later generations, although
they might be proud of their ancestry or pay some form of respect to it, may not nec-
essarily have regarded themselves as belonging to that original identity.
In this respect, I think it is noteworthy that the great majority of Etruscan gen-
tilicia were adapted, where necessary, to the requirements of Faliscan (§7.8.2): indica-
tions that they remained phonetically, phonologically, or morphologically different
are very scarce. An exception are the names in -na, that show a larger number of in-
stances where the name remains unadapted, as happened in the Latin onomasticon. In
general, however, unadapted names are generally found only in the Etruscan inscrip-
285
CHAPTER 7
tions from the area, while there are hardly any instances of Faliscan names appearing
in Etruscan inscriptions from the area (§9.2.2), implying that the adaption of names
worked in one direction only.
Although the majority of gentilicia was therefore of Etruscan origin, the lan-
guage of the area was Faliscan, and families with Etruscan gentilicia on the whole
used Faliscan and Faliscan forms of their names at least in the epigraphic texts, what-
ever they may have spoken or written in other contexts (§9.2.1). It is also noteworthy
that with few exceptions (see below), in contrast to the gentilicia, the majority of the
praenomina in the area was local and not Etruscan, whatever the onomastic back-
ground of the families involved (§7.10.5).
In some cases, however, there are indications that the bearers of Etruscan names
were indeed ‘Etruscan’ in one or more senses of this word. This is most notable in a
group of inscriptions from Corchiano that show a number of Etruscan traits (§9.2.2.3),
the most important ones being
poplia  calitenes | aronto  cesies | lartio  uxor MF 265
ueltur  tetena | aruto MF 266
arute macena | morenez MF 269
poplia | zuconia MF 271
cauio  nomes|ina  maxomo | zeruatronia MF 272
Within this group, not only are all gentilicia Etruscan, and mostly unadapted (§7.8.2),
but the praenomina are predominantly Etruscan, too (§7.10.5); furthermore, the in-
scriptions show linguistic features such as -e in arute and larise and -es/-ez for female
names (§9.2.2.1), orthographical features such as the use of z- (§3.5.3), and epigraphic
features such as the fact that these inscriptions were scratched (§11.1.4.1c), which
may all be considered Etruscan or at least non-Faliscan.
Peruzzi (1990) has shown in his study of the gentilicia in these inscriptions that
the families named may have been recent immigrants from the area of Clusium. Yet
even these families use the Faliscan alphabet, language and textual formulas: it is not
their gentilicia, but the gentilicia occurring in a context of other features that makes it
possible for the modern scholar to regard them as more Etruscan than others.
7.10.4. Praenomina and the family. One bond that can be expressed by praenomina
is that with the family, especially in cases of a praenomen that ‘runs in the family’,
children being named after a relative, often the grandfather (cf. Salomies 1987:378-
88). The Faliscan family tombs and the frequent use of PRAENOMEN GENTILICIUM give
some insight into this usage for the ager Faliscus. A good example is the third-century
tomb of the gens Velminaea at Vignanello, where the following members of the gens
can be identified:
286
THE ONOMASTICON
Gavius cauio uelminẹo | popliai file MLF 308

Iuna son of Titus iuna uelmineo | titio MLF 307
Publius popli[o] | uelmi|no MLF 316
Publius or Publia popl[---]| [u]elṃi[ne---] MLF 317
Quinctus cuicto uelmineo |[---?]uoxie[.]eai MLF 310
Titus son of Iuna tito  uel|mineo  iun|aị i*ice MLF 315
Titus son of Nu. ? tito ᛂ uelmineo | nu i*ice MLF 309
Titus son of Titus tito  uelmineo | titọi MLF 305
Volta uolta  uelmineo MLF 313
Voltius son of Titus uoltio [] uelmineo | titio MLF 312
Even within this small group there are at least three and possibly even five different
men called Titus: furthermore, there is both a ‘Iuna son of Titus’ and a ‘Titus son of
Iuna’, implying that praenomina could skip a generation, as they did in Rome.
A similar picture occurs from the gens Aufilia. The following members of this gens
were buried in a family tomb at Falerii Veteres:
Caesius son of Iuna ḳai[s]i[o  auf]ilio  iun[?eo] MF 51
Gavius cauio  aufilio MF 49
Gavius caui[o]  auf ̣ilio MF 50
Iuna iuna  oufilio MF 48
Iuna or Volta? [---]a  aufi[lio MF 53
Two other members of the gens, apparently brothers, were buried at Corchiano:
Caesius son of Volta ceisio  oufilio | uolθeo MF 276
Gavius son of Volta cauio  oufilio | uolteo MF 275
Finding more than one Gauius is perhaps not very surprising, since this praenomen oc-
curs with great frequency (§7.7.1-2), but three within such a small group may be sig-
nificant. Two other men are called Caesius, which is also a frequent praenomen, al-
though not as frequent as Gavius (§7.7.1-2). Interestingly, of the two sons of Volta from
Corchiano, one is called Gauius and the other Caesius, making it even more probable
that these were the preferred and perhaps hereditary praenomina within the gens. Also,
there appear to be either two Iunae or two Voltae in the family.
Less clear instances are the following:
(a) The gens Spurilia shows both a Marcus son of Gaius (m  spurilius  c  f Lat 237)
and a Gaius son of Marcus (c  spurilius | m  f Lat 238), both from Falerii Novi, one
presumably being the father of the other.
(b) Among the five members of the gens Folcosia whose praenomina are known
(cẹịṣ[i.] | holc[osi] | ar  p[...] MF 140 from Falerii Veteres, and sesto ᛍ | fulczeo LF
329, uoltio | folcozeo | zextoi | fi LF 330, cẹḷio *olcuzeo | ***io LF 332, cesio folcuso
LF 331 from Carbognano-Vallerano) there are perhaps two Caesii.
287
CHAPTER 7
(c) The gens Vicinia shows a closely related Gavius (ca  uecineo [] uoltio LF 224, ca 
uecineo LF 225) and Gavia (cau[ia ] uecin[e]a  uotiliạ LF 222, cauia  uecinea LF
223). In the traditional reading of LF 224, ca  uecineo [] uolti , they were even re-
garded as brother and sister.
(d) The gens Socconia may provide two men called Vel, but the attestations are very
doubtful (uel zu[con]|eo  fe [cupa] MF 56 and [--- ue?]l su|[conio? ---]*ic MF 191).
Note that the gentilicium is adapted from Etruscan Zuχu, and that the praenomen is
Etruscan as well.
7.10.5. Praenomina and ethnic identification. Apart from the familial identification,
praenomina can also be used to express ethnic identification. This is of some interest
in an area like the ager Faliscus, lying as it does on the crossroads of several quite
different cultures and languages, each with their own names.
In §7.7.2 it was established that there is a clear preference for several
praenomina, namely Gauius, Iuna, Volta and its derivation Voltius, and Caesius. Even
when abbreviations and patronymic adjectives are not counted, these five together
make up nearly one-third of all instances of Faliscan praenomina (including abbrevi-
ated praenomina). Now Iuna, Volta and Voltius are peculiar to the area, occurring
nowhere else, while Gauius and Caesius do occur elsewhere, but nowhere with a fre-
quency that came close to the frequency of these names in the ager Faliscus. (Note
that with the exception of vultasi Etr XLII, these names do not even appear in Etrus-
can inscriptions from the area: see §9.2.1.) If the fragmentary picture presented by the
epigraphic sources represents the real distribution of praenomina in the ager Faliscus
during the fourth to the second century BCE, this must mean that someone bearing the
name Iuna, Volta, or Voltius would have been immediately recognisable as someone
from the ager Faliscus, or perhaps even as ‘a Faliscan’, that is, someone who regarded
himself as such.
The persistent choice for a local praenomen like Iuna or Volta must have re-
flected an adherence to some form of specifically Faliscan identity that was clearly
different from an Etruscan, Latin, or Sabellic one. This is even more apparent from
the fact that these names, popular as they were, quickly disappear from the record in
the inscriptions that can be dated to the period after c.240. Iuna recurs only in a patro-
nymic adjective in LF 220, Volta in MLF 367-370, in a patronymic adjective in LF
224, and makes a last appearance in a filiation in LF/Lat 214. The disappearance of
Iuna and Volta may be due to the fact that Latin did not have male praenomina of the
first declension, or it may be due to reasons connected with a changing identity of the
populace, or at least of that part of the populace that left (sepulchral) inscriptions.
Gauius and Caesius likewise disappear from the record, and were perhaps replaced by
their perceived Latin equivalents Gaius and Kaeso.
288
THE ONOMASTICON
On the other hand, the inscriptions from the period after c.240 show praenomina
that either were absent or not very frequent in the Middle Faliscan onomasticon.
These are Aulus, Gaius, Lucius, Marcus, Publius, and perhaps also Sextus. All of
these appear to be associated with Latino-Faliscan and Latin inscriptions from the
area, even though Aulus and Publius are (ultimately) of Etruscan origin. Their fre-
quency in the later inscriptions must be due to people inclined to use the Latin rather
than the Faliscan onomasticon, whether they did so because they were immigrants or
descendants of immigrants from Latium or because they were members of local fami-
lies whose identity was shifting towards a ‘Latin’ one. I do not think that it is neces-
sary to ascribe this to a conscious ‘breach with the past’ on the part of the Faliscan
population (or at least those members of the population that left (sepulchral) inscrip-
tions), but rather to the fact that the area had become part of a larger world that was
Latin, with Rome at its centre: in such a world the identity especially of the families
that made up the administration of the area could no longer be strictly local.
Etruscan praenomina do appear in fairly great number in the area, but apart from
Arruns, they do not appear to have enjoyed great popularity, and even Arruns comes
at best in sixth place in the number of instances of individual praenomina. There are
no indications that Etruscan names were among those that recurred within one family,
or that families with Etruscan gentilicia preferred to use Etruscan cognomina: in most
cases, there are simply too few attestations of a particular gentilicium for this. The
only indications are the dubious recurrence of Vel in the gens Succonia (§7.10.4) and
perhaps aruz  cesịe  aruto MF 257 (below). On the whole, families seem to have
preferred Faliscan praenomina, whether or not the gentilicium is Etruscan, and Etrus-
can praenomina also occur in families that do not have specifically Etruscan gen-
tilicia.
A notable exception to this are the inscriptions from Corchiano, aruz  cesịe 
aruto MF 257, poplia  calitenes | aronto  cesies | lartio  uxor MF 265, and ueltur 
tetena | aruto MF 266, arute macena | morenez MF 269, larise  mar||cna  citiai MF
270.145 In all these cases, there is a combination of Etruscan praenomen with an
Etruscan gentilicium, and in MF 257, 265, 266, the filiation shows that the father had
an Etruscan praenomen as well. These inscriptions also show a number of other
Etruscan features (cf. §7.10.3 and §9.2.3). This could imply that Etruscan praenomina
occurring within families with an Etruscan gentilicium were an indication of an Etrus-
can rather than a Faliscan identity.
An example that subsumes the whole gamut of the Faliscan onomasticon is that of the
gens Clipearia. The gentilicium is a Berufsgentiliz, probably pointing to a fairly mod-
est background as artisans. In the mid- or late fourth century, a potter or painter signed
145
Note that these four inscriptions provide four out of the seven instances of Arruns, one of
the two instances of Veltur, and the only instance of Lars.
289
CHAPTER 7
his work with oufilo  clipeaio  letei  fileo  met  facet MF 470*, showing a praenomen
Aufilus that is quite rare and Italic (if not Faliscan), while his father’s name can only be
matched by the Etruscan gentilicium leθaie in Etr XLVIII. From two tombs near Falerii
Novi, and therefore probably dating to the period after c.240, are three more Clipearii, a
l  clipịaṛ[io] LF 230, a c  clipeaṛ[io] | m  f  harịṣ[ex LtF 231, who (apart from being
a haruspex) was at least ce(n)]|so and rex, and a m  cḷ[i]peario  m [ f LtF 233, whose
name may likewise have been followed by a cursus honorum (all that is left is [---]or).
If these are members of the same gens as Aufilus Clipearius (and I can hardly imagine
that within such a small community they were not, since the gentilicium is quite rare),
they appear to have made a social climb from being a family of shieldmakers and pot-
ters with Italic and Etruscan praenomina to being magistrates at Roman Falerii Novi,
and to have adjusted their choice of praenomina accordingly to the Latin Gaius,
Lucius, and Marcus. In a sense, this is as illustrative of the social and onomastic de-
velopment of the ager Faliscus as Livy’s story of the rise of Tarquinius Priscus (1.34)
is of Rome and Etruria.
290
Chapter 8
Syntax and text structure
Although some attention has been paid to it in the past (e.g. Stolte 1926:22-7, G. Giacomelli
1963:155-68), the syntax of Faliscan is on the whole a rather neglected subject. This is not
surprising, for from a traditional, morphology-based syntactic perspective the data provided by
the Faliscan texts can be described in fairly brief terms and provide little that is of interest
(§8.2-3). There are, however, two other and more promising approaches to the subject. The first
is what can broadly be called the ‘word order’ approach, where it is the order of the constitu-
ents of sentences and word groups that is the subject of study: on this point, Faliscan appears to
align with early Latin (§8.4-7). The second approach is the study of the way in which types of
texts, especially formulaic texts, are constructed. This provides some interesting insights into
the relation between the textual types found in the Faliscan material and comparable material
from the surrounding areas (§8.8-13).
8.1. Syntax and text structure: methodological issues
The major problem in the description of Faliscan syntax is that the data are extremely
limited. This is mainly due to the contents of the inscriptions, which in many cases
consist entirely of names. More elaborate texts containing verb phrases are few, apart
from the formulaic type hec cupat/cupant (cf. §8.10.1).
The only longer inscriptions that show verb phrases and even sentences are the
Early Faliscan inscriptions EF 1-4 on the one hand, and the second-century Latin
inscriptions on the other. Although I have used the former group, I found that trying to
use the latter group as a source of data for a description of a Faliscan syntax severely
biased the outcome and I have therefore excluded this group on methodological
grounds, as it cannot be ascertained whether these, in fact, represent Faliscan syntax.
Not only are these inscriptions in most cases very different from the Faliscan inscrip-
tions with regard to their type and contents, they are also usually of a very late date.
Especially note Lat 217 (c.125), 218 (c.125), 219 (c.120-50), 237 and 238, 240, 250
(106), 251 (late second century), 268 (fourth century, but probably imported), 291
(second century), 296 (an import), 377, 393, Cap 386, 421 (c.150), Cap 431-437, Lat
438 (all from before 211), and 456 (c.100-50). I have likewise excluded Late Faliscan
or Latin 214 (c.150?), not because it is necessarily non-Faliscan, but because it is a
unique text even among the Latin inscriptions and can therefore not be used as data for
a consistent picture of Faliscan syntax.
291
CHAPTER 8
8.2. Nominal morphosyntax
8.2.1. Cases and numbers. As appeared from Chapter 4, the Faliscan material shows
six cases in productive use, namely the nominative (§4.2.1, §4.2.6, §4.3.1, §4.3.6,
§4.5.1, §4.5.3.2), the genitive (§4.2.2, §4.3.6, §4.4, §4.5.2, §4.5.3.3, §4.6.1), the dative
(§4.2.3, §4.3.2, §4.6.2), the accusative (§4.2.4, §4.3.3, §4.5.3.1, §4.6.3), and, attested
only for Early Faliscan, the ablative (§4.3.4) and the vocative (§4.3.5). The ablative
occurs in fossilized form in the adverbs foied MF 59-60 (m Proto-Latin */hō+dē(d)/ or
*/ho-dē(d)/, cf. §6.2.34) and ifra MF 40 (m Proto-Latin */enð(e)rā(d)/). The locative
likewise occurs only in fossilized form in the adverb hec m (Proto-Latin */he+ke/
(attestations in §4.8, §6.2.33). Unfortunately, it cannot be ascertained whether in
(Middle) Faliscan the locative had ceased to be productive, as in Latin (cf. LHS I
pp.148-51, Meiser 1998:129), or remained a productive case, as in the Sabellic
languages (cf. Von Planta 1897:415, Buck 1928:199-200).146
The numbers are two, namely singular and plural: as in the other Italic languages,
there is no dual. In Latin inscriptions, the singular could be used for the plural in the
case of a gentilicium preceded by more than one praenomen, as in m c pomplio no
f CIL I2.30 and q k cestio q f CIL I2.61, and perhaps also q a aidicio CIL I2.2442
(cf. Wachter 1987:390-1).147 An Early Faliscan instance of this usage has been read in
euios  mama zextos EF 1, which was interpreted as ‘Mama (et) Sextus Euius’ by
Herbig (CIE 8079), but this is uncertain at best: a clearer instance seems to be m tito
tulio uoltilio hescuna MLF 346, although the m may be a later addition.
Agreement is as rigidly observed as in every Italic language, with the exception of
two problematic instances. The first is poplia  calitenes | aronto  cesies | lartio  uxor
MF 265, where even in the least strained interpretation, ‘Publia Calitenes, wife to
Arruns Caesius Lars’ son’, the patronymic adjective would be expected to show the
genitive larti instead of the nominative lartio. The second is cau‹i›o  pauiceo  |
ḷ[oc]ịes  cela MF 12, for which the most probable interpretation seems to be ‘the tomb
of Gavius Pavicius, Lucius’ son’, which would be caui  pauici  | ḷ[oc]ịes  cela. In both
cases, the confusion in the use of the cases has been ascribed to Faliscan-Etruscan
language contacts, which is probable at least in the case of MF 265 (see §9.2.2-3). The
possibility of an epigraphic error cannot be excluded, however: MF 265 is known only
from apographs, while the lines of MF 12 may in fact belong to separate inscriptions.
146
M.Mancini’s (2002:28-33) interpretation of lete MLF 285 as a locative /lē ̣tę̄/ ← */lŏtā/
would presuppose that the locative still was productive in Faliscan.
147
Early authors interpreted these forms in -o as duals, but in view of the complete absence of
traces of a productive Italic dual, it is extremely unlikely that the dual.
292
SYNTAX AND TEXT STRUCTURE
8.2.2. Syntactic functions of the cases. The use of the cases expressing the arguments
is without surprises: the nominative is used to mark the subject, the accusative to mark
the object, and the dative to mark the indirect object. The instances are:
(intransitive verb, with nominative subject only:)
uel zu[con]|eo  fe[ cupa] MF 56
[ma]rco  pleina  marcio  man[o]mo  cauiacue | [u]eculia  uoltilia  uentarc[i 
[--- c]elio cesi fi | [---] cauia * * | [he cup?]a MF 94
[--- ]celio[ ---|---]*  hec  cupa[t] MF 95
cạ[u]io  le[ueli]o  cau[i] | hileo  ian[ta  ..]lnia | hec  cupat MF 146
[--- pu]peḷ[i---|---]a he[ cupat] MF 150
ụoḷ[ta : ]**[---] | iatac ̣ue  ḷ[---] | hec : cupạṭ MF 158
[leu]elio [---|---]io ca[--- | he c]up[at] MF 159
[po]plia[  ---| hec ]  cup[a] MF 161
uipia  zertenea  loferta | marci  acarcelini | mate  he  cupa LF 221
ca uecineo[ ]uolti | he cupat LF 224
tito[  ]acarcelinio  | ma  fi pop petrunes ce f | [h]e cu[pa] LF 226
tito : uelmineo | titịọ : fe cupa LF 307
c clipear[io] | m f haraṣp ̣[ex cen]|sor rex q *(*)ẹ[---] | heic cubat LtF
231
ofetios kaios uelos amanos salueto salues seitei (= s{e}ite or s‹ie›te) EF 4
a írpios esú Cap 389
k sares esú Cap 404
(transitive verb, with both nominative subject and accusative object:)
ceres  far *[0-2]e[1-3]tom ... *[3-4]*ad EF 1
eqo ụrneḷ[a ti?]tela ... arcentelom huti[c?]ilom pepara[i EF 1
tele*[1-2?] med fifiked EF 9
oufilo  clipeaio  letei  fileo  met  facet MF 470*
[.]a protacio m f mạcistratu | keset cụestod pi pretod pis LF 242 (a sim-
ilar text, but damaged, is LF 243)
perhaps precono | cui teneṭ | let MLF 361 (if ‘Praeconus qui tenet lect(um)’)
293
CHAPTER 8
(transitive verb, with only the nominative subject expressed; the object (the in-
scribed object itself) can be inferred from the context:)
cauios frenaios faced MF 471*
(transitive verbs, with only the accusative object expressed; the subject is
present in the verb form:)
foied uino pipafo cra carefo MF 59 = foied uino ‹pi›pafo cra carẹ[f]o
MF 60
(transitive verb, with nominative subject, accusative object, and indirect dative
object expressed:)
There are no instances of other uses of the cases than nominative subject - accusative
object - indirect dative object. Carefo in foied uino pipafo cra carefo MF 59 =
foied uino ‹pi›pafo cra carẹ[f]o MF 60 may have had an ablative complement as
in Latin, but in early Latin this verb is also used with the genitive and the accusative
(see LHS II p.83). I do not adopt Renzetti Marra’s (1990:336-7) interpretation of
precono | cui teneṭ | let MLF 361 as ‘Preconio cui (al quale) tocca il letto’: this requires
a construction for teneo that is to my knowledge entirely without parallels in Latin or
any Italic language.
Due to the lack of longer texts, there are hardly any instances of the use of cases in
satellites. Ablatives pramod (and propramod?) occur in propramom  pramed [u]mom
pramod pramed umom  pramod  propramọḍ  pramod umọ[m] EF 2, and possibly in
[--- p]rotacio[ ---|---o]stro - pro[---] LF 244 and [---]*[---|--- c]ues[tor ---|---]ostr[o
---] LF 245 (cf. Renzetti Marra 1990:338), but in all these texts the function of these
forms is entirely unclear. The vocative occurs in the pseudo-subjects salue[to]d uoltene
EF 3 and ues saluete sociai EF 4.
On the word level, the only case attested with any frequency is the possessive
genitive (for which see §8.1.3), which can also be combined with esú ‘I am’, as in
ac ̣iuaiom esú Cap 465, and possibly also in k sares esú Cap 404 (cf. §9.3.2). In a
prepositional phrase, the ablative is attested in pro pramọḍ in EF 2 (see §8.4.4).
8.2.3. The nominative and the genitive in isolation. In the epigraphic material, case-
forms regularly occur in isolation. In the case of the nominative and the genitive, this
can be ascribed to the semantic functions of these cases, and does not require the
assumption of ellipsis of a verb, which I think is necessary in the case of the dative and
the accusative (for which see §8.2.4).
The nominative can of course occur in isolation because it is semantically ‘unmarked’:
in inscriptions, its use in isolation serves as a very explicit and specific identification
mark, used to identify:
294
(a) the deceased, in sepulchral inscriptions: the instances are too numerous to enume-
rate (c.184 instances: for the attestations, see §11.1.4.1).
(b) the owner of the object, in Besitzerinschriften where the possessive relationship
between owner and object is not expressed by the genitive (for attestations see §8.8.1).
(c) the maker of the object, in signatures (see §8.9.1 and §11.1.4.2a), but also in
inscriptions naming the magistrate responsible for the execution of a public work as the
‘maker-by-proxy’ (see §8.11 and §11.1.4.5).
(d) the inscribed object itself: in inscriptions where the object is the Topic and the Focus
consists of the attribute of the object, as in cau‹i›o : pauiceo | ḷ[oc]ịes : cela MF 12,
*[.]pi : uesθi : cela MF 83, caui[ : ]ṭ**(*)[i ]: cela MF 84, [---]fate cela MLF 285 (see
§8.9.3).
(e) an attribute of the inscribed object: mention of the inscribed object (the Topic) is
omitted, and the text consists only of the attribute of the inscribed object (the Focus), as
in sacra MF 127 and the labels canumede [die]s pater cupi‹d›o menerua MF 62.
The isolated genitive involves a different syntactic and semantic perspective. It is in fact
an attributive (possessive) genitive denoting the possessor (the Focus), with an ellipsis
of a noun that designates the inscribed object itself (the Topic), which can be inferred
from the context. Undisputed instances are the Besitzerinschriften in the genitive:
titias MF 201
pupiias MF 305
tulom MF 80
and the dedications in the form of a Besitzerinschrift with a divine owner (the use of the
genitive here may be modelled on Etruscan, see §8.10.1):
apolonos EF 10
loifiṛtato MF 31, loifirtato MF 32 (probably likewise dedicatory)
Ambiguous are the isolated io-stem forms in -i, which can be either genitives or
abbreviated nominatives, serui MF 34-36, ani MF 45, uli MF? 261-262, caui  turi MF
273, uolti  catinei MF 469*, marci  anel[i] MF 472*, anni LtF 63, c pscni Cap 387, c
aci Cap 395, sex | senti Cap 399, sex senti Cap 430:148 see also §9.2.2. In the signa-
tures c ̣ cutri MF 200 (and in the (partly imported) Latin signatures c popil[i] Lat 295,
c popili Lat 296, [c ]popili Lat 478*, l quinti Lat 477*), such ambiguous forms in -i
mark the maker of the object, not the owner: if interpreted as a genitive, this must
denote a very loosely possessive sense: see §8.9.1.
148
For a similar ambiguity in the Etruscoid forms in -es, where nominative and genitive were
apparently homomorphemic, see §9.2.2.4.
295
CHAPTER 8
The isolated first-declension forms in -ai, iunai MF 74, 112, MLF/Cap 475*,
uoltai MLF 367-370, sceiuai LF 379, and perhaps [---]ạltai  MF 109, and the isolated
second-declension forms in -oi, caisioi MF 20, tiroi colanioi MF 69-71 can be either
genitives or datives: I prefer to interpret them as genitives (see §4.4.4 and §8.8.1).
8.2.4. Elliptic sentences, isolated datives, and isolated accusatives. In addition to the
nominal phrases described in the preceding section, there are a number of texts that can
be interpreted as verb phrases with an ellipsis of the verb that shifts the emphasis onto
the semantic functions of the arguments. The clearest instances of this are those texts
that consist of a subject in the nominative and an indirect object in the dative: the object
(identical with the inscribed object itself) is always implicit as it can be inferred from
the context. The ellipsis can even include the subject, leaving the emphasis wholly on
the recipient. The instances of this are as follows:
(a) dedications:
titoi | mercui | efiles MF 113
titoi | mercui | efile MF 114
[t]ito[i] | ṃercu[i] | efiles MF 115
titoi | mercụ[i] | efi[les] MF 116
[titoi | mercui | efil]es MF 117
(with ellipsis of the nominative subject:)
titoi  mercụ[i] MF 118
[t]itoi  mercu[i] MF 119
[ti]ṭoi  mercu[i] MF 120
[ti]toi  mercui MF 121
titoi  ṃ[e]rcui MF 122
[titoi : ?]merc[ui] MF 123
[m]ẹrcui MF 124
ṃercui MF 125
meṛ[cui] MF 126
and possibly [---]ạltai  MF 109
(b) Geschenkinschriften (the forms in -ai and -oi can also be genitives, see §8.2.3):
locia eiṃoi MLF 293
(with ellipsis of the nominative subject:)
iunai MF 74, LF 112, Cap 475*
caisioi MF 20
tiroi colanioi MF 69-71
uoltai MLF 367-370
sceiuai LF 379
296
(c) sepulchral inscriptions of the type ‘X [made this grave] for Y’ (see §8.9.2):
cauio uelminẹo | popliai file MLF 308
perhaps uoltio | folcozeo | zextoi | fi LF 330
perhaps [..] folcosio | *****oi LF 333
possibly [---] zaconiai MF 154
possibly cuicto uelmineo |[---?]uoxie[.]ai MLF 310
A similar ellipsis has been invoked to explain the problematic isolated accusative eitam
EF 5. (Middle Faliscan tulom MF 72 is more likely to be a genitive plural.) Similar
accusatives occur in Latin mirqurios alixentrom CIL I2.553 and uenos diouem prosep-
nai CIL I2.558, and in Praesamnitic tecliiam Ps 16-17 and Oscan spuriieis culcfnam
Cm 27. In these cases, too, ellipsis has been assumed, e.g. by Wachter (1987:111) for
the Latin instances, and by Colonna (1980d) for tecliiam.149 I would rather explain
eitam as due to a confusion in the use of the cases in Etruscan-Italic language contact:
see §9.2.2.
8.3. Verbal morphosyntax
As said in Chapter 5, there are few verb forms, and consequently, little can be said
about the morphosyntax of the Faliscan verb. The only tenses that are attested are
present, perfect, and future, and the majority of the forms are indicative.
The attestations of the present indicative are all of durative verbs (‘to lie’, ‘to be’,
and perhaps ‘to hold’):
uel zu[con]|eo : fe[ cupa] MF 56
[ma]rco  pleina  marcio  man[o]mo  cauiacue | [u]eculia  uoltilia  uentarc[i 
[--- c]elio cesi fi | [---] cauia * * | [he cup?]a MF 94
[--- ]celio[ ---|---]*  hec  cupa[t] MF 95
cạ[u]io  le[ueli]o  cau[i] | hileo  ian[ta  ..]lnia | hec  cupat MF 146
ụoḷ[ta : ]**[---] | iatac ̣ue : ḷ[---] | hec : cupạṭ MF 158
[po]plia[  ---| hec ] cup[a] MF 161
149
Colonna referred to tecliiam as an exclamatory accusative, but his discussion shows that he
means an accusative object in an elliptic invocation ‘(Behold,) a tecliia (I dedicate to thee)!’.
297
CHAPTER 8
uipia  zertenea  loferta | marci  acarcelini | mate  he  cupa LF 221

tito[  ]acarcelinio  | ma : fi pop petrunes ce f | [h]e cu[pa] LF 226
tito  uelmineo | titịọ  fe cupa LF 307
c clipeaṛ[io] | m f haraṣp ̣[ex cen]|sor rex q *(*)ẹ[---]| heic cubat LtF 231
a írpios esú Cap 389
k sares esú Cap 404
perhaps precono | cui teneṭ | let MLF 360
The instances of the perfect indicative all refer to an act in the past:
tele*[1-2] med fifiked EF 9
eqo ụrneḷ[a ti?]tela ... arcentelom hutị[c?]ilom pepara[i EF 1
[.]a protacio m f mạcistratu | keset cụestod pi pretod pis LF 242
[---|--- ma|cistratu]| kese[t duum]|uiru p ̣[reto]|r cues[tor] LF 243
Of the instances of the future tense, one (pipafo/‹pi›pafo) refers to a future act, the other
(carefo/carẹ[f]o) to a future state:
foied uino pipafo cra carefo MF 59 = foied uino ‹pi›pafo cra carẹ[f]o
MF 60
The present subjunctive, imperative, and future imperative are attested only for Early
Faliscan. The instances of the subjunctive are mostly from an unclear context, but are
generally interpreted as having adhortative or injunctive force:
ceres  far *[0-2]e[1-3]tom  *[3-5]uf[1-4]ui[..]m *[3-4]*ad EF 1 (injunction as
part of a prayer?)
douiad EF 1 (context unclear: a loose addition to the text that constitutes an addi-
tional injunction?)
ofetios kaios uelos amanos salueto salues seite (= s{e}ite or s‹ie›te) EF 4 (sub-
junctive joined to the imperative, see below)
tulate tulas urate EF/Etr 385 (interpretation unclear: tulate and tulas are subjunc-
tives, while urate may be an imperative (§5.1.2.23), see below)
perhaps also pramed in propramoṃ  pramed [u]mom pramod pramed umom 
pramod  propramọd pramod umọ[m] EF 2 (context unclear, cf. §5.2.1.15).
The instances of the imperative and the future imperative are the following:
298
saluẹ[to]d uoltene EF 3
ues saluete sociai EF 4
ofetios kaios uelos amanos salueto salues seite (= s{e}ite or s‹ie›te) EF 4
tulate tulas urate Etr/EF 385 (interpretation unclear: tulate and tulas are subjunc-
tives, while urate may be an imperative (§5.1.2.23), see below)
Although morphologically imperatives, salue[to]d, saluete, and salueto do not have real
imperative force: they belong rather in the category of the ‘non-characteristic use of the
imperative in curses, maledictions, and well-wishes’ (see Bolkestein 1980:43-7). It is
perhaps for this reason that in EF 4 the future imperative salueto and the adhortative
subjunctive salues seite (= s{e}ite or s‹ie›te) are apparently equivalent. The only ‘real’
Faliscan imperative would then be urate Etr/EF 385 (if that is not a subjunctive as well,
cf. §5.1.2.23): this imperative, too, is apparently equivalent in sense to the adhortative
subjunctives tulate and tulas. For a functional overlap or equivalence of the adhortative
subjunctive and the imperative in early and vernacular Latin, cf. LHS II p.385.
8.4. Constituent order
8.4.1 The order of the arguments. In the Faliscan inscriptions, the order of the
arguments is consistently SV in phrases with verbs with one argument like cupat and
lecet and the forms of ‘to be’ (for instances, see §8.1.2), and SOV in phrases with verbs
with two or three arguments. The instances of the latter are:
(transitive verb, with both nominative subject and accusative object:)
ceresS  [ far *[0-2]e[1-3]tom ]O ... *[3-4]*adV EF 1
[ mamaS z[e]xtos ]S med O f[.f]ịqodV EF 1
[ eqo ụrneḷ[a ti?]tela ] S ... [ arcentelom huti[c?]ilom ]O pepara[iV EF 1
tele*[1-2?]S medO fifiked V EF 9
praụ[i]osS urnamO  soc ̣[iai] p ̣ordedV karai EF 1
[ oufilo  clipeaio ]S  letei  fileo  metO  facetV MF 470*
[ [.]a protacio ]S m f mạcistratuO | kesetV cụestod pi pretod pis LF 242
(a similar text, but damaged, is LF 243)
(with only the nominative subject expressed:)
[cauios frenaios]S faced V MF 471*
(with only the accusative object expressed:)
foied uinoO pipafoV cra carefoV MF 59 = foied uinoO ‹pi›pafoV cra
carẹ[f]oV MF 60
Several of these phrases are formulaic and may therefore not represent the unmarked
word order. Yet an important argument in favour of SOV being the usual Faliscan order
299
CHAPTER 8
is in fact furnished by just such formulaic phrases as [ mama z[e]xtos ]S medO f[.f]ịqodV
EF 1 and tele*[1-2]S medO fifikedV EF 9, for these have an SOV-order even though they
reflect an Etruscan model miniO zinaceV MAKERS where the order is consistently OVS
(§8.9.1). The only exceptions would be the SVO-order of Renzetti Marra’s (1990:336-
7) interpretation of precono | cui teneṭ | let MLF 361 as ‘Praeconus qui tenet lect(um)’,
and the OVS-order of Vetter’s (1953:301) impossible reading of MF 89 as ------a hac ̣
---a : (-)ṛ--ṭ : maximo ‘...a(m) ha(n)c ...a(m) ...t Maximus’.
In the few instances of nominal sentences with more than one constituent (cf. §8.1.3),
the order of the core constituents is nominative subject - dative complement in
lociaS eimoiC MLF 293
and in the sepulcral inscriptions of the type ‘X [made this grave] for Y’ (cf. §8.9.2):
[ larise  mar||cna ]S  citiaiC MF 270
[ cauio uelminẹo ]S | [ popliai file ]C MLF 308
possibly [ [---]]S  zaconiaiC MF 154
possibly [ cuicto uelmineo ]S | [---?]uoxie[.]aiC MLF 310
perhaps [ uoltio | folcozeo ]S | [ zextoi | fi ]C LF 330
perhaps [ [..] folcosio ]S | *****oiC LF 333
(I reject this interpretation in the cases of [---]ronio  uol[t---|---]a*ome MF 156, tito 
uel|mineo  iun|aị ị*ice MLF 315, and tito ᛂ uelmineo | ịụ i*ice MLF 309.)
In the few Faliscan dedications that consist of RECIPIENTDAT DEDICANTNOM, on the other
hand, this order is reversed:
[ titoi | mercui ]C | efilesS MF 113
[ titoi | mercui ]C | efileS MF 114
[ [t]ito[i] | ṃercu[i] ]C | efileS MF 115
[ titoi | mercụ[i] ]C | efi[les]S MF 116
[ [titoi | mercui ]C | efil]esS MF 117
Comparable Latin inscriptions, too, show a slight tendency to place the name of the
deity first (see §8.10.1), perhaps to give the deity ‘pride of place’.
8.4.2. The position of adjuncts. The location adjunct precedes the verb in the sepul-
chral inscriptions with the formulaic phrase hec cupat/cupant (for which see §8.9.1). In
these phrases, the place of the adjunct is therefore, as in Latin, “somewhere between the
first argument and the finite verb” (Pinkster 1990:181).150
Interestingly, the only exceptions to this order are the two instances where the
formula is also varied in other ways:
150
Note nouios  plautios  med  romai  fecid CIL I2.561, where a formulaic phrase (cf. §8.9.2)
is broken up to insert the adjunct in its correct place.
300
[---]o cicio cicoi  cupat  ifra MF 40 (with ifra instead of hec)

[uo]ltio[  ]ueicọno : lecet : ḥec MF 88 (with lecet instead of cupat)
The former can be explained by assuming extra stress on the adjunct, or by assuming a
contrast (either ‘[here lies Y,] below lies X’ or ‘[not here, but] below lies X’). For the
unusual order of MF 88 I have no explanation. The exceptional placement of the time
position adjuncts in foied uino pipafo cra carefo MF 59 = foied uino ‹pi›pafo
cra carẹ[f]o MF 60, is due to pragmatic reasons, namely the contrast foied : cra.
8.5. Word order within word groups
8.5.1. The position of the adjective. Given the nature of the Faliscan material, the
number of noun phrases containing adjectives is limited. Adjectives only occur in the
more elaborate Early Faliscan texts:
soc ̣[iai] ... karai EF 1
ụrneḷ[a ti ?]tela EF 1
arcentelom hutị[c?]ilom EF 1
quto ... duenom EF 3
titias duenas EF 3
probably far *[0-2]e[1-3]tom EF 1
In all these instances, the order is noun - adjective, which seems to be in accordance
with the order in Latin and the Sabellic languages.151 The only adjectives in the Middle
and Late Faliscan inscriptions are the patronymic adjectives, which always follow the
name (attestations in fig.7.3), but this is due to pragmatic and formulaic reasons.
8.5.2. The position of the attributive genitive. In attributive groups, where one noun
stands in the genitive, the attributive genitive generally precedes the head (genitive -
noun or GN), but the problem is that such groups are virtually always formulaic:
(a) the Middle and Late Faliscan filiation formula FATHERGEN SON/DAUGHTER (36-41
attestations, see fig.7.3);
(b) the Middle and Late Faliscan marital formula HUSBANDGEN WIFE (4-8 instances, see
§7.4.2), with 1-4 counterexamples of WIFE HUSBANDGEN with noun - genitive or NG, the
clearest of which is tanacu[il] | anelia   | uxor  ia MF 101;
(c) the Middle Faliscan designations of freedmen and -women FORMER MASTERGEN
FREEDMAN/FREEDWOMAN in ṭị [] ṭịria lo[?---]|l[e]a  cs  f MF 155 and [---]*i  u[o]ltiai
lo MF 165;
151
This argues against Olzscha’s reading (in Radke 1964:136-7, cf. also Radke 1994) ro]uf[om]
ui[no]m in EF 1 and the interpretation of uipia  zertenea  loferta LF 221 as ‘Vibia, a Sertinian
freedwoman’ (first Schulze 1904:513).
301
CHAPTER 8
(d) the Late Faliscan ‘reversed filiation’, marci  acarcelini | mate LF 221;
(e) the Middle Faliscan ius sepulcrale-formula OWNERGEN cela in cau‹i›o  pauiceo  |
ḷ[oc]ịes  cela MF 12, *[.]pi  uesθi  cela MF 83, caui []t **(*)[i]  cela MF 84, and
[---]fate cela MLF 285;
(f) comparable to (e) is Middle or Late Faliscan caui  tertinei  | posticnu MLF/Cap
474* (‘a statue of Gavius Tertineius’?);
(g) titias duenom duenas EF 2.
The reverse, where the attributive genitive follows the head (noun - genitive or NG)
appears only in the Middle Faliscan filiation formula FATHERGEN (attestations in fig.7.3),
which in my view is a shortened form of the formula FATHERGEN SON/DAUGHTER rather
than as a distinct formula in its own right: see §7.5.2.1.
From these instances, it would appear that in Faliscan the attributive genitive regularly
preceded the noun (genitive - noun or GN). Yet the placing of the genitive in these
groups may be due to pragmatic considerations (giving pre-eminence to the Focus), and
therefore not be unmarked: the limited corpus, too, makes it difficult to draw valid
conclusions. A GN order is in accordance with the other Italic languages: according to
Adams’s (1976:74-7) analysis, which includes early Latin formulaic material, GN was
the predominant order in early Latin (although it later changed to NG), and Rosenkranz
(1933) concluded the same for the Sabellic languages. In Etruscan, the genitive appears
to follow the Head: cf. e.g. the predominance of mi OWNERGEN over OWNERGEN mi (268
and 19 attestations respectively, see §8.8.2), and the counterpart of the Faliscan formula
OWNERGEN cela, eca σuθi/σ́uθi [neσl] OWNERGEN (12 instances, §8.9.3).
8.5.3. The position of appositions. Contrary to the position of the attributive genitive
(§8.5.2), in appositional groups where both nouns are in the same case, the apposition
always follows the head. Examples of this are:
(a) the Middle and Late Faliscan filiation formula FATHERGEN SON/DAUGHTER following
the name (36-41 attestations, see fig.7.3);
(b) the Middle and Late Faliscan marital formula HUSBANDGEN WIFE and WIFE
HUSBANDGEN following the name (5-12 instances, see §7.4.2),
(c) the Middle and Late Faliscan designations of freedwomen in loụṛia | [l]oifirta MF
41 and uipia  zertenea  loferta LF 221, and in the MASTERGEN FREEDMAN/FREEDWOMAN
in ṭị [] ṭịria lo[?---]|l[e]a  cs  f MF 155 and [---]*i  u[o]ltiai lo MF 165;
(d) the Late Faliscan ‘reversed filiation’, uipia  zertenea  loferta | marci : acarcelini |
mate LF 221;
(e) Middle and Late Faliscan titles and cursus honorum (instances in fig.2.2).
This usage does not differ from that of Latin, the Sabellic languages, or Etruscan.
302
8.5.4. Prepositional phrases. Due to the nature of the material, there are hardly any
examples of prepositional groups. Propramọḍ in EF 2 can be read as pro pramọḍ, but in
view of the propramom in the same text, it can also be read as propramọḍ. If read as
pro pramọḍ it is the oldest attestation of pro + ablative (and not pro fileod in CIL
I2.2658, cf. Vine 1993:195-6). All other attestations of prepositions are from Latin
inscriptions (LF/Lat 214, Cap 433, Lat 217-218). There are no attestations of postposi-
tions,152 although it is possible that Faliscan had postpositions at a prehistoric date: see
§8.6.
8.6. Coordination and subordination
On the level of the noun phrase, two or more entities are usually placed in asyndetic
parataxis. Here two groups can be distinguished. The first consists of collegiate
magistrates, which are named asyndetically in a [.]osena uẹ narionio MF 206, and
c**(*)coṇẹo l***(*) ce paui[cio 1-2]so MLF 290, [1-2 ] hirmio m[ f ]ce
tertineo c f preṭ[ores ?---] LF 213 (cf. also Lat 250 from near S. Maria di Falleri).
The second group consists of sepulchral inscriptions pertaining to more than one
person. The majority of these inscriptions have asyndetic coordination (which in a
number of cases may be due to the fact that burials and inscriptions were added later),
but there are a few instances where the names are connected with each other by means
of -cue:
uol[ta  ]**[---] | iatac ̣ue  ḷ[---] | hec  cupat MF 158
[---]***|[---]ocuẹ | [---]ịọ MF 170
[ma]rco  pleina  marcio  man[o]mo  cauiacue | [u]eculia  uoltilia  uen-
tarc[......  hec  cupa]nt MF 80
uolta  uelmineo | fuloniacue MF 313
cf. also m neroni | a f et hlau|elea m f LtF 325
The use of -que in sepulchral inscriptions can be compared to the Etruscan use of -c: in
Latin sepulcral inscriptions, -que is rarely used. This may be due to different methods of
burial: the Etruscan and Faliscan family tombs were more suited for joined burials.
There are no other instances of coordinators, whether on the phrase or on the
sentence level. Coordinators l[o]ụf[ir = ‘or’ (cf. Oscan loufir TB 8) and u[el have been
read by Olzscha in ceres farme(n)tom l(o)ụf(ir) ui(no)m x-kad euios (1965:123) and
ceres far d[el]e[c]tom (?) u[el ro]uf[om] ui[no]m f ̣[in]ḳad (in Radke 1965:136-7). Of
these readings, at least the former is epigraphically impossible. Vetter (1953:283)
proposed to restore a hypotactic coordinator in pepara[i .. ]douiad EF 1. Although
152
I do not adopt Pisani’s (1964:349-51) pramoe = pramoi-e(n) in EF 2 or Rix’s (1993a:86)
suggestion ḥei-e(n) in EF 4.
303
CHAPTER 8
syntactically attractive, such a restoration appears to be epigraphically impossible:

douiad is either an asyndetic hypotaxis or a loose addition to the text. I do not adopt the
sei ... sei in Pisani’s (1964:349) interpretation of EF 4 (§12.3).
8.7. Word order typology
It is rewarding to place the results of sections §§8.4-6 in a typological perspective, not

only because in the diachronic definition of dialect (§1.2) language typology could be
expected to provide important arguments (cf. e.g. Cowrie 1981:204-9), but also because
the point is of some importance for the discussion of formulaic texts (§8.13). Reviewing
the preceding sections according to the major typological criteria first established by
Greenberg (1966), the Faliscan material could be said to show:
(a) SOV-order, both in formulas (even where borrowed from SVO-languages) and in
non-formulaic sentences (§8.5.1);
(b) prepositions (Pr), with no indication of postpositions (§8.5.4);
(c) attributive genitives preceding the head of the noun phrase (GN), except in several
instances of a formula adopted from Etruscan (§8.5.2);
(d) adjectives following the noun (NA) (§8.5.1).
In other word, Faliscan is SOV/Pr/GN/NA and in this respect appears to align with
contemporary Latin as described by Adams (1976): a significant point, for Faliscan and
Latin are slightly unusual in that languages with SOV-order tend to have postpositions
and languages with prepositions tend to have a VSO- or SVO-order. However, in view
both of the fairly slight Faliscan material on which this SOV/Pr/GN/NA is based and of
the general questions about the validity of correlation between these word order
parameters (see e.g. Cowrie 1981:86-93), this conclusion should be regarded with even
more than the usual caution.
If it is felt that there is an ‘SOV - Pr discrepancy’ in Latin and Faliscan that needs
to be explained, this may be done by assuming a shift either in the constituent order or
in the use of pre- and postpositions. In my view, arguments for the latter possibility can
be found more easily, since Latin preserves postpositions such as -cum and tenus
(Adams (1976:88) also pointed to ted endo CIL I2.4 = in te) and also had the genitive
preceding the noun (GN), which is also assumed to be more in accordance with the use
of postpositions. The Sabellic languages, especially Umbrian (-a(ŕ), -e(n)/-em,
-ku(m)/-co(m), -pe(r), -ta/-tu/-to), but also South Picene (-ín) and Oscan (-en), offer
even more material testifying to the existence of postpositions in the Italic languages: if
this material is indeed indicative of a Latin shift from postpositions to prepositions, it
would therefore appear to be of Proto-Latin date. The alternative, that Latin would
originally have had an SVO-order, is more difficult: see also Adams 1976.
304
Insofar as the material presents a usable picture, Faliscan appears to align with
Latin on the point of word order typology, while it appears to differ from the Sabellic
languages, irrespective of whether the documents on which this picture is based date
from before or after the period of Roman expansion into the area.
8.8. Besitzerinschriften
8.8.1. OWNERNOM, OWNERGEN, RECIPIENTDAT. The simplest type of Besitzerinschrift

consists of a name in isolation, either as OWNERNOM or as OWNERGEN (cf. §8.2.2). These
types are so common that it is not necessary to adduce instances from other languages.
The semantic difference between the two is that in the case of the nominative the owner
is named as such, the inscription serving as a very explicit identification mark, while
the use of the genitive points to the existence of a possessive relationship between the
person and the object. The word describing the inscribed object itself is omitted as
being inferable from the context. In the Faliscan inscriptions (as in those in Latin), the
forms in -i that are derived from io-stems can be nominatives as well as abbreviated
genitives (attestations in fig. 8.1): see also §9.2.2.2.
Several Faliscan Besitzerinschriften have isolated forms in -ai and -oi: iunai MF 71, LF
112, MLF/Cap 475*, uoltai MLF 367-370, sceiuai LF 379 and caisioi MF 20, tiroi
colanioi MF 69-71. These are usually interpreted as datives, but this seems to be due in
part to a reluctance to assume a first-declension genitive in -ai for Faliscan (cf. §4.2.2).
There are no attestations of such isolated datives for Latin: the isolated Latin forms in
-ai, -ae, and -o have to my knowledge always been interpreted as first-declension
genitives and as second-declension nominatives in -o(s) respectively. The Sabellic
Besitzerinschriften (primarily Um 31-41, Ps 4-20, Fr 7-8, He 3, Si 14-22, Cp 38-40, Po
86-92, Cm 32-37) are all in the nominative or in the genitive, never in the dative.
The only indications that the isolated Faliscan forms in -ai and -oi can be datives
are: (1) the general similarity of this use of the dative to that in the dedicatory inscrip-
tions in Faliscan, Latin, or the Sabellic languages (cf. §8.10.1); (2) the inscription locia
eiṃoi MLF 293 from near Corchiano, which can be interpreted as an elliptic sentence
with the name of a giver in the nominative (locia) and a recipient in the dative (eiṃoi),
but which can equally well be interpreted as ‘Lucia Aemi (f.)’ with a genitive in -oi (cf.
§4.4.4); and (3) the Etruscan inscription vultasi Etr XLII=Fa 3.4 from Vignanello, an
Etruscan form in -si of the Faliscan name Volta. The Etruscan forms in -si and -(a)le
can be used to designate both by whom and for whom the object was made (Steinbauer
1999:174-6): the latter use corresponds to that of the Indo-European dative. Problemat-
ic, however, is the fact that this inscription is to my knowledge the only one where a
305
CHAPTER 8
form in -si or -(a)le occurs in isolation in a Besitzerinschrift or a Geschenkinschrift: in

most other cases, the use of the -si or -(a)le form is due to a verb, usually mulu-.153
All in all, I regard the evidence in favour of a dative-interpretation of the isolated
forms in -ai and -oi in Faliscan Besitzerinschriften as very slight indeed. If these forms
are datives, they would appear to record a gift, presenting the person named as the
recipient rather than as just the owner: the formula for these inscriptions is therefore
RECIPIENTDAT rather than OWNERDAT, and they should properly be classed as Geschen-
kinschriften rather than as Besitzerinschriften.
8.8.2. Besitzerinschriften of the iscrizioni parlanti-type. The Faliscan Besitzer-

inschriften also comprise several types of iscrizioni parlanti. The simplest types are ego
OWNERGEN (Agostiniani’s type L1) and OWNERGEN ego (Agostiniani’s type L2), attested
for Faliscan in:
eko lartos EF 6
eko kaisiosio EF 7
aịṃiosio eqo EF? 467*
Both types have parallels in the Etruscan formulas mi OWNERGEN (Agostiniani’s type
E1, 268 attestations) and OWNERGEN mi (Agostiniani’s type E2, 19 attestations). The
difference between the types lies in the order in which Topic and Focus are presented.
An extension of the former formula ego OWNERGEN (Agostiniani’s type L1) is the
type ego POSSESSIONGEN OWNERGEN (Agostiniani’s type L5), attested for Faliscan in:
This formula, too, is modeled on an Etruscan type, namely mi POSSESSIONGEN
OWNERGEN (Agostiniani’s type E5, c.28 attestations). In this extended form of the
formula both Faliscan and Latin (in eco urna tita uendias Lat 479†) use the variant
where genitive follows the head: Faliscan and Latin counterparts of the Etruscan type mi
OWNERGEN POSSESSIONNOM (Agostiniani’s type E6, c.20 attestations) are not attested.
The counterpart with the nominative, ego OWNERNOM and OWNERNOM ego (Agostiniani’s
types L3 and L4) is attested only in Latium, namely eqo kanaios CIL I2.474=482†, eqo
p ̣ulp ̣ios CIL I2.479, and eco c antonios CIL I2.462, and in the ager Capenas, namely
eco tulie LF 383
m adicio eco LF 378
Eco tulie is admittedly ambiguous: tulie can be either an Etruscoid nominative/genitive
tulie(s) or a monophthongized genitive in -ie /-ię̄/: probably the former, as there are no
(other) instances of a monophthongized genitive ending (§3.7.7, §4.2.2) from the area.
153
The Geschenkinschriften with forms in -si and -(a)le are: (with mulu-:) OA 3.1, La 3.1, Fa
3.2, Cr 3.10, 12, 13, 15, 16, 18, 20, 23, AT 3.1, 2, Vc 3.2, AS 3.1, 2; (other:) OA 3.9, Ve 3.1,
Fa 3.1, Cr 3.4, 5, 6, 7, 8, Ta 3.1, Vc 3.6, AH 3.4 (with menaχe).
306
first declension second declension third declension
nominative turia MF 22-27 licinio MF 259-260 larise uicina MF 371

iuna MF 73 statio cailio MF 376 larise | uicina MF 372
latria MF 74 cauio : peṭṛọṇẹọ MF 473*
larise uicina MF 371 setorio MLF/Cap 476*
larise | uicina MF 372 ? ueiụeto MLF 464
hermana MF/Etr 264 cauios LF 382
? ṛica MF/LtF 21 ḳ uomanio Cap 388
? aie* (aieạ ?) MF 107 at fertrio Cap 391
ueiuatia MLF 463 f pacios Cap 392
seralia LF 380 st clanidio Cap 394
? apa Cap 457 p iunio Cap 462
dedication apolo MF 65
signature k p aiedies Cap 390

nominative serui MF 34-36
or genitive ani MF 45
uolti catinei
marci : anel[i] MF 472*
uli MF? 261-262
caui  turi MF 273
anni LtF 61
c pscni Cap 387
c aci Cap 395
sex | senti Cap 399
sex senti Cap 430
signature c ̣ cutri MF 200
genitive titias MLF 201 tulom MF 72 [fel]ịcinatiu LF 384
pupiias MLF 304
dedication apolonos EF 10
loifiṛtato MF 31
loifirtato MF 32
genitive iunai MF 74 caisioi MF 20
or dative iunai LF 112 tiroi colanioi MF 69-71
uoltai MF 367-370
sceiuai LF 379
iunai MLF/Cap 475*
dedication ? [---]ạltai MF 109
Fig.8.1. Isolated cases in Besitzerinschriften, dedications, and signatures.
307
CHAPTER 8
These types correspond to the Etruscan formulae mi OWNERNOM and OWNERNOM mi

(Agostiniani’s types E3 and E4, 8 attestations). It is, in fact, surprising that these
formulas should exist at all, for they are in a sense ‘illogical’, as the inscribed object
claims to be a certain person rather than claiming to belong to a certain person (cf.
Colonna 1975:165).
Beside these nominal formulas with ‘I’, there is a second group of iscrizioni parlanti
where a verbal formula with ‘I am’ is used. Two types of this group, OWNERGEN esúm
and OWNERNOM esúm, are attested from the ager Capenas in
ac ̣iuaiom esú Cap 465 (OWNERGEN esúm)
a írpios esú Cap 389 (OWNERNOM esúm)
k sares esú Cap 404 (either OWNERGEN esúm or OWNERNOM esúm)
These types are not modeled on Etruscan formulas, for in the Etruscan formulaic
repertoire the verbal formulas of the Greek models were replaced by pronominal
formulas with mi/mini. Agostiniani (1982:245, 261-2) distinguished two groups of these
verbal formulas: (1) an old tradition of Sabellic formulas derived directly from Greek
models with e„m…, and (2) a new repertoire of Latin sum-formulas, which, from the third
century onwards, replaced the ego-formulas derived from Etruscan models. The type
OWNERGEN esúm, attested in ac ̣iuaiom esú Cap 465 (and in k sares esú Cap 404, if
sares is interpreted as a Sabellic genitive in -es, cf. §9.3.2) is comparable to the Sabellic
type OWNERGEN súm (Agostiniani’s types I1-2,12 attestations)154.
More problematic is OWNERNOM esúm in a írpios esú Cap 389 and k sares
esú Cap 404, for this type has no Sabellic counterpart in Besitzerinschriften.155 There-
fore, the Capenate formula OWNERNOM esúm may well be connected to the earliest phase
of the Latin replacement of the ego-formulas by the sum-formulas as envisaged by
Agostiniani (1982:245): the occurrence of OWNERNOM esúm could then be ascribed to
the persistence of OWNERNOM ego in the area, attested in m adicio eco LF 378 (cf.
Bakkum 1996:3-4).
8.9. Signatures
8.9.1. MAKERNOM, MAKERGEN. The main problem in describing signatures is that when
they consist only of the maker’s name, they take the form of MAKERNOM or MAKERGEN,
and are therefore formally indistinguishable from Besitzerinschriften of the type
154
I have merged Agostiniani’s types I1 and I2, for the distinction between these two is not
based on difference in word order, but on the use of súm/sum in I1 vs. sim in I2. The order is
always OWNERGEN súm/sum/sim: in Latin both sum OWNERGEN and OWNERGEN sum occur.
155
The only Sabellic instance of OWNERNOM sum (Agostiniani’s type I3) is the Oscan sepulchral
inscription vibi[s :] smintiis : vibis : smintiis : sụṃ Cp 4.
308
OWNERNOM and OWNERGEN. They can only be identified as signatures on epigraphic

grounds, usually because they were written on the object before its completion
(§11.1.4.2a). Examples of MAKERNOM are pleina MF/Etr 199, arθ[3-5]rẹ MF/Etr 267, k
p aiedies Cap 390, and the Latin inscription t fourios *[ ]f ̣ Lat 216. The type
MAKERGEN (‘[the work] of ...’?. ‘[from the workshop] of ...’?, ‘[by the hand] of ...’?), is
represented only by forms in -i that can also be abbreviated nominatives: c ̣ cutri MF
200, and the imports c popil[i] Lat 295, c popili meuanie Lat 296, [c ]popili Lat
478*, and l quinti Lat 477*. Note that with the exception of MF 200, none of these
inscriptions is certainly Faliscan. Comparable to the type MAKERNOM are the official
inscriptions with the names of the responsable magistrates (§8.12).
8.9.2. Signatures of the iscrizioni parlanti type. Like the Besitzerinschriften, the
signatures, too, occur in an iscrizione parlante-variant MAKERNOM me MAKE3RD PF
(Agostiniani’s type L12), attested in Faliscan both in the singular and in the plural:
tele*[1-2] med fifiked EF 9
oufilo : clipeaio : letei : fileo : met : facet MF 470*
? also cauios frenaios faced MF 471* (with omission of med?)
mama z[e]xtos : med f[.f]ịqod EF 1
For Latin, there is one attestation, nouios plautios med romai fecid CIL I2.561.
The type is derived from the Etruscan type mini zinace MAKERNOM (Agostiniani E7.3,
2-3 attestations)156, but with an adaptation of the word order of the Etruscan model.
There is one instance where a Latin inscription retains the Etruscan word order
(Agostiniani’s type L9) is med loucilios feced Lat 268 from Corchiano (cf. §9.4.2).
In the original Greek model for such signature formulas the all-purpose verb
™po…hse was used, and this is reflected by the corresponding formulas in Etruscan
(zinace), Latin (feced, fecid), Middle Faliscan (faced, facet), Oscan (úpsed/úpsens, cf.
Rix 1993b), and Venetic (vhagsto Le 128, hvagsto Le 127). The Early Faliscan
instances are exceptional in that fifiked and f[.f]ịqod are forms of fingo, which would
have had a specific meaning applicable only to pottery, which makes the Early Faliscan
formula unique among the signature formulas used in ancient Italy: note that the Middle
Faliscan formula with faced/facet is different in this respect. This oddity is in fact one of
the main arguments adduced in favour of interpreting fifiked and f[.f]ịqod as forms of
facio (Poccetti 2005:21-3): see §5.1.2.7-8.157
156
To the two attestations quoted by Agostiniani’s (1982:202) add mi(ni) mamarce zinace Ve
6.2, unless this is read as a unique (?) type mi mamarce zinace ‘I, Mamarce, made this’.
157
Perhaps this use of fingo is due to a (dialectal?) development in the meaning of the verb,
from specific ‘knead’ → general ‘make’, cf. Greek m£ssw ‘to knead’ : Proto-Germanic
*/makōn/ ‘to make’, both from PIE */meħ2 ́-/ or */meħ2ḱ-/.
309
CHAPTER 8
8.10. Sepulcral inscriptions
8.10.1. DECEASEDNOM and DECEASEDNOM hec cupat. The vast majority of the Faliscan
sepulchral inscriptions are of the type DECEASEDNOM. In these inscriptions it is usually
the formal form of the name that is used, PRAENOMEN GENTILICIUM [FILIATION]
[COGNOMEN] for men (§7.3) and PRAENOMEN GENTILICIUM [FILIATION] for women
(§7.4.1). This type of sepulchral inscription is so common, not only in Faliscan, but also
in Etruscan, Latin, and the Sabellic languages, that I refrain from giving examples, as
these can be found without much trouble. In the case of married women buried without
their husband, the name could be extended with [HUSBANDGEN uxor] or [uxor
HUSBANDGEN] (§7.4.2). This addition has parallels in Etruscan and Latin: there are no
attestations for the Sabellic languages, perhaps because the number of Sabellic sepul-
chral inscriptions of women is limited (§7.4.1 with note 115).
Although this nominal type is the most common among the Faliscan sepulchral
inscriptions, there is also a fairly frequently used verbal type DECEASEDNOM [FILIATION]
hec cupat/cupant. The instances of this type are:
uel zu[con]|eo : fe[ cupa] MF 56
[ma]rco : pleina : marcio : man[o]mo : cauiacue | [u]eculia : uoltilia :
uentarc[i : ..... : hec : cupa]nt MF 80
[--- c]elio cesi fi | [---]: cauia * * | [he cup?]a MF 94
[--- ]celio[ ---|---]* : hec : cupa[t] MF 95
cạ[u]io : lẹ[ueli]o : cau[i] | hileo : ian[ta : ..]lnia | hec : cupat MF 146
ụoḷ[ta : ]**[---] | iatac ̣ue : ḷ[---] | hec : cupạṭ MF 158
[po]plia[ : ---| hec ]: cup[a] MF 161
uoltio uecineo | maxomo | iuneo he : cupat LF 220
uipia : zertenea : loferta | marci : acarcelini | mate : he : cupa LF 221
marcio : acarcelinio | cauia : uecinea | hẹc ̣ cupat LF 223
tito[ : ]acarcelinio : | ma : fi pop petrunes ce f | [h]e cu[pa] LF 226
tito : uelmineo | titịọ : fe cupa LF 307
c clipear[io] | m f haraṣp ̣[ex cen]|sor rex q *(*)ẹ[---] | heic cubat LtF
231
(with variations:)
[---]o cicio cicoi : cupat : ifra MF 40
[uo]ltio[ : ]ueicọno : lecet : ḥec MF 88
This formula can be compared, first of all, to the Etruscan formula DECEASEDNOM θui
cesu/ceśu ‘... lies here’ or ‘... is buried here(in)’ (Steinbauer’s type G7), which is found
310
in twelve inscriptions, mostly from Tarquinii: ramθa : huzcṇai : θui : cesu Ta 1.51,
ramθa : zertnai : θui : cesu Ta 1.52-53 (twice), larθ : velχas : θui : cesu Ta 1.205, [---
spur]inas : sacni : θui : c ̣eseθce (=c ̣es‹u ...›θce?) Ta 5.3, laṛ[is aneini θ]ui ceśu
puiaχ̣ AS 1.393, larθ auclina | ceśu θui AS 1.472, and, slightly differently, apac atic |
saniσ́va θu|i cesu Cr 1.5 and vel : aties : velθurus | lemniσa : celati : cesu Ta 1.66
(unclear are Vc 1.59, Ta 5.4, and Pe 5.3).
Secondly, there are the parallels in the Central Italic sepulchral inscriptions with
cubo: South Picene apaes  qupat[  e]sm|ín  púpúnis  n|ír  mefiín  veia|t  vepetí
MC.1 and apúnis  qupat  a[2-3]**[1-2 ]  [ 5-6 ]  [n]ír AP.3, Vestinian [s]acaracirix
| cibat cerria | licina saluta | salaus MV 7, Marrucinian st loisies paq̣ ecuf ̣
ẹncubat a[?] | salaus MV 8, and Paelignian pes pros ecuf incubat | casnar oisa
aetate | c anaes solois des forte | faber Pg 10. Latin cubo, too, can be used in the
sense ‘to lie in a grave’, although in Republican inscriptions this is found only in the
quite late q brutius | p f quir u | mercator boua | de campo heic | cubat CIL
I2.1259, q cornelius q p l | diphilus cornelia q l | heraes heic cubant CIL
I2.1638, and truttedia hic | cubat CIL I2.2135.158
Even within this small number of instances, these formulas show a high rate of
variation, both in the order of the constituents and in the words themselves: thus beside
hec cupa(n)t, Faliscan has cupat : ifra MF 40 and lecet : ḥec MF 88, South Picene has
both qupat [ e]smín and mefiín  veia|t  vepetí in MC.1, Marrucinian and Paelignian
use ecuf ̣ ẹncubat MV 8 / ecuf incubat Pg 10, with a compound transitive verb, and
Etruscan once has celati : cesu Ta 1.66. This suggests that this type was either not
rigidly formulaic, or perhaps that its variation was due to variations in the burial
ritual.159
It should be noted that the Faliscan instances of the hec cupat/cupant-type are (a)
more frequent than the instances of all comparable Etruscan and Italic types put
together, and (b) far more frequent among the Faliscan sepulchral inscriptions than the
θui cesu/ceśu-type among the vastly greater number of Etruscan sepulchral inscriptions.
The Faliscan hec cupat/cupant-type can therefore be said to constitute a truly ‘Faliscan’
type of sepulchral inscription, irrespective of whether or not it was derived from the
Etruscan θui cesu/ceśu-type or from a common Italic model.
The Late Faliscan, Latino-Faliscan, and Latin sepulchral inscriptions also furnish
several instances of cursus honorum in LF 242-244, 245, 247-249, LtF 231-232, Lat
219, 237-239 (see §2.3.3). As is explained in §11.1.4.1, these were appropriate to the
158
Cf. further Lucr. RN 3.892, and, with ossa as the subject, in CIL I2.1312 and Ovid. Am.
1.8.108 = Ep. 7.162 = Tr. 3.3.76 (a mock-epitaph).
159
For such local variation, cf., e.g., how the habit of marking grave-goods with the word
σ́uθina is very frequent at Volsinii (126 instances), but virtually non-existent elsewhere.
311
CHAPTER 8
roadside burials of Latium rather than to the closed rock-tombs of the ager Faliscus:
they are almost certainly an imitation of Roman burial features after the war of 241.
8.10.2. ‘X [made this grave] for Y’. A different type of sepulchral inscription is
PROCURATORNOM DECEASEDDAT, where the inscription records that one person (whom
Lejeune (1974) labeled the ‘procurateur’) made or prepared a tomb for the deceased:
cauio uelminẹo | popliai file MLF 308
possibly [---] zaconiai MF 154 (unclear)
possibly cuicto uelmineo | [---?]uoxie[.]ai MLF 310 (unclear)
perhaps uoltio | folcozeo | zextoi | fi LF 330 (see below)
perhaps [..] folcosio | *****oi LF 333 (see below)
[rejected: [---]ronio  uol[t---|---]a*ome MF 156, and tito  uel|mineo  iun|aị ị*ice
MLF 315 and tito ᛂ uelmineo | ịụ i*ice MLF 309 (where i*ice has been inter-
preted as a verb, but is perhaps a cognomen)]
PROCURATORNOM DECEASEDDAT and DECEASEDDAT PROCURATORNOM are, of course, well-
known from the Latin sepulchral inscriptions (where the dative could also be expressed
by dis manibus DECEASEDGEN).160 The type is notably absent, however, in the Sabellic
languages, where only South Picene ma kupri koram opsút ani{ni}s rakineḷíṣ
pomp[úne]í AQ.2 shows a comparable formula, and in Etruscan, as a scrutiny of the
forms with -si/-(a)le in the categories Grabinschriften and Bau- und Grabstifterin-
schriften of ET shows. In the case of Etruscan, this may be because the language
lacked a ‘true’ dative (cf. Steinbauer 1999:170-5). More or less comparable is eca :
σuθic : velus : ezpus | clensi : cerine Vc 1.87 (Steinbauer’s type G4).
The problem with the occurrence of the PROCURATORNOM DECEASEDDAT type in Faliscan
is twofold: (a) inscriptions of this type do not appear to fit in with the Faliscan burials in
closed family tombs and the function of the sepulchral inscriptions within this type of
burial (§11.1.4.1), and (b) the interpretation of these texts depends on whether the
Faliscan forms in -ai and -oi can be genitives or must always be datives: a discussion,
where, as said, there is an unwillingness to accept the existence of the genitive in -ai,
and consequently, of that in -oi (§4.4.4).
In my view, Faliscan forms in -ai and in -oi can very well be genitives, and the
forms -oi in LF 330 and 333 certainly give the impression of being genitives in
filiations of the type FATHERGEN [SON/DAUGHTER] (cf. [---]o cicio cicoi  cupat  ifra MF
160
It also occurs frequently in Venetic (Le 63, Le 77-79, Le 102, Le 109, Le 226, and
Le 110bis), where formulas with DECEASEDDAT were popular: cf. DECEASEDDAT (over 20
instances), ego DECEASEDDAT (Le 75ter), TOMBNOM DECEASEDDAT (4 instances), ego TOMBNOM
DECEASEDDAT (5 instances), ego PROCURATORGEN DECEASEDDAT (2 instances), PROCURATORGEN
egoDAT (2 instances), ego PROCURATORGEN TOMBNOM DECEASEDDAT (Le 130).
312
40 and tito  uelmineo | titọi  fe cupa MLF 305). On the other hand, it is likely that the
forms in -ai in MF 270 and MLF 308, and possibly also those in MF 154 and MLF 310
are datives. In the case of popliai file MLF 308, there is some external evidence for a
dative popliai file(ai), for the grave-gifts found in the undisturbed loculus to which the
inscription belonged show that the deceased was a woman rather than a man (Giglioli
1916:68-72), which precludes interpreting popliai file as a metronymic filiation popliai
file(o). Such an interpretation is of course still possible in the cases of larise  mar||cna 
citiai MF 270 and [---] zaconiai MF 154.161 Note, however, that although metronymic
formulas are well-attested for Etruscan, there is no (other) trace of them in inscriptions
from the agri Faliscus and Capenas: neither Cristofani (1988:18) nor Peruzzi
(1990:280) even consider this interpretation for MF 270. They also tend to follow
patronymic formulas, which is not the case here.
8.10.3. OWNERGEN cella and the ius sepulcrale. Another type of sepulchral inscriptions
occurs on the outside of the tombs, namely OWNERGEN cella in cau‹i›o : pauiceo : |
ḷ[oc]ịes : cela MF 12, *[.]pi : uesθi : cela MF 83, caui [:]t **(*)[i] : cela MF 84, and
[---]fate cela MLF 285. As is argued in §11.1.4.1a, these inscriptions refer to the owner
of the tomb, and designate the right to a place of burial rather than the burial itself, even
though the owner will of course have been buried there as well.
The formula OWNERGEN cella is not found in Latin or Sabellic inscriptions, and is
clearly an adaptation of Etruscan formulas like eca σuθi/σ́uθi [neσl] OWNERGEN
(Steinbauer’s type G3).162 Interestingly, this formula is relatively frequent in the nearby
areas of Tarquinii (17 instances: Ta 1.31, 45, AT 1.11, 159, 177, 178 (with ca), 192
(with ta); with neσl AT 1.30, 70, 138, 140, 141, 148, 188; unclear AT 1.146, 149, 163),
of Vulci (14 instances: Vc 1.1, 2, 3, 4, 5, 10, 16, 45, 69, 77, 102, AV 1.5, 8, 14;
OWNERGEN σuθi Vc. 1.102), and of Horta (only AH 1.81): elsewhere, it occurs only
sporadically (4 instances: without eca Pe 1.168, 328; OWNERGEN ta σuθi AS 1.9;
fragmentary Pe 1.688). The Faliscan adaptation, with OWNERGEN preceding cela (cf.
§8.4.2), closely resembles, not the most common Etruscan type eca σuθi/σ́uθi
OWNERGEN from Tarquinii and Vulci, but the unique OWNERGEN σuθi of Vc 1.102, larθia
| larθial anienas σuθi.
Since the Faliscan formula contains no equivalent of the deictic element eca/ca/ta
in the Etruscan formula, it is perhaps not surprising that there are likewise no instances
of a Faliscan adaptation of the Etruscan iscrizione parlante-type mi OWNERGEN
161
In MF 270, the woman is designated by GENTILICIUM, which appears to be the normal form
of the mother’s name in Etruscan metronymic filiations: on the other hand, Etruscan metronym-
ic filiations are usually added after patronymic filiations, which is not the case here.
162
Steinbauer (1999:472, 473) translates σuθi as “Grab(stelle)” because of its connection to
σuθ- (“viell. ,,legen, stellen” ”), and tamera as “(Grab-)kammer”. For cela as a Latin-Faliscan
rather than an Etruscan word, see §6.2.8.
313
CHAPTER 8
σuθi/σ́uθi (Steinbauer’s type G1), found mostly at Volsinii (Vs 1.43, 54, 73, 86, 98,
136, Fe 1.7), and mi σuθi/σ́uθi OWNERGEN (Vc 1.78, Cl 1.946, Fe 1.2, 9, Li 1.1).163
Related inscriptions are those that state the number of loculi in the tomb, as in [---]fate
cela lete zot xxiiii MLF 285, or the right to burial there, as in iii l[.......]|[..........]nai
[?---]|[....]o uxo MF 17 (if = ‘the third lectus ... for ...na’ or ‘three lecti ... for ...na’), and
perhaps precono | cuiteneṭ | let MLF 361 (if = ‘Praeconus here (?) keeps (?) a lectus’),
and most clearly in the Latin inscription l uecilio uo f et | po[l]ae abelese | lectu
i datus | [. ]uecilio l f et plenese | lectu i amplius nihil | inuiteis l c
leuieis l f | et quei eos parentaret | ne anteponat Lat 251. These instances show
that lectus was a terminus technicus for ‘(place in a) loculus’.164
8.11. Dedications
8.11.1. OWNERNOM, OWNERGEN, RECIPIENTDAT. As in the case of the Besitzerinschrif-

ten (§8.7.1), there are three basic types of dedicatory inscriptions, the first just naming
the divine owner (OWNERNOM), the second stressing the possessive relationship between
the inscribed object and the divine owner (OWNERGEN), the third stressing the act of the
dedication itself (DEDICANTNOM RECIPIENTDAT).
The only example of OWNERNOM is
apolo MF 65
Since this inscription was found in a tomb, it has been interpreted as an abbreviated
human name Apollo(...). Dedicatory inscriptions of the type OWNERNOM occur in
Etruscan, however, although infrequently (Steinbauer’s types S1-2, e.g. flere Ar 3.1,
fuflunz Um 4.1-2, tiniia Ta 4.2, tinia AH 4.1, tinia caluṣna Vs 4.7, turan Li 4.1, uni Ta
4.12), and rarely also in Latin (only iuno loucina CIL I2.375, marspiter CIL I2.970,
diana CIL I2.1435, diana af louco CIL I2.2444, although the last two may be instances
of a dative in -a, cf. §4.2.3) and in the Sabellic languages (Marsian esos | nouesedẹ |
p ̣esco pacrẹ VM 5 and Oscan dí{}pa{}tír púḷ[ie]ṿṣ Po 22).
The second type is OWNERGEN, attested for Faliscan in
apolonos EF 10
loifiṛtato MF 31
loifirtato MF 32
163
This type is not discussed by Agostiniani, but is comparable to Agostiniani’s types E6
(Agostiniani 1982:187-97) and E21 (Agostiniani 1982:221-3).
164
M.Mancini (2002:28-33) argues against interpreting let and lete as forms of lectus, and
suggests instead that they represent a */lŏtā/, which, however, has the same meaning. The
corresponding word in Etruscan may have been tunu- or tusu- (cf. Steinbauer 1999:483, 484).
314
As this type is comparable to the Besitzerinschriften in the genitive (but with a divine
owner), there is no need to assume that a word like sacrum was to be understood:
something that is questionable in any case, as sacer appears to be more frequently used
with the dative thạn with the genitive at least in Latin (cf. e.g. the instances quoted
below). OWNERGEN occurs in Latin, although rarely: a contextless genitive of the deity’s
name occurs only in salutis CIL I2.1626a, deum | maanium CIL I2.2117, and uenerus
heruc CIL I2.2297. In the Sabellic languages, the type is found in Praesamnitic toutikej̣
dipọterej̣ Ps 1, Marsian apols VM 7, and Oscan fatuveís Hi 6, iúveís | lúvfreís Fr 5.
A deity’s name in a genitive is not uncommon in Etruscan dedications (Steinbau-
er’s types S3-9), e.g. artmsl Ru 4.3, fuflụnsl paχ(ies) Vc 4.3, [fuflunsl p]aχies Vc 4.4,
fuflunl paχies velclθi Vc 4.1-2, menervas Ve 4.1, menervas Cr 4.1, [m]enrua[s] Cr
4.17, menṛ[vas] Cr 4.18, tuṛ[ns] Ta 4.9, turṇ[s] Ta 4.10, turns Ta 4.11, 13, : unial : Cr
4.8, unial : Cr 4.9. According to Steinbauer (1999:170-1), in Etruscan, where a ‘true
dative’ was lacking, the genitive could fulfill functions that in an Italic language would
have been fulfilled by a dative: the Etruscan type OWNERGEN could therefore also be
described as RECIPIENTGEN. This may explain why three Faliscan dedicatory inscriptions
are of the type OWNERGEN, which is a relatively large number.
In dedications, however, the dedicant wants to stress, not just that the dedicated object
belongs to the deity, but that it was given, by using the elliptic verb phrase RECIPIENTDAT
DEDICANTNOM or even just RECIPIENTDAT (cf. §8.7.1).
titoi | mercui | efiles MF 113
titoi | mercui | efile MF 114
[t]ito[i] | ṃercu[i] | efiles MF 115
titoi | mercụ[i] | efi[les] MF 116
[titoi | mercui | efil]es MF 117
(with ellipsis of the name of the dedicant:)
titoi : mercụ[i] MF 118
[t]itoi : mercu[i] MF 119
[ti]ṭoi : mercu[i] MF 120
[ti]toi : mercui MF 121
titoi : ṃ[e]rcui MF 122
[titoi : ?]merc[ui] MF 123
[m]ẹrcui MF 124
ṃercui MF 125
meṛ[cui] MF 126
and possibly [---]ạltai  MF 109 (if this is a dedication)
Both variants of this type are common in Latin and in the Sabellic languages. In Latin,
RECIPIENTDAT DEDICANTNOM occurs in nine instances (diana mereto | noutrix paperia
CIL I2.45, apoline | l carnius c f CIL I2.2219, dianai opifer | nemorensei | l
315
CHAPTER 8
apuleius l l antio CIL I2.1480, fortunae opse[q] | p peilius l f c caluius p

f | cens CIL I2.1509, diuei | ardeates CIL I2.39, uediouei patrei | genteiles iuliei CIL
I2.1439, [iuno]ne regina | [. an]toni p f l uirgini t f | mag pag CIL I2.1993,
hercolei | tesorus CIL I2.2220, ueneri u[ictrici] | l bombius m f pro i CIL I2.2246),
and DEDICANTNOM RECIPIENTDAT in six instances (l mr fouṛ[---]|menerua[i lindiai] CIL
I2.404, q caecilius cn a q flamini leibertus iunone seispitei | matri reginae
CIL I2.1430, [---]s m f hercolei CIL I2.1579, [---]| tampia l f | diouei CIL I2.2171,
c nrius eros apolline et iouei et neptuno minerua et sis mircurio CIL I2.2233, p
laeuius | fortun[ae primig] CIL I2.2531). In the Sabellic languages, RECIPIENTDAT
DEDICANTNOM occurs in dio#hi [?---]| titidiej [?---] Lu 27, and DEDICANTNOM
RECIPIENTDAT in tanas : niumeriis : | frunter[eí] Sa 27, lúvkis ṿeleí[is] | mefitẹí |
ạravinaí Hi 3, and siviiu magiú | mefit(eí) Hi 4. In the Latin instances, the order
RECIPIENTDAT DEDICANTNOM, which is the order of the Faliscan dedications of this type,
is slightly more frequent.
The shorter form RECIPIENTDAT, with ellipsis of the name of the dedicant, appears
in Latin in 20 instances (iounonei loucina CIL I2.362, apolenei CIL I2.368, fide CIL
I2.369, iunone CIL I2.370, salute CIL I2.373, lebro CIL I2.381, remureine CIL I2.971,
mauortei CIL I2.991, terrai | matri CIL I2.995, [mer]q̣urio CIL I2.2243, meneruae CIL
I2.2194, uenerei erucina[CIL I2.2221, apoline CIL I2.2628, iue CIL I2.2630, uictorie
CIL I2.2631, castorei  podlouqeique | qurois CIL I2.2833, luei CIL I2.2848a, [a]polo |
menerua | [gr]aiua CIL I2.2876, herecle CIL I2.2887b, hercolei CIL I2.2889), and in the
Sabellic languages in nine instances (menervai Pg 8, heretateí Cm 11, mirikui Cm 12,
#enzhi mefiti[ai] Lu 31, [m]efithi | kaporoinnai Lu 32, mefithi | ara#inai Lu 33, meḅithi |
ara#inai Lu 34, zw#hi | pizhi Lu 35, mamertei | mefitanoi Lu 36).
In Etruscan, this type is completely absent, probably because of the lack of a
‘true’ dative (see above). The only text that resembles DEDICANTNOM RECIPIENTDAT is
uras arnθ θufl(θas) σu{u}ris (Cristofani 1990b, Steinbauer 1999:288).
8.11.2. Other types of dedicatory inscriptions. Another type of dedication stresses the
relation between the inscribed object and the deity by the word sacer. For Faliscan, this
type is attested in sacra MF 127, and perhaps in abbreviated form in sa MF? 131 (and
perhaps sa MF? 76, from a tomb). The type occurs with some frequency in Latin, e.g.
fortunai | poblicai | sacra CIL I2.397, sacro matre mursina CIL I2.580, deuas |
corniscas | sacrum CIL I2.975, iunone | loucina | tuscolana || sacra CIL I2.1581,
[---]ole | [tusc]olana | sacra CIL I2.1582, salutei | sacrum CIL I2.1626, iouei libero
s[acr-] CIL I2.1838, sacra lanuio CIL I2.2296, saturno | sacro CIL I2.3375, diouei
mourc | sacr CIL I2.3171, cerer sac CIL I2.3471, herc sa CIL I2.3472a, her sac ad
laue CIL I2.3472b, and uen sac CIL I2.3474.165 In the Sabellic languages the type is
165
These are the fourteen Latin instances of dedications consisting only of the word sacer and
the name of the deity, without any mention of the dedicant.
316
attested for Umbrian in [cupr]as matres pletinas sacru esu Um 17 (the same text in Um
16, 18-19, where sacru is restored) and supunne sacr Um 24, and for Oscan in appel-
lounhi sakoro Me 1-3 and herekleij s(a)k(oro) Lu 26. Leaving aside Umbrian sacre
stahu Um 10 (a boundary stone, not a temple dedication), Faliscan sacra MF 127 is
unique among these inscriptions in omitting the name of the deity: within the sanctuary
where the object was dedicated, it was of course clear which deity was meant.
More or less comparable Etruscan types, where the name of the deity is also often
omitted, are the shorter dedicatory inscriptions consisting only of σuris or tinscvil
‘dedication’, e.g. σuris Pe 0.6, σ́uriś Ar 4.2, sauvcnes : σuris AT 4.1, mi : σ́uris : ca[ Cr
4.12, and tinścvil Co 4.7-9, Ar 3.2, tinia | tinscvil Vs 4.10, tinia : ti[nscvil] Vs 4.11,
tinia : tinscvil Vs 4.13, [--- t]inscvil Vs 4.14, eśta ẓinu herma tinścvil Vs 4.12. A more
direct parallel might be mi cipaχ Etr XVIII from Narce, if Briquel (in Gran-Aymérich
& Briquel 1997) is right in suggesting that cipaχ may mean ‘sacred, consecrated’.
There are no Faliscan dedicatory inscriptions where verb phrases are used. Thulin
(1907:304) interpreted sta MF? 128 (and sta MF? 28) as sta(t) ‘stands as a sacred
object’, but sto by itself cannot be used in this sense: in the text quoted by Thulin as a
parallel, Umbrian sacre stahu Um 10, the ‘sacral sense’ is due to sacre, not to stahu.
The only dedications with a verbal structure from the area are the Latin inscriptions
from S. Maria di Falleri (LF/Lat 214, Lat 217-218, 219) and the ager Capenas (Lat 377
from Ponzano, Cap 421 from Capena, and Cap 431, 433, and 435 and Lat 432 and 434
from Lucus Feroniae), with Latin formulas. The many Etruscan types of dedicatory
inscriptions with verbs like turuce, alice (e.g. Etr VIII=Fa 3.1, Etr IX=Fa 3.3) or
muluvanice (e.g. mulu in Etr XIX=Fa 3.2) have no parallels in Faliscan: note that the
only Etruscan dedication from a temple in Falerii Veteres, anae lauv|cies Etr XXIX=Fa
0.6 from the Tempio Maggiore at Colle di Vignale, apparently only names the dedicant.
8.12. Official inscriptions
Official inscriptions are scarce. Usually, they consist only of the names of the magi-
strates responsible, as in cauio lullio MF 207, cauio latinaio MF 210, a [.]osena uẹ
narionio MF 206, and c**(*)coṇẹo l***(*) ce paui[cio 1-2]so MLF 290. These can
in fact be considered comparable to signatures of the type MAKERNOM (see §8.8.1). A
magistracy is named only [1-2 ] hirmio m[ f ]ce tertineo c f preṭ[ores ?---] LF
213: it is possible that [---]ilio c[ f ---] LF 215 also contained a magistracy. As in the
case of dedications, the more explicit verbal formulas occur only in the Latin inscrip-
tions from the area (Lat 291 from near Corchiano, Lat 456 from Lucus Feroniae).
As an expression from the official language should also be mentioned macistra-
tu | keset LF 242, ma|cistratu?] || kese[t LF 243). This corresponds to the Latin
formulaic official expression magistratum gerere (for instances, see TLL 6.2
317
CHAPTER 8
1939.1-1940.56), and the formula may well have been taken over from Rome or
Latium, as was the use of the cursus honorum itself (§11.1.4.1), and most of the official
vocabulary (§6.3.6).
8.13. Summary of §§8.8-12
When we look at the textual types and the formulas discussed in §8.8-12, the following
tendencies emerge:
(a) General textual types such as OWNERNOM and OWNERGEN in the Besitzerinschriften
(§8.8.1), MAKERNOM and MAKERGEN in the signatures (§8.9.1) and the official inscrip-
tions (§8.12), DECEASEDNOM in the sepulchral inscriptions (§8.10.1), and OWNERNOM and
OWNERGEN in the dedications (§8.11.1) have parallels in Latin, Etruscan and the Sabellic
languages. However, OWNERGEN in the dedications appears to be relatively frequent in
comparison to the occurrence of this type in Latin and the Sabellic languages: this may
be due to Etruscan, where the type had a function equal to that of RECIPIENTDAT in Latin
and the Sabellic languages (§8.11.1). The dedicatory type using sacer has parallels in
Latin and Sabellic, but not in Etruscan, and may be considered Italic (§8.11.2).
(b) The Early Faliscan iscrizioni parlanti generally follow Etruscan models, e.g. in the
case of the Besitzerinschriften of the type ego OWNERGEN, ego POSSESSIONNOM
OWNERGEN and ego OWNERNOM (§8.8.2), although they tend to be adapted to the
Faliscan word order, e.g. in the Besitzerinschriften of the type OWNERGEN ego (§8.8.2).
Alternatively, the Faliscan type reflects an Etruscan variant with a word order that is
closer to the Faliscan one, as in the Besitzerinschrift of the type OWNERNOM ego (§8.8.2)
and the signature formula MAKERNOM me MAKER3RD PF (§8.9.2). The Capenate Besitze-
rinschriften with OWNERGEN sum, on the other hand, appear to be of Sabellic rather than
of Etruscan origin (§8.8.2). The uniquely Capenate type OWNERNOM sum may be due to
third-century Latin replacement of the Etruscan type OWNERNOM ego (§8.8.2).
(c) The Middle Faliscan sepulcral type DECEASEDNOM hec cupat/cupant has parallels
both in Etruscan and in the Sabellic languages, but is used in Faliscan with a frequency
that far exceeds that of its parallels, and can therefore at least in this respect be regarded
as Faliscan (§8.10.1). Another sepulchral type, OWNERGEN cela, is probably modeled on
the South Etruscan inscriptions with eca σ́uθi/σuθi OWNERGEN (§8.10.3), but the word
order is again that of Faliscan rather than that of the most common Etruscan model.
(d) The Latin dedicatory and official inscriptions show textual types with verbal
formulas that do not appear in the Middle or Late Faliscan inscriptions (§8.11.2, §8.12):
they also appear to have influenced the Late Faliscan sepulchral inscriptions (§8.10.2).
318
Chapter 9
Language contact
As said in §1.1, one of the interesting features of Faliscan is the fact that it is found between
areas where Etruscan, Sabellic languages and Latin were spoken. Language contact must have
been frequent, and it can sometimes be traced in the texts by features occurring in Faliscan
inscriptions that can be ascribed to influence from these languages. This chapter opens with a
number of methodological observations on the problems of observing language contact from
texts such as we have for Faliscan (§9.1). In the next sections I discuss the features in the Falis-
can inscriptions that can be ascribed to language contact with speakers of Etruscan (§9.2),
Sabellic languages (§9.3), and Latin (§9.4). A short conclusion is drawn in §9.5.
9.1. Language contact in the ager Faliscus: methodological issues
In the following sections, I have tried to map which features can be ascribed to lan-
guage contact between Faliscan on the one hand and Etruscan, Sabellic languages,
and Latin on the other. The study of language contact involving fragmentarily docu-
mented languages such as Faliscan involves a number of problems, and can only re-
sult in the most tentative of conclusions. Therefore, in spite of excellent discussions of
the problems involved by Adams & Swain (2002), Adams (2003:1-29, 2007:1-36),
and especially Langslow (2002), my remarks on the methodological issues will be
longer than usual. They are largely meant to adapt the framework used by these au-
thors (which is also based on literary texts, sometimes even texts pretending to repre-
sent spoken language, e.g. comedy) to the conditions of the material on which this
study is based.
This material is, in the first place, fragmentary in the sense that it does not even
allow drawing a complete picture of the languages involved in the contact. Worse,
epigraphic material is fragmentary also in another sense, namely in the sense that an
inscription normally represents an individual ‘utterance’, generated as a kind of ‘one-
sided interaction’ directed towards a (conceived) reader. This is a great disadvantage
when studying language contact, for language contact presupposes a spoken interac-
tion between at least two speakers involving more than one language. Therefore, an
epigraphic source that is not consciously conceived as a bilingual text but does show
features of code-switching should be regarded as representing the mental choices of
one bilingual individual for one code or another. Unfortunately, there is no possibility
to check to what degree that individual was bilingual, or in how far the choices from
his or her bilingualism represent those made by the language community as a whole.
319
CHAPTER 9
An even more structural problem with regard to the description of language contact on
the basis of material such as the Faliscan corpus is that all the material used as data in
this study is written, and that written language is only an indirect or derived form of
language use. Writing is always an acquired process, and does not necessarily reflect
spoken language at every turn. Acquiring the art of writing a specific language in-
volves mastering a set of rules associated with the writing of that language and its
written tradition. These rules concern not just the alphabet and its related orthographic
conventions (including e.g. historically motivated spelling) but also conventions about
what is allowed in spoken but not in written language (in effect, level-distinction), and
the construction of certain types of texts, including the formulas belonging to each
type. Each written language therefore has its own individual set of consciously ac-
quired prescriptions that tell the individual how he or she should ‘translate’ a mentally
conceived or spoken utterance into one that is acceptable within the written form of
that language. The process of this ‘translation’ is a conscious application of these
rules, although ‘conscious’ is of course a relative concept here: depending on how
well-versed the writer is in writing a specific language, it may be a process that will
be regarded as hardly conscious at all by the individual in question.
As any user of both spoken and written language can testify, there is often a vast
gap between how we speak and how we write, even if this speaking and writing con-
cerns the same utterance. Some features of spoken language (such as repetition and
other forms of over-marking that are necessary to compensate for the ‘loss’ in the
transmission of a spoken utterance) simply do not enter written language because they
are unnecessary, but others are not used because the rules tell us so, often because the
written language, which is taught and learned, is necessarily more conservative than
the spoken language, and therefore by its very existence produces a level-distinction
between spoken and written language. 166 In how far a written speech utterance resem-
bles this desired standard or, on the other hand, the language as it is actually spoken,
therefore depends (a) on the width of the gap between written and spoken language in
the framework of the writing of that specific language, (b) the degree of knowledge
that the writer has of that framework, and (c) his or her ability to apply these rules.
As an acquired form of language that requires a conscious effort to use, written
language is by its nature more formal than spoken language. Language contact, on the
other hand, is first of all a matter of the spoken language, of the interaction between
speakers of two different languages, and phenomena such as interference and code-
switching are normally the result of unconscious psycholinguistic processes. Before a
166
This does not mean that I assume a Faliscan diglossia, as does R. Giacomelli (1978 pas-
sim) unless this term is (in my view wrongly) applied to the level-distinction between written
and spoken language: see §1.3.2. Since the Faliscan inscriptions are not only written language
but also highly formulaic, I find it difficult to assume that they could represent ‘informal
speech’, as does R. Giacomelli (2006 passim).
320
LANGUAGE CONTACT
feature of language x can occur in a written text of language y, it must therefore not
only pass the threshold into a different language: it must also cross the additional
threshold of the formalization of the speech utterances of language y into the written
form of language y, and therefore either fit in with or in some way circumvent the
rules of writing language y.
This means that observing phenomena that can be classed under the headings of
‘interference’, ‘borrowing’, ‘code switching’ etc. in written material such as the epi-
graphic material on which this study is based can imply (a) that that specific feature
was accepted as permissible in the written form of language y (and that recurrent oc-
currence therefore represents borrowing rather than interference), (b) that the writer
was uncertain or non-cognizant of (certain aspects of) the rules governing the written
form of language y (probably implying that he or she had another language as his or
her first language) and/or (c) that in spite of the writer’s knowledge of the rules of the
written form of language y, language x was what has been called ‘psychologically
dominant’ (cf. e.g. Muysken 1995), i.e., that in the mind of the bilingual individual it
dominated the decisions with regard to the use of the codes of either language to such
an extent as to overrule the rules of the written form of language y.
To be able to assess phenomena that can be ascribed to language contact, it is also
necessary to establish the nature of these language contacts. Unfortunately, for the
period and the material under discussion here, this can only be done in the most gen-
eral way: in effect, little more can be done than to refer in very general terms to such
forms of contact as trade, transhumance, war, colonization, immigration, and inter-
marriage. Any details that can be added to this are finds derived from the archaeologi-
cal and historical sources, for which see §2.1-7. There is usually no way in which
these can be quantified or specified, or connected with specific features of language
contact that might be observable in the texts.
Establishing the nature of the language contact, even in these general terms, also
shows on which level such contact took place. In the case of Faliscan, much of the
contact will have been informal, e.g. with Etruscan-speaking neighbours and in-laws,
with Latin-speaking tradesmen from the colonies at Sutrium, Nepete, and Capena, or
with herdsmen from the Sabellic-speaking interior. Yet there must also have been
language contact on a more formal level, especially at the level of the governing bod-
ies: during the period before the war of 241-240, Falerii was allied with the South
Etruscan city-states, and, if we are to believe Livy (4.23.4-24.2, 4.25.7-8, 5.17.6-10),
even attended the meetings of the Etruscan League at Fanum Voltumnae, while during
the period after 240, there will have been frequent contact with Roman officials, even
if the magistrates themselves were from local families (§2.6.2). Of the former contact,
no trace remains: of the latter, evidence can be seen in a number of official and sepul-
chral inscriptions from the period after 240 (see §9.4.2).
321
CHAPTER 9
A last remark should be made with regard to onomastic borrowing, a point that has
already been touched upon in §1.3.2.2 and discussed in more detail in §7.1.1. The
existence of the individual per se entails the use of his or her name, not only within
the language community where that name originated, but also in another language
community, where it automatically generates a situation where his or her name be-
comes an interferential form. This onomastic borrowing is both vastly more frequent
than the borrowing of other elements of speech, and cannot be treated in the same way
as other observable features of language contact.
9.2. Faliscan and Etruscan
9.2.1. The nature of Faliscan-Etruscan language contact. Generally speaking, the

inhabitants of the ager Faliscus may be assumed to have been in quite regular contact
with speakers of Etruscan: indeed, especially in view of the onomastic material and of
the number of Etruscan inscriptions from the area, it is not unreasonable to assume
that a number of autochthonous inhabitants of the ager Faliscus were speakers of
Etruscan as a first language: the presence of such Etruscan-speaking individuals and
groups in the ager Faliscus has been discussed by Cristofani (1988): see also §9.2.3.
Of those individuals that had Faliscan as their first language and those that had
Etruscan as their first language, a reasonable proportion must have been bilingual,
although it cannot be ascertained to what extent, and whether they acquired both lan-
guages at an early age or were bilingual due to later acquisition of a second language.
In any case, interaction between autochthonous speakers of Faliscan and autochtho-
nous speakers of Etruscan must have been an everyday occurrence.
This interaction between Faliscan and Etruscan did not only take place on the
level of individual speakers. If the roadside inscriptions Etr XXXVIII and XXXIX
can indeed be classed as inscriptions on public works (§11.1.4.5), they show that in
the ager Faliscus Etruscan could be used for an official inscription, even if this may
have been possible only on a local level. The Roman and Greek historical sources
(quoted in §2.5-6) report unanimously that the Faliscans as a city-state were allied
with the Etruscans: as a community, the ager Faliscus had its political roots in the
Etruscan world, which must have meant that speakers of Etruscan must have been
common at government level. When Livy (9.36) describes Fabius’ incognito journey
through the silua Cimina in 310, he even envisages him as a man who was suited for
the job because of his proficiency in speaking Etruscan. 167
167
Livy’s portrayal presents him as a typical ‘elite bilingual’: “Caere educatus apud hospites,
Etruscis inde litteris eruditus erat linguamque Etruscam probe nouerat. habeo auctores uulgo
tum Romanos pueros, sicut nunc Graecis, ita Etruscis litteris erudiri solitos”.
322
LANGUAGE CONTACT
Yet the epigraphic material indicates that the (written) language of the ager Fal-
iscus was in the first place Faliscan, not Etruscan. This can be observed first of all
from the number of inscriptions. During the Early Faliscan period, the Etruscan in-
scriptions are in the majority: there are 19-20 early Etruscan inscriptions from the ager
Faliscus, 18 of which are from Narce (which appears to have been an Etruscan-
speaking town and has yielded no Faliscan inscriptions) and the south-west (Etr I-XV
and XVIII-XX: only Etr XXXII is from Corchiano), against only 9-12 Early Faliscan
inscriptions (EF 1-4, 6-10, 467*, EF/Etr 5 and 385). During the Middle and Late Fal-
iscan periods however, the Faliscan material far outnumbers the Etruscan inscriptions
even in the most generous count: Civita Castellana yields only 7-16 Etruscan or pos-
sibly Etruscan texts (Etr XXV-XXXI, MF/Etr 37, 61, 64, 66-67, 77, 199, MLF/Etr
208-209), Corchiano 10-16 (Etr XXXII-XLI, MF/Etr 256, 264, 267, 279-280,
MLF/Etr 289), the remainder of the ager Faliscus only 2-4 (Etr XLII-XLIII MLF/Etr
356-357), and the ager Capenas only 2 (Etr XLIV-XLV). Another point that shows
that Faliscan was the first language of the area rather than Etruscan is that although
many gentilicia are undoubtedly of Etruscan origin, these names were very often
adapted to Faliscan: see §7.8.2.2,5 and §7.10.3.
9.2.2. Etruscan features in Faliscan inscriptions. During the 150 years of Faliscan
studies (§1.5), a number of features have with more or less justification been ascribed
to Etruscan influence, especially by Herbig and those who closely followed his publi-
cation of the Faliscan material in the CIE (!):
(1) Phonological features. I can see no phonological features that have to be ascribed
to Etruscan. The monophthongization of the Faliscan diphthongs more or less coin-
cides with similar processes in Etruscan (§3.7), but it is very unclear whether this can
(or must) be due to contact with Etruscan. Monophthongization of diphthongs af-
fected most of the Italic languages at one time or other, and can be described as a
‘natural’ phenomenon. Those who wish to do so can regard this as an attestation of a
Sprachbund (§1.3.1.1): if there were any relationship between the developments in
both languages, it would be an attestation of the frequency of language contact. The
Middle Faliscan realisation of /#fV/ as [h] has been ascribed to Etruscan influence
(§3.5.2), but as the Faliscan development preceded the Etruscan development, and the
Etruscan development was restricted to the north and north-east of the Etruscan-
speaking area, this is altogether unlikely.
One feature that may perhaps be observable in the inscriptions is that the Etrus-
can realization of /s#/ may have been ‘stronger’ than in Faliscan, i.e., [s] rather than
the Faliscan [h] or [ʔ] (§3.5.7d). Arguments for this may be the following forms: (a)
the Etruscoid forms in -(i)es, where omission of -s is notably less frequent than in the
Faliscan forms in /s#/ (§3.5.7d); (b) morenez MF 269 and perhaps ạcṛẹẓ MF/Etr 67,
323
CHAPTER 9
which have -ez instead of -es, perhaps representing a ‘strongly’ realized /s#/ (§3.5.3),
although I doubt whether this can be maintained; (c) larise MF 270, MLF 371-372,
reflecting Etruscan Laris, which has been explained by Vetter (1953:317, 325) and
Peruzzi (1990:281 n.9) as a form with an epenthetic [-e] or [-ə] following a ‘strongly
realized’ /s#/, although I would rather regard this form as an accusative larise(m) used
instead of the nominative (see below). 168
(2) Morphological features. On the morphological level, there are several features
that have been regarded as Etruscan:
(a) nominatives in -i. A number of forms in -i have been interpreted as Etruscan
feminine forms. In several cases this interpretation is at least possible:
(1) ca uipi  leueli | filea MF 14 (Thulin (1907:281-2 (Cauipi), Herbig CIE 8075,
Morandi 1982:58-9): here the interpretation ca uipi = Gauia Uibia is plausible, also
because, if uipi leueli is interpreted as a Faliscan genitive, the father is referred to with
PRAENOMEN GENTILICIUM, which is unique in FILIATION (§7.5 with figs.7.3-4).
(2) [---]o cicio  cicoi  cupat  ifra MF 40, where cicoi was interpreted as a rendering
of an Etruscan feminine *cicui by G. Giacomelli (1963:84, 1965a): this appears to be
a way of avoiding having to assume a genitive in -oi, however, since she did not pro-
pose this interpretation for any of the other forms -oi. I would rather interpret this
form as a second-declension genitive (a dative is impossible here): see §4.4.4.
(3) larise  mar||cna  citiai MF 270, where citiai is regarded as a Faliscan nominative
recharacterized with the Etruscan feminine -i by Cristofani (1988:18) and Peruzzi
(1990:278-81): it can also be interpreted as a dative, or possibly a genitive (§8.10.2).
(4) [---]ạltai MF 109, read as ven]eltvi by Colonna (1986:172-3) and as ?]ẹltvi by Rix
(ET Fa 2.19).
An Etruscan interpretation is unlikely in the following cases: (1) titoi |mercui |efiles
MF 113-117 and titoi  mercui MF 118-122 (Herbig 1914, 1923); (2) uolti  catinei MF
469* (Froehner in Lejeune 1952b:115-6); (3) caui  tertinei  | posticnu MLF/Cap 474*
(Herbig CIE 8339, Vetter 1953:308, Pisani 1964:345); (4) uolta | neroni | ca fi MF 15
(Pisani 1964 335-6); (5) serui MF 34-36 (Herbig CIE 814-8016).
(b) nominatives in -u. There are also cases where forms in -u have been interpreted as
Etruscan nominatives in -u: in every case, I think that another interpretation or reading
is more likely: (1) [fel]ịcinatiu LF 384 (Renzetti Marra 1974:351, Briquel 1972:820,
826, Rix ET Fa 2.17), rather a Faliscan genitive plural; (2) posticnu MLF 474† (Lejeune
1952b :114-20), rather a Sabellic first-declension nominative; (3) tuconu MF 85 (Gar-
rucci 1860:270 etc., Vetter 1953:300), but here the Etruscan interpretation appears to be
168
The forms quoted under (b) and (c) would represent a conscious effort to express in written
Faliscan texts a different phonetic realization in spoken Etruscan, which I find hard to believe.
324
LANGUAGE CONTACT
a case of ‘inexplicable, therefore Etruscan’; (4) the uaria lectio fafariu in MF 136 (Le-
jeune (1952b:120 n.1), which is rather to be read as fafarṇ, and (5) ueiụeto MLF 464
(reflecting an Etruscan *uei uetu according to Herbig (CIE 8465)).
(c) Etruscoid forms in -(i)es/-(i)e. I have called these forms ‘Etruscoid’ rather than
Etruscan since they appear to be Etruscan nominatives that originally arose from Italic
forms, those in -(i)es perhaps from Italic nominatives in */-os/ and */-ios/ (with a
weakening of the vowel within Etruscan itself?), 169 and are therefore the result of a
much older contact between Etruscan and the Italic languages: note that they are used
indiscriminately for the masculine as well as the feminine. These forms are used as
the nominative of women’s names in fasies MF 41, satelie MF 42, calitenes MF 265,
morenez MF 269, petrunes LF 226, plenes LtF 231 (and hence apparently used as the
basis of a dative in plenese Lat 251), and as the nominative of men’s names in cesie
MF 257 and uenelịes MF 258 (in [---]uenelịes sapnonia, where uenelies is probably a
patronymic adjective ‘son of Venel’ following a man’s name). In other inscriptions,
they occur where a genitive of a man’s name is expected in ḷ[oc]ịes MF 12, cesies MF
265: see below under (4) on confusion of cases in Etruscan-Faliscan contact. Form in
-ie(s) that can be either nominative or genitive occur in ulties MF/Etr 64 (without con-
text) and eco tulie LF 383 (which can be either ego OWNERNOM or ego OWNERGEN, see
§8.8.2.). Probably a genitive is ạcṛẹẓ MF 67 if this can in fact be read at all. In Etruscan
inscriptions from the area nominatives in -(i)es occur in Etr XI-XV, XXVI, XXIX,
XXXIV, XXXIX, XLIII, XLV, XLVIII.
In all these cases of ‘Etruscan morphology in Faliscan inscriptions’, it should be noted
that the Etruscan form is always the (unmarked) nominative: these forms are therefore
not so much instances of morphological interference, but rather of onomastic borrow-
ing, the morpheme being borrowed as part of the form of the name, not as a morpheme.
There is no indication that Etruscan morphemes were borrowed into Faliscan: in fact,
Etruscan names in -a are regularly declined according to the Faliscan first declension,
and names such as Arnθ and Larθ according to the Faliscan third declension (§4.1), as
might be expected. If larise MF 270, MLF 371-372 is an accusative larise(m) (see
below under (4)), Etruscan Laris appears to have been declined as a consonant-stem170
instead of being declined according to the Faliscan i-stems, which had a nominative in
-is (§4.5.1.5). Similarly, an Etruscoid form in -es apparently constitutes the basis for the
dative plenese in [.]  uecilio  l  f  et  plenese | lectu  i Lat 251.
169
I doubt whether they can be explained as going back to a Italic vocatives in */-(i)e/ (thus
Adams 2007:97-100, with literature): this would explain the co-occurrence of -ie, -e, and -i in
Etruscan, but does not explain the -s unless this is regarded as a subsequent recharacterization
of these forms within Etruscan (after the Italic nominative?).
170
Note that the Latin gentilicium Larisius is apparently derived in the same way.
325
CHAPTER 9
(3) Lexical features. Lexically, there is very little that can be ascribed to Etruscan. A
possible interferential form is Etruscan puia read by Vetter (1953:305) in MF 144,
[.?]ạ*ịạ | lepuia | uoltilia. Although difficult for other reasons, this interpretation is
not unlikely in view of the great number of Etruscan sepulchral inscriptions where the
woman is designated with the word puia: the inscription may have been written by an
Etruscan who thought in the terms of the Etruscan sepulchral formula (a case of
Etruscan being ‘psychologically dominant’). Its occurrence within a formula might
imply that the word is interferential and not a borrowing. Perhaps Etruscan, but unex-
plained, is eitam EF/Etr 5 (cf. below under (4)).
Possibly a borrowing rather than an inferential form (judging by the fact that the
word also occurs in Latin) is the word clipeus/clipeum that underlies the Berufsgen-
tiliz Clipearius, occurring in clipeaio (clipea‹r›io?) MF 470*, clipịaṛ[io] LF 230, cli-
peaṛ[io] LtF 231, and cḷ[i]peario LtF 233. Pace Bréyer (1993:341-2), Faliscan cela in
MF 12, 83, 84, and MLF 285 = Latin cella is in my view not a borrowing from Etrus-
can, but rather a Latin-Faliscan word that occurs as an interferential form in two
Etruscan inscriptions vel  aties  velθurus  lemniσa  celati  cẹsu Ta 1.66 and cela  sal
 θn Vc 0.40 instead of the normal word σuθi/σ́uθi/σ́uθi.
To these instances must be added Radke’s (1965:138) hybrid hutị[p]ilom ‘four-
fold’ in EF 1, from Etruscan huθ + Latin -plum (?), and Pisani’s (1964:341) verb
ip ̣ice in MLF 309 and 315, from an Etruscan ipi ‘urn’ + Etruscan third singular active
past tense -ce. Neither of these is acceptable, especially the former: cf. §1.3.1.2.
(4) Syntactic features: confusion of the cases. A syntactic feature that may be as-
cribed to Faliscan-Etruscan contacts is that in several inscriptions there appears to be a
confusion of the cases. The clearest instance, in my view, is the use of the accusative
for the nominative. I have assumed this in the cases of arute in arute macena | more-
nez MF 269 and of larise in larise  mar||cna  citiai MF 270, and larise uicina MLF
371 and larise | uicina MLF 372. These forms have been explained by Vetter
(1953:316-7, 325) as containing an epenthetic -e [-e] or [-ə] marking a ‘strong’ pro-
nunciation of a word-final consonant that in Faliscan was pronounced so weakly that
it could be omitted in writing: in the case of arute, the cluster /nt#/, in the case of lar-
ise, a ‘strongly realized’ /s#/. In my view these forms are better explained as
aru(n)te(m) and larise(m), that is, as third-declension accusatives with omission of -m
that also occurred in the accusative of the second declension, uino MF 59-60
(§3.5.7a). Since in the nominal forms Etruscan made no morphological distinction
between the nominative and the accusative (Rix 1984:211, Steinbauer 1999:167-70),
it is quite possible that these forms were confused in the contact with an Italic lan-
guage where this distinction was made. Such a confusion might also underly the ‘iso-
lated accusative’ eitam EF/Etr 5 and other such isolated accusatives occurring in Latin
mirqurios alixentrom CIL I2.553 and diouem prosepnai CIL I2.558 from Praeneste, in
326
LANGUAGE CONTACT
Praesamnitic te·cliia·m Ps 16-17, and in Oscan spuriieis culcfnam Cm 27. All these
forms can be ascribed to Etruscan influence, although other explanations, such as el-
lipsis of a verb, have been proposed (see §8.2.3). If this is the correct explanation, the
writers of these inscriptions may be assumed to have been bilinguals with Etruscan as
their first language: see also §9.2.4 for other indications for this.
In two other instances, there appears to have been a confusion of nominative
and genitive, namely cauio  pauiceo | ḷ[oc]ịes  cela MF 12 , apparently ‘the tomb of
Gavius Pavicius, son of Lucius’, and poplia  calitenes | aronto  cesies | lartio  uxor
MF 265, apparently ‘Publia Calidenia, wife of Arruns Caesius son of Lars’. In the
latter case, Peruzzi (1964c:337) suggested that the confusion is due to the Etruscoid
form cesies, as such forms can be used both for the nominative and for the genitive
(see above under (2c)); Pisani’s (1964:337) suggested that it might have arisen from
the fact that in FILIATION both the genitive and the patronymic adjective could be used.
An additional factor may be Unfortunately, neither solution explains MF 12, where
the fault lies not so much in ḷ[oc]ịes as in cauio  pauiceo. It is possible, however, that
the two lines of this inscription in fact belong to separate inscriptions. These cases
show writers who were uncertain of the use of the cases, which implies that they were
bilinguals with Etruscan as first language and an incomplete command of Faliscan.
Related to this point might be the use of dedications where the name of the deity
is expressed in the genitive (apolonos EF 10, loifirtato MF 31, loifiṛtato MF 32): the
relative frequency of this type of dedication might be due to the way the genitive was
used in Etruscan: see §8.11.1.
(5) Textual/formulaic features. A feature of Faliscan inscriptions that is clearly of
Etruscan origin are the iscrizioni parlanti-formulas, which reflect Etruscan models
both in the Besitzerinschriften-formulas ego OWNERGEN eko lartos EF 6 and eko
kaisiosio EF 7, OWNERGEN ego in aịṃiosio eqo EF? 467*, ego POSSESSIONGEN OWNER-
GEN in eco quto *e uotenosio EF 3 , ego OWNERNOM in eco tulie LF 383, OWNERNOM ego
in m adicio eco LF 378 (§8.8.2), and in the signature-formula MAKERNOM me MAKE3RD PF
in tele*[1-2] med fifiked EF 9 and oufilo : clipeaio : letei : fileo : met : facet MF 470*
and mama z[e]xtos : med f[.f]ịqod EF 1 (§8.9.2). However, the Faliscan adaptations of
these formulas show that, where necessary, Etruscan models with a word order that
was alien to Faliscan were either remodelled to suit the Faliscan word order, or that a
choice was made for the Etruscan variant that most closely resembled Faliscan word
order (see §8.8.2 and §8.9.2). Yet the Etruscan formulas form a feature that could be
regarded as an Etruscan borrowing in Faliscan.
Apart from the formulas, most of the Etruscan influence in the Faliscan inscriptions
appears to be connected with onomastic borrowing. The same is true the other way
round: although Faliscan in all probability was the ‘standard language’ (or even the
‘dominant language’) of the area (§9.1), there are virtually no Faliscan features in the
327
CHAPTER 9
Etruscan inscriptions from the area, apart from onomastic borrowing: lar*s ruvries
Etr XIX from Narce, anae lauvcies Etr XXIX and perhaps cnav**es (?) Etr XXVI
from Civita Castellana, vultasi Etr XLII from Vignanello, hvulve·s· Etr XLVIII per-
haps from Mazzano Romano, cavies  uhtav[---] Etr XLV from Lucus Feroniae. Apart
from this, I argued (§6.2.8) that the occasional occurrence of cela in Etruscan inscrip-
tions may be explained as an interferential form due to Etruscan-Latin or Etruscan-
Faliscan contact in South Etruria or even the ager Faliscus itself.
9.2.3. Etruscan-speaking groups in the ager Faliscus. As discussed in §9.2.2, the

Faliscan inscriptions contain several linguistic features that can be ascribed to Etrus-
can, and to these can be added many onomastic features (§7.10.3). There are, how-
ever, two groups of inscriptions that contain more than the usual number of Etruscan
features and can be said to be more ‘Etruscan’ than other inscriptions.
The first of these groups has been treated by Peruzzi (1964c, 1990) and consists of the
sepulchral inscriptions from Corchiano. Since Corchiano was deserted as a result of
the war of 241-240 (§2.6.2), these inscriptions are probably from before this date.171
The inscriptions are presented together here. With each inscription I have also indi-
cated whether it was scratched or painted: as said in §11.1.4.1c, scratching instead of
painting occurs in only 17 inscriptions, appears to have been limited to the northern
ager Faliscus, and is in many cases connected with Etruscan linguistic features.
(a) from tomb 11 of the second necropolis of Il Vallone:
aruz  cesịe  aruto MF 257 (scratched)
[---] uenelịes  sapnonia MF 258 (scratched)
Aruz and aruto represent aru(n)s and aru(n)to(s), where the Etruscan praenomen clearly
shows adaptation to Faliscan morphology (§4.5.1, §4.5.3): note that both the father and
the son have the same Etruscan praenomen (§7.10.4-5). The gentilicium cesịe has an
Etruscoid ending in -ie(s) and recurs in two inscriptions from tomb 11 (below). The
interpretation of uenelịes is unclear (see §9.2.2.2c): it is an Etruscoid form in -ies that
appears to have been used as a patronymic adjective. Sapnonia is a name that has no
parallels elsewhere, but looks as if it is adapted from an Etruscan name *Sap( )nu.
(b) from tomb 7 of the first necropolis of S. Antonio:
poplia  calitenes | aronto  cesies | lartio  uxor MF 265 (painted)
ueltur  tetena | aruto MF 266 (scratched)
larθ  ceises | velusa Etr XXXIV (scratched)
171
Peruzzi (1990) used the gentilicia to trace several of the gentes whose members were bur-
ied at Corchiano to the area around Chiusi. I have voiced some general misgivings against this
method in §7.1.1: note that several of the individuals have typically Faliscan praenomina.
328
LANGUAGE CONTACT
Calitenes, cesies, and ceises are ‘Etruscoid’ nominatives. The gentilicium calitenes may
be derived from an Etruscan toponymic adjective *Calite, but the (Etruscan?) toponym
*Cali/Cale from which this is derived may be local, if it is in fact the old name of mod-
ern Gallese (§7.8.1.35, §6.5.7). With the exception of poplia, all the praenomina are
Etruscan (§7.7.2), although aronto and aruto show adaptation to Faliscan morphology
(§4.1). The first inscription shows a man’s name with FILIATION added, which in the
marital formula HUSBANDGEN UXOR is unique (§7.4.2). The use of the nominative lartio
instead of the genitive may be due to these unique circumstances, or it may reflect a
confusion over the cases (§9.2.3.4). The writer shows a lack of knowledge both of the
correct use of the Faliscan formulas and of the cases, implying that Faliscan may not
have been his or her first language. For the gentilicium Tettena, see below under (f).
(c) from tomb 28 of the second necropolis of S. Antonio:
arute macena | morenez MF 269 (scratched).
larise  mar||cna  citiai MF 270 (scratched).
poplia | zuconia MF 271 (scratched)
Arute and larise are probably accusatives arute(m) and larise(m) used as a nominative
(§9.2.2.4). Macena and mar||cna both represent Marcena: the omission of syllable-final
r in macena also occurs in several Faliscan inscriptions (§3.5.7b); the form mar||cna
may be due to syncopation (§3.6.6) or a graphic shortening. Morenez is an Etruscoid
form in -es: the -z may indicate the ‘strong’ realization of /s#/ in Etruscan (§9.2.2.1).
Citiai is in my view a dative or a genitive (§8.10.2): both Cristofani (1988:18) and
Pisani (1990:280) regard it as a Faliscan nominative citia recharacterized with the
Etruscan ending -i, however. Zuconia is an adaptation of the Etruscan gentilicium Zuχu
aacording to the usual pattern -u → -onius (§7.8.2.4). The Etruscan gentilicium also
occurs at Corchiano, in larisa zuχus Etr XXXII (late sixth century): the Faliscan form is
found at Civita Castellana in uel zu[con]|eo MF 56.
(d) found on the site of the Rio Fratta necropolis:
cauio  nomes|ina  maxomo | zeruatronia MF 272 (scratched)
Nomesina is Etruscan and has a non-rhotacized s (§3.5.3), but the o is a (graphical or
phonological) adaptation to Faliscan. The praenomen cauio is the most frequently used
praenomen in the Faliscan inscriptions (§7.7.2.1): the cognomen, too, is Faliscan, and
recurs in several Faliscan inscriptions (§7.9.1). Zeruatronia is also Etruscan, but is
adapted according to the usual pattern -u → -onius (§7.8.2.4) This name shows the use
of z- that may be due to Etruscan influence (§3.5.3).
(e) from Corchiano, details unknown:
cauio  oufilio | uolteo MF 275 (scratched)
ceisio  oufilio | uolθeo MF 276 (scratched)
329
CHAPTER 9
These inscriptions are the most ‘Faliscan’ of the sepulchral inscriptions from
Corchiano. The only Etruscan feature is the use of θ in MF 276, for which see §3.5.3.
A similar group can be found in the inscriptions from Civita Castellana. This group,
too, has been the subject of a publication by Peruzzi (1964b).
(f) from a tomb in the Valsiarosa necropolis:
fas[i]es  c[ai]sia || louci  teti  uxor and loụṛia || [l]oifirtạ MF 41
cauia  satelie | caui  felic ̣inate | uxor MF 42
In MF 41 the woman has a gentilicium fas[i]es with an Etruscoid ending (§9.2.2.2c):
the gentilicium precedes the praenomen, an order that is unique for the Faliscan ono-
masticon. The names of the husband are Italic, unless Tettius is an adaptation of Tettena,
which occurs in the group from Corchiano (above). In the second inscription, too, the
woman has a gentilicium with an Etruscoid ending, satelie. The gentilicium of her
husband is of Etruscan origin, felic ̣inate = Feliginatis, derived from a toponym
*Feligin- that may well be identical with Umbrian Fulginium/Fulginiae, modern Fo-
ligno (§7.8.1.59, §6.5.9), Both have the praenomen Gauia/Gauius, which is the most
frequent praenomen in the Faliscan inscriptions (§7.7.2.1).
The third inscription from this tomb is very damaged. If the two tiles are to be
read as belonging together, it may read ueṇe[---]na  | ux[or ---] MF 43, which could
show an attestation of an Etruscan praenomen Venel, although the ux[ seems to point
to the sepulchral inscription of a woman buried without her husband (§7.4.2.).
Taken together, these inscriptions show a number of Etruscan features, some even
unique, probably pointing to writers whose first language was Etruscan and whose
command of Faliscan was not perfect, as is shown by uncertainties about the use of
cases and of formulas. Whether they were in fact (recent) immigrants is in my view
unclear: although the gentilicia of the people appearing in the texts are mostly Etruscan,
and the proportion of Etruscan praenomina is unusually high, especially among the
group from Corchiano (cf. §7.10.5), the inscriptions also contain Faliscan praenomina,
and the general tendency to adapt Etruscan gentilicia to Faliscan morphology.172
9.3. Faliscan and the Sabellic languages
9.3.1. The nature of Faliscan-Sabellic contact. Contact between the agri Faliscus
and Capenas on the west bank of the Tiber and the Sabellic-speaking areas on the east
bank may well have been regular, and certainly involved contact connected with trade
172
Peruzzi (1964b-c, 1990) repeatedly points to the unadapted gentilicia in -na, but although
these could be adapted, they very often were not: see §7.8.2.2.
330
LANGUAGE CONTACT
and transhumance pasture, as can be established by archaeological evidence (§2.1.3).

The shrine of Lucus Feroniae in the ager Capenas may have drawn worshippers from
Sabellic-speaking areas (§9.3.2), although Feronia seems to have been a mainly Latin
deity. The story of the Hirpi Sorani (§2.3.4) may even reflect small-scale migrations
from these Sabellic-speaking areas: I cannot see any evidence of large-scale migra-
tions, certainly not during the historic period (§2.4.2, §2.8.2). Neither are there any
indications of conflict between the ager Faliscus and Sabellic-speaking areas on the
east bank or signs of mutual cooperation in the wars against Rome. Although Sabel-
lic-speaking peoples figure prominently in the Roman and Greek historians’ accounts
of Roman history, they play no role in the history of South Etruria. As said in §2.1.1,
the Tiber constituted not only a geographical boundary, but also a boundary between
the political networks on either bank.
Yet the Faliscan and Capenate inscriptions show gentilicia of Sabellic origin,
either recognizable by their derivation or assumed to be Sabellic because they have
parallels only in Sabellic texts. Examples are Aiedius, Alliuaeus, Battius, Blaesius,
Clanidius, Didius, Fertorius, Hirpius, Marhius, Neronius, Pacius, Pescennius,
Petronius, Plarius, Pumponius (?), Sedius/Saedius/Saidius, and perhaps also Tettius
and Vinucius (?): cf. §7.8.1-2. Other gentilicia refer to geographical features that be-
long to the east bank of the Tiber (cf. §7.8.2.5 and §6.5): Feliginas (probably derived
from the Umbrian town Fulginium/Fulginiae), Umbricius and Umbricianus,
Fa(r)farn... (perhaps connected with the name of the Sabine-Latin Farfarus/Fabaris),
and Narionius (possibly connected with the name of the Sabine Nar). Taken together,
these onomastic data could indicate that the Faliscan-Capenate area contained a num-
ber of families that originated in the Sabellic-speaking areas.
Sabellic epigraphic and linguistic features in inscriptions from the agri Faliscus
and Capenas (§9.3.3) are found predominantly in the ager Capenas. This may be due
to any of the following: (a) travel between the Sabellic-speaking area and the agri
Faliscus and Capenas may have used the Tiber-crossing near Lucus Feroniae rather
than the one near Grotta Porciosa; (b) trade routes may originally have followed the
old Via Tiberina, which may have run east of Mount Soracte rather than entering the
ager Faliscus proper (cf. G. Jones 1962:201), a situation that changed only with the
construction of the Via Flaminia in 220; (c) the network of ancient transhumance
routes (tratturi) may have passed through the ager Capenas and the area south of the
Lago di Bracciano rather than through the ager Faliscus with its steep gorges.
9.3.2. Sabellic features in Faliscan and Capenate inscriptions. In a previous publi-

cation, I summarized the Sabellic features in the Capenate inscriptions in one footnote
(Bakkum 1996:4 n.4), and although the discussion here can be allowed some more
space, it is still true that the Faliscan and Capenate texts in fact contain little that can
be ascribed to the Sabellic languages. I concentrate on the linguistic features: the epi-
331
CHAPTER 9
graphical features have been discussed extensively by Briquel (1972:813-45) and

Marinetti (1982b), both of whom have shown that the writing especially in the ager
Capenas shows a number of features that can only be ascribed to Sabellic background.
Two inscriptions in my corpus can be regarded as entirely Sabellic, showing Sabellic
features in the shape and type of the letters used (cf. Rix 1992a, Briquel 1972:831-3),
as well as in their onomastic and linguistic features. The first is setoms  míom | face
Sab 480†, showing the Sabellic development of the cluster /pt/ in setoms (*/sehtomo-/
← Proto-Italic */septomo-/, cf. Meiser 1986:93), in the Endsilbensynkope in setoms,
and in the forms míom (cf. §4.7.2) and face (cf. §5.2.1.6): not surprisingly, it is
classed as paläoumbrisch in ST (Um 4). The second is pa‹qu›is blaisiís Sab 468*,
with the Sabellic names pa‹qu›is (§7.7.1.47) and blaisiís (§7.8.1.29) and Endsil-
benkürzung in pa‹qu›is and blaisiís. Strangely, this inscription has not been included
in ST. However, it is far from certain that these two inscriptions reflect Faliscan-
Sabellic language contact: the first is from the La Tolfa area and was connected with
Faliscan because it contains the ‘arrow-f’, , which may not have been exclusively
Faliscan (§11.2.2.2), while the latter, ascribed by Buonamici (1928:605-6) to South
Etruria, is originis incertae or ignotae. However, both inscriptions do illustrate the
early Sabellic linguistic presence in the larger area of South Etruria.
Other inscriptions, whose provenance is more certain, are not wholly Sabellic, and
may reflect Faliscan-Sabellic language contact. Epigraphically, this is reflected e.g. by
the use of í in atnuíplau... Cap 423 (Briquel 1972:833 n.2) and in a  írpios  esú Cap
389 (Briquel 1972:833-7). In the former inscription, it is apparently used as the sec-
ond element of a diphthong, in the latter, it represents (Sabellic?) /ẹ/. The ú appears in
a  írpios  esú Cap 389, in k  sares  esú Cap 404, and in ac ̣iuaiom or aḷiuaiom esú
Cap 465, where it represents (Sabellic?) /ọ/ (cf. Briquel 1972:833-7). Of the names in
these inscriptions, írpios is clearly Sabellic (§7.8.1.74), while sares has parallels in
Latin and in Vestinian (§7.8.1.137): aḷiuaiom may be derived from the Sabellic gen-
tilicium Allis (§7.8.1.11).
Linguistic features that can be linked with Sabellic languages are few. Faliscan
had some phonological developments that have parallels in the Sabellic languages,
such as the word-internal development of the voiced aspirates (§3.3.3) and mo-
nophthongization of the diphthongs (§3.7), but these can hardly be ascribed as due to
contacts with speakers of Sabellic languages: note that the monophthongization of the
diphthongs in Umbrian and Volscian may not have been contemporary with the de-
velopments in Faliscan.
Morphophonologically, there are a few forms that can conceivably be described
as Sabellic. A first-declension nominative with the Sabellic rounding of /-ā/ to /-ō/
(§4.2.1) may be attested in posticnu MLF/Cap 474*, which has a parallel only in
South Picene postiknam CH.2, and possibly also in sta sediu Cap 466, if this is a
332
LANGUAGE CONTACT
woman’s name corresponding to Latin Sedia/Saedia/Saidia: note that the name Sae-
dius/Saidius has a parallel in Oscan saidiieis Cp 9. Similarly, k  sares  esú Cap 404
may contain a second-declension nominative singular in -es, reflecting an ending
*/-os/ → /-is/ with the Sabellic Endsilbensynkope. It could also be an Etruscoid
nominative in -es, however (cf. §9.2.2). The same interpretation can be given to
aiedies in k  pa  aiedies  Cap 390. The Endsilbensynkope has also been read in
pạrtis MF 79, but this may be a case of incidental syncopation, which occurs also in
Latin (uibis pilipus CIL I2.552, mirquris CIL I2.563, caecilis CIL I2.1028).
Morphologically, there is the form esú that occurs in a  írpios  esú Cap 389, k 
sares  esú Cap 404, and ac ̣iuaiom (or aḷiuaiom) esú Cap 465. This form corresponds
to South Picene esom TE.4 and Praesamnitic esum Ps 4, 5, Hernician esu He 3, and
Umbrian esu Um 18, but it also occurs in pari med esom kom meois sokiois in the
Garigliano inscription and in morai eṣo[m] from the ager Signinus. In §5.2.1.5 it is
assumed that it may have existed in Latin as well as in the Sabellic languages, and that
its occurrence in the ager Capenas can therefore be ascribed to either language: on the
other hand, the Capenate inscriptions where this form occurs show other Sabellic fea-
tures as well. If the form is purely Sabellic, it is an important indication for the pres-
ence of speakers of Sabellic, as the form for ‘I am’ may be regarded as standing on a
far deeper level of the language than the other features discussed here, implying inter-
ference and borrowing also on other levels (cf. §1.3.2.2).
There may, however, be another reason for the occurrence of esú(m) in these
inscriptions: in all three instances, this form occurs in iscrizioni parlanti-formulas:
ac ̣iuaiom (or ac ̣iuaiom) esú Cap 465 in OWNERGEN esúm, a írpios esú Cap 389 in
OWNERNOM esúm, and k sares esú Cap 404 either in OWNERGEN esúm or in OWNERNOM
esúm, depending on the interpretation of sares. The type OWNERGEN esúm is probably of
Sabellic origin (Agostiniani 1982:261-2): the type OWNERNOM esúm I ascribed to the
replacement of the formulas by this type by a new type of Latin sum-formulas (§8.8.2).
The verb form esú(m) may therefore be present in these Capenate inscriptions because it
was (at least in the mind of the individuals who wrote these inscriptions) the form that
was the proper one to use in this formula.
Syntactically, there are very few data apart from these formulaic texts with
esú(m). The Faliscan sepulcral formula hec cupat/cupant has some indirect parallels
in the Sabellic inscriptions (§8.10.1), but can certainly not be regarded as due to lin-
guistic influence from Sabellic-speaking areas: if anything, it is due to the Etruscan
formula θui cesu, but even that is uncertain. In the area of filiation formulas there is k 
pa  aiedies  Cap 390, which may reflect the Umbrian-Volscian filiation formula
where the father’s name is placed after the praenomen instead of after the gentilicium,
‘K. Aiedius son of Pa.’ (cf. §7.5). Note that both the gentilicium aiedies and the
praenomen pa = Pacius are Sabellic.
333
CHAPTER 9
On the lexical level, the words posticnu in MLF/Cap 474*, allegedly from
Falerii Novi, and p ̣esco(m) in Torelli’s (1974:741-6) reading of Cap 431 from Lucus
Feroniae (see below for the complete texts) can be regarded as forms that are probably
due to interference from a Sabellic language, for these words have parallels only in
South Picene postiknam CH.2 and in Marsian pesco VM 5 respectively and show Sa-
bellic phonological features (*/-ā/ rounded to -u in posticnu, pesco(m) ← */perḱ-sk-o-
m/?). If the interpretations posticnu ‘statue’ and pesco(m) = ‘votive offering’ are cor-
rect, the words denote the inscribed objects themselves, and are the central ‘theme’ of
their respective inscriptions; also, in both cases the inscription may be thought of as
having been phrased carefully.
I would say that, although Faliscan-Sabellic language contact may well have been
frequent, the linguistic data point to a discernible Sabellic presence but to little actual
‘influence’. A number of families in the area may have been of Sabellic descent, but
Sabellic languages do not appear to have entered the area with any great effect. This
may imply that speakers of Sabellic languages who settled in the area did so as indi-
viduals or in small groups, not in large-scale migrations. The linguistic features that
can be ascribed to Sabellic languages are all from a few inscriptions: a  írpios  esú
Cap 389, k  sares  esú Cap 404, atnuíplau... Cap 423, ac ̣iuaiom (or aḷiuaiom) esú
Cap 465, sta sediu Cap 466, and these may reflect individuals of Sabellic origin rather
than an overall ‘Sabellic presence’. However, there are two exceptions to this picture:
(a) caui  tertinei  | posticnu MLF 474† (reputedly from Falerii Novi, and therefore
most likely dating from after 240): the inscription is written in the Latin alphabet, the
praenomen is the most frequently occurring Faliscan praenomen, the gentilicium has a
parallel in tertineo LF 213 but is not attested elsewhere, but the word posticnu is par-
allelled only by the much earlier South Picene postiknam CH.2.
(b) p ̣esco ṣal | plaria  t  l | feron  dono || [q]uod  a[fluc] | dedet  libes | m  mereto
Cap 431 in Torelli’s (1974:741-6) reading (from Lucus Feroniae, second half of the
third century): the inscription is Latin both in the epigraphic and in the linguistic fea-
tures, the gentilicium plaria may be Sabellic or Latin, but the word p ̣esco has a paral-
lel only in Marsian pesco VM 5.
9.4. Faliscan and Latin
9.4.1. The nature of Faliscan-Latin contact. The contact between Faliscan and Latin
must of course be divided into two periods: (1) the period before 240, when the ager
Faliscus was an independent geopolitical unit belonging to Etruria, and (2) the period
after 240, when the area came under Roman rule as a consequence of the disastrous
war of 241 (§2.6).
334
LANGUAGE CONTACT
Establishing the nature of the Faliscan-Latin contacts before 241 is difficult. The
Roman and Greek sources concentrate almost exclusively on the wars that Rome
fought with the Faliscans from the late fifth century onwards (§2.5). These wars must
of course have involved contacts in the form of envoys, truces, treaties, etc. (Livy
(5.27) even portrays Faliscan envoys as speaking before the Roman senate), but will
not have been influential in bringing Roman Latin to the area: on the contrary, they
may have strengthened Faliscan ethnic identity and perhaps even have incited ‘anti-
Roman’ or ‘anti-Latin’ feelings, especially as during this period the Faliscans always
sided with their Etruscan allies. Other forms of contact must have existed, however. It
can be assumed that there was contact in the form of trade up and down the course of
the Lower Tiber along the ancient Via Tiberina, as well as along the precursor of the
Via Amerina (§2.1.3); also, from the early fourth century onward, there were Roman
colonies at nearby Sutrium, Nepete, and Capena (§2.5.2). Several inscriptions in the
Latin alphabet have been found at Civita Castellana and Corchiano (see §9.4.2), towns
that were wholly or partly abandoned after the war of 241-240, and these inscriptions
may therefore well belong to the period before 240. Interesting in this respect is the
fourth-century strigilis with med  loucilios  feced Lat 268 from Corchiano, which is
clearly Latin, but contains a unique word order that must be due to the Etruscan model
mini zinace MAKERNOM (cf. §8.9.2), perhaps implying that it was made in a largely
Etruscan-speaking environment, even though the name of the craftsman is Latin.
Traffic from Rome and Latium may have passed through the ager Faliscus and
thence to the Sabine and Umbrian interior long before the Romans gained control of
the Tiber-crossing near the Grotta Porciosa site after the war of 241. The Roman
adoption of the carmina Fescennina may also have belonged to this period, especially
if Fescennium is identified with Narce, as Narce ceased to exist shortly after the war
of 241 (§2.1.2). Other allegedly Faliscan features in Roman society, such as the ius
fetiale and the additions to the lex XII Tabularum (§2.3.3, §2.3.1), if not fictitious,
may have been adopted during the same period: even if this attribution is entirely fic-
titious, it reflects at least a memory of contacts between Rome and the ager Faliscus.
Contact with Latium and Rome, although hard to pin down, may well have been
quite frequent. This is implied by the fact that during the fifth, fourth and early third
centuries several important morphological changes took place that affected both Latin
and Faliscan: the replacement of -osio by -i (§4.4.9) and probably also the replace-
ment of -as by -ai (§4.2.2), and the replacement of the old third singular perfect end-
ing -ed by a new ending -et (§5.2.4e). The spread of such replacements, whether
originating from Latium, Rome, or the ager Faliscus, cannot have been accomplished
without fairly frequent contacts. It may even have been the case that parts of South
Etruria were originally Latin-speaking, and by and large remained so even though the
written language of the area was Etruscan (cf. §2.4.2) during the earlier centuries of
this period.
335
CHAPTER 9
The contact during the period after 240 is a different question. I have already pointed
at the changes that affected the area in §2.6.2, whereby the area was ‘ruralized’ and its
main site replaced by a new Roman settlement. It can be assumed that there was a
substantial Roman or Latin presence in the area from shortly after this war onward. If,
as the sources imply, a large part of the male indigenous Faliscan population had per-
ished in the war (§2.6.1), and were in effect replaced by the speakers of Latin that
now settled in the area, not only in the new Roman Falerii, but also on the lands that
had been ceded to the Romans, this must have changed the proportion of speakers of
Faliscan vs. speakers of Latin substantially. If I assume a population for the area of
perhaps 20,000-22,500 people at best, of which several thousands (hardly the 15,000
named in the sources, but, say, 5,000) die in the war, and an influx of perhaps 6,000
immigrants (for these numbers see §2.7e-f with notes 30-31), this means that over a
quarter of the population of the area was in effect replaced by speakers of Latin.
Even more importantly, the administration of the area and especially of its main
site now came in Roman hands. In how far this meant that Roman Latin became the
‘standard language’ is unclear: although it will have become the language of the ad-
ministration of the Roman parts of the area, the Romans, especially at this period,
nowhere seem to have exercised a ‘language policy’ of forcing people to adopt Latin
(Bakkum 1985). The inscriptions on public works from Falerii Novi offer a rather
confused picture: one, LF 213, is in the Faliscan alphabet and shows Faliscan features,
but another, LtF 215, which may likewise have been a public inscription, is in the
Latin alphabet. The second-century public dedications Lat 216-217 are Latin. In
LF/Lat 214 the alphabet is Faliscan, but the language can be Latin as well as Faliscan.
The ruling class of Falerii Novi may have consisted of local families who could
be trusted not to work against the interests of Rome (§2.6.2), and the inscriptions
show that the members of the new upper class assumed e.g. the Roman use of the cur-
sus honorum in their sepulchral inscriptions (§9.4.2), dropped the use of the patro-
nymic gentilicium in favour of the alternative FATHERGEN [SON/DAUGHTER] which was
the formula used in Latium (§7.5), and started to give their children Latin praenomina
that soon replaced the old Faliscan ones (§7.10.5).
All in all, the contacts with (Roman) Latin during the period after 240 lead to a
quite speedy disappearance of Faliscan as a written language. The reasons for this
have been discussed in §2.8.3, and more will be said on the subject in the next section.
9.4.2. Latin features in Faliscan inscriptions. For the purposes of this study I have
divided the inscriptions in the Latin alphabet from the area into three groups: Latino-
Faliscan, Capenate, and Latin, the first two defined as still containing dialect features
that are in accordance with the linguistic features of the Faliscan inscriptions, the last
term used to denote the inscriptions that show no such features (§11.1.3). The only
author to treat these inscriptions together is Safarewicz (1955:184-90).
336
LANGUAGE CONTACT
Inscriptions in the Latin alphabet have been found throughout the area. From
Civita Castellana are LtF 63, 140, 170-174, 205, and perhaps also MF/LtF 21, and
from Corchiano and surroundings LtF 277-278, 288, 290, 292, 294, 299-301. Of
these, LtF 140, 170-174, 288, and 299-301 are sepulchral inscriptions and therefore
not likely to have been written anywhere else. Some may even predate the war of 241,
as Civita Castellana and Corchiano were abandoned not long after that date (§2.6.2).
From (near) S. Maria di Falleri (Falerii Novi) are MLF/LtF 241, LtF 215, 231-233
(sepulchral), 239, and MF? 254, and from the site at Grotta Porciosa, which survived
into the Roman period, LtF 340-345. Other inscriptions in the Latin alphabet are LtF
325-327 from Carbognano-Vallerano and LtF 328 from Fabbrica di Roma, settle-
ments that may have been located in the part of the ager Faliscus that was not ceded to
the Romans. Inscriptions in the Latin alphabet are therefore found both in the Roman-
administrated part and in the part that was still nominally independent.
Apart from the alphabet, the inscriptions show several other features that do not
occur in the contemporary Faliscan inscriptions. On the phonological level, the diph-
thongs usually show non-monophthongized spelling: beside the already mentioned
loucilios Lat 268 there is fourios Lat 216, claudia Lat 393 (but polae Lat 251), heic
LtF 231, eidus Lat 393 (beside the late idem Lat 456 and the obviously historically
motivated spellings loidos Lat 217, coiraueront Lat 218, coer Lat 456): the old diph-
thong /a/ is spelled as e in pretod LF/Lat 214 and leuia LtF 327, however.
A morphophonological feature is the form of the ending of second-declension
nominative singular. Middle and Late Faliscan inscriptions almost universally show
this ending as -o (§4.3.1, §3.5.7d) and the Latino-Faliscan and Capenate inscriptions
likewise have -o in [---]ilio LtF 215, cḷ[i]peario LtF 233,*(*)coṇẹo LtF 290, ṃunio
LtF 377, and genucilio Cap 434 (§4.3.1). The oldest Latin inscriptions from the area
have -os in loucilios Lat 268 and fourios Lat 216, but from the late third century on-
wards, they generally have -us: spurilius Lat 237, spurilius Lat 238, calpurnius Cap
432, [---]rcius Lat 436, egnatius Lat 291, latrius Lat 218, u]mpricius and [?]aburcus
Lat 219, fuluius Lat 250 (106), didius and uettius Lat 456 (c.100-50). Lat 251 has
lectu(s) (twice). The change of /o/ to /u/ in closed final syllables (§3.6.6) is attested
for the area only in the inscriptions in the Latin alphabet 173 and in [fel]ịcinatiu LF
384, but not in the Late Faliscan inscriptions. With regard to the writing or omission
of -s, I have assumed in §3.5.7d that this may represent different orthographical con-
ventions rather than different realizations of /s#/: in Faliscan as well as in third- and
second-century Latin, /s#/ was ‘weakly ’realized as [h] or [ʔ], but whereas in Latin the
rule seems to have been to write the -s, its omission in Faliscan may have been due to
a convention not to write it (§3.5.7d).
173
Cf. also donom Cap 431 and esú Cap 389, 404, 464, vs. sacru(m) and cuncaptum LF/Lat
214, gonlegium, aciptum, and opiparum Lat 217, and donum Lat 218.
337
CHAPTER 9
In other respects, such as morphology, lexicon, and syntax, there are very few
points of comparison between the inscriptions that can be thought of as representing
Latin and the Faliscan inscriptions. This is due to the fact that the Latin presence in
the area brought with it whole new ways of phrasing specific types of inscriptions.
Thus, public dedications on bronze now appear, such Lat 217, Lat 218, and the curious
Late Faliscan or Latin text menerua  sacru | la  cotena  la  f  pretod  de | zenatuo 
sententiad  uootum | dedet  cuando  datu  rected | cuncaptum LF/Lat 214. The same
change can be observed in the roadside inscriptions that (presumably) mention magis-
trates responsible for their construction or maintenance: the Faliscan type consisted
exclusively of names (MLF 207, 210, Etr XXXVIII, XXXIX), and this is reflected in
the Latino-Faliscan c**(*)coṇẹo  l***(*)  ce  paui[ceo  ru?]so LtF 290. A more
Latin type appears in c  egnatius  s[ex  ]f prata | faciunda  coirauit Lat 291.
This same change in repertoire shows up in the Faliscan sepulchral inscriptions,
that now start to contain cursus honorum in LF 242, 243, 245, 247, 248, 249, LtF 231,
232, (and perhaps LtF 233), Lat 237, 238, 239 (and Lat 219?) which is a feature that
properly belongs to the roadside tombs and monuments of Latium, but quite inappro-
priate for the sealed-off Faliscan family tombs (§11.1.4.1). These inscriptions show
words like quaestor, praetor, censor, and formulas like magistratum gero that have
parallels in Latin. As was said in §6.3.6, the lexical subset pertaining to public office
and the like is entirely Latin. This is of course not because Faliscan did not have
words for magistracies or formulas for official expressions, but because the words and
formulas involved occur only in inscriptions that show Latin influence in the way in
which they are phrased: the extant Faliscan inscriptions are simply not of the type
where such words and formulas were used. It has been assumed that Latin influence in
the ‘official’ vocabulary is also visible in the word efiles in MF 113-117, which may
have been a calque on Latin aediles (G. Giacomelli 1963:243-4): as argued in §6.2.1, I
do not think that it is necessary to assume this.
I already pointed to the change in the formula of filiation, where the use of the
patronymic adjective is associated exclusively with Middle and Late Faliscan inscrip-
tions, while Latino-Faliscan, Capenate, and Latin inscriptions from the area always
have the formula FATHERGEN [SON/DAUGHTER] (§7.5 with figs.7.3-4). Although the pat-
ronymic adjective was still used in Late Faliscan inscriptions (LF 220, 222, 332, 336),
it clearly belonged to the texts written according to the Faliscan rule-set, and disap-
peared when these ceased to be produced.
The changes in the texts during the period after 240 therefore concern not so much
individual features of Faliscan, but the entire way which texts were expressed. The
coming of Latin to the ager Faliscus brought with it a different set of rules for the pro-
duction of written texts, and the rule-set of written Faliscan was not so much influ-
enced as replaced by this new set. This was probably made easier by the fact that Fal-
338
LANGUAGE CONTACT
iscan and Latin were very much alike, so that the rules for written Latin could be ap-
plied to Faliscan without too much difficulty. All in all, this sea-change appears to
have been quite quick, which might be due to the replacement of part of the popula-
tion by speakers of Latin as described above. The Late Faliscan inscriptions are far
fewer in number than the Middle Faliscan inscriptions, and there appears to be no
material that can be dated to after c.150: I would not be surprised if it could be shown
that Faliscan inscriptions disappeared a generation earlier, between 200 and 175.
This, of course, does not mean that Faliscan itself disappeared around that date:
however, the written form of the language that is spoken in the area is the one associ-
ated with Roman Latin. This comes close to the idea of formal = Roman vs. informal
= Faliscan advocated by R. Giacomelli (2006 passim): although I contest that this
distinction existed before 241, it may be quite applicable to the Late Faliscan period.
9.5. Summary of §§9.2-4
All in all, the Faliscan material shows some very interesting points with regard to lan-
guage contact and bilingualism, even though the material is limited. As might be ex-
pected, there is a high degree of onomastic borrowing (cf. chapter 7 passim), espe-
cially from Etruscan. On other levels, interference and borrowing is rather limited, as
might be expected on the basis of what has been said in §9.1.
Etruscan interference shows up most clearly in the ‘Etruscoid’ forms: these are
part of a larger set of interference phenomena where morphemes of the nominative
appear due to onomastic borrowing. There appears to be no interference of borrowing
on deeper levels of the language, and this is not surprising in view of the very differ-
ent grammatical structures of Etruscan and Faliscan. A very interesting feature, which
allows us to draw some conclusions about the degree of bilingualism of some of the
writers, is the uncertainty with regard to the use of cases. This can only be ascribed to
the different use of the cases in Etruscan. The use, in Faliscan texts, of formulas that
are modelled on Etruscan shows that these formulas were not just borrowed, but also
adapted to the structure of Faliscan: this is a case of ‘true’ borrowing by speakers of
Faliscan, not interference by speakers of Etruscan.
Sabellic interference is observable, but limited, not only in a numerical, but also
in a geographical sense, as it shows up mainly in inscriptions from the ager Capenas.
The influence from Latin in the period before 240 is not so much absent as very
hard to detect. In the period after 240, there is a quite rapid change in the written lan-
guage from Faliscan to Latin, indicating that new rules for writing had entered the
area together with the probably quite substantial influx of speakers of Latin. This led
to a rapid disappearance of Faliscan as a written language: if it was still spoken after
150, it was no longer written and therefore is not accessible to us.
339
CHAPTER 9
Resuming what was said in §9.1, the epigraphic material from the ager Faliscus and
Capenas thus shows the co-existence of several sets of rules for the writing of the lan-
guages occurring in the area: a set for writing Faliscan, a set for writing Etruscan, a
set for writing Latin, and perhaps also a set for Sabellic. On the whole, there is re-
markably little cross-over between these rule-sets. Writing in language x entails a dif-
ferent set of rules and conventions from writing in language y, and the individuals that
wrote the inscriptions were apparently well aware of this: perhaps because writing
was probably not an everyday process for everybody, and therefore required a more
conscious effort on the part of the writer, or because the texts that we have are often
very formulaic in nature, obeying even more specific rules than a text that is freely
composed.
The rule-sets of Etruscan and Faliscan were so different that there is hardly any
spread of features apart from onomastic features. In the case of Sabellic features in
Faliscan inscriptions, it is clear that most inscriptions that show Sabellic influence are
Sabellic in more ways than one, and reflect writers that probably had a Sabellic lan-
guage as their first language. The rule-set of Latin on the other hand quickly replaced
the rule-set of Faliscan, and could do so not only because of sociolinguistic factors as
discussed in §2.7, but also because of the close resemblance between the two.
9.5. A note on Faliscan outside the ager Faliscus
In view of the extensive language contacts with Etruscan, Latin, and the Sabellic lan-
guages, it might be expected that, just as their presence may be observed in inscrip-
tions from the ager Faliscus and Capenas, so Faliscan presence might be observed in
inscriptions from areas other than the ager Faliscus and Capenas. There are several
inscriptions where this has been assumed: in the first place, eqo kaṇaios 482†, titoio
483† and neuen deiuo 484†, all from Ardea: this is discussed in §18.3.2. I likewise do
not adopt Lucchesi’s (2005) suggestion that the Satricum-inscription, CIL I2.2832a, is
Faliscan: this is discussed in §18.3. On the other hand, Poccetti (forthcoming) con-
vincingly shows Faliscan influence in inscriptions from the area of Magliano Sabino,
on the east bank of the Tiber opposite the ager Faliscua and Capenas.
340
Chapter 10
Conclusion: Faliscan as a Latin dialect
10.1. Language or dialect?
10.1.1. Drawing conclusions. As said in §1.1, it was the aim of this study to show
that Faliscan is a dialect of Latin, not an independent language closely related to
Latin. This, of course, depends on the definition of dialect that is used, and in §1.2 I
therefore described three possible approaches to this subject: the sociolinguistic
approach, where only extralinguistic factors are taken into account, and the structural
approach, where only intralinguistic factors are taken into account, the latter
subdivided in a synchronic approach, which is the one best applicable to living
languages and dialects, and the diachronic approach, which is the one better
applicable to fragmentarily preserved dead languages and dialects whose general
position within a well-studied language-family is known. In the following sections, I
shall look at each approach in succession and assess the linguistic position of Faliscan
according to definitions of each of these approaches.
As has repeatedly been stated, the major difficulty in assessing the Faliscan
material is the fragmentary state of the material, not only the Faliscan material itself,
but also the Latin and Sabellic material required for a valid comparison. This applies a
fortiori to the Early Faliscan period. Yet in spite of this fragmentary state, I hope to
have shown that a remarkable amount of linguistic data can in fact be derived from
this material, allowing an assessment to be made both from a synchronic and a
diachronic perspective. Other sources, mainly historical, make it possible to draw the
outlines of a very general sociolinguistic assessment. It is therefore possible, in my
view, to draw valid conclusions on Faliscan and the people that spoke it on the basis
of the epigraphic, historical, and archaeological material that we now possess.
10.1.2. The sociolinguistic definition. According to the exclusively sociolinguistic

definition, Faliscan can most definitely be described as a language in its own right,
and not as a dialect. As I have shown in §§2.2-3, the ager Faliscus can be assumed to
have had an identity of its own, recognized as such probably both by the inhabitants
of the area themselves and by the peoples with which they came into contact, as is
shown for instance by the existence of an ethnonym (§2.2.2). This identity manifested
itself most clearly in the independence of the area, both from Rome and from the
Etruscan cities (§2.4.2), but also in a number of local customs and peculiarities (§2.3),
341
CHAPTER 10
some of which semi-linguistic, such as the use of its own alphabet (§11.2) and of its
own praenomina (§7.7.2), or linguistic, such as the use of its own formulas for
specific types of texts (§8.10.1). Each of these features may or may not have been
regarded as relevant to that identity by the inhabitants of the ager Faliscus and those
that came into contact with them. It is noteworthy that this Faliscan identity may not
have been connected with one specific ‘ethnic’ background: the onomasticon shows
that the inhabitants of the ager Faliscus have gentilicia that were Italic as well as
Etruscan in origin (§7.10.3), and the data on language contact of Faliscan with
Etruscan show that a number were probably bilinguals whose first language was not
Faliscan, but Etruscan (§9.2). The same may have been true for the ager Capenas and
the Sabellic languages (§9.3).
The linguistic aspect of this identity must have been expressed most clearly with
regard to the speakers of Etruscan: if the Faliscans were indeed members of the
Etruscan League at the beginning of the fourth century (as Livy presents them, see
§2.5.1), they were in all probability the only member of the League whose first or
main language was not Etruscan. In view of what can be glimpsed of the ‘Faliscan
identity’, I am reasonably sure that if it were possible to explain the modern concepts
of ‘dialect’ and ‘language’ to fourth- or third-century inhabitants of the ager Faliscus,
they would regard what we call Faliscan as a language, and that this conclusion would
be based not so much on an assessment of the structural correspondences and
differences between Faliscan and the languages of Latium or of Umbria, but on a gut
feeling that what they spoke was part of their identity as a Faliscan, especially in their
probably frequent contacts with speakers of Etruscan (§9.2.1), Sabellic languages
(§9.3.1), and Latin (§9.4.1).
Even the facetious wisecrack about ‘language being a dialect with an army’
(§1.2) is true to the extent that the independence, linguistic or otherwise, of the ager
Faliscus was maintained by a number of armed conflicts with Rome during the late
fifth, fourth, and third centuries (§§2.5-6), possibly preceded by similar conflicts with
Veii during the sixth and early fifth centuries (§2.4.2), and that the disappearance of
Faliscan at least as a written language was accelerated by the loss of its army and its
political independence in the war of 241 (§2.6). After this war, there are few signs of
an independent identity, and the area seems to have been Romanized in many
respects, including linguistically, in a relatively short time (§2.8.3, §9.4.2): assuming
that Falerii Novi was founded somewhere between 240 and 220, no more than two
generations appear to have passed before Faliscan disappeared as a written language
(possibly with a brief revival in LF/Lat 214). If it lingered on as a local patois, it was
not what within the sociolinguistic definition must have been a dialect, for during the
second century, Faliscan was no longer an independent variant in its own right, but a
variant that was becoming substandard beside a standard that came to be based on or
to be identical with Roman Latin (§1.2, §2.8.3).
342
FALISCAN AS A LATIN DIALECT
10.2.2. Synchronic comparison. Whatever the conclusions that can be drawn from
the sociolinguistic perspective, what interests me more is whether Faliscan is also a
dialect from what I have called the ‘structural’ perspective. According to the
synchronic approach, Faliscan can be compared on a number of points with the
surrounding languages. For the approach to be as synchronic as the fragmentary
material allows, however, this comparison should be made for one point in time.
A synchronic comparison for the Early Faliscan period is very difficult, since
there are only 10-12 Early Faliscan inscriptions (EF 1-4, 6-10, 467*, and EF/Etr 5 and
385) and the contemporary Latin and Sabellic epigraphic material is likewise scarce.
This has caused many problems in the assessment of the position of Faliscan, for it
has often forced scholars to compare Early Faliscan forms with much later Latin or
Sabellic ones. Three vexed forms in this respect are fifiked EF 9, f[.f]ịqod EF 1, and
ues EF 4. Fifiked and f[.f]ịqod show that Early Faliscan had a reduplicative perfect of
fingo, and this is contrasted with Latin finxi: but finxi is much later, and it is very well
possible to assume that Faliscan dropped the reduplicative perfect of fingo after the
Early Faliscan period, while on the other hand Latin finxi, which is an old aorist, not a
perfect, had replaced an earlier reduplicative perfect (cf. the replacement of
vhevhaked CIL I2.3 by feced CIL I2.4). Likewise, f[.f]ịqod shows an aorist ending
being used in the third plural perfect, while the reconstructed Latin perfect ending is
*/-ēri/, first attested as steterai CIL I2.2832a. Yet the material does not allow us to say
with any degree of certainty whether f[.f]ịqod represents the standard Early Faliscan
form in this respect, or was an occasionally recurring analogical creation beside a
regular perfect ending. The pronoun ues can only be compared to Latin uos, but
neither uos nor ues is attested for the early Latin inscriptions, and the same dichotomy
can be observed in the Sabellic languages, where Umbrian uestra TI VIb.61 shows the
e-vocalism of the Early Faliscan form, but Paelignian uus Pg 9 the o-vocalism of the
Latin form (§4.7.3). It will be clear that the Early Faliscan material is just too little
and too lacunary to allow a comparison with contemporary Latin and Sabellic
material: what, for instance, of douiad EF 1, which can be compared only to much
later Latin duam and Umbrian purtuvitu TI IIa.24, or of umom EF 2 which has a
parallel only in ‘Old Hernician’ udmom He 2, but shows an assimilation /dm/ → /mm/
that is parallelled in Latin?
If the position of Faliscan is to be judged by a synchronic comparison, this
comparison is best made on the basis of the material from the first half of the middle
of the third century, that is, Middle Faliscan, and only on the basis of this material,
without the admixture of Early Faliscan, tempting though it is to fill the gaps in the
data. The Middle Faliscan period provides enough material, and, as it precedes the
period after 240 during which Rome took over the ager Faliscus, may be assumed to
be relatively free from Latin influence. Such a comparison shows that Faliscan is on
most points identical with Latin, and differs greatly from the Sabellic languages.
343
CHAPTER 10
If we look at the Middle Faliscan material from the perspective of a synchronic

comparison both with Latin and with the Sabellic languages, there are different
conclusions on different levels.
From a morphological point of view, I can see no other conclusion than that
Faliscan is a Latin dialect, or, alternatively, that it split off only at a very recent date.
The nominal morphology is remarkably similar (§4.10). Points of difference are
perhaps
(1) the Faliscan first-declension genitive singular (§4.2.2), where I assume that the
ending was changing from -as to -ai at this time, just as it was in Latin, but the
material is ambiguous, as all forms in -ai that can be interpreted as genitives can also
be interpreted as datives. If Faliscan only had -as, this provides a parallel with the
Sabellic languages.
(2) the Faliscan second-declension dative singular, which was still -oi (§4.3.2): it is
unclear whether the Latin ending was already -o, or still -oi. The Sabellic ending was
certainly /-ō/ at the time.
(3) the ending of the third-declension genitive singular, which was -os (§4.5.2),
which, however, may have been the contemporary Latin form as well; it was most
certainly not the Sabellic form at the time, which was a reflex of /-es/ (e.g. -e(r) in
Umbrian, but still -eís in Oscan).
On every other point of nominal morphology where a difference can be established
between Latin and the Sabellic languages, Faliscan sides with Latin (§4.10).
Significantly, Faliscan shows a second-declension nominative plural ending -e in lete
MLF 285, while contemporary Sabellic languages show a reflex of */-ōs/ (§4.3.6),
and of course the ending of the genitive singular, /-ī/ (§4.4.3), where the Sabellic
languages had /-ēs/.
In the personal pronouns, Middle Faliscan shows both ego (eco LF 378, 383)
and met (MF 470*), the latter probably showing signs of a change to me at this time,
which are the same forms (and the same change) as in contemporary Latin (§4.7.1-2):
at least in the case of me the Sabellic languages may have differed, but the only form
to show this is much older.
In the Middle Faliscan verbal morphology, a significant point is the ending of
the third singular perfect -et (facet MF 470*, keset LF 242), as in Latin, in contrast
with the Sabellic languages, which have -ed (§5.2.4.5). A difference, however, is
Faliscan facet MF 471* vs. Latin fecit, where it should be noted that a comparison
with the Sabellic languages shows that Umbrian likewise had a perfect stem /fak-/: the
Latin form /fēk-/ is unique in the Italic languages (§5.3.6). Another interesting form is
esú(m) Cap 389, 404, 465: this form is attested with certainty for the Sabellic
languages, but may have existed in (dialectal) Latin as well, and may therefore have
occurred in Faliscan or Capenate (§5.3.5).
344
From a lexical point of view, too, Middle Faliscan shows no discernible

differences from Latin, and where there are differences between Latin and the Sabellic
languages, Faliscan sides with Latin (§6.2 passim, §6.3). It should be noted that due to
the available data, the lexicon is extremely limited: yet note, for instance, that
Faliscan has filios/filia, as does Latin, while the Sabellic languages show reflexes of
puclom/fu(h)tēr (§6.2.24-25).
A synchronic comparison of the syntax of Middle and Late Faliscan is very
difficult. The morphosyntactic data are too few, the data on word order too much
depending on purely formulaic texts, and the phrasing of the varying types of
inscription too different to allow valid conclusions to be drawn. In my view, no real
synchronic comparison can be made between Middle Faliscan and contemporary
Latin or Sabellic languages on the points of syntax.
Apart from the already mentioned facet MF 470*, the main differences between
Middle Faliscan and Latin that can be observed in a synchronic comparison are
phonological and phonetic. The phonemic system of Middle Faliscan differed from
that of Latin in one significant respect, namely that third century Faliscan had no short
diphthongs (and long diphthongs perhaps only as the case-endings for the dative
singular of the first and second declensions), while it had more degrees of openness in
the front vowels (§3.6.1, §3.7). In mid-third-century Latin, this may have been the
case in some local variants, but certainly not in all, and certainly not in the dialect of
Rome. In a synchronic comparison, this monophthongization and the regular
occurrence of word-internal /f/ are features that Faliscan shared with the Sabellic
languages, the former feature with Umbrian and Volscian especially, the latter with
the Sabellic languages in general. On the other hand, the Middle Faliscan phonemic
system certainly differed from the Sabellic languages in having a labiovelar occlusive
series, as did Latin (§3.4, §3.5.1). Beside this, there is a phonotactic difference: in
Faliscan, /f/ could regularly occur word-internally, as in the Sabellic languages, while
in Latin a word-internal /f/ must have been rare at least in Roman Latin, if perhaps not
in all Latin dialects (§3.3.3).
Differences that were phonetic rather than phonemic were the tendency to
realize word-initial /f/ before a vowel as [h], although it is very probable that this
tendency also occurred in Latin or at least in some local variants of Latin, as is shown
by several epigraphic instances from Praeneste and the forms quoted by the Latin
grammarians (§3.5.2), but not in the Sabellic languages. Another tendency is the one
to drop syllable-final nasals, a tendency that sometimes is observed in Latin
inscriptions, but is more frequent in the Faliscan inscriptions (§3.5.7a). The tendency
to omit word-final -s when it was preceded by a (short) vowel is not so much a
difference with Latin, but occurs in Faliscan with a far greater frequency, perhaps
implying that it may well have been regular in the sense that it was an orthographic
rule (§3.5.7c).
345
CHAPTER 10
Are these synchronic differences enough to regard Faliscan either as Latin or as

non-Latin? This is a difficult question to answer, as in the case of dead languages
verdicts of this kind are hardly ever purely based on synchronic comparison. I would
say that the phonemic differences, which are not just a difference in individual
phonemes but in the phonemic structure are certainly a point that divides Middle
Faliscan from contemporary Latin, but that on the other hand this must be set off
against the vastly greater number of correspondences between the two, not just on the
phonological level, but on other levels as well. In my view, these differences still fall
within the limits of variation that can be observed among the dialect variants of one
and the same language.
Interestingly, the synchronic differences between Middle Faliscan and
contemporary Latin are all points where Faliscan sides with the Sabellic languages,
both in the phonological differences and in the perfect stem /fak-/. Yet on the other
hand the number of differences between Middle Faliscan and the contemporary
Sabellic languages is so great, especially on the morphological and the lexical level,
that the gap between Middle Faliscan and the Sabellic languages is far greater than
that between Middle Faliscan and contemporary Latin.
Would a mid-third century inhabitant of the ager Faliscus, if asked, have
regarded Faliscan as substantially different from Latin? It seems very likely that
inhabitants of the ager Faliscus and Latium would have understood each other without
too much trouble. Such a mutual understanding, however, may be based more on
similarity of the lexicon, which in this case appears to have been very great, and on a
number of known phonological differences that could easily be ‘mentally transposed’
in a language contact situation: a Faliscan might well have known that a medial -f-
usually corresponded to a Latin -b- or -d-.
10.1.3. The diachronic perspective. The diachronic perspective affords us a look not
only at the Middle Faliscan forms, but also at the Early Faliscan forms and even at the
reconstructed forms and developments, for these can be used to fill in gaps in the
attested material and place it in a wider perspective. This means that we can look at a
larger amount of data, and can place these data against the larger background of the
developments as they are reconstructed for Proto-Italic, Proto-Latin, and Proto-
Sabellic. As said in §1.2, most scholars use a combination of the diachronic and the
synchronic perspective in any case.
In the diachronic approach, it is quite clear that Faliscan belonged to the Latin
branch of the Italic family of PIE. There are no indications of separate developments
at the Proto-Italic stage, as is to be expected if Faliscan is assumed to be an Italic
language, and during the post-Proto-Italic period, Faliscan shares most features of
preservation and innovation with Proto-Latin, not with Proto-Sabellic. The clearest
cases are:
346
common Proto-Latin preservations:

(1) phonological: preservation of the labiovelar occlusive series, which in the Sabellic
languages merged with the labial series (§3.4);
(2) morphological: the second-declension genitive singular ending */-oso/, preserved
in Proto-Latin (§4.4.3, §4.4.6), but replaced in Proto-Sabellic by /-es/ (§4.4.7);
(3) morphological: the consonant-stem genitive singular ending */-os/, which is in all
probility original, which in Sabellic was replaced by /-es/ (§4.5.2);
(4) morphological: preservation of the Proto-Italic secondary third plural ending
*/-nd/ (§5.2.4.2), which in Proto-Sabellic was reformed to */-ns/;
common Proto-Latin innovations:
(5) morphological: transfer of the pronominal nominative plural endings */-a/ and
*/-o/ to the first and second nominal declensions (§4.2.6, §4.3.6), while in Proto-
Sabellic the nominal endings were extended to the pronominal declensions;
(6) lexical: the innovations */ɸīlos/ and */ɸīlā/, while Proto-Sabellic preserved
*/puklom/ and */ɸuγ(i)tēr/, reflecting PIE */putlom/ and */dhughitēr/ (§6.2.24-25);
either common preservation or common innovation:
(5) morphological: the accusative */mēd/ of the personal pronoun of the first person
singular, where Proto-Sabellic had */mēom/: depending on which form is
reconstructed for Proto-Italic, this is either a case of common preservation or of
common innovation (§4.7.2).
The only clear instance of a significant and early feature that separates Faliscan from
Latin must also be dated to the period preceding the earliest inscription:
separate Latin and Faliscan innovations:
(6) phonological: the development of the word-internal reflexes of the PIE voiced
aspirates, where the Faliscan reflexes of */bh dh gh/ are /f/, as in the Sabellic
languages, while Latin has /b d / (§3.3.3). The reflex of */gh/ is in my view unclear:
if it is taken to be /g/, this reflects a uniquely Faliscan development: however, it shows
more similarity with the Latin occlusive reflexes than with the Sabellic spirant reflex.
It is in this phonological development that Faliscan shows the clearest early difference
from Latin. This difference cannot be recent and apparently did not have any parallel
developments in Latin. (As said in §3.3.3, the evidence for a similar development in
‘rural Latin’ is slight to the point of being non-existent).
During the historical period, the following shared features can be distinguished:
common Latin-Faliscan innovations:
(7) morphological: the replacement of the second-declension genitive ending /-oso/
by /-ī/ (§4.4.9), although the significance of this feature has been played down due to
a tendency to regard the genitive in /-oso/ as exclusively Faliscan (cf. §18.3.3);
347
CHAPTER 10
(8) morphological: probably also the replacement of the first-declension genitive

singular ending /-ās/ by /-āī/, although the evidence for this is ambiguous (§4.2.2);
(9) morphological: the replacement of the ending /-ed/, used in the third singular
perfect, by /-et/ or possibly by /-it/ (§5.2.4.5).
It should be noted that I regard features that Faliscan shares with Latin as ipso facto
due to Latin or Roman influence in the ager Faliscus. Granted, every development
must start somewhere, but in the case of these developments or replacements there is
no indication where in the Latin-speaking area (which in my view included the ager
Faliscus) they originated. In any case, there is no evidence that any of these features
necessarily originated at Rome, as is sometimes assumed.
Beside these points, there are a number of phonological features of recent date
that Faliscan shares with at least some of the Latin dialects. These are of minor
significance, not per se because they are more recent, but rather (a) because they are
in some cases synchronic phonetic tendencies rather than full-blown phonemic
changes, and (b) because in some cases they also occur in one or more Sabellic
languages.
common recent innovations (some also occurring in Umbrian):
(10) phonological: intervocalic rhotacism (§3.5.3), which is found both in Latin and
in Umbrian: note that Umbrian also had word-final rhotacism, which Latin and
Faliscan did not;
(11) phonological: a development /#fV/ → /#hV/, which is attested also for some
Latin dialects, but not for the Sabellic languages (§3.5.2);
(12) phonological: monophthongization of the diphthongs, which is found both in
Latin and in Umbrian and Volscian: the Faliscan monophthongizations took place
slightly earlier than in Latin, and probably at around the same time as in Umbrian
(§3.7), and affected all diphthongs, again as in Umbrian, while the Latin
monophthongizations show local (and social?) differences in which diphthongs were
monophthongized and at what date.
(13) phonetic realization: a tendency to ‘drop’ (or realize very weakly) syllable-final
nasals and liquids and word-final nasals, liquids, and the sibilant /s/ (§3.5.7): similar
tendencies can be observed in Latin and in Umbrian (where the word-final /s/ had
been rhotacized to /r/, however).
On the morphological level, there are two or three forms that can and often have been
regarded as separating Faliscan and Latin, namely fifiked EF 9 and f[.f]ịqod EF 1: to
these forms must now be added faced MF 471* and facet MF 470*.
Fifiked and f[.f]ịqod have been discussed only shortly in the section on the
synchronic perspective since from a strictly synchronic viewpoint these forms cannot
be evaluated due to a lack of comparable forms from contemporary Latin. From a
348
diachronic perspective, however, they can be described as instances of preservation of

a reduplicative perfect that is found also in Oscan and is therefore probably Proto-
Italic: note the probably Proto-Italic perfect reduplication with /i/ as the reduplicative
vowel of a root containing /i/. That later Latin had finxi is no argument against this
assumption, for finxi goes back to an aorist that may at some point have replaced a
reduplicative perfect *fifigi, especially if reduplicative perfects of roots beginning in
an original voiced aspirate disappeared in Latin (§5.3.7-8). From this perspective, the
Faliscan forms can therefore not be considered a feature that separates Faliscan from
Latin, unless it is argued that Latin never had this reduplicative perfect. However, in
view of the Faliscan form, which reflects a Proto-Italic formation, the reduplicative
perfect must have existed in Proto-Latin as well.
The same explanation can be applied to faced/facet. Faced MF 471* and facet
MF 470* contrast with Latin fecit on a synchronic level: the Latin form is first attested
in feced CIL I2.4. Leaving aside the endings and concentrating of the perfect stems,
Faliscan has /fak-/ and Latin /fēk-/, which is a clear difference. Both perfects,
however, reflect old Ablaut forms of the aorist stem, which apparently replaced an
older reduplicative perfect of a root starting in an original voiced aspirate (§5.3.6): in
this case, however, this perfect is actually attested for Latin in vhevhaked CIL I2.3
(unless the authenticity of the fibula Praenestina is again called into question).
This replacement of reduplicative perfects would therefore be a comparatively
recent innovation, occurring after the time of the earliest inscriptions, and therefore
not indicative of an early split between Faliscan and Latin. Of course, Faliscan and
Latin made different choices in the replacement of the reduplicative perfect /fefak-/,
possibly because both areas were isolated at the time, or, if one wishes to stress the
influence of the Sabellic languages on Faliscan, because Umbrian had /fak-/ as the
perfect stem: it is noteworthy that it was Latin that chose /fēk-/, the only Italic
language to do so.
The ending of f[.f]ịqod EF 1 was likewise not discussed from a synchronic
perspective. It reflects an aorist ending where Latin shows an old perfect ending, but
as I suggested, it is not clear from the Faliscan material whether it was regular in
Faliscan. The form is in itself unique and we simply cannot judge whether in Faliscan
or in Latin third plural perfect forms with aorist endings may have existed either as
regular forms or as occasional by-forms due to analogy with the aoristic third singular
ending. The only attestation of a third plural perfect ending that comes close in time to
f[.f]ịqod is Latin steterai CIL I2.2832a, a form that is likewise not without difficulties,
as it is not a direct reflex of */-ēri/ but probably shows analogical evidence from the
endings of the first and second singular. If f[.f]ịqod is regarded as significant evidence
for a separation of Faliscan from Latin, it should be noted that the form likewise
separates Faliscan from the Sabellic languages, where the corresponding ending was a
Proto-Sabellic innovation */-ens/.
349
CHAPTER 10
10.2. Recent views on the position of Faliscan
10.2.1. Recent views. In §1.5, I named several scholars that have given explicit views
on the status of Faliscan since the appearance of G. Giacomelli’s La lingua falisca
(1963). In the following sections I shall review their arguments and conclusions
briefly, to see where and how these differ from my own. I have not discussed the
statements by earlier authors (cf., however, §1.5), among whom I should name Beeler
(1956:48), who regarded Faliscan as an Italic language independent from both Latin
and the Sabellic languages, a position that I regard as entirely untenable in view
especially of the Proto-Italic and Proto-Latin phonological and morphological
developments (§§3.2-3), and which Beeler (1966:57) himself eventually abandoned.
10.2.2. Campanile: an independent Faliscan. Campanile first gave attention to the

phenomena of Faliscan in ‘Elementi dialettali nella fonetica e nella morfologia del
latino’ (SSL 1 (1961), pp.1-21), where he used the Faliscan data like those of other
Latin dialects, such as Praenestine. In his later monograph, Studi sulla posizione
dialettale del latino (1969), however, he devoted seven pages (pp.85-92) to the
position of Faliscan, where he made his views on the matter more explicit and came to
a different conclusion, questioning the idea that Faliscan is (closely) related to Latin.
First, he admitted that there are a number of features that are common to both
Faliscan and Latin, pointing e.g. to the lexical correspondences cra MF 59-60 and
foied MF 50-60 (both forms, by the way, where a comparison with the Sabellic
languages is impossible as the corresponding Sabellic forms are lacking), uxor MF 17
etc., saluete EF 4, peparai [sic] EF 1, and the morphological correspondences med EF
1, 9, and the future suffix */-bh-/ in carefo MF 59 and pipafo MF 59. He then named a
number of differences: the phonological difference in the development of the voiced
aspirates, the morphological differences in the ending of fifikod [sic] EF 1 and the
formation of douiad EF 1, which compares only to Umbrian purdouitu TI VIa.56 etc.,
finally pointing to the isolated Faliscan lexemes lecet MF 88, lepe in Pisani’s
interpretation of EF 4, and ues EF 4. As unusable for comparison he named the
monophthongization of the diphthongs (§3.7), rhotacism (§3.5.3), which in his view
could both be ascribed to Umbrian influence, and the f-/h- variation (§3.5.2), which he
ascribes to ‘a peripheral band of Latin’, and Bonfante’s (1966) idea that z- represents
a saṇdhi voicing of /#sV-/ (§3.5.3). He concluded:
“Se scartiamo, quindi, i fenomeni che rappresentano innovazioni proprie del falisco o
prestiti più o meno tardi, si resterà piuttosto perplessi innanzi alla conclamata
strettisima parentela fra il latino e il falisco. Una lingua che conosce ues e non uos,
douiad e non det, fifikod [sic] e non finxerunt (o finxere), lecet e non iacet, lepe e non
uiue, mal può, a nostro giudizio, essere valutata come una mera variante locale del
latino; e non possiamo sottrarci all’impressione che gli studiosi, nello stabilire la
350
posizione linguistica del falisco, abbiano attribuito, forse inconsciamente, peso

eccessivo alle caratteristiche meramente conservatrici che esso condivide col latino,
senza valutarne adeguatamente gli aspetti innovatori. Con ciò, naturalmente, non
concludiamo che falisco e latino siano lingue toto caelo distanti; al contrario, esse
presentano varie innovazioni comuni. Ma ciò non dovrà farci dimenticare le
innovazioni che il falisco condivide con altre tradizioni linguistiche e che gli
conferiscono, pur con le doverose riserve che c’impone la povertà delle attestazioni,
una sua precisa anche se modesta autonomia non solo nei confronti dei dialetti italici,
ma anche del latino.” (p.92)
As Campanile on this basis explicitly rejected that Faliscan is a local variant of Latin,
it is worth looking at his arguments. First of all, the argument that Faliscan did not
have uos, det, finxerunt or finxere, iacet, or uiue, is an argumentum e silentio that
cannot be substantiated: all the more so as several of these forms are Early Faliscan,
which, as said in §10.1.3, makes comparison even more difficult. As a result,
Campanile ends up comparing Early Faliscan forms with forms from much later
Latin. I agree that ues beside Latin uos is problematic, but so is Umbrian uestra TI
VIb.61 (which together with ues may point to an earlier e-vocalism) beside Paelignian
uus Pg 9: the division occurs in both branches of the Italic languages, and which of
these languages is conservative and which is innovative on this point cannot be
established on the basis of the present data (§4.7.3). With regard to f[.f]ịqod, this form
does indeed differ from Latin finxi, but as I have argued in §10.1.4, a reduplicative
perfect from fingo may eventually have disappeared in Faliscan as well, just as the
much later Latin form finxi may well have replaced an older reduplicative perfect
*fifigi. The form lepe (in EF 4) is uncertain and assumes the occurrence of a root that
is not attested in either Latin or the Sabellic languages: if only for that reason, I cannot
regard it as attested until it can be substantiated in other ways.
Lecet, the only Middle Faliscan form in Campanile’s list, may very well have
existed in Latin, in view of the nouns lectus and lectica. It is therefore not an
innovation but rather a case of preservation of a verb that had disappeared in Latin:
note also South Picene veia|t MC.1. Moreover, comparing lecet to iacet is an arbitrary
choice: as the inscriptions show, the regular and formulaic Faliscan verb was
cupat/cupant, and this is of course well-attested for Latin.
I certainly agree with Campanile that in order to establish the position of
Faliscan with regard to Latin and the Sabellic languages, one should look at the
innovations as much as at the preservations: in fact, I hope to have done so in the
preceding chapters and especially in section §10.1.4. My conclusion, however, is that
with the exception of the developments of the word-internal voiced aspirates, any
development that can be called a ‘Faliscan innovation’, whether phonological,
morphological, lexical, or syntactic, is either of early date and shared with Latin, or is
of recent, often even Middle Faliscan, date, and in most cases is parallelled by
developments that took place in other local variants of Latin.
351
CHAPTER 10
10.2.3. Solta: convergence with the Sabellic languages? In Zur Stellung der
lateinischen Sprache (1974), pp.45-47, Solta also briefly discussed the position of
Faliscan, from a mainly synchronic perspective. He stressed the correspondences
between Faliscan and Latin, naming especially the preservation of the labiovelars (cf.
§3.4), the perfect pepara[i EF 1 (cf. §5.3.1.12), which, however, cannot be compared
with a corresponding Sabellic form, and the b-future (cf. §5.2.2.2), and also remarked
on the second-declension genitive singular, noting that Faliscan had an older ending in
-osio and a later ending -i that is the same as in Latin (§4.4.3, §4.4.6), without drawing
any conclusion from this: a point he could have elaborated had he known the popliosio
ualesiosio of CIL I2.2832a at the time. As his general conclusion, however, Solta also
saw a convergence of Faliscan with the Sabellic languages:
“Die adäquateste Deutung des faliskischen Sprachhabitus ist wohl die dynamische
Auffassung, wonach sich dieser vom Lateinischen weg zum Umbrischen hinentwickelte.
Die stimmhaften Aspiraten bh, dh sind im Falisk. im Inlaut wie in den ,ital.‘ Dialekten
entwickelt, nicht wie im Lat. Der vielbehandelte Rhotazismus scheint in Falerii älter zu
sein als in Rom und stellt vielleicht das Zwischenglied in der vom Umbrischen
ausgehenden Lauttendenz bei ihrem Vordringen nach Rom dar.” (p.46)
Note that Solta’s conclusion that Faliscan was ‘developing away from Latin towards
Umbrian’ adds a diachronic dimension to his largely synchronic comparison. The first
point that he named was the development of the word-internal reflexes of the voiced
aspirates, where the Faliscan reflexes point to a development that must have been
similar, or identical, to the one in Proto-Sabellic, but that was clearly different from
Roman Latin (§3.3.3). This must have taken place at a prehistoric stage, but cannot be
dated more precisely. The second of the points named by Solta, intervocalic rhotacism,
took place also in Latin and in Umbrian, and as far as this can be established, during the
same period, namely the fourth century (§3.5.3): it can therefore not be treated on a par
with the development of the word-internal reflexes of the voiced aspirates, unless both
are assumed to illustrate a convergence of Faliscan with the Sabellic languages that
went back to a very early period. Solta could have pointed to the monophthongization
of the diphthongs as well (§3.7), for in Faliscan this took place at an earlier date than in
Latin, and the reflexes of the old diphthongs are similar to those in Umbrian. All these
points, however, are phonological, while the points that connect Faliscan and Latin are
also morphological and lexical (see §10.1.3): on these levels, there is no indication that
Faliscan was in any way developing towards Umbrian or any other Sabellic language.
The only morphological development where Faliscan shows a difference with Latin and
a similarity with Umbrian is the perfect /fak-/ in faced MF 471*, facet MF 470*.174
174
Solta’s perspective almost begs the question whether from a synchronic perspective Falis-
can could eventually have become a Sabellic language altogether, just like Trubetzkoy sug-
gested that non-Indo-European languages could eventually become Indo-European (e.g.
Trubetzkoy 1939). At least from a morphological perspective, this seems unlikely.
352
10.2.4. G. Giacomelli: a Latin Faliscan with Sabellic influence. I did intentionally

not start this discussion with G. Giacomelli’s own conclusion in La lingua falisca
(1963). There, she discussed the position of Faliscan only in the most percursory way,
stating her conclusion most clearly as follows:
“Credo invece che non si debba deflettere dalla posizione tradizionale, che distingue il
falisco dal latino, pur tenendo fermi gli stretti legami d’affinità fra i due popoli e le due
lingue: senza negare, d’altra parte, la penetrazione di elementi esterni sia italici sia
etruschi” (p.21)
This conclusion was unfortunately not argued for by a systematic comparison
(whether synchronic or diachronic), although in the linguistic part of the work she
pointed out differences and similarities here and there.
Her conclusion in the article ‘Il falisco’ (in Lingue e dialetti dell’Italia antica
(1978), pp.509-535) was different, and made some finer distinctions:
“Gli elementi che sembrano collegare alla base il latino e il falisco sono veramente
significativi: si riscontrano non solo nel lessico – che manca, si può dire, di elementi
non latini – ma anche nella grammatica: la conservazione della labiovelare [...], i plurali
dei temi in -a (e probabilmente quelli in -o) con desinenza di origine pronominale [...],
l’accusativo del pronome di 1a persona med [...], il futuro in -fo [...].” (p.522)
Here G. Giacomelli rightly pointed to a number of older developments: common
preservation in the case of the labiovelars (§3.4), common Proto-Latin innovation in
the cases of the first and second-declension nominative plural endings (sociai EF 4,
lete MLF 285: see §4.2.6, §4.3.6), and the future in -fo/-bo (carefo MF 59, pipafo MF
59: see §5.2.2.2). The form med EF 1, 9 can be ascribed either to common
preservation of a Proto-Italic innovation, or to common Proto-Latin innovation,
depending on how the Proto-Italic form is reconstructed (§4.7.2).
Yet G. Giacomelli also assumed a convergence with the Sabellic languages:
“Accanto agli elementi che associano la nostra lingua al latino non vanno però
sottovalutati quelli che rivelano un’impronta, sia pure secondaria, delle lingue italiche,
riportandosi alla fonetica e alla morfologia osco-umbra: per la seconda si notano il tipo
di perfetto raddoppiato fifiked [...], forse il genitivo in -osio, certo il dativo in -oi [...];
per la prima citiamo la tendenza a una monottongazione precoce secondo il tipo che si
riscontra nell’umbro (ma anche in parte nel latino rustico e nell’etrusco [...]), la caduta
di consonanti finali [...] e la mancata differenziazione tra posizione iniziale e posizione
interna per quanto riguarda i resultati di BH en DH indoeuropei – ma non di GH, che
dà /h/ all’iniziale e /g/ all’interno [...], in modo probabilmente affine a quello latino.
Quest’ultimo fatto è particolarmente importante per mostrarci la lingua protostorica nel
suo avvicinamento a parlate osco-umbre (probabilmente al sabino): infatti, in
qualunque modo si interpretino questi fenomeni, è certo che almeno per la fonetica non
si sono avuti svolgimenti parallelli e indipendenti, ma un’influenza italica la quale ha
avuto presa fin dal momento in cui le sonore aspirate si sono ridotte a spiranti sonore,
ma non ancora a occlusive sonore, difficilmente riversibili (e non ha avuto presa, per
ragioni che ci sfuggono, nel caso del gutturale).” (p.522-523)
353
CHAPTER 10
With this part of her conclusions I cannot agree at all. The genitive ending -osio and
the dative in -oi are clearly cases of common preservation: even before the publication
of the Satricum inscription’s popliosio ualesiosio CIL I2.2832a, there was more
(presumed) evidence for the existence of an original */-oso/ in Latin than there was
for its existence in any of the Sabellic languages (§4.4.6), while it is clear from numa-
sioi CIL I2.1, duenoi CIL I2.4 and Marius Victorinus’ remark “populoi Romanoi pro
populo Romano scito priores scribere” (CGL 6.17.20), that Latin, too, had a dative in
/-ō/, although this appeared somewhere after the period of the earliest inscriptions,
perhaps earlier than in Faliscan (§4.3.2).
The fact that fifiked EF 9 has a parallel only in Oscan fifikus Cp 37,5 does not
automatically mean that it is a borrowing, even when the form is contrasted with Latin
finxerunt: see §5.3.2.7-8. Note that fifiked EF 9 and f[.f]ịqod EF 1 are used in a
formula that was taken over from Etruscan and that the use of a verb ‘to knead’ in this
formula is unique: Oscan used úpsed/úpsens in such formulas, not fifikus (§8.9.2).
The early monophthongization (§3.7) and the ‘drop’ of word- and syllable-final
consonants (§3.5.7) are phenomena of Middle Faliscan that cannot be treated on a par
with the much earlier development of the word-internal reflexes of the voiced
aspirates (§3.3.3). In the case of the latter, I agree that the Faliscan reflexes point to a
development that was probably similar, if not identical with the Proto-Sabellic
development, and that this is the most important point of separation between Faliscan
and Latin. Monophthongization and omission of word- and syllable-final consonants
also occur in Latin dialects, however, although the monophthongization, especially of
all diphthongs, is nowhere documented as early as in Faliscan.
I find G. Giacomelli’s en masse attribution of Faliscan features to influence
from Sabellic languages very difficult. This virtually denies the possibility of
independent parallel development, attributing similar features to language contact
apparently only because they occur in a nearby language as well. Transfers on the
scale that G. Giacomelli envisaged are impossible without intense language contant,
especially as the influence affects both the phonology and the declensional
morphology (§1.3.2.2). Of such intensive language contact between Faliscan and the
Sabellic languages, the inscriptions show little or no trace: in fact, the epigraphic and
linguistic features in inscriptions from the agri Faliscus and Capenas that can be
ascribed to contact with the Sabellic features are few (§9.3). Furthermore, if the
morphology and phonology of Faliscan are as heavily influenced by the Sabellic
languages as G. Giacomelli suggested, why is there no trace of this influence in the
lexicon, which would be the first level on which such influence would manifest itself,
or in the formulaic texts? The Faliscan lexicon is Latin (§6.3), with only posticnu
MLF/Cap 474* and p ̣esco(m) Cap 431 as possibly Sabellic words, both of which are
interferential forms rather than borrowings (§9.3.2), and traces of Sabellic influence in
the formulaic texts are minimal and limited to the ager Capenas (§8.8.2).
354
10.2.5. Joseph & Wallace: not a Latin dialect. The most influential and best-argued
verdict on the position of Faliscan is surely Joseph & Wallace’s article ‘Is Faliscan a
local Latin patois?’ (Diachronica 8 (1991), pp.159-186). Their conclusions are adopted
by Baldi (The foundations of Latin (1999), pp.170-4), and apparently also by Meiser
(Handbuch der lateinischen Laut- und Formenlehre (1998), p.9-10).
Contrary to most authors on the subject, they make it quite clear what they mean
by the term ‘Latin’, which they limit to the language of Latium and the colonies
sponsored by Rome, thus in effect already ruling out that Faliscan was a Latin dialect.
The term ‘dialect’ they reserve for non-Roman Latin. They also challenge the notion
that shared innovation is necessarily indicative of an original linguistic unity, assuming
that it can also “reflect the areal spread of changes after considerable divergence of the
two languages had taken place” (p.163). I would like to point out that this is exactly the
problem discussed in §1.2: from a purely synchronic perspective, the result of shared
innovation is indistinguishable from unilateral innovation with subsequent spread of the
innovative feature. Therefore, the diachronic perspective should be taken into account
wherever possible in judging such features. Also, Joseph & Wallace indicate nowhere
under which circumstances such a spread could happen or how far the languages can
have diverged for such a spread still to be possible, and this makes it difficult to see
whether or not this is applicable to Faliscan. If this could happen unconditionally, the
value of the diachronic method would in effect be denied.
Joseph & Wallace then point to a number of features that Faliscan shared with
dialect Latin: the monophthongization of the diphthongs (§3.7), the representation of
antevocalic /i/ by e (§3.6.2), the loss of word-final /t/ (§3.5.7c), the loss of word-final /s/
after long vowels (§3.5.7d), the realisation of /#fV-/ as [h] (§3.5.2), the third-declension
genitive in -os (§4.5.2), /e/ represented as i before /rC/ (§3.6.2), the second-declension
genitive in -osio (§4.4.2), and f as a reflex of */bh dh/ in medial position (§3.3.3).175
They conclude that with the exception of the genitives in -osio and -os and the
development of f as a reflex of */bh dh/ these features are all recent and can therefore
not be considered significant for establishing the position of Faliscan, all the more so as
several of these features also show up in the Sabellic languages.
Quite rightly, they attach more value to the older features that bind or separate
Latin and Faliscan: as such, they name the f-future in carefo MF 59, pipafo MF 59
(§5.2.2.2) and the accusative med EF 1, 9 of the pronoun of the first person singular
(§4.7.2) as shared innovations (which were not shared by the Sabellic languages), and
the development of the word-internal reflexes of the original voiced aspirates (§3.3.3) as
a feature that definitely separates the two. They admit that other connecting features
175
I leave out their first-declension dative singular in -a: this is found only in LF/Lat 214,
which may be Latin rather than Faliscan, and which is a feature that, as said in §3.7.6, seems
to have spread through Latin colonies. It may be regarded as Latin rather than Faliscan.
355
CHAPTER 10
may exist, but point rather to two other separating features, namely the third plural
perfect endings, where they see a separation, and to the fact that Faliscan shows a lecet
MF 88 that in Latin was replaced by the new formation iaceo. The former point I regard
as un-evaluable (§10.1.4 and §5.2.4.5), while the latter must have been a relatively
recent and at least post-Proto-Italic innovation in Latin. Their conclusion is:
“The overall force of the evidence brought forth here points towards the conclusion that
Faliscan is not a dialect of Latin. Not only are the features typically adduced in support of
the view of Faliscan as dialectal Latin inadmissible once well-established principles for
sub-grouping of dialects are brought into play, but there are as well significant
innovations that separate Faliscan and Latin off from each other. Some unite all Latin
dialects but exclude Faliscan and others are found only in Faliscan to the exclusion of all
of Latinity. These facts therefore confirm the traditional view that Latin and Faliscan are
distinct languages, though closely related to one another as immediate siblings.” (p.182)
This implies that from a synchronic perspective Faliscan may give the impression of
being a Latin dialect due to a number of shared features, but that these features are due
to recent developments, sometimes shared with (or due to?) the Sabellic languages,
notably Umbrian, while from a diachronic perspective, Faliscan has several older
features that separate it from Latin.
In my view, this conclusion gives too much value to the one phonological feature
that separates Faliscan and Latin, namely the development of the word-internal reflexes
of the voiced aspirates (§3.3.3) and leaves undiscussed a number of morphological
innovations (§10.1.4) shared by Latin and Faliscan, but not by the Sabellic languages,
as well as the fact that the lexicon is overwhelmingly Latin (§10.1.3). These features
cannot be explained by just assuming that spread of an innovation can take place even if
the languages have diverged significantly, without discussing at least the conditions
under which such a transfer can take place, and an attempt to show that these conditions
were fulfilled to some extent. Apart from that, I find it hard to assume that these features
could be ascribed to Latin influence on a (closely related but still quite distinct)
language. All the features named by Joseph & Wallace, recent or not, can be dated to the
period before the war of 241, and there is hardly a trace of Latin having been such a
strong influence before 240.
Also, if Faliscan belongs to the Latin branch of the Italic languages, as Joseph &
Wallace assume and apparently even defend, but is not a Latin dialect, what is it then? If
it is a distinct language, it is surely a language that (as far as can be judged on the basis
of a diachronic analysis of the material) is separated from Latin by one or two features
at best, and that from a synchronic perspective is quite similar to Latin dialects. Some of
these features, but surely not all, can be ascribed to recent developments, and not
necessarily to influence from Latin. In my view, this is not enough to declare Faliscan
an independent language. Among the various local variants of Latin, Faliscan may well
be the one that is most separated from Latin, but the degree of separation still remains
inside the limits of dialect variation.
356
10.2.6. R. Giacomelli: level-distinction. R. Giacomelli, in Ricerche falische (1978),

extensively discussed a number of Faliscan linguistic phenomena, drawing the
following concusion:
“Dall’analisi della fonetica e della morfologia falische che abbiamo condotta sia pur
senza pretesa di completezza nelle pagine di questo lavoro, risulta evidente la gran
quantità di fatti che ab antiquo il falisco mostra di condividere con il latino.
In tale contesto è possibile bensi rilevare la presenza di alcuni fenomeni tipicamente
falischi, accanto ad altri che pur trovando riscontro in osco-umbro, possono tuttavia
agevolmente essere classificati come propri di un ambiente dialettale cui è possibile, a
buon diritto, ascrivere il falisco.
Tuttavia anche fenomeni come gli esiti italici delle medie aspirate ie. ovvero la
generale monottongazione dei dittonghi, che parrebbero con sicurezza avvicinare il
falisco all’umbro, se osservati in una prospettiva geo-linguistica più ampia, inducono a
ridimensionare l’incidenza, nella loro determinazione, della generica soluzione degli
influssi italici di Faleri.
Infine, gli evidenti influssi etruschi ce si palesano nel campo dell’onomastica falisca
sono cronologicamente recenti ed in pratica vengono a sovrapporsi, senza modificarne
minimamente la configurazione dell’insieme, al quadro generale della latinità del
falisco.” (p.67)
As an overall conclusion, I think this is a fair assessment of the data, even without going
into all the data as discussed by R. Giacomelli, where I do not always share his views.
He does not adopt G. Giacomelli’s idea of a Faliscan that is heavily influenced by the
Sabellic languages (cf. §10.2.4), but does not deny that some Faliscan features, such as
the development of the word-internal reflexes of the voiced aspirates, appear to be
Sabellic rather than Latin and can apparently be ascribed to a shared development of
related, but different, languages, in the line of Pisani’s Italic Sprachbund (§1.3.2.1).
In his recent Nuove ricerche falische (2006), the most recent contribution to the
field of Faliscan studies, he by and large draws similar conclusions, but now places the
data in the perspective of level-distinction, a point that had been prominent in his earlier
study as well. In chapters VI (Volgarismi, pp.73-107) and VII (Il falisco, un sub-
standard latino, pp.108-119), he makes a synchronic comparison between Faliscan and
‘vulgar’ Latin, and rightly shows the many similarities between the two. What I cannot
accept is his idea that this non-Roman Latin represents a substandard form of Latin, if
only because I very much doubt if, during the Middle Faliscan period, there was already
a standard of Roman Latin, let alone that this Roman standard was already regarded as
the standard for all Latinity. Roman Latin was itself still very much finding its ground in
establishing a standard, and I wonder if even Romans themselves would have thought
in terms of ‘higher level’ and ‘lower level’ Latin in the third century: if they did so, it
was perhaps as a social, but hardly as a geolinguistic distinction. From the perspective
of ethnic identity, of course, Faliscan was the local standard in the area and will hardly
have been regarded as ‘substandard’ by its speakers, at least before the process of
Romanization was well under way, that is, by the end of the third century.
357
CHAPTER 10
10.3. Faliscan, a Latin dialect
Above, I have discussed several approaches to the question of the position of Faliscan,
and several recent views on the subject. Then what, finally, is my own verdict? I shall
discuss this mainly from the diachronic, but partly also from the synchronic perspective.
I shall take no recourse to the possibility of extensive language contacts and influence
from other Italic languages on the deeper levels of language, even though these might
be adduced: in my view, these are not necessary to explain the features of Faliscan,
although they in no way contradict my conclusion and could even be used to explain
those features that can be regarded as contradicting it.
First, I think that this study once more shows that from a diachronic perspective
Faliscan clearly belongs to the Latin branch of the Italic family: this is shown by a
number of diachronic developments of Proto-Latin date operating on the phonological,
morphological, lexical, and perhaps even syntactic levels where Faliscan sides with
Latin, which have been enumerated in §10.1.4. These developments consist both of
common Proto-Latin innovations and of common preservation of features that in the
Sabellic languages were replaced by Proto-Sabellic innovations. The one great
exception is the development of the word-internal reflexes of the voiced aspirates,
where the development of Faliscan reflects a development similar to that of the Sabellic
languages: this is discussed below.
Second, Faliscan shared with Latin a number of developments that took place at a
more recent date, that is, during the historic period. Notably, Faliscan and Latin shared
several important morphological innovations that did not occur in the Sabellic
languages, namely the replacement of the -osio by -i during the fifth and fourth
centuries, the replacement of -ed by -et during the late fourth century, and probably also
that of -as by -ai during the third century. A shared phonological innovation is the
intervocalic rhotacism during the fourth century: although rhotacism also occurs in
Umbrian, in Latin and Faliscan it was limited to intervocalic position, while in
Umbrian, rhotacism took place also in word-final position.
To the historic period should probably also be ascribed the innovation of the
perfect, in Faliscan to faced/facet /fak-/ and in Latin as feced /fēk-/: in view of Latin
vhevhaked, these were both recent replacements of the reduplicative perfect by stems
that originally belonged to the aorist. Both languages differ in how they reformed the
perfect, Faliscan generalizing /fak-/ (which is also the perfect stem used in Umbrian),
and Latin, /fēk-/. If this replacement was due to a general tendency to replace
reduplicative perfects from verbs with a root in /f-/ that operated in both Faliscan and
Latin, this opens the possibility that Early Faliscan fifiked/f[.f]ịqod was also replaced by
an unattested form of aorist origin: note the original aorist used as a perfect in Latin
finxi. This removes a point of difference between Faliscan and Latin.
358
The more recent developments and tendencies in Middle Faliscan, e.g. the weak
realization of word- and syllable-final nasals and of the word-final sibilant /s/, and the
realisation of /#fV/ as [h], also occurred in other variants of Latin, even if they are more
frequently or better attested for Faliscan. In the case of the monophthongization of the
diphthongs during the Middle Faliscan period, the Faliscan development was slightly
earlier than in Latin. Note that several of these tendencies also occur in Umbrian.
From a synchronic perspective, I could add that the lexicon of Middle Faliscan is
very much the same as that of Latin, and that in all instances where there is a clear
difference between the Latin and the Sabellic lexicon, Faliscan sides with Latin. I can
see no reason to ascribe this to Latin influence, certainly not for the period before 240.
The only older development that separates Faliscan from Latin is the development of
the word-internal reflexes of the voiced aspirates, which unfortunately can be dated
only relatively as predating the earliest texts. Here, Faliscan and Latin are clearly
different, and the difference is ‘old’: as said, I think it at best unproven and at worst
unlikely that there were other Latin dialects that showed the same development and that
what we call the ‘Latin’ development was originally limited to Roman Latin. A far more
recent development is the innovation of the perfect of facio, where Faliscan chose for
/fak-/ and Latin generalized /fēk-/. Are these two differences enough to say that Faliscan
is not Latin? This depends on how far dialects are allowed to vary before they can be
said that they constitute a language of their own. With regard to the ‘grammar’ and the
lexicon, Faliscan is Latin. The differences are mainly matters of phonology, or even
phonetics, most of them of recent or even very recent date. If these developments are
regarded as ‘acceptable’ within the degree of variation that can be expected within the
local variants of one language, Faliscan is, indeed, a Latin dialect.
As the only local variant of the Latin language outside Latium, Faliscan may at
times have been isolated to a varying degree from the remainder of the Latin-speaking
area, and such a period of isolation might explain the separate development of the word-
internal reflexes of the voiced aspirates, and perhaps even the comparatively recent
generalization of /fak-/ as a perfect stem. Yet, if there were such periods of isolation,
Faliscan would be expected to have diverged from Latin to a far greater extent than
appears to be the case. That the divergence remained limited may be due to fairly
frequent contacts with the remainder of the Latin-speaking community. The problem
with Faliscan seems to be that there is a large number of respects in which Faliscan is,
for all intents and purposes, Latin, while there are a small number of features in which
Faliscan is not Latin at all, and that these separating features are separated from each
other in time and are not either all recent or all old, implying that they are not due to
one continuous process of divergence or convergence. Some of these could of course
have arisen due to local variation, as can be expected in a dialect, and especially one
both physically and historically at the periphery of the language of which it was a part.
359
CHAPTER 10
In §2.4.2. I mentioned the possibility that the area where Latin was spoken
extended to the north of the Tiber as well, and disappeared or shrank with the growth
and emergence of Etruscan culture and language. The area in between Latium and the
ager Faliscus remained, to some extent, an area where Latin continued to be spoken
until the Roman conquest made it ‘officially’ Latin again. Perhaps the differences and
correspondences between Faliscan and Latin are to be ascribed to such alternating
stages when Faliscan was now the northernmost exponent of a Latin-speaking
continuum stretching southward all the way to Latium Adiectum, and now separated
from that continuum and in more frequent contact with speakers of the greatly different
Etruscan and the not-so-different Sabellic languages. It might be this now closer, now
looser bond of Faliscan with the rest of Latin to which its many similarities, but its few
but significant dissimilarities with Latin may have to be ascribed.
360

The publication of this book is made possible by a grant from the Machtelt Bolkestein
Fonds.
Lay out: Gabriël C.L.M. Bakkum, Amsterdam

Cover design: René Staelenberg, Amsterdam
ISBN 978 90 5629 562 2

e-ISBN 978 90 4850 875 4
NUR 630
© G.C.L.M. Bakkum / Vossiuspers UvA – Amsterdam University Press, 2009
All rights reserved. Without limiting the rights under copyright reserved above, no part of this
book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any
form or by any means (electronic, mechanical, photocopying, recording or otherwise) without
the written permission of both the copyright owner and the author of the book.
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam,
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het College voor Promoties ingestelde
commissie, in het openbaar te verdedigen in de Agnietenkapel
op dinsdag 10 februari 2009, te 14.00 uur
door
Gabriël Cornelis Leonides Maria Bakkum

geboren te Amsterdam
Promotiecommissie:
Promotor: prof. em. dr. H. Pinkster
Overige leden: prof. dr. Ph. Baldi

prof. dr. M. Gnade
prof. dr. C. Kroon
prof. dr. P.Poccetti
dr. R. Risselada
Faculteit der Geesteswetenschappen

Chapter 11
The epigraphic material
This chapter constitutes the introduction to the presentation of the epigraphic material in
chapters 12-19. In it, I discuss the nature of the epigraphic material, subdivided according to
various criteria such as provenance, age, type, and alphabet (§11.1), the Faliscan alphabet and
orthography (§11.2), and the organization of the way in which the material is presented in
chapters 12-19 (§11.3).
11.1. The Faliscan epigraphic material
11.1.1. General. In the second part of this study, I present the epigraphic material on
which the discussions and conclusions in chapters 2-10 are based. This edition is
intended to show the reasons and motives behind my readings and interpretations of the
individual texts, and hopefully to discard some of the impossible readings and interpre-
tations that still crop up in the literature from time to time. Discussions other than those
necessary to establish the correct or most plausible reading have therefore in many
cases been replaced by a reference to the relevant section of chapters 3-9. The aim of
the presentation was not to provide a fully-fledged epigraphic edition, as will be clear
from the small number of drawings: the edition is intended as a means to an end rather
than as an end in itself, and rather than expanding it even further, I have chosen to add
references to all published photographs and drawings known to me.
The material presented in chapters 12-19 comprises 535 inscriptions from the
ager Faliscus and Capenas, including several inscriptions from other locations or of
uncertain or unknown origin which have been regarded as originating from the area, or
as somehow connected with Faliscan or Capenate. As the material is intended to be the
basis of a comparative study of dialect and language, I have chosen to include every
inscription from the area known to me, whether it is Faliscan, Capenate, Latin, Etrus-
can, or Sabellic, which consists of more than one letter.176 In the following sections I
have subdivided this material according to provenance (§11.1.2), period/alphabet group
(§11.1.3), type (§11.1.4), and alphabet (§11.1.5), with a more detailed discussion of the
criteria.
176
Inscriptions consisting of one letter are not only without linguistic value, it is also unclear
whether they are in fact letters. A cross may be a Faliscan or Latin x (), an Etruscan s̽ (),
the number 10 (), a Faliscan or Etruscan t (or ), or a cross; an arrow may be a Faliscan f
(), an Etruscan χ ( or ), a Sabellic ú ( or ), the number 50 ( or ), or an arrow.
361
CHAPTER 11
Of the 535 inscriptions presented here, 18 are known only through apographs
(MF 20, 65, 88-89, 138-139, 201, 211-212, 265-266, MLF 353-355, LF 335, and Etr
XXXIV-XXXV, and XLIX). Of the remaining 517, I publish 125 from autopsy,
namely EF 1-4, 6-7, 10, EF/Etr 5, MF 14, 35, 59-60, 62, 90-91, 94-96, 98, 100,
102-103, 105, 113-116, 118-127, 132, 136-137, 140-146, 148-154, 158-161, 163-164,
166-170, 175-190, 264, 269-272, 275-276, MF? 128-131, MF/Etr 37, 64, 66, MLF 320,
323-324, 360, LF 220-230, 329-334, 336, LF/Lat 214, LtF 63, 171, 174, 231-233, and
340. The others I have published e prioribus, working from all available photographs,
drawings, and transcriptions of the text in typeset. The only cases in which neither of
these were available are MF 135, LF 246, Lat 250, MLF 358-359, and Cap 394. The
following 111 inscriptions were of little or no linguistic value:
(1) 19 inscriptions consisting only of abbreviations of a praenomen and a gentilicium:
MF 29, 38, MF? 33, Cap 395-397, 400, 415, 419, 424-425, 427-429, 452-455, 461.
These have been used in the chapter on the onomasticon (see §7.1.2), but not elsewhere.
(2) 55 inscriptions consisting only of abbreviations of two or three letters: EF 8, MF 44,
46, 76, 209, 274, 281, 283-284, 294, 373-375, 460, MF? 28, 30, 68, 131, 133, 134, 203-
204, 254-255, MF/Etr 37, 256, MLF 320-323, MF/LtF 241, 252-253, 277-278, LF?
381, LtF 286, Cap 398, 401-402, 405-414, 416-418, 426, 439-451, and Lat 386. Most of
these have only been used in the chapter on the onomasticon (see §7.1.2).
(3) 26 inscriptions consisting only of a few legible letters without word-divisions: MF
55, 104, 106-108, 132, 168, 176-177, 179, 182-190, 192-194, 319, MLF 342-345.
(4) 7 inscriptions that are illegible or so disputed as to be functionally illegible: MF/Etr
287, LtF 288, MF/Etr 61, MLF 356-357, and Cap 422-423.
(5) 4 inscriptions that may be falsifications: MF 335 (known only through an apograph,
and perhaps genuine), MLF 464, Etr XXXI, and an inscription discussed under Etr
XXXIX.
11.1.2. The material divided according to provenance. I have included all the
epigraphic material from before c.100-50 BCE (see §11.1.3) from the ager Faliscus and
the ager Capenas. The extent of these areas I have tried to establish in §2.1.2: broadly
speaking, the ager Faliscus comprised the area enclosed by the Tiber, the Monti Cimini,
the Monti Sabatini, the ridge connecting these to Monte Soratte, and Monte Soratte
itself, while the ager Capenas comprised the area southward from Monte Soratte along
the Tiber to Capena and the shrine of Lucus Feroniae to the crossing of the Tiber near
Monterotondo.
The area therefore includes the towns of Narce, Nepi and Sutri, even though these
towns became dependencies of Veii in the sixth or early fifth century (§2.4.2) and
subsequently came under Roman rule from the early fourth century (§2.5.2). Although
they thus ceased to be a part of the ager Faliscus at an early date, these towns and their
362
THE EPIGRAPHIC MATERIAL
inscriptions could not be omitted: Narce was in fact one of the most important sites of
the area during the Early Faliscan period (§2.4.2), perhaps the site of Fescennium,
which is named as a Faliscan town by the ancient sources (§2.1.2). The area of these
towns provides 24 Etruscan inscriptions (Etr I-XVIII from Narce, Etr XIX from
Mazzano Romano, and Etr XX-XXIV from Nepi), nearly half of the 51 Etruscan
inscriptions presented in chapter 19.
I have also included several inscriptions of uncertain or unknown origin that are
ascribed to the ager Faliscus or Capenas, whether I support this attribution (as in the
cases of 467*-478*) or not (as in the cases of 479†-481†), and three inscriptions from
Ardea (482†-484†) that have for various reasons been ‘associated’ with Faliscan.
Apart from chapters 12 (the Early Faliscan inscriptions) and 19 (the Etruscan
inscriptions), the presentation of the material is ordered by provenance:
Chapters 13-14: Civita Castellana (Falerii Veteres) and surroundings;
Chapter 15: S. Maria di Falleri (Falerii Novi) and surroundings;
Chapter 16: the sites of the northern ager Faliscus (Corchiano, Vignanello,
Fabbrica di Roma, Carbognano-Vallerano, the site at Grotta Porci-
osa, and the area around Gallese and Borghetto);
Chapter 17: the sites of the south-eastern ager Faliscus and the ager Capenas
(Rignano Flaminio, S. Oreste, Ponzano Romano, Civitella S. Paolo,
Fiano Romano, Civitucola (ancient Capena), and the shrine at Lucus
Feroniae);
Chapter 18: (1) inscriptions of unknown or disputed origin that may be from the
ager Faliscus or Capenas; (2) inscriptions from other provenances
that are or have been regarded as Faliscan.
A similar organization is used within chapter 19, where the Etruscan inscriptions are
presented: these are divided into (1) inscriptions from Narce (perhaps the site of
Fescennium) and the south-western ager Faliscus, (2) Civita Castellana (Falerii
Veteres), (3) Corchiano and the northern ager Faliscus, (4) the south-eastern ager
Faliscus and the ager Capenas, (5) and inscriptions of unknown or disputed origin that
may be from the area.
Dividing the area in this way also serves another aim, as it divides the material
into (a) Civita Castellana, the main site during the Middle Faliscan period (chapters 13-
14); (b) S. Maria di Falleri, the main site during the Late Faliscan and subsequent
periods; (c) the northern ager Faliscus, which on the one hand was probably divided up
during the division of the ager Faliscus following the war of 241 (§2.6.2), but which
also shows signs of Etruscan presence at its main site, Corchiano (§9.2.3); and (d) the
south-eastern ager Faliscus and the ager Capenas, where Latin influence was far more
pronounced due the Roman colonization of Capena in the early fourth century (§2.5.2),
but which also shows signs of the presence of speakers of Sabellic languages (§9.3).
363
CHAPTER 11
11.1.3. The material divided according to period/alphabet category. As explained in

§1.4.3, the dating of the inscriptions is very difficult in most cases. This relative lack of
securely dated Faliscan inscriptions also makes it impossible to provide dating criteria
that rely on alphabet or orthography, except in the most general way.
An exception is constituted by the inscriptions on movable objects, especially
pottery or mirrors, where the object itself can usually be dated more or less accurately
on typological grounds. The inscriptions that were added by their makers at the time of
the making itself (signatures) are of course from the same time as the object. Besitzerin-
schriften and other inscriptions on such objects are usually tacitly assumed to have been
added soon after the object was made and therefore to have approximately the same
date, although strictly speaking there is no a priori justification for this assumption.
Most of the Faliscan inscriptions are sepulchral, however, and these are unfortu-
nately far more difficult to date. The sepulchral inscriptions are all from chamber tombs
cut into the steep rock-faces of the area, or into the sides of hollow roads (§11.1.4.1).
Here three main problems affect the dating:
(a) These tombs were often re-used for long periods of time, and when a loculus was
emptied to make place for another burial, previous burials were inevitably disturbed,
grave-goods becoming confused with those of later burials or just left lying around in
the tomb.177 In a number of cases, even the inscribed tiles used to close the loculus were
reused for other burials (§11.1.4). Inscriptions at the entrance of the tombs or on the
walls of the chamber are virtually impossible to date, as it is unclear to which stage of
the use of the tomb they belong.
(b) Over the centuries, many tombs were ransacked, and the chambers cleaned out and
re-used as cattle-stalls, shepherds’ shelters, tool-sheds, or cheese-cellars. Datable grave-
goods thus became separated from the sepulchral inscriptions belonging to the same
burial or tomb, and tiles or inscriptions on the walls of tombs are therefore often without
any datable context.
(c) Most sepulchral inscriptions are written on tiles (cf. §11.1.4.1c), and although
these could of course be dated by thermoluminescence dating, that would give only
the date of the tile itself as a terminus post quem for the inscription: it is not clear if
the tiles that were used to close the loculi were (always) new, or had (sometimes)
already been in use as roof-tiles for, say, half a century.
Public inscriptions fall into two categories. The first group consists of the inscriptions
cut into the rock-face of the side of hollow roads, which usually contain names,
presumably those of magistrates that had the road constructed or maintained: these are
without datable context at all apart from the roads themselves, which in most cases
177
A striking example is the tomb of the gens Velminaea at Vignanello: although found
apparently undisturbed and dating from the third century, it contained remains of a shield of a
type that is associated with burials of the seventh century (Giglioli 1916:64-5).
364
cannot be dated with any accuracy. The public inscriptions on bronze on the other hand
can be dated quite accurately, but only because these are nearly all in Latin and can
therefore be dated according to the epigraphic and linguistic dating criteria used for the
Latin inscriptions.
As said in §1.4.3, I have therefore used a different method of dating, which is
basically an elaboration of the criteria suggested by G. Giacomelli (1978:510-1). In this
method, the inscriptions are divided into larger ‘period groups’ according to historical
and archaeological criteria. Although crude in both its methodology and its criteria, this
way of dating has turned out to be very workable, and the groups arrived at in this way
often coincide with groups that can be defined on the basis of linguistic features. In my
view, this classification can therefore be regarded as a valid tool to subdivide the
corpus. In any case, it remains an open question whether more subtle dating criteria
would produce significantly better or different results. Inscriptions are unique objects,
and, from a linguistic perspective, they represent unique speech utterances: in a
linguistic study, placing them together into larger groups would have been necessary to
provide meaningful results in any case.
The ‘period groups’ into which the inscriptions are divided are the following:
(1) the Early Faliscan group (EF, 10-12 inscriptions) comprises all inscriptions from
before the fourth century. Since these are all inscriptions on pottery, they can be dated
quite well on typological grounds, and form a group that is quite distinct in several other
respects, such as the alphabet and the contents of the inscriptions. The Early Faliscan
inscriptions are EF 1-4 and 6-10 from Civita Castellana and EF 467* of unknown
origin: either Early Faliscan or Etruscan are EF/Etr 5 from Civita Castellana and EF/Etr
385 from Fiano Romano.
All inscriptions in the Faliscan alphabet from the fourth century and later are classed
either as Middle Faliscan or as Late Faliscan. The dividing line between the Middle
Faliscan and the Late Faliscan periods is the war of 241-240 BCE (§2.6), since as a
result of this war several of the more important sites were abandoned, and the influence
of (Roman) Latin can reasonably be assumed to have increased markedly, due to the
foundation of Falerii Novi and the division of the ager Faliscus into a Faliscan- and a
Roman-administrated part (§2.6.2).
(3) the Middle Faliscan group (MF, 185-228 inscriptions) comprises (a) all inscrip-
tions in the Faliscan alphabet from the fourth century or later that have been found at
sites that were abandoned after the war of 241, notably Civita Castellana (Falerii
Veteres) and Corchiano, and their direct surroundings, and are therefore assumed to
date from between the fourth century and c.240. Inscriptions from these sites in the
Faliscan alphabet have always been classed as Middle Faliscan unless there is a positive
reason not to do so. Middle Faliscan also comprises (b) all inscriptions in the Faliscan
alphabet from other sites that can reasonably be dated to the period between the fourth
365
CHAPTER 11
century and c.240. The Middle Faliscan inscriptions are: (a) MF 11-20, 22-27, 31-32,
34-36, 39-43, 45, 47-60, 62, 65, 69-75, 79-110, 113-127, 132 , 135-139, 141-170, 175-
198, 200-201, probably also MF? 28-30, 33, 38, 44, 46, 68, 76, 78, 111, 128-131, 133-
134, 203-204, and possibly also MF? 202 (unclear), from Civita Castellana (Falerii
Veteres), MF 257-260, 263, 265-266, 269-276, and probably also MF? 253-255, 261-
262, 281, and 283-284, from Corchiano; (b) MF 367-375 from Rignano Flaminio, MF
376 from S. Oreste, and MF 469*-473* of unknown origin. Either Middle Faliscan or
Etruscan are MF/Etr 37, 61 (illegible), 64, 66-67, 77, and 199, from Civita Castellana
(Falerii Veteres), and MF/Etr 256, 264, 267, 279-280, 282, and 287, from Corchiano
and surroundings. Either Middle Faliscan or Latino-Faliscan are MF/LtF 21 from Civita
Castellana (Falerii Veteres) and MF/LtF 252, 277-278 from Corchiano.
(4) the Middle or Late Faliscan group (MLF, 57-66 inscriptions) comprises all
inscriptions in the Faliscan alphabet from the fourth century or later that have been
found at sites that continued to exist after the war of 241, and as a consequence cannot
be dated with any kind of accuracy other than that they are from between the fourth and
the second centuries. Inscriptions from these sites in the Faliscan alphabet have always
been classed as Middle or Late Faliscan unless there is a positive reason not to do so.
The Middle or Late Faliscan inscriptions are: MLF 206-207 and 210-212, from the
wider surroundings of Civita Castellana (Falerii Veteres); MLF 285-286, 293, and 297-
298, from the wider surroundings of Corchiano; MLF 302-323 from Vignanello; MLF
324 from Fabbrica di Roma; MLF 338-339 and 346-355 from Grotta Porciosa and the
surrounding area; MLF 358-359 from the area near Gallese and Borghetto (although
very little is known of these inscriptions, they appear to be in the Faliscan alphabet);
MLF 360-362 of unknown northern Faliscan origin; MLF 363-366 from Rignano
Flaminio; MLF 459-60 and 463-464 of unknown Capenate origin (MLF 464 may be a
falsum), and probably also MLF/Cap 474*-476* of unknown origin. Either Middle or
Late Faliscan or Etruscan are MLF/Etr 208-209 from the wider surroundings of Civita
Castellana (Falerii Veteres), MLF/Etr 289 from the wider surroundings of Corchiano,
and MLF/Etr 356-357 from the area between Gallese and Borghetto. Either Middle or
Late Faliscan or Latino-Faliscan are MLF/LtF 241 and 252 from the surroundings of S.
Maria di Falleri (Falerii Novi.).
(5) the Late Faliscan group (LF, 40-41 inscriptions) comprises (a) all inscriptions in
the Faliscan alphabet that have been found at S. Maria di Falleri (Falerii Novi), as that
site appears to have come into existence or prominence only after the war of 241-240
(§2.6.2). Inscriptions from this site in the Faliscan alphabet have always been classed as
Late Faliscan unless there is a positive reason not to do so. Late Faliscan also comprises
(b) all inscriptions in the Faliscan alphabet that can with reasonable certainty be dated to
the period after the war of 241-240. These inscriptions are therefore assumed to date
from between c.240-220 and the middle of the second century. The Late Faliscan
366
inscriptions are: (a) LF 213, 220-230, 232 (partly), 234-236, and 242-249 from S. Maria
di Falleri (Falerii Novi); (b) LF 112 from Civita Castellana (Falerii Veteres), LF 329-
337 from Carbognano-Vallerano (LF 335 may be a falsum), LF 378-380, 382-384, and
probably also LF? 381, from Civitella S. Paolo.
The remaining inscriptions are either in the Latin or in the Etruscan alphabet. Those in
the Latin alphabet are a priori more likely to date from the period after c.240, but in
many cases this date cannot be positively established. Also, there are inscriptions in the
Latin alphabet that are obviously earlier than c.240, such as Lat 268, on a fourth-century
strigilis found at Corchiano, showing that it is dangerous to assume a priori a date after
c.240. The inscriptions in the Latin alphabet have therefore been classed as follows:
(6) the Latino-Faliscan group (LtF, 33-38 inscriptions) comprises the inscriptions in
the Latin alphabet from the ager Faliscus. Inscriptions in the Latin alphabet from the
ager Faliscus have always been classed as Latino-Faliscan (and thus as representing a
local form of Latin) unless they show linguistic features that are not in accordance with
those encountered in the Faliscan inscriptions. The Latino-Faliscan inscriptions are: LtF
63, 140, 171-174, and 205, from Civita Castellana (Falerii Veteres) and surroundings,
LtF 215, 231-233, 239, from S. Maria di Falleri (Falerii Novi) and surroundings, LtF
277-278, 288, 290, 292, 294 and 299-301 from Corchiano and surroundings, LtF 325-
327 from Carbognano-Vallerano, LtF 328 from Fabbrica di Roma, LtF 340-345 from
Grotta Porciosa, and LtF 377 from Ponzano. Either Middle Faliscan or Latino-Faliscan
are MF/LtF 21 from Civita Castellana (Falerii Veteres), and MF/LtF 253, 277, and 278
from Corchiano. Either Middle or Late Faliscan or Latino-Faliscan are MLF/LtF 241
and 252 from the surroundings of S. Maria di Falleri (Falerii Novi). Several Latino-
Faliscan inscriptions consist only of abbreviations (LtF 172, 174, 205, 241, 277-278,
294, 342-345, 12 inscriptions in all) and can therefore not be evaluated linguistically.
(7) the Capenate group (Cap, 72 inscriptions) comprises the inscriptions in the Latin
alphabet from the ager Capenas. Inscriptions in the Latin alphabet from the ager
Capenas have always been classed as Capenate unless they show linguistic features that
are not in accordance with those encountered in the Faliscan inscriptions: ‘Capenate’ is
therefore a counterpart to ‘Latino-Faliscan’. The Capenate inscriptions are: Cap 386-
392 and 394-430 from Capena, Cap 431, 433, 435, 437, 439-455 from Lucus Feroniae,
Cap 457-459 and 461-462 and 465-466 of unknown Capenate origin. Either Middle or
Late Faliscan or Capenare are MLF/Cap 474*-476* of unknown origin. Most Capenate
inscriptions only consist of abbreviations (Cap 386, 395-398, 400-403, 405-412, 414-
419, 424-429, 437-457, 458-459, 53 inscriptions in all) and can therefore not be
evaluated for linguistic features.
(8) the Latin group (Lat, 19 inscriptions) comprises the inscriptions in the Latin
alphabet that show linguistic features that are not compatible with those found in the
inscriptions in the Faliscan alphabet: many of these can be dated with some degree of
367
CHAPTER 11
accuracy to the middle of the second century or later. The Latin inscriptions comprise:
Lat 216-219, 237-238, 240, 250, and 251 from S. Maria di Falleri (Falerii Novi) and
surroundings; Lat 268 (an import), Lat 291 and 295-296 (both imports) from Corchiano
and surroundings; Lat 393 from Capena, Lat 432, 434, 436, 438, and 456 from Lucus
Feroniae; and 477*-478* (both imports) of unknown origin.
It may be argued that the distinction between Latino-Faliscan and Capenate on the one
hand and Latin on the other is too subtle. The distinction between the two groups is
certainly not arbitrary, however, and allows, where necessary, to distinguish between
the inscriptions that can be considered to show dialect features and those that do not.
(9) the Etruscan group (Etr, 51-72 inscriptions) comprises the inscriptions in the
Etruscan alphabet that also show Etruscan features in the morphology, phonology, or
lexicon. These inscriptions, presented separately in chapter 19, are: Etr I-XVIII from
Narce, Etr XIX from Mazzano Romano, Etr XX-XXIV from Nepi and surroundings,
Etr XXV-XXXI from Civita Castellana (Etr XXXI may be a falsum), Etr XXXII-XLI
from Corchiano and surroundings, Etr XLII from Vignanello, Etr XLIII from Rignano
Flaminio, Etr XLIV from Monte Laceto, Etr XLV from Lucus Feroniae, and Etr
XLVI-LI of unknown origin. Either Middle Faliscan or Etruscan are MF/Etr 37, 61
(virtually illegible), 64, 66-67, 77, and 199, from Civita Castellana (Falerii Veteres),
and MF/Etr 256, 264, 267, 279-280, 282, and 287 from Corchiano and surroundings.
Either Middle or Late Faliscan or Etruscan are MLF/Etr 208-209 from the wider
surroundings of Civita Castellana (Falerii Veteres) and MLF/Etr 289 from the wider
surroundings of Corchiano, and MLF/Etr 356-357 from the Gallese-Borghetto area.
Two special cases are LF/Lat 214 and Sab 468*. In LF/Lat 214, the alphabet is
Faliscan: the language, however, is Latin, without any dialect features that are specifi-
cally Faliscan, and the inscription has therefore been classed as LF/Lat. Sab 468* is of
South Etrurian and perhaps Capenate origin, but shows so many Sabellic features that it
can without problems be classed as a Sabellic, perhaps Sabine or Umbrian, inscription.
Added to the edition only to reject their connection with Faliscan are 479† (early Latin,
of unknown South Etrurian origin), 480† (early Sabellic, probably Palaeo-Umbrian,
from the La Tolfa area), 481† (perhaps Sabellic, from Foglia, near Magliano Sabino),
and 482†-484† (Latin, from Ardea).
11.1.4. The material divided according to type. The material can also be divided into
groups according to the type of inscription: sepulchral inscriptions, inscriptions on
moveable objects, dedications, and public inscriptions.
(1) Sepulchral inscriptions. By far the greatest group of Faliscan inscriptions is formed
by the sepulchral inscriptions. These are all from the chamber-tombs cut into the
relatively soft tuff of the steep rock-face of the gorges or the sides of the hollow roads
of the area. These tombs were chambers of varying size, sometimes with a decorated
368
entrance or façade on the outside, or even a porticus or antechamber. Inside the tomb,
the deceased were buried in loculi cut into the walls, which could number well over 30,
after which the loculi were closed with vertically placed roof-tiles (see below under (c)).
For descriptions of such tombs, see e.g. Ward-Perkins & Frederiksen 1957 passim, and
Colonna 1990:127-35 (short overview and typology, with clear illustrations). The
inscriptions show that the tomb or the chamber was designated with the word cela (MF
12, 83-84, MLF 285) = Latin cella, while the loculi or the places in the loculi were
designated as lete (MF 285: the word probably also occurs in MF 17 and perhaps in
MLF 361) = Latin lecti: see also §6.2.8,39.178
(a) Inscriptions on the exterior of the tomb. The first type of sepulchral inscription
encountered in these tombs consists of the inscriptions on the outside of the tomb, either
over or beside the entrance, or in the porticus. Of this type, there are 10 instances: MF
11-12, 13, 79, 83-85 from Civita Castellana (Falerii Veteres), MLF 285 and LtF 288
from the surroundings of Corchiano, and LtF 251 from the surroundings of S. Maria di
Falleri (Falerii Novi). The aim of these inscriptions was apparently to name the first or
most important owner of the tomb, or perhaps more precisely, of its burial rights. They
can consist of a name in the genitive (MF 11, 13) or of a name in the genitive followed
by the word cela ‘the tomb of ...’ (MF 12, 83-84, MLF 285), and thus correspond to the
Etruscan inscriptions with σuθi (see §8.10.3). A few simply consist of one name in the
nominative (MF 79, and probably also LtF 288, possibly also MF 85). Two contain
mention of burial rights: MLF 285 reads [---]fate cela  lete zot xxiiii ‘the tomb of ...fas:
there are 24 lecti’, while Lat 251 is even more elaborate, reading l  uecilio  uo  f  et |
po[l]ae  abelese | lectu  i  datuṣ | [. ]uecilio  l  f  et  plenese | lectu  i  amplius 
nihil | inuiteis  l  c  leuieis  l  f | et  quei  eos  parentaret | ne  anteponat ‘to Lucius
Vecilius son of Volta and to Paula Abellensis, one lectus is given; to ... Vecilius son of
Lucius and to Plenes, one lectus: let no one place anything in front against the wishes of
Lucius and Gaius Laevius sons of Lucius, and those who venerate them as ancestors’.
The inscriptions inside the tomb were placed either on the wall (27 instances) or on the
tiles that closed the loculus (181-184 instances). They had a function that was related to
but slightly different from that of the inscriptions on the outside of the tombs: obvious
though it may seem, they were there to indicate the identity of the deceased – and
hardly anything else. The tombs were family tombs, re-used for generations, and the
inscriptions had to make it clear who was buried were, with regard to burial rights, with
regard to deciding which loculi could be cleared, and perhaps with regard to ancestor-
sacrifices (cf. Lat 251, quoted above). The inscriptions therefore consist of little more
than the names of the deceased (sometimes joined by -cue), often with their filiation,
178
Peruzzi (1967a) interpreted [---]fatecela in MF 285 (see below) as [---] f atecela, which
would give a plausible word *a(n)tecela = *antecella for the porticus, but the text is in my view
to be read with Herbig (CIE 8391) as [---]fate cela ‘...fatis cella’.
369
CHAPTER 11
and sometimes with the formula hec cupa(n)t ‘lie(s) here’ added at the end. In the case
of married women not buried together with their husband, the name of the husband was
often explicitly indicated (cf. §7.4.2).
These inscriptions therefore have a very different function from the roadside
sepulchral inscriptions known e.g. from Latium. Although the Faliscan tombs were
often conspicuous, with decorated façades cut into the rock-face, the inscriptions could
be seen only by those who had business inside the tomb, and as a consequence they did
not have the function of drawing the attention of passers-by. Decoration is therefore
very scarce (occurring only in MF 80 and 89, and LF 223), carmina epigraphica are
absent,179 and the mention of honores in Middle Faliscan inscriptions is limited to MF
90 and perhaps MF 91. In the inscriptions from the period after c.240, there are cursus
honorum, some quite elaborate, in LF 242-243, 245, 247-249, LtF 231 and 233, LtF 232
and 239, and Lat 219 and 237-238, all from S. Maria di Falleri and surroundings: these
are in all probability due to Roman influence.
(b) Inscriptions on the walls of the tombs. The inscriptions on the walls of the tomb
could either be cut or painted, over, under, or beside the loculi to which they belonged.
Cut are MF 40, 47, 82, 86-87, 195-198 from Civita Castellana (Falerii Veteres), MLF
302-303 from Vignanello, and MLF 346 from the surroundings of Grotta Porciosa;
painted are MF 15-19 (the tomb of the gentes Neronia and Firmia) 48-54 (the tomb of
the gens Aufilia), 57, 80-81, and 88-89 from Civita Castellana (Falerii Veteres), and MF
347-353 (the tomb of gens Ara(n)tia) from Grotta Porciosa. Among these inscriptions,
special mention should be made of MF 17,  iii  l[---|---]naị[?---|....]o uxo, which
apparently mentions burial rights ( iii  l[---] = ‘three beds ...’ or ‘the third bed ...’), and
of MF 40, [---]o cicio  cicoi � cupat � ifra, where the usual formula hec cupat has
been adapted to make it clear that the inscription belongs to the loculus underneath.
(c) Inscriptions on the tiles covering the loculi. The majority of the sepulchral
inscriptions (181-184 instances) is made up by the sepulchral inscriptions on one to four
of the tiles that covered the loculus. These roof-tiles (tegulae180) were rectangular slabs
of fired clay with flanges running along on the long sides. In the period of the Faliscan
inscriptions, two main types of clay appear, one a hard reddish pink- to brown-firing
clay, the other a friable yellow-firing one that is very liable to crumbling and flaking.
The measures of these tiles vary from 40-48 cm in width with an average of c.45 cm
(i.e., approximately a Roman sesquipes), by a length of 60-70 cm with an average
c.68 cm (i.e., just over two Roman pedes). For descriptions of such tiles from South
179
Peruzzi (1964d:310-1) unconvincingly tried to interpret the very fragmentary MF 91 as a
carmen epigraphicum.
180
Peruzzi (1964d:310-1), in his interpretation of MF 91 as a carmen epigraphicum,
suggested interpreting the imr[ read by Gamurrini (1883:166) as im(b)r[ = Latin imbrex, but
the imbrex is not the tile, but the semi-circular cover laid over the joint between two tiles.
370
Etruria see R. Bloch 1955:56, 1976:162-3, Ostenberg 1969:103, Murrey-Threipland &

Torelli 1970:85-6, and Potter 1976:162.
The inscriptions on these tiles are mostly on the back (non-flanged) side, espe-
cially when the inscription covered more than one tile. Most were inscribed lengthwise,
i.e. with the text running between the flanges along the length of the tile: only a few
were written across. Usually, these inscriptions are painted, either directly onto the tile,
or on a layer of plaster covering the tiles.181 Of this type of inscription, there are 164-
167 instances from all periods and localities: MF 14, 39, 41-43, 56, 90-108, 136-139,
141-170, 175-194, 211 from Civita Castellana and surroundings, 265, 297-298, 305-
319, MLF 339, 358-360, 364-366, LF 220-230, 232, 234-236, 242-249, 329-337, LtF
140, 171, 173, 231-232, 233, 299-301, 325-328, 341, LtF 172, 174, 232, 239, 340, 342-
345, Lat 237-238, Etr XLIII, and probably also MF 55, 135, and 212 (where complete
descriptions are lacking). In only 17 instances, the inscriptions were scratched into the
tile rather than painted, a custom limited to the northern ager Faliscus: MF 257-258,
266, 269-272, 275-276, and Etr XXXIV-XXXV from Corchiano, MLF 324 from
Fabbrica di Roma, MLF 338 and 354-355 from Grotta Porciosa, MLF 361-362 of
unknown northern Faliscan origin. With the possible exception of MLF 361, these
scratched inscriptions all appear to have been written on one tile each. Interestingly, the
inscriptions that were scratched are often associated with clearly Etruscan linguistic
features (§9.2.3), whereas among the painted inscriptions there are few that show such
features.
A complicating factor from an epigraphic point of view is that these tiles were
sometimes re-used and can therefore have multiple inscriptions. In some cases, the tiles
were re-used for the same loculus (as in LF 222-223 and LF 224-225), but apparently
sometimes tiles were re-used for an altogether unrelated inscription. Re-use could take
the form of (a) using the other side of the tile, as in MF 136-137, 138-139, 144-145,
156-157, and 297-298; (b) washing over the titulus prior with plaster and the painting
the titulus posterior on this second layer, as in MF 90-91, LF 222-223, 224-225, and
228-229, or (c) just painting over the titulus prior, as in MLF 365-366. In the case of
LF/LtF 232-233, the tiles were apparently re-used several times, with the titulus
postumus painted on a coat of plaster that was washed over several tituli priores. This
re-use makes the inscriptions unclear, especially when the titulus prior was washed over
with a new coat of plaster: depending on the state of the plaster, it is either the titulus
prior (as in LF 228-229) or the titulus posterior (as in MF 90-91) that is illegible. To
complicate the matter even further, when the tiles were re-used for the same loculus,
they could be placed back in a different order, as in LtF 231.
181
As many tiles are preserved only as fragments, and in some cases no data are given on the
way the inscriptions were painted, the data are insufficient to specify the material accordingly:
where known to me, such data are given under the individual inscriptions.
371
CHAPTER 11
(d) Exceptional cases of sepulchral inscriptions are Lat 250, which also mentions the
consuls of 106 BCE and Lat 393, which also mentions a date.
(2) Inscriptions on movable objects: The inscriptions on movable objects are mostly
found on pottery. They fall into several categories:
(a) Signatures (19 instances). In some, the maker of the object is explicitly named as
such, as in mama z[e]xtos med f[.f]ịqod in EF 1, tele*[1-2?] med fifiked EF 9, oufilo �
clipeaio � letei � fileo � met � facet MF 470* (all three of the iscrizioni parlanti-type, for
which see §8.9.2), cauios frenaios faced MF 471*, ranazu zinaχe Etr III and c[e]ṛụr �
purφiunạs Etr LI. In other cases, the fact that the inscription is a signature is inferred
from the fact that the inscription was added during the making of the object: ac MF?
111, c ̣ cutri MF 200, pleina MF/Etr 199, vce (?) MF/Etr 256 (a terracotta strigilis), cel
Cap 386, and t  fourios  *[  ]f ̣ Lat 216 (a terracotta mould), l  quinti Lat 477*, [c ]
popili Lat 478*, c  popil[i] Lat 295, and c  popili meuanie Lat 296, and evrs  ci Etr
XXX, all written on pottery objects before they were fired; arθ[3-5]rẹ MF/Etr 267
(stamped on a bronze strigilis) and med  loucilios  feced Lat 268 (engraved on a bronze
strigilis), the latter another example of a signature of the iscrizioni parlanti-type.
(b) Besitzerinschriften (205 instances). This large group of inscriptions mainly consists
of names scratched on pottery items. Such inscriptions have always been interpreted as
Besitzerinschriften unless there are indications that they are to be interpreted otherwise.
The name can be in the nominative, as in MF 22-27, 73-75, 259-260, 371-372, 376,
473*, MF/Etr 264, MLF 463-464, LF 380, 382, Cap 388, 390 (plural) 391-392, 420,
466, 458?, Sab 468*, and Etr V, XI-XV, and XXI-XXII; in the genitive, as in MF 20,
34-36, 45, 58, 69-71, 72? (plural), 109 (or a dedication?), 201, 273, 367-370, 472*,
MF? 261-262, MF/Etr 67, MLF 304, 469*, LF 112, 379, 384 (plural), LtF 63, Cap 387,
399, 413?, 423?, 430, 465 (plural), MLF/Cap 475*, and Etr XVI, XXIII, XXXII,
XXXVII, and probably Etr XXIV; (either nominative or genitive are MF/Etr 64, 279-
280, 282, and Etr XLV), or abbreviated: MF 274, 373-375, MF? 28-30, 33, 38, 44, 46,
68, 76, 111, 131, 133-134, 203-204, 254, 255, 281, 283-284, MF/Etr 37, MF/LtF 253,
277-278, MLF 286, 320-323, 460, MLF? 474*, MLF/LtF 241, 252, LF? 381, Cap 395-
398, 400-403, 405-412, 414-419, 424-429, 439-457, 459, 461, LtF 294, and Etr VI-VII,
and possibly also Etr II. Special cases are locia eiṃoi MLF 293 and vultasi Etr XLII,
both of which appear to contain datives: see §8.8.1 Possibly also Besitzerinschriften are
MF 110, 263, MF? 78, 128-129, 130, 202, MF/Etr 62, 66, 77, 287, Cap 423, LtF 292.
A special group are the Besitzerinschriften of the iscrizione parlante-type, eco
quto *e uotenosio MF 3, eko lartos EF 6, and eko kaisiosio EF 7, aịṃiosio eqo EF 464*,
m adicio eco LF 378, eco tulie LF 383, a  írpios  esú Cap 389, and k  sares  esú Cap
404 , and ac ̣iuaiom esú Cap 465 (the last three with esú(m) = dialectal Latin esum ‘I
am’): see §8.8.2. [The area has also yielded several Etruscan Besitzerinschriften of the
iscrizioni parlanti-type: mi qutun lemausnas Etr III, cnav**es mi Etr XXVI, mi alsi*is
372
mi Etr XL, velelias mi staslar {v?} Etr XLIX, and mlakas sela aska mi eleivana Etr
XLVI and ṃi tafina lazia vilianas Etr XVII, where the type of vase is named as well
(cf. ạcṛẹẓ cat MF/Etr 67 where cat may be an abbreviation of catinus).182]
(c) Dedications. The dedicatory inscriptions (on movable objects or otherwise) are
discussed under (3). Note that in some cases the only indication that an inscription is to
be interpreted as dedicatory is the fact that it has been found in a stips or in the ruins of
a temple or sanctuary.
(d) Part of the decoration (12 instances). In several cases, the inscriptions can be
regarded as part of the decoration of the object. In this category come, first of all, the
paired inscriptions foied  uino  pipafo  cra  carefo  MF 59 and foied  uino  ‹pi›pafo 
cra  carẹ[f]o  MF 60. Other inscriptions that come into this category are the inscrip-
tions that label mythological figures, canumede [die]s pater cupi‹d›o menerua MF 62.
[Etruscan instances of such inscriptions are φerse Etr L (painted on vases), alcestei
atmite Etr XXVII, θevrumines hercle ariaθa vile menrva mine Etr XXVIII, turṃṣ tinia
apulu XXXI, aχle aivas Etr XXXIII, uslẹ*es turan acaviser setlans XLI (engraved on
mirrors), and herkle kukne Etr XXV (engraved on a gem).] More or less into the same
category falls caui � tertinei � | posticnu MLF/Cap 474*, on a bronze statuette base.
(e) More elaborate inscriptions from the Early Faliscan period. Several of the Early
Faliscan inscriptions of the early period are longer and more varied in content. Thus,
ceres far *[0-2]e[1-3]tom *[3-5]uf[1-4]ui[..]m *[3-4]*ad euios mama z[e]xtos
med f[.f]ịqod praụ[i]os urnam soc ̣[iai] p ̣orded karai eqo ụrneḷ[a ti?]tela fitaidupes
arcentelom hutị[c?]ilom pe para[i? .] douiad EF 1 contains at least a maker’s
signature (mama z[e]xtos med f[.f]ịqod) and perhaps dedicatory elements in the sense
that ceres is mentioned, but it also records that the vase was a gift (praụ[i]os urnam
soc ̣[iai] p ̣orded karai). At least partly a Besitzerinschrift is eco quto *e uotenosio titias
duenom duenas salue[to]d uoltene � MF 3. Unclear are propramom prameḍ [u]mom
pramod pramed umom pramod propramọḍ pramod umọ[m] EF 2 and tulate tulas
urate EF/Etr 385, which appear to contain word-plays, and e**azieputilepe ḳapena
rufia ḳalẹptia ues saluete sociai ofetios kaios uelos amanos salueto salues seitei
ofeteqemeneseseie EF 4. [Examples of more elaborate early Etruscan inscriptions from
the agri Faliscus and Capenas are Etr IV, VIII, IX, X, XIX, and XLVII-XLVIII.]
(f) Alphabetaries. The area has yielded two early alphabetaries that are apparently
neither Etruscan nor Latin or Faliscan (for their importance as data on the development
of the Faliscan alphabet see §11.2.2): abcdevzḥθik Etr I and abc ̣ḍevzhθ̣iḳsi*p ̣*q̣χfu Etr
XLIV. MF/Etr 110 (aie*) has been read as acev by Colonna (1990:136) and has been
classed by him and by Rix (ET Fa 9.3) as an Etruscan alphabetary.
182
Perhaps in some way related to this type are eitam EF 5 and tafina Etr XXXVI, both of
which appear to consist only of a word denoting the type of vase.
373
CHAPTER 11
(3) Dedicatory inscriptions. Dedicatory inscriptions from the periods before c.240 are
few, and all from Civita Castellana. The only clear cases are from the temples of Civita
Castellana (Falerii Veteres): apolonos EF 10 and anae lauv|cies Etr XXIX from the
Tempio Maggiore at Colle di Vignale, and the Titus Mercus-dedications MF 113-126
and the cup inscribed sacra MF 127 from the temple at Contrada Celle. Another
possible dedicatory inscription is [---]ạltai MF 109 from the Tempio Maggiore at
Colle di Vignale, and perhaps ace* (aceṿ?) MF/Etr 110, if this is an alphabetary. From
these temples are several cups inscribed with names that could be the names of
dedicants but could also be Besitzerinschriften that predated the dedication of the
object. They are LF 112 from the temples at Colle di Vignale, MF? 133-134 from the
temple at Lo Scasato, and MF? 128-131 from the temple at Sassi Caduti.
Found in tombs at Civita Castellana, but also sometimes interpreted as dedica-
tions are loifiṛtato MF 31 and loifirtato MF 32, interpreted either as the genitive of the
name of a deity Libertas or as dedications on the occasion of enfranchisement, and
apolo MF 65, which has also been interpreted as a (abbreviated?) slave-name. Perhaps
not dedicatory in the stricter sense, but apparently mentioning gods and perhaps calling
on their benevolence are parts of EF 1. Perhaps dedicatory, too, is the Etruscan
inscription mi cipaχ Etr XVIII.
The later periods yield several Latin public dedications from S. Maria di Falleri
(Falerii Novi): LF/Lat 214 to Minerva, Lat 217-218 to the Capitoline Triad, and Lat 219
to Apollo. Private dedications are from the ager Capenas: LtF 377 from Ponzano and
Cap 421 from Capena, both to an otherwise unknown Mars Numesius, and the series of
dedicatory inscriptions to Feronia from the shrine at Lucus Feroniae, Cap/Lat 431, Lat
432, Cap 433, Lat 434, Cap 435, Lat 436, Cap 437, and Lat 438, several of which
(Cap/Lat 431, Cap 435, and Lat 436) were made by freedmen and freedwomen.
(4) Inscriptions on public works. Most of what can be classed as inscriptions on public
works consists of names that are cut into the sides of the hollow roads of the area. These
are probably the names of magistrates responsible for the construction or maintenance
of these roads (cf. Ward Perkins & Frederiksen 1957:141-2), or of the surrounding
fields: Lat 291, c  egnatius  s[ex ] f  prata | faciunda  coirauit, in fact mentions the
reclaiming of pasture-land. Some mentions only one name, like MLF 207 and 210, from
the wider surroundings of Civita Castellana (Falerii Veteres), and Lat 291, and Etr
XXXVIII and XXXIX, from the surroundings of Corchiano, others two, like MLF 206
from the surroundings of Civita Castellana and LtF 290 from the surroundings of
Corchiano. Cristofani (1988:19), pointing to the care with which several of these
inscriptions are written, also attributes a propaganda value to these texts. An interesting
point is that two of these inscriptions, Etr XXXVIII and XXXIX, are written in the
Etruscan alphabet, implying that using this in a public inscription in the ager Faliscus
was, if not an everyday occurrence, at least a possibility. Unfortunately, none of these
inscriptions can be dated with any accuracy. Several other roadside inscriptions are
374
unclear, and may not in fact belong in this group: LtF 205 (abbreviations only),
MLF/Etr 208 and 209 (only two letters), from the surroundings of Civita Castellana,
MLF/Etr 289 (at least one name) from Corchiano, and MLF/Etr 356 and 357 from
between Gallese and Borghetto.
The only other clear instance of a building-inscription is [..] hirmio  m[ f ] ce 
tertineo  c  f  pret[ore(s) ?---] LF 213, which according to Garrucci (1877:199) was
written in mosaic across the entrance of a small building at S. Maria di Falleri (Falerii
Novi): this recalls the Oscan inscription Po 14, cut across the entrance to the cella of the
temple of Apollo at Pompeii. Two other possible building-inscriptions are [---] l*[---]
MF 132, the text in the terracotta frieze from the temple at Contrada Celle, which is
unfortunately too damaged to give any clue to its contents, and [---]ilio  c[  f ?---] LtF
215 from S. Maria di Falleri, which is written on a strip of bronze, a material that in the
ager Faliscus is known only from official inscriptions. A much later building-inscription
from the ager Capenas is first-century Lat 456 from Lucus Feroniae.
11.1.5. The material according to alphabet. The inscriptions from the area can also be
divided according to alphabet. Note that a distinction according to alphabet is an
epigraphic or orthographic distinction, not a linguistic one, although it can serve as such
in an ancillary role to linguistic arguments. The Faliscan alphabet is discussed in detail
in §11.2. The distinguishing features between the Faliscan and the Etruscan and Latin
alphabets are as follows:
(1) Faliscan vs. Etruscan alphabet: The Faliscan alphabet is distinguished from the
Etruscan alphabet (a) by the use of d and o in the Faliscan, but not in the Etruscan
alphabet, (b) by the use of v and χ in the Etruscan, but not in the Faliscan alphabet (θ
occasionally also occurs in the Faliscan alphabet); (c) by the differences in the shape of
the f (Faliscan  : Etruscan ), and of the r (Early Faliscan , Middle and Late Faliscan
: Etruscan �).
(2) Faliscan vs. Latin alphabet: The Faliscan alphabet is distinguished from the Latin
alphabet (a) by the use of b in the Latin, but not in the Faliscan alphabet; (b) by the use
of z in the Faliscan, but not in the Latin alphabet; (c) by the use of cu in the Faliscan,
but of qu in the Latin alphabet to render /k /; (d) by the occasional use of θ in the
Faliscan, but not in the Latin alphabet, (e) by the difference in shape of the a (Early
Faliscan or �, Middle and Late Faliscan � or : Latin ) and the f (Faliscan � :
Latin ) and (f) finally by the ductus, which is normally sinistroverse in inscriptions in
the Faliscan, but normally dextroverse in inscriptions in the Latin alphabet (although
there are exceptions to this (see below), which is why this feature is placed last).
According to these criteria, the inscriptions can be divided as follows:
(1) Inscriptions in the Faliscan alphabet: (a) dextroverse ductus (normal in the earliest
inscriptions): EF 1-4, MF? 129; (b) sinistroverse ductus: EF 6-10, EF? 4647; MF 11-20,
375
CHAPTER 11
MF 22-27, MF? 29, MF 31-32, MF? 33, MF 34-36, MF 39-43, MF? 44, MF 47-54, MF
56-60, MF 62, MF 65, MF 69-71, MF 75, MF 79-91, MF 94-103, MF 105-107, MF
113-127, MF? 130, MF 132, MF 136-139, MF 141-165, MF 167-170, MF 177-178, MF
180-185, MF 193-197, MF 200, MF 257-260, 263, 265-266, 269-276, MF? 284, 367-
372, 376, MF 470*-473*; MLF 206-207, 210-212, 285, 293, 297-298, 302, 303-324,
338-339, 346-355, 360-366, 460, 469*, MLF/Cap 476*; LF 213, 220-230, 232 (partly),
234-235, 242-245, 247-249, 329-337, 378-380, 382-384, LF/Lat 214; Etr XLIII; (c)
probably in the Faliscan alphabet too, but too fragmentary to show any distinguishing
features (all sinistroverse ductus): MF 92-93, 166, 175-176, 179, 186-192; (e) alphabet
unknown but probably Faliscan: LF 246.
(2) Inscriptions that can be read as being in the Faliscan or the Etruscan (but not the
Latin) alphabet: (a) sinistroverse ductus: EF 5, EF/Etr 385; MF? 30, 38, MF 45, MF?
46, MF 55, 72-74, 104, 108-110, MF? 111, 128, 133-134, MF 198, 201, 203-4 MF?
254-255, 261-262, 281; MF/Etr 61, 64, 77, 199, 208-209, 256, 264, 267, 279-280, 282,
356; MLF 286; MLF/Etr 356-357; LF 112, 236, LF? 381.
(3) Inscriptions in the Etruscan alphabet: (a) dextroverse ductus: Etr I-VIII, X, XIX-
XX, XLIV, XLVI-XLVII, MF/Etr 67, 256; (b) sinistroverse ductus: Etr IX, XI-XVIII,
XXI-XLII, XLV, XLVIII-XLI, MF/Etr 37, 66, 264, 267, 279, MLF/Etr 289, 357.
(4) Inscriptions that can be read as being in the Faliscan or the Latin (but not the
Etruscan) alphabet: (a) dextroverse ductus: MF? 28, 68, 76, 78, 131, 283, MF 373-
375, 470*; MF/LtF 21, 253, MLF/LtF 241; (b) sinistroverse ductus: MF? 202; MLF
463-464; MLF/Cap 474*.
(5) Inscriptions in the Latin alphabet: (a) dextroverse ductus: LtF 63, 171-174, 205,
215, 231, 232 (partly), 233, 239, 251, 277-278, 290, 292, 294, 299-301, 325-328, 340-
345, 377, and probably LtF 288; Cap 386-393, 395-412, 414-421, 424-431, 433, 435,
437, 439-455, 457-459, 461-462, 465-466, MLF/Cap 475*, and probably Cap 413 and
422-423; Lat 216-219, 435, 237-238, 240, 250, 268, 291, 295-296, 432, 434, 438, 456,
478*; (b) sinistroverse ductus: LtF 140, MF/LtF 253.
(6) other: (a) Sabellic alphabet 468*; (b) alphabet and/or ductus not reported: MF 135;
MLF 358-359; MLF/LtF 252; MLF/Cap 394; LF 246; (c) illegible: MLF/Etr? 287.
11.2. Alphabet and orthography
11.2.1. The Faliscan alphabet. From the earliest inscriptions onwards, a distinct
alphabet was used, which is found in inscriptions from the seventh century (EF 1-4)
until the mid-second century BCE (LF/Lat 214). Although, like the other alphabets of
ancient Italy, the Faliscan alphabet is derived from a West Greek alphabet, and is
therefore what was once known as a ‘red alphabet’ (after the map at the end of
376
Kirchoff 1887), this alphabet differed both from the Etruscan and from the Latin
alphabet in the letters it contained and in the shape of some of these letters, as has
already been briefly described in the previous section. From the point of ethnic
identity, this will have meant that a text written in the Faliscan alphabet may have
been a marker of this identity: it may even have been regarded as such by its users, as
may appear from its use in LF/Lat 214.
When the Faliscan inscriptions were discovered in the middle of the nineteenth
century (Garrucci 1854, 1860) there was nothing short of marvel at the new orthogra-
phy with its sinistroverse ductus and its until then unknown sign for f, the ‘arrow-f’ .
Many of the early studies were preoccupied with the alphabet, as it had several
features that were alien to the Latin and Etruscan alphabets. Interest in the alphabet
largely ceased after the establishment of the Faliscan alphabet and its orthographic
conventions in the studies of Thulin (1907) and Herbig (CIE). In the last decennia,
much study has been devoted to the development of the earliest alphabets of Central
Italy, which may shed new light on the development of the Faliscan alphabet and its
relations to the other early alphabets of the surrounding areas: in view of the scope of
this study, which is linguistic, I discuss the origin of the alphabet only briefly.
11.2.2. The origins of the Faliscan alphabet. The Faliscan alphabet, from its earliest
occurrences onwards, differed markedly from the Etruscan, and, to a lesser degree,
from the Latin alphabet, not just in the shape of the letters, but also in the letters that it
consisted of. The discussion of the origins of the Faliscan and Latin alphabets has
therefore concentrated on whether these alphabets were developed directly from a
West Greek prototype, independently from the Etruscan adaptations of the Greek
alphabet, or were derived from an early Etruscan alphabet that acted as an intermedi-
ary, and not directly from a Greek model. The differences between the Faliscan and on
the Etruscan alphabets must therefore be taken into account.
(1) The occlusive series. The (West) Greek alphabet from which the Etruscan, Latin
and Faliscan alphabets were all (ultimately) derived must had three sets of signs for
three occlusive series, i.e., the voiceless π τ κ, the voiced series β δ γ, and the voiceless
aspirated series φ θ χ. In addition, it also contained %, which must already have been a
lettre morte as the Greek dialects had by this time long lost the labiovelar occlusives.
The Etruscan, Latin, and Faliscan writing systems all adopted γ as c, which
became the regular sign for /k/ (see below). All three writing systems also adopted κ
as k and % as q, and these signs were originally similarly used to denote /k/. Three
signs for the same phoneme, however, proved to be an unsustainable degree of
redundancy: in Etruscan, q disappeared after the earliest period, in Latin k disappeared
apart from a few standardized abbreviations, while q was used only in the digram qu
to denote /k /, while in Faliscan, q disappeared entirely and k was kept for onomastic
abbreviations, and, later, to denote /g/ (§11.2.5.2).
377
CHAPTER 11
The other two voiced occlusive signs, β and δ, were not adopted in the Etruscan
alphabet, while in the Latin alphabet both signs were adopted as b and d. In the
Faliscan alphabet on the other hand, d was retained but b was not. The most important
reason for this was probably phonological: /b/ must have been one of the rarest
phonemes in Faliscan, for /b/ ← PIE */b/ was very rare due to the rarity of PIE */b/
itself, while /b/ ← PIE */bh/ was absent in Faliscan, where */bh/ developed into /f/
(§3.4). Another reason for the retention of d and the discarding of b may have been
morphological: the difference between /b/ and /p/ was morphologically irrelevant, and
one sign could therefore be used for both phonemes, whereas the difference between
/t/ and /d/ was morphologically relevant, since it formed the distinction between the
primary and secondary endings of the third singular, /-t/ and /-d/ (§5.2.4.1-2).183
The signs φ θ χ that were present in the Greek model were adopted in the
Etruscan, but not in the Latin or the Faliscan alphabet: although θ occasionally occurs
in inscriptions in the Faliscan alphabet, there is no indication that the Faliscan
alphabet retained θ as a lettre morte or an alternative to t or d (§11.2.5.3, §3.5.4).
(2) The letters u v f. Another difference between the Etruscan, Latin, and Faliscan
alphabet were the letters u, v and f. Here, the model alphabet contained a sign υ and a
sign #, and no sign for /f/. The Etruscan alphabet followed this model closely in using
υ as u, and # as v; beside these two, a separate sign was developed for /f/. Latin and
Faliscan differed from Etruscan in that they used υ for both /u/ and / /; they differed
from each other in that Latin used # as f /f/, while Faliscan had a separate sign for f,
the ‘arrow-f’ , probably developed from a variant of . This ‘arrow-f’ has been
regarded as emblematic of the Faliscan alphabet, so much so, in fact, that inscriptions
that contain this sign have been regarded as Faliscan even when the language is
clearly different (e.g. 480† and 481†; the sign has also been read in 479†). It would
appear, however, that the sign was not limited to the Faliscan alphabet: the sign
perhaps occurred already in the ‘Lower Tiber’ alphabet (see below), and may have
been in more general use in the area of the Lower Tiber basin. The Faliscan alphabet,
however, happens to be the only alphabet of which a sufficient number of documents
is preserved to show that here, at least, it was the standard shape of the f.
In view of these differences, it seems almost impossible to assume that the Faliscans
took over their alphabet from the Etruscans: as e.g. Cristofani (1972:478) concluded,
they must have formed their alphabet separately from contacts with the Greeks.
Wachter (1987:14-22), however, convincingly argues against this on the basis of
the so-called ‘C/K/Q-convention’, the orthographic184 convention found in early
183
Probably because the Latin and Faliscan alphabets retained d, the shape of the letter r in
these alphabets was never , as it was in Etruscan, but of always of the types .
184
Wachter in my view rightly assumes that this convention was purely orthographical rather
than due to a desire to render different phonetic realisations of /k/.
378
Etruscan, Latin, and Faliscan inscriptions to use c before e and i, k before a, and q before
o and u. According to him, this convention could only have arisen in Etruscan, and its
occurrence in Latin and Faliscan presupposes that it was taken over, together with the
alphabet, from Etruscan. In his treatment of this matter, the Greek model had π τ κ - β δ γ
- φ θ χ, with % as a lettre morte. These letters were all adopted by the Etruscans when
they adopted the alphabet from the Greeks. Since the Etruscan language did not require β
δ γ, these would have become additional lettres mortes. The Greek model would
probably already have had a convention of using κ before α and % before ο: in fact, the
existence of such a convention in Greek seems to be the only clear reason why % was
preserved at all in the Greek writing system. This convention was not only adopted by
the Etruscans, using k and q, but was actually expanded by using the lettre morte γ,
which now became another (and in fact completely redundant) sign for /k/, namely c.185
This step is unlikely to have occurred in a language where β δ γ did not become lettres
mortes: if the Latin and Faliscan alphabets had been derived directly from a Greek model
with β δ γ, it is very difficult to see why only β δ would have been adopted and a
redundant alternative sign c for /k/ created, while the obvious step would have been to
employ γ as g for /g/. Note also that Latin as well as Faliscan had a phoneme /k /, so that
it would likewise have been a very obvious step to use the lettre morte % to render it:
something which Latin eventually did, but Faliscan did not.
A problem with Wachter’s theory is that if the Latin and Faliscan alphabets were
to be derived from the (South) Etruscan alphabet in this way, there must have been a
stage where the Etruscan alphabet had both the series p t c/k/q and φ θ χ, while it had
not yet dropped the lettres mortes b and d. In addition to this, it must have had both u
and v as well as the lettre morte o, as well as at least two signs for s, namely s and ś
(, which became the model for the Latin and Faliscan x). In short, the earliest
Etruscan alphabet is assumed to have consisted of (at least) a e i o u - p t c/k/q - b d -
φ θ χ - m n l r - s ś z - v, containing at least three lettres mortes, namely o b d, and two
redundant letters, namely c and q.
Although this looks rather strained, there are two early alphabetaries from the
Faliscan-Capenate area that appear to point to the existence of just such an alphabet:
ET Fa 9.1=Etr I from Narce (mid-seventh century)
dextroverse: a b c d e v z ḥ? θ i k
ET Fa 9.2=Etr XLIV from Capena (seventh century)
dextroverse: ạ b c ̣ ḍ e v z h θ̣ i ḳ s i * p ̣? * q? χ f u
Neither the order of the signs nor the choice of signs is in accordance with the
conventional Etruscan alphabet. Pandolfini (in Pandolfini & Prosdocimi 1990:90-4) in
185
According to Wachter (1987:17 n.33), even this part of the C/K/Q-convention may already
have existed in the Greek model, as the oldest reported name of γ was gšmma.
379
CHAPTER 11
fact treats the second alphabetary separately as non-Etruscan, and the first as Etruscan
only because the same vase is inscribed with the Etruscan (?) word ara. Although
both alphabets appear to be incomplete and the second is damaged in the second half,
they give p? k/c/q? bd θχ (and perhaps φ): not only that, they also contain both v and
u, and in addition, an f that appears to be , an early form of the arrow-f. It is very
interesting to see that this f is added at the end of the alphabet (where new signs were
added), while the alphabet still contained v as well.
Signs similar to those appearing in these alphabetaries appear in the seventh-
century Sabellic (‘Palaeoumbrian’) inscriptions from Poggio Sommavilla, Um 2, and
Magliano Sabino, Um 3, and in the La Tolfa inscription Um 4=Sab 480† (c.530-525).
Together with the Narce and Capena alphabetaries, they point to an alphabet used in
the basin of the Lower Tiber which may have been the direct source of the Faliscan,
the Latin, and the early Sabellic alphabet, and which shows several of the features
required by Wachter’s reconstruction. Interestingly, the form of the f differs in these
three alphabets: Latin has a digram (perhaps indicating that this alphabet was
adapted at an earlier stage than the other two, when there was not yet a separate sign
for f), later switching to , while Faliscan has from the earliest inscriptions onwards
and shows no traces of ever having had a digram. The early Sabellic inscriptions,
however, show in Um 4=Sab 480†, but (Etruscan?) in Um 2. If the sign
remained in use in the not epigraphically attested Sabellic languages along the Lower
Tiber, this could explain its occurrence in the (much later) inscription Lat (?) 481†
from Foglia, near Magliano Sabino.
In view of the subject of this study, the importance of this section on the origin of the
Faliscan alphabet is relatively small. Yet the origins and adaptations of the early
alphabets of the Lower Tiber basin show that the ager Faliscus was independent
enough from the Etruscan cultural-linguistic influence to develop an alphabet of its
own, and was likewise independent enough from the remainder of the Latin-speaking
area to have done so separately.
11.2.3. Alphabet and orthography of the Early Faliscan inscriptions. The base
Faliscan alphabet of the archaic period therefore consisted of the letters a c d e f h i k l
m n o p q r s t u x z. 186 For the shape of these letters in the Early Faliscan inscrip-
tions, see fig.11.1. B and θ do not occur in the Early Faliscan texts, and, although they
occasionally occur in the Middle and Late Faliscan texts (§11.2.4, §11.2.5.3), I think
this was due to influence from other orthographic traditions, not because they were
lettres mortes in the Faliscan alphabet.
186
Note that I give the letters in the order of a modern Latin alphabet: there are no Faliscan
alphabetaries that show us whether Faliscan followed the Etruscan or the Latin alphabetical
order, or an order of its own.
380
The Early Faliscan texts show the following orthographic features:

(1) Ductus. The ductus is dextroverse in the oldest inscriptions, but quite early on
(already by the end of the sixth century) it changes to sinistroverse (see fig.11.1), as in
the Etruscan inscriptions. The Latin alphabet made the reverse shift more or less
during the same period.
(2) The C/K/Q-convention. In EF 1 this convention is observed: this inscription has
ceres, soc ̣[iai], arcentelom, karai, as well as f[.f]ịqod and eqo where q in all probabil-
ity represents /g/. Similarly EF 467* has eqo. In EF 3, 4, and 7, the convention is
partially observed: EF 3 has quto, but also eco; EF 4 has sociai, ḳapena, ḳalep̣tia, and
kaios, but also qe in seiteiofeteqemeneses*eie; EF 7 has both kaisiosio and eko. The
convention is not observed in EF 6, which has eko, and in EF 9, which has fifiked. It is
perhaps significant that in most cases where the C/K/Q-convention is not observed,
this involves cases where the phoneme rendered is not /k/, but /g/. Apart from the
unintelligible seiteiofeteqemeneses*eie, all the ‘deviations’ are cases where /g/ has to
be represented: eco EF 3, eko EF 6, 7, fifiked EF 9. This may be part of an early
tendency to represent /g/ in some way. In the Middle and Late Faliscan inscriptions,
there are occasional tendencies to use k for /g/ (§11.2.5.2).
(3) Double letters. There is no indication that the double consonants or long vowels
were expressed in writing. (Note that there are no words where this may be expected.)
(4) Word-division. The use of interpunction in this period is irregular. EF/Etr 5, EF 8,
and EF 10 consist of one word only; EF 3, 4, 6, 7, 9, and 467* consist of two or more
words, but have no interpunction at all, although in EF 3 a double interpunct � was
added to divide the beginning and the end of this circular inscription. EF 1 and EF 2
use a triple interpunct , but there appears to be no recognisable consistency in its use:
ceres 1 far *[0-2]e[1-3]tom 2 *[3-5]uf[1-4]ui[..]m 3 *[3-4]*ad euios 4 mama
z[e]xtos med f[.f]ịqod 5 praụ[i]os urnam 6 soc ̣[iai] p ̣orded karai 7 eqo ụrneḷ[a
ti?]tela fitaidupes 8 arcentelom hutị[.]ilom 9 pe 10 para[i? .] douiad
Interpunct 1 appears to separate the nomen divinum from the rest of the text, while
interpuncts 4-5 and 5-7 appear to enclose sentences. Interpunct 9 on the other hand
appears to separate the main verb from the rest of the sentence (?), while interpunct 10
appears to have been used to separate the reduplicative syllable from the root of the
verb (cf. vhe vhaked CIL I2.3). The aim of interpuncts 3, 6, and 8 is unclear. The
interpunction in this inscription has also been explained as indicating metric cola
(Radke 1994:106-8), but I do not find this convincing. The use of the interpunct in EF 3,
propramom 1 prameḍ [u]mom pramod pramed umom 2 pramod propramọḍ 3 pramod
umọ[m] is unclear.
(4) The use of z. For the use of z in z[e]xtos EF 1, which may or may not render a
specific allophone of /s/, see §3.5.3.
381
CHAPTER 11
dextroverse ductus sinistroverse ductus
1 2 3 4 5 6 7 8 9 10 467* 385
A � � �
B ?
D � �
E � �
F   
I � � � � � � �
L    � � � �
M �
N �
O 
X 
Fig.11.1. The alphabets of the Early Faliscan inscriptions.
382
11.2.4. The alphabet of the Middle and Late Faliscan inscriptions. Between the
Early Faliscan and the Middle Faliscan periods, the Faliscan alphabet undergoes
several changes. Unfortunately, due to the lack of material from the fifth and early
fourth centuries, the process of these changes itself cannot be documented.
(1) Ductus. The most obvious change is the change in ductus. Already during the
Early Faliscan period there appears to be a change from dextroverse to sinistroverse,
and in the Middle Faliscan inscriptions the normal ductus is sinistroverse. There are a
few exceptions to this, most notably the late fourth-century MF 62, where the
dextroverse ductus is regarded as an archaism by Wachter (1987:367-9). Since there
change to sinistroverse ductus started in the sixth century, I doubt whether this is
possible: it may be that the ductus ‘fluctuated’. Several other inscriptions of the
Middle and Late Faliscan periods also show dextroverse ductus, most notably MF?
129 (in the Faliscan alphabet). Other examples are MF? 28, 68, 76, 78, 131, 283, MF
373-375, MF/LtF 21, 253, MLF/LtF 241, and MLF/Cap 474*; all written in what can
be either the Faliscan or the Latin alphabet, although the assumption that the inscription
might be in the Latin alphabet is in several cases based on the dextroverse ductus.
(2) Shapes of the letters. When compared with the Early Faliscan inscriptions, a few
letters have quite different shapes in the Middle and Late Faliscan inscriptions, as is
noted below. The general shape of the Faliscan letters appears to be more rounded,
and not just in the inscriptions that are painted: those that are scratched in pottery
items or cut into the rock, too, seem to emulate more rounded forms, perhaps imply-
ing that the normal way of writing was now the pen rather than the stylus (cf. Cencetti
1957:188). One inscription, LtF 140, in fact very much gives the impression of having
been written with a reed-pen rather than having been painted with a brush. With
regard to the shapes of the individual letters, the following can be said:
A. (The basic shape of the a evolved from Early Faliscan and �, to Middle Faliscan
�, usually written with a slightly convex left side as . This in turn led to the variant
, with the transverse bar sticking out to the left: a very common, if not the com-
monest shape in the Middle Faliscan inscriptions. As this could also be written as ,
the distinction between this letter and the r, , became less obvious, and there are
several inscriptions where and were confused (e.g. in MF 57). In a few inscrip-
tions, such as MF 101, a new type of a appears, , apparently not so much a correc-
tion of r to a but an independent variant, drawn, like , in two strokes, with a
connecting bar. Interestingly, this sign, too, is used as r in MF 59-60, beside the
normal . (2) A second development that affects the a is the occurrence of cursive
forms like , , , , and . Although these have been ascribed to Latin influence,
they appear in inscriptions that show no other signs of Latin influence, and Cencetti
(1957:195-8) has shown that some of these forms may have originated at a very early
date, when direct Latin influence in the ager Faliscus cannot have been very great. If
383
CHAPTER 11
these forms did indeed develop within Faliscan, the development of the Faliscan a in
fact provides another indication for their early date, since they can be derived much
easier from the original � than from . MF 79 and MF 111 in fact show a peculiar
form , which may be an early Faliscan cursive form of the a derived from .
B. As in the Early Faliscan inscriptions, there is no indication of the presence of b in
the Faliscan writing system, with the exception of tito batio MLF 359. Unfortu-
nately, this inscription is very badly documented, and there are no reports of the shape
of this b (or even whether the inscription is written in the Faliscan alphabet at all). On
this basis, it cannot be assumed that Faliscan had preserved b as a lettre morte. The
use of b was apparently so rare that even in the Latin inscription Lat 219 from
between c.120-50, the name Umbricius is spelled as umpricius, perhaps a deliberate
archaism recalling the original Faliscan spelling of the name.
C. The C/K/Q-convention of the Early Faliscan inscriptions having disappeared, c is
the standard sign both for the voiceless dorsal occlusive /k/ and the voiced dorsal
occlusive /g/, and is used in the digram cu for the voiceless labiovelar occlusive /k /
(§11.2.5.1). The shape is always , although in some scratched graffiti it is .
D. No specific developments or features. The shape is � (see also under Θ).
E. The normal e in the Middle and Late Faliscan inscriptions is either (as it had
already been in the Early Faliscan inscriptions) or �. A few inscriptions have , which
could be a simple error were it not for the fact that it recurs several times (in MF/Etr
64, MF 258, MLF 285). Beside these shapes, however, there are three other types of e
that are ‘cursive’. (1) The most numerous of these variants is the cursive e also known
from Latin inscriptions, �� (see Cencetti 1957:190-3). (2) A probably Etruscan form
occurs in MF/Etr 267 and Etr XXXIV (and also in Etr XLV?). Peruzzi (1964c:228)
suggested that this may have formed the basis for ��. (3) A very rare form appears
in MF 146 (which also has an h of the type ) and Lat 483† from Ardea, which on
these grounds has been regarded as Faliscan.
F. The sign for f in the alphabet of the Middle and Late Faliscan inscriptions is always
the ‘arrow-f’, . This letter is in fact one of the distinguishing features between the
Faliscan alphabet (where the f was ), the Etruscan alphabet (where it was ), and the
Latin alphabet (where it was ). The Etruscan type only occurs in Etruscan inscrip-
tions, the Latin only in Latin inscriptions.
[G. Like the alphabet of the Early Faliscan inscriptions, the alphabet of the Middle
Faliscan inscriptions shows no separate sign to render /g/, although there are several
inscriptions where k is used in this way (see under K). Two points of note with regard
to g are adduced by Girard (1989:169): (1) the curious spellings gonlegium, uolgani,
gondecorant in Lat 217, explained by him as due to a Faliscan struggling with the
correct use of an unfamiliar sign (which disregards the fact that in the Faliscan
384
alphabet /g/ could in fact be rendered by k), and (2) the fact that the introduction of g
in the Latin alphabet was ascribed by Plutarch (Quaest. 54) to the same Sp. Caruilius
Ruga who subjugated Falerii in 293 (§2.5.2).]
H. The normal shape of the h in the Middle and Late Faliscan inscriptions is . There
are a few variants, however: (1) some inscriptions show � or , probably a simplifica-
tion of (or perhaps inspired by the shape of the Latin ). It would be surprising if
this sign was much used, however, as it had the same shape as the Faliscan sign for n.
(2) A different (cursive?) h, , is found in MF 146: this is also the only Faliscan
inscription that has an e of the type (see under E).
I. In the Middle and Late Faliscan inscriptions, the sign for i is �, as it had been in the
Early Faliscan inscriptions.
K. The k returns in several inscriptions, now always as and no longer as . Since the
c/k/q-convention has disappeared, its use is now special: see §11.2.5.2.
L. The sign for l is � keeps its old form, although � appears from time to time.
M. The basic shape of the m in the Faliscan alphabet is ��, as opposed to the and
of the alphabet of the Early Faliscan inscriptions. The shape �, although the usual
shape in the contemporary Etruscan inscriptions, is quite rare, occurring in fact only in
MF 269 and MF 272 from Corchiano. Both these inscriptions also show other
Etruscan features: see §9.2.3c-d.
N. The shape of n in the Middle and Late Faliscan inscriptions is �, as opposed to the
or of the Early Faliscan inscriptions. This � is always written upright: the slanting
 only occurs in Latin inscriptions from the area.
O. The letter o is often painted or written in two strokes as , which could be called
‘cursive’. Variants where the o is open at the bottom or the top are therefore found. In
inscriptions that are scratched or cut into the rock, the o is often diamond-shaped or
polygonal. Cencetti (1957:189) regarded this as a distinct cursive type, which is well
possible especially in the more open variants.
P. The normal form of p is , beside rare occurrences of � and .
[Q. The Faliscan alphabet as used in the Middle and Late Faliscan inscriptions did not
contain a q. (As was the case also in the contemporary Etruscan alphabet, and in a
sense also in the Latin alphabet where the q was restricted to the digram qu. The
Faliscan orthography used cu, not qu, for /k /.]
R. In Middle Faliscan, the shape of r is , one of the diffferences between the Faliscan
and the Etruscan alphabet (see note 183). The shape is very similar to that of a, which
was or , and the two signs are sometimes confused (see under a). For the errone-
ous notion that in MF 59-60 represents [z] (Sittig 1932, Belardi 1964), see §3.5.3.
385
CHAPTER 11
S. The s of the middle Faliscan alphabet is always � . Apparent instances of in

Middle and Late Faliscan inscriptions are in every case due to accidents or careless
writing: there is no indication that the sign continued to be used. A very common
feature of Faliscan inscriptions is the reversal of the s to : this in fact occurs so
frequently that I do not enumerate the instances here, but only indicate it under the
individual inscriptions. See also under Z.
T. The sign for t has three forms that all occur frequently, namely , , and . There
does not appear to be a chronology in their use.
Θ. The instances of θ in Faliscan inscriptions are few, and probably due to ortho-
graphic influence from Etruscan. (For the discussion of the possibility that θ rendered
an allophone of /t/, see §3.5.4.) The sign is usually with a central point, probably to
avoid confusion with o (note that in the Etruscan inscriptions from the area, where this
confusion could not arise, the sign usually appears as ), as in θania MF 81, uesθi
MF 83, salθan MF/Etr 77, and arθ[3-5]rẹ MF/Etr 267, but without central point in
uolθeo MF 276: I have suggested that this shape also appears in hạθ̣i MF 13. Three
instances, known only from apographs, are unclear: θanacuil MF 49 (θ given both as
� and as ), [---]nθia MLF 212 (θ given as ), senθia MLF 362 (θ given as ). In
two inscriptions (MF 49 and MF 276) the shape of θ resembles that of d.
U. As in the Early Faliscan inscriptions, the sign for u in the Middle Faliscan alphabet
is . In careless writing, the two strokes may become separated, or they may be
written too close together, so that confusion with x sometimes arises.
Z. The z is found in several inscriptions. Like s, it can be reversed. It is unclear
whether or not the use of z denotes a different sound than [s], or whether the variation
is merely orthographic: see §3.5.3.
11.2.5. Orthographic conventions in the Middle and Late Faliscan inscriptions. In

the Middle and Late Faliscan inscriptions, the following orthographic conventions can
be observed:
(1) The use of cu. In the Middle and Late Faliscan periods, cu is used to render the
labiovelar occlusive: -cue MF 80, -c ̣ue MF 158, -cuẹ MF 170, -cue MLF 313; cuicto
MLF 310 (and perhaps cuitenet MLF 361); (3) in cụestod LF 242, cues[tor] LF 243,
c]ues[tor LF 245, cue[stor LF 247; θanacuil MF 49, tanacu[il] MF 101, θancuil
MLF 347; cua MF 129.
(2) The use of k. Whereas in the Early Faliscan inscriptions k was used in the C/K/Q-
convention, its use in the Middle and Late Faliscan inscriptions appears to have been
twofold:
(a) k was used to render /g/, as in kreco MF 147 = ‘Graecus’, keset LF 242 and kese[t
LF 243 = ‘gessit’, and Ekn in LF 246, most likely a form of the name Egnati-.
386
(b) k was also used in the name ḳai[s]i[o MF 51, and as an abbreviation of a
praenomen k Cap 388, 390, 404, Lat 218, and perhaps Cap 403. In both these cases
the use of k is quite clearly a specific convention, the use of a sign that has no normal
function as a letter.
(c) The use of k in nuikuẹ MF? 202 and [---]*a*kit*ue*a LF 234 is unclear.
(3) The use of θ. Although θ does not appear to have been a letter of the Faliscan
alphabet, it occurs in several inscriptions: hạθ̣i (?) MF 42, θanacuil MF 49, θania MF
81, uesθi MF 83, salθan MF/Etr 77, [---]nθia MLF 212, arθ[3-5]rẹ MF/Etr 267,
uolθeo MF 276, and senθia MLF 362. I assume that this is an orthographic variation
only: see, however, §3.5.4.
(4) The use of z. In a number of cases, z is used instead of s: word-initially in zextos
EF 1 and zextoi LF 330; zot MLF 285; zenatuo LF/Lat 214; in names: zaconiọ MF
153 and zaconiai MF 154; zuconia MF 271 and perhaps zu[con]|eo MF 56;
zeruatronia MF 272; word-internally in zertenea LF 221; fulczeo LF 329, folcozeo LF
330, and *olcuzeo LF 332 vs. folcuso LF 331 and folcosio LF 333; and word-finally in
aruz MF 257, morenez MF 269, and perhaps ạcṛẹẓ MF/Etr 67. There is a possibility
that z represents [z] in at least some of these cases, although I tend to regard most of
them as influenced by Etruscan orthography: this is discussed in §3.5.3.
(5) Doubling of vowel or consonant signs. The doubling of signs to express a long
vowel or a long or double consonant is exceedingly rare. Doubling of vowels is in fact
not attested for inscriptions in the Faliscan alphabet except for LF/Lat 214 (c.150?)
which has uootum: this case can be ascribed to the orthographic influence of contem-
porary Latin. Doubling of consonants likewise appears to be connected with the Latin
rather than the Faliscan alphabet, in cases such as anni LtF 63: the only exceptions are
four cases of doubling of l in uolḷia MF 47, uolḷ[---] MF 86, putellio MF 156, and
lullio MLF 207. R. Giacomelli (2006:91-3) has suggested that this may represent
palatalization: see §3.5.5.3.
(6) Interpunction. Word-division in the Middle and Late Faliscan inscriptions is by �
or , and a few inscriptions also use ‘stroke-interpuncts’ and (i.e., short or long
vertical stroke used as an interpunct). Interpunction is not always used consistently:
several types of interpunction may appear in one text, and interpunction may be used
after one word but omitted after another within the same text. At the end of a line
interpunction is usually omitted.
(7) Line ends. The Middle and Late Faliscan texts appear to avoid breaking off words
at the end of the line, preferring either to start the next word on a new line, or to write
the last letters downward, above or below the line as dictated by the available space.
Words divided over more than one line are found in larise � mar||cna � citiai MF 270
(where the text then continues on the line above it), tito � uel|mineo � iun|aị ị*ice MLF
387
CHAPTER 11
315 and popli[o] | uelmi|no MLF 316, and probably also in MF 56, LF 243, and LtF
231. An actual hyphen has erroneously been read in MF 336.
(8) Graphic contraction. There are a few instances of graphic contraction in the
Middle Faliscan inscriptions. Most of these involve u, which could stand for u(o):
ulties MF/Etr 64 (= u(o)lties?), tuconu MF 85 (= t u(e)conu?), uli MF? 261-262 (=
u(e)li?) perhaps also mar||cna = marc(e)na MF 270 and fulczeo = fulc(o)zeo LF 329,
if these are not a syncopated forms or simply errors. From the ager Capenas are pscni
= p(e)sc(e)ni Cap 387 and fertrio = fert(o)rio Cap 391.
(9) Reversed letters indicating women’s names. In a few inscriptions, women’s
names are marked by reversing the initial letter. The instances are: a � u[eculi]a | ca �
e[c]ṇata � θania MF 81, ca  uecineo | a  mania LF 225, tito [�] acarcelinio : | ma � fi 
op  etrunes  ce  f | [h]e cu[pa] LF 226, ola marcia : sus[?---] LF 227, [---]rcius 
�  l | [fer]oneae | [l] m Cap 436, and a  e**sa Cap 458. Note that with the exception
of MF 81, this feature appears to be associated with the Late Faliscan period and with
the ager Capenas: it may therefore be due to Latin influence.
11.3. The use of the Etruscan and Latin alphabets
In the ager Faliscus and Capenas, the Faliscan alphabet was not the only one in use
(cf. §11.1.5): from the earliest period onward, inscriptions written in the Etruscan
alphabet occur in the area, and it is likely that this was also the first writing system in
the area (cf. §11.2.2).
However, although there are occasional indications of Etruscan orthographic
interference in the use of θ and z in texts written in the Faliscan alphabet, on the
whole both writing systems not only remained quite distinct, but they also appear to
have been quite firmly associated with the languages for which they were developed.
Although this cannot be used as an a priori, a study of the documents in this edition
shows that where both the alphabet and the language of the inscription can be clearly
distinguished, texts that show Faliscan phonological, morphological and lexical
features are virtually always written in the Faliscan alphabet, using Faliscan ortho-
graphic conventions, while the texts that show Etruscan phonological, morphological
and lexical featureas are virtually always written in the Etruscan alphabet, using
Etruscan orthographic conventions. There are very few clear exceptions to this, the
clearest being umrie Etr XLIII, where the alphabet is Faliscan but the phonology and
the morphology Etruscan. However, this is an exception to the rule. The texts show
that these two writing systems were quite firmly linked to their respective languages,
more, perhaps, than would be expected in an area where contacts between speaker and
writers of both languages must have been frequent (§9.2.1).
388
The situation is different where the Latin alphabet is concerned. The Latin
writing system is not present from the earliest date, and neither can the frequent
occurrence of cursive letters in Middle Faliscan inscriptions be counted as an
indication of its presence in the area: these may have arisen independently within
other orthographic traditions, as Cencetti (1957, cf. pp.190-2 on Faliscan) suggested.
Exactly when the Latin writing system came to be used in the ager Faliscus is very
hard to establish (as opposed to assuming that its introductions took place after, and as
a consequence of, the war of 241). The first inscription in the Latin alphabet that can
be dated is med  loucilios  feced Lat 268 on a fourth-century bronze strigilis found at
Corchiano, but this was probably an import. Several Latin inscriptions appear on
imports, and cannot be used to document the introduction of the Latin writing system,
apart from the assumption that people were able to read these texts. The inscriptions
in the Latin alphabet that were written within the ager Faliscus, i.e. sepulchral
inscriptions and roadside inscriptions, cannot be dated with certainty. What, for
instance, are the implications of LtF 140 and 171-174, sepulchral inscriptions written
in the Latin alphabet found at Civita Castellana, a town assumed to have been
(virtually?) abandoned after the war of 241-240? Do these inscriptions imply that the
Latin writing system was used at Civita Castellana before c.240, or, conversely, that
burials around the town continued in the period after c.240, when the introduction of
the Latin alphabet can more easily be imagined? In any case, the Latin writing system
was present in the new Roman Falerii (S. Maria di Falleri), where it was used by a
craftsman signing his work (LtF 216), in public dedications (LtF 217-218), and in
several undated sepulchral inscriptions (LtF 231-233). Exactly when the Latin writing
system completely ousted the Faliscan one is a question that cannot be answered. The
last datable inscription written in the Faliscan alphabet appears to be LF/LtF 214,
which is usually dated to c.150 BCE, but in this inscription the use of the Faliscan
alphabet may already have been an archaism.
11.4. A note on the presentation of the inscriptions
As has been said (§1.1), the present study is a linguistic one. The aim of the edition is
therefore (a) to facilitate access to the material on which the linguistic discussions and
conclusions are based, and (b) to justify and discuss the readings and interpretations
that I have used. The aim has not been to present a fully epigraphic edition. For the
commentary, this means that the discussion usually focuses on what can be read and
what not and what can be used as data and what not.
The edition includes all inscriptions from the ager Faliscus and the ager Cap-
enas that consist of more than one letter, whether in the Faliscan, Etruscan, or Latin
alphabet, and whether assumed to be in the Faliscan, Etruscan or Latin language, from
389
CHAPTER 11
the earliest documents to the early first century BCE, as well as a number of inscrip-
tions of unknown or uncertain provenance that have been regarded as Faliscan. For
the ordering of the inscriptions according to provenance, see §11.1.2; for the division
in ‘period/alphabet groups’, see §11.1.3; for a division according to type of inscrip-
tion, see §11.1.4; for a division according to alphabet, see §11.1.5.
The organization of the presentation in each case consists of the following five
elements (sometimes very briefly, depending on the available data):
(1) Introduction. For each locality, and, in the cases of Civita Castellana and
Corchiano, for each site, a brief overview of the location is given, with references to
its excavation history where this is relevant. As explained in §1.4.5, in some cases the
excavation history and the provenance of the inscriptions may not be beyond doubt.
(2) Description of the object. Each lemma starts with a description of the object and
the way it has been written, where possible with measurements. Note the following:
(a) in the case of tiles, the front is the flanged side and the back the non-flanged side
(cf. §11.1.4.1c). The length is given first, then the width, irrespective of whether the
tile is inscribed across or lengthwise. Across means that the inscription is written from
one flanged side to the other; lengthwise, that it is written between the flanges.
(b) in the case of tile fragments, the maximum height and width are given. If preceded
by the word ‘total’, measures are taken across several adjoining fragments.
(c) in the case of pottery or pottery fragments, the measures given are height and
diameter (), the latter measured at the rim unless indicated otherwise.
(d) in the case of inscriptions painted on or cut in a rock-face, the height and length
given are those of the inscribed surface.
These descriptions are followed, where possible, by an approximate dating, and
remarks on the provenance, history, or authenticity of the item; if an inscription is
known only through apographs, this is also noted here.
(3) Text of the inscription, using the following signs and conventions as given in
Conventions in the representation of epigraphic texts (p.LII).
(4) Discussion of the reading of the text and/or its interpretation. As I said, the aim of
this discussion is to establish the reading of the text and, where possible, enough of its
interpretation to use the text as data in the linguistic discussions in part I (chapters 2-
10). Note that when rendering letters of the inscription in the text, the use of printable
symbols indicates that the letter is of a specific but recognizable type (e.g. , �, or
for a), while the use of a drawing indicates either (a) that the letter is either of a
unique shape or variant, or (b) that I intend to render the drawing made by a specific
editor of the text.
390
(5) References. Each inscription is followed by a bibliography that at least contains

all editions, as well as publications where the object, the inscription, or its contents are
discussed or referred to in a way that is considered relevant. In the reference section,
the following elements and symbols are used:
A large number of inscriptions are published from autopsy. The year of the autopsy is
given, followed by an abbreviation of the museum where the autopsy was done and
the number or numbers of the inscribed objects in the inventory.
Bibliography: Bibliographical references are presented in chronological order. When
at a certain point the data were substantially altered (e.g. by the discovery of a new
part of the text), the bibliography has been divided into sections numbered (I), (II),
etc. References to publications that I was unable to consult are preceded by †: if I had
indirect access to these publications, e.g. because they are quoted or discussed by
other authors, this is explained in the text. References between square brackets [ ]
refer to publications where the inscription is mentioned but no text is given (e.g. in
archaeological discussions of the inscribed object or in catalogues). The word
(autopsy) following a reference indicates that the publication is based on an autopsy
by the author: later publications by the same author are so marked only if a new
autopsy took place, as explained in the text. Numbers between pointed brackets ¢ ²
following a reference represent the number given to the text in that edition.
Since illustrations have been kept to a minimum, references are given to all published
photographs or drawings that I was able to find. Only if to my knowledge no
photographs or drawings have been published are references made to transcriptions,
that is, reproductions of the text in appropriate font types, popular especially among
earlier editors. If an illustration was reproduced in a later edition, this is referred to as
(reproduced in ...) following the reference to the original. In cases where authors
have used a common source, such as the archive of the Soprintendenza, I have used
the sign = to indicate that the photographs in these publications are identical.
391
CHAPTER 11
392
Chapter 12
The Early Faliscan inscriptions
12.1. The Early Faliscan inscriptions as a group
The Early Faliscan inscriptions constitute a separate group with regard both to the
contents of the texts and to their linguistic features. All are inscriptions on pottery, but
several (EF 1-4) are longer than and different from the usual pottery-inscriptions (cf.
§11.1.4.2e). The linguistic features of the Early Faliscan inscriptions are also different,
notably the second-declension genitive singular in -osio (§4.4.2). For the alphabet of the
Early Faliscan inscriptions, see §11.2.3.
The Early Faliscan inscriptions in this chapter, EF 1-4, 6-10, and EF/Etr 5 are all
from Civita Castellana. Two other inscriptions that I regard as Early Faliscan are
seventh-century (?) tulate tulas urate EF/Etr 385 from Fiano Romano (§17.6) and sixth-
century aịṃiosio eqo 467* (§18.2) of unknown provenance. Inscriptions that have at
some time been regarded as Early Faliscan but that I regard rather as Latin are the
Vendia-inscription eco urna tita uendias mamaṛ[cos 6-9 m]ẹḍ ṿḥe[ked] Lat 479†
(§18.3.1) of unknown provenance, and eqo kaṇaios Lat 482† from Ardea (§18.3.2).
Sabellic rather than Faliscan are setums míom | face Sab 480† (§18.3.1) and pa‹qu›is
blaisiís Sab 468* (§18.3.2), both of uncertain South Etrurian origin: see also §9.3.2. For
the contemporary Etruscan inscriptions from the ager Faliscus and Capenas (Etr I-
XXVI, XXXII, XLII, XLIV, and XLVI-XLIX), most of which are from Narce, see
Chapter 19.
12.2. The ‘Ceres-inscription’
1. Scratched c.2¼ times around the shoulder of a reddish-black impasto urn ( shoulder
32 cm, lip base 7 cm) decorated with two winged horses, from tomb LXII/3 of the Le
Colonnette necropolis (Mengarelli in Thulin 1908:255, FI II.2 p.208).187 The shoulder
fragments appear to be the only part of the vase that is preserved (or found?).
187
Gamurrini gave the provenance as “la necropoli di Civita Castellana”, Della Seta named the
necropolis without specifying the tomb. Nogara (in a letter of 14.X.1927, quoted in Stolte
1929:111) stated that he had not found details of its discovery in the inventory of the Museo di
Villa Giulia (presumably in 1903, when he did his autopsy). According to Morandi (1982:54),
the Museo di Villa Giulia acquired the vase in 1891 from the Collezione Feroldi.
393
CHAPTER 12
Gamurrini dated the vase and the inscription on technical and palaeographical
grounds to the sixth century, and this date is accepted by most authors, although
Giglioli not unconvincingly dated the vase on technical grounds to the seventh century:
earlier datings had in fact already been considered by Thulin and Della Seta. Bruhl and
Norden dated the vase rather later, without giving any arguments for this; Safarewicz
(1953:245-6) regarded the text as later than CIL I2.1, but this is based on the readings
farme[n]tom, l[o]uf[ir, and ui[no]m that are doubtful or impossible.
Fig.12.1. Mengarelli’s reconstruction of EF 1, and Herbig’s adjustments.

left: Mengarelli’s drawing of his reconstruction. The shards with ]ịqod praụ[ and ]ilom pe para[ are
drawn more to the left than they are placed in the actual reconstruction. The drawing probably gives their
correct position: note that this affects the lacunas above and below it. (From Herbig CIE 8079)
right: Herbig’s adjustments and restorations. In his drawing, the shards with ]ịqod praụ[ and ]ilom pe
para[ are definitely placed too far to the left. The shards with osurnam soc ̣[ and, underneath it, ]douiad [,
are rightly tilted upwards with regard to how they are placed in the reconstruction. Joining them to the
shard with ]tom *[ in the first line would seem to be impossible, however. (From Herbig CIE 8079)
The vase, found in shards, was reconstructed by Mengarelli and Malvolta in 1907.
There is no need to question the overall correctness of this reconstruction: virtually
identical reconstructions had in fact been made independently by Gamurrini in 1894, by
Nogara in 1903, and by Thulin in 1906. However, there is an unfortunate tendency
among editors of this text to overlook the fact that at several points the reconstruction
was explicitly meant as provisory, as Mengarelli (1908:103) himself explained.
Especially the first part of the text is therefore less certain than it may appear, as will be
explained below.
394
THE EARLY FALISCAN INSCRIPTIONS
However, I have three general remarks with regard to the reconstruction:

(1) The circumference of the vase may have been very slightly larger than it is in the
reconstruction. Since it is hard to establish this without taking the reconstruction apart
or making a virtual model of the shards, I have not pursued this here.
(2) The group of shards with the letters ]ịqod praụ[ and, underneath, ]ilom pe para[,
does not fit exactly onto the shard with ceresfar*, and is probably to be moved slightly
to the left and down (cf. figs.12.1-2).
(3) The shards with the letters ]osurnam soc ̣[ and, underneath it, ]douiad [, should be
moved slightly upwards and tilted c.5-10° counter-clockwise (cf. Herbig’s drawing in
fig.12.1).
Also, several of the lacunas are interdependent, as the shards nowhere join up to form a
complete circumference of the vase. This applies especially to the lacunas between
ceres far*[0-2]e[ and ]tom *[, between ]tom *[ and ]uf[, and between ]uf[ and
]ui[..]m . In each of these cases, every letter that is restored above the minimum amount
possible is to be subtracted from the maximum amount possible for one of the others: in
all, a total of 8-10 letters appears to be missing. Bearing this in mind, I read the text as:
ceres far*[0-2]e[1-3]tom *[3-5]uf[1-4]ui[..]m *[3-4]*adeuios mamaz[e]xtosmed
f[.f]ịqod praụ[i]osurnam soc ̣[iai]p ̣ordedkarai eqoụrneḷ[ati?]telafitaidupes arcen
telomhutị[c?]ilom pe para[i?1]douiad [
Dextroverse, and spiralling downward: the text bends downwards after z[e]xt and after
arcente to pass underneath the turns above it. S is , but in praụ[i]os, soc ̣[iai], and
fitaidupes, a variation found also in the Etruscan inscriptions Etr IV from Narce and
Etr XLVI of unknown provenance; h is , a variant of . The interpunct is triple:
there appears to be no discernable consistency in its usage. G. Giacomelli (1963:41-2)
would rather regard it as a punctuation of sorts, which seems attractive, but in that case
its word-internal use in pe para[i is awkward. Radke (1994:106-7) implausibly
suggests that they divide the text into Saturnian cola. Individual letters are discussed
below: see also §11.2.3 for a general discussion of the Early Faliscan alphabet and
other orthographical features.
As the inscription is usually divided into five ‘phrases’ or ‘lines’, I have divided my
discussion accordingly: it should be noted that I do this to facilitate the discussion,
not because I necessarily agree with this division at every turn.
(1) ceres — *ad: The first letters, ceres far are beyond doubt. This is followed by the
upper left-hand corner of an m, or an n, or possibly a d, although I do not think this
likely. The text then continues on another fragment with an e of which the bottom half
is preserved: it cannot be read otherwise. Whether there is a lacuna between these last
two letters depends both on how the letter after far is read and on the size of the lacuna
in f[ ]iqod directly underneath (see fig.12.2). If that lacuna is left as it is in Mengarelli’s
395
CHAPTER 12
reconstruction, possible readings here are ṃe[, ṇ[.]e[, or ḍ[.]e[: if it is reduced, ṃe[, ṇe[
or ḍe[, and if it is enlarged, ṃ[.]e[, ṇ[.]e[, or ḍ[..]e[. (Pace Radke (1965:134), these
three possibilities are impossible if the lacuna in f[ ]ịqod is left as it is.) Of these nine
possibilities, ṃ[.]e[, ṇ[.]e[ (suggested by Thulin) and ḍ[.]e[ do not look very promising
(unless ḍ[.]e is ḍ[u]e… or ḍ[i]e…): the only real choices are ḍ[..]e[ (Olzscha in Radke
1965:136-7, Radke 1965:134, 1994:105), ṃe[ (the usual reading since Herbig CIE), and
ṇe[ (Gamurrini). As in my view the lacuna in f[ ]ịqod should in fact be reduced rather
than enlarged (see below), I would also dismiss ḍ[..]e, leaving only ṃe[ or ṇe[.
Fig.12.2. Detail of EF 1: the lacunas in f[ ]ịqod and hutị[ ]ilom.

The lacunas in f[ ]ịqod and hutị[ ]ilom both depend on the placing of the shard with ]ịqod . In my view,
the group of shards to the right of the lacuna should be moved slightly to the left and down, although it is
difficult to see how the shard with ]ịqod fits unto the one with ceresfar*[ due to the filler material partly
obscuring the join. Note how the decoration is interrupted for the l of ilom. (Tracing from author’s slide.)
The e[ is followed by a large lacuna to which belong two fragments that cannot
be joined to the other shards, but clearly belong to the first line since they contain parts
of a ridge that ran around the base of the lip.188 The larger of these fragments contains
the letters ]tom followed by the lower half of an l (the usual reading since Herbig CIE,
but Radke (1994:105) expressly rejects this possibility) or a u (Thulin 1908:257,
Olzscha in Radke 1965:137, Radke 1965:134-5, 1994:105). The t is , not as in the
rest of the inscription: Vetter (1953:208) in fact considered reading ẓ, but it is hard to
see what word could possibly result from reading ]ẓom. The smaller fragment reads
188
Ribezzo joined both fragments, which is impossible. In Thulin’s and Ribezzo’s drawings, the
larger fragment is joined to the shards of the second line, which is equally impossible.
396
]uf[: the traces of an o read by Vetter (1939:156, 1953:280) before the u are
non-existent, nor is it possible to read the second letter as a (pace Thulin 1908:257).
The order in which these two fragments are to be placed is not known: the provisory
arrangement in the reconstruction is e[2-3]tom *[4]uf[1-2], but the shards could equally
well be arranged as e[2-5]uf[2-5]tom *. Editors have usually interpreted the word
ending in ]tom as an attribute to far. Suggestions are e.g. ṃe[re]ṭom (Vetter 1953:280),
me[lc]tom or me[lq]tom (Vetter in Knobloch 1958:138),189 ḍ[el]e[c]ṭom (Olzscha in
Radke 1965:136-7, Radke 1965:134 n.1) or d[el]e[k]tom (Radke 1994:105), and
me[le]tom or me[la]tom ‘molitum’190 (Joseph & Klein 1981:294). Pisani, however,
restored farṃẹ[n]ṭom as a noun *farmentum (1946:54, 1964:348), a reading adopted by
G. Giacomelli (1963:41) and Morandi (1982:55), but a derivation with /-smentom/ from
a nominal stem at this date unconvincing, however.191
Following the lacuna that contained these two fragments is ]ui[..]m (with only
the two lower points of the interpunct preserved): the traces of an r seen by Vetter
(1939:156, 1953:280) before ]ui[ are non-existent.
This ]ui[..]m in turn is followed by a lacuna whose size may vary slightly,
depending on the restorations in the second line, and contained three or four letters. On
the left-hand edge of the lacuna the lower half of a shaft is preserved, which may be part
of an i, k, m, n, p, or r, perhaps of an f, q, or t, but not of an a (pace Herbig CIE 8079,
Jacobsohn 1910:3, Ribezzo 1918:56, 1927:151-2, 1930:74, 1936:46), as it is too straight
for this, nor of a d, (pace Herbig 1923:233, Vetter 1925:27, Pisani 1946:50, 1964:347),
as it shows no sign of a bar or branch at the bottom.
On the right-hand edge of the lacuna part of a stroke leaning to the right is visible,
probably part of a k or a u, although a d, p, or r might also be possible. It might perhaps
even be an i, although in the rest of the inscription the i is more vertical and less
inclined than the trace; the l read by Ribezzo (1927:151-2, 1930:74), is impossible, as
this letter is either straight or leans to the left in the rest of the inscription. This
*[3-4]*ad has been taken as a third person singular present subjunctive since Herbig
CIE.192 In that case the only restoration that has been proposed and that fits both the
189
Joseph & Klein (1981:294) rejected these restorations, stating that Vetter had proposed them
as reconstructed forms of multum ‘much’, but Vetter’s discussion clearly shows that he had
intended them as reconstructed forms of the perfect participle multum=mulctum.
190
Watkins (1995a:128) points to the parallel between far me[la]tom and Avestan yauua aš ̣a,
Hittite ZÍD.DA ZÍZ mallan: the same would be true for Vetter’s me[lc]tom or me[lq]tom.
191
For the derivation, Pisani compared frumentum, which he derived from frux, presumably
through */frūgsmentom/, as */frūgmentom/ would have given frugmentum (cf. Bücheler
1905:318-9); it can equally well be derived from the root of fruor, /frū( )-/ m */frūg -/ (cf. DÉ,
LEW s.v. fruor), either through */frū (s)mentom/ or through */frūg( )-smentom/.
192
Radke (1965:136) compared ]*adeuios to a South Picene genitive akinevihi read by him in
AQ.2, but this is now read as a dative rakinevíi. Herbig had at first read ḷẹuios (1910:184).
397
CHAPTER 12
traces and the size of the lacuna is Peruzzi’s p ̣[ore]ḳad ‘porrigat’ (1964a:157), adopted
by G. Giacomelli (1978: 525) and Morandi (1982:55); however, none of the meanings
of porrigo listed in the OLD seem to fit the text.193 In view of the uncertainty about
what precedes, it is also possible, as Peruzzi (1964a:157) noted, to read *[3-4]*ad as a
third person plural present subjunctive in -a(n)d, or even as an a-stem ablative singular
in -ad. Radke (1994:106) is the first to interpret *[3-4]*ad in this way, reading
“a[dkap]iad (= accipiant)” with Herbig (CIE 8079) and Jacobssohn (1910:3), but the
syntax of the sentence as he reads it, Ceres far d[el]e[k]tom u[el ro]uf[om] ui[nom]
a[dkap]iad Euios, is strained.
Not surprisingly, this part of the text has been read in various ways. Herbig read
ṃẹ[l ferc]ṭom : ḷ[o]ụf[om ]ui[no]m :, interpreting this a prayer to Ceres to accept (CIE,
with ạ[dkap]ịad) or to provide ([dou]ịad 1913, ui[ctu]m : [dou]ịad 1923) various gifts,
and these readings and interpretations where adopted by editors such as Buonamici and
Stolte, and with variations, by Ribezzo. They were rejected ey are based were rejected
by Vetter, who read l[o]ụfir ‘Liber’ (1925:27-8: his later lọuf[ir (1939:156) and even
louf[i]ṛ (1953:280) were based on non-existent traces seen by him before ]uf[ and ]ui[),
taking the phrase as a prayer that Ceres should provide spelt and Liber wine. This
interpretation was adopted by all later editors except Ribezzo194 and Radke: for the
latter’s reading of this passage, see below.
The possibility of reading l[o]ụf[ir was doubted by Peruzzi (1964a:156-7), how-
ever, and was rejected by Radke (1965:134-5) as being simply far too small to fit the
lacuna. This I can confirm from autopsy: reading l[o]ụf[ir, let alone lọuf[i]r, is
impossible, and consequently there is no direct mention of Liber in this text, which
makes the interpretation of euios as EÜioj difficult, as is discussed below. (Radke rightly
added that it also reduces the likelihood of the restoration ui[no]m being correct.) It is
necessary to stress this point, since even Watkins (1995a, 1995b) still uses Vetter’s
impossible lọuf[i]r and bases an important part of his discussion of the text on this
reading. The contents of the first part ceres — *ad therefore in my view remain unclear
apart from the fact that ceres is mentioned in association with far that was possibly
specified as ]tom, and that, if *[3-4]*ad is indeed a third singular present subjunctive,
the phrase may well be an invocation.
193
Other proposals have been ạ[dkap]ịad (Herbig CIE 8079, Jacobsohn 1910:3, Ribezzo
1918:56), ạ[dtul]ạd (Ribezzo 1927:151-2, 1930:74), ạ[ddou]ịad (Ribezzo 1936:46), [dou]iad
(Herbig 1913:75, 1923:233, Vetter 1925:27, Pisani 1946:50, 1964:347), f ̣[in]ḳad (Olzscha
1965:123, in Radke 1965:136), [fe]ṛad (Vetter 1953:280, Joseph & Klein 1981:294), and
[pa]ṛad (Joseph & Klein 1981:294).
194
Ribezzo interpreted the text as an injunction to bring offerings, taking ceres either as
‘granum’, one of the gifts (1918:56, 1927:151; 1929:79 n.4), apparently in the accusative (?), or
as the recipient (1930:74 (with l[o]ụf[rom), 1934:226, 1936:46) in the genitive (??).
398
(2) euios — f[.f]ịqod The following letters are euios mamaz, followed by a small
lacuna. On the lower edge of this lacuna both Mengarelli’s and Thulin’s drawing shows
a small trace, seen also by Vetter (1953:280), who read ẓẹxtos, and apparently by G.
Giacomelli (“la e va supplita quasi interamente”, 1963:41). Both Peruzzi (1964a: 152-3)
and Radke (1965:137) deny its existence, however, and I myself have likewise been
unable to find it. Their doubt whether the lacuna could not have contained two letters
seems unjustified. The lacuna is followed by part of an x (the upper right-hand and both
lower ends are preserved), a t (Radke’s alternatives ḳ and ị (1965:137) are impossible)
and os: euios mamaz[e]xtos, or perhaps euios mamazẹxtos.
Reading l[o]ụf[ir ‘Liber’ in the first phrase, after Vetter, has its repercussions for
the way in which euios is interpreted, since the interpretations quoted in the discussion
of the first phrase ceres — *ad all presupposed that euios was to be taken together with
the following mamaz[e]xtos as part of the second phrase. As Herbig (1913:78 n.1) and
Scherer (1956:118) had noted, this had the disadvantage of making the first phrase the
only one not to end in , while at the same time assuming within what was apparently a
closely connected group of names, since euios mama z[e]xtos was interpreted as either
(1) praenomen + gentilicium + cognomen (Herbig CIE 8079 (‘Sextus Mama Euius’),
Ribezzo 1918:56, 1927:150-1, 1930:74, 1934:226, 1936:47, Vetter 1939:155), an
interpretation which can be rejected for a text of this date and which furthermore
requires that f[ ]ịqod is interpreted as a third person singular, or (2) a gentilicium
followed by two praenomina (in which case euios is singular (Meister 1916:101), not a
Sabellic-type nominative plural in /-ōs/, as Norden (1939:206-7) and Vetter (1953:280)
suggested, as this would clash with the Early Faliscan a-stem nominative plural sociai
in EF 4, cf. §4.2.6, §4.3.6) or (3) three praenomina (Herbig 1913:78 n.1, Pisani
1946:50, 1964:347-8). See also §7.2.1 for a discussion of this group of names.
As a consequence of Vetter’s l[o]ụf[ir ‘Liber’, euios was again attached to the
first phrase by G. Giacomelli (1963:41-2) and Peruzzi (1964a:166) and interpreted as
EÜioj, an epithet of l[o]ụf[ir. As was said above, Radke (1965:134-5) is in my view
right in regarding l[o]ụf[ir as impossible and as a consequence, in removing the basis
for the restoration ui[no]m and for the interpretation of euios as EÜioj. If l[o]ụf[ir is
dropped as a correct or even possible reading, maintaining euios as EÜioj (Olzscha in
Radke 1965:137,195 G. Giacomelli 1978:525, Morandi 1982:55) has no other basis than
the frequently adduced association of Ceres with Liber: but the only evidence for this
association at the time of the Ceres-inscription seems in fact to be the reading of
l[o]ụf[ir ‘Liber’ in the Ceres-inscription itself (cf. Le Bonniec 1958:303-4).
195
Olzscha, who had previously (1965:123) interpreted l[o]ụf[ir implausibly as ‘uel’ (cf. Oscan
loufir TB 1,8) to avoid the hyperbaton ‘Liber uinum ...at EÜioj’ (although he was apparently
quite happy to accept the hyperbaton in soc ̣[iai] p ̣orded karai), maintained this interpretation by
reading u[el ro]uf[om] (?) ui[no]m, with an adjective preceding the noun, contrary to the
normal Faliscan word-order (§8.4.1). This reading is maintained by Radke (1994).
399
CHAPTER 12
The inscription then continues with med followed by a lacuna that has on its left
edge the left half of an f and on its right edge the middle part of a shaft (see fig.12.1),
which can be part of an i, t, or z, but not of an a (pace Thulin 1908:258 n.3 and Radke
1965:137), and which is in its turn followed by qod . For this f[ ]ịqod both Buonamici
(1913:40) and Herbig (1913:74-80) independently arrived at an restoration f[if]iqod or
f[ef]iqod ‘finxerunt’; similar restorations had been considered by Thulin (1908:258) and
Herbig himself (CIE 8079). This has been adopted by all editors except Ribezzo, who,
having first adopted f[if]ịqod as ‘fecit’ (1918:248 n.3), later erroneously claimed that it
did not fit the lacuna and proposed the impossible f ̣iqod ‘figat’ (1927:150-1), and
f ̣[e]iqod (‘fingat’ 1930:74-5, 1936:47, ‘figito’ 1934:226). The lacuna is in fact large
enough to allow even the f[i:f]ịqod considered by Lejeune (1955:146 n.9) and Radke
(1965:137). As the fragment containing ]qod cannot be joined perfectly onto the group
of shards that contains f[, the size of the lacuna can be adjusted by a few millimetres,
which affects both the possibilities for the letter between far and e[ in the preceding and
the lacuna in hut*[ ]ilom in the next line. If the lacuna in f[ ]ịqod is enlarged, Radke’s
f[e:f]ịqod (1965:137) becomes possible; if it is reduced, f[ ]qod can be restored only as
f[if]ịqod. F[.f]ịqod and fifiked are usually regarded as perfects of fingo: Lejeune
(1955:148-50) rejected the possibility that they are forms of facio. This view has now
been revived by Poccetti (2005:31-5) and Berenguer & Luján (2005:206-7), however:
see §5.2.1.7-8. For the ending of f[.f]ịqod see §5.1.4e. The names followed by med
show that the phrase is undoubtedly a potters’ signature of the iscrizioni parlanti-type
(see §8.8.2), with a direct parallel in tele*[1-2?] med fifiked EF 9.
(3) praụ[ — karai The interpunct after ]iqod is followed by the clearly legible pra and
the lower half of a u (or possibly of a d or l), after which the text continues on another
fragment (cf. fig.2.1). All editors except Thulin read an i at this point, although the
traces of this letter seem to fall in the joint between both fragments (Nogara in Herbig
CIE 8079), and I doubt whether they are not in fact part of the joint. After the lacuna,
the text continues with osurnam so. This is followed by the leftmost part of a small
circle on the left edge of a lacuna, either c (thus most editors) or possibly q (Thulin
1908:258; Ribezzo 1927:151). The lacuna itself may have contained three or four
letters, depending on the sizes of the lacunas in f[ ]qod, *[ ]*ad, and ụrneḷ[ ]tela.196 On
the other side of the lacuna is the lower half of a shaft, which can be part of an f, i, p, r,
or t, but not of a q (Thulin 1908:258). Herbig restored this as soc ̣[iai]p ̣orded (cf. Thulin
1908:258), which has been accepted by all editors except Ribezzo.197 Note that here is
196
Thulin’s so(q)orded (without lacuna), based on a reconstruction he later discarded, is
impossible, as he himself admits (1908:255, 258).
197
Ribezzo’s soq[uiai (1927:151) and Vetter’s alternatives soc[ruei or soc[ru (1925:26-7) are
epigraphically possible; not so Ribezzo’s so[uai (1918:56, 1930:74) and so[dali (1934:226), and
Vetter’s alternatives soṣ[orei, soṣ[erei, or soṣ[rei (1925:26-7).
400
only one trace between the edge of the lacuna and orded: reading soc ̣[ia]ịporded
(Vetter 1939:155, 1953:280, Pisani 1964:347) is impossible. This the only part of the
inscription where the editors agree to some extent: praụ[i]os urnam soc ̣[iai] p ̣orded
karai ‘Pravius gave the urn to his dear girlfriend’ has been the accepted reading and
interpretation since Herbig, with the exceptions noted above.
(4) eqo — fitaidupes The next phrase starts with the letters eqo, clearly ‘ego’, attested
in Early Faliscan eco EF 2, eko EF 6, 7, and eqo EF 465*: Ribezzo’s interpretation of
eqo as an */ekō/ from the Sabellic pronominal stem /eko-/ (‘ita’ 1918:56, 1934:226;
‘sic’ 1930:74-5), or as ‘e(r)go’ (1927:151) is not convincing,
In the following letters, ụrneḷ[a (thus first Ribezzo 1918:56), only the lower
halves of the u and the l are preserved (the second is omitted by Thulin). The remaining
two or perhaps three missing letters (depending ultimately on the size of the lacunas in
f[ ]ịqod and soc ̣[ ]p ̣orded) are best taken together with the ]tela following the lacuna
and interpreted as an adjective with ụrneḷ[a, as was already proposed by Thulin
(1908:259).198 Since Ribezzo (1934), this ụrneḷ[a 2-3]tela has generally been regarded
as a ‘diminutivo continuativo’ along the lines of Catullus’ turgiduli ocelli (3.18). Of the
restorations along this line, the most attractive is surely Watkins’ ti]tela (1995a:129)
based on eco urna tita uendias Lat 479†, although along with it Watkins unfortunately
resurrected Knobloch’s (1958:137-8) highly fanciful interpretation ‘piggy-bank’ for
urna tita. Other possibilities are Pisani’s pa]tela ‘patula’ = ‘larga, ampia’ (1946:53,
1964:347-8) and Giacomell’s lu]tela ‘lute(ol)a’ (1963:41-3, 1978:525-6), which has
been adopted by Peruzzi (1964a:161), Radke (1965:137), and Morandi (1982:56).
Ribezzo’s arcen]ṭela (1918:56, 1927:151, 1930:74, 1934:226) does not fit the lacuna
(nor the sense, as the urn is anything but bright as silver), and Vetter’s pu]tela ‘parvula’
(1953:280) is based on his very doubtful interpretation of putellio MF 151 as ‘infans’.
The remaining fitaidupes199 is usually divided either as fita idupes or as fitai
dupes.200 Fita or fitai has been connected to (1) Latin fingo and derivations (fitilla
Thulin 1908:259, Herbig CIE 8079, Ribezzo 1918:56, 1927:151, 1930:74, 1934:226,
1936:47; perfect participle fi(c)ta G. Giacomelli 1963:41-3, 1978:525-6, Peruzzi
198
Most early editors let the next word start after the lacuna, reading either telafitai (thus
Gamurrini 1894:340, Thulin 1908:259, Herbig 1913:85 n.2 (both interpreting it as the name of a
goddess), Herbig 1923:233 (“der zu Erde gewordenen (Toten)”), Vetter 1925:30 (first person
singular perfect of an unknown */tela-/) or tela ‘terra’ (thus Herbig CIE 8079, Jacobsohn
1910:3, Buonamici 1913; ex ]tela ‘ex terra’ Stolte 1929:107).
199
Mengarelli’s and Ribezzo’s drawings, Buonamici (1913:45-6), Ribezzo (1918:56,
1927:151-2, 1930:74-5, 1934:226), and Stolte (1929:107), erroneously read an r instead of p.
200
Other divisions are Buonamici’s tela fitaid ures (1913) and Stolte’s ex ]tela fita idu res
(1929:107). Morandi (1982:56) and Watkins (1995a:129 n.5) compare fitaidupes to South
Picene pidaitúpas TE.5.
401
CHAPTER 12
1964a:161-2, Morandi 1982:56), (2) Latin fio (fitum est Andr.29L Stolte 1929:107, fitai
‘exstiti’ Vetter 1953:280; Herbig’s comparison of Umbrian fito TI VIb.11 (CIE 8079)
assumes the same root), and (3) Latin fetus (Pisani 1946:53, 1964:347-8).
Dupes was first interpreted as ‘bipes’ (Thulin 1908:259, Herbig 1913:85 n.2,
1923:233), but this was rejected by Vetter (1925:29-30, 1953:282-3), who connected it
with Latin dupondius (thus also Pisani 1946:51-3, 1964:347-8, and Morandi 1982:56).
Ribezzo (1936:47) took dupes as a present participle *dubens ‘potens’. G. Giacomelli
(1963:41,43, 1978:525-6) and Peruzzi (1964a:163-4) both divide idupes, reading
‘*indupens’ = ‘pesante’ and ‘*indupes’ = ‘impes’ = ‘impetus’ respectively. The
problem with these last suggestions is that they presuppose an i(n)du- (again with a
problematic drop of a syllable-final nasal) at a time when this word at least in Latin still
had the form endo in CIL I2.4.201 Radke’s interpretation (1994:108-9) of idupes as
‘Idubus’, with the ending -pes (instead of the expected *-fos) ascribed to a presumed
Siculian presence in the ager Faliscus, is in my view entirely without merit.202 Likewise
unconvincing is Matzloff (2006:67-75), who divides fit aidupes, with aidupes as ‘qui a
le pied enflé’ (≈ O„d…pouj), translating “[Pravios] a-t-il une crise de goutte? («devient-il
podagre (aidupes)?», «devient-il homme aux pieds enflés?» )”.
(5) arcentelom — douiad The fifth phrase, which appears to be syntactically linked
with the fourth, starts with a clearly legible arcentelomhut. This is followed by a lacuna
whose width again depends on the one in f[ ]ịqod: if that is reduced or left as it is, two
letters are missing here; if it is enlarged, two or three (cf. fig.12.2). Pisani’s view
(1946:53, 1964:348-9), which he incorrectly ascribes to Stolte (1929), that this lacuna
contained only a part of the decoration but no letters is untenable in view of the form of
the horse on the other side of the vase and also of the fact that the decoration is actually
interrupted for the l almost directly following the lacuna, implying that the inscription
was written before the decoration (cf. fig.12.2). A small trace is preserved on the left
edge of the lacuna, consisting of the upper part of a shaft tilted slightly to the left,
belonging to an i (eds.), an u, or an l (Ribezzo 1936:47). The traces of a c or x seen by
Vetter (1939:156, 1942:220) on the right-hand edge of the lacuna are non-existent.
Arcentelom is obviously derived in some way from argentum, either as a
diminutive or as an adjective. Some authors have interpreted both words as nouns, e.g.
Herbig, who interpreted arcentelom as ‘*argentulum’, and his hut*ilom (CIE 8079) as ≈
201
This applies also to Stolte’s interpretation of his misreading idu res (1929:107).
202
Note e.g. (1) that Radke’s parallel, Siculian nanepos in the inscription from Centuripe, has
a different vowel-grade; (2) that assumptions about a Siculian presence in the ager Faliscus
are entirely based on Dion. 1.21.1; (3) that this would be an instance of interference involving
an inflexional suffix without any indications of interference or borrowing from Siculian on the
higher levels of the borrowing hierarchy (§1.3.2.2).
402
Greek fÚtlh?, later (1923:233) suggesting hut[im p]ilom ‘cÚsin p‹lon’: others have
interpreted both as adjectives, e.g. Pisani (1946:51,53-4, 1964:347-8) who read dupes
arcentelom hutilom (or hut*ilom) and interpreted ‘dupondium argentulum fusile’.
Martzloff (2006:68-9) unconvincingly derives arcentelom from arceo in the same way
that adulescentulus is derived from adolesco, and regards it as similar in sense to Greek
¢lexif£rmakon. He continues (2006:75) his translation (quoted above) with “Je tiens a sa
disposition (pe:parai) une petite potion, en guise de petit remède (arcentelom)”.
Fig.12.3. Detail of 1, showing the join between prau[ and ]osurnam soc.̣
The shards with prau[ and ]osurnam soc ̣[ do not fit together, a fact obscured by the large amount of filler
material used in the reconstruction. The line underneath, with pe para[, runs straight into ]osurnam soc ̣[,
whereas ]douiad is written much lower. The shards with ]osurnam soc ̣[ and ]douiad should be moved
slightly upwards and tilted c. 5-10° counter-clockwise (cf. fig.12.1). (Tracing from author’s slides.)
Most editors, however, take one of the words arcentelom hut[ ]ilom as a noun and
the other as an adjective (often again as a ‘diminutivo continuativo’). Reading
arcentelom as a noun and hut*[.]ilom as an adjective does not result in adequate
interpretations for the latter (Herbig 1913:85-6; Buonamici 1913; Vetter 1925:29-30,
hutic ̣ilom 1953:280,283; Stolte 1929:107; hutic ̣ilom ‘gegossen’, from an adjective in
-icius, Knobloch 1958:136; hutị[p]ilom ‘quadruplum’, from Etruscan huθ with a Latin
suffix, Radke 1965:138). It is more promising to take arcentelom as an adjective, with
hut*[1-2]ilom as a noun. The best proposal seems to be G. Giacomelli’s hutị[c]ilom
‘*futicillum’ (1963:41,43, 1978:525-6), a double diminutive, like penis - peniculus -
penicillus (Peruzzi 1964a:163-4). In that case, the change in gender is surprising (G.
Giacomelli 1963:43 n.15), and the word is perhaps rather a genitive plural (Peruzzi).203
Ribezzo’s hut[r]ilom ‘poculum’, ‘operculum’ (1918:56 etc.) and hutlilom ‘futillum,
operculum’ (1936:47) are epigraphically and linguistically impossible.
203
Unclear as to the precise status of both words are Buonamici’s hutifilom (1913) and Radke’s
‘Silberausguß’ (1965:138).
403
CHAPTER 12
The final part of the text is usually read as pe para[i]douiad (with enough space
after douiad to show that this is indeed the end of the text) and interpreted as ‘peperi
duat’ since Herbig (1913:84-7): reading pe paraị at this point (Herbig 1913:85 n.2,
1923:233, Stolte 1929:107, Pisani 1946:50, 1964:347) is unjustified. The reading is not
without problems, however. The width of the lacuna shows that more is missing than
one i, and while pe para[ seems to run straight into the praụ[i]os in the line above it,
douiad is written downwards at an angle to the rest. This difference in direction is less
marked if the shards with ]osurnam soc ̣[ and ]douiad are moved slightly upwards and
tilted 5-10° counterclockwise, as they are in Herbig’s drawing (cf. fig.12.1).
Syntactically, too, the abrupt change from the first person pe para[i to the third
person douiad is awkward: douiad is usually taken either as a loose addition to the text
or as directly dependent on pe para[i (Herbig 1923:233, Vetter 1925:30). Vetter
(1953:283) in fact considered integrating a conjunction, but the lacuna is too small for
this. None of the proposals to avoid these problems are attractive. Peruzzi’s pe par
a[dke]douiad ‘per par accedat’ and pe para[te ke]douiad ‘per parate accedat’
(1964a:164-6) do not result in a really meaningful phrase, and assume an omission of
word-final r that is difficult (§3.5.7b). Ribezzo’s pe (‘-que’ (1918:56, 1930:74,
‘(nem)pe’ 1927:151-2) followed by para--- ‘para(tum)’ (1918:56) or by para[ad]doviat
[sic] (1927:151), para[i]douiad (1930:74-5), or para[:]douiad (1934:226) ‘superaddat’,
‘iuxta addat’ (1936:47), are impossible on various grounds.
Therefore, even though several words can be more or less plausibly isolated, it is
very difficult to make any sense of the last two phrases. The fourth phrase opens with
‘I, a little ... urn’, which appears to be the subject of pe para[i (if to be read thus), but
how pe para[i is to be understood is unclear because of the problems in interpreting
telafitaidupes and arcentelom hutị[.]ilom, which constitute or contain the object of
pe para[i. None of the attempts to interpret this part of the text or similar sentences
resulting from different restorations (Pisani 1946:51-4, 1964:347-9, G. Giacomelli
1963:41-3, 1978:525-6, Peruzzi 1964a:160-6) is in my view really convincing.
(1-5) The text as a whole. Putting together the readings discussed above, I arrive at the
following text, choosing to be too cautious rather than too bold:
┌─────┬───┐
(1) ceres far*[0-2]e[1-3]tom *[3-5]uf[1-4]ui[2]m *[3-4]*ad (invocation?)
(with interdependent second, third and fourth lacunas, where 8-10 letters are
missing in all, and the possibility that ]tom *[ and ]uf[ are in the wrong order)
(2) euios mamaz[e]xtosmedf[.f]ịqod (formulaic potters’ signature)
(3) praụ[i]osurnam soc ̣[iai]p ̣ordedkarai (statement that the vase was a gift)
(4) eqoụrneḷ[ati?]telafitaidupes
(5) arcentelomhutị[.]ilom pe para[i? 1-2]douiad
404
Earlier authors regarded the contents of the text, especially of the first and the last part,
as religious (Thulin 1908:258, Herbig CIE 8079, Buonamici 1913:37-8, Taylor
1923:76-7); some more specifically connected it with the funerary ritual (Herbig 1913,
1923, Stolte 1929:109-111, 1930, Altheim 1931:117-8, Ribezzo 1918:56, 1927:150-1,
1930:74-5, 1934, 1936:46-8).204
Vetter, stressing the third phrase, regarded the text rather as a commemoration of
a festive occasion, with the first part as an invocation of Ceres’ blessing on the recipient
(1925:27-8). This interpretation has been adopted by most later authors (Pisani,
Lejeune, Knobloch, G. Giacomelli, Peruzzi (connecting the first part with the cult of
Bacchus), and, with some hesitation, Radke). Although this has the advantage of being
based on a relatively clear part of the text, it provides no clues for the restoration of the
more fragmentary parts of the text. The fact that the first phrase may be an invocation
while the second phrase is a definitely formulaic signature (§8.9.2), while the third
phrase seems to be of a more personal nature, does not seem to point to any kind of
consistency within the text as a whole. Like many of the early texts, it appears to be
‘highly individual’, and, although using of formulaic phrases, it has no real parallels.
Poetic aspects. The inscription was regarded as metrical by Vetter (1925:27, 1953:
281-2) because of the word-order in soc ̣[iai] p ̣orded karai and because the text can be
divided into five lines of an approximately equal number of syllables. Thus, we have
c.12 syllables in the first line (depending on the reconstruction, e.g. 12 in Vetter’s 1953
rendering ceres far ṃe[re]ṭom l[o]ụf[ir] uinom [fe]ṛad), 11 in the second, 12 in the
third, probably 11 in the fourth, and 10-13 in the fifth, depending on the inclusion or
exclusion of douiad). The same has been maintained by Norden (1939:206) and
Brandenstein (in Pisani 1964:348)205, who apparently analysed the inscription as iambic
senarii. As Peruzzi (1964a:155) states, there is a distinct possibility that the verse is
Saturnian, and it is in fact analysed in this way by Radke (1994:106-8), although his
interpretation of Saturnians is difficult to follow. The main problem with any metrical
interpretation is that it must necessarily be based on more or less complete texts, as it
requires an approximately equal number of syllables per line. Given the great
uncertainties in both reading and interpretation, I do not think that it is possible to arrive
at a metrical interpretation of the text that is in any way justifiable.
204
Stolte (1930) and Altheim compared the horses of the decoration to hippomorphic represent-
ations of Demeter from Arcadia. The type of decoration is a common one at this period,
however, and has probably no significance for the interpretation of the text (Peruzzi 1964:151
n.9, Morandi 1982:56).
205
The latter’s scansion, however, cerés farméntom lóufir vínom dóvjad / evjós mamá zextós
med fífiqód / pravjós urnám socjái pordéd karái / eq’ úrnelá ?telá fitái dupés / arcéntelóm
hutícilom péparai dóvjad, is very implausible, to say the least: note (among several other
oddities) that the reduplicative syllable is apparently long in fífiqód but short in péparai.
405
CHAPTER 12
An elaborate and more attractive discussion of the poetic aspects of the text is
given by Watkins (1995a:126-31, 1995b:45-8), for his discussion focuses more on the
textual than on the metrical basis for a poetical interpretation. Unfortunately, his
interpretation is based on Vetter’s untenable restorations, and this invalidates several of
his conclusions, although perhaps not the overall gist of his discussion. In his view,
Ceres far me[la]tom Louf[i]r ui[no]m p̣[a]ṛad is a variation of a traditional blessing
formula found also in Paelignian dida  uus  deti  hanustu  herentas Pg 9206: a variant
in the sense that the crucial verb of the formula, douiad, was replaced in the first phrase
by a p ̣[a]ṛad that is echoed later in the text by pe:parai (as he reads it). The ‘frustrated
expectation’ in which this results is then resolved at the end, where the expected douiad
is deftly inserted, making the text into ‘ring-composition’ (for his definition of this
term, cf. Watkins 1995a:34-36). Watkins is thus unique in making the difficult and
isolated douiad at the end into one of the cornerstones of his interpretation. Poccetti
(2007:251), too, draws attention to the stylistic features that can be regarded as Faliscan,
abstaining from judgement on the poetic form or metre itself.
From two autopsies in the Museo di Villa Giulia, Rome (inv. 3548). Bibliography: Gamurrini
1894:339-41 (autopsy); Mengarelli 1908 (autopsy); Thulin 1908 (autopsy); Herbig 1910:184-5 (au-
topsy); Jacobsohn 1910:3 ¢11²; Herbig CIE 8079; Herbig 1913:74-8, 80-7; Buonamici 1913:37-49 ¢1²;
[Weege in Helbig 1913:361]; Meister 1916:99-101; [Della Seta 1918:52 (autopsy)]; Ribezzo 1918:
248-9; Taylor 1923:77; Herbig 1923:233; Vetter 1925; Ribezzo 1927:150-1; Buonamici 1929b:511;
Ribezzo 1929a:79 n.4; Stolte 1929; Ribezzo 1930:74-5; Stolte 1930; Giglioli 1930:337; Altheim 1931:
117-8; Ribezzo 1934; Buonamici 1936:424-6; Ribezzo 1936; Vetter 1939a:155-6 (autopsy); Norden
1939:266-7; [Andrén 1940:87]; Vetter 1942:220-1; Pisani 1946; Lejeune 1952a:340-1; Safarewicz
1953:245-6; Bruhl 1953:25; Vetter 1953:279-83 ¢241²; Lejeune 1955:146-8; Scherer 1956:118;
Knobloch 1958; Le Bonniec 1958:303-4; Radke 1962:143; G. Giacomelli 1963:41-4 ¢1² (autopsy);
Pisani 1964:347-9 ¢151²; Peruzzi 1964a:150-67 (autopsy); Radke 1965b (autopsy); Olzscha 1965:123;
G. Giacomelli 1978:525-6 ¢1²; FI II.2 p.208 (autopsy); Joseph & Klein 1981; Agostiniani 1982:150-1
¢591²; Morandi 1982:54-6 ¢8²; Radke 1994 (autopsy); Watkins 1995a:126-31; Watkins 1995b:45-8;
Martzloff 2006:66-74; Poccetti 2007:251. Photographs: Mengarelli 1908:101-2; G. Giacomelli 1963
tav.II; G. Giacomelli 1978:509. Drawings: Mengarelli 1908:103 (reproduced in Herbig CIE 8079 p.23,
Peruzzi 1964:151 fig.2); Thulin 1908 inter pp. 254-5 (reproduced in Herbig CIE 8079 p.22, Morandi
1982:55); Ribezzo 1918:56 fig.2 (reproduced in Ribezzo 1927:151); Lejeune 1952:122.
12.3. The two impasto pitchers
2-4. Two impasto pitchers, reputedly from Civita Castellana, were sold to the Museo di
Villa Giulia in 1921. The first words of 2 were quoted already in 1933 by Pallottino, but
the inscriptions were published in their entirety only in 1935 by Giglioli and Braun.
They were dated to the seventh century by Giglioli, in which case they would be older
than the Ceres-inscription (which is usually treated as the oldest Faliscan inscription).
206
Already Buonamici (1913:38) had pointed to this text as a parallel, but it has largely
disappeared from the discussions of the text by later authors.
406
2. Scratched c.1½ times around the body (letters 4-11 mm high) of a small impasto
pitcher that also contains 3. From their position on the vase, G. Giacomelli (1963)
concluded that 2 is the primary inscription.207
propramom10 prameḍ[u]mo20mpramodpra30medumom pr40amodpropra50mọḍ pr
amod60umọ[m]
Dextroverse. The letters vary in size, and the text is damaged at several points. After
letter 46 the text bends downward to pass under the first line. From my autopsy, I can
almost completely confirm the readings of G. Giacomelli and Peruzzi. They rightly
stated that the text starts with propramom, not with Giglioli’s propramod (adopted by
most authors), Braun’s propramop̣, or Ribezzo’s propramoe. They also rightly read
letter 11 as where others read a lacuna ([e] Giglioli, [:] Ribezzo) or an empty space
(Ribezzo, Pisani, Vetter). Letters 16-17 are a very clear e (certainly not Braun’s and
Ribezzo’s i!) and the upper left-hand corner of a d (hardly Ribezzo’s and Vetter’s p).
This is followed by a lacuna of one letter on the place of letter 18, where [u] is the
obvious restoration: the traces of this u seen by Vetter are non-existent. Letters 26-27
are od, as G. Giacomelli and Peruzzi saw, not Giglioli’s oe, which has been adopted by
most authors.
Following letter 44, Vetter read a double and G. Giacomelli a triple interpunct,
but, like Peruzzi, I can see no interpunct at all at this point. The text then continues with
propram, followed by letters 51-53, the first of which is obviously o; the second is read
as ḍ or d by G. Giacomelli and Peruzzi, and the third is a triple interpunct with one point
missing. Feretti’s drawing accompanying Giglioli’s article has lead to the misreadings
pro pram[od] ẹpramod (Giglioli), pro pram[ed]i (Ribezzo), pro pram[o]e or pro
pram[oe] e pramod (Pisani 1937) and pramọ[ e a ]pramod (Pisani). After um there
appear to be slight traces, so that the final letters 63-64 can be read as umọ[m] (Vetter
and G. Giacomelli) rather than um[om].
Due to the repetition of the various elements, the resulting text can easily be
divided into separate words and word-groups: (1) propramom pramed [u]mom, (2)
pramod pramed umom , and (3) pramod propramọḍ (or pro pramọḍ) pramod
umọ[m]. (Alternatively, the text could be divided into segments according to the
interpuncts, but this does not result in a clearer picture.) Apart from the fact that this is
clearly some kind of polyptotic word-play like e.g. duenom duenas EF 3 and tulate
tulas EF/Etr 385, the interpretation is difficult, to say the least. The main word is
pramo-, occurring in the ablative pramod, and in the masculine or neuter accusative (or
neuter nominative) of a compound propramom. Propramọḍ is usually read as pro
pramọḍ (preposition with noun in the ablative) but in view of propramom, it can also be
read as propramọḍ, an ablative of the compound propramo-. Pramed appears to be an
207
Pisani, too, regarded 2 as earlier, but his argument for this, namely that in 3 the word-final
consonants are missing, is spurious.
407
CHAPTER 12
adverb (Vetter 1939a:147): Pisani, however, interpreted it as a subjunctive of a verb

*pramo, equal in sense to Latin promo. Although I cannot agree to the details of his
interpretation, it does appear to be the only way of reading a verbal form in the text.
(Another way would be to take propramom as a first person singular aorist (or perfect
with an aorist ending?), as has been proposed (and rejected) for tulom MF 72, but this
would clash with the accepted reading pe para[i in EF 1. Knobloch’s (1966:48)
pramo(n)d is morphologically impossible.)
Pramo- /prāmo-/ m PIE */p Ħmo-/ (*/p ħ3mo-/?) recurs in Latin prandium m
*/prām(o)-ed-o-m/ ‘early meal’ or ‘first meal’ (cf. §6.2.58), but the exact meaning of
pramo- and its derivatives in this text can only be guessed at. A meaning ‘breakfast’
(thus Ribezzo (with pro pram[ed]i ‘pro prandio’ and an adjective pramidumom,
pramedumom (‘prandial’?)), and Knobloch) or ‘meal’ (Pisani) does not seem
particularly attractive. Vetter (1939:148) gave the word the meaning ‘good’, but later
does not recur to this unargumented explanation.
Umom has only recently received a convincing interpretation. M. Mancini
(2003:239-41, 2004:205-7) rightly points to the udmom in ‘Old Hernician’ [---]matas
udmom ni hvidas mi kait[sis ---] He 2. This may be the name of a type of vase: Rix
(1998:250-1) explains it as /ud-mom/ ← PIE */ ed-/, comparing Latin unda ← */ud-
nā/, and umom can in fact very well represent */ummom/ ← */ud-mom/ (cf. §3.3.4.3).
This may finally lay to rest the presumed connection between umom and Latin umidus
and umor, which, although rejected already by Braun (in Giglioli 1935:241), keeps
cropping up in the literature. If umom were indeed derived from an */ūg -smom/, the
expected result at this date would rather be */ūksmom/ (cf. iouxmen|ta CIL I2.1). (Pisani
tried to avoid this problem by deriving umidus from */uĦ-sm-/, which seems equally
impossible.) I fail to understand how umom can be explained by “assonanza con vinom
o, meglio ancora, a ragioni onomatopeiche” (G. Giacomelli 1963:45).
It is in my view impossible to interpret the text beyond the vaguest possible sense.
Most of the interpretations that have been proposed are based at least partly on
misreadings, on the impossible connection of umom with umidus, or on various
unsubstantiated interpretations of pramo-. Explaining the text as an injunction to eat all
day (Ribezzo) or to drink wine before, during, and after breakfast (Pisani, Knobloch
1966:48-9) furthermore require a pro used in the sense of ante, while Pisani’s a ]pramo
‘post prandium’ also requires an a used in the sense of post. A similar translation which
avoids these problems is M. Mancini’s (2004:206-7) “al primo pasto (propramod) per
prima una brocca (scil. di vino); al pasto principale (pramod) per prima una brocca; al
primo pasto, al pasto principale per prima una brocca”. For Giglioli, the text was an
invocation referring to “una primizia (di un liquido –vino? latte?– oppure invocazione a
qualche divinità di sorgente?)” (1935:241); Braun (1935:443) referred to the text as a
“formula magica”, which amounts to little more than saying ‘this is unintelligible’.
408
3. Scratched, once around the body (letters 5-9 mm high), close to the foot, of the
impasto pitcher that also contains 2.
ecoquto*e10uotenosiot20itiasdueno30mduenassal40ue[to]duolte50ne�
Dextroverse. The q is , the s . For c before o and q before u, see §11.2.2-3. The fifth
letter, although damaged, is certainly a t (pace Ribezzo’s eco quịo ‘ego cuius?’). Quto is
followed by a vertical stroke (cf. Agostiniani’s drawing). The n read by virtually every
editor apart from Buffa (who read nothing at all here), Braun (interpunct), and Vetter
(ị), is impossible: the reading of the second word can only be quto (thus Dirichs, Buffa,
Pisani (although quto[n] still in Pisani 1934), Braun, Vetter, Peruzzi, Agostiniani).
Quto is an adaptation of the Etruscan qutun occurring e.g. in mi qutun lemaus-
nas Etr III from Narce. It is usually regarded as the nominative of an ōn-stem (thus first
Braun 1935:440-1), cf. cupi‹d›o MF 62 and apolo MF 65, but this poses a problem if
(as most editors do) duenom is interpreted as an adjective to go with quto, for duenom
can only be a neuter nominative. However, quto might be read as quto(m), an adapta-
tion of Etruscan qutun to the o-stem neuters. Colonna (1974:140-1) in fact observed that
Etruscan qutun would best be derived, not from Greek kèqwn, but from a Greek *kîqon,
apparently unaware that this form is attested from Hesychius (kîqa: pot»ria k 4788
Latte). Early Faliscan quto could thus represent /k tom/ or /k ttom/ (or perhaps even
/g tom/ or /g ttom/, cf. Latin guttur and guttus). Another possibility is that quto is to be
read as quto(r) /g tor/ or /g ttor/, a counterpart of Latin guttur, which is also often
derived from kèqwn (or perhaps from a *cÚtwr?). Both readings assume omission of a
word-final consonant in Early Faliscan (cf. §3.5.7), but provide a satisfactory explana-
tion for the neuter duenom. There are several possible ways to divide the text here:
(1) quto‹n› euotenosio (Pallottino, Pisani 1934, Buonamici 1934, Buffa, Giglioli, Braun,
Ribezzo, G. Giacomelli, Devine, Agostiniani), with quto‹n› = Etruscan qutun and a
name euoteno- that is has been connected with euios EF 1, although it is usually not
made clear how: a direct derivation (with as suffix /-tēno-/?) seems unlikely (but cf.
Lejeune 1952b:123), so that some editors have assumed an (equally unlikely) pun, e.g.
‘«quello che tiene da Evio»’ (G. Giacomelli 1963:46, 1978:527);
(2) quto‹n›e uotenosio, with uotenosio = uo‹l›tenosio/uo(l)tenosio, referring to the same
person as uoltene (see below), in which case quto‹n›e could be explained (like arute MF
269 and larise MF 270, and MLF 372 and 373) as having an ‘epenthetic [ ]’ or as being
an accusative in -e(m) used for the nominative (see §9.2.2.1,4);
(3) quto ịeuotenosio (Vetter), with a name ịeuoteno-;
(4) quto neuotenosio (Pisani 1935 etc.), with a name neuoteno-;
(5) quto [n]euo tenosio ‘kèqwn novum Teni’ (Dirichs), taken up by Peruzzi, which
depends entirely on the [n], and assumes preservation of a Proto-Italic */e / (§3.2.5; see
also §18.3.3), and an early omission of word-final -m (§3.5.7a).
409
CHAPTER 12
(6) quto *e (?) uotenosio (quto ne uo(l)tenosio Steinbauer in Rix 1998:250 n.21), with
uotenosio = uo‹l›tenosio/uo(l)tenosio, the same person as uoltene (see below).
My own preference is either (2) or (6), although in (6) it is quite unclear how *e should
be interpreted. Note that regarding uotenosio as the same name as uoltene is only
possible if it is assumed that in uotenosio the syllable-final l was omitted either by
accident (uo‹l›tenosio) or intentionally (uo(l)tenosio): I would then rather regard it as an
error than assuming that it is the omission of a consonant that was weakened in syllable-
final position, as is discussed in §3.5.7b. However the name is read, the name is
certainly a genitive in -osio, not, as Knobloch wants it, an adjective euotenosio(s)
(1954) or euotenosio(m) (1966:48). Note that in the first of these interpretations it is
assumed that quto is masculine, in the second that it is neuter, while stems in ōn-stems
are usually feminine.
The next problem is duenom, which as it stands can hardly be anything else than
an adjective with quto (thus first Braun 1935:443). This is only possible if quto is a
neuter, which would seem to point to quto(m) or quto(r) rather than to an ōn-stem quto
/k ttō/ or /g ttō/, for it is unlikely that the latter would be neuter (thus G. Giacomelli
1978:527). If duenom is a noun, the only possible meaning would be bonum ‘a good
thing’ (TLL 2.2098,42-2099,27): the meaning ‘possession’ is apparently found only in
the (collective) plural (TLL 2.2101,6-81). Even if duenom could mean ‘possession’,
then titias duenom would still clash with eqo quto euotenosio: euotenosio certainly
gives the name of the owner, not of the potter, as Pisani suggested. Peruzzi’s eqo quto
neuo tenosio titias duenom duenas ‘ego kèqwn nouum Teni (et) Titiae bonarum
bonae’, involves a morphologically difficult a-stem genitive plural in -om (cf. §4.3.7).
This whole problem would be solved if duenom could in any way be interpreted as =
Latin donum, but it is impossible to derive duenom from donom /dōnom/ m */doħ3nom/
(or, alternatively, Latin donom /dōnom/ from duenom /d enom/) by any regular
process. Pisani therefore explained duenom as a hyper-Umbrism (!) for donom,
comparing cases where Umbrian ue corresponds to Latin o, which would have been
used to obtain a pun with duenas. This explanation is apparently adopted by G.
Giacomelli (1978) and Morandi: in my view, it is too far-fetched.
The text ends in salu[..]duoltene (certainly not Ribezzo’s salạẹmo or Braun’s
uoltẓne). The last e of uoltene is placed below the line for lack of space, as the (circular)
inscription had already ended up against its own beginning: to mark this point, a double
interpunct was added between uolten and eco. The earliest editors read no more of the
verb than salue.... Vetter’s saluteṃ has no ground at all and made him take uoltene as a
dative with an impossibly early monophtongization of /-ei/ o /-ē/̣ (§3.7.5). G.
Giacomelli and Peruzzi rightly read a d after the lacuna, making it possible to read
salue[to]d, for which cf. saluetod tita (Colonna 1980). G. Giacomelli’s saluẹ [me]d
(1963:44,46, 1978:527) remains an attractive alternative, however, in view of Latin ne
410
med malo statod CIL I2.4. Peruzzi’s salu[o me]d uoltene ‘nonne me saluum uultis?’
(assuming, like his eco quto neuo (see above), that the pitcher replaced one that was
broken) is very difficult, assuming (again) an early omission of -m (§3.5.7a), a uolte
apparently with a secondary ending, and an impossible -ne = nonne (?) affixed not to
the initial but to the final word of the question. Uoltene appears to be a vocative of a
name derived from the frequently occurring Volta, perhaps with a suffix /-ēno-/: cf.
§7.2.2.16. It is not necessary to compare Greek ca‹re kalÒj (Pisani) or to assume that
uoltene is the vocative of a */ tenos/ ‘Geliebter’ (Knobloch 1966:49).
From two autopsies in the Museo di Villa Giulia, Rome (inv. 43110). Bibliography: (I) Pallottino
1933:241 n.1 (autopsy); Pisani 1934; Buonamici 1934:356; Dirichs 1934:26; Buffa NRIE 974; Pisani
1935; – (II) Giglioli 1935:238-41 (autopsy); Braun 1935:436-45; Buonamici 1936; Ribezzo
1936:143-65 (autopsy); Pisani 1937; Norden 1939:266 (autopsy); Vetter 1939a:145-56 (autopsy);
Hofmann 1940:120-2; Pisani 1946:51 n.2; Lejeune 1952b:121-3; Vetter 1953:283-5 ¢242A-B²;
Knobloch 1954; Bassi 1956:51-2 ¢61²; G. Giacomelli 1963:44-6 ¢2a-b² (autopsy); Pisani 1964:349-51
¢152a-b²; Peruzzi 1964a:174 n.78; Knobloch 1966:48-9; Peruzzi 1967b; Devine 1970:112-4; G.
Giacomelli 1978:526-7 ¢2b²; Colonna 1980c:52-3; Agostiniani 1981:389; Agostiniani 1982:150 ¢590²;
Morandi 1982:56-7 ¢9²; Lejeune 1989:67 ¢8²; M. Mancini 2003:240-1; M. Mancini 2004. Photographs:
Giglioli 1935:239 fig.1; Ribezzo 1936:154 fig.A, tav.I; Bassi 1956 tav.XXI; G. Giacomelli 1963 tav.III.
Drawings: Ferreti in Giglioli 1935:240 fig.2 (reproduced in Lejeune 1952b:122 (partly), Morandi
1982:57, Lejeune 1989:67); Gasperini in Agostiniani 1981:389.
4. Scratched c.1½ times around an impasto pitcher (height 15 cm, letters 4-11 mm
high), similar to the one of EF 2 and EF 3. Seventh century.
e**azieput10ilepeḳapen20arufiaḳalẹ30ptiauessal40uetesociai50ofetioskai60osuelosa
ma70nossalueto80saluesseit90eiofeteqem100eneses*eie110
Dextroverse. Several of the letters have very curious forms or appear to be corrections
of other letters, as is explained below. The inscription leaves me with the impression of
having been written by someone who had only a slight grasp of the art of writing, or,
more specifically, who was perhaps not used to writing in this particular alphabet. The
reading of the difficult text has been greatly improved by Prosdocimi’s re-edition of the
text in CivFal, all the more as this is accompanied by excellent photographs, and my
discussion is largely based on his readings and interpretations (see also fig.12.4).
The traces of letters (ut?) read by Vetter before the text are at the very best
accidental scratches. Letter 1 is an e, followed by letters 2-3 that look like and have
been read as pe by most editors (pi Braun 1935:447-8). Prosdocimi, however, interprets
this as an n with a square c inserted underneath, comparing the c in Etr XLIV. The use
of c itself, square or otherwise, remains suprising, however, as elsewhere in the
inscription k is used before a. Letters 8-12 are , to be read with G. Giacomelli and
Prosdocimi as util (ạl Giglioli, χil Braun, ịịl Ribezzo, *l Vetter, itu Pisani 1937, 1946, iẹ
Pisani 1964). This first part of the text defies any attempt at interpretation. Prosdocimi
goes no further than en = ‘in’ and a hesitating adoption of Pisani’s lepe ‘vive’ (for
which, see below).
411
CHAPTER 12
1 2 3 4
5 6
Fig.12.4. Various details of EF 4 as shown in Prosdocimi’s photographs.
(1) Letters 2-3, usually read as pe, but by Prosdocimi as nc. (2) Letters 8-12, read by G. Giacomelli and
Prosdocimi as util. (3) Letter 16, the k of kapena. (4) Letter 27, the first letter of *aleptia. (5) Letters 30-
31, the ep of kaleptia. (6) Letters 104-107, read by Prosdocimi as seṣb. (Author’s drawings after the
photographs in Prosdocimi 1990.)
The next part is usually read as the names of women followed by ues saluete sociai (cf.
§7.2.1). The first is kapena: the first letter of this word, letter 16, is , apparently some
kind of correction (r from c Vetter, k from r or p, or r or p from k Peruzzi, k from r
Prosdocimi). The second is rufia: although the sides of the u are convex, Vetter’s rọfia
is groundless. Both readings presupposes an almost impossibly early
monophthongization of /o / (cf. §3.7.2 and Rix 1993a:85-6). The third name is more
difficult. The first letter (letter 27) is and has been read as a v (Giglioli, Braun; v or u
Pisani), but Prosdocimi’s photograph shows that it is rather a k (thus Ribezzo, Vetter, G.
Giacomelli, and Peruzzi). According to Peruzzi and Prosdocimi it was corrected from
a t (which was in fact an alternative reading proposed by Giglioli). Letters 30-31 are :
the first is usually read either as an e (Braun, Ribezzo, Pisani, Vetter, Peruzzi) or a k
(Giglioli, G. Giacomelli), the second may be an e (s Vetter). According to Prosdocimi,
the first letter is an i with horizontal bars that are probably unintentional, while the
second is p (kalịp ̣tia). I am not so certain with regard to the ị as Prosdocimi, especially
as the distance between this letter and the next is quite large, and would maintain the
possibility of an e. The phrase ues saluete sociai gives a problematic ues (cf. §4.7.3),
and a sociai that shows that at this date the pronominal nominative plural had replaced
the older nominal nominative plural in /- ās/ and /- ōs/ (cf. §4.2.6).
412
What follows is usually read as the name of several men followed by salueto. The
names are clearly legible as ofetios (ofeṭχos Giglioli, ofeiχos Braun), kaios, uelos (velqs
Braun), and amanos, but the relation between these names is not clear: see §7.2.1. Most
editors assume four separate names. Vetter took ofetios as the nominative plural of a
gentilicium, followed by three praenomina, but I doubt if at this date gentilicia already
played any role: Brauns praenomen + gentilicium + patronym + cognomen is certainly
impossible at this date. In all these interpretations uelos is nominative, a thematized
form of Etruscan Vel, although it could also be a genitive uelos (thus Pisani, who
interpreted kaios uelos amanos ‘Gaius Velis f. Amanus’ with the father’s name between
the praenomen and the gentilicium, as is done later in Umbrian and Volscian).
Salueto presents a problem, for, not only would we expect saluetod, as in the
seventh-century inscription from Osteria dell’Osa, saluetod tita (Colonna 1980c), but as
the number of men is plural, we might even expect even saluentod. Prosdocimi
(1990:304-5) attributes the absence of -d it to a different morphology, pointing to
Umbrian, where the original form */-to/ is not differentiated for number either: cf.
§5.2.4.4 and §5.3.1.16-17. Salueto seems to be followed by a similar expression salues
seite. Salues appears to be an i-stem nominative plural in /-ēs/, either a regular plural
from an elsewhere unattested i-stem adjective */sal i-/, or an irregular transfer of the i-
stem ending to another declination. Seite is can hardly be taken in any other way than
equal in sense to Latin sitis, in which case it must be an error either for s{e}ite or for
s‹ei›te: see §5.2.1.18. The greetings themselves, ues saluete sociai, salueto, and salues
seite have parallels in salue[to]d uoltene EF 3 and the inscription from Osteria dell’Osa,
saluetod tita (see Colonna 1980c).
The last part of the inscription starts with ofeteqemene (not Ribezzo’s
ofeiehemene). Peruzzi (1967b:122), finding a sequence qe difficult, read this as φe, but
Prosdocimi rightly chose for qe, pointing to Etruscan silqetenas ET La 2.3 and Latin
qetios CIL I2.2658 (text after Mancini 1979:370). The last problematic part is letters
104-107, , where several of the letters appear to be reworkings or corrections. The
readings of the various editors at this point are, understandably, wildly different (seịṣeie
Giglioli, sevesgseie Braun, seṣịpseie Ribezzo, seis(u)peie Pisani, seis[*]eṣie Vetter
1939a, seis.eṣie Vetter 1953, seị**eie G. Giacomelli 1963, sses**eie G. Giacomelli
1978). Prosdocimi read seṣbeie (his text erroneously has neṣbeie), with a b that would
presumably be a lettre morte: the photograph (tav.XIIc) shows that it is virtually
impossible to read otherwise, although it is equally impossible to see what it should
mean. Prosdocimi cautiously referred to South Picene bie at the end of Sp CH.1.
I mention only to reject Pisani’s e pea zie piẹ lepe kapena rufia uale eti aues
saluete sociai (i)ofet ụos kaios uelos amanos salueto salue sei te iofet eqemene sei
s(u)peie ‘In pia die pie vive, Capena; Rufia, vale et aveas; salvete, sociai! Futuit vos
Gaius Velis f. Amanus. Salveto! Salve, sive te futuit equitabiliter, sive *suppediter’
(thus Pisani 1964:349). Apart from impossible readings pea, piẹ, uale eti aues, uos, sei
413
CHAPTER 12
s(u)peie, this contains several oddities: (1) the absence of -d in the ablatives pea and die
and in the adverbs piẹ, eqemene, and s(u)peie (its omission in salueto is at best an
exception, not the rule); (2) eti side by side with (i)ofet, iofet m */obheti/, as both
should have lost /i#/ by the same rule; (3) aues as a subjunctive of aueo; (4) an
accusative uos side by side with the nominative ues; (5) the active sense of eqemene,
formed with same suffix as the Greek medial participles in -menoj. Of this reading, lepe
(which has no parallel in Italic at all) has been adopted by some recent authors. Rix
(1993a:86) considers the possibility that iofet may be iubet, but (like rufia) this requires
a very early monophthongization of /o /.
From autopsy in the Museo di Villa Giulia, Rome (inv. 43111). Bibliography: Giglioli 1935:241-3
(autopsy); Braun 1935:445-51; Buonamici 1936:424-6; Ribezzo 1936:165-8 (autopsy); Pisani 1937:
238-45; Vetter 1939a:151-6 (autopsy); Pisani 1946:51 n.2; Vetter 1953:285-7 ¢243²; Bassi 1956:51-2
¢62²; G. Giacomelli 1963:44-6 ¢3² (autopsy); Pisani 1964:349-51 ¢152a-b²; Peruzzi 1967b (autopsy); G.
Giacomelli 1978:526-7 ¢3²; Prosdocimi 1990 (autopsy); Rix 1993a:85-6. Photographs: Giglioli
1935:241 fig.3; Ribezzo 1936:162 fig.B, tav.II-III; Bassi tav.XXI,62; G. Giacomelli 1963 tav.IV;
Prosdocimi 1990 tav.I-XII. Drawing: Giglioli 1935:242 fig.4.
12.4. The other Early Faliscan inscriptions
5. The eighth- to seventh-century necropoles of Montarano, situated to the northeast of

the town, on the north side of the Rio Maggiore, are the oldest necropoles of Civita
Castellana, and probably belong to the early settlement on Colle di Vignale. These
necropoles were excavated in 1888-1890 (FI II.2 pp. 4-5), when the following
inscription was found, scratched on a seventh-century olla decorated with two horses,
from tomb LVII/43:208
eitam
Dextroverse. The form of the t, , is unexpected at this date, especially in a dextroverse
inscription (cf. §11.2.3). Vetter, comparing Praesamnitic te·cliia·m Ps 16, 17 and Oscan
spuríieis culcfnam Cm 27, suggests that eitam may be an accusative. This is preferable
to dividing the word as eita m(arci) (Peruzzi, comparing the interpretation of tulom MF
72 as tulo m(arci)). For these problematic ‘isolated accusatives’ see §8.1.4 and §9.2.2.
Peruzzi also suggested that eitam may an adaptation of the Etruscan itan Ve 3.21 and
may denote a type of vase. In an earlier publication (Bakkum 1991) I rejected this
interpretation, but I am now prepared to adopt it on the basis of e.g. tafina Etr XXXIII.
As a parallel, I then suggested fourth-century eiṭma leicunas ET Cm 2.83, 2.84; other
possible parallels are South Picene íitas AP.3 and ẹitah AQ.1. G. Giacomelli’s equation
of eitam with Oscan eitiuvam Po 3, eituam TB 19 etc. is unconvincing.
208
Nogara (in Herbig CIE 8001) erroneously gave the provenance as tomb LVI.
414
From autopsy in the Museo Archeologico dell’Agro Falisco, Civita Castellana (inv. 3519). Bibliogra-
phy: Herbig CIE 8001 (autopsy); Vetter 1953:289 ¢250²; G. Giacomelli 1963:56 ¢22²; Peruzzi 1964a:
169-70; G. Giacomelli 1978:78-82 ¢5²; Colonna 1980d; FI II.2 p.86 (autopsy); Rix ET Fa 0.3; Bakkum
1991 (autopsy); Bakkum 1992:2. Drawings: Nogara in CIE 8001; FI II.2 p. 86.
6-7. Scratched, 6 beside a graffito of a human profile, 7 upside down,209 on a bucchero

cup from tomb LXXXVI/6 of the La Penna necropolis. Sixth century.
ekolartos
ekokaisiosio
Sinistroverse, with s (the apparent in 6 is a slip of the stilus). For the use of k before
o, see §11.2.2-3. In 6, lartos is a genitive: there is no reason to interpret it as a
nominative, as does Colonna (1977). In 7, a chance point in the second o (not in Cozza
& Pasqui’s drawing) led the earliest editors to read kaisiθsio. Herbig was the first to
read kaisiosio (1910) and to interpret this as a genitive (CIE), although he still
considered Danielsson’s kaisi oṣṭio (‘ostiu(m)’=‘poculum’?). His interpretation was not
generally adopted (Buonamici preferred Danielsson’s, and Stolte and Ribezzo regarded
kaisiosio a dittography) until it was confirmed by the publication of uotenosio EF 2 in
1933 and of aịṃiosio EF 467* in 1952. For Knobloch’s suggestion that kaisiosio is an
adjective kaisiosio(s) (1954) or kaisiosio(m) (1966:48) has not been adopted (cf.
§4.4.2). For the formula ego OWNERGEN, see §8.7.1.
From autopsy in the Museo di Villa Giulia, Rome (inv. 759). Bibliography: Cozza & Pasqui
1887a:175 (autopsy); Gamurrini 1887:61-2; †Dennis 1890:161; Gamurrini 1894:340; Herbig 1910:97-8
(autopsy); Jacobsohn 1910:4 ¢12²; Herbig CIE 8163; Buonamici 1913:54-55 ¢6²; [Weege in Helbig
1913:368]; Herbig 1917:33; [Della Seta 1918:59 (autopsy)]; Stolte 1926:28 ¢11²; Ribezzo 1930:98;
Lejeune 1952b:123; Vetter 1953:288 ¢245a-b²; Knobloch 1954; G. Giacomelli 1963:48-9 ¢4a-b²
(autopsy); Pisani 1964:342 ¢146Ca-b²; Peruzzi 1965:149 n.2; Knobloch 1966; Devine 1970:23-4;
Colonna 1977:104; G. Giacomelli 1978:528-9 ¢4a-b²; FI II.2 p.151 (autopsy); Agostiniani 1982:151
¢592²; Lejeune 1989:66-7 ¢6². Photograph: G. Giacomelli 1963 tav.V. Drawings: Cozza & Pasqui
1887a:175 (reproduced in CIE 8163); Herbig CIE 8163; Lejeune 1952b:122; Lejeune 1989:66.
8-9. Scratched, 8 close to, and 9 behind a broken-off handle on a dolium (height 63.8
cm, rim 24 cm on the outside; letters 25-35 mm high) from tomb III of the Ponte
Lepre necropolis.
ud
tele*[1-2?]medfifiked
Sinistroverse, Faliscan alphabet. The letters ud are unexplained. Gulinelli’s edition of
the text (the first based on autopsy since Herbig’s, as well as the first to include
photographs and a drawing) now clearly shows that the first letters are tele and that the
209
When the cup stands on its foot. Vetter read the inscription as dextroverse with an upside-
down a, but the other letters, especially the e, show that this is incorrect.
415
CHAPTER 12
earlier readings heva, heva, veza, vepa, or zvea (Herbig 1913), ḥẹṿa (Herbig 1923,
Pisani), ḥẹṿa (Herbig 1923), -e-a (Vetter), and [...]a (G. Giacomelli) are impossible.
Tele is followed by a shaft with a small side bar that may or may not be intentional: an i,
a, v, or e, according to Gulinelli. These first signs are separated from the rest by a space
of 4 cm that is apparently empty: Gulinelli saw only very slight traces here, but noted
that the distance of the letters elsewhere in the inscription is very regular, which would
make the empty space a strange anomaly. Fifiked has been interpreted as a third
singular perfect of fingo since Herbig, in which case the k may be an early instance of
the occasionally occurring Faliscan custom to use k to denote /g/ (see §11.2.4). It is
unlikely that it is a form of facio (cf. Lejeune 1955:148-50), although this idea has
recently been revided (e.g. Poccetti 2005:31-5, Berenguer & Luján 2005:206-7): see
§5.2.1.7-8. The phrase has a parallel in mama z[e]xtos med f[.f]ịqod EF 1: see §8.8.2.
Bibliography: (I) Herbig 1913:71-4 (autopsy); Herbig 1923:231; Vetter 1953:291 ¢257²; Lejeune
1955:146-8; G. Giacomelli 1963:51-2 ¢11a-b²; Pisani 1964:345-6 ¢148²; Agostiniani 1982:151 ¢593². –
(II) Gulinelli 1996 (autopsy). Photographs: Gulinelli 1996:226 figs.1 and 3. Drawing: Gulinelli
1996:226 fig.2.
10. Scratched under the foot ( 8.5 cm; letters 8-10 mm high) of an Attic black-
varnished cup from the area of the Tempio Maggiore on Colle di Vignale. First quarter
of the fifth century.
apolonos
Sinistroverse, with a �, p , n �, and reversed s ( ). There is enough space both before
and after the word to show that the text consisted of one word only. Just as the genitive
is used in Besitzerinschriften to indicate a human owner, so it is used here to designate
the god as the owner of the cup dedicated to him (Pisani). It is not necessary to assume
that a verb such as sum or an adjective such as sacer is lost or was implied, as does G.
Giacomelli. The use of the isolated genitive in Faliscan dedications is discussed in
§8.10.1. For the worship of Apollo in the ager Faliscus, see §2.3.4.
From autopsy in the Museo Archeologico dell’Agro Falisco, Civita Castellana (inv. 7377). Bibliogra-
phy: Jacobsohn 1910:3 ¢3²; Herbig CIE 8030 (autopsy); Taylor 1923:78; Vetter 1953:292 ¢260²; G.
Giacomelli 1963:59 ¢31²; Pisani 1964:341-2 ¢146A²; Moscati 1983:66 (autopsy); Comella 1986:171
¢28² (autopsy); Degrassi & Krummrey CIL I2.2912; Wachter 1987:397. Photograph: Moscati 1983
tav.XVIII,b-c. Drawing: Herbig CIE 8030 (reproduced in CIL I2.2912); Comella 1986 tav.76,R49.
416
Chapter 13
The inscriptions from Civita Castellana (Falerii Veteres) I
13.1. Civita Castellana and its inscriptions
Like many South Etrurian towns, Falerii was situated on a spit of land enclosed by deep
river-gorges. On the eastern side, a narrow saddle connects the site of the modern town
with the Colle di Vignale, the place of the oldest settlement and probably of the Faliscan
arx. Both hills are surrounded by the wide but steep valleys of the Rio Filetto to the
south and of the Treia to the east, and by the sheer-sided ravine of the Rio Maggiore to
the north. A short history of the town has been given in §2.4-6. The results of the many
excavations at Civita Castellana have mostly been published in FI II.2: a more recent
overview may be found in Moscati 1990.
Civita Castellana yields the great majority of the Faliscan inscriptions, even
allowing for errors in the attribution (cf. §1.4.5). Most of the inscriptions are from the
necropoles along the ancient roads radiating from the town. A smaller number comes
from the various temple sites (§14.2). I have tied the dating of the inscriptions from
Civita Castellana to the apparent desertion of the town after the war of 241 BC (§1.4.4,
§11.1.3), classing all inscriptions from Civita Castellana as Middle Faliscan unless there
are reasons to suppose otherwise. There are indications that some of the temples
remained in use until the second century, and at least two inscriptions from the temple
sites (LF 112 and LtF 131) appear to belong to this period. Families owning tombs near
the old town may also have continued to use these: the sepulchral inscriptions in the
Latin alphabet, LtF 140 and 171-174, may belong to such late burials. Another inscrip-
tion apparently in the Latin alphabet is MF/LtF 21.
13.2. The necropolis of Contrada Le Colonnette
Contrada Le Colonnette lies to the northeast of Civita Castellana, on the other side of
the Rio Maggiore, a little to the north of the station of the Roma-Viterbo railway line.
From there, an ancient hollow road leads down to the Fossa dei Cappuccini and the
temple of Contrada Celle. The discovery of Faliscan remains in this area by Count
Feroldi de Rosa (cf. Del Frate 1898:72-3), including the discovery, in 1873, of the small
fourth- to second-century temple known as the ‘Ninfeo Rosa’ (cf. Eroli 1875, Pasqui
1887:93 n.1, and Blanck 1990), led to an excavation in the necropolis in 1890. This
417
CHAPTER 13
excavation, published in FI II.2 pp.206-10, yielded the ‘Ceres-inscription’ EF 1 (§12.2)

and MF 14; MF 15-19 were discovered by Mengarelli during the 1880s and 1890s (cf.
Thulin 1907:264).
11-12. Cut, 11 to the right (letters c.9 cm high) and 12 to the left (letters c.8 cm high) of
the entrance to a tomb.
uolti teti
cau‹i›o�paui|ceo�
ḷ[oc]ịes�cela
Sinistroverse, Faliscan alphabet. The interpunct in 11 consists of a vertical stroke: see
§11.2.4. The first line of 12 turns around a corner in the rock-face after paui. Following
the lacuna, Thulin’s drawing shows only a small part of a shaft. – Uolti teti may be an
abbreviated nominative (Herbig) or a genitive (Vetter, G. Giacomelli): see §8.8.1. Both
Thulin and Herbig asserted from autopsy that the first word of 12 is caulo. If this is not
an error for the very frequent cauio (the same error of � for � occurs in MF 32), it can be
compared to Etruscan Caule, Latin Caulius (Herbig); Thulin’s and Vetter’s Etruscan
Cavili, and Latin Cavilius, Gavilius, or Gavillus assume a syncope that is not in
accordance with Faliscan phonology (§3.6.6.2). The name Pauicius occurs also in ce 
paui[ceo  ru?]so in LtF 290. The restoration ḷ[oc]ịes goes back to Herbig (CIE),
although similar restorations had been proposed earlier by Thulin and Herbig (1910).
The problem of this inscription is its awkward syntax. Ḷ[oc]ịes, probably a
genitive (see §9.2.2.4), belongs with caulo � pauiceo, ‘Gavius Pavicius [son] of Lucius’:
taking it with cela would give ‘Lucius’ tomb’, but when a person is designated as the
owner of a tomb, the expected onomastic formula would be PRAENOMEN GENTILICIUM
(as in the other inscriptions of this type, MF 84-84 and MLF 285), not PRAENOMEN: cf.
§7.3. Even more problematic is the syntactic relation between the nominatives cau‹i›o �
pauiceo and cela (unless Herbig’s and Buonamici’s very improbable interpretation of
cela as a cognomen is adopted, cf. §7.9.1.5). Thulin pointed to the similar lack of
congruence in poplia � calitenes | aronto � cesies | lartio � uxor MF 265 (cf. §9.2.2.4),
but his alternative, taking caulo � pauiceo (as he read it) as a dative (‘C.P. Lucii (filio)
cella’) is unattractive, as comparable inscriptions have either the nominative or the
genitive (§8.8.1, §11.1.4,1.a). Vetter and G. Giacomelli adopted Thulin’s suggestion
that the lines cauio � pauiceo � and ḷ[oc]ịes � cela belonged to two different inscriptions.
This is not impossible, in spite of the interpunct after pauiceo, but the second text,
ḷ[oc]ịes � cela, can then hardly be complete.210
I have considered combining these inscriptions as
210
“Die Zweite [Zeile] ist jedenfalls mehr zerstört, vielleicht absichtlich ausradiert, als das Grab
den Besitzer gewechselt hat” (Thulin 1907:274).
418
THE INSCRIPTIONS FROM CIVITA CASTELLANA (FALERII VETERES) I
cau¢ i² o � pauiceo � � � � � � � �
uolti tetiňņņŉḷ[oc]ịes � cela �� ░░��ňņņŉ� |� �
(‘Gavius Pavicius. – The tomb of Voltius Tettius son of Lucius.’)

but the epigraphical differences between the two inscriptions are perhaps too great for
this. However, this would solve the problem of the awkward syntax.
Bibliography: Thulin 1907:273-4 ¢7-8² (autopsy); Herbig 1910:101-3 ¢15² (autopsy); Jacobsohn 1910:3
¢9²; Herbig CIE 8076-8077 (autopsy); Buonamici 1913:56 ¢8-9²; Vetter 1953:295 ¢271a-b²; G. Gia-
comelli 1963:74 ¢68,I-IIa-b² (autopsy); Loicq 1965:697. Photograph: G. Giacomelli 1963 tav.VIII.
Drawings: Thulin 1907:273, 274 (reproduced in CIE 8076-8077).
13. Cut over the entrance of a tomb c.10 m to the right of the one of 11-12 is a badly
legible inscription. My reading, which is in effect a guess rather than a certainty, is
based on the drawings by Thulin and Herbig (reproduced in fig.13.1), and on G.
Giacomelli’s photograph, which is unfortunately unclear.
t qṿintoi q hạloi
tetia t � front teti ạtronị
Fig. 13.1. Thulin’s and Herbig’s drawings and readings of MF 13.

left: Thulin’s drawing and reading. (From Thulin 1907:275.)
right: Herbig’s drawing and reading. (From CIE 8078.)
ohạθ̣i
tetiatron (or tetiatronị ?)
Sinistroverse, Faliscan alphabet. Thulin read a very narrow before the first o,
coinciding with a small edge in the stone, but Herbig’s drawing shows no such letter
and he doubted its existence. The o’s are angular and pointed at the bottom, which led
Thulin and Herbig to read the first o as a q in spite of the fact that the Middle Faliscan
alphabet has no separate sign for q (§11.2.2-4). The third letter is , which Thulin
interpreted as a ligature vi (� ), adding a second non-Faliscan letter to the text. Thulin’s
reading, with its many ligatures and qv instead of the usual Faliscan cu,211 is unaccept-
211
Vetter erroneously credited Thulin also with a ligature qv.
419
CHAPTER 13
able to me. The third letter is either an n or an a, but neither ohnoi nor ohaoi makes any
sense. I therefore suggest that the second o, and perhaps also the first, might be a θ. This
letter occurs occasionally in Middle Faliscan inscriptions, and although it is virtually
always written as , there are parallels for the omission of the central point (see
§11.2.4). This gives ọ hạθ̣i or θ̣ hạθ̣i, with hạθ̣i a gentilicium comparable to Latin
Fadius, cf. cauia | hadenia MLF 360. The second line is more legible, except for the
last sign. I assume that this is n; Herbig and G. Giacomelli took the elongated shaft of
this letter as an i written over the n, which would give the expected atroni. If read in this
way, the resulting hạθ̣i and teti atronị are either genitives or abbreviated nominatives,
perhaps rather the former, cf. uolti teti MF 11, if that name is to be read in isolation.
Bibliography: Thulin 1907:274-5 ¢9² (autopsy); Herbig CIE 8078 (autopsy); Buonamici 1913:57 ¢10²;
Herbig 1914a:239 ¢6²; Vetter 1953:296 ¢272²; G. Giacomelli 1963:74-5 ¢69² (autopsy). Photograph: G.
Giacomelli 1963 tav.VIII. Drawings: Thulin 1907:275 (reproduced in CIE 8078); Herbig CIE 8078.
14. Painted in red on a tile (66u40 cm, letters 7-10 cm high) from tomb CXXXVI.
cauipi:leueli
filea
Sinistroverse, Faliscan alphabet. The first line bends downwards at the end, showing
that the inscription occupied just one tile. Conway’s cauiại is an error. Thulin inter-
preted the text as ‘Cauipi Leueli filia’, Herbig as ‘Gaia Vibia Levelii filia’, with uipi as
an Etruscan feminine: see §9.2.2.2a. Vetter took uipi as the genitive of the father’s
praenomen, and this has been adopted by all editors except Morandi (who followed
Herbig). It is the only instance of FILIATION added to of PRAENOMEN and the only
instance of a FILIATION consisting of PRAENOMEN GENTILICIUM (§7.5.1).
From autopsy in the Museo di Villa Giulia, Rome (inv. 3733). Bibliography: Conway 1897:379 ¢334²
(autopsy); Thulin 1907:281-2 ¢15² (autopsy); Herbig 1910:88-9 ¢8² (autopsy); Jacobsohn 1910:3 ¢8²;
Herbig CIE 8075; Buonamici 1913:58 ¢12²; Bormann CIL XI.7516; Vetter 1953:295 ¢270²; G. Gia-
comelli 1963:73 ¢67² (autopsy); Pisani 1964:336 ¢143B²; Dohrn in Helbig/Speier 1969:690 ¢2775²
(autopsy); G. Giacomelli 1978:532 ¢9²; FI II.2 p.207 (autopsy); Morandi 1982:58-9 ¢11². Photograph:
G. Giacomelli 1963 tav.VII. Drawings: Herbig CIE 8075; Morandi 1982:59.
15-19. Painted212 inside a tomb discovered by Mengarelli “vor vielen Jahren” (Thulin
1907:264-5 with plan213) but lost already in 1908 (Herbig CIE 8070-8074). The tomb
belonged to the gentes Neronia (15-16) and Firmia (18-19): the former gentilicium also
occurs at Fabbrica di Roma (LtF 325, 328) and the Grotta Porciosa site (LtF 340), the
latter also at Falerii Novi (LF 213) and at Vignanello (MLF 302).
212
G. Giacomelli’s “scolpite nel sasso e dipinte” is a misinterpretation of Herbig’s “titulus qui a
sinistra parte loculi inferioris [c.q. superioris], qui a latere dextro introitus [c.q. ingressus] ... rupi
insculptus est, calce in saxo pictus”.
213
It is not clear how correct this plan is: the drawing of MF 15 shows two loculi to the right of
the inscription, which is impossible if the plan is correct.
420
15. Painted in white beside a loculus.

uolta
neroni
cafi
Sinistroverse, Faliscan alphabet. The point in neroni is accidental. Neroni is an
abbreviated nominative, not an Etruscan form (as Pisani suggested): see §9.2.2.2a.
Bibliography: Thulin 1907:267-8 ¢4d,α²; Herbig 1910:109 n.1; Jacobsohn 1910:3 ¢6²; Herbig CIE
8070; Buonamici 1913:57-8 ¢11²; Vetter 1953:294 ¢269a²; G. Giacomelli 1963:73 ¢66,I²; Pisani
1964:335-6 ¢143A². Drawing: Mengarelli in Thulin 1907:267 (reproduced in CIE 8070).
16. Painted in red on plaster beside a loculus.

[.......]
neṛ[oni.]
ị[......]
Sinistroverse, Faliscan alphabet. The second line probably consisted of a form of the
gentilicium Neronius, although only a minimal trace of the r is preserved. The third line
will have contained the filiation; if the shaft that is all that is preserved of its first letter
is an i, it may have been ị[uneo], ị[unea], or ị[unai  f].
Bibliography: Thulin 1907:264-8 ¢4d,β²; Herbig CIE 8071; Vetter 1953:294 ¢269a²; G. Giacomelli
1963:73 ¢66,II². Drawing: Mengarelli in Thulin 1907:268 (reproduced in CIE 8071).
17. Painted in red on plaster beside a loculus. The letters of the second line are smaller
than those of the first and the third; perhaps the lines were written at different times.
iiiḷ[.......]
[..........]naị[?---]
[....]ouxo [
Sinistroverse, Faliscan alphabet. The l is , a rare form in the ager Faliscus, and
associated rather with the Latin alphabet. Thulin rightly pointed to LtF 251, where
lectus is used in the sense ‘loculus’ or ‘place in a loculus’ (cf. §6.3.39):  iii  ḷ[.......]
would then be ‘the third lectus ...’ or ‘three lecti ...’: cf. lete zot xxiiii ‘lecti sunt XXIIII’
MLF 285 and perhaps let in MLF 361. The rest is difficult: [---]naị[ can be a dative or a
genitive, but the space after uxo in Mengarelli’s drawing shows that this word is
apparently complete and therefore a nominative, implying that the woman’s name was
likewise a nominative. (Herbig’s uxo[ri] [sic] ignores this space.) The [....]o preceding
uxo is probably the genitive of the husband’s name, e.g. [arut]o or [lart]o (Vetter).
Bibliography: Thulin 1907:265 ¢4a²; Herbig CIE 8072; Vetter 1953:294 ¢269b²; G. Giacomelli 1963:
73 ¢66,III². Drawing: Mengarelli in Thulin 1907:265 (reproduced in CIE 8072).
18. Painted in red on plaster underneath a loculus.

poplia
ḥirmia
421
CHAPTER 13
Sinistroverse, Faliscan alphabet. The first letter of the second line is . Thulin read this
as a reversed n, but in view of the Faliscan attestations of Firmius (§7.8.1.62), Herbig’s
ḥirmia has been adopted by all later editors. A similar h occurs in MF 88 (cf. §11.2.3).
Bibliography: Thulin 1907:266 ¢4b²; Herbig CIE 8074; Vetter 1953:295 ¢269c²; G. Giacomelli 1963:
73 ¢66,V². Drawing: Mengarelli in Thulin 1907:266 (reproduced in CIE 8073).
19. Painted in white underneath a loculus.
Fig.13.2. Mengarelli’s drawing of MF 19.

(From Thulin 1907:266.)
[---]ḥiṛṃeoiu
Sinistroverse, Faliscan alphabet. The letters given in Mengarelli’s drawing are meaning-
less (..ia II ueθiu. Thulin, θure ueθiu. Herbig, ọi-ạiiueoiu. Vetter, ọi.ạIIueoiu. G.
Giacomelli). Taking �� as a damaged m, however, gives [---] ḥiṛṃeo iu , with a
gentilicium as in poplia | ḥirmia MF 18 (or perhaps [---]**ṛịṇeo) followed by a filiation
iu.
Bibliography: Thulin 1907:266-7 ¢4c²; Herbig CIE 8073; Vetter 1953:295 ¢269d²; G. Giacomelli
1963:73 ¢66,IV². Drawing: Mengarelli in Thulin 1907:266 (reproduced in CIE 8073).
13.3. The necropolis of Contrada Celle
Contrada Celle lies to the northeast of Civita Castellana, to the north of the Rio
Maggiore. The necropolis spread to the southeast of the ancient road to the Tiber
crossing near Borghese (Ward Perkins & Frederiksen 1957:152-3). With the exceptions
of MF 39, a chance find, and MF 40, discovered in the 1950s, all inscriptions from this
site were found during the excavation of the necropolis in 1888. These excavations have
been published by Pasqui (1887) and in FI II.2 pp.99-143 with map p.100).
20. Scratched inside a red-varnished cup from tomb LXXVI/50. Known only through
the apograph in FI II.2 and Nogara’s copy of the apograph in the Museo di Villa Giulia
inventory.
caisioi
422
Sinistroverse, Faliscan alphabet. The o contains an accidental point. Caisioi is usually

interpreted as a dative, but it is perhaps rather a genitive (Pisani): see §4.4.4.
Bibliography: Jacobsohn 1910:3 ¢1²; Herbig CIE 8002; Herbig 1914a:239 ¢4²; Lejeune 1952b:125;
Vetter 1953:289 ¢251²; G. Giacomelli 1963:57 ¢23²; Pisani 1964:342-3 ¢146E²; Devine 1970:17-9; FI
II.2 p.126 (autopsy). Drawing: FI II.2 p.126.
21. Scratched near the rim inside a black-varnished plate (height 4.5 cm, 18 cm;
letters 4-7 mm) from tomb 94.
ṛica
Dextroverse: the r is . Lejeune regarded ṛica as a woman’s name, perhaps in the
genitive, but this involves omission of -s after a long vowel, which is rare (§3.5.7d).
Bibliography: Herbig CIE 8020; Lejeune 1952:117; Vetter 1953:294; G. Giacomelli 1963:58-9 ¢29,III²;
FI II.2 p.144 (autopsy). Drawing: Nogara in Herbig CIE 8020.
22-30. The following inscriptions are all from tomb CXIX/89.

22-27. Scratched on six similar red clay saucers ( 12 cm, let. 5-15 mm).
turia
turia
turia
turia
turia
turia
Sinistroverse, Faliscan alphabet. The inscriptions do not appear to be all in the same
hand, or even in the same alphabet: in 26, a is , in the others � (of a more squarish
shape in 24 and 27); in 25 and 26 the r is , in the others . Turius is also attested in
caui � turi MF 273, and probably also in ca : tu MF 38: it is unnecessary to assume a
connection with Greek Turè (as did Herbig).
Bibliography: Thulin 1907:307 ¢60² (autopsy); Herbig CIE 8003-8008 (autopsy); Vetter 1953:290
¢252²; G. Giacomelli 1963:57 ¢24,I-VI²; FI II.2 pp.137-8 (autopsy). Drawing: Pauli in Herbig CIE
8003-8008.
28. Scratched inside a cup ( 13 cm; letters 14-20 mm high) described as red-varnished
in the Museo di Villa Giulia inventories quoted by Thulin, but as black-varnished in FI.
sta
Either sinistroverse with a reversed s and an upside-down t (FI) or dextroverse with an
upside-down a (all other editors). For the interpretation, see MF 29.
Bibliography: Thulin 1907:307 ¢59² (autopsy); Herbig CIE 8013; Vetter 1953:290 ¢254b²; G. Gia-
comelli 1963:58 ¢27²; FI II.2 p.137 (autopsy). Drawing: Thulin 1907:307 (reproduced in CIE 8013).
423
CHAPTER 13
29. Scratched inside a cup (height 4 cm, 12.2 cm, letters 11-16 mm high) described
as red-varnished in the inventory of the Museo di Villa Giulia (quoted by Thulin), but as
black-varnished in FI.
statuo
Sinistroverse, Faliscan alphabet with reversed s. Thulin compared sta MF? 128, which
he read as a verb sta(t) ‘ist aufgestellt’ (in a sacral sense). Statuo could then be “das
Aktivum dazu” (Thulin 1907:307), apparently the equivalent of Greek ¢nat…qhmi, while
sta MF? 28 could be either sta(t) or sta(tuo). This was adopted by Herbig and G.
Giacomelli, but I do not adopt Thulin’s interpretation of sta MF? 128, and regard sta
and stat with Vetter as abbreviations of Statius, and uo as an abbreviated gentilicium.
Bibliography: Thulin 1907:307 ¢58² (autopsy); Herbig CIE 8012; Vetter 1953:290 ¢254a²; G. Gia-
comelli 1963:58 ¢26²; FI II.2 p.137 (autopsy). Drawing: Thulin 1907:307 (reproduced in CIE 8012).
30. Scratched inside a black-varnished Etrusco-Campanian cup named in FI among the

finds from this tomb, but published by Herbig and G. Giacomelli as incertae originis.
ce
Sinistroverse.
Bibliography: Herbig CIE 8294; G. Giacomelli 1963:61 ¢38,III²; FI II.2 p.137 (autopsy). Drawing:
Herbig CIE 8294.
31-32. Usually treated together because of their identical contents are two inscriptions
on black-varnished cups from two different tombs of this necropolis.
31. Scratched on the bottom of a black-varnished cup ( 12.4 cm; letters 6-12 mm
high) from tomb CXX/105.
loifiṛtato
Sinistroverse, Faliscan alphabet. The first i is either written as � (as an error for �) or
damaged by a scratch that makes it look like an �. The r is , an example of the
confusion of (r) and (a) rather than an example of : see §11.2.4. The reading
loifirtato goes back to Herbig (loifia tato Thulin, loifirtatio Nogara). See under 32.
Bibliography: Thulin 1907:306 ¢56² (autopsy); Jacobsohn 1910:3 ¢2a²; Jacobsohn 1911:464; Herbig
CIE 8010 (autopsy); Buonamici 1913:53-4 ¢4²; Vetter 1953:290 ¢253²; G. Giacomelli 1963:57 ¢25,I²;
Pisani 1964:342 ¢146B,a²; FI II.2 p.134 (autopsy). Drawings: Thulin 1907:306 (reproduced in CIE
8010); Herbig CIE 8010.
32. Scratched inside a black-varnished cup (height 8.8 cm, 13.2 cm; letters 5-8 mm
high) from tomb CXXIII/92.
loifirtato
Sinistroverse, Faliscan alphabet. Leaving aside Thulin’s interpretation (based on the
misreading loifia tato), all editors have interpreted loifirtato as the genitive of a noun
424
corresponding to Latin libertas.214 The meaning of the text is unclear. Loifirtato has
been taken as ‘libertatis (ergo)’, a dedication made on the occasion of enfranchisement
(thus Herbig and G. Giacomelli), or as ‘Libertatis’, the name of a deity (Herbig,
Jacobsohn 1911:464, Buonamici, Vetter 1953:410),215 with a genitive as in apolonos EF
10 (cf. §8.11.1) Herbig notes that in the latter case loifirtato could also be a plural
genitive. Problematic is that both interpretations presuppose the presence, in tombs, of
dedications that are apparently not related to the funeral rites. Pisani tried to avoid this
problem by interpreting loifirtato as the genitive of the woman’s name ‘Libertas’ read
by him in MF 41 as l]ọịfiṛṭạ(s), but the correct reading there is l]oifirta ‘liberta’. For the
diphthong, see §3.7.3.
Bibliography: Thulin 1907:305 ¢55² (autopsy); Jacobsohn 1910:3 ¢2b²; Jacobsohn 1911:464; Herbig
CIE 8011 (autopsy); Buonamici 1913:53-4 ¢5²; Vetter 1953:290, 410 ¢253²; G. Giacomelli 1963:57
¢25,II²; Pisani 1964:342 ¢146B,b²; FI II.2 p.132 (autopsy). Drawing: Thulin 1907:305 (reproduced in
CIE 8011); Herbig CIE 8011.
33. Painted in ochre inside a black-varnished plate ( 16 cm, letters 8 mm high) from
tomb CXXXIV/90.
lo�cr
Sinistroverse, Faliscan alphabet. Lo is ‘Lo(ci-)’ (Thulin), not Vetter’s ‘Lo(uci)’, nor
Herbig’s lθ ‘L(ar)θ’. For cr cf. the gentilicium Grae- in crạ[---] MF 141, cre[---] MF
142, cṛ[---] MF 143 (Herbig); kreco MF 147, adduced by Thulin and G. Giacomelli, is
a praenomen.
Bibliography: Thulin 1907:306 ¢57² (autopsy); Herbig CIE 8018 (autopsy); Vetter 1953:290 ¢256b²; G.
Giacomelli 1963:59 ¢30²; FI II.2 p.111 (autopsy). Drawing: Thulin 1907:306 (reproduced in CIE 8018).
34-37. On four vessels from tomb CXXXV/98.

34-36. Scratched inside a black-varnished cup ( 13 cm; letters 8-11 mm high) and two
black-varnished plates ( 17 cm, letters 7-15 mm high and 18 cm, letters c.7 mm
high respectively).
serui
serui
serui
Sinistroverse, Faliscan alphabet. The s is reversed ( ) in 34 and 36; in 35, it is �, but
written so casually that it looks like an . Serui is a genitive or perhaps an abbreviated
214
Herbig also discussed possible interpretations of his loifir tato, ‘Tatus liber’, ‘Tatus liber
(=puer)’, and ‘Liber pater’.
215
Cf. (Latino-)Venetic [---]firtati, published by Innocenti Prosdocimi (1976:269-72), perhaps
connected to the enfranchisement of the Latin citizens by the lex Iulia in 90-89 BCE.
425
CHAPTER 13
nominative: it is not necessary to assume that it is an Etruscan nominative (as Herbig

proposed, cf. §9.2.2.2a). The name can be a praenomen or a gentilicium.
35 from autopsy (Museo di Villa Giulia, inv. 2271). Bibliography: Herbig CIE 8014-6 (autopsy);
Weege in Helbig 1913, p.372; Vetter 1953:290 ¢255²; G. Giacomelli 1963:58 ¢28,I-III²; FI II.2 pp.141-2
(autopsy). Drawing: Herbig CIE 8014-8016.
37. Scratched inside a small cup ( 8.5 mm; letters 7-10 mm high).
arn
Sinistroverse. Etruscan alphabet. The a is ; the r is the Etruscan . The points that
between the letters (arn) are in all probability accidental or decorative. Herbig
compared Etruscan Arn and Latin Arnius, Hirata Etruscan Arnθ and Latin Aruns.
From autopsy in the Museo di Villa Giullia, Rome (inv.2311). Bibliography: Herbig CIE 8017
(autopsy); Vetter 1953:290 ¢256a²; G. Giacomelli 1963:58-9 ¢29,I²; Hirata 1967:35; Rix ET Fa 2.18.
Drawing: Herbig CIE 8017.
38. Scratched before firing inside a black-varnished saucer on a high foot (height 8.7
cm, 12.3 cm, letters 5-6 mm high)
ca�tu
Sinistroverse, Faliscan alphabet. The u is carelessly written as . Tu is an probably an
abbreviation of the gentilicium Turius, cf. caui � turi MF 273 and turia MF 22-27.
Bibliography: Herbig CIE 8019; Vetter 1953:290 ¢256c²; G. Giacomelli 1963:58-9 ¢29,II². Drawing:
Herbig CIE 8019.
39. Painted in red across a tile (letters 10-12 cm high) said to have been found “in
sepulcreto di Celle multis annis ante [sc. before 1912]” (Herbig CIE 8566).
iuna
malio
Sinistroverse, Faliscan alphabet. The gentilicium is Mallius or Ma(n)lius: see §7.8.1.94.
Bibliography: Herbig CIE 8566 (autopsy); Vetter 1953:303 ¢288²; G. Giacomelli 1963:84 ¢87².
Drawing: Nogara in Herbig CIE 8566.
40. Cut over a loculus.216

[---]ociciocicoi�cupat�ifra
Sinistroverse, Faliscan alphabet. The first part of the text is to be read either as
[---]ocicio (G. Giacomelli 1963, 1978), either [---]o cicio ‘...us Cincius’ (for a Cincia
from Civita Castellana, cf. MF 135), or [---]ocrio (with ic � read as r ) which gives a
gentilicium Ocrius (G. Giacomelli 1965, Calzecchi-Onesti). Rix’s u(m)p]ricio is
impossible, see G. Giacomelli 1965. Cicoi is the only form in -oi that cannot possibly
216
Calzecchi-Onesti erroneously called the necropolis ‘Necropoli del Colle’.
426
be taken as a dative. G. Giacomelli interpreted it as a Faliscan transcription of an

Etruscan feminine Cicui (§9.2.2.2a), but I think that it could better be interpreted as a
genitive in -oi of a Cincus or Cicus, as was done by Pisani. For the genitive in -oi, see
§4.4.4. The resulting ‘...us Cincius son of Cincus’ could perhaps be compared to the tul 
tullius  tul  f in CIL I2.1493 and 1497 from Tibur. Cupat � ifra is a unique adaptation of
the normal formula hec cupat: see §8.10.1.
Bibliography: G. Giacomelli 1963:84 ¢88² (autopsy); Pisani 1964:340 n.1; Olzscha 1965:122; Rix
1965:447 n.4; G. Giacomelli 1965a:549-50; Calzecchi-Onesti 1981:181; G. Giacomelli 1978:531-2 ¢8².
Photograph: G. Giacomelli 1963 tav.IX (reproduced in G. Giacomelli 1965 tav.CXXXIVa).
13.4. The Valsiarosa necropolis
The necropolis of Valsiarosa lies to the west of Civita Castellana, to the north of the
modern road to Nepi, near the ancient road leading to the Ponte Terrano. Several tombs
were excavated here between 1886 and 1889 (cf. Cozza & Pasqui 1887c, FI II.2
pp.187-204 with map p.188), but all the inscriptions from this site are from a tomb
excavated in 1900 by Magliulo.217 Although it contained eighteen loculi, only three
sepulchral inscriptions are recorded: it is not stated whether the other loculi had any
inscriptions. The fact that two of the sepulchral inscriptions were still in situ and that
some at least of the grave-goods were recovered (cf. Thulin 1907:257) seems to imply
that the tomb was intact.218 Thulin (1907:257) dated the sepulchral inscriptions to the
fourth rather than the third century. All pertain to apparently unrelated women: Peruzzi
(1964b) assumed that they were women initiates of the Bacchic mysteries, partly
because of the burial of a freeborn woman and a freedwoman in the same loculus (see
MF 41), but there is no convincing argument for adopting this interpretation. The
inscriptions contain several Etruscan features: see §9.2.3f and Peruzzi 1964b.
41. Painted in violet on plaster across four tiles (each c.60uc.49 cm; letters 13 cm high)
found in situ.219
1
fas2[i]es�c3[ai]sia4 loụṛia
1
louc2i�teti�u3xor[ l]4oifirtạ
Sinistroverse, Faliscan alphabet. The last letters of each line (ria and irta) are written
vertically downwards as there was no room to write the line in full on the last tile. Only
217
Several other inscriptions (347-355) ascribed to the Valsiarosa necropolis by Herbig have
now been shown to come from the site at Grotta Porciosa (§16.6).
218
Peruzzi (1964b:142) assumed that the tomb was plundered because of the small number of
grave-goods listed by Thulin: the latter, however, gave only a selection, not an exhaustive list.
219
Herbig (CIE 8190) gave the height of the letters as “m. c. 0,03 a.”, but from Thulin’s drawing
it appears that this must be an error.
427
CHAPTER 13
the bottom parts of the letters ur in loụṛia remain: reading e.g. loục ̣ia would appear to
be impossible, however. The f in fasies, the c and the a in c[..]sia, the i in louci, the
interpunct between louci and teti, and the x of uxor had disappeared already by Thulin’s
time and were restored by him from Magliulo’s apograph. The lacuna in the first line is
large enough for c[ai]sia (which is more in keeping with the ou in louci and loụṛia) or
c[ei]sia: Thulin’s c[e]sia is too short, and Herbig’s c[ae]si(ạ) (1910:195) is untenable
and was abandoned later. Fasies has been interpreted as the genitive of the father’s
name (Thulin, Herbig, Buonamici, Pisani) used as the woman’s patronym (Vetter) or
gentilicium (G. Giacomelli): rather, it is an Etruscoid form in -es (see §9.2.2.2d) of the
gentilicium Fassius (attested for Oscan in fassii[s ZO 1) and placed here before the
praenomen as in Etruscan. Thulin took the inscription as pertaining to two daughters of
Fasi-, namely c[e]siạ louria and l]oifia. A different interpretation was proposed by
Herbig (1910:196). From the triple interpunct that follows c[ai]sia in Thulin’s drawing
and from the fact that louria and l]oifia (as he read the text) were not started on a third
line but written at the end of the first two, he concluded that the fourth tile was added
later. This interpretation became all the more attractive when the last word was read as
l]ọifirtạ by Nogara (in CIE 8190).220 The inscription could then be read as pertaining to
a fasies � c[ai]sia | louci � teti � uxor and to a loụṛia | [l]oifirtạ interred in the same
loculus. This interpretation was rightly adopted by most later authors. Pisani read
l]ọifiṛṭạ(s), interpreting this as a woman’s name ‘Libertas’.
Bibliography: Thulin 1907:257-64 ¢1² (autopsy); Herbig 1910:195-8 ¢37²; Jacobsohn 1910:4 ¢18²;
Herbig CIE 8190; Buonamici 1913:59 ¢14²; Vetter 1953:297 ¢276a²; Knobloch 1954:36; G. Giacomelli
1963:76-7 ¢73,I²; Pisani 1964:337-8 ¢143G²; Peruzzi 1964b. Drawing: Thulin 1907:258 (reproduced in
42. Painted in red on plaster on two tiles (60u44 cm, letters 10 cm high) found in situ.
1
cauia2:satelie
1
caui:fel2ic ̣inate
1
uxor 2
Sinistroverse, Faliscan alphabet. The i in cauia, the interpunct between cauia and
satelie, the second e in satelie, the i in caui, the interpunct between caui and felic ̣inate,
and the r in uxor were all restored by Thulin from Magliulo’s apograph. Magliulo’s
satelie is probably to be preferred to the satelie[s] read by Herbig and all editors after
him, for although Herbig (1910:199) asserts that in Nogara’s drawing there is space to
restore s, this is not borne out by the drawing as published in CIE, and the possibility is
expressly rejected from autopsy by Nogara himself. 221 The form is an Etruscoid name in
220
“Non si vede più il terzo punto dell’interpunzione in principio del tegolo IV, quello in basso,
e probabilmente non esistette mai” (Nogara in Herbig CIE 8190).
221
The final e is based on Magliulo’s apograph: it is unlikely that misread a damaged a ( ) as
an e ( ), as he read the damaged last letter of cauia, which in Nogara’s drawing is , as a.
428
-ie(s), for which see §9.2.2.2d. In the second line, the drawings show the top parts of the
letters i and c (or o, or θ) between fel an inate. Herbig’s felịc ̣inate (CIE), with a
toponymic gentilicium that occurs also in Etruscan (felcinatial Pe 1.485 and 1.1235)
and Latin (Feliginas), has been adopted by all authors, and is now confirmed by
[fel]ịcinatiu LF 384. The genitive ending -e renders either /-es/ or /- ē ̣s/: see §4.5.2.
Bibliography: Thulin 1907:257-64 ¢2² (autopsy); Herbig 1910:198-9 ¢38²; Jacobsohn 1910:4 ¢19²;
Herbig CIE 8191; Buonamici 1913:59 ¢15²; Vetter 1953:297 ¢276b²; G. Giacomelli 1963:76-7 ¢73,II²;
Peruzzi 1964b. Drawing: Thulin 1907:261 (reproduced in CIE 8191); Herbig CIE 8191.
43. Painted in red on plaster along the length of the backs of two tiles (max. 51?u44 and
66u44 cm; letters c.10 cm high).
1 2
ueṇe[ ] na
1 2
ux[or ] [
Sinistroverse, Faliscan alphabet. The tiles, found and described separately by Magliulo,
were joined by Thulin, who read Venena ux. (with the letters ne restored from
Magliulo’s apograph?). A trace of a shaft after the letters ue was in fact seen by Nogara
(in Herbig CIE 8192), who also read the interpunct at the end of the first line. Vetter
emended Thulin’s Venena to ueḷịena, which was adopted by G. Giacomelli. Herbig
interpreted ue(ne)na either as a masculine genitive ue(ne)na(s) preceded on a lost tile by
a woman’s name in the nominative, or as a feminine nominative preceded on a lost tile
by the woman’s praenomen in the first line, and by her husband’s praenomen in the
genitive in the second line. The former interpretation was adopted by G. Giacomelli.
Both interpretations presuppose that a missing tile preceded the text, and this
appears to be impossible in view of the amount of space before ueṇe[ and ux[or shown
in Thulin’s drawing. Vetter (1953:443) took his ueḷịena as a masculine nominative,
apparently reading the text as ‘Veliena (and) wife’ without any missing tiles, but there is
no Faliscan parallel for a woman to be designated by the word uxor alone. I suspect that
a tile is missing in the middle, in which case the text may be read as a woman’s name
with in the second line ux[or, followed by the husband’s name in the genitive. For
ueṇe[, cf. perhaps ue in MLF 206 and the gentilicium nel[n---] LtF 299 and neln LtF
300? Uene[ ]na can be a man’s name, Venel ...na, but in that case it is very difficult to
explain what the remainder of the text (starting with ux[or) may have looked like.
Bibliography: Thulin 1907:257-64 ¢3² (autopsy); Herbig 1910:189 ¢27²; Herbig CIE 8192; Buonamici
1913:60 ¢16²; Buonamici 1935:343; Vetter 1953:298 ¢276c²; G. Giacomelli 1963:76-7 ¢73,III²; Peruzzi
1964b:140. Drawing: Thulin 1907:264 (reproduced in CIE 8192).
44-46. From the same tomb are two or three inscribed vessels. Thulin (1907:307) spoke
of “zwei [Gefässen] mit kleinen Graffiti”, but describes not two, but three. Editors have
tacitly assumed that all these three vessels belonged to this tomb.
429
CHAPTER 13
44. Scratched under a small black-varnished cup ( 10 cm; letters 5-10 mm high).
tur
Sinistroverse, Faliscan alphabet. Cf. turia MF 22-27 and turi MF 273.
Bibliography: Thulin 1907:307-8 ¢61² (autopsy); Herbig CIE 8193; Vetter 1953:289 ¢247²; G. Gia-
comelli 1963:50-1 ¢8². Drawing: Thulin 1907:308 (reproduced in CIE 8193).
45. Scratched inside a black-varnished cup ( 12 cm; letters 6-12 mm high) similar to
that of 44.
ani
Sinistroverse, Faliscan alphabet? The a is written rather carelessly as . A genitive
(Vetter and G. Giacomelli) rather than an abbreviated nominative (Thulin); see §8.8.1.
comelli 1963:50-1 ¢9². Drawing: Thulin 1907:308 (reproduced in CIE 8194).
46. Scratched under the foot of a small plate ( 14 cm; let. 5-7 mm) is
hap
Sinistroverse, Faliscan alphabet? The h is . G. Giacomelli took the elongated p, ,
as a ligature pi (considered also by Herbig), but this seems unnecessary. Hap may
render Fabius (Herbig) with the Faliscan spelling h- for original /f-/ (§3.5.2).
Bibliography: Thulin 1907:308 ¢63² (autopsy); Herbig CIE 8195: Vetter 1953:289 ¢249²; G. Giacomelli
1963:51 ¢10². Drawing: Thulin 1907:308 (reproduced in CIE 8195).
13.5. The La Penna necropolis
The La Penna necropolis is situated to the southwest of the city, south of the modern
road to Nepi. Excavations were conducted at this site in 1886 and 1887, and several
more tombs were excavated in 1888 and 1889 (see Cozza & Pasqui 1887a-b, FI II.2
pp.144-86 with map p.145). Most of the inscriptions from this site, including the cup
with eko lartos EF 6 and eko kaisiosio EF 7 (§12.3), were found during these excava-
tions; only the discovery of MF 55 seem to be undocumented.
47. Cut over a loculus of Cozza & Pasqui’s tomb II.
uolḷia
Sinistroverse, Faliscan alphabet. The third and fourth letters are ��. Most editors read
uollia, but the rarity of the geminated spelling in Faliscan inscriptions (§11.2.4.3) lead
Herbig to read uolụia and Vetter to read uolṭia. The latter is attractive, but may not be
necessary, cf. uolḷ[---] MF 86. See also §3.5.5.3.
Bibliography: Cozza & Pasqui 1887b:263 (autopsy); Conway 1897:375 ¢xl.24²; Herbig CIE 8164;
Vetter 1953:296 ¢273²; G. Giacomelli 1963:75 ¢70². Transcription: Cozza & Pasqui 1887b:263
(reproduced in CIE 8164).
430
48-54. The following inscriptions were painted under and over loculi of Cozza &
Pasqui’s tomb 6, 7, or 8 (cf. Thulin 1907:271, with a drawing showing the inscriptions).
Deecke jr. suspected the inscriptions, and Thulin remarked: “Alles macht den Eindruck,
gleichzeitig von einer Hand gemalt zu sein. Viel schwerer fällt es mir jedoch zu
verstehen, wie jemand auf den Gedanken hätte kommen sollen, die Wände eines leeren
Grabes zu bemalen. Vielleicht rühren die Inschriften von einer Restaurierung in antiker
Zeit her.” (1907:272-3). The tomb was the family tomb of a gens Aufilia or Oufilia.222
48. Painted in red between two loculi (letters c.7 cm high).
iunaoufiliopoplia
Sinistroverse, Faliscan alphabet. The n is reversed. The first a seems to be damaged at
the top. In Lignana’s transcription (from an apograph by Fiorelli), the first p is given as
 and the second as ; in Cozza & Pasqui’s, both are ; in Thulin’s, both are . Ribezzo
impossibly regarded oufilio as a genitive in -io (§4.4.5).
Bibliography: Lignana 1887:202 ¢b²; Cozza & Pasqui 1887b:267 (autopsy); Deecke 1888:151-2 ¢29²;
Conway 1897:375 ¢xl.19²; Thulin 1907:271 ¢6a² (autopsy); Herbig 1910:85-8 ¢3² (autopsy); Herbig
CIE 8167; Jacobsohn 1910:4 ¢13²; Ribezzo 1930:99; Vetter 1953:296 ¢274a²; G. Giacomelli 1963:75-6
¢71,I² (autopsy); Pisani 1964:339 ¢144B²; G. Giacomelli 1978:530-1 ¢7,I²; R. Giacomelli 1978:74-5
¢2,I². Drawing: Deecke jr. in Deecke 1888 Taf.I.

cauioaufilioθanacuil
Sinistroverse, Faliscan alphabet. In Lignana’s and Cozza & Pasqui’s transcriptions (the
former from an apograph by Fiorelli), the θ is  (Danacvil Lignana); in Deecke’s
drawing (and in Thulin’s?), it is . For the use of θ, see §3.5.4. Both transcriptions
show a stroke after cauio. Ribezzo took aufilio as a genitive in -io: see §4.4.5.
Bibliography: Lignana 1887:202 ¢d²; Cozza & Pasqui 1887b:267 (autopsy); Deecke 1888:152 ¢30²;
Conway 1897:375 ¢xl.20²; Thulin 1907:271 ¢6b² (autopsy); Herbig 1910:85-8 ¢4² (autopsy); Herbig
CIE 8168; Jacobsohn 1910:4 ¢14²; Buonamici 1913:58 ¢13²; Ribezzo 1930:99; Vetter 1953:296 ¢274b²;
G. Giacomelli 1963:75-6 ¢71,II² (autopsy); Pisani 1964:339 ¢144C²; G. Giacomelli 1978:530-1 ¢7,II²;
R.Giacomelli 1978:74-5 ¢2,II². Drawing: Deecke jr. in Deecke 1888 Taf.I.
50. Painted under a loculus (letters c.7 cm high).

caui[o]auf ̣iliopoplia
Sinistroverse, Faliscan alphabet. In Lignana’s transcription (from an apograph by
Fiorelli), caui is followed by . ᛌ ; in Cozza & Pasqui’s, it is preceded by a space.
Bibliography: Lignana 1887:202 ¢c²; Cozza & Pasqui 1887b:267 (autopsy); Deecke 1888:152-3 ¢31²;
Conway 1897:375 ¢xl.21²; Thulin 1907:271 ¢6c² (autopsy); Herbig 1910:85-8 ¢5² (autopsy); Herbig CIE
8169; Jacobsohn 1910:4 ¢15²; Ribezzo 1930:99; Vetter 1953:296 ¢274c²; G. Giacomelli 1963:75-6
¢71,III² (autopsy); G. Giacomelli 1978:530-1 ¢7,III². Drawing: Deecke jr. in Deecke 1888 Taf.I.
222
Lignana, Conway, and Deecke read -filio as ‘filius’ preceded by an abbreviated father’s
praenomen, which is implausible; the correct interpretation goes back to Thulin.
431
CHAPTER 13
51. Painted in red under a loculus (letters c.7 cm high).

ḳai[s]i[oauf]ilioiun[?eo]
Sinistroverse, Faliscan alphabet. The first letter is very doubtful: Lignana’s transcription
(from an apograph by Fiorelli) gives it as (), Cozza and Pasqui’s as , which in a
Middle Faliscan inscription may reflect a specific use of k to mark this name, as in Latin
(§11.2.4.3).223 The letters iun were seen only by Herbig.
Bibliography: Lignana 1887:202 ¢a²; Cozza & Pasqui 1887b:267 (autopsy); Deecke 1888:153 ¢32²;
Conway 1897:375 ¢xl.22²; Thulin 1907:272 ¢6f² (autopsy); Herbig 1910:85-8 ¢6² (autopsy); Herbig CIE
8170; Vetter 1953:296 ¢274d²; G. Giacomelli 1963:75-6 ¢71,IV² (autopsy); G. Giacomelli 1978:530-1
¢7,IV². Drawing: Deecke jr. in Deecke 1888 Taf.I.

[---]oụ*[..]o *ạ**[---]
Sinistroverse, Faliscan alphabet. Herbig’s drawing shows that only the top
halves of the letters were preserved. Herbig read this as ọụf ̣[ili]ọ cauị [.f], which has
been adopted by the later authors, but I doubt very much whether this is possible.
Bibliography: Cozza & Pasqui 1887b:267 (autopsy); Herbig CIE 8172; Vetter 1953:296 ¢274f²; G.
Giacomelli 1963:75-6 ¢71,VI² (autopsy); G. Giacomelli 1978:530-1 ¢7,VI². Drawing: Herbig CIE 8172.
53. Painted underneath a loculus (letters c.7 cm high). Seen only by Thulin.
[---]aaufi[lio ?---]
Sinistroverse, Faliscan alphabet. The praenomen may have been [iun]a or [uol]ta.
Bibliography: Thulin 1907:272 ¢6e² (autopsy); Herbig 1910:85-8 ¢7²; Herbig CIE 8173; Vetter
1953:296 ¢274g²; G. Giacomelli 1963:75-6 ¢71,VII²; G. Giacomelli 1978:530-1 ¢7,VII².
54. Painted under a loculus (letters c.7 cm high). The part containg the letters mio had
broken off (and disappeared?) when the inscription was seen by Thulin and Herbig.
puponiofir mio
Sinistroverse, Faliscan alphabet. Both Lignana’s and Cozza & Pasqui’s transcriptions
(the former from an apograph by Fiorelli) give the m and the n as  and  respectively.
According to Cozza & Pasqui, the space between fir and mio is due to the ancient writer
avoiding a ‘sfaldatura del tufo’: it is unnecessary to read fiạ....mio (Deecke, Conway) or
firemio (Thulin). Lignana’s ‘Pomponio filio’ is impossible. Puponius appears to be a
gentilicium used as a praenomen, see §7.7.1.52.
Bibliography: Lignana 1887:202 ¢e²; Cozza & Pasqui 1887b:267 (autopsy); Deecke 1888:153 ¢33²;
Conway 1897:375 ¢xl.23²; Thulin 1907:271 ¢6d² (autopsy); Herbig CIE 8171; Vetter 1953:296 ¢274e²;
G. Giacomelli 1963:75-6 ¢71,V² (autopsy); G. Giacomelli 1978:530-1 ¢7,V². Drawing: Deecke jr. in
Deecke 1888 Taf.I.
223
Note that Lignana had erroneously read k instead of c in MF 59-60.
432
55. On a tile fragment from a tomb “a parte sinistra prope sepulcrum gentis Oufiliae”
(Herbig).
c[---]
Sinistroverse (or dextroverse if held the other way up). Herbig’s transcription,    ,
appears to imply that the c stood close to the edge of the tile.
Bibliography: Herbig CIE 8175.
56-57. The following inscriptions are both from Cozza & Pasqui’s tomb X.
56. Painted in red on plaster on a tile fragment.
1
uelzu[2][con]
1
eo:fe[2][cupa]
Sinistroverse, Faliscan alphabet. The o is . It is unclear whether the ... preceding uelzu
in Cozza & Pasqui’s transcription indicate traces, vacant space, or a missing tile
preceding the text. Herbig read uelzu as a praenomen Volso, which was adopted by all
other editors. I would rather restore the text as uel zu[con]|eo  fe[ cupa], or, assuming
that more than one tile is missing at the end, as uel zu[coneo ---]|eo  fe[ cupa ?---]).
The praenomen uel occurs e.g. in MF 82 (cf. also §7.7.1.80) and the gentilicium
zu[con]|eo in zuconia MF 271: cf. also larisa zuχus Etr XXXII and §7.8.1.148. The use
of the interpunct in Faliscan inscriptions is not so consistent that its absence after uel
and zu[---] can constitute an argument against this. The hypercorrect spelling fe for
he(c) occurs also in MLF 305: see §3.5.2.
Bibliography: Cozza & Pasqui 1887b:272 (autopsy); Conway 1897:375 ¢xl.25²; Herbig 1910:187 ¢25²;
Herbig CIE 8176; Vetter 1953:297 ¢275a²; G. Giacomelli 1963:76 ¢72,II². Transcription: Cozza &
Pasqui 1 887b:272 (reproduced in CIE 8176).

calin[---]
rezo[---]
Sinistroverse, Faliscan alphabet. The r appears to be  rather than : see §11.2.4.2.
Herbig restored the first line as calin[ia, with rezo as a genitive. I wonder whether the
ca is not rather the frequent abbreviation ca = Gauius. As z is more common at the
beginning of words (§11.2.4.3, §3.5.3), rezo[---] is probably to be divided as [---]|re
zo[---]. Zo[---] could perhaps be zo[coneo] or zo[conea], a further adaptation to
Faliscan of the gentilicium zuconia MF 271 (and perhaps zu[con]|eo in MF 56), which
is itself an adaptation of the Etruscan gentilicium zuχu in larisa zuχus Etr XXXII: see
§7.8.1.148.
Bibliography: Cozza & Pasqui 1887b:272 (autopsy); Conway 1897:375 ¢xl.26²; Herbig 1910:187 ¢25²;
Herbig CIE 8177; Vetter 1953:297 ¢275b²; G. Giacomelli 1963:76 ¢72,II². Transcription: Cozza &
Pasqui 1887b:272 (reproduced in CIE 8177).
433
CHAPTER 13
58. Scratched in a small black-varnished cup (height 7.8 cm, rim 11.9 cm, foot 5.6
cm, letters 12-18 mm high), according to Schippa from tomb 10 of the La Penna
necropolis. c.300.
fofiti
Faliscan alphabet. Schippa (1980:48) described the inscription as “con andamento
destrorso”, but his drawing shows the ductus as sinistroverse. The form is either a
genitive or an abbreviated nominative (§8.8.1). Schippa compared the gentilicium
Fuficius, but perhaps f ofiti could be read, with an abbreviated praenomen f (probably
Fertor, see §7.7.1.23) as in f aino MLF 352 from the Grotta Porciosa site and f  pacios
Cap 392 from Capena; a gentilicium Aufitius occurs in CIL VI.6945 from Rome.
Bibliography: Schippa 1980:48 ¢50² (autopsy). Drawing: Schippa 1980 tav.LXI.
59-62. Tombs CXXVII and CXXVIII/86 were connected already in antiquity. From this
double-tomb are the following inscriptions.
59-60. Painted around the central tondi of two almost identical late fourth-century
kylikes, the name-pieces of the Foied-painter, cf. Beazley 1947:106-7.224 Deecke (in
Lignana 1887a:443, 1888:154-5) regarded the inscriptions as falsifications, which led
others to suspect them too (Lignana 1887b:199, Duvau 1889:10, Pauli in Herbig CIE).
The arguments on which this was based were rejected by Spinazzola.
foieduinopipafocracarefo
foieduino‹pi›pafocracarẹ[f]o
Sinistroverse, Faliscan alphabet. Lignana’s and Gamurrini’s transcripts erroneously give
k instead of c (and of f (koied Lignana 1887a).225
Sittig and Belardi took the slight difference between the r’s in carefo, carẹ[f]o, ,
and those in cra, , as an indication that the latter represented a not yet completely
rhotacized /s/, i.e. [z]. Apart from the fact that both types of r occur in other inscriptions
(§11.2.4.2), this is very improbable: phonetic differences are noted at all, this may be
done by means of ‘dead letters’, the introduction of new signs being reserved rather for
phonemic differences. Unfortunately, the idea has started to lead a life of its own since it
was adopted by Bonfante (1966), who questioned the occurrence of rhotacism in
Faliscan (§3.5.3).
224
A third example, without inscription, was published by Gabrici (1912a:73-4). The scene in
the tondi is often described as ‘Dionysus with Semele’ (due to a similar scene on a mirror where
the figures are labelled φuφluns and semla, cf. Gerhard 1843:1.87-8 with Taf.LXXXIII): others
regard it as ‘Dionysus with Ariadne’ (thus first Gabrici 1912a:73-4).
225
Conway (1897:372) notes that in MF 60 “the last word [...] seems to have been complete
when Lignana sketched it”, which is probably due to the inaccuracy of Lignana’s drawings (or
Conway’s misinterpretations of them).
434
The earliest interpretations koi (h)ed ‘qui hoc’ (Lignana 1887a), foied ‘favebit’
(Lignana 1887b), ‘foede’ Gamurrini (in Lignana 1887b:198) are partly based on
misreadings and can be discarded: Cozza & Pasqui’s reading and interpretation has
rightly been adopted by all editors. Foied, from a fossilized ablatival phrase
/hō(d)+dē(d)/: see §6.2.34 (for hypercorrect f- instead of h, cf. §3.5.2).
Whether the variation pipafo/pafo is significant (reduplicated vs. non-reduplicat-
ed) is debated. Assuming that the painter had the intention of making two identical
pieces, the form is an error at least from an epigraphical perspective, and I therefore
regard pafo as an error for ‹pi›pafo. Pisani (and R. Giacomelli) took carefo to mean
‘mancherò, sarò morto’: used in this sense, however, careo always has a complement
like uita, sensu, die, luce, etc., showing that by itself the word could not carry this
meaning (TLL 3 450.19-52). Friedrich regarded the phrase as a Saturnian, an idea
adopted by Vetter and elaborated by Morelli and Poccetti. As Friedrich noted, it is
interesting to find the Saturnian used in poetry of a less serious nature.
From autopsy in the Museo di Villa Giulia (inv. 1674-1675). Bibliography: Lignana 1887a; Gamurrini
1887b; Helbig 1887; Lignana 1887b:196-9; Cozza & Pasqui 1887b:273; Kroker 1888:514-5; Deecke
1888:154-5 ¢34²; Brizio 1889:440; †Dennis 1890:16-7; Spinazzola 1891; Conway 1897:372 ¢312a-b²;
Von Planta 1897:588 ¢318.1-2²; Tambroni 1898:3; †Savignoni 1904:77; Jacobsohn 1910:4 ¢16-17²;
Jacobsohn 1911:466; Herbig CIE 8179-80; Weege in Helbig 1913:370-1; Buonamici 1913:50-3 ¢2-3²;
Friedrich 1917:143-4; Della Seta 1918:74 (autopsy); Ribezzo 1918; Ducati 1927:512; Sittig 1932;
Vetter 1942:221; Beazley 1947:7, 106-7; Vetter 1953:287-8 ¢244a-b²; G. Giacomelli 1963:49-50
¢5,I-II²; †Belardi 1964; Pisani 1964:346-7 ¢150a-b²; [Zosel in Helbig/Speier 1969:686-7 ¢2776²];
Morelli 1974; G. Giacomelli 1978:529-30 ¢5,I-II²; R. Giacomelli 1978:73-4 ¢1²; FI II.2 p.148; Morandi
1982:58 ¢10.A-B²; Poccetti 2007:251-3. Photographs: Herbig CIE 8179-80; Ribezzo 1918:245 fig.1;
Beazley 1947 pl.XXV.4; G. Giacomelli 1963 tav.VI; Moretti 196X:173 fig.136; Morandi 1982 tav.IX,1.
Drawings: Cozza & Pasqui 1887b:273; Spinazzola 1891:14; Morandi 1982:58.
Fig.13.3. Herbig’s tracing of MF/Etr 61.

(From CIE 8178.)
61. Scratched inside a small plate ( 13.3 mm, let. 3-6 mm high) within a graffito of
what appears to be a phallus with legs (thus Gamurrini 1887c:602)226 is an illegible
226
Cozza & Pasqui rather curiously described it as “rappresentante forse un animale qua-
drupede, di cui si espressero soltanto due gambe” (1888b:272). Danielsson (in Herbig CIE)
compared a winged phallus in IG 13.1658 add.
435
CHAPTER 13
sinistroverse graffito (see fig.13.3). Gamurrini read udori.bonues (with a cursive e)

‘odorari bonum est’ (1887c). Helbig (in Gamurrini 1887c:602) took udori as an
ablative; Danielsson (in Herbig CIE 8178) compared Umbrian utur TI IIb.15. The text
defies any attempt at interpretation: if the alphabet is Faliscan at all (which Herbig
doubted), the drawings seem to show u**al*onu*(*)s, which may be an Etruscan
genitive ...al ...s.
Bibliography: Gamurrini 1887b:151 (autopsy); Gamurrini 1887c:602; Cozza & Pasqui 1887b:272
(autopsy); Bormann CIL XI.6707,6; Herbig CIE 8178 (autopsy); Vetter 1953:294; G. Giacomelli
1963:263 ¢IX²; FI II.2 p.148 (autopsy). Drawings: Cozza & Pasqui 1887b:272 (reproduced in CIL
XI.6707,6, CIE 8178); Herbig CIE 8178.
62. Painted in the upper border of a scene on a stamnos227 (letters 3mm high) above the
figures represented in the scene, the name-piece of the Diespiter-painter (Beazley
1947:73, Adembri 1990). c.380-370 (Colonna).228
canumede [die]s pater cupi‹d›o menerua
Dextroverse, but apparently in Faliscan alphabet. M is , r is , the t is . The first and
second a have rounded tops, the third is . Degrassi’s canumede[s (considered also by
Wachter) is impossible, for the distance between the e and the edge of the lacuna is too
great. The next name is usually restored as [die]s pater, but [ioui]s pater (Lommatzsch)
or [iouo]s pater are possible, pace Degrassi: according to Wachter (1987:150-3),
however, the usual form of the nominative in the fourth century appears to have been
Diespater. The third name is cupico, where the second c () is clearly an error for d: the
text is cupi‹d›o, not Girard’s Cupigo.
Although often regarded as Latin, the inscription is in fact Faliscan (as may be
expected in the case of a local product). Wachter points to the co-occurrence of 
and , which is normal in the Faliscan, but not in the Latin alphabet. The omission of -s
in canumede is also more frequent in the ager Faliscus, although not after a long vowel
(§3.5.7c). The a in [die]s pater cannot be regarded as an attestation of the Faliscan
absence of vowel-raising in word-internal syllables, however, as Diespater may well
have been regarded as separate words (§3.6.6.2). Canumede reflects Greek Ganum»dhj,
not Etruscan Catmite or Latin Catamitus (§6.4.3).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 1599). Bibliography: Gamurrini
1887d:231-2 (autopsy); Cozza & Pasqui 1887a:175; Brizio 1889:439-40; †Dennis 1890; Bormann CIL
XI.6708,13; Weege in Helbig 1913:370-1; Della Seta 1918:73; Lommatzsch CIL I2.454; Ducati
1927:512; Diehl 1930:80 ¢778²; Beazley 1947:73; Vetter 1953:288; Safarewicz 1955:186 ¢2²; Degrassi
ILLRP 1228; G. Giacomelli 1963:263 ¢XI²; Torelli RMR pp.55-6 (autopsy); Colonna 1980a:46; FI II.2
p.154 (autopsy); Wachter 1987:367-9; Girard 1989:167; [Adembri 1990]. Photograph: Ducati 1927
vol.II tav.252, no.615. Drawing: Gamurrini 1887d tav.X.
227
Diehl and Vetter erroneously describe it as a ‘Schale’.
228
Safarewicz erroneously dated the vase to the third or second century.
436
63-64. Scratched inside a red-varnished saucer ( 17.5 mm, letters c.12 mm high in 63,
and c.5-10 mm high in 64).
anni
ulties
63 is written dextroverse in uneven and straggling letters, apparently in the Latin
alphabet: note the double n (§11.2.4.3); 64 is written sinistroverse in smaller and neater
letters, with a Faliscan t (). The e in 64 is : a similar e occurs in MF 258 and MLF
285. Herbig connected both words as ‘Ultius Anni (servus)’, but both the difference in
writing and the relative position of the words make this very unlikely. G. Giacomelli
rightly takes the words as Besitzerinschriften written by successive owners (like EF 6
and EF 7). Ulties may stand for u(e)lties or u(o)lties (cf. Schulze 1904:252).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 1650). Bibliography: Herbig
1910:199-200 ¢39²; Herbig CIE 8181; Vetter 1953:293-4; G. Giacomelli 1963:50 ¢6a-b²; Rix ET
Fa 2.20a-b. Drawing: Herbig CIE 8181.
65. Scratched on a patera. Known only through Nogara’s copy of the apograph in the
Museo di Villa Giulia inventory.
apolo
Sinistroverse, Faliscan alphabet. According to Nogara’s transcript, the a is . Vetter,
apparently reluctant to assume a dedication in a tomb, proposed to take the form as an
abbreviated slave-name Apollo..., but apolo is rather the name of the deity (cf. apolonos
EF 10), in the nominative, as if it were a ‘divine Besitzerinschrift’ (cf. §8.11.1). For the
cult of Apollo in the ager Faliscus, cf. §2.3.4.
Bibliography: Herbig CIE 8182; Taylor 1923:78; Vetter 1953:289 ¢246²; G. Giacomelli 1963:50 ¢7²;
Wachter 1987:134. Transcription: Nogara in Herbig CIE 8182.
13.6. The necropolis of Ponte Lepre
Two inscriptions (MF/Etr 66-67) were found in tombs at a locality known as Ponte
Lepre, excavated in 1904 and 1906 by Feroldi di Rosa (cf. Herbig CIE p.111). Ponte
Lepre was located to the west of Civita Castellana beyond the Valsiarosa necropolis, to
the north of the modern road to Nepi. Also from the Ponte Lepre tombs are is the
dolium with ud EF 8 and tele*[1-2?] med fifiked EF 9, and probably also MF 68 and
MF 69-72, which came to the Villa Giulia Museum from the Collezione Feroldi.
66. Scratched inside a saucer (height 4.5 cm, 13 cm, letters 25-30 mm high) from
tomb I, dated between the sixth and the fourth century.
namureṭua
437
CHAPTER 13
Sinistroverse. The r is the Etruscan ; the ṭ is  (x?). Rix reads namures̽ka, with the 
interpreted as s̽ and the u of Herbig’s drawing, , as k, but the original shows that u is
the correct reading. Editors generally regard the inscription as Etruscan because of the r,
but no satisfactory interpretation has been proposed.
CIE 8567 (autopsy); Vetter 1953:294; G. Giacomelli 1963:53-4 ¢14,I²; Rix ET Fa 2.26. Drawing:
Herbig CIE 8567.
Fig. 13.4. Nogara’s drawing of MF/Etr 66.

(From CIE 8577.)
67. Scratched on the bottom of a red-varnished saucer (height 6 cm, 17 cm, letters
c.15 mm high) is an inscription which Herbig hesitatingly read as ạcṛẹẓ cat (dextro-
verse, with reversed e and z), comparing statia catino CIL I2.480. G. Giacomelli
adopted this reading, but interpreted the text as ‘Acrius Cat(ineius)’, comparing catinei
MLF 469*, but there are no Faliscan parallels for a praenomen written in full followed
by an abbreviated gentilicium. Nogara’s drawing seems rather to show a sinistroverse
text (ta*ec*a?).
Bibliography: Herbig CIE 8577 (autopsy); Vetter 1953:294; G. Giacomelli 1963:53-4 ¢14,II². Drawing:
Nogara in Herbig CIE 8577.
68-72. The following inscriptions, on plates originally from the Collezione Feroldi,
were added to the inscriptions from Ponte Lepre by Herbig.
68. Scratched under the foot of a catillus (height 1.5 cm, 11 cm, letters 20 mm high).
ce
Dextroverse.
Bibliography: Herbig CIE 8569. Drawing: Herbig 8569.
69. Scratched on the rim of a small red-varnished plate (letters 6-10 mm high) and on
the rim of a saucer of red clay (letters 5-10 mm high).
tiroicolanioi
tiroicolanioi
Sinistroverse. Presumably Faliscan alphabet. For interpretation and bibliography, see
below under 71-72.
438
71-72. Scratched on a saucer of red clay, 71 on the rim (letters 4-6 mm high), 72 on the
inside (letters c.8-12 cm high).
tiroicolanioi
tulom
Sinistroverse, presumably Faliscan alphabet. Tiroi  colanioi is regarded as dative by all
editors except Pisani (1964:343), who, probably rightly, regarded these forms as
genitive: see §4.4.4. More problematic is tulom. The arrangement of the words and the
difference in writing show that it is a separate text. Herbig (1914:238 n.1) interpreted it
as */(te)tul-o-m/, a first singular perfect from PIE */tel-/ (perhaps rather */telħ2-/) with
the aorist ending as in the Oscan perfect manafum Cp 37,6 man]afum Cp 37,1. This
interpretation was adopted by G. Giacomelli, but cannot be maintained: assuming an
aorist ending in a Middle Faliscan first singular perfect would at least require an
explanation in view of the Early Faliscan first singular perfect pepara[i ‘peperi’, the
accepted reading of pepara[ in EF 1 (which was proposed by Herbig himself). It would
also be quite unexpected for a language or dialect belonging to the Latin branch of the
Italic family, even if there is a parallel in Oscan. Herbig’s alternative, taking tulom as a
noun ‘donum’, is equally problematic: at least in Latin the expected result of a PIE
*/tlom/ or */tħ2om/ would be */tolom/ or */talom/ respectively. It is more likely that
tulom is a name (cf. tulo in MF 151) as was suggested by Lejeune, who proposed tulo m
‘Tullus M(arci f.)’, and by Peruzzi, who proposed tulom ‘Tullum = Tullorum’. The
latter interpretation is now supported by the genitive plural [fel]ịcinatiu LF 384 and
ac ̣iuaiom esú Cap 465. The issue is made even more problematic by the possibility of a
relation between tulom and tulate tulas EF/Etr 385.
Bibliography: Herbig 1914a:237-9 ¢1-2, 3a-b² (autopsy); Herbig 1923:231; Lejeune 1952b:125; Vetter
1953:291 ¢258, 259a-b²; G. Giacomelli 1963:52 ¢12,I-IIIa-b² (autopsy); Peruzzi 1964a:167-9; Olzscha
1965:123-4.
73-78. The following inscriptions occur on plates originally from the Collezione
Feroldi.
73. Scratched inside the rim of a small plate.
iuna
Sinistroverse, alphabet not indicated.
Bibliography: G. Giacomelli 1963:52-3 ¢13,II² (autopsy).
74. Scratched inside the rim of a small plate.

iunai
Bibliography: G. Giacomelli 1963:52-3 ¢13,I² (autopsy).
439
CHAPTER 13
75. Scratched on the rim of a small plate.

latria
Bibliography: G. Giacomelli 1963:52-3 ¢13,III² (autopsy).
76. Scratched in a saucer.

sa
Dextroverse. G. Giacomelli compared the equally dextroverse sa ‘sa(cra)’ MF? 131,
although that is a dedication from a stips.
Bibliography: G. Giacomelli 1963:52-3 ¢13,VI² (autopsy).

salθan
Sinistroverse, Faliscan or Etruscan alphabet. G. Giacomelli read calθan, hesitatingly
comparing calitenes MLF 265, but Colonna’s drawing shows that his salθan is the
correct reading: as an interpretation, he points to śalθn Co 3.1.
Bibliography: G. Giacomelli 1963:52-3 ¢13,IV² (autopsy); Colonna 1993 (autopsy). Drawing: Colonna
1993:298.

ipa
Dextroverse. Cf. perhaps Etruscan ipas ikam Etr X?
Bibliography: G. Giacomelli 1963:52-3 ¢13,V² (autopsy).
13.7. The tombs near the Ponte Terrano
Both the ancient and the modern road from Civita Castellana to S. Maria di Falleri and
Fabbrica di Roma cross the gorge of the Rio Maggiore just to the west of the ancient
town by means of the Ponte Terrano. This impressive bridge still conserves ancient
masonry in its substructure (Dennis 1878:94-5) and may well go back to Faliscan times
(Ward Perkins & Frederiksen 1957:143-5). Across the bridge, the road swings west-
wards, following the ridge between the Rio Maggiore to the south and the Torrente
Purgatorio to the north.
The area on the north side of the gorge, known as Terrano, is riddled with tombs,
especially in the cliff-side to the left of the Ponte Terrano, and on the spit of land
between the Ponte Terrano and the junction of the Rio Maggiore and the Torrente
Purgatorio. Most of these easily visible tombs have been known (and converted for
modern re-use) for a long time, and the inscriptions from this location, especially those
cut in the rock were among the first to be rediscovered in the previous centuries: MF 79
440
was in fact the first Faliscan inscription to be published (in 1726), although it was
regarded as Etruscan for almost 150 years. Later epigraphical finds from these tombs
are MF 88-89, discovered by Mengarelli, and MF 87, apparently found during the
Second World War.
79.229 Cut over and beside the right-hand corner of an arcisolium-like niche (letters
“about a foot in height” according to Dennis (1878:94), but 16-20 cm high according to
Herbig) in a tomb on the right-hand side of the road, immediately to the north of the
Ponte Terrano. Ainsley (in Dennis 1845:139) reported traces of red paint in the letters
(cf. MF 83-85).
leiueliopạrtis
uolti[
Sinistroverse, Faliscan alphabet with reversed s. Lanzi erroneously gave the first word
as leiuilio. The first letters of the second word are a (damaged) p followed by or .230
Garrucci took this as a v used for /u/ (purtis 1860, SIL; pfrtis ‘Purte’ 1864), in which he
was followed by Bormann, Herbig and Jacobsohn. The likelihood both of a v in an
inscription in Faliscan alphabet and of its being used for /u/ is very doubtful, however.
Pace Herbig (CIE), Schneider’s pẹrtis231 (adopted by Deecke, Von Planta, and Conway)
and Vetter’s pạrtis (adopted by G. Giacomelli) are both possible and more attractive: a
similar a appears in MF? 111. The ending of pạrtis is surprising, both in the absence of
o (§3.6.6.2) and in the presence of -s (§3.5.7d): it is perhaps an adaptation of an
Etruscan name. The shape of the niche makes it unlikely that it was closed with tiles,
and the inscription therefore probably ended with uolti, either an abbreviation of a
patronymic adjective uolti(o), or the genitive of the father’s name (§7.5).
Bibliography: Buonarruoti 1726:35-6 (autopsy); Gori 1733:77-8; †Passeri 1740:444: Passeri 1767:130;
Lanzi 1824:392 ¢466²; Dennis 1845:139 (autopsy);232 [Dennis 1848:124-5]; Orioli 1854:XXII; Garrucci
1860:269-70 (autopsy); Garrucci 1864:60; Fabretti CII 2441bis,a; Garrucci SIL 793; [Dennis 1878:94];
Zvetaieff IIM 52; Zvetaieff III 54; Schneider 1886:105 ¢4²; Bormann CIL XI.3160 I,6; Deecke
1888:128-30 ¢1²; Von Planta 1897:588 ¢317²; Conway 1897:374-5 ¢xl.15²; Herbig 1910:190-1 ¢30²
(autopsy); Jacobsohn 1910:3 ¢7²; Herbig CIE 8205; Buonamici 1913:60 ¢17²; Vetter 1953:299 ¢279²;
[Ward Perkins & Frederiksen 1957:145 (autopsy)]; G. Giacomelli 1963:78-9 ¢77² (autopsy); FI II.1
pp.179, 236 (autopsy). Photograph of squeeze: Herbig CIE 8205. Drawings: Buonarruoti 1726
tab.LXXXII.1 (reproduced in Gori 1733 tab. III.1, CII tab. XLIII); Garrucci 1860 tav.G.3; Garrucci
1864 tav. III.4 (reproduced in IIM tab. VIII.6, Deecke 1888 Taf.I); Herbig CIE 8205; FI II.1 pp.179,
236.
229
This was the first Faliscan inscription to be published (in 1726), although it was not the first
Faliscan inscription to be recorded: that honour was reserved (in 1676), for the group LtF 205,
MLF 206-207, MLF/Etr 208-209, and MLF 210.
230
Buonarruoti, Gori, Lanzi, and Orioli gave these letters in reversed order. Buonarruoti, Gori,
and Lanzi also read it instead of ti.
231
Schneider erroneously ascribed the same reading to Garrucci.
232
Dennis erroneously treated the inscription as if yet unpublished.
441
CHAPTER 13
80-81. The following inscriptions were discovered by Garrucci and seen only by him
(“frustra quaesivi”, Herbig CIE 8207-8208).
80. Painted in black on a plaster over a loculus in a tomb on the right-hand side of the
road. Underneath the text runs a decorative border.
[ma]rcopleinamarcioman[o]mocauiacue
[u]eculiauoltiliauentarc[......heccupa]nt
Sinistroverse, Faliscan alphabet. The size of the initial lacuna is nowhere indicated, but,
judging from Garrucci’s drawing, was probably no more than two letters in the first line
and one in the second, assuming that the inscription started at the end of the loculus:
Bormann’s restoration [mar]co (adopted by all later editors except Deecke) is therefore
preferable to Garrucci’s [uenta]rco (adopted by all early editors). In Garrucci’s drawing,
the lacuna between man and mo is rather large for a single o, but the restoration can
hardly be anything else, whether one reads man[o]mo with Garrucci or emends to
max̣[o]mo (Schneider, Deecke, Vetter, Pisani). The emendation is unnecessary, as the
cognomen manomo is attested in MF 149 (see §7.9). In the second line, ]eculia was
thought to be complete by the early editors (Garucci SIL, Schneider, Zvetaieff), but
Bormann compared ]eculia to u[.....]a MF 81 and read [u]eculia, which was adopted by
all later editors.
The restoration of the last lacuna is problematic. The text apparently ended with
hec  cupa]nt, but what preceded this is less clear, especially as it cannot be ascertained
how precise Garrucci’s drawing is with regard to the size of the lacuna. As the woman is
already described with praenomen, gentilicium, and patronymic adjective, uentarc[ can
hardly be part of her name as well unless it is either a cognomen (Herbig) or a second
gentilicium (G. Giacomelli); on the other hand, uentarc[ia  hec  cupa]nt is certainly
too short for the lacuna. Some of the earlier editors, who adopted Garrucci’s [uent]arco
in the first line, restored uentarc[i  uxor (Garrucci 1864, Zvetaieff, Vetter) or just
uentarc[i (Garrucci SIL, Zvetaieff, Schneider Bormann, uentarc[oi Deecke). Apart from
the fact that adopting Bormann’s [mar]co in the first line makes this virtually impossi-
ble, uentarc[i  uxor  hec  cupa]nt is probably be too long for the lacuna, whereas
uentarc[i  hec  cupa]nt is certainly too short. Uentarc[ was probably the name of a
third person: perhaps it should be read as uen tarc[, with an abbreviated praenomen uen
= Ven(el), followed by a gentilicium Tarc[....., e.g. tarc[oneo, tarc[uineo, or tarc[onteo
(cf. Tarcontius in CIL XI.3370 from Tarquinii).233 The careful writing, the decoration,
and the complete onomastic formula make this one of the most elaborate Faliscan
sepulchral inscriptions.
233
Pisani’s cauiacue / [u]eculia  uoltilia  uentarc[... ‘Gaviaque Veculia Voltilia Ventarci f.’ is
unconvincing. In m  tito  tulio  uoltilio  hescuna MLF 346, the patronym precedes the second
gentilicium, so ‘Gavia Veculia Ventarcia vel sim., daughter of Voltius’ might not be impossible.
442
Bibliography: Garrucci 1864:60 (autopsy); Fabretti CII 2441bis,e; Garrucci SIL 796; Zvetaieff IIM 53;
Zvetaieff III 55; Schneider 1886:105 ¢7²; Bormann CIL XI.3160 II.1; Deecke 1888:135-9 ¢7²; Conway
1897:372 ¢314²; Herbig CIE 8207; Buonamici 1913:61 ¢18²; Vetter 1953:299 ¢281a²; G. Giacomelli
1963:79-80 ¢79,I²; Pisani 1964:340. Drawing: Garrucci 1864 tav.III.6 (reproduced in CII tab.XLIII, IIM
tab.VIII.7, Deecke 1888 Taf.I, CIE 8207).
81. Painted in red on plaster over a loculus.

cau[eculi]a
cae[c]ṇataθania
Sinistroverse, Faliscan alphabet. The c of the first ca is dextroverse, indicating a
woman’s name (§11.2.4.3). The e is cursive. The gentilicium has been restored as
u[eculi]a (Herbig CIE, Vetter, G. Giacomelli) from comparison with MF 80. The
restoration e[c]nata was proposed already by Garrucci (1864), but taken up again only
by Herbig. Deecke’s ḷạ[u]ṭ[n]ata ‘liberta’ is impossible. The second letter of θania is ,
a correction of n to a (Garrucci; Zvetaieff Vetter, G. Giacomelli), rather than a ligature
an (Bormann, Deecke, Conway, Herbig), which would give a geminated spelling
θannia that is very rare in Middle Faliscan (§11.2.4.3). Garrucci took the inscription as
pertaining to two women, interpreting θania as a matronym. Other editors took ca in the
second line as a man’s praenomen, since, contrary to the ca in the first line, the c is not
reversed here. Herbig thus took the inscription as pertaining to one woman, but his
explanation of ca  e[c]ṇata as ‘Gai nata’ and of θannia (as he read it) as a cognomen
are unconvincing. Vetter’s interpretation of the second line as containing the names of
three slaves of Gavia Veculia, i.e., Gavius, Egnata, and Thannia, is certainly preferable
to this, and was in fact adopted by G. Giacomelli. This is giving too much importance to
the fact that the c in ca in the second line is not reversed. The use of the reversed initial
to indicate a woman’s name is very rare and inconsistent (§11.2.4.3): apart from that,
the second line may have been written at a different time than the first. Ca in the second
line may therefore be feminine and be taken together with e[c]ṇata (for a gentilicium
Egnatius, see §7.8.1.50), followed by the name of a third woman.
Bibliography: Garrucci 1864:60-1 (autopsy); Fabretti CII 2441bis,f; Garrucci SIL 795; Zvetaieff IIM
54; Zvetaieff III 56; Schneider 1886:105 ¢8²; Bormann CIL XI.3160 II.2; Deecke 1888:139-40 ¢8²;
Conway 1897:372, 527 ¢26*²; Herbig CIE 8208; Vetter 1953:299-300 ¢281b²; G. Giacomelli
1963:79-80 ¢79,II². Drawing: Garrucci 1864 tav.III.7 (reproduced in CII tab.XLIII, IIM tab.VIII.8,
Deecke 1888 Taf.I, CIE 8208).
82. Cut underneath a loculus in a tomb to the left of the Ponte Terrano.234 Letters c.10
cm high.
234
Garrucci described the location as “sulla rupe destra che domina il rio dell’acqua forte
accanto al ponte Terrano” (1860:269, echoed in Deecke 1888:131) and as “nella rupe a sinistra
del ponte Terrano” (1864:60, quoted in Zvetaieff 1884:46, 1886:22). The statements are not
incompatible: leaving Civita Castellana and looking to the left from the Ponte Terrano, the
tombs are on the right side of the gorge.
443
CHAPTER 13
uel[]uis niolna
Sinistroverse, Faliscan alphabet with reversed s. Garrucci’s drawing shows an empty
space between uis and ni, on which neither he nor any other editor remarks.
In view of *[.]pi  uesθi  cela MF 83 and caui[  ]ṭ**(*)[i ] cela MF 84, most
editors take olna as a noun, with uel and uisni as genitives (in the case of uel, apparently
an abbreviated genitive). Olna was explained by Garrucci as ‘ulna’ = ‘loculus’ (“misura
corrispondente invero all’altezza dei loculi” 1860:269), which was apparently adopted
by Zvetaieff and Schneider (as both print olna without a capital), by Deecke as ‘olla’,
and by Vetter as ‘ulna’=‘pulvinus’ (“Übersetzung von etr. hupni”, 1953:299), which
was adopted by G. Giacomelli. I find none of the explanations of olna as a noun
convincing: rather, it is a second gentilicium (as Schulze and Herbig suggested),
perhaps marking the name of a freedman (cf. Rix 1965:376-8). For other instances of a
double gentilicium from the area, e.g. m  tito  tulio  uoltilio  hescuna MLF 346, see
§7.6.
Bibliography: Garrucci 1860:269 (autopsy); Garrucci 1864:60; Fabretti CII 2441bis,b; Garrucci SIL
794; Zvetaieff IIM 51; Zvetaieff III 53; Schneider 1886:105 ¢3²; Bormann CIL XI.3160 I.5; Deecke
1888:131-2 ¢2²; Conway 1897:375 ¢xl.16²; Schulze 1904:73 n.3; Herbig 1910:105 (autopsy); Herbig
CIE 8206; Vetter 1953:299 ¢280²; G. Giacomelli 1963:79 ¢78². Drawings: Garrucci 1860 tav.G.2;
Garrucci 1864 tav.III.5 (reproduced in CII tab.XLIII, IIM tab.VIII.5, Deecke 1888 Taf.I, CIE 8206).
83. Cut over the entrance of a tomb (letters c.21 cm high) to the left of Ponte Terrano.
“Rimae litterarum recentiore tempore gypsatae sunt” (Herbig CIE 8209): cf. MF 84 and
85.
*[.]pi:uesθi:cela
Sinistroverse, Faliscan alphabet with reversed s. Garrucci’s drawing gives the
praenomen as , Thulin’s as , and Herbig’s as . The first letter is usually
read as t, but t[i]pi (Garrucci 1860, 1864, Fabretti, Zvetaieff, Schneider) or t[e]pi
(Bormann Deecke, Conway) is not attested from the ager Faliscus or Capenas, while
Herbig’s [trep]i and Vetter’s and G. Giacomelli’s [ui]pi appear to be incompatible with
the traces. The third letter of the gentilicium is the top half of a reversed s (Herbig): the
earlier uetθi and uezẓθi (Deecke, Conway) can be disregarded. Cela is a noun ‘cella’, as
in MF 12, 84, and MLF 285 (not, as Herbig had thought, a cognomen, cf. §7.9): cf.
§8.10.3
Bibliography: Mommsen 1860:451; Garrucci 1860:270-1 (autopsy); Garrucci 1864:59; Fabretti CII
2441bis,c; Garrucci SIL 791; Zvetaieff IIM 48; Zvetaieff III 50; Schneider 1886:105 ¢2²; Bormann CIL
XI.3160 I,2; Deecke 1888:133-4 ¢4²; Conway 1897:375 ¢xl.18²; Thulin 1907:275-6 ¢10²; Herbig
1910:101 ¢14²; Jacobsohn 1910:3 ¢10²; Herbig CIE 8209; Vetter 1953:300 ¢282²; G. Giacomelli
1963:80 ¢81². Drawings: Garrucci 1860 tav.G.4; Garrucci 1864 tav.III.2 (reproduced in CII tab.XLIII,
IIM tab.VIII.2, Deecke 1888 Taf.I); Thulin 1907:276 (reproduced in CIE 8209).
444
84. Cut over the entrance to a tomb to the left of the Ponte Terrano. Letters c.16 cm
high: “sulci litterarum, olim rubro colore expicti, a recentioribus gypsatae sunt” (Herbig
CIE 8210): cf. MF 83 and 85.
caui[:]ṭ**(*)[i]:cela
Sinistroverse, Faliscan alphabet. The remains of the gentilicium are given by Garrucci
as (five letters), and by Herbig as (four letters). It therefore started
in Tal- or Tri- (Garrucci 1864): Garrucci’s t(i)..pi.. (1877) and Deecke’s ṭ[repi] are
impossible. Herbig, and after him Vetter and G. Giacomelli, read tạḷị,235 but this leaves
some space after the i. The text consists of a name in the genitive followed by cela
‘cella’, as in MF 12, 83, and MLF 285: cf. §8.10.3.
Bibliography: Mommsen 1860:451; Garrucci 1860:271 (autopsy); Garrucci 1864:59; Fabretti CII
2441bis,d; Garrucci SIL 790; Zvetaieff IIM 47; Zvetaieff III 49; Schneider 1886:105 ¢1²; Bormann CIL
XI.3160 I,1; Deecke 1888:132-3 ¢3²; Conway 1897:375 ¢xl.17²; Herbig 1910:101 ¢13² (autopsy);
Herbig CIE 8210; Vetter 1953:300 ¢283²; G. Giacomelli 1963:81 ¢82² (autopsy). Drawings: Garrucci
1860 tav.G.5; Garrucci 1864 tav.III.1 (reproduced in CII tab. XLIII, IIM tab.VIII.1, Deecke 1888 Taf.I);
Herbig CIE 8210.
85. Cut to the left of the entrance of a tomb to the left of the Ponte Terrano. Dennis and
Zvetaieff reported traces of red paint (cf. MF 83 and 84), which were probably ‘recent’
addititions. Letters c.15 cm high.
tuconu
Sinistroverse, Faliscan alphabet. Early editors misread the inscription in various ways
(ticonu Garrucci 1864, SIL; tucθnu Dennis 1887; hucome Del Frate), but the text is
certain, and, according to Herbig (from autopsy), complete. The form is usually
interpreted as an Etruscan genitive in -u(s) (Garrucci, Deecke, Herbig, G. Giacomelli).
Herbig’s alternative, interpreting it as a Faliscan genitive plural tuconu(m) (CIE), is
unattractive, since the expected form would be tucono(m), unless the inscription is
assumed to be Late Faliscan (cf. [fel]ịcinatiu LF 384) Perhaps the text could be read as t
u(e)conu, with a gentilicium as in ueicọno MF 88.
Bibliography: Dennis 1845:139 (autopsy); Dennis 1848:124; Garrucci 1860:270 (autopsy); Garrucci
1864:59-60; Fabretti CII 2453; Garrucci SIL 792; Zvetaieff IIM 49 (autopsy); Zvetaieff III 51; Schnei-
der 1886:105 ¢6²; Dennis 1878:94; Bormann CIL XI.3160 I.3; Deecke 1888:134-5 ¢5²; Conway
1897:527 ¢27*²; Del Frate 1898:77 (autopsy); Herbig 1910:186-7 ¢24²; Herbig CIE 8211; Buonamici
1913:61-2 ¢19²; Vetter 1953:300 ¢284²; G. Giacomelli 1963:81 ¢83² (autopsy). Drawings: Garrucci
1864 tav.III.3 (reproduced in CII tab.XLIII); Zvetaieff IIM tab.VIII.3 (reproduced in Deecke 1888
Taf.I); Herbig CIE 8211.
86. Cut over a loculus in a tomb close to the tomb of MF 85.

uoll[---]
235
G. Giacomelli erroneously has caui[i]t(ali) instead of caui[:]t(ali).
445
CHAPTER 13
Sinistroverse, Faliscan alphabet. The last letters are . Although this can be read as ll,
cf. uolḷia MF 47, a geminated spelling is very rare (§11.2.4.3), and editors have
therefore read uolṇ[---] (Zvetaieff, Bormann, Deecke) or uolụ[---] (Herbig), or
emended to uolt[---] (Vetter).
Bibliography: Zvetaieff IIM 50 (autopsy); Zvetaieff III 52; Bormann CIL XI.3160 I.4; Deecke
1888:135 ¢6²; Conway 1897:375 ¢xl.14²; Del Frate 1898:77 (autopsy); Herbig CIE 8212; Vetter
1953:300 ¢285²; G. Giacomelli 1963:81 ¢84². Drawing: Zvetaieff IIM tab.VIII.4 (reproduced in Deecke
1888 Taf.I, CIE 8212).
87. Cut underneath a loculus (length 78 cm, letters 18 cm high). Third century (G.
Giacomelli).
mareina
Sinistroverse, Faliscan alphabet. Vetter’s proposal ma pleina (in G. Giacomelli
1965:551) is apparently impossible.
Bibliography: G. Giacomelli 1961:321 ¢1² (autopsy); G. Giacomelli 1963:80 ¢80²; Olzscha 1965:123;
G. Giacomelli 1965a:551. Photograph: G. Giacomelli 1963 tav.IX; G. Giacomelli 1965a
tav.CXXXIVb. Drawing: G. Giacomelli 1961:321 fig.1.
88-89. The following inscriptions were painted within a coloured border around a
loculus in a tomb “in Terrano” (Thulin 1907:268), 88 painted in red downwards along
the left side of the loculus, and 89 downwards along the right side and under the same
loculus. They are known only through Mengarelli’s apographs.
[uo]ltio[]ueicọnolecetḥec
[4-5]ạhac****a[?]ạ[?]ṃmaximo
Sinistroverse Faliscan alphabet. Although the c in lecet is and the h in hec , the
last two words of 88 are certainly to be read as lecethec (Thulin’s and Herbig’s leset :
leo makes no sense): the form of the c may be due to the fact that the inscription was
painted vertically, while a similar h occurs in MF 18. For lecet instead of the usual
cupat, cf. §5.3.1.2-3, §6.3.13: it is a form of the verbal root underlying the noun used to
designate the loculus, lectu Lat 251, lete MLF 285, and perhaps l[..........] MF 17 and
let MLF 360. A Sabellic parallel, also occurring beside qupat, is South Picene veia|t in
apaes  qupat [ e]smín  púpúnis  n|ír  mefiín  veia|t  vepetí MC.1.
In the gentilicium, Mengarelli’s drawing shows between ue and no. The word
has therefore been read as ueisc ̣no (Thulin, Herbig), ueiṣọno (Herbig) and uersno
(Vetter, G. Giacomelli). As the fourth letter is identical to the c in lecet (Thulin and
Herbig in fact read both letters as s), and can hardly be taken together with the i as r (as
does Vetter), ueic seems certain. The , standing between ueic and no, can hardly be
anything other than a vowel; it is probably an o, deformed, like the c’s, by the unusual
way of writing.
446
In 89, the first legible letter is , either an x (G. Giacomelli) or an a (Thulin,

Herbig, Vetter). The traces of the next word are hac****a. The inscription then
continues horizontally underneath the loculus with , that is,
either r (Vetter) or a (Thulin), possibly preceded by one very small letter, and followed
by an illegible trace and the upper part of an m. The last word is maximo, with only the
lowest part of the i preserved (Thulin’s mammo is impossible). The interpretations by
Thulin, who divides the inscription into two, reading ..ạ ha...a and .a..t : maṃmo, and
by Vetter, whose ------a hac ̣---a : (-)ṛ--ṭ: maxịmo ‘...a(m) ha(n)c ..am ...t Maximus’
assumes a type of text that is completely without parallels in the Faliscan sepulchral
inscriptions. The masculine cognomen maximo indicates that the preceding text
consisted of a man’s praenomen, gentilicium, and a filiation. Dividing [4-5]ạ hac****a
or [4-5]x̣ hac****a before the h seems the obvious thing to do even though an inter-
punct is absent. This gives a praenomen [4-5]ạ ([uolt]ạ or [iun]ạ) or [se]x̣, followed by
a gentilicium hac****a ([uolt]ạ hac ̣***a: Herbig). The filiation is almost illegible and
is so short that it was probably abbreviated (ạ[i]ṃ ‘Aem(ius)’?).
Bibliography: Thulin 1907:268-70 ¢5a-c²; Herbig CIE 8213a-b; Vetter 1953:301 ¢286A-B²; G.
Giacomelli 1963:81-2 ¢85a-b²; Pisani 1964:340 ¢144G². Drawing: Mengarelli in Thulin 1907:269.
13.8. The tomb near Torrente Purgatorio
Torrente Purgatorio flows along the northern side of Terrano, joining the Rio Maggiore
just west of the Ponte Clementino. In 1881, a fourth- or third-century tomb with
nineteen loculi was discovered on its left bank. It had been plundered in antiquity,
during which the tiles with the inscriptions had been smashed. According to Lucidi, the
discoverer of the tomb and owner of the terrain, the tiles belonging to the principal
loculus were stolen shortly after its discovery (Gamurrini 1883:165).236 Most of the tiles
pertain to a gens Caelia (MF 90-97, 104),237 whose parentage was tentatively recon-
structed by Peruzzi (1964d), and a gens Raeclia (MF 98-100). The latter name was read
by earlier authors only in MF 99 (reiclio), but I read the same name also in MF 98
(reic ̣[lio]) and in MF 100 (reị[cli.]).
90-91. The titulus prior (90) was painted in a dark colour on plaster across the back of
two tiles (70u47 cm, letters c.9 cm high). This inscription was later washed over with
plaster, on which was then painted the titulus posterior (91). As this second coat of
plaster has crumbled away almost completely, 91 is in a far worse state than 90.
236
Gammurrini (whence Bormann, Deecke, and Conway) erroneously placed the Torrente
Purgatorio on the east side of Civita Castellana. This error was corrected by Herbig (CIE
8214-8231 p.39).
237
Vetter (1953:301-2), who read [c]elio : cailio, rendered the praenomen as ‘Gellius’.
447
CHAPTER 13
[1]
[leu]2elio:cai3lio[:...]
[1]
[max]2om[o:]re3x[:..]**
Sinistroverse, Faliscn alphabet. The praenomen was restored as leiu]elio by Thulin, and
the second line as [max]om[o. This was adopted by all later editors, except Vetter (and
Peruzzi), who read the praenomen as c]elio ‘Gellius’, with [max]|omo at the end of the
line. This is impossible, both because there is not enough room for a c to precede ]elio
on the same tile and because this restoration would leave no room for the filiation in the
first line. The gentilicium is cailio; Deecke’s ceilio (adopted by Conway) is a misread-
ing. The gentilicium was probably followed by an abbreviated filiation, of which
nothing is left. In the second line, Thulin read omo:rex, but in the other early autopsies
the second o is invisible. The x of rex is now invisible, but was read in all early
autopsies. This word, read with great uncertainty in 91, is now attested also in rẹ[x in
249 and perhaps also in my reading of LtF 231. Of the end of the line only some badly
legible traces are preserved, which have been read as ...iri (Gamurrini, whence Bor-
mann), ..iai (Conway, erroneously stating that the r and the a have the same form),
///riso or ịṣpṛịṣo (Thulin), ***ṛẹọ̣ (Herbig), and [:m]ạrọ (Vetter). None of this can be
sustained from the traces that are now left.
Below, I present my reading of 91 alongside the readings of the early autopsies. The
tiles are presented in the order in which they were placed in 90, but the inscription is so
damaged that it is impossible to say whether this is correct.
inv. 8154 inv. 8172 Herbig Thulin Gamurrini
ņņņņņņņņņņņņņņņņņņ ņņņņņņņņņņņņņņņņņņņņņņņņņņņņņņ
*oc*[.] [..]*[....] *ic ̣ọ* ..... ...eico.uoc...
uaṭ[..] [.......]* uaṭṛ uap/ ...uatụ..eco...
***xi[ [..]*[...]* ***x* .ṛex: ...ụ.exi......238
*ma*[ [...]*[...]ạ .mạ* aịm ...pal....imr...
Sinistroverse, Faliscan alphabet. The reading is extremely uncertain. In the second line,
the a may be followed by either a t or a p. In the third, the traces that precede the x
could perhaps be an e, but Thulin’s ṛex (“zwar [ist] nur die linke Hälfte des r erhalten”,
1907:278) can no more be read; following the x is i, not Thulin’s . Gamurrini and
Deecke saw traces on the second tile as well; although still visible, they are illegible,
except for the last line, where an r or an a (..ima..? Deecke) can perhaps be recognized.
The only attempt at interpretation is Peruzzi’s,239 who interpreted Gamurrini’s and
Deecke’s readings as a carmen epigraphicum, reading d]eico.uoc[e ‘dico uoce’, atr*
(from Herbig) as a form of ater, eco ‘ego’, exi[ as a form of exire or exitus, and imr[ as
238
Bormann erroneously rendered Gamurrini’s reading as //v.exci//////.
239
Earlier editors limited themselves to names (Deecke: praenomen e.g. num]eico, gentilicium
e.g. uoc[onio; Thulin: uap... cf. Vapusius, aima fem. of Aemus; Herbig: gentilicium ụaṭṛ[io).
448
a form of imbrex or imber. Not all of this is plausible (the reference to an imbrex or an
imber is unclear, and the parallel quoted for mr = /m(b)r/, umrie Etr XLIII, is Etruscan
rather than Faliscan) and the likelihood of such a carmen may be questioned in view of
the function of Faliscan sepulchral inscriptions (§11.1.4.1).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8154+8173). Bibliography:
Gamurrini 1883:166 ¢5² (autopsy); Bormann CIL XI.3162c,4a-b; Deecke 1888:145-6 ¢14a-b²; Conway
1897:373-4 ¢319a-b²; Thulin 1907:277-8 ¢11a-b² (autopsy); Herbig CIE 8214a-b (autopsy); Vetter
1953:301-2 ¢287a,A-B²; G. Giacomelli 1963:82-4 ¢86,Ia-b²; Peruzzi 1964d:310-1. Drawings: Deecke
jr. in Deecke 1888 Taf.I (reproduced in CIE 8214a-b); Thulin 1907:277, 278 (reproduced in CIE
8214a-b); Herbig CIE 8214a-b.
92. Painted in red on plaster along the length of a tile (60u47 cm, letters 10 cm high).
tan[---]
cail[ia ?---]
Sinistroverse, Faliscan alphabet. Herbig’s tan is preferable to the pan of the earlier
editors. The a in the second line is very damaged, but can hardly be anything else than
cail[---]. If the inscription occupied one tile, the first line was probably tan[a (Herbig)
or tan[ia; otherwise, it can be read as e.g. tan[acuil] with cail[ia  ..  fi] in the second
line. Vetter’s tan|cail is meaningless, Pauli’s tan|cṿil (in Herbig CIE 8216) impossible.
Bibliography: Gamurrini 1883:166 ¢9² (autopsy); Deecke 1888:147 ¢18²; Conway 1897:374 ¢xl.4²;
Herbig CIE 8216 (autopsy); Vetter 1953:301-3 ¢287c²; G. Giacomelli 1963:82-4 ¢86,III²; Peruzzi
1964d:310. Drawings: Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8216.
93. Painted in red on plaster on a tile.

laie[---]
cai[lia---]
Sinistroverse, Faliscan alphabet. The first line seems to have contained a man’s name
with a gentilicium in Ie- (for suggestions with regard to this name, cf. §7.8.1.75),
followed in the second line by a woman’s name cai[lia (not Deecke’s and Herbig’s cai
‘Gai’), either a praenomen or a gentilicium (thus Peruzzi).
Bibliography: Gamurrini 1883:166 ¢15² (autopsy); Deecke 1888:149 ¢24²; Conway 1897:375 ¢xl.10²;
Herbig CIE 8217 (autopsy); Vetter 1953:301-3 ¢287d²; G. Giacomelli 1963:82-4 ¢86,IV²; Peruzzi
1964d:311-2. Drawing: Deecke jr. in Deecke 1888 Taf.I (reproduced in CIE 8217).
94. Painted in red on plaster across the back of a tile (67u46 cm, letters 10-15 cm high).
[1]
[---c]2ẹliocesifi
[1]
[---]2cauia * *
[1]
[---]2a
Sinistroverse, Faliscan alphabet. The e in cesi is cursive (): the gentilicium can
therefore be c]ẹlio, ca]ilio or ce]ilio. In the second line, Herbig’s cauia (read errone-
ously as icruir by the earlier editors) is followed by two traces that may be the traces of
449
CHAPTER 13
a titulus prior that were seen by Thulin (in Herbig CIE 8215). The last line, first read by
Pauli (in CIE), is [---]a (Herbig), e.g. the formula [hecup]a or the end of the woman’s
gentilicium or patronym. Peruzzi, restoring the first line as ce.c]elio.cesi.fi, regarded the
text as a new epitaph for the deceased of MF 90, made after the death of his wife cauia.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8180). I succeeded only in
finding the fragment containing the letters fi and several traces of the second line. Bibliography:
Gamurrini 1883:166 ¢4² (autopsy); Bormann CIL XI.3162c,3; Deecke 1888:145 ¢13²; Conway
1897:373 ¢318a²; Herbig CIE 8215 (autopsy); Vetter 1953:302 ¢287b²; G. Giacomelli 1963:82-4 ¢86,II²;
Peruzzi 1964d:310. Drawings: Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8215.
95. Painted in red on plaster along the length of the back of two damaged tiles (the first
max. 30u45 cm, the second max. 30u45 cm; letters 13-15 cm high).
[1]
[---]2celio3[---]
[1]
[---]2*:hec:c3upa[t?]
Sinistroverse, Faliscan alphabet. The earliest editors treated the tiles as separate
inscriptions celio (or celioi) utpos and ...upa. Thulin, however, rightly read hec:c instead
of utpos (cf. Herbig’s drawings) and joined it to the tile with upa.240 The lost first line
on the second tile probably contained the man’s filiation. At the beginning of the second
line, Thulin read a trace , probably part of a, on a shard that I was unable to find. At
the beginning of the text, one tile is missing, which contained the man’s praenomen in
the first line, and in the second line perhaps a woman’s name (---]ạ). Cupa[t?] is
therefore perhaps plural rather than singular.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8162+8174). I did not succeed in
finding the shard of the first tile with the trace at the beginning of the second line. Bibliography:
Gamurrini 1883:166 ¢3+8² (autopsy); Bormann CIL XI.3162c,5; Deecke 1888:146-7 ¢15+17²; Conway
1897:374 ¢xl.1²; Thulin 1907:279 ¢12² (autopsy); Jacobsohn 1910:4 ¢21²; Herbig CIE 8218 (autopsy);
Vetter 1953:302 ¢287e²; G. Giacomelli 1963:82-4 ¢86,V²; Peruzzi 1964d:311. Drawings: Deecke jr. in
Deecke 1888 Taf.I; Thulin 1907:279 (reproduced in CIE 8218); Herbig CIE 8218.
96. Painted in red on plaster along the length of the back of a tile (68u43 cm, letters c.15
cm high in the first line, the second line slightly smaller).
1
iunace[2][lio---]
1
arutielia[2][?---]
Sinistroverse, Faliscan alphabet. The gentilicium can be read as ce[lio or ce[ilio
(Herbig; ce[ilio Jacobsohn, ce[lio Vetter, G. Giacomelli). In the second line, the u
appears to have been corrected from a t, whose sidestroke, apparently obliterated on
purpose, is still vaguely visible. Arutiela was first proposed by Pauli (in Herbig CIE
8221). A second tile will have contained, in the first line the remainder of the gen-
tilicium and the man’s filiation, and in the second perhaps the woman’s filiation or -cue.
240
Thulin’s drawing gives a false impression of a great difference in size between the two tiles:
in my measurements, the sizes come to 45 and 43 cm respectively.
450
1883:166 ¢1² (autopsy); Bormann CIL XI.3162c,2; Deecke 1888:142-3 ¢10²; Conway 1897:373 ¢316²;
Jacobsohn 1910:4 ¢22²; Herbig CIE 8221 (autopsy); Vetter 1953:302 ¢287g²; G. Giacomelli 1963:82-4
¢86,VIII². Drawing: Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8221.
97. Painted in red on plaster on a tile fragment.241 Letters 11 cm high.

[---c]elio[---]
[---]r*i[....]*
Sinistroverse, Faliscan alphabet. The (cursive) e, , was seen only by Gamurrini (c]elio
Deecke, ...elio Conway; ce]lio Herbig, G. Giacomelli). He read the second line as rpi a
(e.g. tu]rpi[li]a Deecke), but the letter following the r, which is  in Deecke jr.’s
drawing and ᛧᚮ in Herbig’s, has also been read as a z (nobo]rzi[ni]a Deecke,
he]rzị[ni]a Thulin in Herbig CIE 8220, no]rzị[ni]a Herbig). Gamurrini’s a was not seen
by Deecke jr.; Herbig’s drawing shows an (illegible) letter. Gamurrini and Pauli (in
Herbig CIE) joined this tile to the one of 99, which is impossible.
Bibliography: Gamurrini 1883:166 ¢3² (autopsy); Deecke 1888:144 ¢12²; Conway 1897:373 ¢318b²;
Herbig CIE 8220 (autopsy); G. Giacomelli 1963:82-4 ¢86,VII². Drawings: Deecke jr. in Deecke 1888
Taf.I; Herbig CIE 8220.
98. Painted in red along the length of the back of a damaged tile (41.5* u 49 cm, letters
of the first line 12-13 cm high, those of the second line c.10 cm high).
[1]
[---]2reic ̣[lio]
[1]
[---]2ṃaxoṃ[o]
Sinistroverse, Faliscan alphabet. The first line has been read as reio by all editors
(ma]reio Deecke), but as only the right half of the c is left, it is also possible to read
reic[lio (cf. reiclio in MF 99). In the second line, the first letter was virtually obliter-
ated by the mortar used to keep the tile in place. The second letter is a, not r (Gamurrini,
Conway); of the last legible letter, only one vertical shaft is left. Thulin read saxoi,
Herbig saxola (adopted by Vetter and G. Giacomelli), but Deecke’s m]axom[o is
certainly not impossible: as there is not enough place for a (complete) m on this tile, it
will have been partly written on the tile missing at the beginning of the text. This tile
will have contained the man’s praenomen in the first line and his filiation in the second.
The missing letters of reic ̣[lio] and ṃaxoṃ[o] were written on the missing part of the
tile; it is not necessary to assume that another tile is missing at the end.
1883:165-7 ¢7² (autopsy); Deecke 1888:147 ¢16²; Conway 1897:374 ¢xl.2²; Herbig CIE 8226 (autopsy);
Vetter 1953:303 ¢287l²; G. Giacomelli 1963:82-4 ¢86,XIII². Drawings: Deecke jr. in Deecke 1888 Taf.I;
Herbig CIE 8226.
241
Herbig (from Pauli) gives the size of this fragment as “m. 0,45 a. x 0,695 l.”, which
corresponds to the size of a complete tile (§11.1.4.1c).
451
CHAPTER 13
99. Painted in red on plaster along the length of the back of a damaged tile (max. 46u42
cm, letters 10-14 cm high).
[1]
[---]2reiclio[?]
[1]
[---]2cẹsiḷị[a]
Sinistroverse, Faliscan alphabet. The text is largely that of Herbig (earlier editors read
seiclio|icasilio). What remains of the second letter of the second line is and may be a
damaged e, which would give cẹsili[a], rather than the damaged a () of the earlier
editors: cf. the e or a in MF 109. The last letter of the second line, , was read as n by
Herbig and Thulin (in Herbig CIE 8222), but is perhaps rather li (Pauli in Herbig CIE
8222). The last letter of cesiḷị[a will have stood on the missing part of the tile; it is not
necessary to assume that another tile followed the text. A tile missing at the beginning
will have contained the praenomen of the man in the first line and his filiation in the
second. Torelli proposed to read (p)reiclio (cf. Praecilia Setoriana in CIL XI.3181 from
near Fabbrica di Roma (?), but the fact that in MF 100 the gentilicium stands at the
beginning of the line pleads against this. Gamurrini and Pauli (in Herbig CIE 8222)
joined this tile to the one of MF 96, but that appears to be impossible.
Bibliography: Gamurrini 1883:166 ¢2² (autopsy); Bormann CIL XI.3162c,1; Deecke 1888:143-4 ¢11²;
Conway 1897:373 ¢317²; Herbig CIE 8222 (autopsy); Vetter 1953:302 ¢287h²; G. Giacomelli 1963:82-4
¢86,IX²; Torelli 1967:536. Drawings: Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8222.
100. Painted in red on plaster across the back of a damaged tile (max. 48u19 cm, letters
c.8 cm high).
[......]
reị[cli.]
m*[....]
cus[....]
ma[....]
Sinistroverse, Faliscan alphabet. A trace of the first line was seen by Thulin (in Herbig
CIE 8224) and Herbig. The second line is read as re[ by all editors, but part of a shaft is
visible after the e, and in view of reiclio in MF 99 and reic ̣[lio in MF 98, it is possible to
read reị[clio] or reị[clia]. Of the last letter of the third line only a shaft is preserved. In
the fourth line, the last letter (first read by Herbig) is certainly an s: it is impossible to
read cup ̣[at. The last line is ma[---] (Deecke) rather than mr.... (Gamurrini) or mọ[---]
(Thulin in Herbig CIE 8224). The fact that the inscription (uniquely) consists of five
lines implies that it probably occupied only one tile, e.g. [marco] | reị[clio] | ma[rcio],
followed by a woman’s name cus[....] | ma[  uxo]: alternatively, it would have been
part of a very long text indeed.
1883:166 ¢13² (autopsy); Deecke 1888:149 ¢22²; Conway 1897:374 ¢xl.8²; Herbig CIE 8224 (autopsy);
G. Giacomelli 1963:83-4 ¢86,XI². Drawings: Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8224.
452
101. Painted in red on plaster along the length of the back of a tile (68.5 u 47 cm, let.
7-13 cm).
tanacu[il]
anelia
uxoria
Sinistroverse, Faliscan alphabet. The first a is , the cross-over form between  and 
found also in MF 59-60, the second and third a are ; the fourth and fifth a and the r are
both , an instance of the confusion of  (a) and  (r): see §11.2.4.2. Of the first letter,
part of the top is missing, but Deecke’s tanacu[il] is certainly right (panacu Thulin).
The two interpuncts at the end of the second line are unique. The third line was first
seen by Thulin, who read uxor.ia/, which was adopted by G. Giacomelli. Herbig
(whence Jacobsohn and Vetter) read uxor.iṛ, but an abbreviation ia of a man’s name
occurs also in MLF 302 and LtF 341.
Bibliography: Gamurrini 1883:166 ¢12² (autopsy); Bormann CIL XI.3162c,6; Deecke 1888:148-9 ¢21²;
Conway 1897:374 ¢xl.7²; Thulin 1907:280-1 ¢14² (autopsy); Jacobsohn 1910:4 ¢23²; Herbig CIE 8223
(autopsy); Vetter 1953:302 ¢287i²; G. Giacomelli 1963:82-4 ¢86,X². Drawings: Deecke jr. in Deecke
1888 Taf.I; Thulin 1907:280 (reproduced in CIE 8223); Herbig CIE 8223.
102. Painted in red on plaster across the back of a tile (57 u 45 cm, let. 9-11 cm).
[1]
[---]2iena:ụ[3][---]
[1]
[---]2ono:ux[3][or]
Sinistroverse, Faliscan alphabet. The u in the first line could also be a tilted l. Deecke
took [---]iena as an Etruscan gentilicium, ụ[---] as the beginning of the father’s name in
the genitive (or of a patronym) and [---]ono as a cognomen, followed by ux[orcue. This
reading was adopted by Herbig and G. Giacomelli. There are no parallels in the Faliscan
inscriptions for a woman to be designated by just the word uxor, instead of a name.
Vetter’s interpretation, to take the whole inscription as referring to a woman, with
[---]ono as the ending of genitive of the husband’s name (e.g. petr]ono), is more
attractive. In both interpretations it is assumed that the inscription occupied three tiles.
1883:166 ¢11² (autopsy); Deecke 1888:148 ¢20²; Conway 1897:374 ¢xl.6²; Herbig CIE 8227 (autopsy);
Vetter 1953:303 ¢287m²; G. Giacomelli 1963:83-4 ¢86,XIV². Drawings: Deecke jr. in Deecke 1888
Taf.I; Herbig CIE 8227.
103. Painted in red on plaster across the upper part of the back of a damaged tile
(71umax. 41 cm, letters c.11 cm high).
1
nut*[2][---]
Sinistroverse, Faliscan (?) alphabet. The letters start close to the edge of the tile: it is
unclear if another tile preceded it. The last letter has been read as r (Gamurrini) and as a
(Deecke jr., Pauli and Thulin in Herbig CIE 8225): at present, little more than a vertical
453
CHAPTER 13
shaft is left (ṇut* Herbig). If the tile was of standard width, no more than two letters can
have followed on the same tile. Herbig and Vetter considered interpreting the word as
‘nutrix’. G. Giacomelli rightly rejected this, as the Middle Faliscan equivalent of Latin
nutrix, older noutrix in CIL I2.45, would have been notrix (§3.7.2). Peruzzi defended
Herbig’s interpretation, regarding nutrix as a Roman Luxuslehnwort. It is not certain,
however, if even in contemporary Roman Latin /o/ had already developed so far
towards /ō ̣/ as to be written u (Wachter (1987:313-3) thinks it was possible).242 Fur-
thermore, it would be the only Faliscan sepulchral inscription where an occupation is
mentioned. I think it is safer to read nut*[ as a name, perhaps as nu t*[---] (cf. nu in
MLF 309 and perhaps also in MF? 202).
1883:165-7 ¢10² (autopsy); Deecke 1888:148 ¢19²; Conway 1897:374 ¢xl.5²; Herbig CIE 8225 (au-
topsy); Vetter 1953:302 ¢287k²; G. Giacomelli 1963:82-4 ¢86,XII²; Peruzzi 1964d:312. Drawings:
Deecke jr. in Deecke 1888 Taf.I; Herbig CIE 8225.

[---]ila[---]
[---]ic[---]
Sinistroverse, Faliscan (?) alphabet. For the first line, Deecke proposed a connection
with Etruscan Veila; Herbig considered dividing ---i la---. The second line could be
ra]ic[li-, re]ic[li-, or perhaps r]ec[li-, if the shaft is half of a cursive e ().
Bibliography: Gamurrini 1883:166 ¢14² (autopsy); Deecke 1888:149 ¢23²; Herbig CIE 8228; G.
Giacomelli 1963:83-4 ¢86,XV². Drawing: Deecke jr. in Deecke 1888 Taf.I (reproduced in CIE 8228).
105-108. Four more fragmentary tiles were seen in 1887 by Deecke jr. in the Museo di
Villa Giulia. Only two (MF 105 and 107) were later seen by others. It is not recorded on
what grounds they were ascribed to the Torrente Purgatorio tomb, or why they did not
appear in Gamurrini’s description: MF 105 is well legible even today, and if its state is
representative of the whole group, Gamurrini could hardly have overlooked these
fragments if they had been discovered together with those published by him in 1883.
105. Painted in red on plaster on a two fragments of a tile (total size 56 u 26 cm; letters
c.10 cm high).
[---]celio[---]
Sinistroverse, Faliscan (?) alphabet.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8219, olim 8199). Bibliography:
Deecke 1888:150 ¢25²; Conway 1897:374 ¢320²; Thulin 1907:279 ¢13² (autopsy); Herbig CIE 8219
(autopsy); Vetter 1953:301-3 ¢287f²; G. Giacomelli 1963:83-4 ¢86,VI². Drawings: Deecke jr. in Deecke
1888 Taf.I; Thulin 1907:279 (reproduced in CIE 8219); Herbig CIE 8219.
242
Note, however, that shortly after 241, a T. Furius, perhaps an Latin immigrant craftsman at
Falerii Novi, still spelled his name as fourios (Lat 215).
454
106. On a tile fragment (presumably painted in red on plaster).

[---]cisi[---]
[---]ipo[---]
Sinistroverse, Faliscan alphabet with reversed s: the sign for this letter is given as ,
which is used for t in MLF 347, 350, and probably 351. Deecke compared Etruscan
Cisie; Herbig considered dividing the second line as ---i po---. Perhaps cisi is an error
(by the painter or by Deecke jr.) for c‹e›si with cursive e, .
Bibliography: Deecke 1888:150 ¢26²; Conway 1897:375 ¢xl.11²; Herbig CIE 8229; G. Giacomelli
1963:83-4 ¢86,XVI². Drawing: Deecke jr. in Deecke 1888 Taf.I (reproduced in CIE 8229).
107. Painted in red on plaster on a damaged tile (max. 50-38u47 cm).

[---]c ̣fli[---]
[---]ạfc[---]
Sinistroverse, Faliscan alphabet: the ạ is . Deecke proposed to read ‘G(ai) f(ilius)’ in
the first line and ‘A(uli) f(ilius)’ in the second. Pauli (in Herbig CIE 8230) read
ạ[n]ẹḷiọ|i? ap ̣c ̣, comparing anelia in MF 98, but it is impossible to see how this reading
can be derived from his drawing, which appears to suggest that the reading is
[---]*[.]ẹụi*[-- ---]ị[.]ap ̣c ̣[---].
1963:83-4 ¢86,XVII². Drawing: Deecke jr. in Deecke 1888 Taf.I (reproduced in CIE 8230); Pauli in
Herbig CIE 8230.
108. On three tile fragments (presumably painted in red on plaster: an autoptic descrip-
tion is lacking).
[---]a[---]
[---]iac[---]
[---]cal[---]
Sinistroverse, Faliscan (?) alphabet. Herbig’s suggestions [---]iac[ue ‘...ia-que’ and
cal[inia ‘Calinia’ are possible, but hardly more than guesses.
1963:83-4 ¢86,XVIII². Drawings: Deecke jr. in Deecke 1888 Taf.I (reproduced in CIE 8231); Pauli in
Herbig CIE 8231.
455
CHAPTER 13
456
Chapter 14
The inscriptions from Civita Castellana (Falerii Veteres) II
14.1. The inscriptions from the temples
14.1.1. The dedications. The inscriptions from the temples of Civita Castellana stand
apart from the other Faliscan inscriptions, which are virtually all sepulchral. Some can
be identified as dedications (apolonos EF 10 (§13.3), the Titus Mercus-dedications MF
113-126, and perhaps also MF 109); dedicatory, too, is sacra MF 127 and anae lauvcies
Etr XXIX (§19.3). Others are Besitzerinschriften rather than dedications (MF 110, 112,
MF? 128-131, and 133-134): apart from their provenance, there are no indications that
these objects were dedications. Unfortunately illegible is MF 132, from the frieze of the
temple of Contrada Celle.
14.1.2. The temples of Colle di Vignale. Colle di Vignale lies to the north-east of the
site of Civita Castellana. Like the latter, it is part of the tuff plateau surrounded by the
gorges of the Treia to the south and the Rio Maggiore to the north-west, the north, and
the east. Vignale is the site of the earliest settlement at Civita Castellana, which goes
back to the archaic period. Two temples, known as the Tempio Maggiore and Tempio
Minore, were built here in the sixth century. Although the habitation was abandoned in
the third century, probably as a result of the war of 241, a few second-century votives
indicate that the temples remained in use for some time afterwards (Moscati 1983:79).
The site was excavated in 1895-1896 under the auspices of Pasqui and Mengarelli. The
results of these excavations remain largely unpublished except for discussions by
Moscati (1983, 1990) and Carlucci (1995); see also FI II.1 p.381 with fig.121 p.191. At
least some of the epigraphical material belongs to the finds from the Tempio Maggiore,
as appears from a manuscript Catalogo degli oggetti scavati nello scavo di un tempio
sull’altura di Vignale, nella proprietà del nobile Sig. Cav. Rocco Trocchi (1896?)
quoted by Moscati (1983:81-7), which mentions EF 10 (§12.4), MF 109 and 110
(below), and Etr XXIX (§19.3).
109. Painted in red on the fragmentary bottom of a patera (letters 14-17 mm high).
[---]ạltai [
Sinistroverse. The quadruple interpunct may point to an early date. Of the first letter,
only two downward-slanting sidebars remain, but it is doubtlessly the same sign as the
457
CHAPTER 14
fourth letter. This is , probably an a of the cursive type described in §11.2.4.2,

although some editors regard is a a carelessly written : Thulin (in Herbig CIE 8031)
read u]ẹltẹi, comparing the curious e’s in MF 146, (§11.2.4.2); Herbig read ṿltvi.
Colonna and Rix both regarded the inscription as Etruscan, reading ven]ẹltvi and ?]ẹltvi
respectively. If the inscription is a Faliscan votive inscription, the form may be a
first-declension dative singular of a deity’s name or epithet.
Bibliography: Herbig CIE 8031 (autopsy); Vetter 1953:292 ¢261²; G. Giacomelli 1963:59 ¢32²;
Moscati 1985:87; Comella 1986:172-3 ¢33² (autopsy); Colonna 1993:299; Rix ET Fa 2.19. Photo-
graph: Comella 1986 tav.70c. Drawing: Herbig CIE 8031.
110. Scratched under a brown-varnished patera ( 9.5 cm, letters 7-8 mm high).
aie*
Sinistroverse. The last letter can be an a (aieạ Herbig) or a squarish o, but Nogara’s
apograph (in CIE 8032) shows : an aieạ or aieọ could (with some difficulty) be
interpreted as a name (cf. Stolte 1928:289, Hirata 1967:32-3). Colonna’s photograph
and Pandolfini’s drawing point rather to a v: Colonna, Pandolfini, and Rix in fact read
acev and place the text among the alphabetaries. The second letter is an i rather than a c,
however, and an a or o not unlike the last letter occurs in Cap 375 and 388.
Bibliography: Herbig CIE 8032 (autopsy); Vetter 1953:292 ¢263²; G. Giacomelli 1963:60 ¢34²;
Moscati 1985:87; Comella 1986:172 ¢32² (autopsy); Colonna 1990:136 (autopsy); Pandolfini &
Prosdocimi 1990:94 ¢III.35²; Rix ET Fa 9.3. Photograph: Colonna 1990 tav.Ib. Drawings: Herbig CIE
8032; Comella 1986 tav.78,R89; Pandolfini & Prosdocimi 1990:94.
111. Stamped inside a small black-varnished saucer ( 6 cm, letters 7 mm high).

ac
Sinistroverse. The a is , a rounded variant of  (see §11.2.4.2): a similar a appears
in MF 79. The name is stamped, and is probably the name of the potter.
Bibliography: Herbig CIE 8033 (autopsy); Vetter 1953:292; G. Giacomelli 1963:60 ¢35²; Comella
1986:172 ¢31² (autopsy). Photograph: Comella 1986 tav.66c. Drawings: Herbig CIE 8033; Comella
1986 tav.77,R76.
112. Scratched on a small vase (letters 8-13 mm high). Third to first century (Herbig).
iunai
Sinistroverse, Faliscan alphabet. The a has the form , which is normally used for r, but
in several later inscriptions also for a (§11.2.4.2). Either genitive (Pisani) or dative, see
§4.2.3 and §8.8.1. As iuna is a man’s name, not a god’s, both interpretations are
incompatible with a dedicatory inscription.243
243
Andrén (1940:87), erroneously assuming that iunai could “represent the name of Iuno”, uses
the inscription to substruct his theory that the Tempio Maggiore was the famous Faliscan temple
of Iuno. This theory is adopted by Riis (1981:55), without reference to this inscription.
458
THE INSCRIPTIONS FROM CIVITA CASTELLANA (FALERII VETERES) II
Bibliography: Herbig CIE 8548; Vetter 1953:292 ¢262²; G. Giacomelli 1963:59-60 ¢33²; Pisani
1964:344 ¢146Gd²; Moscati 1985:66; Comella 1986:172 ¢30² (autopsy). Photograph: Comella 1986
tav.67b. Drawings: Nogara in CIE 8548; Comella 1986 tav.77,R76.
14.1.3. The temple ruins ‘ai Sassi Caduti’. The temple ruins ‘ai Sassi Caduti’ lie at
the bottom of the gorge of the Rio Maggiore, between Contrada Le Colonette to the
west and the Colle di Vignale to the east. A private excavation conducted between
August 1901 and February 1902 yielded a large quantity of architectural terracottas as
well as a number of votives, including several inscribed vessels. Although few of the
excavation results have been published (cf. Mengarelli 1911, Della Seta 1917:166-77,
Andrén 1940:104-21), the finds indicate that the temple went back to the fifth century
and remained in use until the Roman period, surviving, like other temples at Civita
Castellana, the war of 241. The theonym Titus Mercus in the votive inscriptions from
the stips and the discovery of part of a terracotta statue of a wing-footed youth244 have
led to the conclusion that the temple was dedicated to Mercury or a similar deity.
113-126 (the ‘Titus Mercus-dedications’). The following inscriptions are attributed to

a third-century stips by Mengarelli (in Thulin 1907:297). This date has been adopted by
all later editors except Dohrn, who dated the inscriptions to the fourth century. Most of
these inscriptions, MF 113-124, are executed in a uniform style, painted in identical
black-varnished Etrusco-Campanian cups (height c.12 cm), as titoi | mercui | efiles (MF
113-117), titoi  mercui (MF 118-122), or mercui (MF 123-124). They may have been
produced in series as ready-made ex-votos to be sold to visitors to the temple, or
perhaps even as dedications imposed by the aediles as a sanction (see below). The
different shapes of the s,  in MF 113 and 115, but  in MF 118, suggest that at least
two people were employed in their production.
113. Painted in yellow on the bottom of a black-varnished ‘Etrusco-Campanian’ cup.
] titoi [
] mercui [
] efiles [
Sinistroverse, Faliscan alphabet. The s is .
From autopsy in the Museo di Villa Giulia, Rome (inv. 12544). Photographs: Torelli 1981:224 fig.97;
Morandi 1982 tav.IX.2; Comella 1986 tav.66d. Drawings: Thulin 1907 between pp.298-9 ¢36² (repro-
duced in CIE 8036); Morandi 1982:59; Comella 1986 tav.77,R77.
244
The statue is not the cult-statue, but may well have occupied a prominent position, possibly
as the central acroterion (see Mengarelli in Thulin 1907:297, 1911:66, Herbig 1914a:241-2,
Della Seta 1918:166, Andrén 1940:116-7 with pl.44 nr.143, and Melis in Sant p.113).
459
CHAPTER 14

] tito [
] mercui [
] efile [
Sinistroverse, Faliscan alphabet. The space after tito is too large for Jacobsohn’s tito[i].
The occurrence of tito side by side with titoi has been interpreted as an attestation of a
change towards a dative in -o (Thulin 1907:303), but it may be no more than a graphical
error (§4.3.3). Efile shows a very rare omission of -s after a long vowel, which may
likewise be an error (§3.5.7d).
From autopsy in the Museo di Villa Giulia, Rome (inv. 12545). Photograph: Comella 1986 tav.68d.
Drawings: Thulin 1907 between pp.298-9 ¢40² (reproduced in CIE 8037); Comella 1986 tav.79,R96.

[t]ito[i]
]ṃercu[i]
]efiles[
Sinistroverse, Faliscan alphabet. Only the left shaft of the m is preserved. The s is .
From autopsy in the Museo di Villa Giulia, Rome (inv. 12548). Photograph: Comella 1986 tav.68b.

] titoi [
] mercụ[i]
]efi[les]
Sinistroverse, Faliscan alphabet. Of the u, only the top left-hand corner remains.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12546). Photograph: Comella 1986 tav.67a.
117. Painted in yellow a fragment of a black-varnished ‘Etrusco-Campanian’ cup.

[titoi]
[mercui]
[efil]es [
Sinistroverse, Faliscan alphabet with reversed s. Thulin took this shard and 123 together
as [titoi] | merc[ui | efil]es, but they have rightly been separated by all later editors.

]titoimercụ[i]
Sinistroverse, Faliscan alphabet. Of the u, only the top half remains.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12547). Photograph: Comella 1986 tav.68a.
460

[t]itoimercu[i]
Sinistroverse, Faliscan alphabet.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 12549). Drawings: Thulin 1907
between pp.298-9 ¢44² (reproduced in CIE 8041); Comella 1986 tav.79,R97.

[ti]toimercu[i]
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 12550). Drawings: Thulin 1907

[ti]toimercui [
From autopsy in the Museo di Villa Giulia, Rome (inv. 12551+12553). Photograph: Comella 1986
tav.67c. Drawings: Thulin 1907 between pp.298-9 ¢44² (reproduced in CIE 8043); Comella 1986
tav.78,R90.

]titoiṃ[e]rcui [
Sinistroverse, Faliscan alphabet. Of the m, only the top left-hand corner remains.
From autopsy, one part in the Museo dell’Agro Falisco, Civita Castellana (inv.12552), the other in the
Museo di Villa Giulia, Rome (inv. 12554). Photograph: Comella 1986 tav.66a. Drawings: Thulin 1907

[---?]merc[ui]
Sinistroverse, Faliscan alphabet. There is a small space between the c and the edge of
the lacuna (merc Herbig and Vetter), but in view of the standardized form of these
inscriptions, merc[ui or [titoi]merc[ui is not impossible. The space under the line is
empty: it is impossible to read [titoi] | merc[ui | efil]es (Thulin, adding 117).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 12555). I saw only the second
shard, with the letters erc. Drawings: Thulin 1907 between pp.298-9 ¢39² (reproduced in CIE 8046);
Comella 1986 tav.77,R70.

[m]ẹrcui [
Sinistroverse, Faliscan alphabet. Of the e, only the top right corner remains.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12556). Photograph: Comella 1986 tav.67d.
461
CHAPTER 14
125. Scratched (not painted) on a fragment of a black-varnished cup.

]ṃercui [
Sinistroverse, Faliscan alphabet. Of the m, only part of the left shaft remains.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12561). Photograph: Comella 1986 tav.69c.
Drawings: Thulin 1907:299 (reproduced in CIE 8048).
126. Scratched on the handle of a vessel.

meṛ[cui]
Sinistroverse, Faliscan alphabet. On the edge of the lacuna a vertical shaft is just visible,
and meṛ[cui] is therefore preferable to Herbig’s me[rcui (which has been adopted by all
later editors).
From autopsy in the Museo di Villa Giulia, Rome (inv. 12566). Photograph: Comella 1986 tav.69b.
Drawings: Thulin 1907:299 (reproduced in CIE 8049).
Thulin proposed to take titoi mercui as a dative of a deity. Efiles, at first regarded as the
name of the dedicant (thus still Buonamici) was equated by Erman with Latin aedilis
(§6.2.1), and the interpretation, adopted by most authors, has since become ‘the aediles,
to Titus Mercus’. The word efiles has been regarded as a calque on Latin aediles by G.
Giacomelli and later authors, but I doubt if this is necessary. If the functions of the
Faliscan aediles were similar to those of their Roman counterparts, Vetter and Combet
Farnoux (1980:137-142) may be right in taking the inscribed vessels as dedications by
trespassers of aedile regulations or as by the aediles themselves aere multatico. This
would explain the connection with Mercus or Mercury, the fact that the names of the
aediles are not mentioned, and that the majority of the vessels were apparently produced
as a series of ready-made dedications. The interpretation is not without problems,
however. Although mercui can be the name of a god connected, like Mercurius, to
Latin merx etc. and to Oscan mirikui Cm 12, amirikum Cm 13, and amiricatud TB 22,
the human praenomen of the god is surprising.245 Most editors therefore tried to infer
the existence of a god Titus from the sodales Titii (and the aues titiae from which Varro
(L 5.85) derived their name), or connected Titus with Mutinus Titinus. Solmsen took
Titos Mercus as the Italic equivalent of ithyphallic Hermes. A discussion of the various
theories is given by Combet Farnoux, who himself plausibly posits an adjective */tito-/
‘propitious, prosperous’, which could be used both as a human praenomen and as an
epithet of Mercury. G. Giacomelli suggested that the dedications were made by the
aediles on behalf of a member of their college who was called Titus Mercu.
A second problem is that interpreting mercui as a fourth-declension dative
involves quite large morphophonological problems, for which see §4.6.2. It is especially
this second problem that has lead to the proposal of alternative interpretations.
245
The Aius in Aius Locutius adduced by Vetter is not convincing as a human name. The
evidence for a Titus Sanquus (Radke 1965b:215, 280) is dubious (Prosdocimi 1965:586).
462
Herbig, who doubted Thulin’s interpretation already in the CIE, took titoi mercui
as an Etruscan feminine ‘Tita Merconia’, with efiles first (1914) as the genitive of the
husband’s name, and later (1923), following Erman, as ‘aedilis’. This was not adopted
afterwards, especially on account of the occurrence of a female aedile. Others interpret
mercui as a form of a Mercuuius. As there are no Faliscan attestations of /u/ being
written as u, or of a drop of intervocalic //, this is difficult, although not impossible.
Jacobsohn interpreted titoi mercui as an abbreviated nominative, but neither ‘Titoius
Mercuvius. Epillius’ nor ‘Titoius. Mercuvius Epillius’ (name of the god + name of the
dedicant, both in the nominative) is convincing.
Mercui has also been taken as a genitive of *Mercuuius. In that case, if titoi is a
dative, it can only be a common noun, as Altheim interpreted it (‘to the *titos of Mercu-
vius’): Jacobsohn’s ‘to Titus of Mercuuius’ is probably impossible, as such a syntagma
would require an adjective, not a genitive (cf. e.g. names like Picus Martius). Although
Altheim’s interpretation of *titos as ‘phallus’ was adopted by Koch, Kroll, Vahlert, and
Andrén, his further expansion of the meaning to ‘Genius’ has been doubted by many
and has been rejected at some length by Combet Farnoux (1980:118-133). New
evidence in support of his interpretation seems to be provided by the use of */tito-/ in
the South Picene inscriptions, titú|i AP.1, titúí TE.5 (also titiúí CH.2, and perhaps
titienom TE.3?). If titos or titios could have a meaning like ‘genius’, this need not
necessarily be derived from an original meaning ‘phallus’, however: Combet Farnoux’s
*/tito-/ ‘propitious, prosperous’ might well constitute a better basis (cf. the Latin (di)
manes : manis or manus = bonus).
Titoi and mercui have also both been taken as genitives. The first to do so was
Jacobsohn, who interpreted ‘Titoii Mercu(v)i’ as the name of the god in the genitive
(followed by the dedicant in the nominative, which, as he himself admits, is awkward).
Pisani interpreted also efile(s) as a genitive, interpreting ‘Titi Mercuvi aedilis’. Al-
though I agree that a genitive titoi for the o-stem /tito-/ is not impossible (§4.4.4), the
side-by-side occurrence of an ending -oi for the o-stem genitive titoi and an ending -i
for the io-stem genitive mercui is implausible. (It would be even more remarkable in
Pisani’s view, where -oi and -i are different stages of a phonological development of
one and the same morpheme.).
In my view, none of these interpretations offers an attractive alternative for taking
mercui as a dative, problematic though the morphology may be. Efile(s) is then
probably a nominative plural rather than singular; an indication for this is the fact that
the omission of word-final -s, which is almost universal after short vowels but virtually
absent after long vowels, is attested here only once, in 114, which probably contains an
error also in tito.
Bibliography: Thulin 1907:296-303 ¢36-48² (autopsy); Nogara 1907:156 n.1; Bartholomae 1910:9 n.1;
Jacobsohn 1910:3 ¢4-5²; Jacobsohn 1911:464-5; Mengarelli 1911 (autopsy); Herbig CIE 8036-49
(autopsy); Solmsen 1912:9-10; Buonamici 1913:55-6 ¢7²; Herbig 1914a:240-6; Erman 1918; Della Seta
463
CHAPTER 14
1918:176-7 (autopsy); Herbig 1923:234; Stolte 1926:51, 53; Altheim 1930:44-71; Kroll 1931:975-82;
Pisani 1933:624 n.1; Vahlert 1933:982; Koch 1934:429-30; Andrén 1940:104-5; Lejeune 1952b:125;
Vetter 1953:292-3 ¢264a-m, 265a-b²; Pisani 1955:322; Camporeale 1956:#; Campanile 1961:7; G.
Giacomelli 1963:54-5 ¢15,I-XII, 16,I-II² (autopsy); Pisani 1964:343 ¢146F,a-c²; Radke 1965b:213;
Dohrn in Helbig/Speier 1969:726-7 ¢2820² (autopsy); Devine 1970:17-8; G. Giacomelli 1978:530
¢6,I-XII²; G. Giacomelli 1978:75-8 ¢4,I-XII²; Combet Farnoux 1980:113-69; Morandi 1982:59 ¢12²;
Moscati Sant p.113; Comella 1986:165-9 ¢1-14² (autopsy).
127-131. The following inscribed vessels were also found in the temple precinct. Some
may also have belonged to the stips, as was suggested by Thulin (1907:303).
127. Scratched on the bottom of a black-varnished cup (letters 5 mm high).
] sacra [
Sinistroverse, Faliscan alphabet. G. Giacomelli suggested that the name of the god may
have stood in the lacuna, but sacra is followed by a space of at least one letter. (In
Thulin’s drawing the space is smaller than it is in reality.) In a text like this, placed in a
temple precinct, it would of course be clear which god was meant, and its name could
therefore be omitted.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12562). Bibliography: Thulin 1907:303 ¢49²
(autopsy); Herbig CIE 8050 (autopsy); Vetter 1953:293 ¢266a²; G. Giacomelli 1963:55 ¢17²; Comella
1986:169 ¢15² (autopsy). Photograph: Comella 1986 tav.66b. Drawings: Thulin 1907:303 (reproduced
in CIE 8050); Comella 1986 tav.77,R68.
128. Scratched inside a black-varnished cup (letters 17-19 mm high) is

] sta[?---]
Sinistroverse, Faliscan alphabet. Thulin interpreted this inscription, together with sta
MF 28 and statuo MF 29 as instances of stare used in a sacral sense, ‘to stand (as a
dedication)’. This interpretation was adopted by Vetter, who compared Volscian statom
VM 2, and by G. Giacomelli. The only valid parallel for Thulin’s interpretation is
Umbrian sacre  stahu Um 10, however, where the ‘sacral sense’ depends not on stahu,
but on sacre. Sta is rather the abbreviation of Statius (cf. Vetter on MF 28-29).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 12565). Bibliography: Thulin
1907:304 ¢52² (autopsy); Herbig CIE 8052; Vetter 1953:293 ¢266c²; G. Giacomelli 1963:56 ¢19²;
Comella 1986:196 ¢17² (autopsy). Drawings: Herbig CIE 8052; Comella 1986 tav.78,R83.
129. Scratched inside a black-varnished cup (letters 5 mm high).

[---]*[5-7] cua [
Dextroverse, but the spelling cu rather than qu points to the Faliscan rather than the
Latin alphabet. The traces consist of the lower halves of three shafts placed close
together, perhaps an m. Cua is apparently an abbreviation, perhaps, as Thulin, thought,
of a name like Quaelius or Quartus. The latter possibility is considered also by Herbig.
G. Giacomelli rejected these proposals, but apparently for no other reason than that she
thought that they were proposed, not as solutions of an abbreviation, but as restorations
464
(which is certainly impossible). It should be noted that names derived from the numer-
als 1-4 were very rare at this time, and that the letters cua do not stand in the position
where a praenomen might be expected.
(autopsy); Herbig CIE 8053 (autopsy); Vetter 1953:293 ¢267²; G. Giacomelli 1963:56 ¢20a-b² (au-
topsy); Comella 1986:169-70 ¢18² (autopsy). Photograph: Comella 1986 tav.69a. Drawings: Thulin
1907:305 (reproduced in CIE 8053); Comella 1986 tav.79,R104.
130. Scratched inside a black-varnished cup (letters 8-10 mm high).

] poe[?---]
Sinistroverse, Faliscan alphabet? O and e are written as a ligature . A similar ligature
oe has been read in LtF/Lat 171. The form is apparently an abbreviation of a name.
From autopsy in the Museo di Villa Giulia, Rome (inv. 12563). Bibliography: Herbig CIE 8054
(autopsy); Vetter 1953:293 ¢268²; G. Giacomelli 1963:56 ¢21² (autopsy); Comella 1986:170 ¢19²
(autopsy). Photograph: Comella 1986 tav.69d. Drawing: Herbig CIE 8054.
131. Painted on the bottom of a black-varnished cup (letters 18-25 mm high).

] sa [
Dextroverse. The dextroverse ductus may indicate that the inscription belongs to the
dedications from the period after 241. The form is clearly an abbreviation, either of
sa(cra) (cf. sacra MF 128), or of the name of a dedicant.
(autopsy); Herbig CIE 8051 (autopsy); Vetter 1953:293 ¢266b²; G. Giacomelli 1963:55 ¢18² (autopsy);
Comella 1986:169 ¢16² (autopsy). Photograph: Comella 1986 tav.68c; Drawings: Thulin 1907:303
(reproduced in CIE 8051); Comella 1986 tav.79,R95.
14.1.4. The temple in Contrada Celle. The temple of Contrada Celle, the ruins of
which are visible even today, may have been the famous Faliscan temple of Juno
described by Ovid (Am. 3.13, where he describes a visit to the deity’s festival and the
procession): see Le Bonniec 1980. For literature on the temple complex and its
excavations, see the literature named in the bibliography to 132.
132. From the cella of the temple of Celle are three shards246 of a terracotta decorative
revetment, perhaps a frieze circling the cella. The shards show a boy’s or a woman's
head, above which, in a bare strip, the inscription is painted in red paint.
[---]l*[---]
Sinistroverse. Only the lower half of the l remains. It is impossible to make any
conjecture as to the contents of the inscription; it may have been the name of the figure,
or it may have been part of a much larger (dedicatory?) inscription.
246
Andrén’s drawing contains a fourth fragment showing more of the hair.
465
CHAPTER 14
From autopsy in the Museo di Villa Giulia, Rome (inv. 3790). Bibliography: [Pasqui 1887:95-6
(autopsy)]; [Weege in Helbig 1913:346 ¢1785,k² (autopsy)]; [Della Seta 1918:205 (autopsy)]; [Andrén
1940:91 ¢II.8² (autopsy)]; [(Sprenger &) Bartoloni 1977:72]; [Corsini in PrItal pp.193-4 ¢123² (au-
topsy)]; Fortunati in Sant. p.112 (autopsy); [De Lucia Brolli 1991a:37]. Photograph: PrItal p.193
fig.123 = Sant p.112 fig.5.2B,1 = De Lucia Brolli 1991a:38 fig.26. Drawing: Andrén 1940:91 fig.24.
14.1.5. The temple of Lo Scasato. The temple ruins of Lo Scasato may date to the Late
Faliscan period: unfortunately, they have yielded only a few inscriptions.
133. Scratched on one of the sides of a 10 cm high pyramid-shaped loomweight

ca
Sinistroverse, but apparently with reversed a ().
Bibliography: Herbig CIE 8564. Drawing: Herbig CIE 8564.
134. Scratched on an “oggetto a forma cilindrica di tubo, di uso incerto, con molti fori
nelle pareti, forse per sostegno di vasi mentre si dovevano cuocere” (Nogara in Herbig
CIE 8565).
se
Sinistroverse, with reversed s.
Bibliography: Herbig CIE 8565. Drawing: Herbig CIE 8565.
14.2. Civita Castellana, origin unknown
A great number of inscriptions has been ascribed to Civita Castellana without further
specification or identification of their site of origin. The majority of these inscriptions
will probably have been found at or near Civita Castellana, either as chance finds or in
clandestine or badly documented (semi)private excavations, but the possibility that
inscriptions from other locations were later erroneously ascribed to Civita Castellana as
the central site of the area cannot be excluded (cf. §1.4.5). This is of some importance
for the dating of the material, for the criterion on which the inscriptions from Civita
Castellana are dated to the Middle Faliscan period, namely the fact that this site was
(largely) abandoned after 241, does not hold for other sites. Tiles from other locations
(and perhaps from Civita Castellana itself, cf. §14.1) may therefore date from the Late
rather than the Middle Faliscan period. Candidates for this are e.g. the inscriptions in
Latin alphabet (LtF 171-174).
The inscriptions that are ascribed to Civita Castellana but not to any specific
location within or around the town fall into three groups. The most numerous are the
sepulchral inscriptions on tiles, subdivided into: (1) tiles on which a gentilicium can be
identified (§14.2.1, MF 135-139 and 141-157, and LtF 140), (2) tiles containing
466
comprehensible parts of text but without identifiable gentilicia (§14.2.2, MF 158-170),

with a subgroup of (3) tiles written in the Latin alphabet (§14.2.3, LtF 171-174) and
(4) the fragmentary texts (§14.2.4, MF 175-194).247 They are followed by other two
groups, the detached loculus-inscriptions (§14.2.5, MF 195-198) and the inscriptions on
pottery (§14.2.6, MF/Etr 199 and MF 200-201, MF? 202-204).
14.2.1. Inscriptions on tiles containing identifiable gentilicia. The following tiles

contain more or less identifiable gentilicia.
135 (Cincius). Colonna mentions an unpublished tile from Civita Castellana with the
text cincia. As far as I have been able to ascertain, this tile has remained unpublished to
this day: neither can I say if cincia is the whole text, or only a part.
Bibliography: Colonna 1972a:446; Renzetti Marra 1974:351.
136-139 (Fa(r)far-). Two inscriptions containing a gentilicium in Fafar-, perhaps

connected to the potamonym Farfarus, see §6.5.1.
136-137. The titulus prior is painted in red across the front of a damaged tile (max.
32u48 cm; letters 10-11 cm high).
poplia
fafarṇ
Sinistroverse, Faliscan alphabet. The interpuncts before poplia read by Herbig and Pauli
jr. (in Herbig CIE 8237a-b) but not by Thulin (in Herbig CIE 8237a-b) are invisible and
may well never have existed. The second line ends in , with the left shaft written
against the inside of the tile’s flange. This is read as iu written upside down in boustro-
phedon by Herbig (fafarịụ = fafari u(xor) or fafarịụ(s)), Lejeune (poplia fafariu, with an
Etruscan ending), Vetter (...:poplia/iu/fafar ‘Publia, Iu(na), Farfarus’?), and G. Gia-
comelli. None of their interpretations is really attractive, however, and there seems to be
no objection against taking it as n ( ), with the left shaft detached as it had to be painted
‘around the corner’. This would give a gentilicium Fa(r)farn..., either comparable to
Etruscan gentilicia of the type Perperna or connected with the potamonym Farfarus
(§6.5.1). The inscription may have continued in a third line on the missing part of the
tile.
247
It could be argued that these should have been included among the inscriptions from the
necropoles in chapter 13, but as there are no clear data on their provenance, there is a possibility
is that they are in fact from other locations: MLF 347-355, for instance, ascribed by Herbig (CIE
8196-8204) to the Valsiarosa necropolis at Civita Castellana, have since been shown to belong
to the site at Grotta Porciosa.
467
CHAPTER 14
The titulus posterior is painted on plaster across the back of the tile (letters 12-14 cm
high).
[---]lio*[?---]
[---]so*[---]
Sinistroverse, Faliscan alphabet. The contents of the first line, now badly legible, may
consist of the end of a man’s name. At the end of the second line there appear to be
some traces (possibly a u or an m?).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8208). Bibliography: Herbig CIE
8237a-b (autopsy); Lejeune 1952b:120 n.1; Vetter 1953:304 ¢293²; G. Giacomelli 1963:86 ¢94a-b².
Drawings: Herbig CIE 8237; Thulin in CIE 8237 (mirrored).
138-139. Known only from apographs by Pauli jr. The titulus prior is painted in red
across the front of a tile fragment.
ucro[---]
**[---]
Sinistroverse, Faliscan alphabet. The text starts at the edge of the tile, and a tile may be
missing in front. The second line contains only three shafts (i.ụ --- Herbig). Ucro is
regarded as a gentilicium by G. Giacomelli and Calzecchi-Onesti.
The titulus posterior is painted in black across the back of the tile fragment.
f ̣af ̣[---]
Sinistroverse, Faliscan alphabet. The f’s are both damaged at the top and could be read
as t’s. F̣af ̣[---] could be the beginning of the same gentilicium as in 137 (Herbig).
Bibliography: Herbig CIE 8281a-b; Vetter 1953:307 ¢311²; G. Giacomelli 1963:92 ¢115,I²; Calzecchi-
Onesti 1981:181. Drawing: Pauli jr. in CIE 8281a-b.
140 (Folcosius). Painted in black across the front of a damaged tile (47u27-30 cm,
letters 6-9 cm high). The letters are so slender that they give the impression of having
been written with a reed-pen and look similar to those of MLF 337, from the tomb of
the gens Folcosia at Carbognano-Vallerano (whence MLF 329-337).
cẹịṣ[i.]
holc[osi]
arp[...]
Sinistroverse, although the alphabet appears to be Latin, with  for h, and cursive a ()
and e (). The first line is , probably cẹ (), followed by vague traces of vertical
lines. The last letter of the third line is  (arp ̣[ineo Herbig, disregarding the interpunct),
an  (ar.f ̣? Vetter) is impossible. The inscription apparently occupied only one tile, in
which case there is not enough room to restore holc[osio] or holc[osia]: the gentilicium
may have been abbreviated to holc[osi], or the last letter may have been written
underneath the line in what is now the missing part of the third line.
468
8256 (autopsy); Vetter 1953:306 ¢309²; G. Giacomelli 1963:91 ¢112². Drawing: Herbig CIE 8256.
141-143 (Graecius?). The following tiles appear to contain the same gentilicium. The
variation between crạ[---] 141 and cre[---] 142 would point to a name with an original
diphthong /ai/ (crạ[i---]) o /ē/ (cre[---]), cf. §3.7.6. In that case, Thulin’s Graecius is
not impossible (cf. the praenomen kreco MF 147 and lo : cr MF 33): see §7.8.1.71 for
this name.
141. Painted in red on plaster on a tile fragment (37u30 cm, letters 10-11 cm high).
[---]crạ[i---]
[iu?]nẹo
Sinistroverse, Faliscan alphabet. All that is left of the last letter of the first line is the
shaft, which could be read as a or e. What remains of the penultimate letter of the
second line are the upper halves of two shafts standing close together, probably a
cursive e (---eo Herbig) rather than an i (...io Thulin).
1907:282 ¢17² (autopsy); Herbig CIE 8260 (autopsy); G. Giacomelli 1963:91-2 ¢113,IV². Draw-
ing: Herbig CIE 8260.
142. Painted in red along the length of the back of a damaged tile (max. 45u45 cm,
letters 8-10 cm high).
[---]iocre[---]
Sinistroverse, Faliscan alphabet. Herbig read cr---, but Thulin’s cre.. is beyond all
doubt: the e is clearly visible.
1907:283 ¢19² (autopsy); Herbig CIE 8261 (autopsy); G. Giacomelli 1963:91-2 ¢113,V². Draw-
143. Painted in red on white plaster along the length of the back of a tile fragment (max.
34u37 cm, letters c.10 cm high).248
[---]ocṛ[---]
[---]leoc[---]
Sinistroverse, Faliscan alphabet. Of the r, only the bottom part of the shaft is visible. Of
the first letter of the second line, a small oblique stroke remains that can only be the
sidestroke of an l. The letters ]leo are probably part of a patronym; c[---] may have
contained the name of a second person.
8257 (autopsy); G. Giacomelli 1963:91-2 ¢113,I². Drawing: Nogara in CIE 8257.
248
Herbig erroneously describes the letters as painted on the front of the tile.
469
CHAPTER 14
144-145 (Lepuius?). The titulus prior is painted in red across the front (letters 7-8 cm
high) of a fragmentary tile (max. 30u44 cm).
Fig. 14.1. Various drawings of the first line of MF 144.

left: Thulin’s drawing. (From Thulin 1907:287.)
middle: Nogara’s drawing. (From Herbig CIE 8243a.)
right: Author’s drawing (tracing from author’s slides).
[.?]ạ*ịạ
lepuia
uoltilia
Sinistroverse, Faliscan alphabet. The main group of shards contains only the lowest
parts of the letters of the first line, before which one letter may still be missing. The first
two traces are almost certainly an a: the bottom part of the left-hand shaft is slightly
curved, exactly like the a at the end of the second line (which is quite different from
Thulin’s drawing). The next three traces were restored by Thulin as ta from a small
fragment containing the upper halves of these letters; the last trace he read, with great
hesitation, as i, interpreting ‘Lepuia Voltius’ daughter (made this grave) for Ianta’. This
was adopted by Vetter (iạṭaị) and G. Giacomelli (ịạtai). Nogara (in Herbig CIE
8243a-b) noted that Thulin’s fragment did not join onto the others and suspected that it
did not belong to this inscription at all.249 The last two letters, however, are certainly not
ai. The very last trace shows the same slight curvature that characterizes the a, and
together with the trace preceding it forms the bottom part of an a. The shaft preceding
this a, which stands quite close to it, can be an i or a t. Of the preceding letter, only the
bottom half of a shaft is preserved, separated from the traces that surround it by rather
more space than is suggested by the drawings by Thulin and Nogara. Possible readings
are at least f, i, p, and t, perhaps also n. None of these letters seems to make sense,
however. Unfortunately, I was unable to find Thulin’s fragment myself. I have tried to
fit the fragment as it appears in the drawings in various positions above the first line, to
see if it would fit with the last traces (giving [?]ạ*ta) or with the first (giving ta*ịa, cf.
249
Herbig published this fragment again separately as CIE 8280, reading it as ---ṭẹ---.
470
θania MF 81), but neither appears to be possible. The text probably consisted of a (now
illegible) praenomen, followed by a gentilicium lepuia and a patronym uoltilia. The
possibility that lepuia is to be read as le, an abbreviation of a man’s praenomen in the
genitive, followed by Etruscan puia ‘wife’ was rejected already by Herbig, but revived
by Vetter: see §9.2.2.3. Not only would it be a case of Etruscan interference within a
formulaic phrase HUSBANDGEN WIFE where uxor was the standard word (§7.4.2), it
would also be the only case where this formula precedes FILIATION (apart from the
exceptional and easily explained case cauia  uxo  a  f LF 242).
The titulus posterior is painted in red on plaster across the back (letters c.13 cm high) of
the tile fragments.
[u]ọlt[---]
Sinistroverse, Faliscan alphabet. Only a very small part of the o is visible. Possible
restorations are u]ọlt[a, or any derivation of this name like [u]ọlt[io, [u]ọlt[ilio, etc.
1907:287 ¢25² (autopsy); Herbig CIE 8243a (part of which=CIE 8280), 8243b; Vetter 1953:305
¢299A-B²; G. Giacomelli 1963:87-8 ¢100a-b². Drawing: Thulin 1907:287 (reproduced in CIE
8243a-b); Nogara in CIE 8243a; Herbig CIE 8243b and 8280.
146-148 (Laeuius, Laeuilius). Three inscriptions contain gentilicia that can connected
either to leuia LtF 328 and leuieis Lat 251 (probably Laeuius, §7.8.1.81) or to leueli MF
14 (probably Laeuilius, §87.8.1.82).
146. Painted in red on plaster across the backs of three tiles: the first 30u33 cm
(apparently part of a tile of a different size, but without doubt belonging to this inscrip-
tion), the second 46u60 cm, the third 43u43 cm (“Von dem dritten ist nur ein jetzt in
Kalk eingefasstes Fragment des Bewurfs erhalten”, Thulin 1907:283). The second tile
seems to contain traces of an earlier inscription.
1
ca[u]2io:le[u3eli]o:cau[i]
1
hil2eo:ian3[ta:..]lni[a]
1
hec:2cupat3[:....]ta[
Sinistroverse, Faliscan alphabet. The e has the rare form (§11.2.4.2, occurring also in
Lat 482†), which can still be read clearly and is beyond any doubt; h has the likewise
curious form . Herbig restored le[ui]o in the first line, but as this would leave no
room for the restoration of a gentilicium in the second, Thulin’s le[ueli]o has been
adopted by all later editors. In view of the size of the letters this might just (but only
just) be possible without assuming that a tile was missing between the second and the
third: as it is, there is hardly any space left for the woman’s gentilicium or patronym.
(Thulin proposed ci]lni[a, and that is about as much as could be fitted into the lacuna.)
471
CHAPTER 14
The third line may have contained the name of a third woman (e.g. a daughter
ian]ta or a son uol]ta?). As the inscription gives every sign of being written all at one
time, cupat must be read as a plural cupa(n)t.
1907:283-4 ¢21² (autopsy); Herbig CIE 8240 (autopsy); Buonamici 1913:62-3 ¢20²; Vetter 1953:304
¢296²; G. Giacomelli 1963:87 ¢97²; Pisani 1964:340 n.1; Peruzzi 1965:276-8. Drawing: Thulin
1907:283 (reproduced in CIE 8240).
147. Painted in red along the length of the back of a tile (69u47 cm, letters 13-15 cm
high).
kreco[---]
iataleue[lia]
Sinistroverse, Faliscan alphabet. The first letter is , which was read as k by Thulin and
by Nogara (in Herbig CIE 8241), but Herbig and others after him, distrusting a k (not
before a) in a Middle or Late Faliscan inscription, regarded it as a c corrected from a
shaft. However, the Late Faliscan inscriptions from Pratoro (LF 242, 243, and 246) now
show that k could be used to represent /g/, which is exactly the way in which it is used
here: see §11.2.4.2. G. Giacomelli regarded creco as a cognomen and assumed that a
tile was missing before the text, but Graecus can very well be a praenomen, as Peruzzi
suggested: see §7.1.1.26. The second line probably contained the gentilicium leue[lia
(§7.8.1.82) although it is also possible to read leue[lif ‘Laeuili f.’ with the praenomen
Laeuilius (§7.7.1.33) or perhaps even leue[a with the gentilicium Laeuius (§7.8.1.32).
Peruzzi (1965:276-8) regarded iata  leue[lia as a daughter of ca[u]io  leu[el]io and
ian[ta  ..]lnia of MF 146, in which case iata  leue[lia would then apparently have been
named after her mother.
Bibliography: Thulin 1907:282 ¢16² (autopsy); Herbig CIE 8241 (autopsy); Vetter 1953:304-5 ¢297²;
G. Giacomelli 1963:87 ¢98²; Peruzzi 1965:276-8. Drawing: Thulin 1907:282 (reproduced in CIE
8241); Herbig CIE 8241.
148. Painted in red along the length of the back of a damaged tile (65umax. 35 cm,
i[un]alẹ[---]
mesio
Sinistroverse, Faliscan alphabet. Thulin read lạ[---] in the first line and ṃạsio in the
second, but the original supports rather the lẹ[---] and ṃẹsio read by Nogara (in Herbig
CIE 8249) and Herbig. The gentilicium may have been either lẹ[uio or lẹ[uelio (cf.
§7.8.1.81-82); mesio is either a patronym or a second gentilicium (§7.7.1.43).
1907:291-2 ¢30² (autopsy); Herbig CIE 8249 (autopsy); Vetter 1953:306 ¢305²; G. Giacomelli 1963:89
¢106². Drawing: Thulin 1907:292 (reproduced in CIE 8249); Nogara in CIE 8249.
472
149-151 (Pupellius or Pupilius). The following tiles contain a gentilicium Pupellius or

Pupilius: see §7.8.1.129).
149. Painted in red on two or three fragmentary tiles (38 u c.115 cm, letters 8-9 cm
high).
1
[u]ol2ṭapupelio
1
[m]an2o[m]o
Sinistroverse, Faliscan alphabet. The fragment containing ol cannot be joined directly
onto the others, but clearly belongs to the beginning of the first line where it was placed
by Herbig. The t is nearly complete: almost the entire shaft is visible. An apparently
unpublished fragment, numbered 8184 like the others, contains , i.e. an a followed
by part of a shaft. Although it cannot be joined onto the other shards, both the colour
and structure of the tile and the lettering make clear that it belongs to this inscription. If
placed at the beginning of the second line, the cognomen is [m]ano[m]o, not
[max]o[m]o as it is usually restored.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8184). The fragment with lio
could not be found. Bibliography: Thulin 1907:285 ¢22² (autopsy); Herbig CIE 8232 (autopsy); Vetter
1953:303 ¢289²; G. Giacomelli 1963:85 ¢89². Drawing: Thulin 1907:285 (reproduced in CIE 8232).
150. Painted in dark red on white plaster across the back of a damaged tile (40u48 cm,
letters c.11 cm high).
[1]
[---p2u]peḷ[3][i---]
[1]
[---2.]ahe[c[3]upa(t)]
Sinistroverse, Faliscan alphabet. Of the l, only a very small trace of the lower half is
preserved. The first p of pu]pel[io must have stood on a tile that is now missing. The
rest of this tile probably contained an (abbreviated) praenomen in the first line. At the
end of the text, another tile is missing, which contained the last letters of he[ cupa(t).
The inscription can be read as pertaining either to a woman only, with the first line
ending in the abbreviation of a father’s praenomen followed in the second line by file]a,
or to a man and a woman, in which case the first line ended in the man’s affiliation and
the beginning of the second line contained the name of the woman (---]a).
1907:285 ¢23² (autopsy); Herbig CIE 8233 (autopsy); Vetter 1953:303 ¢290²; G. Giacomelli 1963:85
¢90². Drawing: Thulin 1907:285 (reproduced in CIE 8233).
151. Painted in red on the back of a fragmentary tile (tot. 63u36 cm, letters 11-12 cm
high). Thulin mentioned six fragments that he assumed belonged to a second tile of the
inscription: three of these can in fact be joined to the end of the first line.
473
CHAPTER 14
Fig. 14.2. Thulin’s and my own drawings of MF 151.

left: Thulin’s drawing. (From Thulin 1907:291.)
right: Author’s drawing from author’s slides and drawings.
1
tulopup ̣[2][elio ?---]
1
ịuneo
Sinistroverse, Faliscan alphabet. Tulo is followed by p, as was proposed by Herbig, not
by Thulin’s ti: what Thulin read as a t is in fact an interpunct with slight drip-mark. The
next letter is probably a large u: a trace of its top right corner is visible on the fragment
that also contains the top of the p. This is followed by a letter of which only the middle
part of a shaft is visible, perhaps p (cf. pupelio MF 149) or t (cf. putellio MF 152). After
that, the text was continued on another tile (assuming the tile has the standard length of
c.68 cm). Fragment 8227 may have belonged to this second tile, as Thulin proposed;
fragment 8231 (ai*), however, probably belonged to another inscription: as it consists
of the upper right-hand corner of a tile, it could be placed only in the first line, but
neither pup ̣ai*--- nor puṭai*--- gives a plausible text.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8200+8202+8224+three frag-
ments without numbers (+8227+8231?)). The fragment containing the letters un could not be found.
Bibliography: Thulin 1907:291 ¢29a-b² (autopsy); Herbig CIE 8250 (autopsy); Vetter 1953:306 ¢306²;
G. Giacomelli 1963:89-90 ¢107a-b². Drawing: Thulin 1907:291 (reproduced in CIE 8250).
152 (Putellius?). Painted in red along the length of the front of a fragmentary tile (tot.
73u max. 42 cm, letters 7-8 cm high).
1
uolti[o]marc[2][---]
1
putellio [2][
Sinistroverse, Faliscan alphabet. Marc is written against the edge of the tile: pace
Vetter, it belongs to the same line as uolti[o]. The text was probably continued on an
second tile: G. Giacomelli’s marc[i is impossible if the text is assumed to have
occupied one tile only. In the second line, Thulin and Herbig read puịellio, but Vetter’s
putellio is certain: part of the sidebar of the t is visible. Putellio provides a rare instance
of geminated spelling (§11.2.4.3, §3.5.5.3). Vetter interpreted putellio as ‘infans’,
related to Oscan puklum Cp 37 and Paelignian puclois Pg 5. Although adopted by G.
474
Giacomelli (‘filiolus’) and considered probable by Rix, I hesitate to adopt it: see
§6.2.62. Vetter’s idea that the main text began with marc, with uolti[o] | puṭellio as
marginal addition is implausible: the text start at the edge of the tile, not half-way down
its surface. Uoltio is rather a praenomen followed by a gentilicium marc[--- (for which
cf. §7.8.1.97-98.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8182+8187+8188+8189
+8991+8201+one fragment without number). Bibliography: Thulin 1907:286 ¢24² (autopsy); Herbig
CIE 8235 (autopsy); Vetter 1953:303-4 ¢292²; G. Giacomelli 1963:85 ¢92²; Rix 1964:448. Draw-
ings: Thulin 1907:286 (reproduced in CIE 8235); Nogara in CIE 8235.
153-154 (Sacconius). Two tiles containing a gentilicium Sacconius.

153. Painted in red on plaster across the back of a damaged tile (49u40 cm, letters 10-12
cm high).
[1]
[----]2cais[io]
[1]
[----]2zaconiọ
The first line was read as c ̣aṛ[co] by Herbig, but the original has , i.e. is, not r. The n
is written under the second line, the i vertically under the n, and the o to the right and
lower than the i, indicating that this was the last tile of the inscription. The interpuncts
imply that one or more tiles are missing at the beginning, although the text appears to be
complete: perhaps it was an addition to an already existing inscription, like MF 41.
1907:294 ¢32² (autopsy); Herbig CIE 8253 (autopsy); Vetter 1953:306 ¢308²; G. Giacomelli 1963:90
¢110². Drawings: Thulin 1907:294 (reproduced in CIE 8253); Herbig CIE 8253.
154. Painted in red on plaster along the length of the back of a tile (56u48 cm, letters 10
cm high).
[1]
[---]2zaconiai
Sinistroverse, Faliscan alphabet. The line slopes downwards, and the final i is placed
slightly lower down than the other letters. As Bonfante noted from a photograph by
Torelli, the word is preceded by an interpunct that does not appear in Thulin’s drawing.
The interpuncts and the case, either a dative or a genitive (§8.10.2), make it probable
that words are missing in front on other tiles.
8252 (autopsy); Vetter 1953:306 ¢307²; G. Giacomelli 1963:90 ¢109²; Bonfante 1966:4. Draw-
ing: Thulin in CIE 8252.
155 (Tirrius). Painted in red on plaster along the length of the back of a tile (70 u 48
cm, let. 15-18 cm).
1
ṭị[]ṭịrialo[2][?---]
1
l[e]acsf [2][
475
CHAPTER 14
Sinistroverse, Faliscan alphabet. The reading of especially the beginning of both lines is
very unclear: Herbig’s drawing in particular shows little more than vague vertical traces
there. Nevertheless, Thulin read the text as tḥeịriạlo|lẹa:cs:f and Herbig as
ṭị[.]ṭịriạlo|l*(-)ạ:cs:f. The latter isolated lo as lo(ferta), as in loferta LF 221: this is
rejected by Vetter and G. Giacomelli, as the freedwoman would then be named before
her mistress. It is impossible, however, not to read lo as an independent word unless one
either reads lo|l[e]a, assuming that the inscription pertained to two women, or assumes
that a second tile is missing at the end. The latter possibility appears to be contradicted
by Herbig’s drawing, where the end of the first line bends downwards slightly, indicat-
ing that this was the last tile of the inscription.
Bibliography: Herbig CIE 8248 (autopsy); Vetter 1953:306 ¢304²; G. Giacomelli 1963:89 ¢105².
Drawing: Thulin in CIE 8248; Herbig CIE 8248.
156-157 (-ronius). Painted in red on the front of a damaged tile (59u45 cm).
[1]
[---]2roniouol3[t---]
[1]
[---]2a*ome 3[
Sinistroverse, Faliscan alphabet. A tile is certainly lost at the beginning, containing the
praenomen and part of the gentilicium; if uol is not an abbreviation (as Colonna
assumed), another tile is missing at the end, which contained the remains of the
patronym (uol[tio or uol[tilio). The second letter of the second line is , either an x or a
t (]a*ome Renzetti Marra). Colonna reads it as a dative m]axome, interpreting ‘...ronius
son of Volt... (made this grave) for ... Maxuma’. There are no certain Faliscan instances
of monophthongization in the ending of the dative (§3.7.6.), nor of a woman with a
cognomen (§7.9): Colonna ascribed these oddities to Latin influence. Another objection
is the fact that his photograph shows a clear difference in colour between both lines,
implying that they were not written at the same time. Perhaps me is an abbreviation of
Maesius (cf. mesio MF 148): the space before the m seems to be wider than those
between the other letters of the second line. In that case, however, it would seem to be
preceded by a man’s name [---]axo or [---]ato.
The titulus posterior is painted in red across the back of the tile.
iuuiụị[---]
Sinistroverse, Faliscan alphabet. The gentilicium was read as uili by Colonna, who
interpreted it as genitive of a gentilicium Vilius which he somehow derived from uelos
in EF 4. The left-hand stroke of the letter he reads as l, however, is very long and
touches the upper end of the i. The reading is probably rather uiụị[---] or perhaps
uiṇ[---] with a reversed n: the latter could be the beginning of a gentilicium like
Vinucius.
Bibliography: Colonna 1972:446-7 ¢57-8² (autopsy); Renzetti Marra 1974:351. Photograph: Colonna
1972 tab.LXXX nrs.57-8. Drawing: Colonna 1972:446-7.
476
14.2.2. Inscriptions on tiles containing identifiable parts of texts. The following

tiles contain no identifiable gentilicia, but the general build-up of the text can in most
cases be established.
158. Painted in red on plaster across the back of two tiles (67u47 cm and max. 34u43
cm respectively; letters 10-11 cm high).
1
uol[ta]2**[3][---]
1
iatac2̣ ueḷ[3][---]
1
heccu2pạṭ[3][?---]
Sinistroverse, Faliscan alphabet. In the first line, only the lower parts of the letters has
been preserved. The first word is uol[ta] or perhaps uol[tio], followed by the lower
halves of two shafts. In the second line, only a trace of the c is left; the last letter can be
either l (Thulin) or u (Herbig). The traces of the last letters of the third line read by
Thulin and Herbig have now disappeared. One or two missing tiles at the end contained
the gentilicia and affiliations.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8173+ 8212+8224). Bibliogra-
phy: Thulin 1907:287-8 (autopsy); Herbig CIE 8242 (autopsy); Vetter 1953:305 ¢298²; G. Giacomelli
1963:87 ¢99². Drawing: Thulin 1907:287 (reproduced in CIE 8242); Nogara in CIE 8242.
159. Painted in red on two tile fragments (max. 37u 40 cm, let. 11 cm).
[leu]elio[---]
[---]ioca[---]
[hec]up[at]
Sinistroverse, Faliscan alphabet. The a is reversed (r Thulin). – Restoring [---]elio as a
praenomen, [---]io as a gentilicium, and ca[---] either as ca[uif] or as ca[uia] results in
an inscription of one tile; restoring [---]elio as a gentilicium (pup]e(l)[io G. Giacomelli)
would make the text considerably longer.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8197 and 8219?). The right-hand
shard (inv. 8219) does not resemble Thulin’s drawing, but has traces of letters on both sides, those on
the back (ọ?) painted on plaster, which fits Herbig’s (but not Thulin’s) description; the left-hand shard
(inv. 8197) is easily recognizable, but here the letters on the back are painted directly on the tile, in
accordance with Thulin’s, but not with Herbig’s description. Bibliography: Thulin 1907:282 ¢18²
(autopsy); Herbig CIE 8234 (autopsy); Vetter 1953:303 ¢291²; G. Giacomelli 1963:85 ¢91². Draw-
ing: Thulin 1907:282 (reproduced in CIE 8234).
160. Painted in red on plaster on the back of a tile fragment (40 u 32 cm, letters c.7 cm
high).
[p]ọpliạ
Sinistroverse, Faliscan alphabet. Of the o, only the lower half remains, which could be
read as a u. After this, the line bends downwards, showing that this was the last tile of
477
CHAPTER 14
the inscription. Traces of paint are visible after the i, but it is not clear whether they
were part of an a or of an o; being the last (or perhaps only) word of the text, [p]ọpliạ is
perhaps more likely than [p]ọpliọ.
8238 (autopsy); Vetter 1953:304 ¢294²; G. Giacomelli 1963:86 ¢95². Drawing: Herbig CIE 8238.
161. Painted in red on plaster across the back of a damaged tile (43u33 cm, letters c.8
cm high).
[po]plia[---]
[hec]cup[a]
Sinistroverse, Faliscan alphabet. Reading cup[at] instead of cup[a] assumes that the
text ran over two tiles, in which case the second tile probably contained the woman’s
gentilicium or affiliation in the first line.
8239 (autopsy); Vetter 1953:304 ¢295²; G. Giacomelli 1963:87 ¢96². Drawing: Thulin in CIE 8239.
162. Painted in red on the back of a fragmentary tile (17u40 cm, letters 5-7 cm high).
[1]
[---ma]2x ̣ọmoᛂᛂuoltilio
Sinistroverse, Faliscan alphabet. The traces of the first letters, , point to xo rather than
to no (man]ọmo Herbig, who also considered max]ọmo) or p (Thulin). The vertical
strokes after ma]x̣ọṃ o, read by Thulin as ii or e, were interpreted by Herbig as inter-
puncts: this seems to be correct, a double stroke-interpunct is unique (§11.2.4.3). Of the
t, only the shaft is preserved. The last three letters are written under the line in boustro-
phedon. A preceding tile will have contained the praenomen and the gentilicium.
Bibliography: Thulin 1907:290 ¢27² (autopsy); Herbig CIE 8244 (autopsy); Vetter 1953:305 ¢300²; G.
Giacomelli 1963:88 ¢101². Drawings: Thulin 1907:290 (reproduced in CIE 8244); Nogara in CIE 8244.
163. Painted in red on plaster on a tile fragment (max. 43u43 cm, letters 12-13 cm).
1
[u]olṭ[.[2]--- [3?]---]
1
[u]oltịḷ[2][i---[3?]---]
Sinistroverse, Faliscan alphabet. Of the t in the first line, only a part of the shaft is left.
Of the last letters of the second line, only the upper parts are preserved. The fragment is
the upper right-hand corner of the tile and therefore gives the beginning of the text: the
first letter of each line can be restored on the same fragment, where the plaster has
crumbled away. The inscription occupied probably three tiles, containing part of the
gentilicium in the first line.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8203+numberless fragment).
Bibliography: Herbig CIE 8245 (autopsy); G. Giacomelli 1963:88 ¢102². Drawing: Herbig CIE 8245.
478
164. Painted in red along the length of the back of a damaged tile (max. 52u26 cm,
[---u]oltio:
[---]o:
Sinistroverse, Faliscan alphabet. The lines end in the interpuncts read by Thulin, not in
the i’s read by Herbig. [U]oltio may be a praenomen or a patronym.
1907:283 ¢20² (autopsy); Herbig CIE 8246 (autopsy); Herbig 1914:239 ¢5²; Vetter 1953:305 ¢301²; G.
Giacomelli 1963:88 ¢103². Drawings: Thulin in CIE 8246; Herbig CIE 8246.
165. Painted in red on the front of a tile of which three fragments are preserved (24u13
cm and 39u19 cm, with a loose fragment of 16u9 cm; letters 5-9 cm high).
[---]*i:u[o]ltiailo
Sinistroverse, Faliscan alphabet. A small trace seems to be visible before the first i. –
Thulin read u[o]ltiailo, which he regarded as a hypercorrective form for uoltielo, but
there is no parallel for ai used in this way; Herbig isolated lo as an abbreviation of lo-
ferta. Reading [---]*i as ---ai, which is far from certain, he arrived at three possible
interpretations. The first, taking both forms in -ai as datives, with uoltiai as a patronym
(‘for ...a, daughter of Volta, freedwoman’), assumes that freedwomen could have a
patronym.250 The second, taking the forms as dative and genitive respectively (‘for ...a,
freedwoman of Voltia’), is certainly better: G. Giacomelli’s objection that women are
never designated by a patronym alone carries little weight in view of the possibility that
uoltia is also a praenomen (§7.7.1.86). The third possibility is to take both forms in -ai
as genitives (‘of ...a, freedwoman of Voltia’): for the genitive in -ai, see §4.2.2.
comelli 1963:88-9 ¢104². Drawings: Thulin 1907:290 (reproduced in CIE 8247); Nogara in CIE 8247.
166. Painted in red on plaster on a tile fragment (max. 23u30 cm, letters c.11 cm high).
[---]cẹla[---]
[---] iun[---]
Sinistroverse, Faliscan alphabet. Herbig read the first line as cẹḷạ, which he interpreted
as a cognomen, followed by a patronym ịuṇ[eo. If the first line is indeed cẹla, this is
rather a noun (cela / iun( ) ‘cella Iunae’ G. Giacomelli); note, however, that cela
always refers to the tomb, not to the loculus (§6.3.8). In any case, cẹla cannot be
considered certain. The two strokes that Herbig read as e are thinner than those of the
250
G. Giacomelli dismisses this as absurd, but her alternative, taking uoltiai as an adjectively
used gentilicium (‘a Voltian freedwoman’?) is at least equally unattractive.
479
CHAPTER 14
other letters and very close together. They could well be a badly painted i. Apart from
that, the text is fragmentary, and several letters may be missing before cẹla.
8251 (autopsy); G. Giacomelli 1963:90 ¢108². Drawing: Herbig CIE 8251.
167-168. Painted in red (letters 10 cm high) on a tile fragment (max. 30u32 cm).
[---]io:uolti[---]
Sinistroverse, Faliscan alphabet. Herbig read ---]iọ uolọ[---, but the original shows an
interpunct, and ti where Herbig read ọ. The second word is therefore in all probability a
patronym uolti[o or uolti[lio.
Painted in red (letters 10 cm high) on the other side of the fragment (30u32 cm).
[---]*ṛ[---]
[---]ạṛp[---]
[---]**[---]
Sinistroverse, Faliscan alphabet. The r in the first line can also be read as a; the second
line may also be read as [---]ṛạp[---]. Of the third line, only vague traces are preserved.
8258a-b (autopsy); G. Giacomelli 1963:91-2 ¢113,IIa-b². Drawing: Herbig CIE 8258.
169. Painted in red on plaster along the length of the back of a damaged tile (max.
37u32 cm, letters 10 cm high).
[---ar]utor[---]
Sinistroverse, Faliscan alphabet. The interpunct read by Herbig between the o and the r
is not visible. The restoration is Vetter’s. Another possibility may be s]uto r[ or s]utor[
(cf. suto Lat 250).
8267 (autopsy); Vetter 1953:306 ¢310²; G. Giacomelli 1963:91-2 ¢113,VIII². Drawing: Herbig CIE
8267.
41u41 cm, letters 8-10 cm high).
[1]
[---]2***
[1]
[---]2ocuẹ
[1]
[---]2ịọ
Sinistroverse, Faliscan alphabet. The traces of the first line seen by Thulin and Herbig
cannot be read any more. The second line was read as ...ocua... by Thulin, with a name
in Cua- as he had read also in MF 129. Herbig read ---]ocue, assuming a coordination
of two names by –cue, but this would be the only instance in which the second name
480
would also be a man’s. The third line was read as ị by Herbig, but the traces to point to
two letters ([---]ịọ or [---]ịạ).
1907:296 ¢35² (autopsy); Herbig CIE 8268; G. Giacomelli 1963:91-2 ¢113,IX². Drawings: Thulin
1907:296 (reproduced in CIE 8268); Herbig CIE 8268.
14.2.3. Tiles inscribed in the Latin alphabet. The following tiles are inscribed in the
Latin alphabet. It is unclear whether this points to a date after c.240. Another tile that
may be inscribed in the Latin rather than the Faliscan alphabet is MF 140.
171. Painted in red on the back of a damaged tile (40u47 cm, letters 13 cm high).251
[---]2ọc ̣eᛍf
Sinistroverse, but Latin alphabet. The interpunct consists of a small vertical line
(§11.2.4.3). The first three letters are . Herbig, comparing poe[?---] MF 130, written
in a similar way, read this as an abbreviated name poe (adopted by Vetter and G.
Giacomelli). The second and third letter, however, are rather ce (Thulin, Pauli in Herbig
CIE 8236), which makes a preceding [---]p difficult, as this can hardly be the end of a
gentilicium or praenomen. It seems better to read the first letter with Pauli as [---]o; a
similar o occurs in MF 88 (where it was probably due to the fact that the painter was
writing a vertically placed line).
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8222/8168). Bibliogra-
phy: Thulin 1907:290 (autopsy); Herbig CIE 8236 (autopsy); G. Giacomelli 1963:86 ¢93². Draw-
172. Painted in red along the length of the back of a tile (58.5u44 cm; letters 10-15 cm
high).
[1]
[---]2cfmo[3][---]
[1]
[---]2mosot[3][---]
The first line is written dextroverse in Latin alphabet. The second line was read by
Thulin as mosox (sinistroverse), which he considered restoring also in the first line
(mo[sox), assuming a Latin-Faliscan bilingue. Nogara (in Herbig CIE 8266) and Herbig
read the second line upside down as ---losom (sinistroverse). G. Giacomelli’s mosot
(sinistroverse) is probably correct, but unclear: is sot the same as zot ‘sunt’ MLF 285?
The difference between both lines is due to ancient re-use of the tile (§11.1.4.1c).
Bibliography: Thulin 1907:295-6 ¢34² (autopsy); Herbig CIE 8266 (autopsy); G. Giacomelli 1963:92
¢114². Drawing: Thulin 1907:295 (reproduced in CIE 8266).
251
“In calce superiniecta, sed ubique detrita et deleta alterius inscriptionis item rubro colore
pictae incertissima vestigia aegre cernuntur” (Herbig CIE 8236). Nothing now remains of these
traces, or indeed of the plaster on which they were painted.
481
CHAPTER 14
173. Painted in red on plaster252 along the length of the back of a tile (69u46 cm, letters
c.15 cm high).
[1]
minos[2][---]
[1]
cmurụ[2][?---]
Dextroverse, Latin alphabet. Thulin’s drawing shows a rounded u followed by a vague s
at the end of the second line (mino.sa | c.muru.(s)), but Herbig’s drawing shows only
two vague vertical lines, which he read as ụ. Mino is clearly mino(r), perhaps a
woman’s name: see also §3.5.7b for the spelling Mino(r). A tile is probably missing at
the end: Pauli and Thulin assumed that this was LtF 174.
Bibliography: Thulin 1907:292-4 ¢31² (autopsy); Herbig CIE 8254; Bormann CIL XI.7517; Vetter
1953:307 ¢312a² (autopsy); G. Giacomelli 1963:91 ¢111a². Drawings: Thulin 1907:292 (reproduced in
174. Painted in red on plaster along the length of the back of a tile (69u46 cm, letters
c.15 cm high).
] decon[
]a.f [
Dextroverse in Latin alphabet with cursive e () and a (). The text is doubtlessly
incomplete; the extant letters could perhaps be read as D. Econ[…] or Dec. On[…], or
possibly De. con[iunx], followed in the second line by a less doubtful A(uli) f(ilius).
The tile was joined to LtF 173 by Pauli (in Herbig CIE 8254-8255), Thulin, and Bor-
mann. I doubt whether this is possible, but it does not in any case yield a better text.
Thulin, the only editor of those who joined the tiles to give an interpretation, read
mino.sa decon | c.muru.(s)a.f., an epitaph of two brothers Mino Deconius and
C. Murronius, sons of Sa- mino, however, is a woman’s name.
1907:292-4 ¢31² (autopsy); Herbig CIE 8255 (autopsy); Bormann CIL XI.7517; Vetter 1953:307 ¢312a²
(autopsy); G. Giacomelli 1963:90-1 ¢111b². Drawing: Thulin 1907:292 (reproduced in CIE 8255);
Herbig CIE 8255.
14.2.4. Tiles containing only fragments of texts. The following tiles are clearly
inscribed, but the preserved fragments of the texts are too short to establish either their
contents or the structure of the text.
175. Apparently unpublished is the inscription on a tile without number in the depot of
the Museo Archeologico dell’Agro Falisco, Civita Castellana. The tile, damaged at the
252
Thulin (1907:292) described both this inscription and MF 174 as “unmittelbar auf der rauhen
Aussenseite aufgemalt”: according to Herbig and Vetter both were painted on plaster, which is
certainly wrong in the case of MF 174.
482
upper end and in the lower left-hand corner, measures 47 cm (length, incomplete) by 45
cm (width). Painted in red on plaster across the back is:
[1]
[---] 2 [
[1]
[---] 2 [
[1]
[---] 2ro [
Sinistroverse, Faliscan alphabet. The height of the r is 9 cm, that of the o 7 cm. There
are no traces of lines preceding this (although there is room for two lines), nor of any
letters following the text, so that at least one tile is missing at the beginning. Cf. [---]ro 
[---] MF 178.
176-194. The following fragments were first published by in the CIE. Except in the last
four cases, I republish them from autopsy, and I have been able to improve upon the
readings of Herbig and Nogara here and there.
31u46 cm, letters c.10 cm high).
[1?]
[---]2?la[---]
[1?]
[---]2?mạ[---]
Sinistroverse, Faliscan alphabet. The first letter of the first line, of which the bottom
part has been preserved, is certainly an l, not the c read by Herbig; the second letter of
the second line is in all probability an a. The text starts near the edge of the tile, and
another tile may be missing in front.
8259 (autopsy); G. Giacomelli 1963:91-2 ¢113,III². Drawing: Herbig CIE 8259.
177. Painted in red on plaster on a tile fragment (20u23 cm, letters 11 cm high).
[---]*[---]
Sinistroverse. Only the right half of the letter is preserved: it can be read as either ọ[---]
or c ̣[---]. If the fragment is held the other way up, the reading is [---]o[---].
8262 (autopsy). Drawing: Herbig CIE 8262.
178. Painted in red on plaster on the back of a tile fragment (16-22u34 cm, letters 10-12
cm high).
[..]roc ̣[---]
Sinistroverse, Faliscan alphabet. The letters are squarish, with a diamond-shaped o ( ).
Of the c, only the right half is preserved: it could also be read as o.
8263 (autopsy); G. Giacomelli 1963:91-2 ¢113,VI². Drawing: Herbig CIE 8263.
483
CHAPTER 14
179. Painted in red on plaster along the length of the back of a tile (60u45 cm; letters 9
cm high).
[---]el[---]
Sinistroverse. Herbig read vl, but this is doubtful. The letters are on the lower right-hand
corner of the tile, and other lines as well as another tile may have preceded the text.
8263a. Drawing: Nogara in CIE 8263a.
180. Painted in red on plaster on the back of a tile fragment (33u37 cm, letters c.10 cm
high).
[---]iaụe[---]
Sinistroverse, Faliscan alphabet. The letters [---]ia are part of a woman’s name, and
ụe[---] (the u can no more be read) a gentilicium, a patronym, or the genitive of a man’s
name followed by filia or uxor.
8264 (autopsy); G. Giacomelli 1963:91-2 ¢113,VII². Drawing: Herbig CIE 8264.
181. Painted in red on plaster on a damaged tile (max. 35u40 cm, letters c.9 cm high).
[---]ioo[---]
Sinistroverse. Herbig found only one fragment of this tile, which he read as ---o.o---,
but a second (inv. 8230), containing part of a shaft, can be joined to this on the right-
hand side.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 8230+8211). Bibliogra-
phy: Herbig CIE 8265 (autopsy). Drawing: Herbig CIE 8265.
182. Painted in red on plaster on a damaged tile (max. 47u47 cm, letters c.10 cm high).
[---]o[---]
[---]ups[---]
It is not possible to read c]upa[t in the second line.
8269 (autopsy); G. Giacomelli 1963:91-2 ¢113,X². Drawing: Herbig CIE 8269.
183. Painted in red on plaster across the back of a tile fragment (max. 61u39 cm, letters
c.17 cm high)253.
[---]**o[---]
[---]nia[?---]
Sinistroverse. The first line, read by Herbig, has now disappeared. The second line may
contain the end of a woman’s name.
253
Herbig described it as “a dextra parte fracta” (CIE 8270), but the damage is on the left side.
484
8270 (autopsy); G. Giacomelli 1963:91-2 ¢113,XI². Drawing: Herbig CIE 8270.
184. Painted in red on a tile fragment (max. 28u20 cm, letters c.7 cm high).
[---]lio[?---]
Sinistroverse.
8271 (autopsy); G. Giacomelli 1963:91-2 ¢113,XII². Drawing: Herbig CIE 8271.
185. Painted in red on plaster on a tile fragment (max. 25u22 cm, letters 7 cm high).
[---]io[---]
Sinistroverse, Faliscan alphabet. The i, read by Herbig, has largely disappeared.
186. Painted in red on plaster on a tile fragment (max. 29u28 cm, letters 13 cm high)
[---]a[---]
187. Painted in red on plaster across the front of a tile fragment (max. 39u15-25 cm,
letters 12 cm high).
[---]i*[---]
Sinistroverse, Faliscan alphabet. The trace stands noticeably lower than the i.
188. Painted in red on plaster across the back of a tile fragment (max. 33u12-28 cm,
letters 8 cm high) is
[---]ạ*[---]
Sinistroverse, Faliscan alphabet. Herbig read ṛụ, but the first letter, , might also be an
a or an n. The second letter is now illegible.
189. Painted in red on plaster on a tile fragment (max. 11u19 cm, letters 8 cm high)
[---]n[---]
Sinistroverse.
485
CHAPTER 14
190. Painted in red on plaster on the back of a tile fragment (max. 47u16 cm, letters
12 cm high).
[---]p[---]
Sinistroverse.
191. Painted in red on a tile fragment.

[1]
[ue?]lsu[co[2?]neo
[1]
[---]*ịạ [
Sinistroverse, Faliscan alphabet. Herbig hesitatingly restored the first line as a
praenomen ue]lsu, comparing his uelzu in MF 56: I would rather restore ue]l su[coneo
as in MF 56, where I would read uel zu[con]|eo. The second line is given as in
Pauli’s drawing.
Bibliography: Herbig CIE 8282; G. Giacomelli 1963:92 ¢115,II². Drawing: Pauli jr. in Herbig CIE
8282.
192. Painted in red on a tile fragment.

[---]i[---]
[---]ia[---]
Sinistroverse.
Bibliography: Herbig CIE 8283; G. Giacomelli 1963:92 ¢115,III². Drawing: Pauli jr. in Herbig CIE
8283.
193. Painted in red on plaster on the back of a tile (65u43 cm, letters c.11 cm high).
[---] oc[---] or [---]so [
Sinistroverse. The tile can also be held the other way up, in which case the text can be
read as [---]so.
Bibliography: Herbig CIE 8275 (autopsy). Drawing: Nogara in Herbig CIE 8275.
194. Painted on plaster on the back of a tile fragment (max. 26u41 cm).
ọ*ẹ
Sinistroverse. The reading is Herbig’s.
Bibliography: Herbig CIE 8276 (autopsy). Drawing: Herbig CIE 8276.
14.2.5. Loculus-inscriptions. The following inscriptions were originally cut in the

ledges between the loculi, like e.g. MF 40, 47, 79, 82, and 86: see also §11.1.4.1b.
These ledges were detached and moved to the Museo di Villa Giulia (probably some-
where during the 1880s or 1890s) where they were seen by Nogara in 1903.
486
195. Cut in a strip of tuff (102u22 cm, letters 10-12 cm high).

cauioarutlo
Sinistroverse, Faliscan alphabet. For the unusual syncope in arutlo, see §3.6.6.2, but cf.
also Etruscan arutleσa AS 1.227.
Bibliography: Herbig 1910:183-4 ¢20²; Herbig CIE 8285; Buonamici 1913:63 ¢21²; Vetter 1953:307
¢313²; G. Giacomelli 1963:92 ¢116². Drawing: Nogara in CIE 8285.
196. Cut in a strip of tuff (17u78 cm, letters 12 cm high).

uoltaia
Sinistroverse, Faliscan alphabet. The usual reading is uoltaia ‘Voltaea’ (cf. Lejeune
1952b:118-9), but uolta ia ‘Volta Ia(nti f.)’ is not impossible. For the spelling ai before
a vowel, as in latinaio MLF 210 and perhaps in frenaios MF 471*, see §3.7.6.
Bibliography: Herbig CIE 8287; Vetter 1953:307 ¢315²; G. Giacomelli 1962; G. Giacomelli 1963:93
¢118,I². Drawing: Nogara in CIE 8287.
197. Cut in a strip of tuff (54u12 cm, letters 10 cm high).

cauioṣ*[---]
Sinistroverse, Faliscan alphabet. Nogara’s drawing showes a shaft after the o, on the
edge of the lacuna. Herbig suggested cauioṣị[o, but a Middle Faliscan genitive in -osio
is implausible (cf. §4.4.2): the letters after cauio are rather the beginning of a gen-
tilicium.
Bibliography: Herbig CIE 8286; Vetter 1953:307 ¢314²; G. Giacomelli 1963:93 ¢117². Draw-
ing: Nogara in CIE 8286.
198. Cut in a strip of tuff (20?u47 cm, letters 12-13 cm high).254

iuna
Bibliography: Herbig CIE 8288; Vetter 1953:307 ¢316²; G. Giacomelli 1963:93 ¢118,II². Drawing:
Nogara in CIE 8288.
14.2.6. Inscriptions on pottery. The following inscriptions were stamped or

scratched under various pieces of pottery: beside their attribution to Civita Castellana,
no information is given with regard to their discovery.
254
Herbig (CIE 8288) gives the size of the block as as “m. 0,10 a. x 0,47 l.” and that of the
letters as “m. 0,12-0,13 a.”: either one of these statements must be an error. His drawing, from a
squeeze by Nogara, shows the proportion of height : length of the block as c. 1 : 2.5; if the
height of the block is assumed to be 20 cm, this would fit both the height : length proportion and
the size of the letters as given by Herbig.
487
CHAPTER 14
199. Stamped on the bottom of a black-varnished saucer ( 15 cm, letters 3 mm high).

pleina
Sinistroverse, Faliscan alphabet. The n is reversed. As the name is stamped, it is perhaps
the name of the potter rather than of the owner.
Bibliography: Herbig CIE 8289; Vetter 1953:326 ¢353a²; G. Giacomelli 1963:60 ¢36,I². Draw-
ing: Nogara in CIE 8289.
200. Stamped on the bottom of a black-varnished saucer ( 11.5 cm, letters 4 mm

high).
c ̣cutri
Sinistroverse, Faliscan alphabet. The first c is shown as . The text gives the name of
the potter, either as an abbreviated nominative or as a a genitive (§8.9.1).
Bibliography: Herbig CIE 8290; Vetter 1953:326 ¢353b²; G. Giacomelli 1963:60 ¢36,II². Drawing:
Nogara in CIE 8290.
201. Scratched vertically downwards on the neck of an amphora (height 17 cm, rim
7.5 cm). Known only through an apograph by Della Seta.
titias
Dextroverse. Della Seta’s apograph as published by Herbig (CIE 8585) reads :
if correct, the co-occurrence of  with dextroverse ductus might point to an early date.
Titias is a clearly a genitive. Herbig curiously stated: “Titulus graeci quandam speciem
praebet, sed heros Tit…aj, Tit…hj [...] hic nihil habet, quod agat” (CIE 8585).
Bibliography: Herbig CIE 8585; Vetter 1953:326 ¢354²; G. Giacomelli 1963:61 ¢39²; Pisani 1964:344
¢146Ga²; Rix ET Fa 2.27. Transcription: Della Seta in CIE 8585.
202. Scratched under the foot of a red-figure amphora (height 36 cm, letters 13-16 mm
high) is
nuikuẹ
Dextroverse. The last letter, , is either a cursive e (thus G. Giacomelli) or ii. The
occurrence of k is unexpected, especially before u; assuming that it represents /g/
(§11.2.4.3) does not make the text any clearer. Herbig regarded the letters as non-
Faliscan, and doubted whether they might not in fact be numerals.
Bibliography: Herbig CIE 8291; G. Giacomelli 1963:60-1 ¢37². Drawing: Nogara in CIE 8291.
203. Scratched on the bottom of a saucer ( 11.2 cm, letters 16-22 mm high) is
iun
The u is .
Bibliography: Herbig CIE 8292a; G. Giacomelli 1963:61 ¢38,Ia,II². Drawing: Herbig CIE 8292a.
488
204. Scratched on the bottom of a small plate ( 18 cm, letters 6-7 mm high).
cs
Sinistroverse, with reversed s.
Bibliography: Herbig CIE 8295; G. Giacomelli 1963:61 ¢38,Ia,II². Drawing: Herbig CIE 8295.
14.3. Inscriptions from the surroundings of Civita Castellana
205-210.255 The following texts were cut in the sides of an ancient hollow road
descending towards the Rio Calello just above where it flows into the Rio Maggiore, to
the west of Civita Castellana (see Ward Perkins & Frederiksen 1957:141-2, G. Gia-
comelli 1961:321-5 with map p.322, Quilici 1990:197-208 with map fig.2 between
pp.200-1).
The early editions (Deecke 1888:156, Herbig CIE 8333, and Vetter 1953:310)
were all based on an incomprehensible apograph in a letter by Suarez of 16.IV.1676 in
the MS Vaticanus Latinus 9140 f.328r (reproduced in CIE 8333 and G. Giacomelli
1965b:155).256 The interpretations given by these earlier scholars are not discussed
here: for a discussion, see Di Stefano Manzella 1995:219-21. The inscriptions were
rediscovered in the 1950s during the survey of the British School and subsequently
republished by G. Giacomelli (1961, 1963:69-71), who also devoted an article to the
apograph (G. Giacomelli 1965b). Hollow roads such as these were constantly recut,
deepened, or widened (cf. Quilici 1990), and 206, 207, and 209 in all probability
contain the names of magistrates that had such reconstructions carried out. Other
instances of such inscriptions are MLF/Etr 289, LtF 290, and Lat 291: §11.1.4.5.
205. Deeply cut in the rock on the left side of the road, c.2 m above the level of the
original road (length 4.26 m, letters 28-36 cm high). “Nel 1987 la scritta è stata
malamente rubricata di bianco da un amatore, assieme alla terza oltre ricordata [= 206]”
(Quilici 1990:205 n.19).
furctpcefiuei
Dextroverse, Latin alphabet. Di Stefano Manzella, the only author to propose an
interpretation, suggests ‘Furc(ulam) T(itos) P(---) c(ensor) ef(fodi) i(ussit) ve¢h²i(culis)’
255
These were the first Faliscan inscriptions to be recorded (in 1676), although they were not
the first to appear in print: that honour was reserved for MF 79 (in 1726).
256
Deecke (1888:156), Herbig (CIE 8333), Vetter (1953:310), and G. Giacomelli (1963:69)
ascribed the apograph to Suarez. Bormann (CIL XI p.477, p.1323) described it as ‘ignota manu’,
noting that another apograph in the codex (‘f.76=92’) was made by Ciampino. G. Giacomelli
later ascribed the apograph to Dell’Arena (1965b, 1978:534), noting that it was not in Suarez’
hand.
489
CHAPTER 14
or ‘Fur(culae) T(itos) P(---) coiravit ef(fodiendum)/ ef(ficiendum) i(ter) ve¢h²i(culis)’,

elaborating Cristofani’s suggestion that furc is a noun and ef a verb, probably effodio.
Although possible, this depends on whether or not furca or furcula can mean ‘gola’,
‘passaggio’, which in turn depends on the exact meaning of the toponym Furcae
Caudinae. A major point is why the text should have been abbreviated: although
abbreviating the more formulaic elements of building-inscriptions is of course com-
monplace, it is unexpected in what is apparently a unique phrase. Here, even the names
are irrecognizably abbreviated, which is strange: although Di Stefano Manzella
(1995:224) rejects G. Giacomelli’s suggestion that these inscriptions were electoral
propaganda, being named in a building-inscription may well have had an added
propaganda value for the magistrate in question, as Cristofani noted. The alternative is
that the inscription consists of names: cf. e.g. Furcilia in CIL XI.3855 from Saxa Rubra,
and Veianius in CIL XI.3805 from Veii, and the “fratres Veianii ex agro Falisco” (Varro
R 3.16.10). G. Giacomelli compared ef to the efiles in MF 113 etc.
Bibliography: Ward Perkins & Frederiksen 1957:141 (autopsy); G. Giacomelli 1963:69-71 ¢62,I²
(autopsy); G. Giacomelli 1965b; G. Giacomelli 1978:533-4 ¢11,I²; †Moscati 1985a:128-9 (autopsy);
Cristofani 1988:20; †Gasperini 1989:69-71; Quilici 1990:205-7 (autopsy); Di Stefano Manzella 1996.
Photographs: Ward Perkins & Frederiksen 1957 pl.XLIb; †Moscati 1985a; Quilici 1990 tav.IIc.
Drawing: G. Giacomelli 1965b:155; Quilici 1990:199 fig.1,D.
206. Cut in the rock on the left-hand side of the road, c.40 m past 205, at c.1.5 m above
the original road-level (length 3.54 m, letters 28 cm high). In 1987 the letters were
painted in (see 205).
a[.]osenauẹnarionio
Sinistroverse, Faliscan alphabet. The photographs show that letters could be missing at
the beginning of the text ([l]a or [m]a?) and appear to show that the ninth letter is o
rather than e. If ue is correct, this is an abbreviation (perhaps of the praenomen Venel)
rather than Vel with (unparallelled) omission of -l. Ward Perkins & Frederiksen’s
provisory reading ...mac ena u[.]nerionio can be disregarded, as can Pallottino’s …
macenavenerionio (in Ward Perkins & Frederiksen 1957:141-2, based on their photo-
graphs). Di Stefano Manzella reads ++ Nosena.
Bibliography: Ward Perkins & Frederiksen 1957:141-2 (autopsy); G. Giacomelli 1963:69-71 ¢62,II²
(autopsy); G. Giacomelli 1965b; G. Giacomelli 1978:533-4 ¢11,II²; †Moscati 1985:# (autopsy);
Cristofani 1988:20 (autopsy); Quilici 1990:205-7; Di Stefano Manzella 1995:224. Photographs: Ward
Perkins & Frederiksen 1957 pl.XLIa; †Moscati 1985:#; Quilici 1990 tav.IIa. Drawings: Reynolds &
Meiggs in Ward Perkins & Frederiksen 1957:142; G. Giacomelli 1965b:155; Quilici 1990:199 fig.1,G.
207. Cut in the rock, 70 cm from 206, at 120-150 cm above the actual road-level (length
86 cm, letters 6-10 cm high).
cauio lulḷio
490
Sinistroverse, Faliscan alphabet. There seem to be traces in the space between the two
words. Of the third l, only the left-hand shaft is preserved. The spelling with a double
consonant is very rare (#): Di Stefano Manzella proposes to read lulio or iulio.
Bibliography: G. Giacomelli 1961:323 ¢2² (autopsy); G. Giacomelli 1963:69-71 ¢62,III²; Olzscha
1965:123; G. Giacomelli 1965b; G. Giacomelli 1978:533-4 ¢11,III²; Quilici 1990:205-7 (autopsy); Di
Stefano Manzella 1995:224. Drawing: G. Giacomelli 1961:323 fig.2; G. Giacomelli 1965b:155.
208-209. Cut in the rock on the right side of the road near 206-207, but higher above the
actual road-level (length. 43 cm, letters 8-13 cm high).
puiatụ
Sinistroverse, Faliscan alphabet. G. Giacomelli (1963) and Olzscha compared Etruscan
names like Puina.
Cut in the rock, 40 cm from 208 (length 28 cm, letters 7-10 cm high).
n*
Sinistroverse, with reversed n. The second letter is illegible. Di Stefano Manzella,
apparently reading ptu, regards it as a contracted form p(uia)tu.
Bibliography: G. Giacomelli 1961:323-4 ¢3a-b² (autopsy); G. Giacomelli 1963:69-71 ¢62,IV-V²;
Olzscha 1965:123; G. Giacomelli 1965b; G. Giacomelli 1978:533-4 ¢11,IV-V²; Di Stefano Manzella
1995:224. Drawings: G. Giacomelli 1961:323 fig. 3a, 324 fig.3b; G. Giacomelli 1965b:155; Quilici
1990:199 fig.1,G.
210. Cut in the rock, 54 m past 208-209, at c.5 m above the actual road-level.
cauio latinaio
For the preservation of /a/ before a vowel, see §3.7.6.
Bibliography: G. Giacomelli 1961:324-5 ¢4² (autopsy); G. Giacomelli 1962 :363-367; G. Giacomelli
1963:69-71 ¢63²; Quilici 1990:205-7 (autopsy); Di Stefano Manzella 1995:224. Photograph: Quilici
1990 tav.Va. Drawings: G. Giacomelli 1961:324 fig.4; Quilici 1990:199 fig.1,C.
211. Painted (“painted on plaster”, Conway 1897:373) over two tiles found in situ in a
tomb 3 km outside Civita Castellana in the direction of S. Maria di Falleri, presumably
along the ancient road connecting both sites. Known only through an apograph by
Mariani, the discoverer of the tomb.
cmecio:a[---]
cesilia
Sinistroverse, Faliscan alphabet. The a in the first line is the beginning of a longer word,
not an abbreviation (as Vetter and G. Giacomelli took it): Garrucci, the only editor to
have seen the apograph, calls the inscription a “monca leggenda” (1864:62) and read
C Mecio . A... Cesilia.
Bibliography: Garrucci 1864:62; Fabretti CII 2441bis,g; Garrucci SIL 797; Zvetaieff IIM 55; Zvetaieff
III 57; Schneider 1886:105 ¢5²; Bormann CIL XI.3162a; Deecke 1888:140-2 ¢9²; Conway 1897:373
491
CHAPTER 14
¢315²; Herbig CIE 8332; Vetter 1953:309-10 ¢321²; G. Giacomelli 1963:93 ¢119². Drawing: Mariani in
Garrucci 1864 tav.IV.2 (reproduced in CII tab.XLIII, III tab.VIII.9, Deecke 1888 Taf.I, CIE 8332).
212. An inscription, presumably on a tile, from the Piani di Castello near Castel S. Elia,
lying on the ancient road between Civita Castellana and Nepi (cf. Ward Perkins &
Frederiksen 1957:136-40). Known only through an apograph by Manzielli.
[---]nθia
Sinistroverse, Faliscan alphabet. The θ is shown as  in Manzielli’s apograph: it may
have been  in the original, or this may be one of few instances from the ager Faliscus
where the central point was omitted (§11.2.4.2). In any case, Deecke’s noia can be
disregarded: cf. iata : senθia MFL 362. For the use of θ, see §11.2.4 and §3.5.4. There is
no reason to read [ar]nθia[l], as does Rix.
Bibliography: Deecke 1888:156 ¢35²; Herbig CIE 8428; G. Giacomelli 1963:271 ¢LII²; Rix ET Fa 0.9.
Drawing: Manzielli in Deecke 1888 Taf.II (reproduced in CIE 8428).
492
Chapter 15
The inscriptions from S. Maria di Falleri (Falerii Novi)
15.1. Falerii Novi
After the war of 241 BC, a new Falerii, probably a Roman colony (cf. §2.6.2), was
build in the open tableland c.3 miles to the west of Civita Castellana, where the
buildings of the medieval abbey of S. Maria di Falleri and the hamlet of Faleri still
preserve the ancient name. The new town was fortified with heavy walls that to a large
extent are still standing today. Part of the area was excavated in the nineteenth century
by the proprietor of the terrain, Sebastiani. See Ward Perkins & Frederiksen
1957:155-62 with map p.156 fig.26 and aerial photograph pl. XXVI, De Lucia Brolli
1991a:48-63 with map p.49 fig.35, and especially Di Stefano Manzella 1979.
As the founding of Falerii Novi can be dated to the period shortly after 241 BC,
perhaps the decade 230-220 BC (§2.6.2-3), the Faliscan inscriptions from the town and
its surroundings are all classed as Late Faliscan or Latino-Faliscan, and there are indeed
indications that immigrants from Latium may have settled at the new town. The town
itself has yielded one Late Faliscan inscription (LF 213), one Latino-Faliscan inscrip-
tion (LtF 215), four Latin ones (Lat 216-219), as well as LF/Lat 214, which appears to
be a Latin inscription in the Faliscan alphabet: with the exception of Lat 216, these are
all official inscriptions. The tombs from the neighbourhood of the town provide a
number of Late Faliscan, Latino-Faliscan, and Latin sepulcral inscriptions (220-251).
Two inscriptions of unknown origin, caui  tertinei  | posticnu MLF/Cap 474* and l 
quinti Lat 477*, may also be from Falerii Novi: see §18.2.
15.2. Inscriptions from the town
213. In a mosaic across the entrance of a small building “litteris fere palmaribus”
(Garrucci 1877:199). The inscription was discovered shortly before 1870 but apparently
destroyed soon afterwards (cf. Garrucci 1870:33, SIL 808).
[..]hirmiom[f]cetertineocfpret[ores ?---]
Sinistroverse, Faliscan alphabet. The [f] may have disappeared during an ancient
repair (“post praenomen patris nullum indicium apparuit litterae deperditae, nisi quod
tessellae ibi erant temere positae”, Garrucci SIL 808). Of the t in pret[, only the upper
493
CHAPTER 15
part was preserved. Pret[ has been read as an abbreviation (Garrucci SIL, Deecke,
Jacobsohn), but may well be restored to pret[ores (Bormann, Vetter, G. Giacomelli).
The language of this text shows Faliscan features in the omission of word-final -s in
hirmeo and tertineo and in the monophthongization of /a/ to /ę̄ / in ce and pret[ores, and
the antevocalic e in hirmeo: see §3.6.2. (The antevocalic e in tertineo probably repre-
sents /ē ̣/ ← //, cf. the genitive tertinei MLF/Cap 474*.)
Bibliography: Garrucci 1870:33 (autopsy); Garrucci SIL 808; Zvetaieff IIM 67; Zvetaieff III 69;
Schneider 1886:106 ¢24a²; Bormann CIL XI.3156a; Deecke 1888:162-3 ¢38²; Conway 1897:377 ¢323²;
Jacobsohn 1910:4 ¢24²; Herbig CIE 8343; Vetter 1953:307 ¢317²; G. Giacomelli 1963:69 ¢61². Draw-
ing: Garrucci 1870 tav. nr.6 (reproduced in IIM tab.VII.3 (whence CIE 8343), Deecke 1888 Taf.II).257
214. Engraved on a bronze tablet (52u18 cm, letters 2 cm high), originally fastened
with rivets, from which it would appear to have been torn or broken. The right half was
found in 1860 near the Porta Cimina. Garrucci, in whose possession it came, showed it
to Henzen, from whose notes it was then published by Mommsen without Garrucci’s
consent. The other half was found in 1870, reputedly together with 215 and 217-218,
after which the inscription was published in its entirety by Garrucci. Date: c.150.
menerua sacru
lacotenalafpretod de
zenatuosententiaduootum
dedet cuandodaturected
cuncaptum
Sinistroverse, Faliscan alphabet. The space between menerua and sacru is intentional.
The l in lacotena and the r in pretod are damaged but certain. Bréal (whence Dessau)
read the second la as lr, but this cannot be justified.
Menerua is dative, the usual construction with sacer in both Latin and Sabellic,
not a genitive menerua(s) (thus Deecke, Herbig 1910, and G. Giacomelli). These
datives in -a seem to have spread from Latium and the Latin-speaking colonies (see
Villar 1986). If in these forms the -a represents /-ē/, as has been proposed (see §3.7.6),
their spread is not due to morphophonological interference, but only to the spread of a
different and specific spelling. The -d in pretod has been regarded as an erroneous use
of the Etruscan r (Bréal, Schneider, Bormann, Dessau, Buonamici, R.Bloch, Ernout,
Degrassi, De Rosalia) and as a saṇdhi with the following d- (Deecke, Herbig CIE,
Lommatzsch, Warmington, Vetter, G. Giacomelli, Pisani), but the new attestations of
the same spelling in cuestod pi pretod pis LF 242 show that it may rather represent a
weakened /-r/, probably a weak alveolar flap [ ] (Peruzzi 1997), although M. Mancini
suggests it may be an attempt to render the Latin realization or /r#/: see §3.5.7b. The
formula de | zenatuo sententiad may well be Latin.
257
G. Giacomelli erroneously ascribes the apograph to Zvetaieff.
494
THE INSCRIPTIONS FROM S. MARIA DI FALLERI (FALERII NOVI)
The interpretation of cuando  datu  rected | cuncaptum is debated. Many

editors (e.g. Deecke 1888, Herbig CIE, Buonamici, Warmington, and G. Giacomelli)
have followed Bréal in assuming that datu has the sense of uotum (“quand il a été fait, il
a été correctement conçu”, 1881:492). Cuncaptum is usually taken in a similar sense as
in the expression concipere uerba ‘to phrase a vow’ (see Norden 1939:92-6). The
phrase as a whole has been compared to Liv. 36.2.3 (first Dessau), where M’ Atilius,
vowing to institute the ludi magni, is represented as saying “quisquis magistratus eos
ludos quando ubique faxit, hi ludi recte facti, donaque data recte sunto”. Wachter
discussed the text in the light of this passage. First, he rightly dismissed Vetter’s
datu(r), for if the phrase from Livy indeed constitutes a parallel, the only possibility
with datu(r) would be quando datur, recte conceptum esto, with a future sense that is
impossible in the light of uootum dedet in the previous sentence. According to Wachter,
cuncaptum refers not only to the phrasing of the vow but also to the actual dedication of
what was vowed (as in sacrum and sacrificium concipere): the phrase would then mean
‘es ist aufgrund eines seinerzeit gebilligten Gelübdes dargebracht worden’ as well as ‘es
ist richtig und mit den richtigen Worten dargebracht worden’. Together, this would be
more or less ‘es ist unter in jeder Hinsicht richtigen Voraussetzungen dargebracht
worden’. I find this strained: if cuncaptum can refer to the dedication itself, I do not see
any reason to assume that it refers to the phrasing of the vow at the same time.
This is apparently the youngest inscription in the Faliscan alphabet, and the main
subject of discussion is in how far the inscription was, in fact, still Faliscan. This is all
the more interesting as it is a public dedication, and therefore apparently reflects the
language used by local magistrates. The inscription cannot be much older than c.150
BCE, as is shown both by the double oo in uootum and by the u’s in sacru, uootu, datu,
and cuncaptum also point to the second century. Orthographically, it is almost entirely
Faliscan: note the alphabet, the ductus, and the use of z- in zenatuo (cf. §11.2.5.2,
§3.5.3):258 the only non-Faliscan features are the double o in uootum, and perhaps the -a
for /-ē/ in menerua. Linguistically, the matter is more difficult.
The following linguistic features could potentially be regarded as Faliscan
(arranged in decreasing order of likelihood):
(1) the -d in pretod, cf cụestod pi pretod pis LF 242, although there is a Latin
parallel in opeinod deuincam ted CIL I2.547: related to this are the cases where -r was
omitted in Faliscan and in Latin (as Peruzzi (1997:63) rightly notes);
(2) the omission of -s in zenatuo, which is in accordance with Faliscan custom of
omitting -s after a short vowel virtually without exception, although there are of course
abundant parallels for this in Latin inscriptions (see §3.5.7d);
258
The use of cu rather than qu cannot be regarded as a separate Faliscan feature (as does
Freeman 2002:252), as the Faliscan alphabet did not have a q: the use of cu is therefore a
corollary of the use of the Faliscan alphabet.
495
CHAPTER 15
(3) the monophthongization of /a/ in pretod, which is regular in Faliscan, although

there are instances of this in contemporary Latin inscriptions (see §3.7.6);
(4) the omission of -m in sacru and datu, which occurs in Faliscan, but of course in
Latin as well (§3.5.7a);
(5) the fourth-declension genitive ending -uo(s) in zenatuo, which is attested only from
Latin, but may have existed in Faliscan, where the consonant-stem genitive ending was
-os (§4.5.2);
(6) the u in sacru, uootum, datu, and cuncaptum appears to be Latin, but could con-
ceivably be second-century Faliscan (§3.6.6.1): note [fel]ịcinatiu MLF 384;
(7) the lack of vowel raising in cuncaptum (see §3.6.6.1), although cuncaptum is rather
a re-analysis of conceptum after the simple verb (perhaps intended as an archaism);
(8) the -d in sententiad and rected is unexpected at this date both in Faliscan (cf. ifra
MF 40) and Latin, and may be an archaism (§3.5.7c): as the -d occurred originally in
Faliscan as well as in Latin, it could be an archaism in either.
As is clear from this list, most of these features occur in Latin as well, and could
without problems be regarded as dialect Latin. The only exception seems to be the -d in
pretod, although here too, the Faliscan parallels are limited to one inscription. (Adams
(2007:106-7) also points to cuando used in a temporal sense as a lexical regionalism.)
How can it be decided what the language of this inscription actually is?
In many cases, this discussion has been obscured by too much attention being
given to the fact that the inscription is written in the Faliscan alphabet. As Adams
rightly notes in the case of this inscription, one cannot argue for the language of a text
on the basis of the alphabet it was written in. Yet the fact that the inscription was
written in the Faliscan alphabet reflects a choice made by those who had it engraved, a
choice to present the text as Faliscan or rather, in a Faliscan context, whether this choice
was made out of local patriotism or, as I suspect, out of a sense of tradition. It could
even be regarded as an expression of ethnic identity, in which case, as noted, it is very
interesting to find this in an official inscription.259
The question can therefore be posed differently: if the inscription was not written
in Faliscan alphabet and the origin was not known, would it have been regarded as
Faliscan? I greatly doubt this: if the inscription were even ascribed to Etruria, this
would probably be on account of the name cotena rather than on the basis of its
linguistic features. It is possible, however, that in this inscription the influence from
Latin is relatively great due to its being an official text of a type that is unattested for
Faliscan and may be Latin in origin (§9.4.1-2).
259
I wonder if the inscription was perhaps intended to replace an older Faliscan inscription, the
replacement being drafted in more contemporary language, as in the case of the Umbrian
Tabulae Iguuinae.
496
In my view, the difficulty underlying the recent discussions on the language of

this inscription is at least partly due to the fact that the difference between Faliscan and
Latin was gradual, especially in this later period, when the language that was spoken in
the ager Faliscus was a mixture of a more general ‘rural Latin’ and features derived
from Faliscan. Any text from this period is not either Latin or Faliscan, but both
Faliscan and Latin to a certain degree. Calling this text Faliscan (as do Peruzzi,
Mancini, and Freeman) or Latin (as do Wachter, Vine (?), and Adams) means nothing
more than to give more relative weight to either ‘side’ of the text. If anything, this
inscription shows that the language of local public inscriptions in the ager Faliscus was
becoming more and more indistinguishable from a more general ‘rural Latin’.
From autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 24425). Bibliography:
(I) Mommsen 1860:452-3; Garrucci 1860:266-9 ¢1² (autopsy); Detlefsen 1861a:199; Garrucci
1864:61-2 ¢1²; Fabretti CII 2441. Drawing: Garrucci 1860 tav.F.1 (reproduced in Garrucci 1864
tav.IV.1, Fabretti CII tab.XLIII). – (II) Garrucci 1870:25 (autopsy); Fabretti CII Suppl. I 2441 add.;
Garrucci SIL 559; Bréal 1881 (autopsy); Deecke 1881:237; Zvetaieff IIM 68 (autopsy); Zvetaieff III 70;
Schneider 1886:106 ¢17²; Bormann CIL XI.3081; Deecke 1888:156-61 ¢36²; Conway 1897:376 ¢321²;
Dessau ILS 3124; Herbig 1910:185 (autopsy); Herbig CIE 8340; Buonamici 1913:81-4 ¢53²; Lom-
matzsch CIL I2.365; Ribezzo 1920:79; [Taylor 1923:74]; Warmington 1940:80-1 ¢71²; Bloch
1952:181-2; Vetter 1953:308-9 ¢320²; Safarewicz 1955:185 ¢1²; Ernout 1957:37-8 ¢63²; Degrassi
ILLRP 238; G. Giacomelli 1963:68-9 ¢59²; Pisani 1964:352-3 ¢154²; Peruzzi 1966a:126; G. Giacomelli
1978:532-3 ¢10²; De Rosalia 1978:39, 91-2 ¢17²; Villar 1986:48; Degrassi & Krummrey CIL I2.365
add.; Wachter 1987:448-53; Vine 1993: 108-9; Peruzzi 1997; Freeman 2002; Mancini 2002:33-42;
Adams 2007:100-7; Clackson & Horrocks 2007:119-20. Photograph: Peruzzi 1966a tav.I between
pp.128-9 [reproduced in Peruzzi 1995:62]. Drawings: Garrucci 1870 tav. no.1; Zvetaieff IIM tab.VII.4
(reproduced in Deecke 1888 Taf.II, CIE 8340, CIL I2.365).
215. Engraved on a strip of bronze from S. Maria di Falleri, reputedly found together
with LF/Lat 214 and Lat 217-218.
[---]ilioc[f ?---]
Dextroverse, Latin alphabet. Cf. m  spurilius  c  f | bis  q[  ]duouir Lat 238 from a
tomb near the town. The absence of word-final -s may be a Faliscan feature.
Bibliography: Garrucci SIL 810 (autopsy); Zvetaieff IIM 69; Zvetaieff III 71; Bormann CIL XI under
3081; Deecke 1888:161 ¢37²; Conway 1897:377 ¢322²; Herbig CIE 8342; Vetter 1953:308 ¢318²; G.
Giacomelli 1963:69 ¢60². Drawing: Garrucci SIL 810.
216. Scratched before firing on the back of the right half of a mold for a woman’s head
of terracotta.260 C.240-220 (Di Stefano Manzella).
tfourios*[]f̣
Dextroverse, Latin alphabet. Only the bottom parts of the last letters remain; the father’s
name may either ṭ or p ̣ (II Safarewicz; the shafts are omitted by G. Giacomelli). The
260
Deecke erroneously describes it as a ‘Thonschale mit Junokopf’. It is unclear whether the
woman represents Juno (as was surmised by Garrucci, Fabretti, Zvetaieff, and Deecke).
497
CHAPTER 15
presence of -s and ou are Latin rather than Faliscan (cf. Latin med  loucilios  feced Lat
268 from Corchiano): T. Furius may have been a recent immigrant from Latium (Di
Stefano Manzella).
Bibliography: Garrucci 1864:69 (autopsy); Fabretti CII 2440bis; Garrucci SIL 512; Zvetaieff IIM sub
70; Zvetaieff III sub 72; Bormann CIL XI.6708,26; Deecke 1888:217 ¢101²; Lommatzsch CIL I2.473;
Safarewicz 1955:186 ¢3²; G. Giacomelli 1963:263 ¢XII²; Di Stefano Manzella 1979:72. Degrassi &
Krummrey CIL I2.473 add. Drawings: Garrucci 1864 tav.VII.1a-c (reproduced in CII tab. XLIII, Di
Stefano Manzella 1979:72 fig. 27).
217-218. Engraved on a bronze table (28u9 cm), reputedly found together with 214-215
(letters 9-12 mm high). Minasi261 suggested that 217 was part of an inscription that was
originally larger but was recut when the bronze was reused for 218. There seems to be
no reason to assume this: both inscriptions appears to be complete. Courtney, too,
argued for re-use, noting that the table must have been fastened against a wall, so that it
would never have been possible to read both sides at once.
gonlegiumquodestaciptumaetateiaged[ai]
opiparumaduiitamquolundafestosquedies
queisoueisaastutiesopidqueuolgani
gondecorantsaip[i]sumecomuiuialoidosque
ququeihucdederun[ti]nperatoribussummeis
uteisesedlubent[isbe]neiouentoptantis
On the other side is engraved 218 (letters 9 mm high).
[di]oueiiunoneimineruai
falescequeiinsardiniasunt
donumdederuntmagistreis
llatriuskfcsalu[e]nauoltaif
coiraueront
Dextroverse, Latin alphabet. My text follows Wachter, who in 217 read uiitam where
all other editors have read ueitam (except Minasi (from autopsy) and recent editors like
Vine, Courtney, and Kruschwitz), and who in 218 restored [di]ouei rather than [i]ouei.
The nt of sunt are written as a ligature , and in 218, the n[t of dederun[t are assumed
to be a similar ligature. Noteworthy is the odd use of g in gonlegium and gondecorant
(but c in comuiuia and q in quolundam and ququei), and uolgani. Girard ascribes this to
an uncertainty of the Faliscans, when using the Latin alphabet, where to use g, which
was absent from their own alphabet, but this is implausible, especially during the Late
Faliscan period, the Faliscans seem,on occasion to have used k to render /g/: see
§11.2.4.2.
261
For Minasi’s publication, I have had to rely on the ample quotations in Peruzzi 1966a.
498
I discuss these texts only briefly, as the language is Latin without specifically
Faliscan dialect features: for detailed discussions, see Kruschwitz 2002:127-38,
Wachter 1987:441-8 and Peruzzi 1966a. Both inscriptions can be dated to the second
half or even the last quarter of the second century. Wachter names the following
orthographical criteria for this: (1) the double-spelling in uiitam and aastutitieis, first
attested c.140; (2) the use of q before u in ququei, first attested (apart from the archaic
inscriptions) c.125; (3) the use of ligatures in sunt and dederun[t, first attested c.150.
Except for a few features that can easily be explained as archaisms (e.g. oi in coirau-
eront and loidosque, -d in opidque and sesed), the linguistic features of the inscriptions
are compatible with this: note e.g. u in closed word-final syllables and ae in aetatei; see
also Peruzzi 1966a:135-50. Arguments for the relative dating of the two inscriptions are
derived from the fact that one of the rivet-holes was placed lower in order to avoid
mineruai, suggesting that 217 was written before 218 (see also Wachter 1987:441-2).
A debated question is the identity of the falesce  quei  in  sardinia  sunt, and, if
the inscriptions are (more or less) contemporaneous, whether they were identical with
the ququei. Peruzzi (1966a:127-8, 160) rightly compared the dedication by the ququei
to the dedications of other ‘guilds’ from Praeneste, where in fact we find a dedication
by the coques atriensis (CIL I2.1447); on the other hand, he compared the falesce  quei 
in  sardinia sunt to the Italici qui Deli negotiantur in the Latin inscriptions from Delos
(CIL I2.2232-2259). In spite of these parallels, the identity of both our groups remains
obscure. Peruzzi (1966a:161-2) hesitatingly suggests that these Faliscans may have
been inhabitants of a Sardinian colony founded under the Gracchi. Wachter (1987:443)
suggests that a group of cooks, forced by (Greek?) competition to emigrate to Sardinia
(why?), made the first dedication to invoke the gods’ blessing on their enterprise, and
the second when they had succeeded in building up new and prosperous existences a
few months (why?) later. As some had been forced to find other employment, by that
time they no longer referred to themselves as the ququei, but as the falesce  quei  in 
sardinia  sunt. Girard suggests that these falesce were clients of the Sp. Carvilius
Maximus who campaigned Sardinia in 234 and who may have been a son or a grandson
of the Sp. Carvilius Ruga who subdued Falerii in 293. Although this is based more on
facts than Wachter’s implausible suggestions, it would appear to be impossible in view
of the date of the inscriptions. From the point of view of ethnic identity, it is worth
noting that this is the only certain instance of the use of the ethnonym Faliscus by
Faliscans, and that it occurs in the context where this would be expected, namely a
group of Faliscans outside their own territory (§2.2.2).
Further points of interest in this inscription are, in 217, the possible use of
Saturnians, although these seem to be so freely used that a metrical interpretation is
difficult (Massaro in fact called the metrical liberties ‘anarchic’), the elevated poetic
language (see Peruzzi 1966a:128-35), and the identity of the inperatoribus  summeis,
who may be the same as the Capitoline triad in 218 (see Peruzzi 1966a:124-5).
499
CHAPTER 15
Bibliography: Garrucci 1864 (autopsy); Garrucci 1871b:259; Garrucci SIL 557-558; Zvetaieff IIM
70a-b; Zvetaieff III 72a-b; Schneider 1886:107 ¢30-31²; Bormann CIL XI.3078a-b; Deecke 1888:193-6
¢62²; Conway 1897:380 ¢335²; †Minasi 1894:221 (autopsy); Lindsay 1897:67-9 ¢XXXVIII²; Dessau
ILS 3083; Herbig CIE 8341 (autopsy); Buonamici 1913:85-9 ¢55²; Bormann CIL XI.3078 add.;
Bücheler 1921:2-3 ¢2²; Taylor 1923:79-80; Bormann CIL XI.7483a-b; Marx 1928:123; Zmigryder-
Konopka & Rozenberg 1936; Warmington 1940:124-5 ¢151a-b²; Vetter 1953:309; Safarewicz
1955:186; Bassi 1957:69-70; Degrassi ILLRP 192; Linderski 1958:47-50; G. Giacomelli 1963:264-5
¢XIVa-b²; Peruzzi 1966a; Till 1976:24-27, 313; Pulgram 1978:205-7; De Rosalia 1978:39, 88-91 ¢16²;
Degrassi & Krummrey CIL I2.364 add.; Wachter 1987:441-8; Girard 1989:167-9; Vine 1993:271;
Zucca 1994:1488; Courtney 1995:34-5, 204-7; Kruschwitz 2002:127-38; Massaro 2007:128-9.
Photographs: Herbig CIE 8341 (reproduced in CIL I2.364, CIL XI.7483a-b); Bassi 1957
tav.XXXII.89a-b; Degrassi ILLRP tab. 93a-b; Peruzzi 1966a tav.II-III; Kruschwitz 2002:129-130
Abb.8-11. Drawing: Garrucci 1870 tav. no.1 (reproduced in Deecke 1888 Taf. III).
219. Incised on a travertin base (25u34.5 (incomplete)u 6 cm; letters 4.5 cm high), recut
on the left, probably as a result of ancient reuse. The first editions were based on the
apographs made in 1822 by Amati, and in 1865 by Mommsen: only in 1979 was the
text published with a photograph, by Di Stefano Manzella. c.120-50.
[.u]mpriciuscf
[ ]aburcusq
[ap]olineidat
Dextroverse, Latin alphabet. As Deecke suggested, it is not necessary to restore aburcus
([c]aburcus Bormann, Conway, Lommatzsch, Degrassi, G. Giacomelli). The p in
[u]mpricius may be a reminiscence of the Faliscan spelling. For the gentilicium, cf.
upreciano in MLF 363 and MLF 364, and umrie Etr XLIII, all from the same tomb
near Rignano Flaminio, and a P. Umbricius Rufus in CIL XI.3254 from Sutri.
Bibliography: (I) Mommsen & Henzen CIL I.1543a; Garrucci 1864:74. Transcription: Mommsen &
Henzen CIL I.1543a. – (II) Mommsen 1865:313 (autopsy); Garrucci SIL 1879; Bormann CIL XI.3037;
Deecke 1888:214 ¢83²; Conway 1897:371 n.1; Dessau ILS 3217; Lommatzsch CIL I2.1991; [Taylor
1923:78]; Safarewicz 1955:186; Degrassi ILLRP 47; G. Giacomelli 1963:263 ¢X². Drawing: Momm-
sen 1865 tav.R.3; – (III) Di Stefano Manzella 1979:81-4 ¢25²; Degrassi & Krummrey CIL I2.1991 add.
Photograph: Di Stefano Manzella 1979:82 fig.34.
15.3. Inscriptions from tombs near the town
220-233 (the ‘tombe Guidi’). The following inscriptions were discovered in situ in
1851 in two tombs near Falerii Novi, by Guidi, who copied them and then removed
them to ransack the graves, leaving the tiles in the tomb. From his apographs the
inscriptions were published by Orioli, the first Faliscan inscriptions to be published as
such.262 The tomb was visited in 1856 by Garrucci, who salvaged what was left and
262
An unpublished apograph of 205-210 had already been made in or shortly before 1676; MF
79 had already been published in 1726, but as Etruscan.
500
republished the inscriptions from autopsy (1860); only later did he discover the tituli
priores and published these as well (1864). After his death the tiles disappeared until
1912, when they were found by Nogara “in horreis musei Vaticani di Scultura” (Herbig
CIE 8344-8353 add.) and relocated to the Museo Gregoriano Etrusco Profano.
220. Painted in red on white stucco over three tiles, the third of which is known only
through Guidi’s apograph263 (the remaining tiles measure 58u41 cm and 65u45 cm;
letters 10-13 cm high). The loculus was reopened for the burial of Gavia Aconia:
afterwards, the letters on the juncture of the tiles were repainted, and the fourth line
added. Having been painted on dry plaster, these additions have largely disappeared.
1
uo2ltiouec3ineo
1
ma2xomo 3
1
iun2eohecu3pat
1
caạ2conia 3
Sinistroverse, Faliscan alphabet. Pace Garrucci (1864:65, SIL 800), Orioli’s ca aconia
and Detlefsen’s (1861b:205) caaconia deserve the benefit of the doubt over Garrucci’s
carconia ‘Gargonia’ (adopted by all editors after him) even if there is no interpunct
between the words. The gentilicum Aconius, attested from Latin inscriptions from the
ager Faliscus may also be read in *(*)coṇẹo LtF 290, aco[n]ẹọ LtF 327 and aco[]nio
LtF 341: see §7.8.1.5. The placing of the patronymic adjective after the cognomen is
exceptional, and probably due to the fact that the cognomen was not yet a fixed element
in the onomastic formula: see §7.9.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20814+ 20816); the
tiles, expecially the second, are now much damaged, and parts are missing. Bibliography: (I) Orioli
1854:XXII ¢10². – (II) Garrucci 1860:273 ¢7² (autopsy); Detlefsen 1861a; Detlefsen 1861b (autopsy);
Garrucci 1864:65; Fabretti CII 2451; Garrucci SIL 800; Zvetaieff IIM 58 (autopsy); Zvetaieff III 60;
Schneider 1886:106 ¢20²; Bormann CIL XI.3159,8; Deecke 1888:170-1 ¢42²; Conway 1897:378 ¢327²;
Jacobsohn 1910:5 ¢29²; Herbig CIE 8347; Nogara in Herbig CIE 8347 add. (autopsy); Buonamici
1913:65 ¢25²; Vetter 1953:311 ¢322d²; G. Giacomelli 1963:94-6 ¢121,IV²; Peruzzi 1963b:440-1, 445;
Pisani 1964:339 ¢144D²; G. Giacomelli 1978:534-5 ¢12,IV². Drawings: Garrucci 1860 tav. G.7;
Garrucci 1864 tav. V.3 (reproduced in CII tab.XLIII, IIM tab. IX.2a, Deecke 1888 Taf.II, CIE 8347);
Zvetaieff IIM tab.IX.2 (reproduced in CIE 8347).
221. Painted in red on plaster along the back of part of a tile and two complete tiles.
Nogara, in 1912, found only fragments of the first and second tiles, which was indeed
all I could find (max. 34u25 cm, max. 28-37u20-26 cm; letters c.10 cm high).
1
uip2iazertene3aloferta
1
ma2rciacar3celini
1
ma2tehecupa 3
263
“Extant adhuc omnes tegulae (m. 0,64 a.); secunda in quinque partes diffracta est” (Nogara
in Herbig CIE 8347 add.). This apparently refers to the two tiles salvaged by Garrucci rather
than to the three tiles of the original inscription.
501
CHAPTER 15
Sinistroverse, Faliscan alphabet. Schulze, Vetter, and Knobloch maintained that

zertenea is an adjective with loferta, ‘a Sertinian freedwoman’, but I doubt very much
whether at this date the gentilicium could still be used in this way: cf. §4.4.11. The
normal word-order in both Faliscan and Latin is noun - adjective (§8.5.1.), and in Latin
this order is maintained when the gentilicium is used adjectivally, as in uia Appia or
pons Aemilius. An instance of adjective - noun would be noutrix paperia CIL I2.45, if
interpreted as ‘a nurse of the gens Papiria’. Zertenea is either Vibia’s original gen-
tilicium, or the gentilicium of the gens where she was a slave and which she adopted as
a freedwoman.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20818+20819).
Bibliography: (I) Orioli 1854:XXII ¢11². – (II) Garrucci 1860:272-3 ¢6² (autopsy); Mommsen
1860:451-6; Detlefsen 1861a; Garrucci 1864:66; Fabretti CII 2452; Garrucci SIL 802; Zvetaieff IIM 60
(autopsy); Zvetaieff III 62; Schneider 1886:106 ¢22²; Bormann CIL XI.3159,5; Deecke 1888:164-7
¢39²; Conway 1897:377 ¢324²; Von Planta 1897:588 ¢319²; [Schulze 1904:513]; Jacobsohn 1910:5
¢25²; Herbig CIE 8344; Nogara in Herbig CIE 8344 add. (autopsy); Buonamici 1913:63-4 ¢22²; Vetter
1953:310-1 ¢322a²; Knobloch 1954:36-7; G. Giacomelli 1963:94-6 ¢121,I²; Pisani 1964:338 ¢143H²;
Peruzzi 1963b:442-3; G. Giacomelli 1978:534-5 ¢12,I²; Mancini 1981. Drawing: Garrucci 1860 tav.G.6
(reproduced in Garrucci 1864 tav.VI.1, CII tab.XLIII, IIM tab.IX.4, Deecke 1888 Taf.II, CIE 8344).
222-223. The titulus prior was painted in red on plaster across the backs of four tiles. I
was able to find only the first tile (67u47.5 cm; letters c.10 cm high): Nogara, in 1912,
still found all four, although apparently in a severely damaged condition. All that I
could read on the tile was ca, the first letters of 222.
1
cau[i2a]uecin[e] 3auotili4ạ
1
mac2iacace3liniu4xo 4
The titulus prior was later washed over and repainted with the titulus posterior, inside a
crude decorative border.
1
mar2cioac3arceli4nio
1
cau2iau3ecine4a
1 2
hẹ c ̣cupat 3 4
Sinistroverse, Faliscan alphabet. The third line is written small. Early editors read hiu;
others conjectured hi[c] (Fabretti), ḥẹ (Zvetaieff, Conway, Pisani), hẹc ̣ (Herbig, Vetter),
or hec (G. Giacomelli). The word is now illegible. Gavia Vicinia died first, and a new
inscription was made when her husband was interred: cupat is a plural, cupa(n)t.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20800). Bibliogra-
phy: (I) Orioli 1854:XXI ¢5². – (II) Garrucci 1860:274-5 ¢11² (autopsy); Detlefsen 1861a. Drawing:
Garrucci 1860 tav.H.11. – (III) Garrucci 1864:66 (autopsy); Fabretti CII 2446; Garrucci SIL 803a-b;
Zvetaieff IIM 61 (autopsy); Zvetaieff III 63; Schneider 1886:106-7 ¢23a-b²; Bormann CIL
XI.3159,2a-b; Deecke 1888:167-8 ¢40²; Conway 1897:377-8 ¢325a-b²; Jacobsohn 1910:5 ¢26-27²;
Herbig CIE 8345a-b; Nogara in Herbig CIE 8345a-b add. (autopsy); Buonamici 1913:64 ¢23²; Vetter
1953:311 ¢322b,A-B²; G. Giacomelli 1963:94-6 ¢121,IIa-b²; Peruzzi 1963b:435-40; Pisani 1964:338-9
¢144A,a-b²; G. Giacomelli 1978:534-5 ¢12,IIa-b²; Mancini 1981. Drawing: Garrucci 1864 tav.VI.2
(reproduced in CII tab.XLIII, IIM tab.IX.5a, Deecke 1888 Taf.II, CIE 8345a-b); Zvetaieff 1885
tab.IX.5 (reproduced in CIE 8345a-b).
502
224-225. The titulus prior was painted in red on plaster (letters c.8-10 cm high) across
the backs of three tiles (69u48 cm, 47u66 cm, and 49u68.5 cm). The tiles were then
washed over and repainted with the titulus posterior (letters 9-12 cm high). In the
process, the tiles were rearranged and what was originally the third tile placed in front,
so that the titulus prior, when discovered, appeared as uolticauecineo | ania
hecupatma (cf. Garrucci 1864).264 Why the inscription was repainted is unclear, as both
inscriptions appear to refer to the same persons. The titulus prior is
1
cau2ecineo3[]uoltio
1
hecu2pat[]m{e}3ania
Sinistroverse, Faliscan alphabet. At the end of the first line a trace is clearly visible after
uolti, as the drawing by Garrucci (1964) shows, and several scholars have therefore
read uolti (Garrucci 1864, Fabretti, Zvetaieff, Deecke, Bormann, Jacobsohn, Herbig).
The tile shows , however, and reading uoltio is justified. In the second line, there is a
letter on the edge of the second tile after the m: editors have read miania (Orioli) or
meania (Garrucci 1864 and all later editors). The letter does not appear to be part of the
repainted inscription (as in MF 220), and is probably an error.
The titulus posterior is:
1
cauec2ineo
1
cama2nia
Sinistroverse, Faliscan alphabet with cursive e (). The c of ca  mania is reversed,
indicating a woman’s name (§11.2.5.9).
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20793+20789+
26794). Bibliography: (I) Orioli 1854:XXII ¢8². – (II) Garrucci 1860:274 ¢8² (autopsy); Detlefsen
1861a. Drawing: Garrucci 1860 tav. H.8. – (III) Garrucci 1864:65-6 (autopsy); Fabretti CII 2449;
Garrucci SIL 801; Zvetaieff IIM 59a-b (autopsy); Zvetaieff III 61a-b; Schneider 1886:106 ¢21a-b²;
Bormann CIL XI.3159,7a-b; Deecke 1888:171-3 ¢43a-b²; Conway 1897:378 ¢328a-b²; Jacobsohn
1910:5 ¢30a-b² Herbig CIE 8348; Nogara in Herbig CIE 8348 add. (autopsy); Buonamici 1913 :65-5
¢26²; Vetter 1953:311-2 ¢322e,A-B²; G. Giacomelli 1963:94-6 ¢121,Va-b²; Peruzzi 1963b:445; G.
Giacomelli 1978:534-5 ¢12,Va-b². Drawing: Garrucci 1864 tav.V.4 (reproduced in CII tab. XLIII, IIM
tab. IX.3a, Deecke 1888 Taf. II, CIE 8348); Zvetaieff 1885 tab. IX.3 (reproduced in CIE 8348).
226. Painted in red on plaster across the backs of four tiles, the fourth of which was
missing already when Zvetaieff saw the tiles in 1880 (the remaining three measure
63u48.5 cm, 65u48 cm, and 68u47.5 cm; letters 8-15 cm high, but slightly less high in
the second line, and only c.5 cm high in the third line).
1
tito[]2acar3celini4o:
1
ma2fipop3petrune4scef
1
[h]ecu[pa] 2 3 4
264
Fabretti’s uolti cauecineo | ania cauecineo | camania is a curious mixture of the titulus
posterior and the rearranged titulus prior.
503
CHAPTER 15
Sinistroverse, Faliscan alphabet. The n is reversed. Garrucci read pop as part of the first
name (Maci. Pop Garrucci 1860, Pop = ‘Pop(lilia tribu)’265 Garrucci 1864), but it is
clearly part of the second, as was seen already by Mommsen and Detlefsen. This
second name is a later addition: the letters are smaller, the interpunct is single instead of
double, and the e’s in petrunes and ce are cursive. (The ‘sagging’ of the line after pop in
Garrucci’s and Zvetaieff’s drawings is non-existent.) The initials of pop and petrunes
are reversed, indicating a woman’s name (§11.2.5.9), but pop  petrunes was first
interpreted as such only by Herbig. The last line, written in small letters and with a non-
cursive e, seems to belong to both inscriptions. Pisani read ẹ instead of [h]e, but there
are no certain parallels for the omission of h- in Faliscan (§3.5.2).
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20790+20805
+20794). Bibliography: (I) Orioli 1854:XXI ¢2-3². – (II) Garrucci 1860:275-6 ¢12² (autopsy); Momm-
sen 1860:199; Detlefsen 1861a; Detlefsen 1861b (autopsy). Drawing: Garrucci 1860 tav.H.12. – (III)
Garrucci 1864:66-7 (autopsy); Fabretti CII 2443-4; Garrucci SIL 804; Zvetaieff IIM 62 (autopsy);
Zvetaieff III 64; Schneider 1886:106 ¢24²; Bormann CIL XI.3159,1: Deecke 1888:169-70 ¢41²; Conway
1897:378 ¢326²; Von Planta 1897:588 ¢320²; Jacobsohn 1910:5 ¢28²; Herbig CIE 8346; Nogara in
Herbig CIE 8346 add. (autopsy); Buonamici 1913:64-5 ¢24²; Vetter 1953:311 ¢322c²; G. Giacomelli
1963:94-6 ¢121,III²; Peruzzi 1963b:442,445; Pisani 1964:338-40 ¢144E²; G. Giacomelli 1978:534-5
¢12,III². Drawings: Garrucci 1864 tav.VI.3 (reproduced in CII tab.XLIII, IIM tab.X.1a, Deecke 1888
Taf.II, CIE 8346); Zvetaieff IIM tab.X.1 (reproduced in CIE 8346).
227. Painted in red on plaster along the backs of three tiles. Of the third tile, the left
half, with the letters us, is known only through Guidi’s apograph, and the right half had
disappeared when Zvetaieff saw the inscription in 1880. The first tile is still intact
(69u47.5 cm; letters 13-17 cm high): the second (54u40 cm) is broken in many pieces.
1
pola2marc3ia:sus[?---]
Sinistroverse, Faliscan alphabet with reversed s. The l is . The p is reversed to
indicate a woman’s name (§11.2.5.9). Sus was explained by Garrucci as a cognomen
‘Sus’, but later editors rightly assumed that letters were missing (Herbig 1910 (cogno-
men?), Jacobsohn, Vetter (cognomen), G. Giacomelli (husband’s name)). Perhaps s
ux̣[or] ‘S(exti) uxor’ could be read, with the second s, , read as x, cf. the x in LtF 301.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20804+20817); I was
unable to find the upper left-hand and the lower left- and right-hand corner of the second tile. Bibliog-
raphy: (I) Orioli 1854:XXII ¢6². – (II) Garrucci 1860:274 ¢9² (autopsy); Detlefsen 1861a; Garrucci
1864:65; Fabretti CII 2447; Garrucci SIL 799; Zvetaieff IIM 57 (autopsy); Zvetaieff III 59; Schneider
1886:106 ¢19²; Bormann CIL XI.3159,4; Deecke 1888:174-5 ¢45²; Conway 1897:379 ¢330²; Herbig
1910:187-8 ¢26²; Jacobsohn 1910:5 ¢32²; Herbig CIE 8350; Nogara in Herbig CIE 8350 add. (autopsy);
Buonamici 1913:67 ¢28²; Vetter 1953:312 ¢322g²; G. Giacomelli 1963:94-6 ¢121,VII²; Peruzzi
1963b:435-40; G. Giacomelli 1978:534-5 ¢12,VII². Drawings: Garrucci 1860 tav.H.9 (reproduced in
Garrucci 1864 tav.V.2, CII tab.XLIII, IIM tab. IX.1a, Deecke 1888 Taf.II, CIE 8350); Zvetaieff IIM
tab.IX.1.
265
Falerii belonged to the Horatia (§2.6.2); until the Social War the Poplilia comprised only
Anagnia and the Hernici (Taylor 1960:49-52). If the citizenship was an individual grant, one
would expect one of the South Etrurian tribes (Arnensis, Sabatina, Stellatina, or Tromentina).
504
228-229. The titulus prior (letters 14-17 cm high) was painted on plaster across the
backs of four tiles (65u47.5 cm, 65.5u47 cm, 64u47.5 cm, 64u44 cm).
1
[------2-----3----4---]
1
[---]ṃ2arc3io
The first line is completely covered by the plaster on which is written the titulus
posterior. The first legible letter in the second line is , probably an m (cf. marcia
227): there is no reason to read larcio (Deecke 1888, Conway, Herbig, Pisani, G.
Giacomelli).266
The titulus prior was later covered with plaster on which was painted the titulus
posterior in red (letters 15-19 cm high).
cesu2lati3perili4atef
Sinistroverse, Faliscan alphabet with cursive e (). The c is reversed to indicate a
woman’s name (§11.2.4). Tiperilia renders Latin Tiberilia in Faliscan alphabet: the
Faliscan form would have been *Tiferilia (cf. tif MLF 459?). The f, , was read as a
ligature fe(lia) or fe(ilia) by Garrucci (1864), but the ‘sidebars’ appear to be nothing
more than random slips of the brush.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20803+20796+
20790+20806). Bibliography: (I) Orioli 1854:XXI ¢4². – (II) Garrucci 1860:276-7 ¢13² (autopsy);
Detlefsen 1861a; Detlefsen 1861b (autopsy). Drawing: Garrucci 1860 tav.H.13. – (II) Garrucci
1864:64-5 (autopsy); Fabretti CII 2445; Garrucci SIL 798; Zvetaieff IIM 56 (autopsy); Zvetaieff III 58;
Schneider 1886:106 ¢18a-b²; Bormann CIL XI.3159,6a-b; Deecke 1888:173-4 ¢44²; Conway 1897:378
¢329²; Jacobsohn 1910:5 ¢31²; Herbig CIE 8349; Nogara in Herbig CIE 8349 add. (autopsy);
Buonamici 1913:66 ¢27²; Vetter 1953:312 ¢322f²; G. Giacomelli 1963:94-6 ¢121,VIa-b²; Peruzzi
1963b:443; Pisani 1964:338-40 ¢143I²; G. Giacomelli 1978:534-5 ¢12,VIa-b². Drawings: Garrucci
1864 tav.V.1 (reproduced in CII tab.XLIII, IIM tab.VII.11a, Deecke 1888 Taf. II, CIE 8349); Zvetaieff
IIM tab.VIII.11 (reproduced in CIE 8349).
230. Painted in red on plaster along the back of two tiles, the second known only
through Guidi’s apograph (the other one measures 66u46.5 cm; letters 9-12 cm high).
1
lclipị2aṛ[3?][io]
Sinistroverse, Faliscan alphabet. The l is . The p is followed by the lower half of a
shaft at the edge of the tile: in spite of Garrucci’s drawings, nothing more can be read or
restored on this tile, as Zvetaieff’s drawing shows. After this, Garrucci restored an a
from Guidi’s apograph; in view of clipeaṛ[io] LtF 231 and cl[i]peario LtF 232 and
clipea‹r›io MF 470* I wonder whether a second shaft might have been overlooked, in
which case clipeaṛ[io with a cursive e would be possible (thus Vetter). Following the r,
Garrucci restored a vertical shaft from Guidi’s apograph. The early editors read clipiai:
Deecke was the first to read clipiar ‘Clipear(ius)’, a solution hinted at already by
266
Pisani’s description shows that he was unaware that the line belongs to a titulus prior.
505
CHAPTER 15
Detlefsen (1861b). Herbig read ḷcḷipịp ̣*(--), which suggests that letters were missing
after the ones restored from Guidi’s apograph.
phy: (I) Orioli 1854:XXII ¢7². – (II) Garrucci 1860:274 ¢10² (autopsy); Detlefsen 1861a; – (II) Gar-
rucci 1864:67-8 (new autopsy); Fabretti CII 2448; Garrucci SIL 805; Zvetaieff IIM 63 (autopsy);
Zvetaieff III 65; Schneider 1886:106 ¢25²; Deecke 1886:201-2; Bormann CIL XI.3159,3; Deecke
1888:1756 ¢46²; Conway 1897:379 ¢331²; Herbig CIE 8351; Nogara in Herbig CIE 8351 add. (au-
topsy); Vetter 1953:312 ¢322h²; G. Giacomelli 1963:94-6 ¢121,VIII²; Peruzzi 1963b:435; G. Giacomelli
1978:534-5 ¢12,VIII². Drawings: Garrucci 1860 tav.H.10 (reproduced in Garrucci tav.1864 VI.4, CII
tab.XLIII, IIM tab.X.2, Deecke 1888 Taf.II, CIE 8351); Zvetaieff IIM tab.X.2; Nogara in CIE 8351
add.
231. Painted on plaster along the backs of two tiles and across a half-tile. Nogara, in
1912, found only the first two tiles, the second of which was broken in two; I was able
to find only a half-tile whose form corresponds to that of the half-tile of this inscription
(68.5 u max. 41 cm), but which no longer contains any legible letters. I present this
inscription in two ways, as I think the text as presented in previous editions is based on
an erroneous rearrangement of the tiles in antiquity.
Arrangement of the tiles as found, as in Garrucci’s apograph:

tile A tile C tile D (=B?)
––––––––––––––––––––––––––––
1 2
cclipeaṛ heic 3plenesqf
1 2
mfharịṣp ̣ cubat 3
1 2 3
sorex q*(*)ẹ
Suggested original arrangement of the tiles before the interment of Plenes:
tile A tile B tile C
––––––––––––––––––––––––––––
1 2 3
cclipeaṛ [io] heic
1 2 3
mfharịṣp ̣ [exce] cubat
1
sorex q*(*)ẹ 2[---] 3
Dextroverse, Latin alphabet. The third tile, which is painted in a different way and in a
different hand, is clearly a later addition that should be read separately: it is useless to
try to read all three tiles as one text, as did many of the earliest editors.
The arrangement of the tiles in all editions except mine follows Guidi’s apograph,
which depicted the tiles in situ. I believe that it is necessary to assume that this was not
the original arrangement, but that the inscription originally consisted of a tile (A), a
half-tile (B), and another tile (C), with an inscription pertaining to C. Clipearius alone.
When these tiles were removed for the interment of Plenes, the first tile (A) and the
506
third (C) were replaced as the first and the second of ‘new’ inscription, and a half-tile
with the epitaph of Plenes was added at the end (D): this may have been same as the
half-tile from the original inscription (B). Irrespective of whether my suggestion is
adopted, the inscription occupied three lines on two (or one) tiles, and was then
continued by two lines on another: a similar arrangement as in LF 243 and LF 247.
The first line on the first tile consists of cclipea (with ), followed by the lower
half of a shaft. The earliest editors read a dative or a nominative plural clipeai. This has
rightly been abandoned in favour of clipear[io] (thus first Herbig), but in that case a tile
must be missing between the first and the second tile, as Garrucci’s drawings show that
it is impossible to restore [io] either on the first or on the second tile.
The second line is mfhara, followed by traces of what are apparently two letters,
given by Garrucci as (Zvetaieff’s and Nogara’s drawings show only vague traces at
the point of te second letter). After this, Garrucci restored an a from Guidi’s apograph,
which in his drawing he places between the first and the second tile; Orioli placed it on
the second tile (acubat), which is impossible. The earliest editors read harai, but
already Garrucci (1864) read haracna. The interpretations of this were mainly based on
the interpretation of the erroneous clipeai in the first line. If we read clipeaṛ[io],
harac ̣ṇa can only be a second gentilicium, but this appears to be indicative of the
libertus (Rix 1963:376-8) and would be incompatible with the fact that the man was
cen]|sor (as I read the sor in the third line).267 Editors pointed to harisp in LtF 232:
Deecke (1886) interpreted haracna as hara-gnā ‘haruspex’, and Nogara in fact read
haraṣp ̣ ‘harasp(ex)’, which has since then become the accepted reading. Since there are
no parallels for the abbreviated honores in (Latino-)Faliscan inscriptions apart from the
q in Lat 237-238, it is preferable to read harasp[ex, which, again, requires a second tile.
This assumption has interesting repercussions for the third line. This starts with
sorex, which Garrucci at first interpreted as a cognomen Sorex (cf. his Locerta in LF
221 and Sus in LF 227). Detlefsen, however, extended Garrucci’s interpretation of
harisp | sor in LtF 232 as ‘harisp(ex) Sor(rinus)’ to this inscription, comparing the
augur Soranus mentioned by Cicero (Div. 1.47.105). Deecke (1886) connected sorex to
the name of Soracte, which was also known as Sorax (Porph. in Hor. Carm. 1.9.1-2)
and its priesthood of the Hirpi Sorani (for which cf. §2.3.4). The very doubtfully
attested (to say the least) Faliscan sorex now became a standard ingredient of the
Faliscan vocabulary: so much so that Peruzzi even proposed an etymology for it that
was not connected to the one on which the word was based in the first place (namely
*/sor-ag-s/ ‘lot-shaker, cleromantis’, adducing Liv. 22.1.11 for Faliscan cleromancy). If
the existence of an original second tile is assumed, however, the second line could
easily be read as cen]|so rex. Instead of a hypothetical sorex parallelled only by the
267
I have considered reading a woman’s name h (?) arati[a] (cf. MLF 348-351). The order,
man’s name – woman’s name – man’s cursus honorum, would be paralleled by LF 249.
507
CHAPTER 15
fragmentary sor in LtF 232, we would then have cen]|so, an office attested without
ambiguities in censo in LtF 232, and a rex as in MF 88 and perhaps in MF 249 (re[x]).
The letters following sorex were given by Garrucci first (1860) as , but later (1864)
as . The q is often interpreted as q(uaestor) (as in Lat 237-238), but what follows
is unclear: Vetter read c(ensor) VII, but it is unlikely that a man could be censor seven
times unless the office was quite different from its Roman counterpart (Peruzzi).
phy: (I) Orioli 1854:XXI ¢1². – (II) Garrucci 1860:277-9 ¢14² (autopsy); Detlefsen 1861a; Detlefsen
1861b (autopsy); Ritschl 1862:96-8; Mommsen & Henzen CIL I.1311; Garrucci 1864:67-8 (autopsy);
Fabretti CII 2442; Garrucci SIL 806; Zvetaieff IIM 64 (autopsy); Zvetaieff III 66; Schneider 1886:107
¢26²; Deecke 1886:201-2; Bormann CIL 11.3159; Deecke 1888:191-2 ¢61²; Conway 1897:379 ¢333²;
Herbig CIE 8352; Nogara in Herbig CIE 8352 add.; Buonamici 1913:67-9 ¢29²; Lommatzsch CIL
I2.1988-9; Vetter 1953:312-3 ¢322i²; Safarewicz 1955:186 ¢2²; G. Giacomelli 1963:94-6 ¢121,IX²;
Peruzzi 1963b:435-40; G. Giacomelli 1978:534-5 ¢12,IX². Drawings: Garrucci 1860 tav.H.14 (repro-
duced in Ritschl 1862:97-8 fig.L); Garrucci 1864 tav.VI.5 (reproduced in CII tab.XLIII, IIM tab.X.3a,
Deecke 1888 Taf.III, CIE 8352); Zvetaieff IIM tab.X.3 (reproduced in CIE 8352); Nogara in CIE 8352
add.
From these inscriptions, the following family tree has been reconstructed (reading the
first line of LF 224 as ca  uecineo[  ]uolti  rather than as ca  uecineo[  ]uoltiọ):
Voltius Vicinius Iunae f. Maxumus
Vibia Sertinia l. u Gavia Aconia
Ň ňņņņņņņņ┴ņņņņņņņņņņņ┐
Marcius Acarcelinius u Gavia Vicinia Gavius Vicinius
Ň u Gavia Mania
Titus Acarcelinius
u Publia Petrones Caesii f.
The pola marcia in LF 227 and the ṃarcio in LF 228 are sometimes grafted onto this
stemma as descendants of Marcius Acarcelinius, but the patronym from Marcius would
be *Marcilius, not Marcius (§7.5.2.2). Both may have belonged to a gens Marcia, or
may have been children of a Marcus of whom we have no record. Perhaps this Marcus
was the father of the slave-born Marcius Acarcelinius, who would then have received
the patronym Marcius as his praenomen: note that a praenomen Marcius is not attested
elsewhere. Peruzzi (1963b:441-6) pointed out how the stemma illustrates the social
climbing of the Acarcelinii: Marcius, the son of a freedwoman, perhaps born while she
was still a slave, since no mention is made of his father, marries the daughter of a pater
familias whose imposing nomenclature and cognomen Maxumus clearly mark him as
special. Their son marries a daughter of the Petronii, a gentilicium repeatedly encoun-
tered in the names of magistrates in inscriptions from Central Italy. The gentilicium
Acarcelinius is a new formation, and Peruzzi suggested that it was derived from a
toponym; A. Mancini (1981) derived it from an */akarkelom/ that would be equivalent
to (but not necessarily identical with) the */okrikelom/ reflected by Latin Ocriculum
and Umbrian */okriʃlom/ implied by Etruscan ucrislane Cl 1.2609, 2611-2613 etc.
508
The family relations of the L. Clipearius in LF 230 and the C. Clipearius in LtF
233 cannot be reconstructed. Peruzzi (1963b:435) claims that the L. Clipearius in LF
230 died before the C. Clipearius M. f. in LtF 231, because the epitaph of the former is
written in Faliscan and that of the latter in Latin alphabet, but the use of the two
alphabets is not so strictly chronologically distributed that it can be used as an argument
in this way. Deecke (1888:177) in fact used the argument in the reversed direction, as
an illustration of how, over the generations, the originally Latin Clipearii became
successively more Faliscanized. The family appears to have been Faliscan, however: an
older member of the gens, oufilo  clipeaio  letei  fileo, now appears in MF 470*.
232-233. The following inscriptions were from a tomb adjacent to the one that yielded
215-231. The titulus posterior, 233, was painted in red on plaster along the back of two
tiles. Where the plaster has crumbled away, traces of several previous inscriptions can
be read, painted directly on the tile surface. These tituli priores are presented here as
232. Nogara, in 1912, found both tiles, broken but still complete: I was unable to find
the bottom left corner of the first extant part (38u37 cm) and the right part of the second
tile (extant part 50u48 cm).
Fig.15.1. Garrucci’s drawing of LF/LtF 232 and LF 233.

(From Herbig CIE 8353)
2
[?--------] [?---------]
2
harisp anco ma
2
sor * censo
mcḷ[i] 2peariom[f]
[----------2------------]
[----------2-------]or [
The titulus posterior, 233, was written dextroverse in the Latin alphabet with cursive a
() and e (). The last word ]or may be quaest]or, praet]or, cens]or, or ux]or. Nothing
now remains of this inscription, the plaster on which it was written having crumbled
away apparently already in 1864 (“Ho cercato di sotto all’intonico sul quale si legge
MClipeario, se si fosse un anterior leggenda, ma di essa ivi non vi ha vestigio alcuno”,
509
CHAPTER 15
Garrucci 1864:62) and certainly by 1880, when Zvetaieff drew the tiles. The tituli
priores and the titulus postumus were read together by the earliers editors (thus still
Deecke); the first to separate them clearly is Conway. Whether the tituli priores belong
together is a different question.
The tituli priores on the first tile are dextroverse in the Latin alphabet with a
curious h, , which has no parallels in other inscriptions from the area. Harisp is
written against the edge of the tile: reading harisp[ex] would require another tile. The
next line contains only sor (probably cen]|sor, as in LtF 231: reading this would also
require a second tile): Garrucci (1864) also read two shafts near the edge of the tile,
which have been interpreted as a cursive e or as a numeral II (Vetter, who took them
together with censo on the second tile), but neither Zvetaieff nor Nogara saw these
shafts, and from what remains of this part of the tile it appears that they are drip-marks.
Of the tituli priores on the second tile, anco and ma are sinistroverse, ma being written
lower and smaller (the early editors erroneously read dextroverse umbr), while censo is
dextroverse. The disappearance of the plaster that originally covered the space after the
last letter shows that censo, not censor, is to be read. In my view, harisp and sor belong
to the same text, while censo, anco, and ma belong to (three?) different inscriptions.
From autopsy in the Museo Gregoriano Etrusco Profano, Città del Vaticano (inv. 20815+ 20813).
Bibliography: (I) Orioli 1854:XXII ¢9². – (II) Garrucci 1860:277-9 (autopsy); Detlefsen 1861a; Ritschl
1862:96-8; Mommsen & Henzen CIL I.1312. Drawing: Garrucci 1860 tav.H.15 (reproduced in Ritschl
1862:97-8 fig.M). – (III) Garrucci 1864:63 (autopsy); Fabretti CII 2450; Garrucci SIL 807; Zvetaieff
IIM 65 (autopsy); Zvetaieff III 67; Schneider 1886:106-7 ¢27²; Deecke 1886:201-2; Bormann CIL
XI.3518; Deecke 1888:190-1 ¢60²; Conway 1897:379 ¢332²; Herbig CIE 8353; Nogara in Herbig CIE
8353 add. (autopsy); Buonamici 1913:69 ¢30²; Lommatzsch CIL I2.1988-9; Vetter 1953:313-4
¢323,A-B²; Safarewicz 1955:185 ¢1²; G. Giacomelli 1963:94-6 ¢122²; Peruzzi 1963b:436-40, 443.
Drawings: Garrucci 1864 tav.IV.3 (reproduced in CII tab.XLIII, IIM tab. VIII.10a, Deecke 1888
Taf.III, CIE 8353); Zvetaieff IIM tab.VIII.10 (reproduced in CIE 8353).
234-241 (‘tomba degli Spurilii’). In 1891, several inscriptions in Faliscan and Latin
alphabet were discovered in one of three third- or second-century tombs in Contrada
Regoletti, c.1 km to the south of S. Maria di Falleri. Gamurrini’s apographs contain
some peculiar letter forms, which may at least partly be due to the severe damage
sustained by the plaster on which the inscriptions were painted (cf. Gamurrini 1891:49).
Vetter (1953:310) in fact called the apographs useless.
234. Painted in red on plaster on two tiles.
| | |
[---?]2*a*ḳit*3ue*af
Sinistroverse, Faliscan alphabet. The k is unexpected: assuming that it represents /g/
(§11.2.4.2) does not provide a better text. A tile is probably missing at the beginning.
Bibliography: Gamurrini 1891 ¢1² (autopsy); Herbig CIE 8335; Bormann CIL XI.7500a; G. Giacomelli
1963:93-4 ¢120,I². Transcription: Gamurrini 1891:49 (reproduced in CIE 8335, CIL XI.7500a).
510
235. Painted in red on plaster on two tiles.

 | |
1
ceịsịa*e2[0-4?]i*iace
Sinistroverse, Faliscan alphabet with cursive e (). Gamurrini’s  on the place of the
third and the fifth letters are perhaps to be read as i’s. The size of the lacuna is not
indicated, but can apparently be anything between zero and four letters. The second
sign after the lacuna is perhaps l or n. According to Gamurrini, no tiles are missing.
Bibliography: Gamurrini 1891 ¢2² (autopsy); Herbig CIE 8336; Bormann CIL XI.7500b; G. Giacomelli
1963:93-4 ¢120,II². Transcription: Gamurrini 1891:49 (reproduced in CIE 8336, CIL XI.7500b).
236. Painted in red on two tiles.

ᛧ| |
1
cạu2[---]ịsi
Sinistroverse, Faliscan alphabet? The gentilicium is ru[---] or au[---]. The size of the
lacuna is not indicated. The last word is ca]ịsi, ce]ịsi, or c]ẹsi (with a cursive e).
Bibliography: Gamurrini 1891 ¢3² (autopsy); Herbig CIE 8337; Bormann CIL XI.7500c (autopsy); G.
Giacomelli 1963:93-4 ¢120,III². Transcription: Gamurrini 1891:49 (reproduced in CIE 8337, CIL
XI.7500c).
237. Painted over two tiles “con belle e grandi lettere” (Gamurrini 1891:49)
1
mspu2riliuscf
1
bisq[]2duouir
Dextroverse, Latin alphabet. For the cursus honorum, cf. §2.3.3.
Bibliography: Gamurrini 1891 ¢4² (autopsy); Bormann CIL XI.7501 (autopsy); Vetter 1953:310.
Transcriptions: Gamurrini 1891:49; Bormann CIL XI.7501.
238. Painted on one tile is

cspurilius
mfiiiiqii
Dextroverse, Latin alphabet. Gamurrini’s transcription (but not Bormann’s) shows the
second i of spurilius as small and written within the angle of the l. The second line was
read by Gamurrini as ‘Marci filius et quaestor iterum’, but a filiation and a honor are
hardly connected by et. Vetter read III [sic] q(uaestor) II [uir] (which requires a third
tile). According to Bormann, the i’s are drip-marks rather than letters.
Bibliography: Gamurrini 1891:49 ¢5² (autopsy); Bormann CIL XI.7502 (autopsy); Vetter 1953:310.
Transcriptions: Gamurrini 1891:49; Bormann CIL XI.7502.
239. Painted in red on a tile.

mpani[---]
Dextroverse, Latin alphabet. The gentilicium may have been Panicius (cf. Schulze
511
CHAPTER 15
1904:203), or pani[---] may be a misreading for paui[---], cf. pauiceo 12, paui[cio 290.
Bibliography: Gamurrini 1891:49 ¢6² (autopsy); Bormann CIL XI.7503 (autopsy). Transcriptions:
Gamurrini 1891:49; Bormann CIL XI.7503.
240. Painted in red on a tile (the second line in smaller letters according to Bormann).
[---]prae[tor ?---]
[---? duum]uir[?---]
Dextroverse, Latin alphabet. The second line, given as ИR/// by Gamurrini, was read as
uir by Bormann. For the cursus honorum, cf. §2.3.3.
Bibliography: Gamurrini 1891:50 ¢7² (autopsy); Bormann CIL XI.7504 (autopsy). Transcriptions:
Gamurrrini 1891:50; Bormann CIL XI.7504.
241. Scratched on a small Etrusco-Campanian cup.

up
Dextroverse, with .
Bibliography: Gamurrini 1891:50; Herbig CIE 8338; G. Giacomelli 1963:62 ¢41,I². Transcription:
Gamurrini 1891:50 (reproduced in CIE 8338).
242-249 (‘tomba dei Protacii’ and ‘tomba dei Tertinei’). The following inscriptions
are from two tombs in the necropolis at località Pratoro, to the southeast of S. Maria di
Falleri (see Renzetti Marra 1990:328). The first was discovered in April 1971, when it
was blown open with dynamite by clandestine excavators: it yielded 72 tiles, 44 of
which contained inscriptions. From this tomb, the ‘tomba dei Protaci’, are LF 242-246.
A second tomb, discovered in September 1973, when it was likewise opened by
clandestine excavators, appears to have contained c.35 tiles, inscribed in the Faliscan
and the Latin alphabet: five of these were stolen on the day the tomb was discovered.
From this second tomb, the ‘tomba dei Tertinei’, are LF 247-249.268 The publication of
the material from these tombs has to my knowledge not progressed beyond the
inscriptions presented here.
242. Painted in red on plaster across the backs of four tiles (each c.58u47 cm; letters
c.7,5 cm high, but those of the third line are larger, c.7.5-12 cm).
1
[.]apr2otacio3mfmạ4cistratu
1
kesetcụ2estodpi3pretod4pis
1
cau2iaux3oaf 4
Sinistroverse, Faliscan alphabet with reversed s. The praenomen can be restored as the
268
Reputedly from the same tomb (Renzetti Marra 1990:329) is a much later Latin inscription,
which I quote here from AÉ 1982 (no.292): D(is) M(anibus) s(acrum). | Pisiuanae (mulieris)
l(ibertae) Vitalini, | uix(it) annis XXX m(ensibus) II, C(aius) Luta|tius C(aii) f(ilius) Hor(atia
tribu) Velitius Ianu|arius, Pisiuana (mulieris) l(iberta) Felicula | patrona, C(aius) Caesonius
C(aii) f(ilius) | Priscus uir coniugi dulcissimae fec(erunt).
512
very frequent [c]a or the less frequent [l]a (probably not as [m]a, as Marcus is here
abbreviated as m). Keset shows the use of k to represent /g/ (§11.2.4.2). The formula ma
cistratu | keset corresponds to the Latin expression magistratum gerere (cf. TLL 6.2
1939.1-1940.56): Renzetti Marra notes that this need not imply that it was taken over
from Latin (and certainly not that the words or the perfect formation of keset did not
pre-exist in Faliscan): on the other hand, the lexicon pertaining to officials and magis-
trates is virtually completely Latin (§6.3.6). The -d in cụestod and pretod is found also
in pretod LF 214 and probably renders a weakened /-r/, an alveolar flap [ ]: see §3.5.7b.
The third line, which is slightly larger, may have been added later: it provides the only
instance of a woman being described as uxor instead of as HUSBANDGEN uxor, since the
husband had already been named in the other Faliscan inscriptions, HUSBANDGEN WIFE is
used only when the woman was not buried together with her husband: see §7.4.2.
Bibliography: Renzetti Marra 1990:333-4, 339-40 ¢P Iabcd² (autopsy); Mancini 2002:35-42. Drawing:
Renzetti Marra 1990:334.
243. Painted in red on plaster across the back of a damaged tile (63u43 cm; letters c.8
cm high).
1-2 3
]kese[ probably: [..protac] kese4[tduum]
1-2 3
]uirup[ [io..fma] uirup4[reto]
1-2 3
]rcues[ [cistratu] rcues4[tor]
Sinistroverse, Faliscan alphabet with reversed s. The r has the form , which is
normally used for a (§11.2.4.2). The word kese[t and the elements of the cursus
honorum show that the text was similar to that of LF 242, but with the honores arranged
in decreasing order of importance. The arrangement of the text is peculiar. The position
of kese[t in the first line precludes the restoration [.protacio.fmacistratu]kese[t, as
this would require an inscription of five or six tiles, which would be too large to fit the
loculus, and too large for even the most extensive cursus to fill the other lines: compari-
son with the other cursus (see §2.3.3) shows that before duum]uiru one can expect at
most rex and censor (and perhaps a priesthood, as in LtF 231), and between duum]uiru
and cues[tor only pretor, while some of the honores may have been followed by a
numeral or words like pis. Assuming a reversed formula kese[t macistratu leaves
similar gaps in the cursus, whether a tile is assumed to be missing at the end or not:
[.protacio.f]kese[tmacis] or [.protacio.f]kese[t]
[tratu.....duum]uirup[reto] [macistratuduum]uiru*
[---------------]rcues[tor]) [---------------]rcues
[tor]
It would therefore appear that the inscription ran for three lines across two tiles and then
for another three across two other ones. I have assumed similar arrangements in LF 247
and LtF 231. For the cursus honorum, see §2.3.3.
Bibliography: Renzetti Marra 1990:333-4,340 ¢P 9² (autopsy). Drawing: Renzetti Marra 1990:333.
513
CHAPTER 15
244. Painted in red on plaster along the length of the back of a tile (67u47 cm; letters
c.12 cm high).
[---p]rotacio[---]
[---o]stro-pro[---]
Sinistroverse, Faliscan alphabet with reversed s. The use of - as an interpunct is unique.
For o]stro cf. ]ostr[o in LF 245. Renzetti Marra reads po]stro . pro[ with a po]stro
‘dietro’, referring to a double interment, followed by “un pro[ elemento architettonico
di cui è data la localizzazione”. I doubt whether this is possible. The syncopation in
po]stro is not a regular in Faliscan phonology (§3.6.6.2): also, pro[---] would appar-
ently refer to (the place in) the loculus, while placing a body before another is described
by anteponat in Lat 251. There appears to be no Latin word that fits the text except
no]stro or uo]stro;269 the ocurrence of ostr[o after the cursus honorum in LF 245 shows
that it is not a cognomen. Pro[---] may be part of a woman’s name, e.g. pro[tacia].
Bibliography: Renzetti Marra 1990:338 ¢P 12² (autopsy). Drawing: Renzetti Marra 1990:338.
245. Painted in red on plaster on a tile fragment (max. 30u26 cm; letters c.8-9 cm high).
[-----]*[---]
[---c]ues[tor---]
[---]ostr[o---]
Sinistroverse, Faliscan alphabet with reversed s. For the cursus honorum, see §2.3.3.
For the interpretation of [---]ostr[o---], see LF 244.
Bibliography: Renzetti Marra 1990:338 ¢P 15² (autopsy). Drawing: Renzetti Marra 1990:338.
246. Renzetti Marra mentions an “Ekn = Egnat[ testimoniata nella tegola P 19”
(1990:339), which provides another instance of the Late Faliscan use of k to render /g/
(cf. §11.2.5.2).
Bibliography: Renzetti Marra 1990:339 ¢P 19² (autopsy).
247. Painted in red on plaster across the back of a tile (68u47 cm; letters 13 cm high).
[1]
[---]2fcue[3][stor]
[1]
[---]2pret[3][or]
[1]
[---]2duum[3][uir]
[1]
[---]2*ar*[3][---]
Sinistroverse, Faliscan alphabet. The arrangement of the cursus honorum seems to
indicate that the man’s praenomen and gentilicium followed by the filiation, occupied
three or four lines on a missing tile, after which the text continued with three more lines
on two more tiles. I have assumed a similar arrangement for LtF 231 and LF 242. The
last line may have contained the name of a woman. For the cursus honorum, see §2.3.3.
269
The only candidates from a formal point of view are austrum/ostrum, claustrum/clostrum,
plaustrum/plostrum, and rostrum.
514
Bibliography: Renzetti Marra 1990:332-4 ¢T 20² (autopsy). Drawing: Renzetti Marra 1990:332.
248. Painted in red on plaster across the back of a tile (70u48 cm; letters c.8-9 cm, those
of the third and fourth line being smaller).
[--------]er[---]
[---]o[]spuṛ[ilio]
[.fpre]toriiduu[muiru]
[---------]****[---------]
Sinistroverse, Faliscan alphabet with reversed s and cursive e. The contents of the first
line are unclear: the main epitaph seems to have started in the second line, which is
slightly larger than the others. It contained the name of a man [---]o[ ]spuṛ[ilio with a
gentilicium occurring also in Lat 237-238. The size of the letters suggests that if only
one tile is missing at the end, the filiation must have stood at the beginning of the third
line. The cursus honorum then consisted only of pre]tor  ii and duu[muiru]: see §2.3.3.
Bibliography: (autopsy) Renzetti Marra 1990:332 ¢T VIb². Drawing: Renzetti Marra 1990:332.
249. Painted in red on plaster across the back of a tile (58u55 cm; letters c.13-14 cm
high, but those of the third line only c.6 cm).
[---]*ocf
[---]*liacf
[duu]ṃuirurẹ[x]
Sinistroverse, Faliscan alphabet with cursive e (). The arrangement of the text, with
apparently a man’s name in the first line, a woman’s name in the second, and a cursus
honorum in the third, is unique. Of great importance is the word rẹ[x (of the e only the
upper parts of the shafts are preserved), which has also been read in MF 90, and can in
my view also be read in LtF 231. For the cursus honorum, see §2.3.3.
Bibliography: Renzetti Marra 1990:333 ¢T 22² (autopsy). Drawing: Renzetti Marra 1990:333.
250. Also from Pratoro (‘tenuta De Ferro’) is a Latin inscription painted on a tile. AÉ
1982 (no.286) gives the text as Pu(blius) Fuluius C(aii) f(ilius) | C(aii) n(epos) Suto(r)
diem | suo obiet a(nte) d(iem) X k(alendas) Dec(embres), | C(aio) Atilio (et) Q(uinto)
Seruio (pour Seruilio?) co(n)s(ulibus). The omission of -r in Suto and -m in suo, and the
e in obiet are in accordance with Faliscan as well as with ‘rustic Latin’. Seruio for
seruilio is an error rather than a palatalization /l/ o // (as Renzetti Marra explains it):
see §3.5.5.3. The consuls are those of 106 BCE.
Bibliography: †Pulcini 1974:138; [Renzetti Marra in G. Giacomelli 1978:508] †Di Stefano Manzella
1981:156 ¢40²; Renzetti Marra 1990:327,329. Drawing: Renzetti Marra in †Pulcini 1974:138
.
251. Cut in the back wall of the portico of a tomb (height c.110 cm; letters 12-14 cm
high) along the ancient road between Falerii Novi and Falerii Veteres near the Fosso dei
Tre Camini, i.e. also from Pratoro. The inscription was detached and brought to the
515
CHAPTER 15
Museo di Villa Giulia somewhere between 1878 (when Bormann saw it in situ) and
1898 (when, according to Herbig, Pauli saw it in the museum).
lueciliouofet
po[l]aeabelese
lectuidatuṣ
[.]ueciliolfeṭplenese
lectuiampliusnihil
inuiteislcleuieislf
etqueieosparentaret
neanteponat
Dextroverse, Latin alphabet. The first four letters of amplius are written as ligatures
 and   . Henzen’s reading (used by Dennis and Noël des Vergers),270 was substan-
tially improved upon by autopsies by Garrucci (for which cf. also Michaelis) and
Ritschl. (Safarewicz erroneously has e[.] in the first line and polae in the second.). –
Lectus is apparently the terminus technicus for the loculus (§6.3.39): cf. lete 285, let
361, and perhaps l[.......] MF 17. Plenese is the dative of the Etruscoid feminine plenes
found in LtF 231 from near S. Maria di Falleri: see §9.2.2.2d. Abelese, on the other
hand, rather reprents Abelle(n)sei. Parentaret is parentarit = parentauerit. The phrase
amplius ... anteponat can only be interpreted as a prohibition against unauthorized
burials in the reserved loculi (Henzen, Garrucci 1860:280-1), although the syntax is
obscure due to the double negation nihil ... ne (for which cf. LHS II pp.802-5), and by
the absence of a subject for anteponat, probably to be solved by assuming an implicit
nemo as the subject. Warmington, the only editor who provides a translation, increased
the difficulties by reading Amplius nihil | inviteis L. C. Levieis L. f. | et quei eos
parentaret; | ne anteponat ‘Nothing further may be done against the wish of Lucius
Levius and Gaius Levius, sons of Lucius, and of him whose duty it is to make sacrifices
to their souls. Let no one place a body in front of these dead’. The language is (late)
second-century Latin without specifically Faliscan dialect features, except perhaps for
the omission of -s in lectu, which would fit Faliscan well (§3.5.7d).
Bibliography: [Dennis 1844 (autopsy)]; Henzen 1844 (autopsy); Dennis 1847:132 (autopsy); Garrucci
1860:279-81 ¢16²; Michaelis 1862:346* (autopsy); Ritschl 1862:52; Noël des Vergers 1864:144 n.1;
Garrucci 1864:73-4 (autopsy); Mommsen & Henzen CIL I.1313; Garrucci SIL 1880; Dennis
1878:99-100; Schneider 1886:107 ¢29²; Bormann CIL XI.3160 (autopsy); Deecke 1888:211-2 ¢81²;
†Lattes 1895:242; Conway 1897:381 ¢336²; Herbig CIE 8334 (autopsy); [Weege in Helbig 1913:381];
Buonamici 1913:84-5 ¢54²; Lommatzsch CIL I2.1990; Warmington 1940:44-5 ¢94²; Vetter 1953:310;
Safarewicz 1955:186 ¢4²; G. Giacomelli 1963:263-4 ¢XIII². Photograph: Herbig CIE 8334; FI II.1
p.397 fig.238. Drawing: Brunn in Ritschl 1862 tab.LIX,B.
270
The publication of this inscription earned the then still unknown Henzen the praise of Emil
Braun: “[Er hat] eine sehr schöne Arbeit gemacht und sehr unangenehme Schwierigkeiten über-
aus glücklich gelöst. Er wird ein sehr guter Epigraphist werden.” (letter dated 13.V.1844,
quoted in Kolbe 1979:535).
516
252. Scratched in a bucchero plate ( 9 cm) found “prope Falerios novos (strada
provinciale NOG.)” (Herbig CIE 8354).
la
Ductus and alphabet are not given.
Bibliography: Herbig CIE 8354; G. Giacomelli 1963:62 ¢41,II².
517
CHAPTER 15
518
Chapter 16
Corchiano and the northern ager Faliscus
16.1. The sites of the northern ager Faliscus
Of the sites in the northern ager Faliscus, Corchiano is the largest. In spite of its obvious
local importance, its history or indeed its ancient name is unknown (see §2.1.2). The
site at Corchiano seems to have been abandoned in the third century, in all likelihood as
a result of the war of 241 (Ward Perkins & Frederiksen 1957:116). The inscriptions
from Corchiano (§16.2, 253-301) are therefore categorically classed as Middle Faliscan
unless there is evidence pointing to a different period. A substantial number of inscrip-
tions are from the smaller settlements of the northern ager Faliscus, at Vignanello
(§16.3, 302-323), Fabbrica di Roma (§16.4, 324-328), Carbognano-Vallerano (§16.5,
329-337), the site at Grotta Porciosa (§16.6, 338-356), and the area near Gallese and
Borghetto (§16.7, 356-359). Contrary to the sites at Civita Castellana and Corchiano,
there are clear archaeological and epigraphical data that point to a continued habitation
after the war of 241-240. The inscriptions from the late tombs of the gens Velminaea at
Vignanello (MLF 305-323) and of gens Folcosia at Carbognano-Vallerano (LF
329-337), and the presence of inscriptions in Latin alphabet at Fabbrica di Roma (LtF
325-328) and the Grotta Porciosa site (LtF 340-345) imply that these sites survived into
period after 240. Vetter (1953:278) regarded most of the inscriptions from the smaller
sites as belonging to this later period: in my view, this is true of at least 19 out of the 61
inscriptions from the northern ager Faliscus (LF 328-337, and LtF 325-327 and
340-345). The ceding to Rome of half the ager Faliscus after the war of 241 BC will
certainly have had considerable effect on the language of the surviving sites in that area.
It cannot be established with which of the smaller sites was included in the area ceded
to Rome: it is likely that this comprised the Grotta Porciosa site with its strategic
location near the crossing of the Tiber, but perhaps not the sites at Vignanello and
Carbognano-Vallerano (§2.6.2).
16.2. Corchiano and surroundings
16.2.1. Corchiano. Corchiano is the largest site of the ager Faliscus after Civita
Castellana, and the main site of the north-western ager Faliscus. It has been named as a
possible candidate for Fescennium (§2.1.2). Like many South Etrurian sites, it is located
519
CHAPTER 16
on a plateau surrounded on three sides by steep gorges, in the case of Corchiano those
of the Rio Ritello and Rio delle Pastine to the north and the Rio Fratta to the south,
meeting on the east side of the town. On the western side the resulting plateau, known
as ‘Il Vallone’, was fortified with a trench and a wall (see Buglione di Monale 1887a
with map tav.III, Ward Perkins & Frederiksen 1957:111-8, and FI II.2 pp.214-9 with
map II.1 p.53). It seems to have been abandoned in the third century, probably a result
of the war of 241 (Ward Perkins & Frederiksen 1957:116). Most of the inscriptions
from the town itself were discovered during the excavations of the necropoles on the Il
Vallone plateau in the 1880s and 1890s; others have been discovered at various dates in
the environments of the town (cf. Bazzichelli 1885, Cozza 1886, Buglione di Monale
1887a, and FI II.2 pp.214-321). The inscriptions from Corchiano show a relatively large
number of Etruscan features (Cristofani 1988, Peruzzi 1964c, 1990: see §9.2.3.
16.2.2. The necropoles of the Il Vallone plateau. The tombs at Corchiano are spread
out over the Il Vallone plateau to the west of the site of the Faliscan town. The various
parts of what is in fact one large necropolis are usually referred to as the first necropolis
of Il Vallone, excavated in 1886 (FI II.2 pp.245-56), the second necropolis of Il Vallone
or Fondo Piergentili necropolis, excavated in 1887 (FI II.2 pp.257-71), part of which,
referred to as the third necropolis of Il Vallone, was excavated in 1893 or 1894 (FI II.2
pp.272-81), the first necropolis of S. Antonio or Fondo Marcucci necropolis, excavated
in 1886-1888 (FI II.2 pp.283-96), and the second necropolis of S. Antonio or ‘scavi
Perez’, excavated in 1892-1893 (FI II.2 pp.297-304).
253-255. Scratched on vessels from tomb 14 of the first necropolis of Il Vallone.

ae
an
ni
Sinistroverse, with cursive a () and e ().
Bibliography: Cozza 1886:155 (autopsy); Herbig CIE 8383c-d,g; FI II.2 p.253 (autopsy). Transcrip-
tion: Cozza 1886:155 (reproduced in CIE 8383c-d,g).
256. Stamped, together with a running horse, on a terracotta strigilis apparently from
tomb 18 of the first necropolis of Il Vallone (cf. FI).
vce
Dextroverse, Etruscan alphabet. The v is .
Bibliography: [Weege in Helbig 1913:379 (autopsy)]; twice in FI II.2 pp.254-5 (autopsy). Transcrip-
tions: FI II.2 p.254 (twice).
257-258. From tomb 11 of the second necropolis of Il Vallone.
520
THE INSCRIPTIONS FROM CORCHIANO AND THE NORTHERN AGER FALISCUS
257. Scratched along the length of a damaged tile (max. 46u42 cm; letters 2.5-3 cm
high).
aruzcesịearuto
Sinistroverse, Faliscan alphabet with reversed s. Aruz with -z represents [arrũ:s] rather
than [arrũ:ts]: for the use of -z, cf. §3.5.3. In cesịe, the first e is shown as . The
reading of the penultimate letter of this word, in Nogara’s drawing, is debated:
Herbig, Jacobsohn, and G. Giacomelli, with some hesitation, read cesve, but Vetter and
Cristofani rightly doubt the likelihood of a v in an inscription otherwise in Faliscan
alphabet. Vetter, assuming that the lower sidebar of this letter was accidental, proposed
cesp ̣e (i.e.,  instead of ); assuming that both sidebars are accidental ( instead of )
would give cesịe, for which cf. cesies or ceises MF 265 and ceises Etr XXXIV, both
from Corchiano. Herbig and Vetter considered joining this tile to MF 258. See §9.2.3a.
Bibliography: Herbig 1904b:518-9 ¢60² (autopsy); Herbig 1910:90-1 ¢10²; Jacobsohn 1910:6 ¢42²;
Herbig CIE 8392; Vetter 1953:318 ¢332²; G. Giacomelli 1963:103 ¢138²; Peruzzi 1964c; FI II.2 p.264
(autopsy); Cristofani 1988:18-9; Peruzzi 1990. Drawings: Nogara in Herbig 1904b Taf.IV (reproduced
in CIE 8392); FI II.2 p.264.
258. Scratched along the length of a tile271 (66u48 cm; letters only 3-4 cm high).
[1]
[---]2uenelịessapnonia
Sinistroverse, Faliscan alphabet with reversed s. The drawings show the e as  (cf. the
similar e in MF/Etr 64 and MLF 285), perhaps a slip. The last two letters are written
under the line in boustrophedon. According to Herbig (1904), the t, a  with a very
slight sidebar, could also be an i; all editors have read uenelṭes, however, usually
regarding this as a gentilicium derived from the Etruscan praenomen Venel. Vetter
(1953:443) hesitatingly interpreted this uenelṭes as a patronymic adjective, which may
not be impossible, although in that case the formation is non-Faliscan (§7.5.2). Reading
uenelịes on the other hand gives a form that could very well be a Faliscan patronymic
adjective or a gentilicium derived from Venel with the ‘Etruscoid’ ending -ies (§9.2.2).
Sapnonia is a clearly a woman’s gentilicium, although the name itself is not attested
elsewhere: Peruzzi suggests that it may be connected to Sabin-. See also §9.2.3a.
A tile is certainly missing at the beginning of the text. Herbig, Nogara (in Herbig
CIE), and Vetter considered the possibility that this might be MF 257, but in view of the
different shapes of the e, I hesitate to adopt this view; neither is it clear how the text
resulting from joining the two should be interpreted (‘Arruns Caesies son of Arruns
Veneltes/Venelies’? ‘Arruns Caesies son of Arruns son of Venel’?).
Bibliography: Herbig 1904b:519 ¢61² (autopsy); Herbig 1910:195 ¢36²; Jacobsohn 1910:6 ¢45²; Herbig
CIE 8393; Buonamici 1935:344; Vetter 1953:318 ¢333²; G. Giacomelli 1963:103 ¢139²; Peruzzi 1964c;
FI II.2 p.264 (autopsy); Cristofani 1988:18-9; Peruzzi 1990. Drawing: Nogara in Herbig 1904b Taf. IV
(reproduced in CIE 8393).
271
The FI author erroneously refers to the object as an ‘embrice’.
521
CHAPTER 16
259-260. Scratched inside272 two Campano-Etruscan cups (height 8 cm, 12 cm;

letters 3-4 mm high) from tomb 12 of the second necropolis of Il Vallone.
licinio
licinio
Sinistroverse, Faliscan alphabet. In 259 l is , in 260 it is .
Bibliography: Herbig CIE 8395-8396 (autopsy); Bormann CIL XI.8124,18; Lommatzsch CIL I2.2657;
Vetter 1953:324 ¢346²; Safarewicz 1955:185; G. Giacomelli 1963:64 ¢49,I-II²; FI II.2 p.266 (autopsy).
Drawing: Herbig CIE 8395-8396.
261-262. Scratched inside a plate (height 55 mm, 125 mm; letters 2.5 mm high) and
a black-varnished saucer (height 7 mm, 123 mm; letters 2 mm high) ascribed to the
second necropolis of Il Vallone by Herbig, but not mentioned in FI.
uli
uli
Sinistroverse, Faliscan (?) alphabet. Herbig read the text as u(e)li, which G. Giacomelli
interpreted as the genitive of *Velus, a thematized form of the Etruscan praenomen Vel
(cf. uelos in EF 4?): a genitive *u(e)li(s) is unlikely, as the consonant-stem genitive
ending is -o(s) (§4.5.2). Perhaps uli is a genitive or abbreviation of Vollius (uolḷia MF
47, uoll[---] MF 86) or Ullius (cf. Schulze 1904:426).
Bibliography: Herbig CIE 8394 and 8587 (autopsy); Vetter 1953:324 ¢345a-b²; G. Giacomelli 1963:64
¢48,I-II². Drawings: Nogara in CIE 8394 and 8587.
263-264. The following inscriptions are from tomb 22 (Benedetti’s tomb 3) of the ‘third
necropolis of Il Vallone’. From the same tomb is a mirror with aχle aivas Etr XXXIII.
Fig.16.1. Herbig’s tracing of MLF 263.

(From CIE 8415.)
263. Scratched in a small cup or saucer with a high, decorated rim (height 4 cm, rim
7 cm; letters 5-8 mm high).
cẹsitfere
272
Thus Herbig, from autopsy; Bormann and Lommatzsch place the inscription under the foot.
522
Sinistroverse, Faliscan alphabet. The second letter is either an e (Herbig CIE) or an a

(Thulin, Herbig 1910): it cannot be taken together with the s as an r (as does Danielsson
in Herbig CIE). The fifth letter is certainly a t (Thulin, Herbig 1910), not an f (Herbig
CIE). This renders impossible Herbig’s cẹsif ̣ : fere = fere cesi f(ileo) : (CIE, adopted by
Vetter and G. Giacomelli), a reading which is in any case improbable, as the words are
arranged in two-thirds of a circle with the empty one-third between fere and cesit. I
wonder whether cẹsit could not be a verb, cf. keset LF 242, kese[t LF 243, but I am at a
loss how this would make a coherent text.
From autopsy in the Museo di Villa Giulia, Rome (inv. 6395). Bibliography: Thulin 1907:270
(autopsy); Herbig 1910:193 ¢33² (autopsy); Herbig CIE 8399; Vetter 1953:325 ¢348²; G. Giacomelli
1963:62 ¢42²; FI II.2 p.275 (autopsy). Drawing: Herbig CIE 8399.
264. Scratched under a black-varnished saucer (height 4 cm, 12.8 cm; letters 6 mm
high) from a woman’s grave (Cristofani). First half of the fourth century.
hermana
Sinistroverse, Etruscan alphabet. The h is (reversed?) ; the a’s are . The text is
generally regarded as Etruscan on account of the Etruscan r, ; morphologically, it
could as well be Faliscan.
Bibliography: Herbig 1910:189 ¢28² (autopsy); Herbig CIE 8400; Vetter 1953:325 ¢348a²; G. Gia-
comelli 1963:62 ¢43²; FI II.2 p.275 (autopsy); Cristofani 1988:17,23 ¢12²; Rix ET Fa 2.13. Draw-
265-266. The following inscriptions are from tomb 7 (Benedetti’s tomb IX) of the first
necropolis of S. Antonio.
265. Painted on two tiles. Known only through apographs by Helbig and Manzielli, and
Nogara’s copy of the apograph in the Museo di Villa Giulia inventories.
popliacalitenes
arontocesies
lartiouxor
Sinistroverse, Faliscan alphabet. Helbig’s apograph has popia (thus Lignana, Conway,
Bormann, and Pisani) and ceises (thus Lignana); most editors read poplia and cesies on
the strength of the apographs by Manzielli and Nogara (ceisies Conway). According to
Helbig’s apograph, the first t is , but the others ; in Nogara’s, they are all . The
syntax of this text is awkward. The simplest solution is to take lartio as a unique
instance of a patronymic adjective within the formula HUSBANDGEN WIFE (thus Gamur-
rini in Lignana, Deecke, Bormann, Vetter, G. Giacomelli, and Peruzzi): see §7.4.2 and
§9.2.2.4. Buonamici’s alternative, in which Publia is the daughter of Arruns Calitenes
and wife of Lars or Lartio Caesius, requires two instances of postponed praenomina, for
which there are is only one Faliscan parallel, fasies  caisia MF 41.273 The problem,
273
Entirely improbable is Lignana’s ‘Poplia di Calitene, Arunzio di Cesio, Lartio e la moglie’.
523
CHAPTER 16
however, is that lartio is a nominative where a genitive would be expected: I very much
doubt that lartio is a genitive lartio(s) from a nominative larti, as Herbig, Vetter, and G.
Giacomelli interpreted it. Although it is not impossible, an Etruscan or Etruscoid
nominative larti would be expected to be declined as an i-stem, taking a genitive
larte(s) (§4.5.2) just like felic ̣inate(s) MF 42. Perhaps the nominative lartio was used by
mistake because the cesies that precedes it could be used both as nominative and as gen-
itive (Peruzzi 1964c:337, §9.2.2.4). Pisani’s idea that the patronymic adjective lartio
was used here instead of the genitive because it is an alternative to the genitive in
FILIATION is very perhaps possible, but not attractive: if cesies | lartio  uxor is to mean
‘the wife of Lars Caesius’, one would of course expect lartia and not lartio.
Bibliography: Lignana 1887a:199-201; Deecke 1888:185-7 ¢56²; Conway 1897:383 ¢345²; Herbig
1910:91-101 ¢12²; Jacobsohn 1910:6 ¢39²; Herbig CIE 8387; Buonamici 1913:75 ¢43²; Bormann CIL
XI.7513; Ribezzo 1930:98-9; Vetter 1953:318 ¢334²; G. Giacomelli 1963:101 ¢132²; Pisani 1964:337
¢143F²; Peruzzi 1964c:229-32; FI II.2 p.288 (autopsy); Cristofani 1988:17-8; Peruzzi 1990:282.
Transcriptions: Helbig in Lignana 1887:199 (reproduced in Deecke 1888 Taf. III, CIE 8387);274
Nogara in CIE 8387.
266. Scratched on a tile. Known only through apographs by Helbig and Manzielli.
uelturtetena
aruto
Sinistroverse, Faliscan alphabet. In Helbig’s apograph the t is ; in the others, it is is .
A is given as . Tetena may be the same name as Tettius in MF 41: see §7.8.1.12-153
and §9.2.3b,f.
Bibliography: Lignana 1887a:201; Deecke 1888:188 ¢57²; Conway 1897:527 ¢28*²; Herbig 1910:91
¢11²; Jacobsohn 1910:6 ¢38²; Herbig CIE 8388; Buonamici 1913:76 ¢44²; Vetter 1953:318 ¢335²; G.
Giacomelli 1963:101-2 ¢133²; Pisani 1964:336 ¢143C²; FI II.2 pp.288-9 (autopsy); Cristofani
1988:17-8; Peruzzi 1990:282. Transcriptions: Helbig in Lignana 1887:201; Manzielli in Deecke 1888
Taf. III (reproduced in CIE 8388).
267. Stamped on the handle of a bronze strigilis (length 23 cm) from tomb 13 of the
second necropolis of S. Antonio.
arθ[3-5]rẹ
Sinistroverse, Etruscan alphabet. The last letter, , is very reminiscent of the hitherto
unique e in Etr XXXIV, , from tomb 7 of the first necropolis of S. Antonio: see
§11.2.4. I wonder whether this inscription might be related to arnθial uṛ[4-5?] Etr LI,
also purported to be from Corchiano. If so, the text could be read as arθ[ial u]rẹ (and
Etr LI perhaps as arnθial uṛ[es mi?]). The text may therefore be Etruscan rather than
Faliscan.
Bibliography: FI II.2 p.300 (autopsy). Drawing: FI II.2 p.300.
274
Deecke’s edition is based on Manzielli’s apography, but his drawing on Lignana’s transcript.
524
268. Engraved on the handle of a bronze strigilis (length c.21 cm) from tomb 22 of the
second necropolis of S. Antonio. Fourth century.
medlouciliosfeced
Dextroverse, Latin alphabet. The unparallelled placing of med is probably due to
influence from the Etruscan model mini zinace ... (see §8.9.2, §9.4.1): med is certainly
not an abbreviated praenomen (as Lommatzsch and Bormann took it). The use of the
Latin alphabet at this date points to an import, as does the presence of -s (§3.5.7d) and
perhaps the use of ou (§3.7.2), although the position of med makes me doubt whether
Wachter (1987:369) is richt in stating that “die Inschrift [könnte] hinsichtlich der
sprachlichen Merkmale ohne weiteres aus Rom stammen”. This is now confirmed by
oufilo  clipeaio  letei  fileo  met  facet MF 470* (mid- or late fourth century) and
cavios frenaios faced MF 471* (late fourth or early third century), which show that in
contemporary Middle Faliscan faced was used rather than feced.
Bibliography: Kretschmer 1912 (autopsy); Meister 1916:96-7; Dessau 1916:32.CXLV ¢9444²; Della
Seta 1918:86 (autopsy); Lommatzsch CIL I2.2437; Bormann CIL XI.8130,1; Diehl 1930:77 ¢721²;
Lommatzsch CIL I2.2437 add.; Safarewicz 1955:186; Vetter 1953:327; Degrassi 1963:357 ¢1251²; G.
Giacomelli 1963:265-6 ¢XIX²; Coarelli in RMR p.311 ¢464² (autopsy); FI II.2 p.302 (autopsy); Agostin-
iani 1982:151 ¢594²; Wachter 1987:369. Photographs: Kretschmer 1912 Taf. between pp.104-5
(reproduced in CIL I2.2437); RMR tav.LXXV. Drawing: FI II.2 p.302.
269-271. The following inscriptions are from tomb 28 of the second necropolis of
S.Antonio.275
269. Scratched crudely across the front of a tile (57.5 u 50.5 cm; letters 5-12 cm high).
arutemacena
morenez
Sinistroverse, Faliscan alphabet. The first six letters are written notably larger, with the
line swerving upwards: the remaining letters were written smaller, as realization
dawned on the maker that he or she was not going to succeed in fitting all the letters of
macena onto the first line; the last a of macena is in fact written above the line. The first
m () and both n’s () are reversed. For the nominative arute, probably rather an
accusative arute(m) used as a nominative than a form with an ‘epenthetic e’, see
§9.2.2.1,4. Macena is ‘Ma(r)cena’ (Cristofani and Peruzzi), cf. mar||cna in 270. For the
use of -z in morenez, see §11.2.4, §9.2.2.1; for the feminine morenez, see §9.2.2.2c.
From autopsy in the Museo di Villa Giulia, Rome (inv. 9551); a small piece containing the top part of
the z is missing. Bibliography: Herbig 1910:89-90 ¢9² (autopsy); Jacobsohn 1910:6 ¢40²; Herbig CIE
8384 (autopsy); Buonamici 1913:75 ¢42²; Vetter 1953:316 ¢329²; G. Giacomelli 1963:100 ¢129²
(autopsy); FI II.2 p.303 (autopsy); Cristofani 1988:18; Peruzzi 1990:280-1; Rix ET Fa 1.1. Draw-
275
Dohrn gave the provenance of MF 271 erroneously as Falerii Veteres.
525
CHAPTER 16
270. Scratched across the front of a tile (58 u 57 cm; letters 7-14 cm high).
cnacitiai
larisemar
Sinistroverse, Faliscan alphabet. The interpuncts are odd: whereas the rest of the text is
scratched in sometimes rather jagged strokes, they appear to have been drilled into the
surface, resulting in neatly rounded holes. The s is very long and thin. The r at the end
of the second line, smaller than most of the letters of the first line, is raised and tilted
backwards, which Herbig and Vetter took as an indication that the first line should be
read after the second, which gives larise  mar||cna  citiai, a much better reading than
cna  citiai | larise  mar. Why the lines should be so arranged is entirely unclear: not
only is the arrangement unique in the Faliscan sepulchral inscriptions on tiles, but there
is more than enough room to write the letters cna  citiai underneath larise  mar rather
than above them. The resulting mar||cna would then be ‘Marc(e)na’ (Cristofani,
Peruzzi), cf. macena ‘Ma(r)cena’ in MF 269. Larise is in my view an accusative
larise(m) used as a nominative, rather than a form with an ‘epenthetic -e’ marking an
/s#/ that was realized more strongly than in Faliscan (Vetter, Peruzzi): see §9.2.2.4.
Citiai is either a dative (thus Herbig, Vetter, and G. Giacomelli) or a genitive: see
§8.10.2. It is not necessary to assume that it is an Etruscan feminine nominative (as do
Cristofani and Peruzzi): see §9.2.3c.276
1910:193-4 ¢34² (autopsy); Herbig CIE 8386; Vetter 1953:317 ¢331²; G. Giacomelli 1963:101 ¢131²; FI
II.2 p.303 (autopsy); Cristofani 1988:18; Peruzzi 1990:278-81; Rix ET Fa 1.2. Drawing: Herbig CIE
8386.
271. Scratched along the length of the front of a tile (64.5u47.5; letters 7-11 cm high).
poplia
zuconia
Sinistroverse, Faliscan alphabet. Zuconia is an adaptation of Etruscan zuχu, attested at
Corchiano in zuχus Etr XXXI and perhaps in zu[con]|eo MF 56 from Civita Castellana.
From autopsy in the Museo di Villa Giulia, Rome (inv. 9552). Bibliography: Herbig 1910:185-6 ¢23²
(autopsy); Jacobsohn 1910:6 ¢43²; Herbig CIE 8385; Vetter 1953:317 ¢330²; G. Giacomelli 1963:100
¢130² (autopsy); [Dohrn in Helbig/Speier 1969:674-5 ¢2752²]; FI II.2 p.303 (autopsy); Cristofani
1988:18; Peruzzi 1990:278-9. Photograph: G. Giacomelli 1963 tav.VII. Drawing: Herbig CIE 8385.
16.2.3. The Rio Fratta necropolis. The third-century Rio Fratta necropolis lies to the
east of Corchiano; it was excavated in 1911 (see Gabrici 1912c).
272. Scratched across the front of tile (68 u 47 cm; let. 3-8 cm) found in 1894 on the
site of the Rio Fratta necropolis.
276
Cristofani erroneously gives the praenomen as Cauie.
526
cauionomes
inamaxomo
zeruatronia
Sinistroverse, Faliscan alphabet with reversed s. Zeruatronia has a parallel in Se-
ruatronia in CIL X.8230 from Capua. For the use of z-, see §11.2.4. See §9.2.3c.
1910:189-90 ¢29² (autopsy); Jacobsohn 1910:5 ¢44²; Herbig CIE 8378 (autopsy); Buonamici 1913:73
¢38²; Bormann CIL XI.7515; Vetter 1953:316 ¢328²; G. Giacomelli 1963:98 ¢127²; Cristofani 1988:19.
Drawing: Herbig CIE 8378.
273-274. Scratched, 273 to the left and 274 to the right of a graffito of a human profile
inside a red-varnished plate (height 5 cm, 14 cm; letters 10 mm) from tomb 3.
cauituri
ct
Sinistroverse, Faliscan alphabet. In 273, t is ; in 274, it is . Caui  turi is genitive
(Vetter, G. Giacomelli) or abbreviated nominative (§8.8.1). The gentilicium occurs also
at Civita Castellana (turia MF 22-27, and probably also tur MF 44).
Bibliography: Herbig CIE 8592; Gabrici 1912c:82-3 (autopsy); Buffa NRIE 992; Vetter 1953:325
¢349²; G. Giacomelli 1963:64 ¢50². Drawings: Nogara in CIE 8592; Gabrici 1912c:83.
16.2.4. Corchiano, specific provenance unknown. Several inscriptions have (rightly

or wrongly) been ascribed to the town without further particulars of their discovery
being known.
275-276. Two tiles that may to have belonged to the graves of two brothers;277 the
gentilicium is attested also from a family tomb at Civita Castellana (MF 48-53).
275. Scratched along the front of a tile (71 u 48.5 cm; letters c.6 cm high).
cauiooufilio
uolteo
Sinistroverse, Faliscan alphabet. The letters are squarish; the o’s are .
(autopsy); Jacobsohn 1910:6 ¢41a²; Herbig CIE 8397; Vetter 1953:319 ¢337a²; G. Giacomelli 1963:104
¢140,I²; Pisani 1964:336 ¢143D,a²; [Dohrn in Helbig/Speier 1969, pp.674-5 ¢2752²]; R.G. Giacomelli
1978:75 ¢3,I². Drawing: Herbig CIE 8397.
276. Scratched along the front of a tile (64u46 cm; letters 2-6 cm high).
ceisiooufilio
uolθeo
277
Dohrn gave the provenance of these tiles erroneously as Falerii Veteres.
527
CHAPTER 16
Sinistroverse, Faliscan alphabet with reversed s. The θ is almost -shaped and was in
fact read as d by Herbig, Jacobsohn, and Pisani, but in view of the general angular
shape of the letters of this inscriptions, this is not necessary: see, however, §3.5.4.
(autopsy); Jacobsohn 1910:6 ¢41b²; Herbig CIE 8398; Vetter 1953:319 ¢337b²; G. Giacomelli 1963:104
¢140,II²; Pisani 1964:336 ¢143D,b²; [Dohrn in Helbig/Speier 1969, pp.674-5 ¢2752²]; R.G. Giacomelli
1978:75 ¢3,II². Drawing: Herbig CIE 8398.
277-284. Scratched on vases in the Crescenzi collection are several inscriptions:

277. Scratched on the rim of a saucer (height 6.5 cm, 16 cm; letters 3-4 mm high).
cau
Dextroverse. Herbig and G. Giacomelli unnecessarily considered reading cal.
Bibliography: Herbig CIE 8383i,a; Vetter 1953:324 ¢347²; G. Giacomelli 1963:63 ¢46,Ia². Drawing:
Nogara in CIE 8383i,a.
278. Scratched in a plate (height 2 cm, 8.5 cm).

la
Dextroverse. The a is .
Bibliography: Herbig CIE 8383k. Drawing: Nogara in CIE 8383k.
279-280. Scratched i plate (height 2.5 cm, 12.5 cm) are

acre
ame
Sinistroverse. In 279, the r is ; in 280, the e is .
Bibliography: Herbig CIE 8383l,a-b; G. Giacomelli 1963:63 ¢46,IIa-b². Drawing: Nogara in CIE
8383l, a-b.
281. Scratched under a small bucchero vase ( 5 cm).

ta
Sinistroverse. The a is .
Bibliography: Herbig CIE 8383m. Drawings: Nogara in CIE 8383m.
282. Scratched a third- or second-century plate (height 5.5 cm, 14.9 cm)
aṃẹ
Dextroverse? The a is , the m (a similar m occurs in MLF 317), the e .
Bibliography: Herbig CIE 8383p; G. Giacomelli 1963:63 ¢46,III². Drawings: Nogara in CIE 8383p.
283. Scratched under a small bucchero vase ( 6 cm).

ta
Dextroverse.
Bibliography: Herbig CIE 8383q. Drawing: Nogara in CIE 8383q.
528
284. Scratched in a plate (height 4 cm, 9.8 cm) is

ue
Sinistroverse.
Bibliography: Herbig CIE 8383r. Drawing: Nogara in CIE 8383r.
16.2.5. The surroundings of Corchiano. From the wider surroundings of Corchiano

are several inscriptions that may date either from before or after the war of 241-240,
and have therefore been classed as Middle or Late Faliscan, unless there are reasons to
date them otherwise.
285. About 1.5 km north of Corchiano, at an ancient crossing of the Rio della Tenuta (or
Rio Merlese) known as Puntone (Dennis, G. Giacomelli), Pontone (Buglione di
Monale, FI), or Ponte (Ward Perkins & Frederiksen) del Ponte, are the remains of a
small Faliscan settlement (see Ward Perkins & Frederiksen 1957:123-7 with map p.122
fig.20), probably a dependency of Corchiano, with which it was connected by a
Faliscan predecessor of the Via Amerina. Like Corchiano, it appears to have been
deserted after the war of 241 (Ward Perkins & Frederiksen 1957:125-7). Cut in the back
wall of the porticus of a tomb to the left of the Rio della Tenuta.
[---]fatecelaletezotxxiiii
Sinistroverse, Faliscan alphabet. The third e is  (cf. the similar e in MF 64 and 258);
the z is ᚮ. The second interpunct was read after celale by Buglione di Monale and
Nogara, but after cela by G. Giacomelli. According to Nogara (in Herbig CIE), the four
strokes at the end are shallower than the other letters and may be later additions. Herbig
interpreted [---]fate as a genitive ‘...fatis’. Peruzzi divided letezotxxiiii as lete zot xxiiii
‘lecti sunt XXIIII’ (for the omission of syllable-final /k/ before /t/, see §3.5.7c). His al-
ternative for the first part, [---] f atecela ‘... f(ilii) antecella’, however, depends on a
space between f and ate that appears to be non-existent. M. Mancini, interprets lete as
the locative of a */lotā/ with a similar meaning: see, however, §8.2.1 and §3.7.6.
Bibliography: [Dennis 1848:157 (autopsy)]; [Dennis 1878:120]; Buglione di Monale 1887:28 (au-
topsy); [Deecke 1888:48]; Herbig 1910:192; Herbig CIE 8391 + add.; Vetter 1953:323 ¢342a,2²; [Ward
Perkins & Frederiksen 1957:125]; G. Giacomelli 1963:102-3 ¢136² (autopsy); Peruzzi 1967a; M.
Mancini 2002:28-33. Photograph: FI II.1 p.414 fig.263. Drawing: Nogara in CIE 8391 add.
286. Scratched on the bottom of a small saucer from tomb 2 of the tombs discovered
and excavated in 1916 at Contrada Lista, c.2 km north of Corchiano. From the same
tomb is the saucer with the Etruscan inscription mi alsi*is mi Etr XXXVI.
la
Sinistroverse.
Bibliography: Bendinelli 1920:29 ¢35² (autopsy); Buffa NRIE 979; G. Giacomelli 1963:63 ¢44,II².
Transcription: Bendinelli 1920:29.
529
CHAPTER 16
287. Scratched on the bottom of a cup from an isolated tomb on the Fosso del Ponte
delle Tavole to the south of Corchiano, explored perhaps c.1893, is an incomprehensible
inscription: incomprehensible, in fact, that Herbig and FI each present it with a
different side up:
Fig.16.2. Herbig’s tracing of MLF 287.

(From CIE 8400a.)
The ductus is apparently sinistroverse, but the letters and contents are unclear.
Bibliography: Herbig CIE 8400a (autopsy); FI II.2 p.318 (autopsy). Drawing: Herbig CIE 8400a.
288. Cut above and to the right of the entrance in the portico of a tomb in Contrada
Musalè, to the left of the Fosso delle Pastine is a dextroverse inscription in Latin
alphabet with cursive e (letters of the first line 11-13 cm high, of the second line 7-9 cm
high), according to G. Giacomelli from the third or second century. The inscription was
published twice, and the two very different accounts leave me at a loss as to what the
text is, apart from the fact that the first letters are mhaedua. G. Giacomelli (1961) read
m.haedua.c.f.g ̣a|( )q̣ue[.]mac ̣ḷoịne(u.)c(u)a, later (1963) changing her reading of the
middle part to ga|[uia]que maclọịne(a), and that of the end to c.(f.)a ‘figlia di Caio’.
Vetter’s magoḷneos or magoḷneos (in G. Giacomelli 1963 and 1965) is impossible
according to G. Giacomelli (1963). The same inscription was published again as
m.ḥaeduaonoa | mimạnoineṣao by R. Giacomelli, who erroneously described it as
unedited. He implausibly connected haedua to the Gallic tribe of the Haedui.
Bibliography: (I) G. Giacomelli 1961:325-6 ¢5² (autopsy); G. Giacomelli 1963:266 ¢XX²; G. Gia-
comelli 1965:551. Drawing: G. Giacomelli 1961:325 fig.5. – (II) R. Giacomelli 1977:63-8 (autopsy).
Photographs: R. Giacomelli 1977:61-3 figs.1-3. – (III) Marinetti 1982:36.
289. Cut in the back wall of a tomb close to where the ancient road to Gallese crossed
the Rio delle Pastine (cf. Ward Perkins & Frederiksen 1957:169) is an intelligible
inscription given by Nogara as
̣̣̣ laris:ṃ:ṛφ̣c ̣χ̣a
̣̣̣
̣ ̣̣
̣ ̣̣ uạịẹṣịṿṣist
Sinistroverse, Etruscan alphabet? G. Giacomelli read the  as ọ. Only the praenomen
laris is clear: Rix reads the remainder of the first line as mạrc ̣ẹṇạ.
Bibliography: Herbig CIE 8588; Vetter 1953:323 ¢342a,3²; G. Giacomelli 1963:103 ¢137²; Rix ET
Fa 1.5. Transcription: Nogara in CIE 8588.
530
290. Cut in the wall of the hollow road leading down to the gorge of the Rio delle
Pastine near the Madonna delle Grazie (cf. Ward Perkins & Frederiksen 1957:169).278
c**(*)coṇẹol***(*)cepaui[ceoru?]so
Dextroverse, Latin alphabet with cursive e. Pasqui’s drawing shows the first two words
cnaecomio, but from the photograph, I am inclined to read either cṇạcoṇẹo (cf.
aco[n]ẹọ LtF 329 and acon[?]io LtF 343) or cp ̣ṛẹcoṇẹo (cf. precono in MLF 361).
What follows is uncertain: Pasqui’s drawing has l.il.o., either a filiation (lf̣ilịo?
lọc ̣ilịo?) or a cognomen. The second name is largely obliterated: Pasqui’s drawing
gives cepaui**so, but the photograph shows that the number of missing letters was
larger, probably three to five. The gentilicium was probably paui[ceo, cf. pauiceo MF
12; ]so is probably a cognomen (cf. perhaps ruso in MLF 318?): see §7.9.1. The names
are those of local magistrates, see §11.1.4.5.
Bibliography: Pasqui in FI II.1 p.45 (autopsy). Photograph: FI II.1 p.414 fig.262. Drawing: Pasqui in
FI II.1 p.45.
291. Cut in the rock c.1.5 km south of Corchiano along the ancient road to the settle-
ment at the crossing of the Rio Cruè (cf. Ward Perkins & Frederiksen 1957:117 with
maps p.112 fig.18 and p.119 fig.19, and FI II.2 p.217). Second century.
cegnatiuss[ex]fprata
faciundacoirauit
Dextroverse, Latin alphabet. The reading egnatius goes back to Nogara (in Bormann;
ponatius Gamurrini, FI). The lacuna can only be restored as s[ex]; G. Giacomelli’s
s[exti.] is impossible (Sp. [f.] Di Stefano Manzella). The expression prata facere was
something of a terminus technicus, cf. e.g. Varro R 2 praef. 4, Iustin. dig. 39.3.3.2.
Bibliography: Gamurrini 1887a:62 (autopsy); Lommatzsch CIL I2.1992; Bormann CIL XI.7505;
Safarewicz 1955:186 ¢6²; Degrassi ILLRP 1263; G. Giacomelli 1963:266 ¢XXI²; FI II.2 p.217 n.32
(autopsy); Degrassi & Krummrey CIL I2.1992 add. Photograph: FI II.1 p.407 fig.249. Drawing: FI II.2
p.217 n.32.
292-296. Along the ancient road between Corchiano and the settlement on the Rio Cruè
(Ward Perkins & Frederiksen 1957:117 with maps p.112 fig.18 and p.119 fig.19, and FI
II.2 pp.217, 320), a series of tombs was excavated in 1887 at Contrada La Selvotta (see
FI II.2 pp.320-1). The stamps on some of the tiles (cf. FI II.2 p.321) show that the
tombs were in use until the Imperial period.
292. Scratched under a plate from the first tomba a fossa.
anslrụfi[?---]
Nogara’s transcription is …, which Herbig read as ‘[…]ans  L  Rụfi …’.
278
FI II.1 p.414 gives the location as “in cima alla cava di Valle Spigliara”; Pasqui located Valle
Spigliara on the right side of the stream and the road with the inscription on the left.
531
CHAPTER 16
Bibliography: Herbig CIE 8381b; FI II.2 p.320 (autopsy). Transcription: Nogara in CIE 8381b.
293. Scratched inside a cup from the first tomba a camera.

lociaeiṃoi
Nogara had read  lociae ṭịtoi, but this would contain an ae that is unparal-
leled in the inscriptions in Faliscan alphabet, and Herbig (CIE) rightly read eiṃoi (cf.
aịṃiosio eqo EF 467*). This is generally interpreted as a dative, making the inscription
unique in giving the names both of the giver and the receiver (§8.8.1); if eiṃoi is a
genitive (§8.8.1), the inscription would present a likewise unique instance of a filiation
formula in a Besitzerinschrift (§7.5.1).
Bibliography: Herbig CIE 8381e; Herbig 1914a:239-40; Lejeune 1952b:125; Vetter 1953:324 ¢344²; G.
Giacomelli 1963:63-4 ¢47²; FI II.2 p.320 (autopsy). Transcription: Nogara in CIE 8381e (reproduced
in Herbig 1914a:240).
294. Scratched under a black-varnished cup from the third tomba a camera
at
Dextroverse, Latin alphabet.
Bibliography: FI II.2 p.321 (autopsy). Drawing: FI II.2 p.321.
295-296. Scratched before firing on a Megarian bowl from the second tomba a fossa. FI
mentions only no.297, but Siebourg, and apparently also Baudrillart (“environs de
Corchiano” 1889:288), ascribe a second bowl to this location. c.230-150 (Baudrillart).
cpopil[i]
cpopili meuanie
Dextroverse, Latin alphabet. The l is ᚳ. FI gives 295 as cpopili. In 296, Baudrillart read
popil., but Siebourg’s popili is certainly right. Meuanie is locative, not Siebourg’s
Mevanie((n)sis): cf. the ablative (?) ocriclo ‘Ocriculum’ on other bowls from this
workshop (e.g. CIL I2.421a-b). The inscriptions are clearly imports. See also Lat 478*.
Bibliography: Baudrillart 1889:288-9; Siebourg 1897:43-4 ¢4, 6²; Bormann CIL XI.6704,2d, 3a;
Lommatzsch CIL I2.419d, 420a; Lommatzsch CIL I2.420a add.; FI II.2 p.320 (autopsy); Degrassi &
Krummrey CIL I2.420a add. Drawings: Baudrillart 1889 pl.VII; Hülsen in Siebourg 1897:42; Siebourg
1897:42; FI II.2 p.320.
297-301. A series of tiles was found in 1953 at a spot called Puntone del Pero and
published by G. Giacomelli among the inscriptions from Corchiano and its territory.
297-298. Painted in yellow along the length, 297 of the front, 298 of the back of a tile.
iuna
iuna
Bibliography: G. Giacomelli 1963:98-100 ¢128,Ia-b² (autopsy). Photograph: G. Giacomelli 1963
tav.X.
532
299. Painted in white on the front of a tile.

1
uonel[2][n---]
Dextroverse in Latin alphabet. The l is ᚿ (§11.2.4). G. Giacomelli read [---]uo as the end
of a gentilicium, assuming a lost preceding tile that contained the praenomen and the
beginning of the gentilicium. In that case, however, nel[n---] would have to be filiation
(in view of its position in 300 it seems impossible that it is a cognomen), which is
difficult as there are no praenomina in Nel( )n-, Nael( )n-, or Nil( )n-. Neln[---] is
therefore rather a gentilicium, and uo is the abbreviation of Volta.
Bibliography: G. Giacomelli 1963:98-100 ¢128,II² (autopsy); Rix 1964:447 n.4; G. Giacomelli
1965:550. Photograph: G. Giacomelli 1963 tav.X (reproduced in G. Giacomelli 1965 tav.CXXXVa).

[1]
[---]2aneln f
[1]
[---]2uxoohi*[..]
Dextroverse, Latin alphabet. The e is  , the l ᚿ: Rix proposed to read neron[, but this
is impossible according to G. Giacomelli (1965). The f is placed between both lines. In
the second line, the i is followed by the upper part of a shaft, perhaps an l (G. Gia-
comelli). One tile is certainly missing at the beginning. Uxo, combined with [---]a,
shows that the inscription contained the name of a woman. The arrangement seems to
be: a praenomen [---]a starting on a tile missing in front followed a gentilicium
neln|[---] continued on the second line of the missing tile; then, still on the missing tile,
the (probably abbreviated) name of the husband in the genitive followed uxo and by an
affiliation ohi*[..] | f. This can perhaps be read as ohiḷ[i] | f ‘Oufili f(ilia)’ with a
praenomen Oufilus (cf. aufilo aratio MLF 348 and §7.7.1.12): the use of h for f would
then be a (hypercorrect) extension of its use for original f in word-initial position: see
§3.5.2.
Bibliography: G. Giacomelli 1963:98-100 ¢128,III² (autopsy); Rix 1964:447 n.4; Olzscha 1965:122-3;
G. Giacomelli 1965:550. Photograph: G. Giacomelli 1963 tav.X (reproduced in G. Giacomelli 1965
tav. CXXXVb).

[....]nea*a
[u]xor ia*
maoṣcin*
Dextroverse, Latin alphabet with cursive a and e. According to G. Giacomelli, the
penultimate letter of the first line looks like “una g stravolta e rovesciata” (1963:99). In
the second line, ia is placed apart from and slightly higher than [u]xor; what follows
this is completely unclear. In the last line, G. Giacomelli read oxcinu, but her photo-
graph and description (“una x stranamente rotondeggiante nella parte sinistra”, 1963:99)
seem to point to an s () rather than an x. It is unclear whether any tiles were missing in
533
CHAPTER 16
front. G. Giacomelli read the text as two inscriptions, the first giving a woman’s name
consisting of a (missing) abbreviated praenomen and a gentilicium ending in ]nea,
followed by the abbreviated name of her husband and [u]xor, and the consisting of the
remainder of the text. Rix proposed to read the first of these inscriptions as ]neron[i..] /
ian[ta... / u]xor, but according to G. Giacomelli (1965), this is impossible.
Bibliography: G. Giacomelli 1963:98-100 ¢128,IV² (autopsy); Rix 1964:447 n.4; G. Giacomelli
1965:550. Photograph: G. Giacomelli 1963 tav.X (reproduced in G. Giacomelli 1965 tav. CXXXVc).
16.3. Vignanello
Although excavations at this location go back to the 18th century (cf. Giglioli
1916:37-8), little is known of the Faliscan settlement at Vignanello, which was
probably one of the smaller settlements of the north-western ager Faliscus. As said in
§16.1, the third-century tomb of the gens Velminaea (from which come MLF 305-323)
implies that the settlement was still inhabited after the war of 241 (cf. §2.5.2). The
inscriptions are all from tombs at Contrada Molesino, to the west of the town, on the
slope between the modern road to Vallerano and the Piano della Cupa, near the tunnel
of the Roma-Viterbo railway (Giglioli 1916:37-8 with map p.39 fig.1). Most of these
tombs were discovered undisturbed in 1916 and excavated in the same year: only
Giglioli’s tomb a (from which come MLF 302-303) appears to have been explored
before that date.
302-303. The following inscriptions are from Giglioli’s tomb a.

302. Cut over a loculus.
ịạfirmiatitia
Sinistroverse, Faliscan alphabet. The traces of the first two letters are : Vetter’s ia,
adopted by G. Giacomelli, is certainly preferable to Giglioli’s [he:?]. Only the upper
points of the interpuncts are preserved. The point in firmia is probably a chance trace.
The last a is of the type : see §11.2.4.2. The gentilicium Firmius is also attested at
Civita Castellana (ḥirmia MF 18, ḥiṛṃeo MF 19, firmio MF 54) and S. Maria di Falleri
(hirmio LF 213).
Bibliography: Giglioli 1916:38 (autopsy); Nogara 1916; Herbig 1923:231 ¢1²; Vetter 1953:320 ¢338a²;
G. Giacomelli 1963:105 ¢143,I² (autopsy). Drawing: Giglioli 1916:38.
303. Cut between two loculi.

popliacocelia
Sinistroverse, Faliscan alphabet. The p is .
Bibliography: Giglioli 1916:38 (autopsy); Nogara 1916; Herbig 1923:231 ¢2²; Vetter 1953:320 ¢338b²;
G. Giacomelli 1963:105 ¢143,II² (autopsy). Drawing: Giglioli 1916:38.
534
304. Scratched on the bottom of a fourth- or third-century saucer from Giglioli’s tomb
2.
pupiias
Sinistroverse, Faliscan (?) alphabet with reversed s. The a is . Giglioli’s transcript
showed a lacuna of one letter between the a and the s. Vetter interpreted pupia[.]s as a
genitive (an interpretation adopted by all later editors), assuming that Giglioli’s lacuna
did not, in fact, contain any letters: the lacuna is in fact omitted by Pisani, Cristofani,
and Rix, and Gulinelli’s drawing now shows definitively that there is indeed no lacuna
between a and s. On the other hand, she adds a new problem by reading pupiias, with a
double i that is without parallel in any inscription from the ager Faliscus. I have
considered reading pupẹas with cursive e, but the inscription seems rather too early for
this: neither is it possible to read puplias.
̣
Bibliography: Giglioli 1916:62 (autopsy); Buffa NRIE 987; Vetter 1953a:326 ¢352²; G. Giacomelli
1963:64 ¢51²; Pisani 1964:344 ¢146G,b²; Cristofani 1988:20, 24 ¢18²; Rix ET Fa 2.21; Gulinelli 1995
(autopsy). Drawing: Gulinelli 1995.
305-323 (the ‘tomba dei Velminei’). Giglioli’s tomb 3 was found intact, with the
inscriptions still in situ: inscriptions MLF 316-319 were discovered only after the tiles
had been removed and cleaned, and their location in the tomb is unknown (see Giglioli
1916:77-8). Most inscriptions contain a gentilicium uelmineo: as this is never written as
uelminio, it may have ended in /-ē ̣os/ ← /-os/, or in /-ę̄()os/ ← /-āos/ rather than in
/-ios/ (cf. §3.7.6, §7.8.1.165). The tomb was dated to the third century by Giglioli.
305. Painted in red across the front of a tile (c.65uc.42 cm; letters c.4 cm high).
titouelmineo
titọifecupa
Sinistroverse, Faliscan alphabet. The last part of each line (the letters ineo and ecupa) is
written vertically downwards. T is , but the second t is . In the second line, Herbig’s
tit[io] (a patronymic adjective) has been adopted by all later editors, but Giglioli’s
drawing appears to have titọi, which, if correct, would be an unambiguous case of a
genitive in -oi (§4.4.4). Unfortunately, according to G. Giacomelli, all that can now be
read of the second line is tit.
Bibliography: Giglioli 1916:65-6 (autopsy); Nogara 1916; Herbig 1923:231-2 ¢3²; Vetter 1953:320
¢339a²; G. Giacomelli 1963:105-7 ¢144,I² (autopsy); Pisani 1964:336-7 ¢143E². Drawing: Giglioli
1916:65 fig.20.
306-307. The titulus prior, 306, is painted in white across the front of a tile.
cauia
[---]
[---]
535
CHAPTER 16
Sinistroverse, Faliscan alphabet. Giglioli mentions traces of a second and a third line
that do not appear his drawing. The tile was later used as the first tile of 307, painted in
red across the front of two tiles.
1
iuna2uelmineo
1
titio 2
Sinistroverse, Faliscan alphabet. The o of uelmineo is written under the line.
Bibliography: Giglioli 1916:66 (autopsy); Nogara 1916; Herbig 1923:231-2 ¢4²; Vetter 1953:320
¢339b,A-B²; G. Giacomelli 1963:105-7 ¢144,IIa-b² (autopsy). Drawing: Giglioli 1916:66 fig.21.
308. Painted in white across the front of a tile.

cauiouelminẹo
popliaifile
Sinistroverse, Faliscan alphabet. The middle parts of both lines (elmi and aif) are
written vertically downwards, the last parts (neo and ile) upside down in boustro-
phedon. The last word is read either as file[ai] (Giglioli, Nogara, Herbig) or as file(ai)
(Vetter, G. Giacomelli). According to Giglioli, the funeral gifts indicate that the
deceased was a woman, which excludes popliai file[o] ‘son of Publia’: see §8.10.2.
¢339c²; G. Giacomelli 1963:105-7 ¢144,III² (autopsy); Pisani 1964:340-1 ¢145A². Photograph: Giglioli
1916:68 fig.23.
309. Painted in white on along the length of three tiles (the first max. c.25uc.42 cm, the
others c.67uc.42 cm; letters 10-17 cm high).
1
ti2toᛂuel3mineo
1
nu2i*ice 3
Sinistroverse, Faliscan alphabet. Editors generally read titoi, but I regard the stroke after
tito rather as an interpunct: see below. The o of uelmineo is written under the line: it
contains a stroke that is probably accidental. The second line, , has
been read as p...ice (Giglioli), nuip ̣ice (Herbig),279 nụidice (Vetter), and nụiḍice (G.
Giacomelli): comparison with 315 shows that it ended in ip ̣ice or iḍice, and that the
preceding letters are probably an abbreviation of the father’s praenomen.
Herbig interpreted his ip ̣ice as an originally reduplicative perfect i(m)pice(t) =
*impigit m */in-pepag-/ from impingo ‘to fasten (a tile) upon (a loculus)’, with uelmi-
ne|o | ṇ ‘N. Volminius’ as the subject and titoi as a dative. Later editors rejected this,
assuming the Middle and Late Faliscan outcome of */in-pepag-/ would be *i(m)pace(t)
rather than *i(m)pice(t) (cf. §3.6.6), although most maintained the interpretation of i*ice
as a verb, however. Stolte took ip ̣ice as a perfect /in-pīgit/ from an *impingo ‘to paint
(an inscription) upon (a tile)’ (an idea rejected by Herbig); Ribezzo and Peruzzi took
279
Pisani’s impossible ṇ... ipice appear to be an erroneous rendering of this reading.
536
ip ̣ice as an Etruscan verbal form in -ce, the former deriving it from Latin ibi (‘collo-
cavit’), the latter from Etruscan ipi ‘olla sepolcrale’. I find none of these proposals
particularly attractive (cf. §6.2.38), and would rather read tito ᛂ (with a stroke-
interpunct) uelmineo | nu ‘Titus Velminaeus son of Nu.’ (or ịụ ‘son of Iu(na)’) and
interpret i*ice either as an intransitive verb similar in sense to cupat, or as a cognomen,
although the latter is difficult, cf. §7.9.1.3.
Bibliography: Giglioli 1916:72-3 (autopsy); Nogara 1916; Herbig 1923:231-2 ¢15²; Stolte 1926:61; Ri-
bezzo 1931b:192; Vetter 1953:320 ¢339d²; G. Giacomelli 1963:105-7 ¢144,IV² (autopsy); Pisani
1964:341 ¢145C,a². Drawing: Giglioli 1916:73 fig.30.
310. Painted in white along the length of three tiles (the first two c.65 u c.45 cm, the
third c.35 u c.45 cm; let. 7-10 cm).
1
cuicto2 uelmin3eo
1
[---?]2uoxie[.]3eai
Sinistroverse, Faliscan alphabet. Although the second line is unclear, the contents of the
text seem to have been similar to those of 308, with a man’s name in the first line
followed by a woman’s name in the dative (or the genitive?) in the second. The only
proposal for restoration is Herbig’s uo.x.fe[l]eai.
¢339e²; G. Giacomelli 1963:105-7 ¢144,V² (autopsy). Drawing: Giglioli 1916:73 fig.31.
311. Painted along the length of the front of a tile (c.55 u c.45 cm; letters 10-12 cm
high).
sextia
¢339f²; G. Giacomelli 1963:105-7 ¢144,VI² (autopsy). Drawing: Giglioli 1916:74 fig.32.
312. Painted in white along the lenth of the front of two tiles (c.65uc.45 cm; letters 8-10
cm high).
1
uoltio[]2uelmineo
1
titiosce2ụa
Sinistroverse, Faliscan alphabet. The penultimate letter is : Giglioli, Nogara, and
Herbig read sceua, Vetter rex a.... The latter is impossible according to G. Giacomelli,
who adopted Giglioli’s alternative scexa, noting from autopsy that the letter is now
illegible. Sceua, however, now has a parallel in sceiuai LF 379.
¢339g²; G. Giacomelli 1963:105-7 ¢144,VII² (autopsy); Torelli 1967:536-7. Drawing: Giglioli 1916:75
fig.35.
537
CHAPTER 16
313. Painted in white along the length of the front of two tiles (c.65uc.45 cm; letters
c.10 cm high).
1
uoltau2elmineo
1
fuloni2acue
¢339h²; G. Giacomelli 1963:105-7 ¢144,VIII² (autopsy); Pisani 1964:340 ¢144F². Drawing: Giglioli
1916:75 fig.36.
314. Painted along the length of the front of a tile (c.60uc.45 cm; letters 6-10 cm high).
cauia loriea
Sinistroverse, Faliscan alphabet. The last part of the line (the letters riea) is written
vertically downwards due to lack of space. For loriea, cf. loụṛia MF 41.
¢339i²; G. Giacomelli 1963:105-7 ¢144,IX² (autopsy). Drawing: Giglioli 1916:76 fig.37.
315. Painted in white along the length of the front of a tile.

titouel
mineoiun
aịị*ice
Sinistroverse, Faliscan alphabet. The third line has been read as iun|aluaice (Giglioli),
iun|aịịp ̣ice (Herbig), iun-|(-)ali-ice or iun|[e]ọụḍice (Vetter), iun|ai [ip]ice (Pisani), and
iun|aị( )**ice (G. Giacomelli). Most editors interpret this text in the same way as 309,
with tito : uel|mineo as the subject and iun|aị as a dative to go with a verb ip ̣ice. Iun|aị
is probably rather a filiation (§8.10.2); for interpretations of i*ice, see 309.
Bibliography: Giglioli 1916:76 (autopsy); Nogara 1916; Herbig 1923:231-2 ¢16²; Stolte 1926:61; Ri-
bezzo 1931:192; Vetter 1953:321 ¢339k²; G. Giacomelli 1963:105-7 ¢144,X² (autopsy); Pisani
1964:341 ¢145C,b². Photograph: Giglioli 1916:77 fig.39.
316. Painted in upward-slanting lines across the front of a tile (c.67uc.45 cm; letters
5-10 cm high).
popli[o]
uelmi
no
Dextroverse, but apparently in Faliscan alphabet. P is ; the e stands within the u; the m
is reversed. Uelmi|no is clearly an error for uelmi|n‹e›o.
¢340d²; G. Giacomelli 1963:105-7 ¢144,XI² (autopsy). Drawing: Giglioli 1916:78 fig.41.
317. Painted across the front of a damaged tile (c.60uc.40 cm; letters 7-10 cm high).
1
popl[2][---]
1
[u]elṃi[2][ne---]
538
Sinistroverse, Faliscan alphabet. The second line is unclear. The third legible letter was
first read as e, but Vetter read it sideways as an m (a similar sign in MLF 282 was als
read as m by Herbig), which would give [u]elṃi[ne-]. This would require a second tile:
Vetter’s suggestion that this was 319 is implausible in view of the different shapes of
the letters and of the fact that 319 is not painted across the tile, but lengthwise.
¢340a²; G. Giacomelli 1963:105-7 ¢144,XIII². Drawing: Giglioli 1916:78 fig.42.
318. Painted on the front of a tile (max. c.60uc.50 cm; letters from c.8-10 cm to c.15-20
cm high).
1
cauio[2][---]
1
ruso[2][?---]
Sinistroverse, Faliscan alphabet with reversed s. Giglioli, Nogara, and Herbig took ruso
as a gentilicium; Vetter, and, apparently, G. Giacomelli, assumed a tile was missing at
the end that contained the gentilicium in the first line, with a cognomen ruso as in M.
Occius Ruso in CIL XI.3254 I.13 from Sutri.
¢340b²; G. Giacomelli 1963:105-7 ¢144,XIV² (autopsy). Drawing: Giglioli 1916:78 fig.43.
319. Painted on a tile (c.60uc.45 cm; letters c.11 cm high) is

[1]
[---]2o
[1]
[---]2no
Sinistroverse, Faliscan alphabet? The letters are followed by an empty space; Herbig’s
o... | no... is therefore impossible. Giglioli (apparently holding the tile the other way
up?) read [---]no |[---]o (dextroverse, with reversed n). At least one tile is missing in
front; Vetter implausibly suggested that this was the tile of 317.
¢340c²; G. Giacomelli 1963:105-7 ¢144,XII² (autopsy). Drawing: Giglioli 1916:78 fig.40.
320-323. From the same tomb are several small plates with abbreviations:
ce
ce
ue
ca
Sinistroverse.
320 and 323 from autopsy in the Museo dell’Agro Falisco, Civita Castellana (inv. 26090 and 26266).
Bibliography: Giglioli 1916:81-2 (autopsy). Transcriptions: Giglioli 1916:81-2.
539
CHAPTER 16
16.4. Fabbrica di Roma
The Faliscan settlement at Fabbrica di Roma is another of the minor centres of the
northwestern ager Faliscus of which little is known: the presence of inscriptions in Latin
alphabet (LtF 325-328) seems to indicate that it survived the war of 241 (§2.1.2). Of the
inscriptions from this site, MLF 324 was found during the excavation, in 1888, of one
of three tombs discovered that year at Poggio or Monte delle Monache, a height to the
south of the town (Pasqui 1889, Ward Perkins & Frederiksen 1957:158); the prove-
nance of the others appears to be unknown. Two inscriptions of uncertain origin (MLF
360-361), may also be from Fabbrica.
324. Scratched across the front of a tile (45u65 cm; letters only 2.5 cm high) from a late
fourth-century tomb at Monte delle Monache.280
cauiolatrio
Sinistroverse, Faliscan alphabet. The first o is damaged, but it is unnecessary to
consider i, f, or p (as does Pasqui), or to read caui (as does Conway).
From autopsy, Museo di Villa Giulia, Rome (inv. 8241). Bibliography: Pasqui 1889 (autopsy);
Conway 1897:372 ¢313²; Herbig CIE 8370 (autopsy); [Della Seta 1918:103 (autopsy)]; Bormann CIL
XI.7514; Vetter 1953:316 ¢325²; [Ward Perkins & Frederiksen 1957:158]; G. Giacomelli 1963:98
¢125². Drawing: Herbig CIE 8370.
325. Painted in red on plaster along the back of a tile (78u42 cm; letters 11-14 cm high).
mneroni
afethlau
eleamf
Dextroverse, Latin alphabet. Hl is a ligature  (t.l Thulin); this use of h is probably a
hypercorrect extension of its use for (original) f in word-initial position before a vowel
(§3.5.2), perhaps implying that the inscription may have been written by someone who
was not well acquainted with Faliscan orthography. Thulin joined this tile to that of LtF
326, but Herbig (CIE) rightly rejected this because of the differences in the sizes of the
tiles and the letters. The gentilicium Neronius is attested also in LtF 328 from Fabbrica
di Roma, at Civita Castellana in MF 15 and 16, and at the Grotta Porciosa site in LtF
340; Flauilius is not attested elsewhere in the area.
Bibliography: Thulin 1907:268 (autopsy); Herbig 1910:109 n.1 (autopsy); Jacobsohn 1910:6 ¢46b²;
Herbig CIE 8374; Buonamici 1913:72 ¢37²; Bormann CIL XI.7519 (autopsy); Lejeune 1952b:118;
Vetter 1953:316 ¢327d²; G. Giacomelli 1963:265 ¢XVI². Drawing: Herbig CIE 8374.
280
Conway erroneously described the inscription as painted and ascribed it to the La Penna
necropolis at Civita Castellana.
540
326. Painted in red on plaster on a tile (69u47 cm; letters 13-15 cm high).
[---] cuba
[---]nte
Dextroverse, Latin alphabet. Thulin unconvincingly joined this tile to that of 325. The
two lines are probably not to be read together, as did Thulin (cubante with an epenthetic
[-e], cf. §9.2.2.1) and Vetter (cuba-/nt (h)e(c), requiring a inversion of the usual formula
(§8.10.1) and an omission of word-initial h-, for which see §3.5.2). The second line
could be aru]nte, but even then the overall arrangement of the text remains unclear.
Bibliography: Thulin 1907:268 (autopsy); Herbig CIE 8375 (autopsy); Bormann CIL XI.7520 (au-
topsy); Vetter 1953:316 ¢327e²; G. Giacomelli 1963:265 ¢XVIII². Drawing: Herbig CIE 8375.
327. Painted in white on the front of two tiles (both 68u46 cm; letters of the first tile 12
cm high, of the second tile 14 cm high).
1 2?
staco[ ]**uei
1 2?
leuia[ ] tf
Dextroverse, Latin alphabet. The traces of the first two letters on the second tile are
(*u or *ụ editors); the i, omitted by Bormann, is (ṭ Herbig, Vetter, and G.
Giacomelli). The first tile can be read separately as ‘St. Acus vel sim.; Levia’, but the
tiles may also be joined, as was done by Herbig, who read st.aco[n]eo.uet(-) |
leuia[.s]t.f(ilea) ‘St. Aconius Vet...; Livia, daughter of Statius’. A similar case of aco or
aconio is LtF 341 from the Grotta Porciosa site; cf. the possible reading ạcoṇẹo in LtF
290 from near Corchiano.
Bibliography: Herbig CIE 8372 and 8273 (autopsy); Bormann CIL XI.7518, 7522 (autopsy); Vetter
1953:316 ¢327a-b²; G. Giacomelli 1963:265 ¢XV,i-ii². Drawings: Herbig CIE 8372-8273.
328. Painted in white on the front of a tile (65u45 cm; letters 22-25 cm high).
1 2?
[se]xne[ ]ro[---]
Dextroverse, Latin alphabet. The x was read as t by Herbig, but in a text in Latin
alphabet this seems unlikely. The tiles were joined by Herbig (albeit with some
hesitation) and all later editors, and read as [se]xnero[nio], with a point that is
probably a chance trace (cf. neroni MF 15). The gentilicium Neronius is also attested in
LtF 325 from Fabbrica di Roma, at Civita Castellana in MF 15 and 16, and at the Grotta
Porciosa site in LtF 340.
Bibliography: Herbig CIE 8376-8377 (autopsy); Bormann CIL XI.7521 (autopsy); Vetter 1953:316
¢327c²; G. Giacomelli 1963:265 ¢XVIIa-b². Drawing: Herbig CIE 8376-8377.
541
CHAPTER 16
16.5. Carbognano-Vallerano
329-337 (the ‘tomba dei Folcosii’). Late in 1880 or early in 1881, a tomb was discov-
ered at Contrada I Quarti, c.3 km north of Carbognano (cf. Dressel 1881, Bazzichelli
1881, and Ward Perkins & Frederiksen 1957:158). Dressel dated the tomb and its
contents to the second half of the second century,281 which if correct would be interest-
ing from a linguistic point of view, as none of the inscriptions show traces of Latin
influence; Bazzichelli dated it more generally to the period after 241. The contents of
the tomb where transported to the newly-founded Museo Civico at Viterbo shortly
afterwards, where they have been part of the collection since 1886 (Emiliozzi
1986:131). Most of the inscriptions from this tomb pertain to members of a gens
Folcosia, a gentilicium found also at Civita Castellana (cẹịṣ[i.] | holc[osi.] | ar f ̣ [...]
LtF 140). The two descriptions of the tomb vary with regard to the number of inscribed
tiles. Dressel published nine inscriptions; Bazzichelli on the other hand describes the
tiles both as “ventisette grandi tegole, otto [my italics] delle quali scritte in caratteri
etruschi con color nero: una di questi è in frantumi, ed altre pure sono state spezzate”
and as “le tegole scritte, che sono dieci [my italics], quattro delle quali ben conservate e
con iscrizioni. In altre tre si notano resti di scrittura, in una veggonsi appena tre lettere,
due hanno appena tracce di scrittura” (1881:136).
329. Painted in black across the front of a tile (68u48 cm; letters 5-7 cm high).
sestoᛍ
fulczeo
Sinistroverse, Faliscan alphabet. The s is , an  with a small accidental stroke at the
top, not the  read by Dressel (who thence regarded this inscription as the oldest). A
small stroke after sesto in Dressel’s drawing, now invisible, is a ‘stroke-interpunct’
(§11.2.4) rather than an i (sestọị Conway). Only the lower half of the u is preserved:
comparison with the o’s shows that it is a u (Deecke, Conway, and Herbig), not an o
(Vetter and G. Giacomelli). The z is ; the second e is . The deceased is either the
father or the son of the Voltius Folcosius of 330, depending on the interpretation of zex-
toi in that text. For the use of z-, see §11.2.4. For sesto vs. zextoi 330, see §3.5.7c.
From autopsy in the Museo Civico, Viterbo (inv. 92/93). Bibliography: Dressel 1881 ¢5² (autopsy);
Zvetaieff IIM 76; Schneider 1886:106 ¢14²; Zvetaieff III 78; Danielsson in Pauli 1887:123 (autopsy);
Bormann CIL XI.3162b,6; Deecke 1888:178-9 ¢47²; Conway 1897:382 ¢337²; Herbig CIE 8357
(autopsy); Buonamici 1913:69 ¢31²; Vetter 1953:314 ¢324a²; G. Giacomelli 1963:96-7 ¢123,I² (au-
topsy); Devine 1970:17-8. Drawing: Dressel 1881 between pp.156-7 (reproduced in IIM tab.XI.5,
Deecke 1888 Taf.II); Herbig CIE 8357.
281
“L’ipogeo [...] spetta alla prima meta incirca del settimo secolo di Roma” (Dressel
1881:159), i.e. approximately 150-100 BCE. Pisani (1964:341) quotes this as “(VII sec.? così
Dressel)”.
542
330. Painted in black on the front of a tile (66u47.5 cm; letters 4-7.5 cm high).
uoltio
folcozeo
zextoi
fi
Sinistroverse, Faliscan alphabet with cursive e (). The z is reversed, . The early
editors adopted Jordan’s interpretation of zextoi as a genitive, which I find not unattrac-
tive; since Schmidt, it has generally been interpreted as a dative: see §4.4.4 and §8.10.2.
For the use of z-, see §11.2.4.
From autopsy in the Museo Civico, Viterbo (inv. 87). Bibliography: Dressel 1881 ¢1² (autopsy);
Jordan 1881:510-1; Deecke 1881:237; Zvetaieff IIM 71; Schneider 1886:105 ¢9²; Zvetaieff III 73;
Bormann CIL XI.3162b,1; Deecke 1888:180 ¢48²; Von Planta 1897:588 ¢321²; Conway 1897:382
¢338²; Schmidt 1905:31; Herbig 1910:194; Jacobsohn 1910:5 ¢34²; Herbig CIE 8358 (autopsy);
Buonamici 1913:70 ¢32²; Herbig 1914a:237; Lejeune 1952b:125; Vetter 1953:314 ¢324b²; G. Gia-
comelli 1963:96-7 ¢123,II² (autopsy); Pisani 1964:341 ¢145B²; Devine 1970:17-8. Drawings: Dressel
1881 between pp.156-7 (reproduced in IIM tab. XI.1, Deecke 1888 Taf.II); Herbig CIE 8358.
331. Painted in black along the front of a tile (68u47 cm; letters 3-5 cm high).
cesiofolcuso
Sinistroverse, Faliscan alphabet with reversed s and cursive e (). The third letter was
read as p by Dressel (whence all early editors), but Herbig’s s, adopted by Vetter and G.
Giacomelli, is fully justified. The penultimate letter is , which has been read as s
(Vetter, G. Giacomelli, whose reading, from autopsy, I adopt) and as i: in both readings
the result is an error for *folcusio. Dressel’s folcno (?) and Deecke’s (1888) folcusio are
impossible. It is unclear whether the traces above the line belong to a lost first line
(Dressel, whence all editors until Conway).
Zvetaieff IIM 77; Schneider 1886:106 ¢15²; Zvetaieff III 79; Danielsson in Pauli 1887:124; Bormann
CIL XI.3162b,7; Deecke 1888:181-2 ¢50²; Conway 1897:382 ¢340²; Herbig CIE 8360 (autopsy);
Buonamici 1913:71 ¢34²; Vetter 1953:314 ¢324d²; G. Giacomelli 1963:96-7 ¢123,IV² (autopsy).
Drawing: Dressel 1881 between pp.156-7 (reproduced in IIM tab.XI.6, Deecke 1888 Taf.III); Herbig
CIE 8360.
332. Painted in black along the front of a tile (63.5u41 cm; letters 3-10 cm).
cẹḷio*olcuzeo
***io
poplia e
uelcẹị f
Sinistroverse, Faliscan alphabet. Only a few traces near the left edge of the tile now
remain. In the first line, all editors read cạio or caio, but Herbig’s drawing ( )
indicates cẹḷio (with reversed l as in the fourth line). Of the first letter of the gentilicium,
only a small trace remains; the penultimate seems to be e (i Danielsson, Herbig, Vetter,
543
CHAPTER 16
G. Giacomelli). The second line, given as by Dressel and as by Herbig,

has been read as caio (Bormann), c ̣ạio (Deecke 1888), or cạio (Buonamici), or as ...ẹṭio
(Danielsson), c ̣ẹtio (Herbig), cẹṭio (G. Giacomelli), but these forms can neither be
patronymic adjectives nor cognomina: perhaps it is possible to read ṣẹx̣ṭ‹i›o: the traces
would appear to allow at least ṣẹx̣to. The e against the edge of the tile at the end of the
fourth line is a continuation of the fourth line (f|e): there is no reason to adopt Lejeune’s
read poplia[cu]e. The last two letters of the name in the fourth line are given as by
Herbig, whose uelcẹị (with reversed l), adopted by G. Giacomelli, is attractive. The
other proposals (uelcịạ Danielsson, uolc...f Dressel, Zvetaieff; valci...f Schneider,
u?ịe??.f Deecke 1888, uezcia Vetter) appear to be impossible.
From autopsy in the Museo Civico, Viterbo (inv. 93bis). Bibliography: Dressel 1881 (autopsy);
Zvetaieff IIM 75; Schneider 1886:106 ¢13²; Zvetaieff III 77; Danielsson in Pauli 1887:123 (autopsy);
Bormann CIL XI.3162b,5; Deecke 1888:180-1 ¢49²; Conway 1897:382 ¢339²; Herbig CIE 8359
(autopsy); Buonamici 1913:70-1 ¢33²; Lejeune 1952b:119; Vetter 1953:314 ¢324c²; G. Giacomelli
1963:96-7 ¢123,III² (autopsy). Drawings: Dressel 1881 between pp.156-7 (reproduced in IIM tab. XI.4,
Deecke 1888 Taf. II, CIE 8359); Herbig CIE 8359.
333. Painted in black along the front of a tile (68 u 47.5 cm; let. 3.5-7 cm). Dressel was
uncertain as to which side of the tile was the upper or the lower, eventually deciding by
a deposit of mud on the inside of one of the flanges. This may be wrong: I have
therefore included my drawing (fig. 16.4) either way up
[..]folcosio or *****oi
*****oi [..]folcosio
If the tile is held as Dressel proposed, the inscription is sinistroverse: if it is held the
other way up, dextroverse. In either case, the alphabet is Faliscan, but in the second case
the letters of folcosio appear more normal:
Fig.16.3. Author’s drawings of MLF 334.

left: Tracing of the inscription in the orientation proposed by Dressel.
right: Tracing of the inscription read the other way up.
Nothing remains of the praenomen preceding folcosio, read as anc... (Schneider), ụ

(Deecke 1888), c ̣ẹ (Herbig, G. Giacomelli), and ẹc ̣ (Vetter). The gentilicium seems to
have been written with f (Deecke 1888, Vetter) rather than h (Herbig, G. Giacomelli).
544
The line ending in -oi is incomprehensible (uo uictoi Schneider; ḷoụ ṭiṭoi Herbig,
whence Buonamici and Vetter, ḷou ṭitoi G. Giacomelli). Because of this -oi, the text is
usually interpreted as ‘... Folcosius (made this grave) for ...’ (§8.10.2): if the forms in
-oi can be genitive, as I assume (§4.4.4), it may be a filiation. In both interpretations this
line would be the second, implying that Dressel’s orientation of the tile is correct.
From autopsy in the Museo Civico, Viterbo (inv. 90). Bibliography: Dressel 1881 (autopsy); Deecke
1881; Zvetaieff IIM 78; Schneider 1886:106 ¢16²; Zvetaieff III 80; Bormann CIL X sub 3162b; Deecke
1888:182-3 ¢52²; Conway 1897:383 ¢342²; Herbig CIE 8361 (autopsy); Buonamici 1913:71 ¢35²; Her-
big 1914a:239; Lejeune 1952b:125; Vetter 1953:314-5 ¢324e²; G. Giacomelli 1963:96-7 ¢123,V²
(autopsy); Drawing: Dressel 1881 between pp.156-7 (reproduced in IIM tab.XI.7, Deecke 1888 Taf.III,
334. Painted in black along the front of a tile (66.5u45 cm; letters 7-10.5 cm high).
cauia
uetulia
Sinistroverse, Faliscan alphabet. Cf. 335.

Deecke 1881:237; Zvetaieff IIM 72; Schneider 1886:105 ¢10²; Zvetaieff III 74; Bormann CIL
XI.3162b,2; Deecke 1888:184 ¢54²; Conway 1897:383 ¢344²; Jacobsohn 1910:5 ¢36²; Herbig CIE 8363;
Vetter 1953:315 ¢324g²; G. Giacomelli 1963:96-7 ¢123,VII² (autopsy). Drawing: Dressel 1881 between
pp.156-7 (reproduced in IIM tab.XI.2, Deecke 1888 Taf.III); Herbig CIE 8363.
335. “Un tegolone oggi perito, poichè rotto in più pezzi fu dallo scopritore gettato via
come inutile, avea, secondo la copia fattane dallo stesso colono, la seguente inscrizione:


Non è difficile emendare la trascrizione in


ed avremmo quindi uno stretto congiunto della sopra” (Dressel 1881:157). I find this
story strange. Is it likely that Jannoni would have discarded a broken inscription when
according to Bazzichelli he preserved a number of broken tiles, both inscribed and
plain? And if he found the tile already broken, why would he puzzle it together to make
the apograph, only to discard the original? And if the tile was not broken when he found
it, why would he have made an apograph?282 Maybe an inaccurate sketch of 334 was
mistaken (by whom?) for an apograph of another, presumably ‘lost’, inscription.
Bibliography: Dressel 1881; Zvetaieff IIM 73; Schneider 1886:105 ¢11²; Zvetaieff III 75; Bormann CIL
XI.3162b,3; Deecke 1888:183 ¢53²; Conway 1897:383 ¢343²; Jacobsohn 1910:5 ¢35²; Herbig CIE 8364;
Vetter 1953:315 ¢324h²; G. Giacomelli 1963:96-7 ¢123,VIII². Transcription: Jannoni in Dressel
1881:157 (reproduced in IIM p.63, Deecke 1888 Taf. III, CIE 8364).
282
The similarity of the names is not suspect: cf. cauio  uecineo LF 224 and LF 225 and cau[ia
 ]uecin[e]a LF 222 = cauia  uecinea LF 223, all from one tomb near S. Maria di Falleri.
545
CHAPTER 16
336. Painted in black on the front of a tile (67u45 cm; letters 10-14 cm high).
titomarhio
uoltilio
Sinistroverse, Faliscan alphabet. H is ; the letters io are written vertically downwards
under the line: the point after uoltilio serves to separate the word from these letters. The
early editors read the h as , an e combined with a ‘vertical hyphen’ (mare|io Dressel,
Danielsson, Schneider) or an i (mareiio Jordan, Deecke 1881, Zvetaieff, Bormann);
marh|io was first read by Deecke (1888). Marh|io may perhaps stand for Marcius
(Herbig, Hirata 1967:60-1), but cf. Campano-Etruscan marhies Cm 6.1 (G. Gia-
comelli): see §7.8.1.99.
From autopsy in the Museo Civico, Viterbo (inv. 88). Bibliography: Dressel 1881 (autopsy); Jordan
1881:511-2; Deecke 1881:237; Zvetaieff IIM 74; Schneider 1886:106 ¢12²; Zvetaieff III 76; Danielsson
in Pauli 1887:123; Bormann CIL XI.3162b,4; Deecke 1888:182 ¢51²; Conway 1897:382 ¢341²; Herbig
CIE 8362 (autopsy); Buonamici 1913:72 ¢36²; Vetter 1953:315 ¢324f²; G. Giacomelli 1963:96-7
¢123,VI² (autopsy); Stuart-Smith 2004:60-1. Drawing: Dressel 1881 between pp.156-7 (reproduced in
Jordan 1881:511, IIM tab.XI.3, Deecke 1888 Taf.III); Herbig CIE 8362.
337. Painted in black on a tile (68.7u47 cm; letters 9.8-10.7 cm high).283

[1][---?] 2pol[3][---]
Sinistroverse, Faliscan alphabet. Possibly an abbreviation p (or [po]p, which would
require a missing tile that contained the letters [po]) followed by a gentilicium ol[---]
(cf. perhaps olna MF 82); hardly ol[cosi---], as there are no certain instances of
omission of word-initial h- in Faliscan (§3.5.2). Deecke’s f]ol[cozeo is impossible.
Frustra quaesiui284 in the Museo Civico, Viterbo, on two occasions in 1990. Bibliography: Dressel
1881 ¢8² (autopsy); Zvetaieff IIM 79; Zvetaieff III 80; Bormann CIL X sub 3162b; Deecke 1888:184
¢55²; Conway 1897:383; Herbig CIE 8365 (autopsy); Vetter 1953:315 ¢324i²; G. Giacomelli 1963:96-7
¢123,IX² (autopsy). Drawing: Herbig CIE 8365.
16.6. The site at Grotta Porciosa
The Grotta Porciosa site lies c.3 km to the south-east of Gallese and c.2 km to the
north-west of Borghetto, between the Fosso delle Rote and Rio Fratta to the north and
the Fosso di S. Silvestro to the south. The remains were first described by Dennis
(1878:120-2) as the probable site of Fescennium (but cf. §2.1.2); other surveys have
appeared in FI II.1 p.48-52 with map p.53 (‘pagus del Comunale o di Lucciano’) and in
Ward Perkins & Frederiksen 1957:171-6 with map p.175 fig.28. The site seems to have
283
In Herbig’s drawing, the letters are thin and reminiscent of those of cẹịṣ[i.] | holc[osi.] | ar f̣
[...] MF 140 from Civita Castellana.
284
The tile was seen by Herbig in the Museo Civico in 1903. G. Giacomelli, who also saw these
tiles, does not mention that it was missing, while noting that she failed to find 329 and 332.
546
been a major settlement of the north-eastern ager Faliscus, owing its importance
perhaps to the proximity of the Tiber crossing. The inscriptions in Latin alphabet imply
that it continued to exist after the war of 241, perhaps because the Tiber crossing was
then connected to the Via Flaminia (§2.5.2), cf. Cifani 2002. Of the inscriptions from
this location, MLF 338-345, attributed to tombs discovered “ad oppidulum Gallese in
regione quae Sa Lucia vocatur, secundum viam” (Herbig CIE p.88), are in all probabil-
ity from the tombs along the ancient road running through the S. Lucia estate directly to
the north of the Grotta Porciosa site (Ward Perkins & Frederiksen 1957:171). From the
tombs on the north side of the Rio Fratta is a rock-cut sepulcral inscription MLF 346.
Another group of inscriptions (MLF 347-355) was discovered, according to Magliulo
(in Herbig CIE p.35), during an excavation in 1890 by a ‘sign. Kamelli’ at ‘Grotta
Pulciosa’. Although placed among the inscriptions from the Valsiarosa necropolis at
Civita Castellana by Herbig (CIE 8196-8204), the rediscovery of some of the inscrip-
tions in or shortly before 1974 shows that they indeed belonged to the site at Grotta
Porciosa.
338. Scratched across on the back of a tile (69u48 cm; letters 6-9 cm high).
tana
lartia
Sinistroverse, Faliscan alphabet. The first t, of which the top half is preserved, was ,
the second . Lartia is perhaps a gentilicium rather than a patronym.
Bibliography: Herbig CIE 8401 (autopsy); [Della Seta 1918:105 (autopsy)]; Vetter 1953:322 ¢341a²;
[Ward Perkins & Frederiksen 1957:171]; G. Giacomelli 1963:104 ¢141²; Cristofani 1988:20. Draw-
339. Painted in red on the back of a tile (69u48 cm; letters 5.5-9 cm high).
ueltur
ortecese
Sinistroverse, Faliscan alphabet. The last e is written under the line upside down in
boustrophedon; Cristofani’s Orteces appears to be impossible. Herbig interpreted
ortecese as *Horticensis; Vetter connected it with Horta, modern Orte, to the northeast
of the ager Faliscus. G. Giacomelli (1963:209) rejects these derivations, among other
reasons because in Faliscan word-initial h- is never omitted (§3.5.2). The name may
well be connected to urtcsnas Etr XXXV from Corchiano (Colonna): both names point
to a toponym *Hortica/*Horticum, perhaps the original name of Corchiano, the oldest
attested form of which is Orclanum (from *Horticulanum?): see §6.5.11, §7.8.1.108.
Bibliography: Herbig 1910:191-2 ¢31² (autopsy); Jacobsohn 1910:6 ¢47²; Herbig CIE 8402; Buonamici
1913:76-7 ¢46²; [Della Seta 1918:105 (autopsy)]; Vetter 1953:322 ¢341b²; [Ward Perkins & Frederiksen
1957:171]; G. Giacomelli 1963:104 ¢142²; Cristofani 1988:20; Colonna 1990:136. Drawing: Nogara in
CIE 8402.
547
CHAPTER 16
340. Painted in red along the length of the back of a tile (68u47 cm; letters 7.5-11 cm
high).
cnero ni
Dextroverse, Latin alphabet with cursive e (). The last two letters (erroneously read
together as m by Bormann) are written slightly apart, near the edge of the tile. The
gentilicium Neronius is also attested in LtF 325 and 328 from Fabbrica di Roma and in
MF 15 and 16 from Civita Castellana.
8404 (autopsy); [Della Seta 1918:105 (autopsy)]: Bormann CIL XI.7524 (autopsy); Vetter 1953:323
¢341e²; [Ward Perkins & Frederiksen 1957:171]; G. Giacomelli 1963:266 ¢XXIII². Drawing: Herbig
CIE p.89.
341. Two tiles, the first (68u46 cm; letters 9-14.5 cm high) painted in red on plaster
lengthwise along the front, the other (68u45 cm; letters 11.5-16 cm high) painted in red
lengthwise directly onto the front of the tile.
1 2?
maco[ ]nioia*
1 2?
rutilce[ ]iliacọ*
Dextroverse, Latin alphabet with cursive a () and e (). The o in the second line of the
second tile, omitted by Bormann and Safarewicz, is . Both lines end in vague traces.
The first tile could be read by itself as ‘M. Acus Rutil(us), Cae(sii f.)’, but despite the
differences in the ways in which they are painted and in the size of the letters, the tiles
are usually read together as ‘M. Aconius Ia(nti) f(ilius) Rutil(us); Caelia co(niunx)’. Co
= ‘co(niunx)’ (Herbig) is unexpected, however, since both in the Faliscan and in the
Latin inscriptions, the normal or formulaic word is uxor (§7.4.2). In Vetter’s cẹ ‘Cae.
(f.)’ the e would have a different form than the cursive e of the first tile. A similar case
of aco or aco[n]eo occurs in LtF 327 from Fabbrica di Roma, cf. also the possible
reading ạcoṇẹo in LtF 290.
Bibliography: Herbig CIE 8403a-b (autopsy); [Della Seta 1918:105 (autopsy)]; Bormann CIL
XI.7523a-b (autopsy); Vetter 1953:322-3 ¢341c-d²; Safarewicz 1955:186 ¢9²; [Ward Perkins &
Frederiksen 1957:171]; G. Giacomelli 1963:266 ¢XXII,i-ii². Drawing: Herbig CIE 8403a-b.
342. Painted in white across the back of a tile (70u50 cm; letters 15-21 cm high) “con
avanzi di una riquadratura a colore giallo” (Museo di Villa Giulia inventories, quoted
by Herbig).
[---]ln[---]
Dextroverse, Latin alphabet. Herbig read l( ) n(eroni) or, apparently holding the tile the
other way up, p( ) n(eroni).
Bibliography: Herbig CIE 8405 (autopsy); Bormann CIL XI.7526c (autopsy); [Ward Perkins &
Frederiksen 1957:171]. Transcription: Herbig CIE 8405.
548
343. Painted in red across the back of a tile (68u48 cm).

[---]roc ̣[---]
Dextroverse, Latin alphabet; ░░ Herbig. The c could perhaps be an o.
Bibliography: Herbig CIE 8406 (autopsy); Bormann CIL XI.7526a (autopsy); [Ward Perkins &
Frederiksen 1957:171]. Transcription: Nogara in CIE 8406.
344. Painted in white on a tile fragment (letters 17 cm high).

[---]er[---]
Dextroverse, Latin alphabet. Bormann read ero.
Bibliography: Herbig CIE 8407 (autopsy); Bormann CIL XI.7525 (autopsy); [Ward Perkins &
345. Painted in red on plaster across the back of a tile (67u45 cm).
[---]*f ̣[---]
Dextroverse, Latin alphabet. The trace is shown as ( ; the f, , could also be an e.
Bibliography: Herbig CIE 8408 (autopsy); Bormann CIL XI.7526b (autopsy); [Ward Perkins &
346. Pasqui’s description of the location of this inscription is unclear: it apparently

belongs to one of the tombs in the north side of the gorge of the Rio Fratta, opposite the
Grotta Porciosa site (cf. Ward Perkins & Frederiksen 1957:176). “Una tomba tuttora
aperta e in balia delle intemperie che presto le danneggeranno ancora di più. Trovasi
essa molto più a valle delle precedenti in luogo detto il Pontone di Costanzo. E` incava-
ta sul masso, con entrata in piano, stretta e poco regolare. La porticella arcuata mette as
una camera rettangolare larga m. 5,45 lunga m. 2,20 entro il quale furono deposti i
cadaveri entro ventiquattro loculi chiusi da tegoli e dentro a piccolo loculo che doveva
contenere un ossario. Sopra al loculo a destra della fila più alta nelle pareti di fronte è
incisa a lettere regolari alte mm. 55:” (FI II.1, p.68)
mtitotuliouoltiliohescuna
Sinistroverse, Faliscan alphabet. Colonna finds the m problematic: the most obvious
interpretation seems to be to take both m and tito as praenomina, followed by a
gentilicium, a patronym, and a second gentilicium in the singular,285 possibly because
the m was added later (as in m  t  u  genucilio Cap 435). Such second gentilica belong
to the onomastic formula of freedmen: see §7.6. Hescuna may be a derivation from the
same root as the Faliscan toponym Fescennium, as Colonna suggested (§6.5.10).
Bibliography: Pasqui in FI II.1 p.67-8 (autopsy); Colonna 1990:123 n.52. Drawing: Pasqui in FI II.1
p.68 (reproduced upside down in Colonna 1990:124 fig.4).
285
Also in m c pompilio CIL I2.30, q k cestio CIL I2.61, and q a aidicio CIL I2.2442.
549
CHAPTER 16
347-353 (‘tomba degli Arati’). The following inscriptions were painted between the
loculi of a tomb excavated in 1890 by a ‘sig. Kamelli’ and Magliulo, and were known
for more than 80 years only through Magliulo’s apographs. These were emended by
most editors because of their peculiar letterforms. In 1974, however, the tomb and six of
the inscriptions were rediscovered during an excavation of the Centro Cattolico
Archeologico Romano near the Grotta Porciosa site. The photographs published by
Renzetti Marra (1974) show that Magliulo’s apographs were quite accurate, his main
mistake being the rendering of the peculiar t, , as , which had been read as c.
347. Painted in white between two loculi (length 170 cm; letters 15-17 cm high).
tanncuilaratia
Sinistroverse, Faliscan alphabet. A is , t is . Tanncuil is an error for tan‹a›cuil.
Bibliography: (I) Jacobsohn 1910:4 ¢20²; Herbig CIE 8198; Vetter 1953:298 ¢278a²; G. Giacomelli
1963:78 ¢76,I². Drawing: Magliulo in CIE 8198. – (II) Renzetti Marra 1974:355-7 ¢1² (autopsy).
Photograph: Renzetti Marra 1974 tav.LVIII,a.
aufiloaratio
Sinistroverse, Faliscan alphabet. The t is .
Bibliography: (I) Jacobsohn 1910:4 ¢20²; Herbig CIE 8199; Vetter 1953:298 ¢278b²; G. Giacomelli
1963:78 ¢76,II². Drawing: Magliulo in CIE 8199. – (II) Renzetti Marra 1974:355-7 ¢2² (autopsy).
Photograph: Renzetti Marra 1974 tav.LVIII,b.
349. Painted in white between two loculi (length 90 cm).

cauioaratio
Sinistroverse, Faliscan alphabet. The t is .
Bibliography: (I) Jacobsohn 1910:4 ¢20²; Herbig CIE 8201; Vetter 1953:298 ¢278d²; G. Giacomelli
1963:78 ¢76,IV². Drawing: Magliulo in CIE 8201. – (II) Renzetti Marra 1974:355-7 ¢4² (autopsy).
Photograph: Renzetti Marra 1974 tav. LIX,a.
350. Painted in white between two loculi (length 90 cm; letters 15 cm high).
titoartio
Sinistroverse, Faliscan alphabet. The first t is , the second and third are . Artio is an
error for ar‹a›tio.
Bibliography: (I) Jacobsohn 1910:4 ¢20²; Herbig CIE 8203; Vetter 1953:298 ¢278f²; G. Giacomelli
1963:78 ¢76,VI². Drawing: Magliulo in 8203. – (II) Renzetti Marra 1974:355-7 ¢6² (autopsy). Photo-
graph: Renzetti Marra 1974 tav. LIX,c.
caisiotirio
Sinistroverse, Faliscan alphabet with reversed s. For the gentilicium, cf. MLF 358.
Bibliography: (I) Herbig CIE 8202; Vetter 1953:298 ¢278e²; G. Giacomelli 1963:78 ¢76,V². Drawing:
Magliulo in CIE 8202. – (II) Renzetti Marra 1974:355-7 ¢5² (autopsy). Photograph: Renzetti Marra
1974 tav.LIX,b.
550
352. Painted in white between two loculi (length 70 cm; letters 15 cm high).
f aino
Sinistroverse, Faliscan alphabet. The distance between the f and anio is usually
disregarded, and the whole read as one word connected with Latin Faenius. The
photograph clearly shows a space after the f, and Renzetti Marra rightly reads f aino: cf.
f pacios LtC 392. In Magliulo’s apograph, the first letter of the second word is  (ạ
Vetter), but Renzetti Marra’s photograph shows that Herbig’s a is definitely correct.
Bibliography: (I) Herbig CIE 8200; Vetter 1953:298 ¢278c²; G. Giacomelli 1963:78 ¢76,III². Drawing:
Magliulo in CIE 8200. – (II) Renzetti Marra 1974:355-7 ¢3² (autopsy). Photograph: Renzetti Marra
1974 tav.LVIII,c.
353. This inscription was ascribed by Magliulo to the same tomb, but was not found
when this was re-excavated.
16.4. Magliulo’s apograph of MF 353.

(From CIE 8204.)
oct*i[....]uoltili
Sinistroverse, Faliscan alphabet? Similar letters are found in MLF 333. Like Latin
names in Oct-, the praenomen, probably abbreviated to oct, is derived from octo
(Herbig in fact read oc ̣ṭọ ‘Octo’) or octauus (oct(a)ui?). Uoltili is probably a patronymic
adjective rather than a gentilicium, in which case the gentilicium can only be very short
(e.g. [aino], cf. aino in MLF 352 from the same tomb?). Uoltili is an abbreviated
nominative or a genitive: it is not necessary to restore uoltili[o] (thus Herbig). I fail to
see how Vetter’s ụcsọṛ [caui] uoltili can be derived from Magliulo’s apograph; also,
there are no Faliscan examples of a woman being described as ‘the wife of ...’ without
her own name being mentioned (§7.4.2), or of cs for x.
Bibliography: (I) Herbig CIE 8204; Vetter 1953:29 ¢278g²; G. Giacomelli 1963:78 ¢76,VII². Drawing:
Magliulo in CIE 8204 – (II) Renzetti Marra 1974:355-7 ¢7² (autopsy).
354. Scratched on a tile found in 1890 during the excavation by Kamelli and Magliulo.
Known only through Magliulo’s apograph.
titopolafio
Magliulo’s apograph gives the p as . Herbig doubted the likelihood of a gentilicium
polafio (CIE) and (1914) read tito p ̣ola fio ‘Titus Pola filius’ (=‘Titus Pola jr.’) with fio
“wohl sicher über *fiius aus *fil’ius” (1914b:251). This was adopted by G. Giacomelli
551
CHAPTER 16
with pola as a genitive pola(s) (cf. §4.2.2, §3.5.7c) and by G. Giacomelli. The palatali-
sation of /l/ on which these interpretations are based appears to have no parallels in the
Faliscan material, however (§3.5.5.3): spellings such as hileo MF 161 and filea MF 14
show that the word was /fīlios/ and not /fīlos/: see §3.6.2. It is therefore better either to
adopt Magliulo’s tito polafio (cf. the Masofius and Patrufius adduced by G. Giacomelli)
or to emend to pol‹fa›io, a derivation of names in Pulf- (Hirata): see §7.8.1.123.
Bibliography: Herbig CIE 8196; Herbig 1914b:251; Herbig 1923:231; Vetter 1953:298 ¢277a²; G.
Giacomelli 1963:77 ¢74²; Hirata 1967:68; G. Giacomelli 2006:92. Drawing: Magliulo in CIE 8196.
355. Scratched on a tile found in 1890 during the excavation by Kamelli and Magliulo.
Known only through Magliulo’s apograph.
manileo
Sinistroverse: Magliulo’s apograph gave the m as  and the n as , and the l reversed as
. Either manileo ‘Manilius’ or m anileo ‘M. Annilius’ vel sim..
Bibliography: Herbig CIE 8197; Vetter 1953:298 ¢277b²; G. Giacomelli 1963:77-8 ¢75². Draw-
ing: Magliulo in Herbig CIE p.35.
16.7. The area around Gallese and Borghetto
356-357. In a ‘tenuta Paciano’ between Gallese and Borghetto, Nogara found two
inscriptions (“alter m. 0,69 a. × 0,52 l., alter 0,45 a. × 0,88 l.” Herbig CIE 8598-8599)
cut in the right-hand wall of a cuniculus leading to the bottom of the gorge of the Rio
delle Rote. Ward Perkins & Frederiksen (1957:173-4) identified the spot with a tomb
c.350 m to the north of Casale Paciano, 2 km to the west of the Grotta Porciosa site, but
refer to the inscriptions as ‘inscribed tiles’. In Nogara’s transcriptions the texts are:
�� ̣ ueiila ��░�� me[.]uela
���̣�� ioiti ���̣░ u[.]suae:rfa
��̣�� pima ���̣�� zeuoc*na
The second inscription is read as me[ ]uela | u[ ]svae : dfa | *evoc*na by G. Giacomelli.
Both are incomprehensible; the latter may end in an Etruscan gentilicium in -na.
Bibliography: Herbig CIE 8598-8599; [Ward Perkins & Frederiksen 1957:173-4, 176]; G. Giacomelli
1963:72 ¢65a-b². Transcriptions: Nogara in CIE 8598-8599.
358-359. Two inscriptions that are described as “dipinti, se ho ben interpretato, su tegoli
sepolcrali” (G. Giacomelli 1977:68). Apart from the text, and the fact that they were
found near Borghetto (cf. the anonymous reference in SE 41 (1973), pp.541-2), no data
are given.
cailiotirio
titobatio
552
If the alphabet is Faliscan, the use of b is indeed surprising (§11.2.2-4) and would
probably point to Latin or Sabellic influence. The gentilicium tirio occurs also in
caisio  tirio MLF 351 from the nearby Grotta Porciosa site; G. Giacomelli for some
reason regarded it as a patronym. He compared batio to Battius in CIL VI.6740 and
6757 from Rome.
Bibliography: R.G. Giacomelli 1977:68-9; Marinetti 1982:36.
16.8. Northern ager Faliscus, exact provenance unkown
360. Painted in red across the back of a tile (67u48 cm; letters 7-8 cm high). The tile is
broken in two: Thulin reports that according to the Museo di Villa Giulia inventories
one half was found at Fabbrica and the other at Gallese (i.e., near the site at Grotta
Porciosa).286
cauia
hadenia
Sinistroverse, Faliscan alphabet. There is no trace of the interpunct read by Herbig
(whence all later editors) before hadenia: as the inscription is very well preserved, it is
more likely that it never existed than that it has disappeared since Herbig saw it in 1903.
1907:294-5 ¢33² (autopsy); Jacobsohn 1910:6 ¢46a²; Herbig CIE 8371 (autopsy); [Della Seta 1918:103
(autopsy)]; Vetter 1953:316 ¢326²; G. Giacomelli 1963:98 ¢126². Drawings: Thulin 1907:295 (repro-
duced in CIE 8371); Herbig CIE 8371.
361. Scratched along the length of the back of a damaged tile. Perhaps from the area of
Fabbrica di Roma (Renzetti Marra).
] precono[
] cuiteneṭ[
] let [
Sinistroverse, Faliscan alphabet. The first line ends in the lower half of a shaft, placed
lower than the rest of the line. It is probably a t, to be read with tene in the following
line. The third line is followed by a vacant space of seven or eight letters; why Renzetti
Marra reads let[ is not explained. Tenet would appear to be ‘tenet’, let an abbreviation
the same word that occurs in the plural as lete in MLF 285: pace Mancini, this is in all
probability the same word as Latin lectu in Lat 251 (see §6.3.39). It is tempting to
interpret cui as relative a pronoun (§4.9), but this presupposes a merger of /ē ̣/ (m */o
e/) with */ī/ that is not attested even in the Late Faliscan inscriptions (§3.7.5) and
286
“Wenn diese Angabe richtig ist, hat wohl eins von beiden im Packsatttel eines Esels als
Ballast die Reise von einen Ort zu dem anderen gemacht” (Thulin 1907:295).
553
CHAPTER 16
would have been very surprising even in a Latin inscription until c.150 BCE. If cui is a
relative pronoun at all, it is probably a nominative (e.g. “‘Preconio il quale occupa il
letto’”, Renzetti Marra 1990:337). I cannot adopt Renzetti Marra’s suggestion that it is a
locative (“‘Preconio qui/dove occupa il letto’”, 1990:338): in Faliscan, ‘here’ is he(c) or
fe(c), and relative ‘where’ will hardly have been cui in view of Latin ubi, Oscan puf Po
34-37, and Umbrian pufe TI Ib.33, VIa.8, VIb.50, VIIa.43. The third possibility given
by Renzetti Marra, taking cui as a dative ‘cui’ (“‘Preconio cui (al quale) tocca il letto’”,
1990:388), requires an unparallelled construction for tenet.
Bibliography: Renzetti Marra 1990:336-7 ¢B,1² (autopsy); Rix 1993a; M. Mancini 2002:28-33.
Drawing: Renzetti Marra 1990:336.
362. Scratched on a tile seen in 1907 by Bormann in the Veroli collection, Caprarola
(letters 8-10 cm high). Vetter gave the vicinity of Corchiano, Vignanello, or Carbog-
nano as a likely provenance, as other pieces in the Veroli collection were from that area.
iatasenθia
Sinistroverse, Faliscan alphabet. Veroli’s drawing shows the s as an  that appears to be
a slip for the normal . The θ is shown as , which in the ager Faliscus is very rare: cf.
§11.2.4. Cf. [---]nθia MLF 212. Note the omission of syllable-final /n/ in iata, but not
in senθia.
Bibliography: Herbig CIE 8586; Vetter 1953:323 ¢342²; G. Giacomelli 1963:97-8 ¢124². Draw-
ing: Veroli in CIE 8586.
554
Chapter 17
The south-eastern ager Faliscus and the ager Capenas
17.1. The inscriptions from the south-east
As said in §2.1.1, the exact borders between the ager Faliscus and the ager Capenas
cannot be established. If the alphabet and the language of the inscriptions is any clue,
the most significant non-Etruscan site to the west of Mount Soracte, Rignano Flaminio
(§17.2, MLF 363-375 and Etr XLIII), was Faliscan rather than Capenate, and the same
seems to be true for the area around Monte Soratte and the land lying between it and the
Tiber, to which belong S. Oreste (§17.3, MF 376), Ponzano (§17.4, LtF 377), Civitella
S. Paolo (§17.5, LF 378-384), and Fiano Romano (§17.6, EF/Etr 385). The inscriptions
from these sites can not be connected to the war of 241 BCE and have therefore all been
classed as Middle or Late Faliscan or Capenate (cf. §11.1.3), unless they can be dated in
other ways. The ager Capenas (at least in the linguistic, but perhaps also in the geopo-
litical sense) would then have occupied only a relatively narrow strip along the west
bank of the Tiber, from the Tiber crossing near Lucus Feroniae in the south to some
unknown point southeast of Mount Soracte. A description of the ager Capenas is given
in G. Jones 1962 with map pl.XLIII. Its linguistic history seems to have been defined by
influences from the surrounding areas: see §9.3.3. Apart from the inscriptions presented
in this chapter, three inscriptions of unknown origin may also be Capenate: pa‹qu›is
blaisiís Sab 468*, iunai MLF/Cap 475*, and setorio MLF/Cap 476*.
17.2. Rignano Flaminio
Rignano Flaminio lies on the western slopes of Soracte, close to the Via Flaminia and
its precursors, the main connection between the ager Faliscus and the south. It is not
clear whether it belonged to the ager Faliscus or the ager Capenas; the inscriptions at
least are all either Faliscan (MLF 363-375; MLF 371-372 have been regarded as
Etruscan) or Etruscan (Etr XLIII) The inscriptions from this site were all found in 1911
during the excavation of a group of tombs discovered that year on the south-east side of
the nearby Monte Casale (cf. Gabrici 1912b:75).
363-366. The following inscriptions were found in tomb 1. As this had been filled in
with debris from other tombs (Gabrici 1912b:75), not all the inscriptions found in this
555
CHAPTER 17
tomb need necessarily have belonged there. From the same tomb is umrie XLIII. The
gentilicia in this tomb, Umbrius and Umbricianus, are probably derived from the
ethnonym of the Umbrians (§7.8.1.158-159, §6.5.18), which is interesting in view of
the Sabellic epigraphical and linguistic features in the ager Capenas.
363. Painted in red on a fragmentary tile (letters 7.5-17 cm high).
[iu]naupreciano
Sinistroverse, Faliscan alphabet. The p is a heavily splayed and tilted , as it is in
364 from the same tomb; the r is  rather than  (§11.2.4.2). The letters ia are written
vertically downwards under the line and the letters no upside down in boustrophedon
underneath the line for lack of space. The gentilicium occurs also in MLF 364: cf. also
umrie XLIII from the same tomb, and u]mpricius Lat 219 from S. Maria di Falleri.
Bibliography: Herbig CIE 8430; Gabrici 1912b:75-6 (autopsy); Buonamici 1913:80 ¢51²; [Della Seta
1918:104 (autopsy)]; Vetter 1953:324 ¢343b²; G. Giacomelli 1963:107-8 ¢145,I² (autopsy). Drawing:
Nogara in CIE 8430; Gabrici 1912b:76 fig.2.
364. Painted in red on a fragmentary tile (letters 10-15 cm high).

[---]upreciano
Sinistroverse, Faliscan alphabet. The p is a heavily splayed and tilted , as in 363 from
the same tomb; the r is  rather than  (§11.2.4.2). The final o is written under the line
for lack of space. For the gentilicium, see under 363.
Bibliography: Herbig CIE 8431; Gabrici 1912b:75-6 (autopsy); Buonamici 1913:80-1 ¢52²; [Della Seta
1918:104 (autopsy)]; Vetter 1953:324 ¢343a²; G. Giacomelli 1963:107-8 ¢145,II² (autopsy). Drawing:
Nogara in CIE 8431; Gabrici 1912b:75 fig.1.
365-366. The titulus prior was painted in red (“rubro colore paene evanido”, Herbig
CIE 8429) on twelve tile fragments (letters 15-16 cm high).
[---]are*[---]
Sinistroverse, Faliscan alphabet. The trace at the end is the bottom of an e or an a.
Apparently painted directly over this, without covering the titulus prior with a layer of
plaster (which is apparently unique, see §11.1.4.1c), was the titulus posterior (letters
12-15 cm high), also in red (“rubro colore melius conservato”, Herbig CIE 8429).
[---]uịnụ[---]
Sinistroverse, Faliscan alphabet. Gabrici read unil (dextroverse, with reversed n). The i
is thin and stands close to the u. Of the last letter, only the bottom halves of two shafts
are preserved: these may have belonged to separate letters. Perhaps uịnụ[---] is a
gentilicum uịnụ[cio ‘Vinucius’?
Bibliography: Herbig CIE 8429a-b; Gabrici 1912b:75-6 (autopsy); [Della Seta 1918:104 (autopsy)];
Vetter 1953:324 ¢343d²; G. Giacomelli 1963:107-8 ¢145,IVa-b² (autopsy). Drawing: Nogara CIE
8429a-b; Gabrici 1912b:76 fig.4.
556
THE SOUTHEASTERN AGER FALISCUS AND THE AGER CAPENAS
367-375. The following inscriptions are from the fourth- or third-century tomb 5.
367-370. Scratched in four black-varnished saucers: (367) height 4.2 cm, 10 cm,
letters 4-5 mm high; (368) height 65 mm, 108 mm, letters 6-9 mm high; (369) height
57 mm, 133 mm, let. 8-15 mm high; (370) height 60 mm, 140 mm, letters 7-8 mm
high. The first is reminiscent of fifth-century ware (Gabrici 1912a:78-9).
uoltai
uoltai
uoltai
uoltai
Sinistroverse, Faliscan alphabet. Genitive (thus Pisani) or dative: see §8.8.1.
Bibliography: Herbig CIE 8435-8438; Gabrici 1912b:79-81 (autopsy); [Della Seta 1918:104 (au-
topsy)]; Buonamici 1913:77-8 ¢48²; Vetter 1953:325 ¢350²; G. Giacomelli 1963:65 ¢52,I-IV²; Pisani
1964:344 ¢146G,c². Drawing: Nogara CIE 8435-8438; Gabrici 1912a:79,81.
371-372. Scratched in two black-varnished plates (height 5.5 cm, 13.5 and 14.5 cm
respectively; letters 7-12 mm high).
larise uicina
larise
uicina
Sinistroverse, Faliscan alphabet. In 372, the first a is squarish. The e is upside down and
() in 371 and upside down and reversed ( or ?) in 372. Rix erroneously gives both
texts as larise | uicina. The -e in larise (also in larise MF 270) has been regarded as an
epenthetic [e] or [ə] after an /s#/ that was pronounced more strongly than in Faliscan
(§3.5.7d), but is perhaps rather an accusative larise(m) used as a nominative: see §9.2.2.
Bibliography: Herbig CIE 8439-8440; Gabrici 1912b:80 (autopsy); [Della Seta 1918:104 (autopsy)];
Buonamici 1913:78 ¢49²; Vetter 1953:325 ¢351a-b²; G. Giacomelli 1963:65 ¢53,I-II²; Cristofani
1988:21; Rix ET Fa 2.22-23. Drawing: Nogara in CIE 8439-8440; Gabrici 1912b:80.
373-375. Scratched under three black-varnished saucers (height 7 cm, 17 cm, letters
18 mm high; height 5.5 cm, 14.5 cm, letters 10 mm high; height 6 cm, 15.5 cm,
letters 10 mm high respectively):
la
la
la
Sinistroverse. In 373, a is , in 374 a is  (but not, it would appear, a v); 375 is
(uo Gabrici, ụọ? Herbig), with a sign very similar to the first o in Cap 388.
Bibliography: Herbig CIE 8441-8443; Gabrici 1912b:81; G. Giacomelli 1963:65 ¢54,I-III². Drawing:
Nogara in CIE 8441-8443; Gabrici 1912b:81.
557
CHAPTER 17
17.3. S. Oreste
376. Scratched in an early to mid-fourth-century red-varnished saucer (height 5.5 cm,

rim 15.8 cm, foot 9 cm; letters 10 mm high).
statiocailio
Sinistroverse, Faliscan alphabet. The l is . The gentilicium recurs in MF 90-97 from
what was apparently a family tomb of the gens Caelia at Civita Castellana.
Bibliography: Colonna 1976b (autopsy); Marinetti 1982:36; Photographs: Colonna 1976b:119
figs.1-2. Drawing: Colonna 1976b:119 fig.3.
17.4. Ponzano Romano
377. Cut on a tuff base (18.5287u38u8.5 cm), damaged at the top, from Contrada Il
Brecceto near Ponzano,288 in the area between Mount Soracte and the Tiber.
[.]ṃuniorecena*
numesiom[art]ẹ
ddlm
Dextroverse, Latin alphabet with cursive e (). Of the initial m, only a trace of the right-
hand shaft remains; of the letters uniorec, only the lower halves are preserved. The o in
the second line is diamond-shaped and open on the lower side. Gatti’s Numesio
M[art]i[o] (adopted by Wachter), for which cf. the dedication mar  popi st  f  n  mart
 d  d  me Cap 420, is preferable to such deities as Munibregenai (Kretschmer), Munio
Recena (Dessau), [se]munib(us) (Egger in CIL I2.2436), or [ter]munib(us) regena-
t(ibus) (Vetter; doubted by Prosdocimi). Ṃunio is the dedicant rather than the deity: the
gentilicium Munius occurs also in CIL XI.3941 from Capena. Restoring m[art]i[o],
however, is impossible, unless the stone is damaged, which from Gatti’s description
does not appear to be the case. As there is not enough space to restore m[arte]i, I read
m[art]ẹ, taking the shaft at the end of the line as part of a cursive e. The similar shaft at
the end of first line is usually read as i, but this makes no sense: perhaps it is not a letter
at all, in which case we would have a cognomen (or a second gentilicium?) recena.289
Bibliography: Gatti 1906 (autopsy); Kretschmer 1917:137-9; Bormann CIL XI.7762; Dessau 1916:32.C
¢9231²; Diehl 1911:14 ¢138²; Lommatzsch CIL I2.2436; Diehl 1930:17 ¢160²; Vetter 1953:330-1 ¢361²;
Degrassi ILLRP 290; Prosdocimi 1962:758 n.4; Briquel 1972:823 n.7; Degrassi & Krummrey CIL
I2.2436 add.; Wachter 1987:440. Drawing: Gatti 1906:61.
287
Gatti (1906:61) gave the height as “m. 0,75”, probably an error for ‘m. 0,175’.
288
Kretschmer erroneously gave the provenance as Capena.
289
Reading [o]rṭịcena, (cf. orticese MLF 339 and urtcsnas Etr XXXV) seems impossible.
558
17.5. Civitella S. Paolo
378-382. The following inscriptions were found in 1959 at località Monte Verde near
Civitella S. Paolo. They are all from the third- or second-century tomb VI.
378. Scratched on the bottom of a red-varnished saucer (height 4 cm, rim 12.7 cm,
foot 7.6 cm; letters 7-9 mm high).
madicioeco
Sinistroverse, Faliscan alphabet. The a is . Torelli divided madicio as m adicio,
probably rightly. The formula OWNERNOM ego is until now only attested for Latin
inscriptions, OWNERGEN ego only for Faliscan inscriptions: see §8.9.2.
Bibliography: Torelli 1967:536 ¢1² (autopsy); Briquel 1972:820, 826; Renzetti Marra 1974:350 ¢I²;
Agostiniani 1982:149 ¢588². Drawing: Torelli 1967:53.
379. Scratched on the bottom of a reddish-brown varnished saucer (height 2.7 cm,
rim 7.4 cm, foot 3.8 cm; letters 9-10 mm high).
sceiuai
Sinistroverse, Faliscan alphabet. The a is . For the name, cf. sceua MLF 314. Torelli
found it strange to find the owner designated by the cognomen alone: I rather suspect
that Scaeua was a female praenomen, corresponding in sense to the male praenomen
Laeuius: see §7.7.1. Sceiuai is either genitive or dative, probably the former: see §8.8.1.
The -ei- represents /ę̄/ ← /ai/, although the ending is still written -ai, indicating that the
here the diphthong (/-āi/) or the disyllabic /-āī/ developed differently: see §3.7.6.
Bibliography: Torelli 1967:536-7 ¢2² (autopsy); Renzetti Marra 1974:350 ¢II². Drawing: Torelli
1967:53.
380. Scratched on the outside of a brown-varnished plate (height 1.7 cm, rim 11.7
cm, foot 4.1 cm; letters 7-12 mm high).
seralia
Sinistroverse, Faliscan alphabet. The a is .
Bibliography: Torelli 1967:537 ¢3² (autopsy); Renzetti Marra 1974:35 ¢III². Drawing: Torelli 1967:53.
381. Scratched inside a black-varnished saucer (height 4.9 cm, rim 13.1 cm, foot 5.1
cm; letters 10 mm high).
iun
Sinistroverse.
Bibliography: Torelli 1967:537 ¢4² (autopsy); Renzetti Marra 1974:351 ¢IV². Drawing: Torelli
1967:53.
559
CHAPTER 17
382. Scratched in a red-varnished saucer (height 6 cm, rim 16 cm, foot 8.7 cm; letters
8-14 mm high).
cauios
Sinistroverse, Faliscan alphabet. The a is ; the o is slightly diamond-shaped. This is
one of the very few Faliscan instances where word-final -s after a short vowel is written
out: see §3.5.7d.
Bibliography: Torelli 1967:537-8 ¢5² (autopsy); Renzetti Marra 1974:351 ¢V². Drawing: Torelli
1967:53.
383-384. The following inscriptions are from the third- or second-century tomb XV.
383. Scratched in a red-varnished saucer (height 3.9 cm, rim 12.3 cm, foot 7 cm;
letters 10 mm high).
ecotulie
Sinistroverse, Faliscan alphabet. The first e is , the o . Tulie is either an a-stem
genitive, with -e representing /-ę̄/ (thus Torelli, but cf. §3.7.6), or an Etruscoid nomina-
tive or genitive tulie(s) (see §9.2.2.2d). In the former case, the formula is ego OWNERGEN
as attested for the Latin inscriptions, in the latter, it could also be ego OWNERNOM
attested for the Faliscan inscriptions (§8.9.2). In view of the preservation of the
diphthong in the ending of sceiuai MF 379, I prefer to take tulie as an ‘Etruscoid’ form.
Torelli’s alternative, to take tulie as a nominative plural, is impossible in view of eco.
Bibliography: Torelli 1967:538 ¢1² (autopsy); Renzetti Marra 1974:351 ¢I²; Agostiniani 1982:150
¢589². Drawing: Torelli 1967:538.
384. Scratched in a brown-varnished saucer (height 6 cm, rim 12 cm, foot 5.1 cm).
Letters 8 mm high.
[fel]ịcinatiu
Sinistroverse, Faliscan alphabet. Of the first i, only the lower half is legible. The a is ,
the t . The integration is Torelli’s, after felic ̣inate in MF 42. The editors regard the
ending as Etruscan, but it is rather the regular i-stem genitive plural ending with an
early, but not impossible, closing of the vowel (§36.6.1). A parallel for a Besitzerin-
schrift consisting of the genitive plural may be provided by tulom MF 72, if this is
interpreted as ‘Tullum = Tullorum’.
Bibliography: Torelli 1967:538-9 ¢2² (autopsy); Renzetti Marra 1974:351 ¢II²; Briquel 1972:820, 826;
Rix ET Fa 2.17. Drawing: Torelli 1967:538.
560
17.6. Fiano Romano
385. Scratched on the shoulder of a small impasto amphora. Sixth or fifth century
according to Paribeni, but first half of the seventh century according to Briquel.
tulate tulas urate
Sinistroverse, Etruscan (?) alphabet with reversed s. The r is . Tulas is written as tul as
on both sides of the handle, but is certainly one word. The text, which appears to be
some kind of word-play, is regarded as Etruscan by most editors, but is not included in
ET. An interpretation based on the Italic languages has been proposed by Pisani, who
connected tulate and tulas with Latin tollo and urate with Oscan urust ‘dixerit’ TB 14,
16, taking the text as an injunction to drink and chatter (“‘sopportate - sopporta!
chiacchierate’”). Although this interpretation is not without its problems, it is certainly
possible: see §5.3.2.20-21,23.
Bibliography: Paribeni 1927; Buonamici 1929a; Buffa NRIE 986; Pisani 1943:262-3; Vetter 1953:291;
G. Giacomelli 1963:268 ¢XXXVI²; Pisani 1964:347; Pallottino TLE 25; Briquel 1972:815-8. Photo-
graph: Paribeni 1927:370 fig.1. Drawing: Paribeni 1927:370 fig.2 (reproduced in Briquel 1972:815
fig.6).
17.7. Civitucola (ancient Capena)
The site of ancient Capena is the Colle di Civitucola or Colle del Castellaccio, c.4 km to
the north of modern Capena (called Leprignano until the 1930s, and occurring under
that name in the older literature on the subject). Like many South Etrurian sites, ancient
Capena lies on a spur of land enclosed between river gorges, in this case those of the
Fosso di Vallelunga or Fosso dell’Olio to the north and that of a smaller stream to the
south, which meet on the east of the town (see Paribeni 1905, 1906a-b, Stefani 1953
with map p.3, and G. Jones 1962:134-5 with maps pp.130-1). The early finds (Cap
386-393) are from Contrada S. Martino, to the north of Civitucola; most of the other
inscriptions were found during the excavations in 1905 and 1906 at Contrada le Saliere,
to the west of Civitucola on the same ridge (Cap 395-420 and Cap 421), and at Monte
Cornazzano, to the south of Civitucola (Cap 422). The provenance of the other
inscriptions (Cap 394 and 423-430) is undocumented. An Etruscan inscription from the
area is the alphabetary Etr XLIV.
386-393. The following inscriptions were found at Contrada S. Martino (Henzen).

386-387. Repeated four times on a black-varnished cup (height 6 cm, 13.5 cm) is a
decorative stamp showing a dolphin and the inscription (letters c.1 mm high).
cel
561
CHAPTER 17
Dextroverse, Latin alphabet. Herbig considered reading ḷeụ. Scratched inside the cup is
cpscni
Dextroverse, Latin alphabet with reversed s. Pscni = p(e)sc(e)ni, a genitive or perhaps
an abbreviated nominative (§8.8.1) of Pescennius.
Bibliography: Henzen 1864:147 (autopsy); Fabretti CII 2453bis,a; Garrucci SIL 816 (autopsy);
Zvetaieff IIM p.66 ¢e²; Deecke 1888:203-4 ¢69²; Bormann CIL XI.6706,4 (autopsy); Lommatzsch CIL
I2.476,4; Herbig CIE 8449 (autopsy); Buonamici 1913:89 ¢56²; Vetter 1953:327 ¢356a²; G. Giacomelli
1963:266 ¢XXIV²; Briquel 1972:822-3; Degrassi & Krummrey CIL I2.476,4 add.; Wachter 1987:440
n.999. Drawing: Palmieri in Fabretti CII tab.XLIII; Garrucci SIL tab.II.3; [e prioribus Deecke 1888
Taf.III]; Herbig CIE 8449 (reproduced in Briquel 1972:822 fig.8).
388. Scratched in a black-varnished cup (h. 6 cm, 15 cm; let. 6-11 mm).
ḳuomanio
Dextroverse, Latin alphabet. A is ; the o’s are polygonal and open at the bottom (cf.
Cap 389, and LtF 377 from nearby Ponzano Romano). The first o looks very similar to
the a or o in MLF 375. The first letter, , was read as t . c ., and interpreted either as
two praenomina followed by a singular (or a very unlikely dual!) gentilicium uomanio
(Henzen, Garrucci, Deecke, Schulze), or as a praenomen t followed by a paternal
praenomen c, with an affiliation between the praenomen and the gentilicium, as in
Umbrian and Volscian (Garrucci, Deecke). Buonamici read p ̣[·], Lommatzsch and
Herbig k ., which has been adopted by all later editors. The gentilicium Vomanius has
been connected to the potamonym Vomanus (cf. §7.8.1.182, §6.5.4).
Bibliography: Henzen 1864:147-8 (autopsy); Fabretti CII 2453bis,c; Garrucci SIL 815 (autopsy);
Zvetaieff IIM p.65 ¢d²; Deecke 1888:200 ¢65²; Bormann CIL XI.6706,5; Conway 1897:384 ¢347²;
Schulze 1904:117; Herbig CIE 8450 (autopsy); Buonamici 1913:89-90 ¢57²; Meister 1916:99-100;
Lommatzsch CIL I2.476.5; Vetter 1953:327 ¢356b²; G. Giacomelli 1963:266 ¢XXV²; Briquel 1972:824;
Degrassi & Krummrey CIL I2.476,5 add. Drawings: Henzen 1864:147; Palmieri in Fabretti CII tab.
XLIII; Garrucci SIL tav. II.5; [e prioribus Deecke 1888 Taf. III]; Herbig CIE 8450 (reproduced in
Briquel 1972:824 fig.9).
389. Scratched in a black-varnished cup (h. 6 cm, 15 cm; let. 9-14 cm.)
aírpiosesú
Dextroverse, Latin alphabet. The o is polygonal and open at the bottom, as in Cap 388.
The second letter is  and has been read as ś by most editors; Briquel convincingly
read it as í, which results in a name írpios that recalls the Faliscan Hirpi Sorani (§2.3.4):
see §7.8.1.74, §9.3.2. The word-final -s is written out, which is very rare in the Faliscan,
but not in the Capenate inscriptions (§3.5.7.d). The last letter is , which was usually
read as χ: as in the case of k  sares  esú Cap 404, Briquel gave this letter the value ú
and read esú, which must be interpreted as esú(m) /esom/ = sum (Colonna 1983b:58,
independently Bakkum 1996): see §5.3.1.5. The resulting formula, OWNERNom sum, is
unparallelled in Besitzerinschriften but provides few difficulties: see §8.8.2.
Bibliography: Henzen 1864:148 (autopsy); Fabretti CII 2453bis,b; Garrucci SIL 814 (autopsy);
Zvetaieff IIM p.66 ¢f²; Deecke 1888:200-1 ¢66²; Bormann CIL XI.6706,6; Lommatzsch CIL 476,6;
562
Herbig 1910b:184-5; Herbig CIE 8451 (autopsy); Buonamici 1913:90 ¢58²; Vetter 1953:327 ¢356c²; G.
Giacomelli 1963:267 ¢XXVI²; Briquel 1972:833-7; Degrassi & Krummrey CIL I2.476,6 add.; Colonna
1983b:58-61; Wachter 1987:440 n.999; Bakkum 1996; Mancini 1997:27-30. Drawings: Henzen
1864:148; Palmieri in Fabretti CII tab.XLIII; [e prioribus Deecke 1888 Taf.III]; Herbig CIE 8451
(reproduced in Briquel 1972:832 fig.10).
390. Scratched, before firing according to Henzen, under the foot of a black-varnished
cup (height 6.5 cm, 15.5 cm; letter 3-7 mm high).
kpaaiedies
Dextroverse, Latin alphabet. The first a is , the second . The e’s are . Some
editors have read the text as referring to two brothers K. and Pa. Aiedius (Schulze,
Vetter, Degrassi, G. Giacomelli) or to three brothers K., P., and A. Aiedius (Henzen).
Aiedies would then be an instance of the second-declension nominative plural in -es,
which is especially frequent at the end of lists (including those consisting of multiple
praenomina, cf. Bakkum 1994), but for this it is too early. Others have assumed that the
text refers to only one person ‘A. Aiedies K. f.’ (Garrucci, Zvetaieff, Bormann),
‘K(aeso) Pa(quii filius) Aiedius’ (Deecke, Wachter), or ‘pa . aiedies . k .’ (Herbig). In
this interpretation, aiedies is a Sabellic nominative singular in -ies (cf. sares in Cap 404)
with the filiation between the praenomen and the gentilicium, as in Umbrian and
Volscian (Briquel). Both the praenomen pa = Pacius and the gentilicium Aiedius are
Sabellic rather than Latin.
Bibliography: Henzen 1864:149-50 (autopsy); Fabretti CII 2453bis,g; Garrucci SIL 811 (autopsy);
Zvetaieff IIM p.65 ¢a²; Deecke 1888:199-200 ¢64²; Bormann CIL XI.6706,1; Conway 1897:383 ¢346²;
Von Planta 1897:588 ¢323²; Schulze 1904:116-7; Herbig CIE 8453 (autopsy); Buonamici 1913:91-92
¢60²; Lommatzsch CIL I2.476,1; Vetter 1953:328 ¢358a²; Cencetti 1957:196-7; G. Giacomelli 1963:267
¢XXVIII²; Briquel 1972:830; Degrassi & Krummrey CIL I2.476,1 add.; Wachter 1987:440 n.999;
Bakkum 1994:20-2. Drawings: Henzen 1864:149 (reproduced in CII tab.XLIII); Garrucci SIL tab.II.4
(reproduced in CIL XI.6706,1, CIL I2.476,1, Cencetti 1957:196 fig.15); [e prioribus Deecke 1888
Taf.III]; Herbig CIE 8453 (reproduced in Briquel 1972:832 fig.10).
391. Scratched in a black-varnished cup.

atfertrio
Dextroverse, Latin alphabet. The a is . At may be Attus. Fertrio is probably to be read
as fert(o)rio = Fertorius, derived from the praenomen Fertor, which may occur in the
area in the abbreviated form f (e.g. in Cap 392, see further §7.7.1.23). There appear to
be no attestations of the gentilicium Feretrius suggested by Deecke.
Bibliography: Henzen 1864:147 (autopsy); Fabretti CII 2453bis,i; Garrucci SIL 812 (autopsy); Zve-
taieff IIM p.65 ¢b²; Bormann CIL XI.6706,2 (autopsy); Deecke 1888:201-2 ¢67²; Conway 1897:384
¢348²; Lommatzsch CIL I2.476,2; Herbig CIE 8454 (autopsy); Buonamici 1913:92 ¢61²; Vetter
1953:328 ¢358b²; Degrassi ILLRP 1232; G. Giacomelli 1963:267 ¢XXIX²; Briquel 1972:824-5;
Degrassi & Krummrey CIL I2.476,2 add. Drawing: Herbig CIE 8454 (reproduced in Briquel 1972:824
290
fig.9).
290
The drawing in Deecke 1888 (Taf.III) reproduces Henzen’s transcription.
563
CHAPTER 17
392. Scratched around the foot of a black-varnished cup (height 3 cm, 8.5 cm; letters:
7-12 mm high).
fpacios
Dextroverse, Latin alphabet with reversed s. The letters ac are accidentally written
together as , hardly the paḳios suggested by Briquel. Henzen’s f . paci is a misread-
ing. The f probably stands for Fertor, as was proposed by Garrucci (see §7.7.1), not for
the Faustus proposed by Henzen. It occurs also in f aino MLF 352 from the Grotta Por-
ciosa site and perhaps in fofiti or f ofiti MF 58 from Civita Castellana.
Bibliography: Henzen 1864:146 (autopsy); Henzen 1865:266-7; Fabretti CII 2453bis,h; Garrucci SIL
813 (autopsy); Zvetaieff IIM p.65 ¢c²; Bormann CIL XI.6706,3; Deecke 1888:202 ¢68²; Lommatzsch
CIL I2.476,3; Herbig CIE 8455 (autopsy); Buonamici 1913:92 ¢62²; Vetter 1953:328 ¢358c²; Degrassi
ILLRP 1233; G. Giacomelli 1963:267 ¢XXX²; Briquel 1972:825; Degrassi & Krummrey CIL I2.476,3
add. Drawings: Palmieri in Fabretti 1867 tab. XLIII (reproduced in Deecke 1888 Taf. III); Herbig CIE
8455 (reproduced in Briquel 1972:824 fig.9).
393. Scratched on a small urn (height 7.5 cm).

claudiacf
adiiieidussext
Dextroverse, Latin alphabet. The language is the Latin of the inscriptions of the second
century BCE.
Bibliography: Henzen 1864:145-6 (autopsy); Fabretti CII 2453bis,k; Garrucci SIL 1883; Bormann CIL
XI.3961a; Herbig CIE 8456 (autopsy); Lommatzsch CIL I2.1987. Transcriptions: Henzen 1864:145
(reproduced in CII 2453, SIL 1883, CIL XI. 3961a, CIL I2.1987); Herbig CIE 8456.
394. Briquel mentions an inscription from ‘S. Marino’ (given the context, probably an
error for ‘S. Martino’) that would shortly be published by Di Giuseppe:
stclanidio
There are no other attestations of a gentilicium Clanidius.
Bibliography: Briquel 1972:825; Degrassi & Krummrey CIL I2.2903c.
395-419. The following inscriptions were found during excavations at Contrada Le

Saliere in 1905 and 1906 (cf. Paribeni 1905, 1906b).
395. Scratched inside a black-varnished etrusco-campanian saucer from tomb VI.
caci
Dextroverse, Latin alphabet with . Aci is a genitive or abbreviated nominative of a
gentilicium Accius or Acilius. The latter occurs in CIL XI.7531 from Falerii Novi and
CIL XI.7768 from Fiano Romano.
Bibliography: Paribeni 1905:304; Paribeni 1906b:288, 482; Herbig CIE 8495; Bormann CIL
XI.8124,2; Lommatzsch CIL I2.2496,1; G. Giacomelli 1963:267 ¢XXXI,i². Transcriptions: Paribeni
1905:304; Paribeni 1906b:288, 482 (reproduced in Herbig CIE 8495).
564
396. Scratched, near the foot, on a black-varnished vessel from tomb VI.
caue
Dextroverse, Latin alphabet with ; the last letter is a cursive e (), not Paribeni’s ii.
Bibliography: Paribeni 1905:304; Paribeni 1906b:288; Herbig CIE 8496; Bormann CIL XI.8124,6;
Lommatzsch CIL I2.2496,5; Vetter 1953:328 ¢359a²; G. Giacomelli 1963:268 ¢XXXI,ii²; Briquel
1972:824-5. Drawing: Paribeni 1905:304 (reproduced in Paribeni 1906b:483, Herbig CIE 8496,
Briquel 1972:824).
397. Scratched in a black-varnished cup from tomb VII.

clanu
Dextroverse, Latin alphabet. The l appears to have been written within the c; the a is .
G. Giacomelli read clanu: cf. perhaps clanidio Cap 394?
XI.8124,4; Lommatzsch CIL I2.2496,3; G. Giacomelli 1963:268 ¢XXXII,i²; Briquel 1972:822-3.
Drawing: Paribeni 1905:306 (reproduced in Paribeni 1906b:482, CIE 8497, Briquel 1972:822 fig.8).
398. Scratched on a black-varnished vessel from tomb VII.

cau
Dextroverse, Latin alphabet. The a is .
XI.8124,7; Lommatzsch CIL I2.2496,6; G. Giacomelli 1963:268 ¢XXXII,ii². Drawing: Paribeni
1905:306 (reproduced in Paribeni 1906b:482, Herbig CIE 8498).
399. Scratched on the body of an olla without handles, from tomb VII.
sex
senti
Dextroverse, Latin alphabet. The e is cursive (). Apparently due to a confusion with
sex  senti Cap 429, Paribeni read sex. senti: G. Giacomelli and Briquel made the same
error. The form is a genitive or abbreviated nominative (§8.8.1). The gentilicium occurs
also in senθia MLF 362 from the northern ager Faliscus, and perhaps also in [---]nθia
MLF 212 from the surroundings of Civita Castellana.
XI.8124,1a; Lommatzsch CIL I2.2496,10a; G. Giacomelli 1963:268 ¢XXXIII²; Briquel 1972:822-3.
Drawing: Paribeni 1906b:483 (reproduced in Briquel 1972:822 fig.8).
400. Scratched on the neck of an urceus from tomb XXX.

csab
Dextroverse, Latin alphabet. The a is . The name is obviously Sab(in-), reflecting the
Latin form: the Faliscan form would have been *Saf(in-), see §6.5.17.
XI.8124,15; Lommatzsch CIL I2.2496,8; Briquel 1972:825. Drawing: Paribeni 1905:326 fig.7 (repro-
duced in Paribeni 1906b:483).
565
CHAPTER 17
401-402. Scratched on two Etrusco-Campanian vases from tomb XXXIII.

si or is
Dextroverse, but it is unclear whether the graffito is to be read as  or as  (thus
Herbig).
ha
Bibliography: Paribeni 1905:327-8; Paribeni 1906b:312; Herbig CIE 8511 and 8514; Bormann CIL
XI.8124,11. Transcription: Paribeni 1905:327-8 (reproduced in Paribeni 1906b:312, CIE 8511,
and8514).
403. Scratched on a late terracotta cup from tomb XXXV.

kape
Dextroverse, Latin alphabet. The a is . The third letter is : editors have usually read
kale (‘Gallus’? Herbig, G. Giacomelli), apparently with an upside-down l; Vetter read
kaue, apparently with an upside-down u.
Bibliography: Paribeni 1905:329; Paribeni 1906b:313, 481; Herbig CIE 8515; Buonamici 1913:93
¢65²; Bormann CIL XI sub 8124; Lommatzsch CIL I2.2496,9; Vetter 1953:329 ¢359c²; G. Giacomelli
1963:268 ¢XXXIV,i²; Briquel 1972:832-3. Drawing: Paribeni 1905:329 (reproduced in Paribeni
1906b:481, CIE 8515, Briquel 1972:832 fig.10).
404. Scratched on an Etrusco-Campanian cup from tomb XXXV.

ksaresesú
Dextroverse, but with Faliscan a and r . The k is . The last letter is , which
is was usually read as  χ: as in the case of a  írpios  esú Cap 389, Briquel gave the
letter the value ú. The resulting esú is esú(m) /esom/ = sum (Colonna 1983b:58,
independently Bakkum 1996): see §5.3.1.5. The formula, OWNERNOM sum, is unparal-
lelled (cf. also on Cap 389), but provides few difficulties: see §8.8.2. The ending of
sares is apparently a Sabellic nominative with Endsilbenkürzung, cf. aiedies in Cap
390.
Bibliography: Paribeni 1905:329; Paribeni 1906b:313, 481; Herbig CIE 8516; Lommatzsch CIL
I2.2496,9; Vetter 1953:329 ¢359d²; G. Giacomelli 1963:268 ¢XXXIV,ii²; Briquel 1972:832-7; Colonna
1983b:58-61; Wachter 1987:440 n.999; Bakkum 1996. Drawing: Paribeni 1905:329 (reproduced in
Paribeni 1906b:481, CIE 8516, Briquel 1972:832 fig.10).
405-409. Scratched on five black-varnished cups from tomb XXXVIII .

ca
ca
ca
ca
pan
566
Dextroverse. The a is . The last text, , has also been read as ran (Paribeni
1905:335, Lommatzsch): the first letter in fact looks more like r than like p: for a
praenomen R- see Salomies 1987:87 n.233.
Bibliography: Paribeni 1905:335; Paribeni 1906b:318; Herbig CIE 8518-8521, 8523; Bormann CIL
XI.8124,3,12-14; Lommatzsch CIL I2.2496,2. Drawing of 409: Paribeni 1905:335 (reproduced in Pari-
beni 1906b:318, CIE 8523). Transcriptions of 405-408:
410. Scratched on a saucer from tomb XLI.

cn
Dextroverse.
Bibliography: Paribeni 1905:338; Paribeni 1906b:320; Herbig CIE 8524; Bormann CIL XI.8124,9. tr.
Paribeni 1905:338 (reproduced in Paribeni 1906b:320, CIE 8524).
411-412. Scratched on a black-varnished saucer and a cup from tomb XLIV.

ua
mc
The first inscription is dextroverse, with  (read upside down by Paribeni and Herbig);
the second sinistroverse.
Bibliography: Paribeni 1905:340; Paribeni 1906b:322; Herbig CIE 8528 and 8530. Drawing: Paribeni
1905:340 (reproduced in Paribeni 1906b:322, CIE 8528 and 8530).
413-415. Scratched on a saucer from tomb XLV is an inscription that Paribeni gave as
�����
“‘L(uci) Ani’ ripetuto due volte” (1905:342). Bormann (in Paribeni 1905:484) sug-
gested reading ��� anṇiạnini (=anṇiạni{ni}?) which was adopted by Herbig
and G. Giacomelli.
Two other saucers from this tomb bear inscriptions:
hui
tps
Dextroverse. Hui may be h ui, with h for an original /f-/, as in Faliscan (§3.5.2): an
abbreviated praenomen f, probably Fertor (§7.7.1.23), occurs in Cap 392, MLF 352,
perhaps in MF 58. Ps is probably P(e)s(cenni-), cf. pscni Cap 387.
Bibliography: Paribeni 1905:342; Paribeni 1906b:324; Herbig CIE 8532-8533; Lommatzsch CIL
I2.2496,7; G. Giacomelli 1963:268 ¢XXXV². Transcriptions: Paribeni 1905:342; Paribeni 1906b: 324
(reproduced in CIE 8532-8533, CIL I2.2496,7).
416. Scratched on a saucer from tomb XLIX.

an or na
The text is given by Paribeni as � or , probably rather the latter.
Bibliography: Paribeni 1905:343; Paribeni 1906b:326; Herbig CIE 8535. Transcription: Paribeni
1906b: 326 (reproduced in CIE 8535).
567
CHAPTER 17
417-418. Scratched on two black-varnished saucers from tomb LVIII.

eno
cn
Dextroverse.
Bibliography: Paribeni 1905:352; Paribeni 1906b:335; Herbig CIE 8538-8539; Bormann CIL
XI.8124,8. Transcriptions: Paribeni 1905b:352 (reproduced in Paribeni 1906b:335, CIE 8538-8539).
419. Scratched on a black-varnished skyphos from tomb LXVIII (tomb LXIX Herbig)
cap
Dextroverse, Latin alphabet. The a is , the p .
Bibliography: Paribeni 1905b:361; Paribeni 1906b:342; Herbig CIE 8540; Bormann CIL XI.8124,5;
Lommatzsch CIL I2.2496,4. Transcription: Paribeni 1905:361 (reproduced in Paribeni 1906b:342, CIE
8540, CIL XI.8124,5, CIL I2.2496,4).
420. Scratched on a black-varnished cup from tomb XX (height 18.5 cm, 13 cm).
manio
Dextroverse, Latin alphabet with reversed n. The i appears to have a very small bar at
the top, which led Bandinelli to read anto = Anto(nius).
Bibliography: Bendinelli 1920:131; Coarelli & Mangani RMR p.311 ¢463²; Degrassi & Krummrey CIL
I2.2903a. Photograph: Coarelli & Mangani RMR tav. LXXVI. Drawings: Bendinelli 1920:131;
Degrassi & Krummrey CIL I2.2903a (from the photograph in RMR tav. LXXVI).
421. Incised on a bronze oinochoe (h. 18.5 cm) found between 1909 and 1912 in tomb
180 (215) at Contrada Le Saliere. The same tomb contained an amphora with a consular
dating of 160 BCE. Second half of the second century (Wachter).
marpopistfnmartddme
Dextroverse, Latin alphabet. The reading is Kretschmer’s: Della Seta (whence Stefani)
read ste ‘Ste(llatina tribu)’ instead of stf. He interpreted n.mart as ‘N(onis) Mart(iis)’,
but it is probably rather an abbreviation of the name of the same deity that appears in
LtF 377 from Ponzano as numesio  m[art]ẹ. Popi is Popius rather than Popilius.
Bibliography: Kretschmer 1917:139; Della Seta 1918:335 (autopsy); Lommatzsch CIL I2.2435;
Warmington 1940:70-1 ¢34²; Degrassi ILLRP 249; Stefani 1958:177; Briquel 1972:826; Degrassi &
Krummrey CIL I2.2435 add.; Wachter 1987:440 n.1000. Drawing: Degrassi & Krummrey CIL I2.2435
add.
422. Scratched on a vase from tomb LXXI at Monte Cornazzano’

ΓYIHLILI
Dextroverse. The reading is Paribeni’s. No interpretation has been proposed for this
unintelligible graffito. “Quid subsit, non video.” (Herbig CIE 8541).
Bibliography: Paribeni 1906b:346; Herbig CIE 8541. Transcription: Paribeni 1906b:346.
568
423-430. The following inscriptions are ascribed to Capena, without further specifica-
tion, and without a reason being given for the attribution. Several of these inscriptions
were described only by Garrucci and have unfortunately vanished since.
423. Scratched around the foot of black(-varnished?) cup.
Fig.17.1. Garrucci’s drawing of Cap 423.

(From CIE 8452.)
Garrucci hesitatingly read M. Spauri(os) ...i, Deecke ...? śnuśpaurn.... Later editors
adopted Bormann’s atnuśpauri. All these readings assume that  () is to be reas as
ś. Briquel, giving the sign its South Picene value í, hesitatingly proposed atnuípluri....
Bibliography: Garrucci SIL 818 (autopsy); Deecke 1888:206-7 ¢74²; Bormann CIL XI.6706,7; Lom-
matzsch CIL I2.476,7; Herbig CIE 8452; Buonamici 1913:90-1 ¢59²; Vetter 1953:327-8 ¢357²; G.
Giacomelli 1963:267 ¢XXVII²; Briquel 1972:833 n.2. Drawing: Garrucci SIL tab.II.6 (reproduced in
CIL XI.6706,7, CIL I2.476,6, CIE 8452).
424-425. Scratched on two black-varnished cups.

cuel
tno
Dextroverse, Latin alphabet with cursive e (). Uel is perhaps Vel(min-), cf. Velminaeus
in MLF 305, 307-310, 312-313, 315-317 from Vignanello.
Bibliography: Garrucci SIL 822, 826 (autopsy); Deecke 1888:208, 208-9 ¢76,77d²; Bormann CIL
XI.6706,10.3, 6706,10.7; Lommatzsch CIL I2.476,10.3, 476,10.7; Herbig CIE 8466-8467 (autopsy).
Drawing: Herbig CIE 8466-8467.
426-429. Scratched on four small cups.

an
cca
lca
cuo
Bibliography: Garrucci SIL 820, 823, 825 (autopsy); Deecke 1888:208-9 ¢77a-c²; Bormann CIL
XI.6706,10.1, 10.6, 10.4; Lommatzsch CIL I2.476,10.1, 10.4, 10.6; Herbig CIE 8471, 8478-8480.
569
CHAPTER 17
Drawing of 425: Herbig CIE 8471. Transcriptions of 426-428: Garrucci SIL 820, 823, 825 (reproduced
in CIL XI.6706,10.1, 10.6, 10.4, CIL I2.476,10.1, 10.4, 10.6, CIE 8478-8480).
430. Scratched on the neck of a pithos, in 1909 in the Sammlung Reimer, Hamburg.
sexsenti
Dextroverse, Latin alphabet with cursive e. Genitive, or perhaps an abbreviated
nominative: see §8.8.1. For other instances of the gentilicium Sentius, see the identical
text Cap 399.
Bibliography: Ballheimer 1909:29-30 ¢45² (autopsy); Lommatzsch CIL I2.2496,10b. Drawing:
Ballheimer 1909:29.
17.8. The shrine of Lucus Feroniae
The sanctuary of Feronia was discovered in 1952 in the southern ager Capenas, near
Scorano (see Taylor 1920, Foti 1953, Andreae 1957, and G.Jones 1962:191-201). Most
of the dedications (Cap 431, Lat 432, Cap 433, Lat 434, Cap 435, Lat 436, Cap 437-
438) are written in a Latin that shows very few dialectal features, except for the
consistent spelling of the name of the deity as Feronea instead of Feronia in Lat 432,
Lat 434, Cap 435, Lat 436, and perhaps Cap 438. This may well have been a manner-
ism going back to an older local spelling with e: cf. §3.6.2. One dedication, cavies 
uhtav[ies ?---] Etr XLV, is written in Etruscan but shows a Sabellic (?) spelling ht in
the name. With the exception of the building inscription Lat 456, all inscriptions appear
to predate the destruction of the shrine by Hannibal in 211 BCE (cf. Liv. 26.11).
431. Cut in two contiguous sides of a small sandstone basis (6.5u3 cm; letters 5-10 mm
high). Second half of the third century.
R. Bloch & Foti Torelli
ņņņņņņņņņņņņņņņņņņņņņņņņņņ ņņņņņņņņņņņņņņņņņņņņņņņņņņ
ṭescoṣal | uoda[rria] p ̣escoṣal | [q]uoda[fluc]
plariatl | dedetlibes plariatl | dedetlibes
ferondono | mmereto ferondono | mmereto
Dextroverse, Latin alphabet. The a is . The first r is , the second . Of the first and
the sixth letters, only the lower halves are preserved. The text has been read in two
entirely different ways. R. Bloch & Foti assumed that the lines continue round the
corner of the block, reading ṭeṣco saluod Arria Plaria T(iti) l(iberta) dedet libe(n)s
Feron(eae) donom mereto. This, however, presents two major problems: (a) since Arria
is nowhere attested as a praenomen, the name Arria Plaria would have to be a double
gentilicium (cf. §7.6), but apparently without any praenomen; (b) the formula donum
dedit libens merito is split in two. Both these problems are solved by Torelli’s
570
(1974:741-6) very different reading. Reading each side separately, he gave the text as
P̣esco. Ṣal(uia) | Plaria T. l. | Fero(niae) don(o) || [q]uod a[f luc(o)] | dedet libes |
m(erente) mereto. This keeps the formula donum dedit libens merito more or less intact,
and instead of Arria has a more common female praenomen Saluia, which according to
Kajava (1995:69 n.88) is especially common as a name of slaves (and therefore of
freedwomen, as in this inscription). Although Torelli’s reading is clearly preferable and
is adopted by Wachter, it is not without problems, especially the occurrence of an
apparently Sabellic word p ̣esco(m), for which cf. Marsian pesco VM 5 (and Umbrian
persklum TI Ia.1 etc.), in an otherwise Latin inscription: see §6.2.54, §6.3, §9.3.2.
Bibliography: R. Bloch 1952:622-5 (autopsy); R. Bloch & Foti 1953:63-71 (autopsy); Foti 1953:16
(autopsy); AÉ 1953 p.60 ¢195²; Andreae 1957:273 ¢1²: Degrassi ILLRP 93a; Jones 1962:193 n.91;
Briquel 1972:824-5; Torelli RMR pp.333-4 ¢482² (autopsy); Torelli 1974:741-6; Moretti 1975:145
¢141²; Poccetti NDI 225; Degrassi & Krummrey CIL I2.2867; Wachter 1987:440-1. Photo-
graphs: R. Bloch 1952 fig.2a-c; R. Bloch & Foti 1953 before p.65, figs.1-3; Degrassi ILLRP tab. 42a-b;
Torelli 1974 tav.CXXX. Drawings: R. Bloch & Foti 1953:66 (reproduced in Briquel 1972:824 fig.9);
Torelli 1974 tav. CXXX (reproduced in CIL I2.2867).
432. Cut in the front of the pediment of an aediculum (9u38 cm). Third century.
lcalpurnius
[fe]roneadonomerite
Dextroverse, Latin alphabet. The p is  , the n , and the l . The gentilicium
Calpurnius occurs also in CIL XI.3943 from Capena. For the antevocalic e in [fe]ronea,
see above ad init. Poccetti’s NDI gives the lines in reversed order.
Bibliography: R. Bloch 1952a:625 (autopsy); R. Bloch & Foti 1953:71-2 (autopsy); AÉ 1953 p.60
¢196²; Foti 1953:16; Andreae 1957:274 ¢2²; Degrassi ILLRP 93b; Briquel 1972:822-3; Coarelli RMR
p.334 ¢483² (autopsy); Moretti 1975:173 ¢39²; Poccetti NDI 226; Degrassi & Krummrey CIL I2.2868;
Wachter 1987:441. Photographs: R. Bloch 1952a:625 fig.2d; R. Bloch & Foti 1953 before p.65 fig.4;
Degrassi 1965:31 tab. 41; Coarelli RMR tav. LXXXV; Drawing: Briquel 1972:823 fig.8.
433. Cut in a stone base (8.5u7.5u13.2 cm; letters c.22 mm high) damaged at the left.
Late third century.
[---]no
[---]afluc
[---]mere
Dextroverse in Latin alphabet; the last e is cursive (). Degrassi & Krummrey inte-
grated [---]no(s) | [Feronea] af luc(o) | [dedet] mere(to); Torelli proposed [Fero(niae)
do]no | [dedet] af luc(o) | [libes] mere(to), apparently without the name of a dedicant. It
is unclear what af luc(o) would have meant.
Bibliography: R. Bloch 1952a:625 (autopsy); R. Bloch & Foti 1953:72-3 (autopsy); AÉ 1953 p.60
¢197²; Andreae 1957:274 ¢3²; Torelli 1974:745 n.10; Moretti 1975:175 ¢43²; Degrassi & Krummrey
CIL I2.2869. Photographs: R. Bloch 1952a:625 fig.2e; R. Bloch & Foti 1953 before p.65 fig.5 (repro-
duced in CIL I2 tab.11 fig.1); Moretti 1975 tav.39.
571
CHAPTER 17
434. Cut on a stone base (4.9u7.6u3.8 cm; letters c.0.8 cm high) damaged at the top.
Third century.
[------------------]
feronea dono[?]
lubensmerto[?]
dedit
Dextroverse, Latin alphabet. The l is . The first line, which stood on the missing part
of the base, contained the name of the dedicant. Merto is a graphical contraction
mer(e)to. For the antevocalic e in feronea, see above. For the dative in -a, see §3.7.6.
Bibliography: Moretti 1975:175 ¢44²; Degrassi & Krummrey CIL I2.2869b. Photographs: Moretti
1975 tav. 38; Degrassi & Krummrey CIL I2 tab.143 fig.5.
435. Cut in a round stone base (height 10.5 cm, 10 cm; letters 2.5 cm high) found in
1970. Late third century.
mtugenuciliosenl
feroneaidedit
Dextroverse, Latin alphabet. The l is . The first two initials are larger (3.5 cm) and
were added later (hence the muddled syntax, with the singular genucilio and dedit).
Moretti suggested that sen is perhaps a cognomen rather than a praenomen. Note
genucilio with antevocalic i beside feroneai with antevocalic e: see above ad init.
Bibliography: Moretti 1975:173-4 ¢40²; Degrassi & Krummrey CIL I2.2869a. Photographs: Moretti
1975 tav.38; Degrassi & Krummrey CIL I2 tab. 143 fig.4.
436. Cut in a round stone base (h. 9 cm, 13.5 cm; let. 1.3 cm). Late third century?
[---]rciusll
[fer]oneae
[l] m
Dextroverse, Latin alphabet. The l is . The first l is reversed: Moretti interpreted it as
‘l(ibertae)’, Degrassi & Krummrey as ‘L(uciae)’, which I prefer. As the lines appear to
have been were centred, the last line was probably [l] m rather than [d d l] m. Note
[---]rcius with i beside [fer]oneae with e: see above ad init.
Bibliography: Moretti 1975:152 ¢156²; Degrassi & Krummrey CIL I2.2869c. Photograph: Moretti
1975 tav.36; Degrassi & Krummrey CIL I2 tab.143 fig.6.
437. Scratched on a shard of a vase (38u9u20 mm R. Bloch & Foti; 9u8 cm AÉ).
fe
Probably the abbreviation of the name Feronia.
Bibliography: R. Bloch 1952:625 (autopsy); R. Bloch & Foti 1953:73; Foti 1953:16; AÉ 1953 p.60
¢198²; Andreae 1957:274; Degrassi & Krummrey CIL I2.2910.
572
438. Scratched on an impasto fragment (max. 5u11 cm). Third century.

[---]*ea**[---]
Since in the other dedications antevocalic e for i appears only in the name of Feronia
(see above ad init.), [---]*ea* is probably [fero]ṇeaẹ or [fero]ṇeaị.
Bibliography: Moretti 1975:152 ¢154²; Degrassi & Krummrey CIL I2.2910a.
439-455. Scratched on a number of fragments of various vessels found in a building

destroyed in the late third century.
mr
mr
mr
mr
mr
mr
mr
mr
mr
mr
cm
lma or lna
lma or lna
mp ̣
de
Dextroverse, Latin alphabet. In 452 and 453, the last letters are , which can be read
as ma or as na. In 454, only the shaft of the last letter is left. In 455, the e is cursive ().
Bibliography: Moretti 1975:156-60 ¢7-22²; Degrassi & Krummrey CIL I2.2910b.1-17.
456. Cut on a stone base (19u112u27.5 cm; letters first line 5-6 cm, second line 4-4.5
cm, third line 3.5-4 cm high) and coloured in with red. Early first century (Fraschetti).
cdidiustfmuettiusmfduomuiri
quinqstatuasfornicesquedd
faciundumcoeridemqueprobauerun[?t]
Dextroverse, Latin alphabet.
Bibliography: Moretti 1975:104-5 ¢141²; Fraschetti 1977:317 n.1; Degrassi & Krummrey CIL I2.3338b.
Photograph: Moretti 1975 tav.28; Degrassi & Krummrey CIL I2 tab.144 fig.3.
573
CHAPTER 17
17.9. Ager Capenas, provenance unknown
457-462. These inscriptions were published by Garrucci as Capenate, but without data
with regard to the circumstances of their discovery. Most have not been seen since.
457. Scratched in a small vase is
apa
Dextroverse, Latin alphabet with cursive .
Bibliography: Garrucci SIL 819 (autopsy); Gamurrini 1880 ¢829²; Deecke 1888:205-6 ¢72²; Bormann
CIL XI.6706,9; Lommatzsch CIL I2.476,9; Herbig CIE 8458; Vetter 1953:330 ¢360e²; G. Giacomelli
1963:268-9 ¢XXXVII,ii²; Rix ET Fa 2.24. Transcription: Garrucci SIL 819 (reproduced in Deecke
1888 Taf. III, CIL XI.6706,9, CIL I2.476,9, CIE 8458).
458. Scratched on a small vase.

cae**sa
Dextroverse, Latin alphabet (?). The text is given as : the first letter is
probably a reversed c, perhaps indicating a woman’s name (cf. §11.2.5.3); Deecke read
it as a t. The second word is unclear (e[z]p ̣sa Deecke, euḷsa Herbig).
Bibliography: Garrucci SIL 821 (autopsy); Deecke 1888:208 ¢75²; Bormann CIL XI.6706,10.2;
Lommatzsch CIL I2.476,10.2; Herbig CIE 8460; G. Giacomelli 1963:269 ¢XXXVIII,i². Transcription:
Garrucci SIL 821 (reproduced in CIE 8460).
459. Scratched on a small vase.

au cau
panur
Dextroverse, Latin alphabet. Au and cau are praenomina; Deecke suggested that panur
is a slave-name panur(co) = Panoàrgoj, cf. N. Munitor Panurcus in CIL XI.3166 from
Falerii Novi.
Bibliography: Garrucci SIL 1882 (autopsy); Deecke 1888:217 ¢103²; Bormann CIL XI.6706,8;
Lommatsch CIL I2.476,8; Herbig CIE 8461; G. Giacomelli 1963:268 ¢XXXVIII,ii². Transcription:
Garrucci SIL 1882 (reproduced in CIL XI.6706,8, CIL I2.476,8, CIE 8461).
460. Scratched in a small vase.

tif
Sinistroverse, Faliscan alphabet. Tif is often connected with Etruscan Tiφile, Latin
Tifilius ‘D…filoj’ (Deecke and Briquel), but may represent *Tif(eri-) or *Tif(erili-), the
Faliscan equivalent of Latin Tiberius or Tiberilius (cf. tiperilia LF 229).
Bibliography: Garrucci SIL 824 (autopsy); Zvetaieff IIM p.66 ¢g²; Deecke 1888:205 ¢71²; Bormann
CIL XI.6706,10.5; Conway 1897:384; Lommatzsch CIL I2.476,10.5; Herbig CIE 8457; Vetter 1953:330
¢360f²; G. Giacomelli 1963:268 ¢XXXVII,i²; Briquel 1972:820,823. Transcription: Garrucci SIL 824
(reproduced in IIM p.66, Deecke 1888 Taf.III, CIL XI.6706,10.5, CIL I2.476,10.5, CIE 8457, Briquel
1972:820 fig.7).
574
461. Scratched in a small vase.

trpe
Dextroverse, Latin alphabet. The e is cursive (). Tr is probably Trebius, pe perhaps
Pe(scenni-), cf. pscni Cap 387.
Bibliography: Garrucci SIL 827 (autopsy); Gamurrini 1880 ¢829²; Deecke 1888:206 ¢73²; Bormann
CIL XI.6706,10.8; Lommatzsch CIL I2.476,10.8; Herbig CIE 8459 (autopsy); G. Giacomelli
1963:268-9 ¢XXXVII,iii²; Briquel 1972:825. Transcriptions: Garrucci SIL 827 (reproduced in Deecke
1888 Taf. III, CIL XI p.1165, CIL I2 p.422); Herbig CIE p.105.
462-464. Published by Herbig in the CIE as “tituli inediti, qui quo anno et quibus
territorii locis inventi sint nescio” (CIE p.105).
462. Scratched before firing on the neck of an urceolus (height 21 cm; letters 3-6 mm
high).
piunio
Dextroverse, Latin alphabet (?) with . The o is (iuniạ? Herbig).
Bibliography: Herbig CIE 8463 (autopsy); Vetter 1953:329 ¢360b²; G. Giacomelli 1963:269 ¢XL²;
Briquel 1972: 821-2. Drawing: Herbig CIE 8463 (reproduced in Briquel 1972:822 fig.8).
463. Scratched under a saucer (height 2 cm, 12.2 cm; letters 6-7 mm high).
ueiuatia
Sinistroverse, Faliscan alphabet. The a is . Most editors adopt Herbig’s interpretation
‘Veius Vatia’. Buffa hesitatingly identified NRIE 991 (not in ET) which he read as tei
uṛ tiṇ, with this inscription.
Bibliography: Herbig CIE 8464 (autopsy); Buonamici 1913:93 ¢64²; Buffa NRIE 991; Vetter 1953:330
¢360c²; G. Giacomelli 1963:269 ¢XLI,i²; Briquel 1972:820, 822. Drawing: Herbig CIE 8464 (repro-
duced in Briquel 1972:820 fig.7).
464 (falsum?). Scratched under the foot of a red-varnished saucer (h. 3 cm, 7.5 cm;
letters 7 mm high). According to Herbig, the saucer wore a label saying ‘L’iscrizione è
evidentemente falsificata dal Mancinelli che l’ha malamente ricopiata dall’altra vera
no.9290’ (= Cap 463) when he saw it in 1903.
ueiụeto
Sinistroverse, Faliscan alphabet. The t is . Herbig regarded the text as a transcription
of an Etruscan *uei uetu ‘Veius Vettonius’.
Bibliography: Herbig CIE 8465 (autopsy); Vetter 1953:330 ¢360d²; G. Giacomelli 1963:269 ¢XLI,ii²;
Briquel 1972:820. Drawing: Herbig CIE 8465 (reproduced in Briquel 1972:820 fig.7).
465-466. Two inscriptions that were first published among a number of other finds
from the necropoles of ancient Capena. Further documentation on the date and location
of their discovery appears to be lacking. As Colonna notes, the writing is similar to that
of the Latin pocula deorum, which would point to a date of c.300 or slightly later.
575
CHAPTER 17
465. Scratched on a black-varnished saucer (height 5 cm, 9 cm) similar to the one of
466.
ac ̣iuaiomesú
Dextroverse, Latin alphabet. The a is cursive ; the c could also be read an l. The ú is
. The name is a gentilicium derived either from Accius (cf. aci Cap 395) or from Allius
or Alius, by means of a suffix /-āo-/ that also appears in early Latin karkavaios CIL
I2.2917a. The form is genitive plural, as in [fel]ịcinatiu LF 384 and probably in tulom
MF 72. For esú = esú(m) ‘I am’, which occurs also in Cap 389 and 404, see §5.3.1.5.
Bibliography: Pollak 1906:55 (autopsy); Colonna 1990a ¢a²; M.Mancini 1997:28. Drawing: Pollack
1906:55 (reproduced in Colonna 1990a:463).
466. Scratched on a black-varnished saucer (height 5 cm, 9 cm) similar to the one of
465.
stasediu
Dextroverse, Latin alphabet. The a is ; the e and the u are written upside-down: the u
could perhaps also be a cursive o (with an open bottom). Probably Sta. Sediu: if this is a
woman’s name, the -u may well represent the Sabellic first-declension nominative
singular /-ō/: see §9.3.2.
Bibliography: Pollak 1906:55 (autopsy); Colonna 1990a ¢b². Drawing: Pollack 1906:55 (reproduced in
Colonna 1990:463).
576
Chapter 18
Inscriptions claimed as Faliscan or Capenate
18.1. The inscriptions claimed as Faliscan or Capenate
The inscriptions presented in this chapter constitute a mixed bag of texts that have for
varying reasons been regarded as Faliscan or Capenate. I have divided them into three
groups according to whether or not I consider this attribution valid; my criteria for this
are discussed under each individual inscription.
The first group (§18.2, 467*-478*), are inscriptions that are originis incertae uel
ignotae but possibly or probably Faliscan or Capenate: of these, 467* is very probably
Early Faliscan, while 468* is an almost entirely Sabellic inscription, perhaps of
Capenate origin; 469*-473* are in all probability Middle Faliscan, and 474* probably
Middle or Late Faliscan; 475*-476* are either Middle or Late Faliscan or Capenate;
477*-478* are Latin inscriptions found, but probably not made, in the ager Faliscus.
The second group (§18.3, 479†-484†) consists of the inscriptions that are
probably not Faliscan or Capenate. The first group (§18.3.1) are originis incertae uel
ignotae: 479† is probably archaic Latin, 480† is Sabellic (‘Palaeoumbrian’), and 481†
is unintelligible. The second group (§18.3.2), 482†-484†, are all from Ardea, and their
association with Faliscan has given rise to a persistent idea that Ardeatine and Faliscan
were in some way related. I do not believe that there are grounds to warrant this: the
issue is discussed at the end of section §18.3.2. In the final section (§18.3.3) I discuss
Lucchesi’s (2005) suggestion that the Satricum-inscription CIL I2.2832a is Faliscan.
18.2. Inscriptions that are probably or possibly Faliscan or Capenate
467* (Early Faliscan?). Scratched on the bottom of a bucchero cup (height 5.2 cm,
rim 12.9 cm, foot 7 cm; letters 6-14 mm high) acquired in Rome in 1889 by Froehner.
Sixth century.
aịṃiosioeqo
Sinistroverse. The second, third, and fourth letters are . Lejeune (1952:124) read
a[im]iosio, considering also aṇṇosio and aṇịṇiosio (with reversed n). Vetter proposed
aṇạiosio, which is adopted by Agostiniani. Lejeune and G. Giacomelli regarded the
inscription as Faliscan, probably because of the genitive in -osio, but this is now attested
also from Latium (popliosio ualesiosio CIL I2.2832a from Satricum): see §4.4. The
577
CHAPTER 18
sinistroverse ductus is found in the ager Faliscus at this date (in EF 6-10), but not
universally (sixth-century EF 1 is dextroverse). The formula ego OWNERGEN (here
uniquely in the reversed order OWNERGEN ego) is as yet attested only from the ager
Faliscus, Latin and Capenate inscriptions having ego OWNERNOM (§8.9.2). The name
Aemius is known only from the ager Falicus (eiṃoi MLF 293, possibly ạ[i]ṃ MF 89)
and from Venetic (·‹a›·imo·i· Le 26). Although the criteria are inconclusive, I give this
inscription the benefit of the doubt and treat it as Early Faliscan.
Bibliography: Lejeune 1952b:120-6 (autopsy); Pisani 1953:320; Vetter in Knobloch 1954:40; Pisani
1955:320; G. Giacomelli 1963:66 ¢56²; Pisani 1964:342; Agostiniani 1982:155 ¢603²; Lejeune 1989:67.
Photographs: Lejeune 1952b pl.XV-XVI. Drawings: Lejeune 1952b:121; Lejeune 1989:67.
468* (Sabellic, but of Capenate origin?). Scratched inside a black-varnished cup,

according to Buonamici from South Etruria.
pa‹qu›isblaisiís
Dextroverse. The third and fourth letters, and , have usually been read as l and p
respectively, but they differ from the other l ( ) and p ( ), and Marinetti rightly reads
them as an upside-down u (cf. the similar u in 477†) and a q (cf. the similar q in 482†).
Since there are no names in Pauq-, Marinetti proposes to read paụq- as an error for
paqu- (cf. Latin Paquius, Oscan pak#hij Lu 40); this attractive solution also accounts
for the unexpected use of q before i. The last letters of this word are and have been
read as is (G. Giacomelli, Arena, Briquel) or s (Colonna), or as r (p ̣aḷpr Buonamici) or e
(palpe Vetter), which is impossible. The penultimate letter of the second word, , is
usually read as an o, but its square form is odd: Buonamici, comparing the h in EF 1,
read blaisihs. Marinetti gave the sign the value í it has in the South Picene alphabet, and
this interpretation is now supported by Rix’s (1992:249-50) reading of in 477† as í.
Buffa, who read pa(canml) aisiθs, called the text Faliscan without giving any
reasons for this attribution. Vetter, too, regarded the inscription as Faliscan, because of
shape of the e in his untenable reading palpe, , which he compared to the in MF
146 and in Lat 482† from Ardea (but regarded as Faliscan, see §18.3.2). According to
Arena, the inscription looked Faliscan or Greek “was die sprachliche Eigentümlich-
keiten betrifft” (1967:115 n.1), unfortunately without elaborating this. Such arguments
as there are seem to speak against a Faliscan connection. The shape of the l’s and the
use of b are un-Faliscan, while the ‘square í’ is South Picene, although possibly
originating in South Etruria or the ager Capenas (Briquel 1972:830-6, Rix
1992:249-51). The Endsilbensynkope is likewise Sabellic, not Faliscan, and so are the
names.
Bibliography: Buonamici 1928:605-6 (autopsy); Buffa NRIE 981; Vetter 1953:359 ¢513²; G. Gia-
comelli 1963:262 ¢VI²; Arena 1967:114-5; Briquel 1972:831-3; Colonna 1980b:67-8 ¢38² (autopsy);
Marinetti 1982b; Degrassi & Krummrey CIL I2.2917c. Photograph: Colonna 1980b pl. 18.4 (repro-
duced in Marinetti 1982 tav.LVIIa). Drawings: Buonamici 1928:606 (reproduced in Briquel 1972:832
fig.10); Buffa NRIE tav.XII; Colonna 1980b:67 fig. 11 (reproduced in Marinetti 1982:365); Degrassi &
Krummrey CIL I2.2917c.
578
INSCRIPTIONS OF UNCERTAIN, UNKNOWN, OR NON-FALISCAN OR NON-CAPENATE ORIGIN
469* (Middle Faliscan). Scratched inside a black-varnished cup (height 5.7 cm, rim
11.3 cm, foot 5.6 cm; letters 12-20 mm) acquired in Naples in 1900 by Froehner.
uolticatinei
Sinistroverse, Faliscan alphabet. Froehner (quoted in Lejeune 1952b:115-6) regarded
the inscription as Etruscan, but in that case the use of o and of u instead of v would be
surprising (cf. vultasi Etr XLII). Lejeune and G. Giacomelli both regard the inscription
as Faliscan. This is probably correct, the arguments in favour being both epigraphical
(the form of the t, , and the sinistroverse ductus, which at this date is regular in the
ager Faliscus), and onomastical (the praenomen Voltius, which occurs with some
frequency in Faliscan inscriptions, but is very rare elsewhere). The form is probably a
genitive, although the possibility of an abbreviated nominative cannot be excluded
(§8.8.1).
Bibliography: Lejeune 1952b:115-20 (autopsy); G. Giacomelli 1963:66 ¢55². Photographs: Lejeune
1952b pl. XIII-XIV. Drawing: Lejeune 1952b:116.
470* (Middle Faliscan). Painted carefully on the underside of the foot of a small vase
in the shape of a rooster (height 20.5 cm, letters 0.5 cm high) in the Steinhardt Collec-
tion, New York. No information is provided about the provenance of the object, but in
view of the fabric and the ornamentation, there can be no doubt that it is of Faliscan
origin, as is the inscription. Mid- or late fourth century.
oufiloclipeaioleteifileometfacet
Sinistroverse. The alphabet is the usual Faliscan alphabet of the period, except for the l
in oufilo, which according to Wallace “appears to be a character in the shape of an
upsilon; traces of a left oblique bar are visible.” (2005:176). Clipeaio is probably an
error for clipea‹r›io, cf. clipịaṛ[io LF 230, clipeaṛ[io LtF 231, and cḷ[i]peario LtF 233
from S. Maria di Falleri. The praenomen oufilo is attested from several sites in the ager
Faliscus (see §7.7.1.12). The name of the father, in the genitive letei, is unattested
elsewhere. Wallace hesitatingly compares Greek L»taoj and Etruscan leθae
(2005:178), but I think letei may well be the Faliscan genitive of *Letaeus, a Latiniza-
tion of the Etruscan name leθaie in Etr XLVIII from Mazzano Romano.
The inscription provides several points of linguistic interest. The perfect facet,
parallelled in faced MF 471*, shows a perfect stem /fak-/: see §5.3.1.6. The third person
singular perfect ending with -et instead of -ed (cf. faced MF 471*) was already known
from keset in LF 242. Met is the only attestation of the accusative of the pronoun of the
first person singular in Middle Faliscan (Lat 268 is an import), although med was
known already from Early Faliscan EF 1 and 9. I am inclined to view the t in this form
as an error due to the fact that -d was already disappearing in the nouns and pronouns,
and was replaced in the verbal endings by -t: see §3.5.7c. For the formula, see §8.9.2.
Bibliography: Wallace 2005 (autopsy); Poccetti 2005; De Simone 2006. Photographs: Wallace
2005:175-6 figs.1-2.
579
CHAPTER 18
471* (Middle Faliscan). Painted on the body of a small column crater (height 31.5 cm,
rim 21.5 cm, foot 12.5 cm; letters 1,5-3 cm high), acquired in 1997 in Basle by the
Museo Arqueológico Nacional, Madrid. Late fourth to mid-third century.
cauiosfrenaiosfaced
Sinistroverse, Faliscan alphabet. The praenomen is undoubtedly Faliscan (§7.7.1.25);
the gentilicium is not attested elsewhere. The only parallel for word-final -s after a short
vowel being written is cauios Cap 382. Berenguer & Luján draw attention to the
conservation of ai, for which cf. uoltaia MF 196 and latinaio MLF 210: see §3.7.6 and
G. Giacomelli 1962. The Middle Faliscan perfect faced has a parallel in facet MF 470*
(quo vide). Linguistically, it is interesting to find the ending -ed more or less contempo-
rary with facet in MF 470*. For the formula, see §8.9.2.
Bibliography: †Olmos Romera 2003; Olmos Romera 2004; Berenguer & Luján 2004; Berenguer &
Luján 2005; Wallace 2005; Poccetti 2005; De Simone 2006. Photographs: Berenguer & Luján
2004 :213, 222.
472* (Middle Faliscan). Scratched on the outer side of a pottery fragment (letters 10-
13 mm high) of Faliscan make (Stanco, with references). Late fourth or early third
century.
marcianel[i ?---]
Sinistroverse. The ductus, the shapes of the letters, and the names are in accordance
with the Faliscan provenance of the piece. Marci is probably a genitive; alternatively, it
could be an abbreviated nominative of Marcius. In both cases, the gentilicium is to be
restored as anel[i. It seems unlikely that any more text followed this.
Bibliography: Stanco 2001 (autopsy). Photograph: Stanco 2001 tav.LVc. Drawing: Stanco 2001:483.
473* (Middle Faliscan). Painted in black under the foot of a Genucilia-plate ( 14,5
cm; letters c.12 mm high), auctioned in 2002 in Paris. Late fourth or early third century.
cauiopeṭṛọṇẹọ
Sinistroverse, Faliscan alphabet. The a is  for a, another instance of the confusion of 
and  (§11.2.4).291 The gentilicium is so damaged as to have almost disappeared:
Briquel restores pe[troneo], probably rightly.
Bibliography: Tajan 2002:59 ¢436²; Briquel 2002 (autopsy). Photographs: Tajan p.59; Briquel 2002
tav. XXXVIa-b. Drawings: Briquel 2002:402.
474* (Middle or Late Faliscan or Capenate?). Engraved along two sides of the upper
surface of a triangular bronze statuette base (length of the sides 9.7 cm, letters 13 mm
high). Ritschl ascribed the inscription to Praeneste, but Mommsen & Henzen (CIL I
291
“Inscription en alphabet étrusque peinte sous le pied: «CRVIO : G…» … Inscription effacée
en fin de mot sinon complet.” (Tajan 2002:59).
580
p.255) claimed that “elementorum quorundam forma et interpunctio duplex a Latina

consuetudine abhorrent”, and Garrucci ascribed it to Falerii, where it was reported to
have been found by Sebastiniani. The object, lost after Garrucci’s death, was rediscov-
ered by Lejeune in the Froehner Collection in Paris.
cauitertinei|posticnu
Dextroverse, Latin alphabet. In Ritschl’s and Garrucci’s drawing, the top of the o is
interrupted for the foot of the statuette. Caui  tertinei is genitive of Gavius Tertineius
vel sim. (thus Lejeune, reviving the interpretation of Bronisch and Tambroni), not an
Etruscan nominative (thus Herbig, Vetter, and Pisani292). Posticnu has been interpreted
variously as an Etruscan nominative or genitive (Lejeune): it is in fact a Sabellic first-
declension nominative with a lexical parallel in South Picene ombriíen  aḳren 
postiknam ... dúnoh  defia  CH.2. The meaning ‘statue’, which has been suggested for
the South Picene word, would fit both contexts: see §6.3.57 and §9.3.2.
Bibliography: Ritschl 1859:382-3; Ritschl 1862:30; Mommsen & Hensen CIL I p.255; Garrucci
1864:69 (autopsy); Fabretti CII 2440ter; Garrucci SIL 809; Zvetaieff IIM 66; Zvetaieff III 68; Schneider
1886:107 ¢28²; Bormann CIL XI.3157; Deecke 1888:197-8 ¢63²; Bronisch 1892:85; Conway
1897:381-2 ¢xli.a²; Von Planta 1897:588 ¢322²; Tambroni 1903:217; Herbig 1910:105-10 ¢19²;
Jacobsohn 1910:5 ¢33²; Herbig CIE 8339; Buonamici 1913:78-80 ¢50²; Lejeune 1952b:114-20 (au-
topsy); Vetter 1953:308 ¢319²; G. Giacomelli 1963:67-8 ¢58²; Pisani 1964:345 ¢147². Photograph:
Lejeune 1952b pl.XII. Drawings: Ritschl 1862 tab.XXXVI,B (whence CIL I p.255); Garrucci 1866
tav.IV.2 (reproduced CII tab.XLIII, IIM tab.VII.2, CIE 8339).
475* (possibly Middle Faliscan or Capenate?). Scratched on a small one-eared cup in

the Museum für Kunst und Gewerbe (Hamburg). Vetter (1953) assumed Capenate
provenance, as several pieces from the excavations at Contrada S. Martino had ended
up in the Museum by way of the Sammlung Reimer (cf. Ballheimer 1909).
iunai
Dextroverse. The n is . Vetter’s reading iunae, on which he assigns the inscription to
the second century, is a misreading of Von Mercklin’s drawing, which shows the last
letter as . The dextroverse ductus is not Middle or Late Faliscan, but could be
Capenate. The form iunai can be either genitive or dative (§8).
Bibliography: Von Mercklin 1930:92 (autopsy); Von Mercklin 1935:317-8 ¢15²; Vetter 1942:220;
Vetter 1953:328 ¢359b²; G. Giacomelli 1963:262-3 ¢VII². Photograph: Von Mercklin 1935 tav.
XLV.15. Drawing: Von Mercklin 1930:92 (reproduced in Von Mercklin 1935:317).
476* (possibly Middle or Late Faliscan or Capenate?). Scratched on a vessel bought

in Rome by Garrucci. “Puto originis esse sabinae et fortasse Capenatem” (Garrucci SIL
817, without giving any reasons for this assumption).
setorio
292
Pisani erroneously rendered caui tertinei as ‘Vibia Tertinia’.
581
CHAPTER 18
Sinistroverse. The s is given by Garrucci as , the t as , and the r as . Ductus and

letter-shapes, as well as the omission of word-final -s (§3.5.7d) and syllable-final r
(§3.5.7b), are all compatible with a Capenate or Faliscan provenance; cf. Praecilia
Setoriana in CIL XI.3181 from near Fabbrica di Roma (?).
Bibliography: Garrucci SIL 817 (autopsy); Gamurrini 1880 ¢828²; Zvetaieff IIM p.66 ¢h²; Deecke
1888:204 ¢70²; Bormann CIL XI.6706,11a; Conway 1897:384; Herbig CIE 8462; Buonamici 1913:92
¢63²; Lommatzsch CIL I2.476,11; Vetter 1953:329 ¢360a²; G. Giacomelli 1963:269 ¢XXXIX²; Briquel
1972:820-1. Transcription: Garrucci SIL 817 (reproduced in IIM p.66, CIE 8462, Deecke 1888 Taf.III,
CIL XI.6706,11a, CIL I2.476,11, Briquel 1972:820 fig. 7).
477* (Latin). Scratched before firing on a Megarian bowl. Late third or early second
century (Siebourg).
lquinti
Sinistroverse, Latin alphabet. The l is , the t . The form is either a genitive (Siebourg)
or an abbreviated nominative (§8.8.1). Fiorelli and Bormann gave the provenance as
Montefiascone, Contrada S. Lorenzo: Bormann, however, identified the bowl with one
in the Museo Municipale in Arezzo, whose provenance is given as S. Maria di Falleri
by Gamurrini293 (in CIL XI.6704,6), as Civita Castellana by Siebourg, and as Civita
della Chiana by the inventory of the Museo (thus G. Giacomelli). They are clearly
bowls from the same mould, as G. Giacomelli suggested: this would explain why
Siebourg described ‘his’ bowl as unpublished. Whether this bowl is from the ager
Faliscus is unclear. For an object at Arezzo, Civita della Chiana is a more likely
provenance than the ager Faliscus. Even if the bowl is from the ager Faliscus it is not
necessarily a local product. The sinistroverse ductus and the shape of the n are Faliscan
rather than Latin, but the alphabet, qu and the rendering of /nkt/ as nt are non-Faliscan
(cf. cuicto 310).
Bibliography: (I) Fiorelli 1883:434 (autopsy); – (II) Siebourg 1897:47-8 ¢13² (autopsy); – (III)
Bormann CIL XI.6704,6 (autopsy); Lommatzsch CIL I2.424; Safarewicz 1955:185; G. Giacomelli
1963:263 ¢VIII². Photograph: Siebourg 1897:46. Drawing: Siebourg 1897:46.
478* (Latin). Scratched before firing on a Megarian bowl. “Fundort nicht bekannt. Jetzt
in Arezzo im Museo municipale” (Siebourg 1891); “ex Civita Castellana [comparavit
GAMURRINIUS et collocavit in museo Arretino]” (Bormann CIL XI.6704,2e). Similar
bowls from the same workshop were found near Corchiano (295-296). c.230-150
(Baudrillart 1889:288-9).
[c]popili
Dextroverse, Latin alphabet. The p is , the l ᛧ; the o is open at the bottom. Genitive or
abbreviated nominative (thus Siebourg).
Bibliography: Siebourg 1897:44 ¢5² (autopsy); Lommatzsch CIL I2.419e; Bormann CIL XI.6704,2e.
Drawing: Siebourg 1897:42.
293
Lommatzsch and G. Giacomelli erroneously ascribed this autopsy to Garrucci.
582
18.3. Inscriptions that are probably not Faliscan or Capenate
18.3.1. Inscriptiones originis incertae uel ignotae. The following inscriptions are of
unknown and uncertain origin and probably not Faliscan.
479† (the ‘Vendia-inscription’, probably early Latin). Scratched on the lower part of
the body of an impasto pithos (height (incomplete) 35 cm, shoulder 45 cm; letters
15-25 mm high). Late seventh to mid-sixth century.
ecournatitauendiasmamaṛ[cos 6-9 m]ẹḍṿḥe[ked]
Dextroverse, with reversed s (). The alphabet shows no specifically Faliscan letter-
forms; the use of vh rather than f is non-Faliscan. Tita uendias is often interpreted as a
very early instance of a woman’s name consisting of praenomen and gentilicium (or
patronym?), in which case the lack of an ending in tita presents a problem. Pallottino
(1951:399) and Colonna (1980c:52) assumed a Gruppenflexion, the name being de-
clined as a unit tita-uendia- (as happens in Etruscan) but if the text is some form of
Latin, this is difficult. Lejeune took the words as a genitive, tita(s) uendias, but this
assumes a very early omission of -s, and after a long vowel, too (§3.5.7d).294 I am
inclined to take tita as the adjective */tito-/ ‘propitious, prosperous’ posited by Combet
Farnoux (1980:150-60),295 though perhaps in a less ponderous sense. I cannot take
serious Pisani’s interpretation of the text as a jocular claim that the urn is a teat (tita)
providing Vendia with wine, or Knobloch’s idea that an urna tita is a piggy-bank (with
an onomatopoetic adjective tita – at a time when minted coin had yet to be invented).
Following this, Pallottino read mamar[c ‘Mamerci (filiae/uxoris)’ (erroneously
mamarc[ 1951:398; mamar[c first Pallottino in Pisani 1953b:425). The letter before the
lacuna is similar to the d in uendia, which lead Pisani (1953b:425) to read mamad[eded
‘Mama dedit’ (adopted by Pallottino 1954; mamaḍ[*****]ḍẹ** G. Giacomelli), but the
r, maintained by Peruzzi (mamaṛ[ce), is confirmed from autopsy by Colonna
(mamar[cos).
After a lacuna of ten to twelve letters, the text continues as . The first trace
is the top of an e or a v; if the appendage of the following letter also belongs to this
letter, it can also be read as n. The second letter is a d (Pisani, Peruzzi, G. Giacomelli)
rather than an r (Pallottino), the third is an e (Pallottino, G. Giacomelli) or a v (Peruzzi,
Colonna, Prosdocimi, Agostiniani; Pisani’s f is a different transcription of the same
sign). This is followed by an h and a trace that is the upper left-hand corner of an a, an
e, a d, an m, an n, a v, or perhaps an h, a p, or an r. Pisani read fifico]ndf̣ẹị*[ ‘finxerunt
294
G. Giacomelli erroneously ascribes this interpretation to Pallottino (1951).
295
In saluetod tita from Osteria dell’Osa, tita is certainly a name (Colonna 1980c:52), but this
does not constitute an argument against the simultaneous existence of an adjective */tito-/.
583
CHAPTER 18
Fi...’, but this requires assuming a very early instance of  for f and the ei being written
as  = , whereas all the other letters are written quite distinctly. Peruzzi’s
m]ẹḍṿḥe[ked ‘me fecit’ has rightly been adopted by all editors, although Hartmann is
doubtful and considers also ]ẹṛẹḥạ[ or ]ṇḍf̣ẹị[ as possible alternatives. Combined with
his mamar[ce or Colonna’s mamar[cos, this still leaves leaves six to nine letters to be
restored, which perhaps constituted the name of a second potter, as in mama z[e]xtos
med f[.f]ịqod EF 1. I have considered reading mamaṛ[ce or mamaṛ[cos med fifik]ẹd, but
this would leave the following ẹḥ*[---] or ṿḥ*[---] unexplained. Hartmann notes that
the d in what is usually read as m]ẹḍ is in fact , which could also be read as an f (),
although he rightly concludes that this does not mean that the inscription is Faliscan.
Since two of the shards were found by Mengarelli and were in the Museo di Villa
Giulia before 1936 (the others were acquired independently by the Principe Massimo),
Pallottino gave the provenance of the vase as Cerveteri, where Mengarelli had worked
in the first quarter of the century. When Lejeune regarded the inscription as Faliscan,
this attribution was adopted by Pisani (1953b) and Pallottino (1954), the latter now
giving several arguments against his former attribution (the vase represented a type
found throughout South Etruria; the clay was different from that used at Cerveteri) and
in favour of the Faliscan attribution (Mengarelli had worked at Falerii as well; other
pieces in the Principe Massimo’s collection were of Falisco-Capenate provenance). Not
everyone adopted this attribution, however: Cerveteri is still the provenance given in
Cristofani 1990, while for Colonna, both the vase and the inscription are from Latium.
Lejeune’s arguments for the attribution, namely the occurrence of tita, which he
connected with titias EF 2, and the omission of -s in this word, are inconclusive: names
of the tita-group are widely attested, and the omission of -s in Early Faliscan is
uncertain, however frequent it may be in Middle and Late Faliscan (cf. §3.5.7d).
Peruzzi, like Colonna, regarded the language of the inscription as South Etrurian Latin.
The dextroverse ductus is, at this date, both Faliscan and Latin (§11.2.3); the form of
the s, , and the use of c before o are attested in Faliscan texts only from the fifth
century onwards (cf. §11.2.3-4). Note that the recently published Middle Faliscan
perfect forms facet MF 470* and faced MF 471* make it very unlikely that the Early
Faliscan form was ṿḥe[ked. All in all, the evidence for the inscription being from the
ager Faliscus, let alone Faliscan, is very slight indeed.
Bibliography: Pallottino 1951 (autopsy); Lejeune 1952b:120-1; Pisani 1953a:328; Pisani 1953b;
Scherer 1953: 116; Pallottino 1954; Knobloch 1958:137-8; G. Giacomelli 1963:261 ¢I²; Peruzzi 1963a;
Pisani 1964:349; Colonna 1980c:51-2 (autopsy); †Prosdocimi 1981; Agostiniani 1982:149, 242-3
¢587²; Prosdocimi 1983:LXV; Cristofani 1990:101 ¢4.4²; Silvestri 1993:106; †Solin 1999:385:
†Urbanová 1999:478; Hartmann 2005:29-32, Photographs: Pallottino 1951:399 figs.1-5; Colonna
1980c tav.X.3-6; Cristofani 1990:101; Hartmann 2005:29-33 Abb.27-32. Drawing: Pallottino 1951:398
(reproduced in Peruzzi 1963a:90).
480† (Sabellic, probably Palaeoumbrian?). Scratched on the handle of a bucchero

krater (height 28 cm, rim 24 cm, shoulder 27.5 cm; letter height 4-12 mm) from tomb
584
XVII of the Il Ferrone necropolis in the La Tolfa area (cf. Colonna 1983:574 n.2, with
literature). The inscription was added after firing, but appears to have been planned
beforehand, as the other handle was decorated, while the one with the inscription was
left blank. Date: c.560.
setums  míom | face
Dextroverse. Rix’s reading, which is followed here, is clearly preferable to Colonna’s
mośm : smutes | face, and shows that the language of the inscription is undoubtedly
Sabellic: in ST it is classed as a Palaeo-Umbrian text. Setums, reflecting Proto-Italic
*/septomo-/, shows the Sabellic Endsilbensynkope and a complete disappearance of /p/
before /t/, as occurred later in Umbrian (Rix 1992:247-8). Míom, too, is clearly Sabellic:
Faliscan has med or met, whereas míom has similar formations only in the Sabellic
languages: South Picene tíom TE.5 (nom.?), Umbrian tiu TI IIa.25 etc., and Oscan tiú
Sa 31 and siom TB I.5 etc. The perfect stem in face has parallels in the Sabellic
languages in, among others, the Umbrian future perfects fakust TI IV.31, fakurent TI
Ib.34, and facurent TI VIIa.43: it is now also attested for Faliscan in the recently
published Middle Faliscan inscriptions oufilo  clipeario  letei  fileo  met  facet 470*
and cauios frenaios faced 471*: see §5.3.1.6. For the formula, see §8.9.2. The only
reason for connecting this inscription with the Faliscan corpus is the shape of the f, ,
but the use of this sign was not constricted to Faliscan: see §11.2.2-3.
Bibliography: Colonna 1970:668 n.6; Colonna 1983a; De Simone 1983; Bakkum 1992:2; Rix 1992a;
Rix ST Um4. Photographs: Colonna 1983a tav.CVIII-CIX. Drawings: Colonna 1983a:576 fig.3, 578
fig.5 (repoduced in Rix 1992:243 fig.1).
481†. Cut in a strip of sandstone (letters c.3.5 cm high), originally probably part of the
ledge between two loculi, now set in the wall of the church of S. Serena at Foglia, near
Magliano Sabino. The use of local stone implies that the inscription was made locally.
[---]c ̣ịụifahls[?---]
Sinistroverse. The first letter is , probably a c, although an x does not seem impossi-
ble. According to Firmani, the letters iu () could be read together as n (). From his
drawing it is not clear whether the inscription ended with the s. I doubt if the text is
Faliscan, as Firmani suggested. The only Faliscan feature is the shape of the f, , but the
use of this sign was not limited to Faliscan: see §11.2.2-3. The sequence ah looks
Umbrian rather than Faliscan: cf. cavies  uhtav[---] Etr XLV from Lucus Feroniae, but
see also §3.5.7c.
Bibliography: Firmani 1977:116 (autopsy); Firmani 1979:119. Drawing: Firmani 1977:115 fig.30e
(reproduced in Firmani 1979 p.118 fig.2,f).
18.3.2. The ‘Faliscan’ inscriptions from Ardea. The following inscriptions are from
the necropoles of Ardea, but have, for various reasons, been regarded as Faliscan. This
question is addressed at the end of this section.
585
CHAPTER 18
482†. Scratched under a bucchero vase ( 9 cm; letters 5-15 mm) found on the site of
Pasqui’s necropolis a (cf. Barnabei 1882:71, Pasqui 1900:54-6) at Ardea.296 Seventh or
early sixth century according to Briquel and Colonna (1976a), although Colonna later
(1980b) ascribed it to the second half of the sixth century.
eqokaṇaios
Dextroverse, with reversed s ( , a very slender ). The n is : Prosdocimi (in Agostin-
iani) suggests kaụịaios (with upside-down u). Wachter’s suggestion Cavidios (i.e., kaui-
dios) is impossible: the letter following the n is certainly an a.
Dressel’s interpretation ‘ego K(aeso) Annaeus’ long remained unchallenged,
although already Mommsen (in Dressel) and Gamurrini (1894) doubted the likelihood
of such an abbreviated praenomen at this date. Since Colonna (1980b) raised this point
again, kaṇaios has become the established reading. For the Latin formula ego
OWNERNOM, see below ad fin. and §8.8.2: Wachter’s interpretation of kaṇaios as a
possessive adjective (“‘ich (bin) ein zur Familie ... gehöriges Gefäß’”) is interesting in
the light of what has been said in §4.4.11: see Colonna 1983b:55-7.
Herbig regarded the ductus, the use of q to render /g/, and the spelling of aṇaios
(= ‘Annaeus’) with a single n, as Faliscan elements. The dextroverse ductus occurs also
in several early Latin graffiti (see below), q for /g/ is also found in eqo CIL I2.479 and
2917c, and Colonna (1980b) has plausibly connected kaṇaios with (Latin) names in
Can-. There is no reason to assume that the inscription is Faliscan.
Bibliography: Barnabei 1882 (autopsy); Dressel 1882 (autopsy); Fiorelli 1882:273-4; Mommsen CIL
X.8336,1 (autopsy); Gamurrini 1887a:62; Gamurrini 1894:340; Herbig 1910:184,192; Diehl 1911:64
¢621²; Lommatzsch CIL I2.474; Ryberg 1940:121 ¢31²; Lejeune 1952b:123; Vetter 1953:331-2 ¢363²;
Cencetti 1957:195-6; Ernout 1957:53 ¢112²; G. Giacomelli 1963:261 ¢II²; Untermann 1964:178;
Briquel 1974:37; Colonna 1976a:372; Colonna 1980b:66 ¢36²; Agostiniani 1982:152 ¢596²; Colonna
1983b: 55-7; Degrassi & Krummrey CIL I2.474 add.; Wachter 1987:92-3; Silvestri 1993:107-8.
Photograph: Colonna 1980b:174 pl.18.3. Drawings: Barnabei 1882:71 (reproduced in Fiorelli
1882:273, CIL X.8336,1, CIL I2.474, Cencetti 1957:195 fig.13); Lejeune 1952b:122; Colonna 1980b:66
fig.10.
483†. Scratched under a red-varnished cup on a high foot (height 8.5 cm, rim 10.5
cm) found together with 484† in the third-century tomb q of necropolis c (Pasqui
1900:54-6). Vetter (1955:3 n.3) dated the tomb one or two generations earlier, on the
basis of his earlier dating of 484†.
titoio
Sinistroverse. The t is . The inscription was without argumentation regarded as
Faliscan by Herbig (and Jacobsohn), presumably on account of the form of the t and the
sinistroverse ductus, both normal in contemporary Faliscan inscriptions, but then
unparallelled in Latin inscriptions apart from the occurrence of sinistroverse ductus in
484†. Herbig interpreted titoio as a nominative after Etruscan tituie in an inscription
296
Ernout (1957:53) erroneously gave the provenance as Falerii.
586
published by Buonamici (1931:410); cf. also Etruscan titui Pe 1.622-624, 627. Pisani
and Dirichs interpreted titoio as a genitive in -oio /-oo/ m -osio /-oso/ (§4.4.10). For
those who (unlike Pisani) regarded the genitive in -osio as exclusively Faliscan, this
interpretation provided an additional argument for a connection between Ardeatine and
Faliscan, though not for regarding the inscription itself as Faliscan, as there the genitive
was -i at this time (thus e.g. Safarewicz and Wachter). As discussed in §4.4.5, the
possible existence of a genitive in -oio in Latin or Faliscan is at best extremely ques-
tionable, and in my view untenable. The possibility that titoio is a possessive adjective
titoio(m) (as has been suggested by Vetter and Hamp) can be excluded at this date.
Hamp in fact curiously suggests that Ardeatine preserved the genitive in -osio (an
assumption based exclusively on the traditional interpretation of titoio as a genitive in -
oio), but that under the influence of the surrounding areas where -i was used, this was
adapted to “a less startling adjectival -o-io(m)” (Hamp 1981:231).
Bibliography: Pasqui 1900:59 (autopsy); Herbig 1910:181, 184 ¢21² (autopsy); Jacobsohn 1910:6 ¢48²;
Pisani 1933:624 n.1; Pisani 1934:295; Dirichs 1934:22; Lejeune 1952b:124-5; Safarewicz 1953:246;
Vetter 1953:332 ¢364a²; Pisani 1955:322; Vetter 1956:1-2; G. Giacomelli 1963:261-2 ¢III²; Pisani
1964:342-3 ¢146D²; Devine 1970:22; Hamp 1981:230; Lejeune 1989:68. Photograph: Lejeune 1989
between pp.64-5.
484†. Scratched on the bottom of a red-varnished plate (height 5 cm, 15 cm; letter
height 3-6 mm in Thulin’s drawing), found together with 483† in tomb q of necropolis
c, dated to the third century by Pasqui (1900:54-6). Vetter (1955:3 n.3) dated the
inscription one or two generations earlier because of the letters, the double interpunct,
and the sinistroverse ductus.
neuendeiuo
Sinistroverse; the first e is , the others (thus Thulin; Pasqui’s and Lommatzsch’s
transcripts give all e’s as ). Thulin treated the inscription as Faliscan (with neuen as a
form of Latin Naevenna = Etruscan Cnevne), probably on account of the form of the e,
parallelled only in MF 146. The attribution was adopted by Herbig and Jacobsohn, and
by Lommatzsch, who interpreted ‘ne ven(das); deivo(m)’. This interpretation was in
turn adopted by Diehl and Warmington (who, however, translated ‘for a god’, appar-
ently regarding deivo as a dative), and still considered by G. Giacomelli. Lommatzsch
later even discarded the text from the CIL because it was Faliscan (CIL I2.455 add.);
later editors have regarded the inscription, with various degrees of hesitation, as Latin.
Pisani’s interpretation (1943:259) of the text as a dedication to the di novensides has
been adopted by virtually all later editors, especially when it was elaborated by Vetter
(1956); only Weinstock still maintained Thulin’s interpretation.
Bibliography: Pasqui 1900:59 (autopsy); Thulin 1907:308 ¢64² (autopsy); Herbig 1910:194; Jacobsohn
1910:6 ¢49²; Lommatzsch CIL I2.455+add. p.714; Diehl 1930:79 ¢754²; Lommatzsch CIL I2.455 add.;
Warmington 1940:74-5 ¢57,x²; Pisani 1943:259; Weinstock 1952:155; Vetter 1953:332-3 ¢364b²;
Safarewicz 1954:102-3; Safarewicz 1955:185; Vetter 1956; G. Giacomelli 1963:262 ¢IV²; Pisani
1964:346 ¢149²; Camporeale 1967:70; Hamp 1981:228-30; Degrassi & Krummrey CIL I2.455 add.;
Wachter 1987:99, 374. Drawing: Thulin 1907:308.
587
CHAPTER 18
Although the attribution of these three inscriptions to Faliscan has been rejected in each
individual case by more than one author, the impression that Ardeatine and Faliscan are
in some way related or ‘close’ seems to persist independently. I therefore review here
again the possible arguments for an Ardeatine-Faliscan connection, mainly those
proposed by Wachter (1987:99).
First, there are the epigraphical similarities between the inscriptions from Ardea
and the Faliscan inscriptions. In the case of eqo kaṇaios 482†, Wachter noted that it has
a dextroverse ductus, like the Early Faliscan inscriptions, whereas the contemporary
Latian inscriptions are sinistroverse. His point of comparison, however, are the longer,
partly ‘official’ early Latin inscriptions CIL I2.1, 4, 2658, 2832a, 2833, and 2833a.
Leaving aside the fact that even in this group there are two instances of dextroverse
ductus (CIL I2.2832a and 2833a), it seems to me that better material for comparison is
provided by the contemporary Latin Besitzerinschriften, where dextroverse ductus is
quite common (e.g. CIL I2.479, 2830, 2832, 2916g,d,e,i, 2916k,a,c,d, 2917a, and
2917b). Another argument might be the shape of the s, which is  (Wachter gives it as
) and seems to have parallels in the ager Faliscus (EF 1 and 4, cf. §12.2-3) but
apparently not in Latium.
Titoio 483† and neuen  deiuo 484†, too, contain epigraphical features associated
with the ager Faliscus rather than Latium. The most striking is probably the sinistro-
verse ductus, which is without clear parallels in contemporary Latin inscriptions,
although it should be noted that among the Faliscan inscriptions themselves there is a
small amount of inscriptions where the ductus is dextroverse instead of the usual
sinistroverse (§11.1.5). Interesting, too, are the form of the t in 483†, , which is the
normal form in the contemporary Faliscan alphabet, but is rare or absent in the Latin
alphabet, and the form of the first e in 484†, , paralleled only in MF 146. The
epigraphical correspondences between the Ardeatine and the Faliscan inscriptions are
therefore undeniable, but their implications are unclear: I rather suspect that if the
inscriptions had surfaced without the circumstances of their finding being known, they
would have been ascribed on epigraphical grounds to the ager Faliscus.
The linguistic arguments for a ‘Faliscan-Ardeatine connection’ are even fewer:
(1) The occurrence of eu in neuen 484†. If the interpretation of neuen as a dialectal
form of nouem is correct, this word would appear to show a retention of PIE */eu/. This
presupposes that the merging of PIE */eu/ with */ou/ did not occur at the Proto-Italic
stage, as is usually assumed, but took place later and separately in the various Italic
languages – or indeed dialects (§3.2.5). Wachter points out that eu is found also in Early
Faliscan euios in 1 and ieuotenosio in 3 (as he reads it), while on the other hand early
Roman has ou in iouxmen|ta and iouestod in CIL I2.1, and in iouesat in CIL I2.4.
Faliscan and Ardeatine, in his view, both retained /e/ longer than Roman.297 Even if
297
Note that iouestod and iouesat may go back to an old o-grade */iouos-/ (DÉ s.v. iūs), and
that in G. Giacomelli’s (1963:41-44) interpretation of EF 1, which Wachter appears to follow,
588
this were true, however, the comparison between the two is fairly slight, for even if in
Early Faliscan /e/ was retained until later than the Roman inscriptions CIL I2.1 and 4,
by the third century it had certainly merged with /o/ and thence monophthongized to
/ō ̣/ (§3.7.2). Vetter (1953:333) in fact quoted the inscriptions neuna  fata CIL I2.2846
and neuna  dono CIL I2.2845 from Lavinium as parallels for the neuen, and not a
Faliscan inscription. Wachter’s comparison would show nothing more than that /e/
was rounded at different times in the different Latin dialects, but thist is not an argu-
ment in favour of a specific connection between Faliscan and Ardeatine – rather the
opposite.
(2) The possibility that titoio 483† is a genitive in /-oo/ m /-oso/. This interpretation
is extremely debatable, to say the least (§4.4.10); if it is adopted, it should be noted that
in the contemporary Faliscan inscriptions the ending of the genitive is -i, and that any
connection between Ardeatine and Faliscan based on an Ardeatine genitive titoio must
therefore be diachronic. In that case, an Ardeatine genitive -oio /-oo/ m /-oso/ is an
argument for a specific connection between Ardeatine and Faliscan only if /-oso/ is
regarded as exclusively Faliscan, and the argument goes back to a time when this was
the prevalent opinion (§4.4.1). If /-oso/ was at any one time the (general) Latin ending,
the fact that a reflex of this ending is found both in Ardeatine and in Faliscan cannot in
any way constitute an argument for an Ardeatine-Faliscan connection: and the occur-
rence of -osio in the inscription from Satricum (CIL I2.2832a) indicates precisely this.
(3) The omission of word-final consonants in titoio in 483† and deiuo in 484†. The
omission of word-final -s and -m, although virtually universal in Middle Faliscan
inscriptions (cf. §3.5.7d,a), occurs too often in Latin inscriptions from other locations to
count as a criterion.
From an epigraphical point of view, there are similarities between these inscriptions
from Ardea and those from the ager Faliscus. From a linguistic point of view, I see no
indications for a specific connection between the two, apart from the general similarities
between what are, in my view, dialects of the same language.
18.3.3. The Satricum-inscription as Faliscan. In a recent article, Lucchesi (2005) has

suggested that the Satricum-inscription (CIL I2.2832a) may itself be Faliscan. Her
arguments (and my objections) are the following:
(1) Publius is attested in the form Poplios only in Faliscan inscriptions. This is true
(except, of course, for the Satricum-inscription itself), but this is probably due to the
fact that Faliscan is the only Latin dialect that provides a large number of relatively
early inscriptions. The argument, of course, depends on the way the name is derived: if
there was a (real or perceived) connection with populus, or a connection with Etruscan,
euios is a Greek loan (EÜioj) and therefore can provide no information about the Italic develop-
ment of PIE */eu/.
589
CHAPTER 18
where the pronomen is attested as puplies Vs 1.29 etc. (§7.7.1.51), there is a very good
chance that the contemporary Latin spelling of the name was likewise Poplios. Lucchesi
makes no reference to the fact that in the ager Faliscus, the name is not attested for
Early Faliscan and is in fact attested for men only in MLF 316 and perhaps in MLF 317,
and in abbreviated form in LF 337, Lat 250, and Cap 409 and 462. Its popularity as a
man’s name therefore appears to be late, and perhaps due to Latin influence. The name
is certainly not a typical Faliscan name (cf. §7.7.2), and occurs also in Latin and
Etruscan. Note that this is an onomastic argument rather than a linguistic one.
(2) The gens Valeria, or at least P. Valerius Publicola, is of Faliscan origin. This
argument depends on the identification of the popliosio ualesiosio of the Satricum-
inscription with P. Valerius Publicola, suggested by Versnel (1980) at a time when the
genitive in -osio was still regarded as exclusively Faliscan by many scholars, as well as
on the likelihood of P. Valerius Publicola being a Faliscan, for which Lucchesi only
refers to very indirect evidence (the myth of ‘Valeria Luperca’ in [Plut.] Vit. Min. 35).
Note that in spite of the large number of gentilicia attested for the ager Faliscus and
Capenas, there is no attestation of Valerius or its pre-rhotacist predecessor Valesios. The
argument is, again, not a linguistic one.
(3) The inscription is dextroverse, like the Early Faliscan inscriptions, while the early
Latin inscriptions are sinistroverse. This, however, is true only for the earliest Early
Faliscan inscriptions (EF 1-4): the others (EF 6-10, 467*, and EF/Etr 385) show that the
change to a sinistroverse ductus was already under way during the sixth and early fifth
centures, not “perhaps in the 5th or at the beginning of the 4th” (p.263).
(4) Steterai is assumed to be Faliscan, apparently because it is a reduplicative perfect
that is not attested for Latin. I do not quite understand why it would then be Faliscan,
since this perfect is likewise unattested for Faliscan: this is in fact an illustration of the
problems of synchronic comparison for the early periods discussed in §10.1.2. Or is it
Faliscan because Faliscan had other reduplicative perfects that Latin did not have, such
as fefiked EF 9 / f[.f]ịqod EF 1? Interestingly, Lucchesi does not discuss the ending of
steterai, which is most certainly not attested for Faliscan, where only f[.f]ịqod EF 1
shows, not a perfect ending, but an aorist ending: a difference that is usually regarded as
one of the major features separating Faliscan from Latin (§5.2.4.5, §10.1.2-3).
(5) -osio is not a possessive genitive: this harks back to the criticism of Untermann
(1964) on the assumption that a genitive in -osio was replaced by a genitive in -i, which
I discussed in §4.4.9. I do not see how this could be an argument for or against the
Satricum-inscription being Faliscan.
(6) The greatest obstacle against a Faliscan attribution of the Satricum-inscription, is, of
course suodales, since this is usually derived from PIE */suedhħ1-/, which would give
*suefales or *suofales as the expected Early Faliscan form. Lucchesi shows that
suodales can also be derived from PIE */sued-/, which would remove this obstacle, and
590
I think that this is possible. In that case, however, the form suodales might just as well
be Latin instead of Faliscan.
(7) The circumstances of the find show that the inscription may have come from
another location. This, of course, is true: the stone of the Satricum-inscription was later
reused in the Satrican temple of Mater Matuta, and its original location is unknown.
Unless, of course, it is assumed that the stone was dragged all the way from Falerii (or
any other location), I do not see how this could be an argument for the language of the
Satricum-inscription being anything else than (local) Latin.
(8) The fictile decoration of the Satrican temple may be related stylistically to that of the
Faliscan temples. I see no reason to doubt this, but all that this argument does is suggest
contact between the Satrican and the Faliscan area. I am quite happy to assume that
such contacts existed, but I can see no linguistic traces of it in the Satricum-inscription.
In my view, all these arguments amount to very little more than a ‘it is possible that ...’.
Unfortunately, so are a lot of things. If the Satricum-inscription is anything else than
(local) Latin, there must be some positive, preferably linguistic, argument on which to
base this assumption. In fact, if the inscription were anything else than local Latin, the
first candidate would probably be Volscian, as was suggested by Coleman (1986:120-
2): see note 67. It is unfortunate that Lucchesi’s suggestion has been embraced rather
enthusiastically by R. Giacomelli (2006:25-7 et alibi).
591
CHAPTER 18
592
Chapter 19
The Etruscan inscriptions
19.1. The presentation of the Etruscan inscriptions
Although the Etruscan inscriptions from the agri Faliscus and Capenas fall outside the
primary scope of this investigation, they cannot be omitted, as Etruscan was one of the
languages spoken and written in the area (cf. §9.2). This chapter contains the forty-four
Etruscan inscriptions from the ager Faliscus and Capenas (Etr I-XLIV), together with
seven inscriptions that are originis incertae vel ignotae (Etr XV-LI). Inscriptions that I
regard as Faliscan or as possibly Faliscan are not presented here, but in the correspond-
ing sections of chapters 12-18: references to these inscriptions are given in the introduc-
tions to the relevant sections of this chapter. The Etruscan inscriptions are presented in
the same way as the Faliscan, except that epigraphic and interpretational discussions
have been kept to a minimum, as the material is presented as an addition to the corpus
presented in chapters 12-18.
19.2. Narce and the south-western ager Faliscus
The south-western ager Faliscus, with its centre at Narce (near Calcata), yields the
largest number of Etruscan inscriptions (Etr I-XXIV) from any single area within the
area of the ager Faliscus (24 out of a total of 44-51). Virtually all these inscriptions are
from the archaic period: at the same time, the area yields no Faliscan inscriptions at all.
This is of course understandable in the light, first, of the role of Narce as the dominant
Etruscan town of the area until the sixth century, closely connected to Veii, and, second,
of the early Latinization of the area after the founding of a colony at Nepete (modern
Nepi) in the early fourth century (see §2.5.2 and Cristofani 1988:16-7). For the ex-
cavations at Narce, see Barnabei 1894a:21-6, Cozza 1894, and Potter 1976:7-16).
I-II. Scratched under the body of an impasto cup (height 11 cm, rim 13.1 cm; letters
5-14 mm high) from tomb 7/LVIII of the necropolis at Contrada Morgi, Narce (Pasqui
1894:516 with plans fig. 204 and Barnabei 1894a fig. 3,O). Mid-seventh century.
abcdevzḥθik
ara
593
CHAPTER 19
Dextroverse, Etruscan alphabet. The ḥ is  (cf. Rix 1992:25), while the θ is . For the
b and the order of the letters, which occurs also in Etr XLIV from Capena, see
Bundgård and especially Pandolfini & Prosdocimi.
Bibliography: [Pasqui 1894:521 (autopsy)]; Gamurrini 1894:321-2, 327-41 ¢1² (autopsy); Lattes 1895:
498-501; †Lattes 1908:310 ¢9²; Herbig CIE 8414 (autopsy); [Weege in Helbig 1913:377 (autopsy)];
[Della Seta 1918:96 (autopsy)]; Neppi Mòdona 1926:502-3; Buonamici 1932:111-2; G. Giacomelli
1963:270 ¢XLV²; Bundgård 1965:27-29 ¢2²; Cristofani 1988:21 ¢1²; Pandolfini & Prosdocimi 1990:21-
2 ¢I.2²; Rix ET Fa X.3=Fa 9.1+0.1; Rix 1992:251. Photographs: Montelius 1904 pl.328,5 (reproduced
in Buonamici 1932 tav.IV fig.7); Neppi Mòdona 1926:502 fig.6; Bundgård 1965:27,29 figs.13-15;
Cristofani 1988 pl.Ia-b; Pandolfini & Prosdocimi 1990 tav.II-III; Drawings: Gamurrini 1894:321
fig.165 (reproduced in CIE 8414), 322 fig.165a; Herbig CIE 8414; Pandolfini & Prosdocimi 1990:22.
III-VII. The following inscriptions are from tomb 2/LX of the third necropolis south of
Pizzo Piede, Narce (Pasqui 1894:474-5 with plans fig.196 and Barnabei 1894a fig.3,J).
Fig.19.1. Gamurrini’s and Nogara’s drawings of the middle part of Etr IV.
left: Gamurrini’s drawing (from Gamurrini 1894:325-5 fig.167b).
right: Nogara’s drawing (from CIE 8415).
III-IV. Scratched, III on the neck, IV on the body of an impasto oenochoe (height 21
cm, body 12.3 cm; letters 3-8 mm high). Third quarter of the seventh century.
miqutunlemausnasranazuzinace
erunaletaseruepninaitaleṭạm*(*)upesitatatuθacetu[5-6]ṭaθine[?---]
Dextroverse, Etruscan alphabet. The s is : in IV, it is reversed in erunaletas (but not in
ṭạm*(*)upes). The a and the u of lemausnas (lemnesnas Gamurrini and Lattes; lemnas-
nas Torp) are written very close together and have been treated by some editors as a
ligature. In IV, Herbig, G. Giacomelli, and Pallottino read epninaṛtale after Nogara’s
drawing, instead of epninaitale. Following this is the damaged part reproduced in
fig.19.2. The first three letters are probably ṭạm; what follows has been read as e*u
(Gamurrini, Rix), ẹiu (Torp, Pallottino), eịṇ (Cristofani), or ṿịu (Herbig and G. Gia-
comelli). Nogara’s drawing shows a trace on the edge of the first lacuna in IV, shaped
like the top of a t, or perhaps a z. For the word qutun, cf. the discussion on Early
Faliscan quto in EF 3 (§6.2.30, §12.3).
Bibliography: [Pasqui 1894:476 (autopsy)]; Gamurrini 1894:322, 342-3 ¢3² (autopsy); [Karo 1896:5];
Lattes 1896:33-38 ¢7-8²: Torp 1906:8-10 ¢2²; Cortsen 1908:84, 86-7; †Lattes 1909:64 ¢4²; Bugge/Torp
1909:24; Danielsson 1910:98-101; Herbig CIE 8415a-b (autopsy); [Weege in Helbig 1913:377
594
THE ETRUSCAN INSCRIPTIONS
(autopsy)]; [Della Seta 1918:96 (autopsy)]; Goldmann 1928:217; Stoltenberg 1956:30-1 ¢IXa-b²; G.
Giacomelli 1963:270 ¢XLVIa-b²; Pallottino TLE 28a-b; De Simone 1968:109; Colonna 1974:140;
Colonna 1975a:181-4; Agostiniani 1982:68 ¢87²; Colonna 1987:62 n.48; Cristofani 1988:22 ¢4²;
Colonna 1990:125; Rix ET Fa X.1=Fa 2.1+6.2+0.2. Drawings: Gamurrini 1894:325-6 fig.167 (repro-
duced in Montelius 1904 pl. 328,3a, CIE 8415), fig.167a (reproduced in Montelius 1904 pl.328,3b, CIE
8415), fig.167b (reproduced in Montelius 1904 pl.328,3c, CIE 8415); Nogara CIE 8415.
V. Scratched under the foot of a plate ( rim 27 cm, foot 7.6 cm; letters 16-26 mm
high). Second half of the seventh century.
misaza
Dextroverse, Etruscan alphabet with reversed s. For z = [ts] or [tʃ] (Agostiniani), cf. lazi
Etr XI-XV = larti and lazia Etr XVII = lartia.
Bibliography: Pasqui 1894:477 (autopsy); Gamurrini 1894:343 ¢4² (autopsy); Lattes 1895:501-3 ¢2²;
Herbig CIE 8416 (autopsy); [Weege in Helbig 1913:377 (autopsy)]; [Della Seta 1918:96 (autopsy)]; G.
Giacomelli 1963:270 ¢XLVII²; Agostiniani 1982:68 ¢86²; Colonna 1983b:53-4; Cristofani 1988:22 ¢5²;
Rix ET Fa 2.2. Drawings: Gamurrini 1894:327 fig. 168; Herbig CIE 8416.
VI-VII. Scratched on a fragment of the rim of a plate similar to that of V (letters 12-14
mm high).
her
Dextroverse, Etruscan alphabet. The h is : oed Gamurrini.
Together with this, Herbig published a second plate (height 5.8 cm, 14.7 cm) with the
same inscription scratched around its foot (letters 10-14 mm high).
her
Sinistroverse, Etruscan alphabet. The h is . The provenance of this plate is not given;
the inventory numbers suggest they were not found together. Cf. hermana MF 265.
Bibliography: Pasqui 1894:477 (autopsy); Gamurrini 1894:343-4 ¢4² (autopsy); Herbig CIE 8417-8418
(autopsy); [Weege in Helbig 1913:377 (autopsy)]; G. Giacomelli 1963:270 ¢XLVIII,i-ii²; Colonna
1990: 125. Photograph: Colonna 1990:142 tav.IIa. Drawings: Gamurrini 1894:328 fig.169; Herbig
CIE 8417-8418.
VIII. Scratched c.3½ times around the foot of an impasto cup (letters 7-13 mm high)
from tomb 38/LIV of the necropolis of Monte Cerreto, Narce (Pasqui 1894:505-7 with
plans fig.201 and Barnabei 1894a fig.3,M). Mid-seventh century.
mialiquauvilesialeṣpuraθeṿnalθiainpeinmlerusiaterimlaχutaziχuχemlaχta
anazinace
Dextroverse, Etruscan alphabet with reversed s (). Of the letters between auvilesi and
pura, only the tops have been preserved: they have been read as al ..  (Gamurrini). aḷạ
Torp, alat  (Lattes), ales (Herbig, Goldmann, G. Giacomelli),  ales (Cristofani) and
ales (Rix). The third letter of θe*nalθia is  and has been read both as an a (Gamurrini,
Lattes, and Torp) and as a v (Herbig, Goldmann, G. Giacomelli, and Pallottino).
595
CHAPTER 19
Bibliography: [Pasqui 1894:509 (autopsy)]; Gamurrini 1894:322-4, 341-2 ¢2² (autopsy); Lattes 1896:
10-25 ¢5²; Torp 1905:32-3; Torp 1906:8; Bugge/Torp 1909:24; Herbig CIE 8413 (autopsy); [Weege in
Helbig 1913:377 (autopsy)]; [Della Seta 1918:95 (autopsy)]; Goldmann 1928:216; Stoltenberg 1956:30
¢VIII²; G. Giacomelli 1963:270 ¢XLIV²; Pallottino TLE 27; Pfiffig 1969:199, 235; Colonna
1975a:181-4; Agostiniani 1982:68 ¢85²; Cristofani 1988:21 ¢2²; Colonna 1990:125 n.57; Rix ET
Fa X.2=Fa 3.1+6.1. Photographs: Cristofani 1988 pl.IIa-c. Drawings: Barnabei 1894b:255 fig.117a;
Gamurrini 1894:323 fig.166 (reproduced in Montelius 1904 pl. 328,1a, CIE 8413), 324 fig.166a
(reproduced in Montelius 1904 pl.328,1b); Herbig CIE 8413.
IX-X. The following inscriptions are from the necropolis of Monte in Mezzo ai Prati,
Narce (Pasqui 1894:540-5 with plans fig.210 and Barnabei 1894a fig.3,V).
IX. Scratched on the rim of a bucchero kylix (height 12.8 cm, 27 cm; letters 6-10 mm
high) from tomb 1/LXIII. Last quarter of the seventh century.
[---]ḳalikeapạminikara
Sinistroverse, Etruscan alphabet. The text starts with the bottom half of a k (thus first
Gamurrini).298 The letter preceding mini is , which Rix reads as a.
Bibliography: [Pasqui 1894:542 (autopsy)]; Gamurrini 1894:324-7, 344 ¢6² (autopsy); Lattes 1896:2-10
¢4²; [Ghirardini 1900:175,187,190]; Thulin 1908:258; [Grenier 1912:385]; Herbig CIE 8411 (autopsy);
[Della Seta 1918:97 (autopsy)]; G. Giacomelli 1963:269 ¢XLII²; Pallottino TLE 26; Agostiniani
1982:68 ¢88²; Cristofani 1988:22 ¢6²; Rix ET Fa 3.3. Drawings: Gamurrini 1894:329-30 fig.170
(reproduced in Montelius 1904 pl.328,4a, CIE 8411), fig.170a (reproduced in Ghirardini 1900:185-6
fig.60), fig.170b (reproduced in Montelius 1904 pl.328,4b); Herbig CIE 8411.
X. Scratched two times around the foot of a bucchero cup (height foot 6 cm, foot 12.7
cm; letters 3-13 mm high: the words ipas ikam are written smaller) from tomb
15/LXX. Late seventh or early sixth century.
·i·pas· ·i·ka·m·
a·rnunaturaniriasekaseletakalemθasvainiastaa·χavisu·r·a·lχuname·a·χaχuname
iθavusvaka·i·tasemlecivaθeneikania
Dextroverse, Etruscan alphabet. The interpuncts are given in various ways, especially
by the earlier editors. In the case of the a’s in arnuna, aχavisur, and alχuname, the
interpuncts stand underneath rather than between the letters.299 In iniasta, the s is  (x
Gamurrini, Torp; i (i.e., ·. = ·i·) Herbig, G. Giacomelli).
Bibliography: [Pasqui 1894:545 (autopsy)]; Gamurrini 1894:344-6 ¢7² (autopsy); Lattes 1896: 25-33
¢6²; Torp 1906:4-8 ¢1²; Bugge/Torp 1909:120-1; Herbig CIE 8412 (autopsy); [Weege in Helbig
1913:378 (autopsy)]; [Della Seta 1918:98 (autopsy)]; Vetter 1926:279; Vetter 1939b:160-1; Buonamici
1942:296-8; Durante 1953; Slotty 1955:195; G. Giacomelli 1963:269-70 ¢XLIII²; Pallottino TLE 29;
Cristofani 1988:22 ¢7²; Colonna 1990:125 n.57; Rix ET Fa 0.4. Photographs: Montelius 1904 pl.328,2;
Herbig CIE 8412 (reproduced in Buonamici 1932 tav.XLVIII fig.83); Cristofani 1988 pl.IIIb. Draw-
ings: Gamurrini 1894:332 fig.171 (reproduced in CIE 8412, Colonna 1990:124 fig.5), 333 fig.171a;
Buonamici 1942: 296-8 figg.22-27.
298
The trace is omitted altogether by Cristofani, who read [---]alike.
299
Rix also punctuates the i of iθavusvaka (and erroneously reads aθeneicania with a c).
596
XI-XV. Scratched under five Etrusco-Campanian plates ( 18, 17.5, 13, 12.5, and 8 cm
respectively) from tomb 18 of the Il Cavone necropolis at Monte Li Santi, Narce
(Pasqui 1894:456 with plans fig. 192 and Barnabei 1894a fig. 3,G). Fifth century.
laziveiane·s·
laziveiạne·s·
laziveiane·s·
laziveiane·s·
laziveiaṇes·
Sinistroverse, Etruscan alphabet. The second a in XII is . In XV, the s has only one
interpunct. For zi = [ts] or [tʃ] (Cristofani), cf. lazia Etr XVII = lartia, and mi saza Etr V.
Bibliography: [Pasqui 1894:461 (autopsy)]; Herbig 1910:194-5 ¢35² (autopsy); Herbig CIE 8419-8423;
Buonamici 1913:77 ¢47²; Buonamici 1935:343; Vetter 1939b:160; Buonamici 1942:299-300; Slotty
1955:28 ¢48²; G. Giacomelli 1963:271 ¢L²; Cristofani 1988:22 ¢8²; Rix ET Fa 2.6-10. Drawing: Herbig
CIE 8419-8422 (reproduced in Buonamici 1942:299 fig.28).
XVI. Scratched inside a red-varnished cup (height 6.8 cm, 17.5 cm; letters 8-20 mm
high) from Narce. Second half of the fifth century.
velθarusvelanas
Sinistroverse, Etruscan alphabet. The letters are carelessly written: the first v is , the
second , first e , the second . The θ is diamond-shaped, without central point.
The only parallels for the a in velθarus are velθạr[nal] AT 1.71 and velθaruσ́a Cr 1.22:
in all other instances of this name, it is velθur or veltur (more than 125 attestations in
ET): cf. velθurusi in Etr XIX.
Bibliography: Buonamici 1941 (autopsy); G. Giacomelli 1963:271 ¢LIV²; Cristofani 1988:22 ¢9²; Rix
ET Fa 2.12. Drawing: Buonamici 1941:370 fig.3.
XVII. Scratched in a black-varnished saucer ( 18 cm) found in a cistern during

Mengarelli’s excavations in 1933 in the habitation at Pizzo Piede, Narce. First half of
the third century.
ṃitafinalaziavilianas
Sinistroverse, Etruscan alphabet. Of the first letter, the last branch is left (ṃi, not
Pallottino’s and Cristofani’s [m]i). The f is , as in Etr XXXVI. The form tafina
occurs also in tafina Etr XXXVI: elsewhere, it is always θavhna/θahvna, θafna, or
θapna (attestations are given under Etr XXXVI). For lazia = [ts] or [tʃ] (Cristofani
1988:16), cf. lazi Etr XI-XV = larti, and also mi saza Etr V.
Bibliography: Buonamici 1935:341-2 (autopsy?); Vetter 1940:135; Slotty 1955:195; G. Giacomelli
1963:270 ¢XLIX²; Pallottino TLE 30; Colonna 1974:133-6 ¢7²; Agostiniani 182:69 ¢89²; Cristofani
1988:23 ¢10²; Rix ET Fa 2.14. Drawing: Buonamici 1935:341.
597
CHAPTER 19
XVIII. Scratched inside the rim of a bucchero oinochoe (height 20.8 cm) from Narce.
C.550-520.
micipaχ
Sinistroverse, Etruscan alphabet. The reading of the CVA is mi cilaχ, but the drawing in
Gran-Aymérich & Briquel seems to support their reading cipaχ. Cipaχ recalls the
Etruscan toponymic adjectives in -aχ: Briquel, comparing zilaχ : zilaθ, tentatively
suggests a connection between cipaχ and cepen/cipen, so that cipaχ may have had a
meaning ‘sacred’, ‘consecrated’.
Bibliography: CVA Louvre 23 p.91; Gran-Aymérich & Briquel 1997 ¢50². Photograph: CVA Louvre
23 pl. 36,8. Drawing: Gran-Aymerich & Briquel 1997:429.
XIX. Painted in white around the body of an ovoid crater of red impasto (h. 21 cm,
body 12.3 cm) from tomb 98 of the Principe Del Drago’s excavation of the necropo-
lis on the northwest side of Monte Soriano, near Mazzano Romano (cf. Pasqui
1902b:612). Third quarter seventh century.
Fig.19.2. Pasqui’s drawing of the damaged part of Etr XIX (enlarged).

(From Pasqui 1902b:613 fig.9.)
mimulularicesip[....]ṃḷ[...]*s**naiesiclinsivelθurusilar*sruvries
Dextroverse, Etruscan alphabet. Mulularice is an error for mulu‹v›a‹n›ice. Pasqui read
the middle as p*[..]*mlạp̣ị[..]svunaiesi. Other readings are p*-(-)**ṃḷ*p ̣ị-*ṣṿụnaiesi
(Herbig), p*...*ml*pi...s**naiesi (Pallottino), p*( )**ṃḷ*( )ṣvụnaisesi (G. Giacomelli),
p[ ]si[...]naiesi (Cristofani), and p[----]ṃ ḷ[---]-s--naiesi (Rix). The trace in lar*s as
given in Pasqui’s drawing may be the bottom half of an a. Why Cristofani reads c[li]nsi
and lar[...]uvries I do not know: the letters clinsi and lar*sruvries appear to be certain.
Bibliography: Pasqui 1902b:613-5 (autopsy); Herbig CIE 8426; G. Giacomelli 1963:271 ¢LI²; Pallot-
tino TLE 32; Agostiniani 1982:69 ¢90²; Cristofani 1988:22 ¢3²; Colonna 1990:125 n.57; Rix ET Fa 3.2.
Drawings: Pasqui 1902b:613 fig.9 (reproduced in Montelius 1904 pl.330,14, CIE 8426), 614 fig.10
(reproduced in CIE 8426); Nogara in CIE 8426.
XX. Stamped on a tile fragment (letters 3.9-4.5 cm high) found close to a tomb near the
Minchione Bridge over the Fosso dell’Isola, c.8-9 km from Nepi (cf. Polidori
1977:296).
micusulpuiunal
598
Dextroverse, Etruscan alphabet. The m is , the n .

Bibliography: Polidori 1977 ¢27² (autopsy); Rix ET Fa 1.6. Photograph: Polidori 1977 tav.XLV.
Drawing: Polidori 1977:297.
XXI-XXIII. The following vessels were found in tomb I of a group of tombs excavated
in 1889 at Vigna Pentriani (‘Villa Pentriani’ CVA Italia 3 p.24), località S. Paolo, Nepi.
XXI. Scratched under the foot of an Attic kylix, black-figured within and red-figured
without (height 14 cm, rim 32.5 cm, foot 12 cm; letters 11 mm high). C.520.
cẹncu
Sinistroverse, Etruscan alphabet. Of the e, only the end of the lowest bar is preserved.
Bibliography: [Della Seta 1918:107 (autopsy)]; CVA Italia 3 pp.23-4; [Beazley 1963:67 ¢5²]; Colonna
1972b:445-6 ¢56² (autopsy); Rix ET Fa 2.4. Photographs: CVA Italia 3 tav.46,1-3; Colonna 1972b
tav.LXXIX. Drawing: Colonna 1972b:445.
XXII-XXIII. Scratched under the foot of an Attic red-figured kylix by the Euaion Painter
(height foot 0.5 cm, foot 11 cm; letters 10 mm high). Mid-fifth century.
umu cacas
Sinistroverse, Etruscan alphabet. Both words are probably names: Colonna rightly
rejected a connection of umu with umom EF 3.
Bibliography: [Della Seta 1918:108 (autopsy)]; [Beazley 1963:792 ¢56²]; [Zanker in Helbig/Speier
1969:679 ¢2759² (autopsy)]; Colonna 1972b:444-5 ¢55² (autopsy); Rix ET Fa 2.11a-b. Photo-
graph: Colonna 1972b tav.LXXIX. Drawing: Colonna 1972b:444.
XXIV. Scratched under the foot of an Attic black-varnished skyphos (height 9.5 cm,
foot 7.5 cm; letters 7-10 mm high) from Nepi, località Fosso del Cardinale. C.475-450.
veka
Sinistroverse, Etruscan alphabet.
Bibliography: Beranger & Fortini 1978 ¢108² (autopsy); Rix ET Fa 0.5. Photograph: Beranger &
Fortini 1978 tav.LXVI. Drawing: Beranger & Fortini 1978:355.
19.3. Civita Castellana (Falerii Veteres)
Judging from the number of inscriptions, Etruscan does not seem to have played an
important role at Civita Castellana itself. According to the most generous count, Civita
Castellana yields only 17 inscriptions that can be regarded as Etruscan, against c.200
Faliscan inscriptions. Of these 17, I regard three as Faliscan ([---]ạltai  MF 107,
[---]nθia MF 138, and titias MLF 205) and four as at possibly Faliscan (eitam EF/Etr 5,
arn MF/Etr 37, ulties MF/Etr 64, aie* MF/Etr 108). Three others (u**al*onu*(*)s
MF/Etr 61, namureska MF/Etr 66, tuconu MF 85) are incomprehensible. Of the seven
that I do consider Etruscan, five (Etr XXV, XXVII-XXVIII, XXX-XXXI) were
599
CHAPTER 19
inscribed by the craftsmen who fashioned the objects and may have been written
elsewhere. This leaves the dedication Etr XXIX and possibly also the Besitzerinschrift
Etr XXVI as the only certainly Etruscan inscription that were certainly written at Civita
Castellana:300 to these might be added EF/Etr 5 and MF/Etr 108, and perhaps also
MF/Etr 61.
XXV. Engraved on a carneole scarab (148u106u96 mm) depicting Hercules attacking

Cycnus with his club, from Civita Castellana. First half of the fifth century.
herkle kukne
Sinistroverse, Etruscan alphabet. Herkle is written upside down.
Bibliography: Babelon 1899:32-3 ¢85²; De Simone 1968:71, 86; Zazoff 1968:40 ¢41²; Rix ET Fa
G.1a-b. Photographs: Babelon 1899 pl.VI,85; Zazoff 1968 Taf. 13,41.
XXVI. Scratched after firing on the inside of a small red-varnished bowl ( rim 16.5
cm, height 4.5 cm) allegedly from the Le Colonnette necropolis. Fifth or fourth century.
cnav**es mi
Sinistroverse, Etruscan alphabet with reversed n and m. The a is ; Gulinelli draws
attention to a similar a in CIE 626 from Chiusi (not in ET?). The damaged part can be
read as “forse una i o ad una doppia i o ad una e in seguito corretta dallo scriba”.
Bibliography: Gulinelli 1995a (autopsy). Photograph: Gulinelli 1995a tav.XL. Drawing: Gulinelli
1995a:319.
XXVII. Engraved on a mirror ( 16.5) from Civita Castellana, depicting Alcestis and
Admetus embracing. Fourth century.
alcestei atmite
Sinistroverse, Etruscan alphabet.
Bibliography: [Gerhard/Klügmann & Körte 1897:217 ¢9² (autopsy)]; Richter 1915:278-80 ¢802²
(autopsy); Buffa NRIE 972; Mansuelli 1947:57; De Simone 1968:14-5,29; Rix ET Fa S.1. Photo-
graph: Richter 1915:279. Drawing: Gerhard/Klügmann & Körte 1897:217.
XXVIII. Engraved in the border of a mirror ( 19 cm), in all probability from (near)
Civita Castellana (Borie 1898:51), depicting Hercules and Minos beside the slain
Minotaur, with Ariadne, Iolaus, and Minerva standing by. Fourth century.
θevrumines hercle ariaθa vile menrva mine
Sinistroverse, Etruscan alphabet. De Simone (1968:74) erroneously read herc[le].
Bibliography: Borie 1898 (autopsy); Körte 1900; Buffa NRIE 1049; Pallottino TLE 755; De Simone
1968:24, 66, 74, 81, 95; Rix ET Fa S.2. Drawing: Borie 1898 pl. I-II; Körte 1900:165.
300
Etr XLVII, published by Rix (ET Fa 6.3) as from Civita Castellana, is originis incertae.
600
XXIX. Cut along two sides of a small peperino base (3.5u3.5u13 cm according to
Herbig) from the area of the Tempio Maggiore at Colle di Vignale, Civita Castellana
(see §14.1.1). Fourth or third century.
anaelauv|cies
Sinistroverse, Etruscan alphabet. Most editors read [l]auvcies with Herbig, but the left
half of the l appears to be visible (lauvcies Rix). The v is , as in Etr XXXIV and XLI.
Bibliography: Herbig 1910:192 ¢32² (autopsy); Herbig CIE 8029; G. Giacomelli 1963:66-7 ¢57²;
Moscati 1983:87; Comella 1986:171 ¢29² (autopsy); Cristofani 1988:13; Rix ET Fa 0.6. Photographs:
Comella 1986 tav.58b-c. Drawings: Thulin in CIE 8029; Herbig CIE 8029; Comella 1986 tav.58b-c.
XXX. Scratched on a ceramic applique (height 18.5 cm, letters 9 mm high) found in
tomb CXXXIII of the Le Colonnette necropolis. Dated to c.325-250 by Michetti.
evrs·ci
Sinistroverse, Etruscan alphabet. The interpretation is unclear: Maras suggests a
possible connection with Greek names in EÙru-.
Bibliography: [FI II.2 p.211 (autopsy)]; Michetti 2003:255-258 ¢658² (autopsy); Michetti & Maras
2003 ¢79² (autopsy). Photographs: Michetti 2003 tav.XII, tav.CXXXVIII; Michetti & Maras 2003
tav.XXXIV. Drawings: Michetti 2003:378 fig.47 ; Michetti & Maras 2003:378.
XXXI (falsum?). Engraved on a mirror ( 10 cm) with handle, depicting Mercury and
Jupiter standing beside a seated Apollo, reputedly found in 1910 in a tomb at Civita
Castellana. Noll regarded the mirror as a bad copy of a silver mirror described by
Ducati (1927:448, with photograph tav.213 fig.524). Late fourth or early third century.
turṃṣ tinia apulu
Sinistroverse, but apulu dextroverse; Etruscan alphabet. Turṃṣ is apparently badly
engraved: Deonna (1915:321 n.3) read  Turuns.
Bibliography: Deonna 1915:321 (autopsy); [Deonna 1919:137]; Noll 1932:159-60; De Simone
1968:21. Photograph: Deonna 1915:322 fig.14 (reproduced in Deonna 1923 pl.43, Noll 1932:160-1
figs.101-102).
19.4. Corchiano and the northern ager Faliscus
Most Etruscan inscriptions from the Middle Faliscan period originate from the north-
western ager Faliscus, especially its centre at Corchiano. Apart from those given here,
possibly Etruscan are hermana MF 265 and the largely illegible MF/Etr 289; sometimes
included, too, but in my view rather Faliscan with Etruscan features (§9.2.2-3), are
arute macena | morenez MF 269 and larise  marc||na  citiai MF 270 from Corchiano,
as well as the purely Faliscan inscription pupiias MLF 304 from Vignanello. Signifi-
cantly, the Etruscan inscriptions from Corchiano include several sepulcral inscriptions
601
CHAPTER 19
(Etr XXXIV-XXXV) and roadside inscriptions (Etr XXXVIII-XXXIX) that are

unlikely to have been written anywhere else. The roadside inscriptions are particularly
interesting, as they show that Etruscan could be used in what were apparently inscrip-
tions on public works (§11.1.4.5). On the whole, Corchiano not only yields a relatively
large number of Etruscan inscriptions, but also, quite frequently, Etruscan features
within Faliscan inscriptions. This material has been the subject of studies by Cristofani
(1988) and Peruzzi (1990): see §9.2.3.
XXXII. Scratched on the outside of a fragmentary bucchero vessel (“a forma di

scodello” Cozza 1886:155; “[tazza] a calice” FI II.2 p.252), from tomb 11 of the first
necropolis of Il Vallone, Corchiano. Late sixth century (Colonna).
larisazuχus
Sinistroverse. The gentilicium zuχus occurs in its Faliscan adaptation in zuconia MF
271, also from Corchiano, and perhaps in zu[con]|eo MF 56 (from Civita Castellana.
Bibliography: Cozza 1886:155 (autopsy); Herbig 1910:185 ¢22² (autopsy); Herbig CIE 8382; G.
Giacomelli 1963:63 ¢45²; Peruzzi 1964c:227; FI II.2 p.252 (autopsy); Cristofani 1988:23 ¢11²; Colonna
1990:120; Peruzzi 1990:278; Rix ET Fa 2.15. Transcription: Cozza 1886:155 (reproduced in CIE
8382).
XXXIII. Engraved on a bronze mirror ( 16.6 cm), depicting Aias and Achilles, from
tomb 22 (Benedetti’s tomb C or 3) of the ‘third necropolis of Il Vallone’ at Corchiano.
From the same tomb are cẹsit  fere MF 264 and hermena MF 265. Fourth century.
aχle aivas
Sinistroverse, Etruscan alphabet. Aχle is written upside down.
Bibliography: [Weege in Helbig 1913 :379 (autopsy)]; Della Seta 1918:84 (autopsy); Mansuelli
1943:501-3 ¢6² (autopsy); Mansuelli 1947:50; De Simone 1968:12,34 (autopsy); FI II.2 p.276 (au-
topsy); Rix ET Fa S.3. Photograph: Mansuelli 1943 tav. XXXV. Drawing: Mansuelli 1943:502 fig.4.
XXXIV. Scratched on a tile from tomb 7 (Benedetti’s tomb IX) of the first necropolis
of S. Antonio at Corchiano. Known only through apographs by Helbig and Manzielli,
and Nogara’s copy of the apograph in the Museo di Villa Giulia inventory. From the
same tomb are poplia  calitenes | aronto  cesies | lartio  uxor MF 265 and ueltur 
tetena | aruto MF 266.
larθ  ceises
velusa
Sinistroverse, Etruscan alphabet. The s’s are reversed (). In Manziell’s apograph the e
has the form , which also occurs in arθ[3-5]rẹ MF 268 from Corchiano: see §11.2.5.
The first letter of the second line is in Manzielli’s apograph and  in Helbig’s and
Nogara’s: editors used to read as celusa (interpreted as a cognomen by Herbig, who
602
compared the Latin cognomen Gelussa), and Buonamici), but velusa, first suggested by
Gamurrini (in Lignana 1887a:201), has become the accepted reading since Rix
(1963:162 n.32). A similar v occurs in Etr XXIX and XLI.
Bibliography: Lignana 1887a:201; Deecke 1888:188-9 ¢58²; Conway 1897:527 ¢29*²; Herbig CIE
8389; Buonamici 1913:74-5 ¢41²; Vetter 1953:319 ¢336²; G. Giacomelli 1963:102 ¢134²; Rix 1963:162
n.32; Peruzzi 1964c:229-30; FI II.2 p.288 (autopsy); Cristofani 1988:23 ¢15²; Peruzzi 1990:280,282;
Rix ET Fa 1.3. Drawing: Deecke jr. (from Manzielli) in Deecke 1888 Taf. III (reproduced in CIE
8389).
XXXV. Scratched on a tile found in situ in tomb 11 of the first necropolis of S.

Antonio, Corchiano. Known only through apographs by Manzielli and the author of the
FI description, and Nogara’s copy of the apograph in the Museo di Villa Giulia
inventory.
larθ
urtcsnas
Sinistroverse, Etruscan alphabet. The θ is . Herbig’s urχosna, after Deecke jr.’s copy
of Manzielli’s apography, has been adopted by most editors (still urχosnas in ET).
Cristofani and Colonna rightly read urtcsnas with FI: cf. Nogara’s urtssnas. For the
name, cf. ortecese MLF 339 (thus Colonna, probably rightly): the name may be derived
from a toponym *Ortica or *Orticum (see §7.8.1.108 and §6.5.11).
Bibliography: Deecke 1888:189 ¢59²; Pauli 1891:105-6; Conway 1897:527 ¢30*²; Herbig CIE 8390;
Buonamici 1913:74 ¢40²; Vetter 1953:323 ¢342a,1²; G. Giacomelli 1963:102 ¢135²; FI II.2 p.290
(autopsy); Cristofani 1988:23 ¢14²; Colonna 1990:135-6; Rix ET Fa 1.4. Drawings: Deecke jr. (from
Manzielli) in Deecke 1888 Taf.III; FI II.2 p.290 (reproduced in Colonna 1990:136 fig.15).
XXXVI. Scratched inside a black-varnished saucer (height 6.5 cm, 18.3 cm; letters
14-20 mm high) from tomb 19 of the first necropolis of S. Antonio. C.350-310.
tafina
Sinistroverse, Etruscan alphabet. The f is , as in Etr XVII. Also scratched inside the
cup are t v, t f, and a cross (FI mentions only the cross, placing it under the foot). Tafina
occurs only here and in Etr XVII: elsewhere, it is θavhna (Cr 2.5), θahvna (Cr 2.6,
2.20, 2.29), θafna (Cr 2.51, Vs 2.7, AV 2.5, Cl 2.26) or θapna (Vc 2.52, Po 2.21, Co
3.1).
Bibliography: [FI II.2 p.295 (autopsy)]; Colonna 1990:118-20 (autopsy). Photograph: Colonna 1990
after p.140 tav.Ia. Drawing: Colonna 1990:119 fig.2.
XXXVII. Scratched after firing on a shard of the bottom of a black-varnished cup

purported to have been found at Corchiano (Fondo Crescenzi). Third or second century.
arnθialuṛ[4-5]
Sinistroverse, Etruscan alphabet. The θ is . Of the gentilicium, only a u and the lower
603
CHAPTER 19
part of an r are preserved, followed by a lacuna that contained 4 or 5 letters. Gulinelli

compares North-Etruscan urinate; I would rather point to larθ | urtcsnas Etr XXXV
from Corchiano. Perhaps this inscription is connected with arθ[3-5]rẹ MF 267, also
from Corchiano, in which case the text could be arnθial uṛ[e +3?] (arnθial uṛ[es mi]?).
Bibliography: Gulinelli 1995a:320-1 ¢4² (autopsy). Photograph: Gulinelli 1995a tav. XL. Drawing:
Gulinelli 1995a: 320.
XXXVIII. Cut, c.4 m above the ground, c.24 m past the church of S. Egidio, on the
left-hand side of the Via di S. Egidio, a cutting leading down to the ancient crossing of
the Rio Fratta near Corchiano (cf. Ward Perkins & Frederiksen 1957:116-7 with fig.17,
Quilici 1990:208-19 with figs.3-4). Dini et al. give the length as 1.80 m, but as they
describe only the last five letters, it may be up to twice as long. Fourth or third century.
[u]elturθanae
Sinistroverse, Etruscan alphabet. Cristofani’s and Rix’s [---]θ anae is based on the
incomplete edition by Dini et al.; Quilici reads *****turθanae, and his photograph
appears to show also the l and the e. A gentilicium Θanae is not attested elsewhere.
Bibliography: Dini et al. 1985:69; Cristofani 1988:23 ¢16²; [Colonna 1990:120 n.40]; Quilici 1990:212
(autopsy); Rix ET Fa 0.8. Photograph: Quilici 1990 tav.Vb. Photograph: Dini et al. 1985:69 fig.48.
XXXIX. Cut, about halfway up the road (letters 33-36 cm high according to Dennis and
Ward Perkins & Frederiksen, but 42 cm high according to Buglione di Monale), on the
right-hand side of the Via della Cannara, a cutting leading up from the ancient crossing
of the Rio Fratta near Corchiano (see Ward Perkins & Frederiksen 1957:116-7 with
map fig.17, Quilici 1990:208-19 with map fig.4). Fourth or third century.
larθvelarnies
Sinistroverse, Etruscan alphabet. This inscription is assumed to have been the model for
the falsum CIE 8380, larθ velarnies (with an r of the shape , apparently based on the
Faliscan ?), on a kantharos reputedly from Civita Castellana. Colonna (1997) has now
published an inscription of unknown origin that appears to be contemporary with Etr
XXXIX and also reads larθ velarnies, but has the normal Etruscan .
Bibliography: Dennis 1848:155-6 (autopsy); Dennis 1878:119; Buglione di Monale 1887a:27 (au-
topsy); Buglione di Monale 1887b; Deecke 1887; [Deecke 1888:156]; Herbig CIE 8379 + add.;
Buonamici 1913:73 ¢39²; Ward Perkins & Frederiksen 1957:116-7 (autopsy); G. Giacomelli 1963:72
¢64²; Peruzzi 1964c:228; Moscati 1980:93; FI II.2 p.217 n.32 (autopsy); Dini e.a. 1985:69-71; Cristo-
fani 1988:23 ¢17²; Quilici 1990:208-9,217-9 (autopsy); Rix ET Fa 0.7. Photographs: Ward Perkins &
Frederiksen 1957 pl. XXXIb; Moscati 1980 p.112 fig.78; Drawings: Dennis 1848:156 (reproduced in
Dennis 1878:119); Nogara in CIE 8379 add.
XL. Scratched under the foot of a yellow saucer (height 6 cm, 16.5. cm) from tomb 2
at Contrada Lista, c.2 km north of Corchiano. Fourth century. From the same tomb is la
MLF/Etr 286.
604
mialsi*ismi
Sinistroverse, Etruscan alphabet. The unidentified letter is , probably a t, perhaps a p
or an n. The name alsiṭis could be toponymic and connected to Alsium.
Bibliography: Bendinelli 1920:30 (autopsy); Diehl 1930:78 ¢737²; Buonamici 1932:83; Buffa NRIE
980; Vetter 1953:326-7 ¢355²; G. Giacomelli 1963:63 ¢44,I²; Pallottino TLE 31; Agostiniani 1982:69
¢91²; Cristofani 1988:23 ¢13²; Rix ET Fa 2.16. Drawing: Bendinelli 1920:30.
XLI. Engraved on a badly corroded bronze mirror ( 17 cm), depicting Aphrodite and
a bearded man embracing, with Hephaestus and a naked youth standing by, found
probably in tomb 6 of a group of tombs to the southeast of Corchiano, explored in 1893
by Benedetti (cf. FI II.2 pp.313, 318). Fourth or third century.
uslẹ*es turan acaviser setlans
Sinistroverse, Etruscan alphabet. The name of the youth is almost illegible in Man-
suelli’s drawing: he compared him to the assistant of Hephaestus on the mirror in
Gerhard/Klugmann & Körte (1897, Taf.49), who is there called tretu (Ta S.8). Ambros-
ini suggests reading usleṭes or usleθ̣es: her drawing also shows that acaviser (with v of
the shape , as in Etr XXIX and XXXIV), not aχviser (ET), is the correct reading
Bibliography: Herbig CIE sub 8412 (autopsy); Buffa NRIE 971; Mansuelli 1943:504-6 ¢7² (autopsy,
erroneously describing it as unpublished); FI II.2 p.318; Rix ET Fa S.4; Ambrosini 1995 (autopsy);
Maras 2002 (autopsy). Drawings: Mansuelli 1943:505 fig.5 (reproduced in Ambrosini 1995:182 fig.1);
Ambrosini 1995:183 fig.2; Maras 2002:468.
XLII. Scratched in large letters on the bottom of a red-varnished cup from a well in the
habitation at Vignanello (Giglioli’s pozzo 1). Fifth to third century.
vultasi
Sinistroverse, Etruscan alphabet. Vulta-, occurring only here, is the Etruscan rendering
of the Faliscan name Volta (§7.7.1.84). The form vultasi may provide an argument in
favour of interpreting the Faliscan forms in -ai and -oi as datives: see §8.7.1.
Bibliography: Giglioli 1924:249 (autopsy); Cristofani 1988:24 ¢19²; Rix ET Fa 3.4. Drawing: Giglioli
1924:249.
19.5. The southeastern ager Faliscus and the ager Capenas
The Etruscan inscriptions from the south-eastern ager Faliscus and the ager Capenas are
very few in number. To those published here, some editors add tulate tulas urate EF/Etr
385 from Fiano Romano, which may be Etruscan, and larise uicina MLF 371 and
larise | uicina MLF 372 from Rignano Flaminio, [fel]ịcinatiu LF 384 from Civitella S.
Paolo, all of which I regard rather as Faliscan, as well as apa Cap 459, reputedly from
the ager Capenas, which may equally well be Faliscan or Latin. The only certainly
Etruscan inscription from the ager Capenas dating from after its colonization early in
605
CHAPTER 19
the fourth century (cf. §2.5.2) is Etr XLV, which appears to show Sabellic influences.
XLIII. Painted in red on a fragmentary tile (letters 12-17 cm high) from tomb 1 of the
group of tombs excavated in 1912 on the south-east side of Monte Casale, near Rignano
Flaminio. Fourth or third century. From the same tomb are [iu]na  upreciano MLF
363, [---] upreciano MLF 364, [---]are*[---] MLF 365, and [---]uinu[---] MLF 366.
umrie
Sinistroverse, Faliscan alphabet. I regard this inscription as Etruscan because of the
non-Faliscan mr: forms in -ies also occur as in otherwise Faliscan inscriptions
(§9.2.2.2).
Bibliography: Herbig CIE 8432; Gabrici 1912b:75-6 (autopsy); Buonamici 1913:80-1 ¢52²; [Della Seta
1918:104 (autopsy)]; Vetter 1953:324 ¢343c²; G. Giacomelli 1963:107-8 ¢145,III² (autopsy); Rix ET
Fa 1.7. Drawings: Nogara in CIE 8432; Gabrici 1912b:76 fig.3.
XLIV. Scratched on the body of a small amphora (height 35 cm, body 7.5 cm; letters
9-25 mm high) found in 1907 at Monte Laceto, near modern Capena. Seventh century.
abc ̣ḍevzhθ̣iḳsi*p ̣*q ̣χfu
Dextroverse, with reversed s. The u is placed under the end of the line. Several of the
letters have unusual forms: see fig.19.4. Pandolfini regards the alphabet not as Etruscan,
but rather as comparable to that of several other early inscriptions such as Sab 480† and
the inscription of Poggio Sommavilla. She notes that the shape of the f and of the u also
occur in the Early Faliscan, but not in the contemporary Etruscan inscriptions. Although
this is undoubtedly correct, the alphabet is certainly not Faliscan because of the
presence of b, v, θ, and χ; also, the shape of the d and of the h are certainly not those of
the Faliscan alphabet. The alphabet may be representative of the time when the
alphabets of the Lower Tiber basin were still developing: see §11.2.2.
Bibliography: Herbig CIE 8547 (autopsy); †Paribeni 1913:69-70; Neppi Mòdona 1926:503; Buonamici
1932:112-5; Vetter 1953:329; Briquel 1972:815-6; Pandolfini & Prosdocimi 1990:90-4 ¢App.1²; Rix
ET Fa 9.2. Photographs: †Paribeni 1913:70 figs.1-2; Neppi Mòdona 1926:503 fig.7; Buonamici 1932
tav.V fig.8; CVA Roma (Museo Preistorico L. Pigorini) fasc.1 tav.X; Pandolfini & Prosdocimi 1990
tav.XLVI. Drawings: Herbig CIE 8547; Pandolfini & Prosdocimi 1990:91.
XLV. Scratched inside a small black-varnished cup from the votive depot discovered in
1952 at the temple of Lucus Feroniae. Third century.
caviesuhtav[---]
Uhtav[---] (not Moretti’s VHTAṾ[ies]) shows a Sabellic ht m */kt/: Colonna pointed to
the occurrence of the name at Perugia (uhtave Pe 1.638, 1.639, 1.891, uhtaves Pe 1.817,
1.1267, uhtavial Pe 1.115; uθavi Pe 1.756, uθavis 1.1264).
Bibliography: Moretti 1975:145-7 ¢142²; Colonna 1976c; Cristofani 1988:13; Rix ET Fa 2.25. Photo-
graph: Moretti 1975 tav. 36.
606
19.6. Etruscan inscriptions incertae uel ignotae originis
The attribution to the ager Faliscus of the following inscriptions is probable only in the
case of Etr L-LI, part of the painted decoration of a vase of Faliscan workmanship, and
possible in the case of Etr XLVIII. The provenance of XLVII is uncertain. There
appears to be no reason why Etr XLVI and XLIX should be ascribed to the ager
Faliscus.
XLVI. Scratched on the shoulder of a bucchero aryballos (height 5.8 cm, 4.5 cm
according to Buonamici, but height 5 cm, 4 cm according to Buffa in Buonamici
1935:356, 1938:319). Third quarter of the seventh century.
mlakasse·laaskamieleivana
Dextroverse, Etruscan alphabet. The s is  in mlakas and aska, but in se·la, a variation
occurring also in EF 1 and Etr IV (cf. §11.2.2).
Bibliography: Fabretti CII 2614 quater (autopsy); Bugge/Torp 1909:134; Cortsen 1935:152; Buonami-
ci 1935:356-7; Buonamici 1938; Poupé 1963:245-7 ¢32²; Pallottino TLE 762; De Simone 1968:27;
Maggiani 1972; Agostiniani 1982:140 ¢32²; Rix ET Fa 2.3. Photographs: Buonamici 1938 tav.
LXI,1-4; Poupé 1963 pl.XXV,3; Drawing: Buffa in Buonamici 1935:356.301
XLVII. Scratched on a bucchero aryballos (height 6.5 cm). The provenance is unclear:
“genannt werden die Umgebung (wohl im weiteren Sinn) von Civita Castellana und
von Bagnorea; nach einer Notiz von Nogara stammt es aus Grotte S. Stefano (Station an
der Eisenbahnlinie Viterbo-Attigliano, 10 km von der Station Montefiascone)” (Herbig
1913a:165). Maggiani (1980:404) quoted a letter of 21.X.1908 in the archives of the
Soprintendenza Archeologica per la Toscana, where the provenance is given as
“Fidene, sul territorio di Fiano romano o meglio Leprignano”. C.630-620.
miaraθialeziχuχe
Dextroverse, Etruscan alphabet. The θ is .
Bibliography: Nogara 1909:196; Herbig 1913a:165-8 (autopsy); Buffa NRIE 734; Poupé 1963:244-5
¢31²; Pallottino TLE 278; De Simone 1970b:124; [Bonamici 1974:39 ¢45²]; Colonna 1975a:181-4;
Maggiani 1980; Agostiniani 1982:106 ¢328²; Rix ET Fa 6.3. Photographs: Herbig 1913a Taf.I,1-4;
Poupé 1963 pl.XXV,2; Bonamici 1974 tav.XXIIa-c.
XLVIII. Scratched on the handle belonging to a bucchero kantharos or kyathos, now in

a private collection in Rome, reputedly found near Mazzano Romano. C.570-560?
leθaiemulvanicemi·nehvulve·s·
Sinistroverse, Etrucan alphabet. Naso reads hvulve·s, but his drawing clearly shows a
301
Buonamici 1938:318 fig.4 only reproduces Fabretti’s transcription.
607
CHAPTER 19
second point following this letter. For hvulve·s· (Latin Fulvus), cf. hvuluves Ve 3.9 and
hvuluves̽ Ve 3.30, and also vhulvenas Vs 1.100. Leθaie is probably the same name as
the Faliscan genitive letei in MF 470*: on this name, see also Vetter 1948:67-8.
Bibliography: Naso 1993. Photograph: Naso 1993 tav.XLVI. Drawing: Naso 1993:263.
XLIX. Scratched on a loomweight. Known only through apographies by Pfister (see

Garrucci 1860:243, Buonamici 1940:397-8). The only reason for including the inscrip-
tion among those of Faliscan origin (as do G. Giacomelli and Rix in ET) is that Garrucci
quoted it as a parallel for the reversed s’s in Faliscan inscriptions. Sixth century (?).
veleliasmistaslarv
Sinistroverse, Etruscan alphabet. The s’s are reversed (). Most editors read velenas,
with  read as n, but as the m is , the use of  for n is unexpected, and Rix is probably
right in reading li instead. The v at the end may have been a monogram.
Bibliography: Garrucci 1860:243; Buonamici 1940:397-9; G. Giacomelli 1963:271 ¢LIII²; Rix
1963:39, 257; Pallottino TLE 33; Agostiniani 1982:70 ¢93²; Rix ET Fa 2.5. Drawing: Conestabile (?) in
Buonamici 1940:398 fig.1.
L-LI. Painted in white on a Faliscan stamnos (height 31.6 cm, rim 20 cm), L on the
front near a running Perseus, LI on the back between two Gorgons. Mid-fourth century.
φerse
c[e]ṛụr  purφiunạs
Sinistroverse, Etruscan alphabet. LI is badly damaged. Pfiffig read c*ṛ[ụ]ṛ
purφịụn[ạ]s ‘cerur [‘Werke, Gebilde’] des Purphiuna’ (cẹṛ[u]ṛ: purφ--n-s Rix). The
use of cerur in this formula is apparently unique: the usual word is acil.
Bibliography: WeltGr p.50 ¢3² (autopsy); [Brommer 1956:159 ¢A.3²302] ; Schauenburg
1960:49-50; CVA Heidelberg 2 pp.29-30 (autopsy); [Brommer 1980:282]; Pfiffig 1965; Rix ET
Fa 7.1a-b. Photographs: Schauenburg 1960 Taf. 22,1-2; CVA Heidelberg 2 Taf. 67,2-3. Drawing: CVA
Heidelberg 2 p.29 (reproduced in Pfiffig 1965: 102 fig.54a, 103 fig.52b).
302
From the data given by Brommer I am not certain that this is indeed the same vase.
608
Dutch summary
Nederlandse samenvatting
Het onderhavige proefschrift bestaat uit een editie (deel 2) en een taalkundige analyse
(deel 1) van het epigrafisch materiaal daterend van vóór de eerste eeuw voor de
christelijke jaartelling uit het gebied dat in de oudheid bekend stond als de ager
Faliscus, rond de stad Falerii, het tegenwoordige Civita Castellana (ong. 50 km ten
noorden van Rome). Centraal hierbij staat de vraag of de taal die in deze inscripties
wordt aangetroffen en wordt aangeduid als ‘Faliscisch’, ook daadwerkelijk als een
aparte taal binnen de Italische tak van de Indo-Europese taalfamilie dient te worden
beschouwd, of als een dialect van het Latijn.
Voor het volgende is het nodig om hier de periodisering van de Faliscische inscripties te
noemen: Vroegfaliscisch (Early Faliscan of EF), d.w.z. alle inscripties in Faliscisch
alfabet van voor de vierde eeuw voor de christelijke jaartelling; Middenfaliscisch
(Middle Faliscan of MF), d.w.z. alle inscripties in Faliscisch alfabet die toegeschreven
kunnen worden aan de periode tussen het begin van de vierde eeuw en de oorlog van
241-240 voor het begin van de christelijke jaartelling; Laatfaliscisch (Late Faliscan of
LF), d.w.z. alle inscripties in Faliscisch alfabet die toegeschreven kunnen worden aan de
periode ná de oorlog van 241-240 voor het begin van de christelijke jaartelling; Middel-
of Laatfaliscisch (Middle or Late Faliscan of MLF), d.w.z. alle inscripties in Faliscisch
alfabet waarvan niet duidelijk is of ze Middelfaliscisch dan wel Laatfaliscisch zijn;
Latinofaliscisch (Latino-Faliscan of LtF) en Capenatisch (Capenate of Cap), d.w.z. alle
inscripties in Latijns alfabet die wél taalkundige kenmerken tonen die in
overeenstemming zijn met die van de Faliscische inscripties; Latijns (Latin of Lat),
d.w.z. alle inscripties in Latijns alfabet die geen taalkundige kenmerken tonen die in
overeenstemming zijn met die van de Faliscische inscripties; en tenslotte Etruscisch
(Etruscan of Etr), d.w.z. alle inscripties die duidelijk in het Etruscisch geschreven zijn.
In hoofdstuk 1 (Introduction, pp.1-18) worden allereerst (̣1.1, pp.1-3) drie algemene

redenen gegeven die het Faliscisch tot een interessant onderwerp van taalkundige studie
maken : (1) het is met het Romeins en het Praenestinisch het best-geattesteerde Latijnse
dialect; (2) de epigrafische documenten voor het Faliscisch zijn relatief oud; (3) het
gebied waar het Faliscisch gesproken werd lag tussen gebieden waar respectievelijk
Etruscisch, Latijn, en Sabellische talen werden gesproken, en is als zodanig een
interessant onderwerp van studie vanuit het oogpunt van taalcontactstudies. Aan het
Faliscisch zijn daarom in de loop van de laatste 120 jaar meerdere grotere publicaties
gewijd. Een overkoepelende studie van het Faliscisch waarin betoogd wordt dat het
Faliscisch een dialect van het Latijn is, ontbrak echter tot op heden.
609
DUTCH SUMMARY – NEDERLANDSE SAMENVATTING
Vervolgens (§1.2) worden enkele methodologische punten besproken met betrekking tot
de definitie van het begrip ‘dialect’ zoals die gehanteerd kan worden bij het onderzoek
naar fragmentarisch bewaarde dode talen zoals het Faliscisch. Allereerst wordt deze
definitie in tweeën gesplitst in (1) de strict sociolinguistische definitie, waarbij vooral
naar buitentalige factoren wordt gekeken, en het criterium voor het taal/dialect-
onderscheid vooral benaderd wordt vanuit de beleving van de sprekers zelf en van
groepen waarmee zij regelmatig in contact staan, en (2) de structurele definitie, waarbij
vooral naar binnentalige factoren wordt gekeken en het voornaamste criterium de mate
van verschil in de taalstructuur zelf is. De sociolinguistische benadering is ipso facto
niet toepasbaar op talen waarvoor de buitentalige gegevens zo schaars zijn als voor het
Faliscisch.
Met betrekking tot de structurele benadering wordt opgemerkt dat deze op haar
beurt opgedeeld kan worden in een strict synchronische benadering, gebaseerd op
synchronische geolinguistische vergelijking (traditioneel resulterend in een
isoglossenkaart), en een diachronische benadering gebaseerd op een diachronisch
‘stamboommodel’ van de betrokken talen. Bij deze laatste benadering worden vooral
gevallen van gemeenschappelijk of afzonderlijk behoud van aspecten van de taal, en
van gemeenschappelijke of afzonderlijke vernieuwing van aspecten van de taal tegen
elkaar afgewogen. Betoogd wordt dat bij een fragmentarisch bewaarde dode taal een
combinatie van de synchronische en de diachronische benadering het beste resultaat
oplevert, onder het voorbehoud dat de synchronische benadering altijd een vergelijking
tussen meer dan twee talen of dialecten dient te omvatten, en dat binnen de
diachronische methode de verschillende gevallen van behoud of vernieuwing ten
opzichte van elkaar ‘gewogen’ dienen te moeten worden.
Verdere methodologische punten (§1.3) hebben betrekking op taalcontactsituaties in het
oude Italië. Gesteld wordt dat in de studie van de talen van het oude Italië het
verschijnsel taalcontact nog te veel ad hoc gebruikt worden om hinderlijke
anomaliteiten te verklaren, en te weinig als zelfstandig punt van taalkundige studie. Het
bestaan van een Italische Sprachbund zoals voorgesteld door bv. Pisani en van een
Faliscisch-Latijnse diglossie zoals voorgesteld door R. Giacomelli wordt sterk in twijfel
getrokken, en in deze studie verder niet gebruikt. In dit kader worden tevens de
begrippen ‘interferentie’ (interference) en ontlening (borrowing) besproken in de
context van fragmentarisch bewaarde talen, en drie factoren die van invloed zijn op
ontlening: (1) de structurele verschillen tussen de betrokken talen, (2) de
ontleningshiërarchie (borrowing hierarchy), en (3) buitentalige (sociolinguistische)
factoren. Voorgesteld wordt hierbij onder andere dat de ontleningshiërarchie een
zodanig sterk gegeven is dat deze in de studie van fragmentarische talen ook in
voorspellende zin kan worden gebruikt, zodat aanwijsbare gevallen van ontlening op de
dieper gelegen niveau’s van de taal als indicatie kunnen dienen van ontlening op de
minder diep gelegen niveau’s zelfs als deze niet in het materiaal geattesteerd is.
610
Vervolgens (§1.4) worden een aantal punten besproken met betrekking tot het
Faliscische epigrafische materiaal. Dit materiaal bestaat uit 535 inscripties, waarvan
ong. 355 bruikbare taalkundige gegevens opleveren.. Het merendeel bestaat uit
grafinscripties; de overige inscripties staan vooral op aardewerk. (Het materiaal wordt
uitgebreider besproken in hoofdstuk 11, zie hieronder.) De datering van het materiaal is
bijzonder moeilijk: de grafcultuur van het gebied, met kamertombes die soms eeuwen
achtereen in gebruik waren en in later tijden dikwijls gebruikt werden als stal of schuur,
zorgt ervoor dat veel inscripties niet aan dateerbare grafgiften te koppelen zijn. Dit
maakt ook een datering op epigrafische of orthografische gronden moeilijk. Als
oplossing voor dit probleem wordt in deze studie een verdeling gebruikt die gebaseerd
is op periodisering naar vindplaats en op het gebruikte alfabet (zie hierboven). Voorts
wordt betoogd dat het indelen van inscripties op taal niet mag worden uitgaan van het
gebruikte alfabet, aangezien dit geen taalkundig argument is. Tenslotte wordt een kort
woord gewijd aan de betrouwbaarheid van het materiaal. Slechts enkele inscripties
(inscripties 335, 464, XXXI, en een inscriptie besproken onder XXXIX) zijn als
falsificaties te beschouwen. Wel kan de archeologische context betwijfeld worden: dit
hangt samen met het zgn. ‘Villa Giulia-schandaal’ uit de jaren 1898-1907. Dit lijkt
echter niet van doorslaggevende invloed te zijn op het onderwerp van deze studie.
Hoofdstuk 1 sluit af met een kort overzicht van de geschiedenis van de studie van het
Faliscisch (§1.5). Aanvankelijk was niet meer bekend dan de opmerking van Strabo
(Geographica 5.2.9): œnioi d'oÙ TurrhnoÚj fasi toÝj Falšriouj, ¢ll¦ Fal…skouj, ‡dion
œqnoj: tinèj dè kaˆ toÝj Fal…skouj pÒlin „diÒglwsson (‘Sommigen zeggen dat de inwoners
van Falerii geen Etrusken zijn, maar Falisken, een eigen volk: sommigen ook dat de
Falisken een stadstaat zijn met een eigen taal’). Hoewel verschillende inscripties al
vroeg werden opgetekend (inscripties 205-210, waarvan een afschrift bekend is uit
1676) of zelfs gepubliceerd (inscriptie 79, gepubliceerd in 1726), begint de
daadwerkelijke studie van het Faliscisch in 1860 met de publicatie van een aantal
grafinscripties uit de omgeving van Falerii Novi (nrs. 220-233).
In de daaropvolgende 150 jaar kunnen de volgende richtingen in het denken over
de taalkundige positie van het Faliscisch onderscheiden worden: (1) het Faliscisch was
taalkundig onafhankelijk van het Latijn en sterk beinvloed door de Sabellische talen
(Deecke, Die Falisker, 1888); (2) het Faliscisch stond taalkundig dichterbij het Latijn
maar was sterk beinvloed door het Etruscisch (Herbig, Corpus inscriptionum
Etruscarum II.2.1, 1912); (3) het Faliscisch stond taalkundig dichtbij het Latijn, maar
was een aparte taal en geen dialect (G. Giacomelli, La lingua falisca, 1963). In
recentere werken wordt het Faliscisch in toenemende mate als taalkundig zeer dichtbij
het Latijn staand beschouwd, hoewel de meeste wetenschappers de term ‘dialect’ niet
willen gebruiken: de belangrijkste publicatie binnen deze richting is het artikel van
Joseph & Wallace, ‘Is Faliscan a local Latin patois?’ (Diachronica 8 (1991), pp.159-
186). Deze recente visies worden besproken in hoofdstuk 10 (zie hieronder).
611
Hoofdstuk 2 (The ager Faliscus and its inhabitants, pp.19-52) bestaat uit een
bespreking van het historiografisch en archeologisch bronnenmateriaal dat betrekking
heeft op de ager Faliscus en de bewoners ervan.
Allereerst (§2.1) wordt de fysieke omvang van het gebied vastgesteld. Aan de
oostkant wordt het gebied begrenst door de Tiber, aan de noorden noordwestkant door
de Monti Sabatini, en aan de zuidwestkant door de Monti Cimini. Aan de zuidkant lijkt
de belangrijkste natuurlijke grens de bergrug die de Monti Cimini verbindt met Monte
Soratte aan de zuidoostkant van het gebied. Deze rug zou mogelijk de oude grens van
het gebied kunnen zijn geweest. In de loop van de vijfde eeuw echter breidde de macht
van de Etruscische stad Veii zich echter uit ten noorden van deze rug en lag de grens
van de ager Faliscus meer naar het noorden. De steden in de ager Faliscus die genoemd
worden door auteurs uit de oudheid zijn de hoofdstad Falerii, gelegen op de plaats van
het moderne Civita Castellana, maar tegen het einde van de tweede eeuw voor de
christelijke jaartelling vervangen door een Romeinse stad met dezelfde naam die ong. 5
km meer naar het westen lag, bij het huidige S. Maria di Falleri, en Fescennium, dat
vermoedelijk geidentificeerd dient te worden met het huidige Narce.
Belangrijk voor de studie waren de routes door het gebied. Vanwege de vele steile
ravijnen waren deze beperkt en volgden een duidelijk patroon, waarbij vele van de
wegen samenkwamen bij Falerii. Directe routes naar het noorden waren er niet of
nauwelijks: het gebied was open naar het zuiden, richting de Etruscische stad Veii en de
Latijnse stad Rome, en had slechts één route naar het westen, tussen de Monti Sabatini
en de Monti Cimini door, en één route naar het oosten, die de Tiber overstak bij het
tegenwoordige Gallese en vandaar verder landinwaarts liep naar de gebieden van de
Sabijnen en de Umbriërs. Deze ligging was van zowel economisch belang (gezien de
handelsroutes langs de Tiber en naar het binnenland) en van strategisch belang: na de
verovering van Veii door Rome aan het begin van de vierde eeuw voor de jaartelling lag
het gebied op cruciale niet door Rome gecontroleerde routes door en uit Zuid-Etrurië.
In de hierboven aangehaalde passage uit Strabo wordt gesteld dat de Falisken een ‘eigen
volk’ en een ‘stadstaat met een eigen taal’ vormden. Gezien de context dient deze
uitspraak relatief te worden geïnterpreteerd als ‘anders dan de rest van Etrurië’. In het
kader van deze studie worden de Falisken gezien als een ethnos (§2.7), hetgeen
gedefiniëerd wordt als ‘een groep die historische banden heeft met het gebied dat zij
bewoont, een min of meer identificeerbare eigen taal en cultuur heeft, en hun eenheid
en verschil van andere groepen erkent door een bewustzijn van hun eigenheid dat tot
uitdrukking komt in een volksnaam’. Vervolgens wordt kort geschetst hoe de Falisken
aan deze definitie voldoen. Het belang van een dergelijke definitie voor deze studie ligt
vooral in het feit dat ethniciteit niet een absoluut, maar een relatief gegeven is, en dat
hoe deze ethniciteit gedefiniëerd werd kon verschillen afhankelijk van de periode en de
andere groepen waarmee de Falisken in contact stonden. Een korte bespreking volgt
van de moeilijkheden bij de implementatie van een dergelijke definitie van identiteit.
612
Vervolgens (§2.3) wordt geprobeerd een aantal van de factoren in te vullen die
relevant kunnen zijn geweest voor een Faliscische identiteit. Onder de archeologisch
definiëerbare factoren wordt allereerst genoemd de kenmerkende grafcultuur van in de
rotsen uitgehouwen kamertombes die dienden als soms eeuwen achtereen in gebruik
zijnde familietombes. Voorts worden aspecten van de structurering van de maatschappij,
de inrichting van het bestuurssysteem, en de religie besproken: in veel gevallen blijkt
het moeilijk om concrete uitspraken te doen, hoewel de historische en epigrafische
bronnen zeker gegevens geeven op deze punten.
Hierop volgt een vrij uitgebreide bespreking van de geschiedenis van de ager Faliscus
(§2.4-6). Gesuggereerd wordt dat in de zesde en vijfde eeuw voor de christelijke
jaartelling de ager Faliscus onder een zekere druk moet hebben gestaan van de
opkomende Etruscische stad Veii. De grootste Faliscische nederzetting van die tijd,
Narce, lijkt geheel en al Etruscisch te zijn geworden, zonder dat dit echter de eigen
identiteit van de ager Faliscus bedreigde. Voor de volgende eeuwen wordt grotendeels
de beschrijvingen van de Romeinse en Griekse geschiedschrijvers gevolgd, m.n. Livius.
Falerii en de ager Faliscus komen hierin in beeld vanaf het einde van de vijfde eeuw
voor de christelijke jaartelling, wanneer Rome in oorlog raakt met Veii, waarvan Falerii
dan een bondgenoot is. Na de val van Veii wordt ook Falerii door de Romeinen
bedwongen (in 392 voor de christelijke jaartelling?): vanaf die tijd blijft het gebied een
bondgenoot van de Romeinen die zich echter bij meerdere gelegenheden samen met
andere steden van Zuid-Etrurië, vooral Tarquinii, tegen Rome keert. Desondanks wordt
Falerii door de Romeinen keer op keer opmerkelijk coulant behandeld.
Hierin komt verandering in 241-240 voor de christelijke jaartelling, als de
Romeinen aan het einde van de Eerste Punische Oorlog de ager Faliscus nogmaals
aanvallen. Na deze Romeinse overwinning wordt de ager Faliscus in twee helften
verdeeld, waarvan er een direct door de Romeinen bestuurd wordt, vanuit het kort
daarop gebouwde nieuwe Romeinse Falerii. Gekoppeld aan het vrijwel geheel
verdwijnen van alle grote Faliscische nederzettingen (Falerii en Corchiano), de
versnippering van het gebied, en de toestroom van Latijnstalige immigranten, moet dit
grote gevolgen hebben gehad voor de samenstelling van de bevolking. Vanaf deze tijd
loopt het aantal Faliscische inscripties terug: de laatste (inscriptie 214) dateert
vermoedelijk van ong. 150 voor de christelijke jaartelling.
Aansluitend (§2.7) worden een aantal sociolinguistische factoren besproken die van
invloed zijn geweest, en wordt geprobeerd een inschatting te maken van de mate van
invloed die elk van deze factoren gehad heeft op het voortbestaan van het Faliscisch.
Hierbij wordt gekekeken naar de economische status van het gebied, de status van de
eigen taal, functionele distributie van de verschillende talen, de bevolkingsgrootte (die
geschat wordt op maximaal 20.000-22.500), de verdeling van de bevolking over het
gebied, en tenslotte de rol van de familie en huwelijksbanden buiten de eigen groep.
613
Tenslotte (§2.8) wordt een korte samenvatting gegeven waarin betoogd wordt dat
het voortbestaan van het Faliscisch in de tijd van de uitbreiding van de Etruscische
cultuur wellicht toe te schrijven was aan de gunstige ligging van het gebied dat
enerzijds tamelijk ontoegankelijk was vanuit het westen en het noorden, maar
anderzijds gelegen was op een cruciaal kruispunt van handelsroutes, en dat het
verdwijnen van het Faliscisch na de Romeinse verovering in 241-240 voor de
christelijke jaartelling welhaast onvermijdelijk was gezien de ingrijpende sociale
veranderingen die deze verovering met zich meebracht.
Hoofdstukken 3-8 zijn monografisch vormgegeven hoofdstukken over het Faliscisch

beschouwd vanuit diverse deelgebieden van de taalkunde: de fonologie, de morfologie
van naamwoorden, voornaamwoorden, en werkwoorden, het lexicon, het onomasticon,
en de syntaxis. Elk van deze hoofdstukken bestaat uit een bespreking van voor dat
deelgebied relevante methodologische kwesties en problemen, een analyse van het
Faliscische materiaal, en een deelconclusie.
Hoofdstuk 3 (Phonology, pp.53-116) begint met enkele methodologische punten

(§3.1), waarin met name gesproken wordt over de aannames op het gebied van de
reconstructive van het Proto-Indo-Europees en van de stamboom van de Italische talen.
In deze studie wordt uitgegaan van een tamelijk strict divergerend stamboom-model,
waarbij de eerste fase van de ontwikkeling van de Italische talen gezien wordt al seen
Proto-Italisch stadium waaruit zich vervolgens een Proto-Latijn en een Proto-Sabellisch
ontwikkelen. Een belangrijk punt voor de vraagstelling van deze studie is waar het
Faliscisch zich op deze stamboom bevindt: ontwikkelde het zich als een zelfstandige
taal naast het Proto-Latijn en het Proto-Sabellische, ontwikkelde het zich als een
zelfstandige taal uit het Proto-Latijn, of kan het als dialect van de Latijnse groep
beschouwd worden? De rol van convergentie wordt in de discussies in dit hoofdstuk
zoveel mogelijk geminimaliseerd.
Vervolgens (§3.2) worden de ontwikkelingen in de Proto-Italische fase besproken, zoals
bv. de ontwikkeling van de laryngalen, het samenvallen van */e/ met */o/, en de
vroegste ontwikkelingen van de Proto-Indo-Europese stemhebbende geaspireerde
occlusieven, die zich gedurende deze periode ontwikkelen tot (stemhebbende of
stemloze) spiranten. De aanname is dat Faliscisch als behorende tot de Italische
taalfamilie op dit punt in geen opzicht afwijkt van de andere Italische talen, en deze
aanname blijkt gerechtvaardigd.
Hierna worden de ontwikkelingen binnen het Proto-Latijn besproken (§3.3),
waarbij de aanname is dat het Faliscisch waar er verschil is tussen de ontwikkelingen
van het Proto-Latijn en het Proto-Sabellisch zich aansluit bij het Proto-Latijn. Hoewel
deze aanname in veel gevallen gerechtvaardigd blijkt, is dit niet het geval waar het de
verdere ontwikkelingen van de stemhebbende geaspireerde occlusieven betreft. Hier
614
toont het Faliscisch een ontwikkeling die eerder identiek lijkt te zijn aan die in de
Sabellische talen, waarbij de Proto-Indo-Europese fonemen */bh dh/ zich in word-
interne positie ontwikkelden tot een foneem dat weergegeven werd als f (vermoedelijk
[ ] of [β]), terwijl in het Latijn deze fonemen in dezelfde positie uiteindelijk
samenvielen met /b d/. Dit moet als een van de belangrijkste verschillen tussen het
Faliscisch en het Latijn worden beschouwd, is het enige duidelijk aantoonbare verschil
tussen beiden in de diachronie van de fonologie, en leidt tevens tot een belangrijk
verschilpunt in de synchronische vergelijking van de fonemische en fonotactische
systemen van beiden. Hoewel dikwijls wordt aangenomen dat eenzelfde ontwikkeling
zich voordeed in andere dialecten van het Latijn, zoals het Praenestinisch, blijkt
daarvoor nauwelijks een bewijs te bestaan: de zeldzame gevallen van een word-interne f
in (dialect-)Latijn kunnen net zo goed worden toegeschreven aan interferentie of
ontlening van woorden uit een Sabellische taal.
Een complicerende factor is voorts de ontwikkeling van het Proto-Indo-Europese
fonemen */gh gh/, die in alle andere Italische talen zich in woord-interne positie
ontwikkelden tot een foneem dat weergegeven werd als h (vermoedelijk [x]), maar in
het Faliscisch lijkt te worden weergegeven k, q, of c, een notatie die eerder op een
occlusief (/g/) dan op een spirant lijkt te wijzen. Een dergelijke ontwikkeling zou echter
niet alleen tegengesteld zijn aan de Faliscische ontwikkeling van woord-interne */bh dh/
(maar wel in de lijn liggen van de Latijnse ontwikkeling van woord-interne */bh dh/ tot
/b d/), maar ook anders dan de ontwikkeling van word-interne */gh gh/ in alle andere
Italische talen. Het materiaal laat geen duidelijke conclusie toe.
Een synchrone vergelijking van de fonemische systemen van Faliscisch en Latijn in de
derde eeuw voor de christelijke jaartelling (§3.4) laten enkele verschillen zien, waarvan
de belangrijkste zijn de zeldzaamheid van /b/ en de grotere frequentie van /f/ in woord-
interne positie in het Faliscisch, en de verdwijning van de tweeklanken als gevolg van
monoftongisering, een proces dat zich in het Faliscisch eerder voltrok dan in het Latijn.
Vervolgens wordt een overzicht gegeven van de ontwikkeling van de Faliscische
medeklinkers gedurende de Vroeg-, Middel-, en Laatfaliscische periodes (§3.5). Enkele
belangrijke deeldiscussies betreffen:
(1) de realisatie van /f/ aan het woordbegin als [h], een verschijnsel dat zowel uit de
epigrafische als uit diverse literaire bronnen bekend is, maar zich niet tot het Faliscisch
alleen lijkt te hebben beperkt, aangezien er ook voorbeelden zijn uit andere Latijnse
dialecten, en (in een latere periode) in het Etruscisch;
(2) het wel of niet plaatsvinden van woord-intern rhotacisme in het Faliscisch: de
twijfels in de literatuur ten aanzien van een Faliscisch rhotacisme blijken grotendeels op
een misinterpretatie van lettervormen te berusten;
(3) de mogelijke palatalisaties van consonanten gevolgd door /i/ of //: suggesties in
deze richting in de literatuur worden grotendeels in twijfel getrokken;
615
(4) het weglaten van medeklinkers aan lettergreep- of woordeinde: betoogd wordt dat
de schrijvers van het Faliscisch een grote neiging hadden tot het weglaten in het schrift
van nasalen aan het lettergreep- en woordeinde, en een nog grotere neiging tot het
weglaten van /s/ aan het woordeinde, doch dat het in beide gevallen eerder gaat om
variatie in de geschreven weergave van fonetische realisaties dan van fonemische
ontwikkelingen. Geopperd wordt dat het vrijwel consequent weglaten van /s/ aan het
woordeinde na een korte klinker (in ong. 97% van alle gevallen) wellicht een
orthografische regel van het Faliscisch is geweest, dat dan in dat opzicht verschilde van
het Latijn.
Eenzelfde overzicht van de ontwikkeling van de Faliscische medeklinkers en
tweeklanken gedurende de Vroeg-, Middel-, en Laatfaliscische periodes (§3.6-7) levert
de volgende belangrijker deeldiscussies op:
(1) in het systeem van de klinkers vond een toenadering tussen /e/ en /i/ plaats, waarbij
beide waarschijnlijk werden gerealiseerd als [ ]: voor een vergelijkbare toenadering van
/o/ en /u/ zijn minder sterke aanwijzingen;
(2) in tegenstelling tot wat gewoonlijk in de literatuur wordt aangenomen, toont het
Faliscisch wel degelijk gevallen van verzwakking of sluiting van woord-interne klinkers
die toe te schrijven zijn aan een (prehistorische) beginklemtoon;
(3) alle korte tweeklanken van het Faliscisch verdwenen al tijdens de Middelfaliscische
periode door monofthongisatie, hetgeen leidde tot een synchronisch verschil tussen de
fonemische systemen van het Faliscisch en van het Latijn, waar de tweeklanken
grotendeels pas later gemonoftongiseerd werden: in deze vroege monofthongisatie lijkt
het Faliscisch op het Umbrisch en het Volscisch;
(4) een opvallende ontwikkeling is die van de tweeklank /o/ tussen twee labialen tot
/o/ (bv. loifiṛtato in inscriptie 31, = Lat. libertatis): deze ontwikkeling moet ook hebben
plaatsgevonden in het Latijn, maar vond niet plaats in de Sabellische talen;
(5) er zijn aanwijzingen voor een tweeklank -ui in de dativus enkelvoud van de vierde
declinatie, wellicht gevormd naar analogie van de (lange) tweeklanken -ai en -oi in de
dativus enkelvoud van de eerste en tweede declinatie.
Als voornaamste conclusie van hoofdstuk 3 (§3.8) wordt gesteld dat de diachronische
ontwikkeling van de fonologie van het Faliscisch op zeer veel punten overeenkomt met
die van het Latijn, en dat het Faliscisch over het algemeen aansluit bij het Latijn op die
punten waar er verschil bestaat tussen het Latijn en de Sabellische talen. Verschillen
tussen Faliscisch en Latijn zijn over het algemeen van recente datum en hebben
betrekking op fonetische of fonemische tendensen die zó universeel zijn dat de
betekenis er van gering is. De grote uitzondering hierop is de ontwikkeling van de
stemhebbende geaspireerde occlusieven in woord-interne positie in het Faliscisch: op
dit punt toont het Faliscisch een vroege afwijking van het Latijn en een ontwikkeling
die overeenkomt met die in de Sabellische talen.
616
Hoofdstuk 4 (The nominal and pronominal inflections, pp.117-154) behandelt de

morfologie van de naamwoorden en de voornaamwoorden. Methodologische kwesties
die aan de orde komen (§4.1) betreffen de mate waarin Etruscische namen wel of niet
aangepast werden aan de morfologie van het Faliscisch, en de meerduidigheid van veel
(contextloze) Faliscische vormen.
Vervolgens wordt de morfologie van de eerste (§4.2) en tweede (§4.3) declinatie
behandeld. Deze levert weinig verrassingen op: de morfologie van het Faliscisch sluit
nauw aan bij die van het Latijn, ook op punten waar het Latijn verschilt van de
Sabellische talen zoals de nominativus enkelvoud van de eerste declinatie (Faliscisch en
Latijn -a , Sabellische talen -o) en de nominativus meervoud van beide declinaties
(Faliscisch en Latijn -ai en *-oi (Middelfaliscisch -e), Sabellische talen -as en -os). Een
discussiepunt is de genitivus van de eerste declinatie: hoewel de literatuur het er over
eens is dat het Faliscisch hier de uitgang -as kende, is men over het algemeen niet
bereid de (meerduidige) Faliscische vormen op -ai als genitivus te interpreteren. Hier
wordt gesteld dat een dergelijke interpretatie wel degelijk mogelijk is, en dat het
Faliscisch waarschijnlijk eenzelfde vervanging van de uitgang -as door -ai heeft gekend
als het contemporaine Latijn. In de tweede declinatie valt de dativus op -oi op, die voor
het Latijn wel verondersteld mag worden, maar niet eenduidig geattesteerd is.
Bovendien wordt betoogd dat sommige van de Faliscische vormen op -oi wellicht
geïnterpreteerd kunnen worden als genitivus.
Het belangrijkste punt van de morfologie van de tweede declinatie is echter
ongetwijfeld de genitivus enkelvoud (§4.4). Al tientallen terug werd vastgesteld dat het
Faliscisch toonde van een genitivus op -osio, een voortzetting van de Proto-Indo-
Europese uitgang */-oso/, terwijl uit het Latijn alleen de genitivus op -i bekend was,
waarvan de oorsprong omstreden is. (In de Sabellische talen was de oorspronkelijke
uitgang van de genitivus enkelvoud van de tweede declinatie al tijdens de Proto-
Sabellische periode vervangen door de uitgang van de derde declinatie.) De aanname
was dat dit een morfologisch punt was waarop Faliscisch en Latijn duidelijk van elkaar
verschilden, en leidde tot de al genoemde aanname dat Faliscisch in de eerste declinatie
geen genitivus op -ai kende, die immers gevormd was naar analogie van de genitivus op
-i. De publicatie van de Vroeglatijnse Satricum-inscriptie (CIL I2.2832a) in 1978 toonde
echter aan dat ook het Latijn aanvankelijk een genitivus op -osio had, en dat de
vervanging van -osio door -i betrekkelijk recent moest hebben plaatsgevonden in zowel
Faliscisch als Latijn.
De verschillende theorieën ten aanzien van de relatie tussen -osio en -i, de manier
waarop en de redenen waarom -osio vervangen werd door -i worden uitgebreid
besproken. Met name wordt scherp stelling genomen tegen het idee dat -i een directe
fonologische ontwikkeling van -osio zou zijn geweest: in plaats daarvan wordt ervan
uitgegaan dat -i teruggaat op een Proto-Indo-Europees */-i /, waarbij gesuggereerd
wordt dat dit */-i / misschien gelijk te stellen is met de nominativus onzijdig meervoud
617
*/- 2/ van de bijvoeglijk naamwoorden op */-os/ en derhalve teruggaat op de

collectieve vorm van een bezittelijk bijvoeglijk naamwoord (bv. */tull 2/ → */tullī/ ‘de
Tullische zaken’). Ook aan deze verklaring kleven echter grote bezwaren. Als onderdeel
van deze discussie wordt ook al het relevante materiaal uit de andere talen van het oude
Italië besproken: hierin wordt ook het materiaal uit de niet-Italische talen zoals
Venetisch, Lepontisch, Messapisch, Siculisch, en Elymisch betrokken, aangezien hier in
de literatuur argumenten aan zijn ontleend. Sommige van deze talen lijken een
vergelijkbare ontwikkeling te hebben doorgemaakt als het Faliscisch en het Latijn.
In de bespreking van de overige declinaties (§4.5-6) zijn de volgende deeldiscussies van
belang:
(1) de genitivus enkelvoud van de derde declinatie toont ook in de Middelfaliscische
inscripties nog een duidelijk onderscheid tussen de consonant-stammen, waar de
uitgang -o(s) is, en de i-stammen, waar de uitgang –e(s) is: hierbij wordt gesteld dat ook
in het contemporaine Latijn de uitgang bij de consonant-stammen zeer waarschijnlijk
-os was;
(2) de dativus enkelvoud van de vierde declinatie gaat uit op -ui: deze uitgang is op
geen enkele manier in overeenstemming te brengen met wat bekend is van de
corresponderende uitgangen in de overige Italische talen en is waarschijnlijk binnen het
Faliscisch gevormd door analogie met de uitgangen -ai en -oi van de eerste en tweede
declinaties.
De voornaamwoorden (§4.7-9) zijn slecht gedocumenteerd, met uitzondering van de
persoonlijk voornaamwoorden (§4.7). Het Faliscisch toont daarbij eco en med (later ook
met) als nominativus en accusativus van het persoonlijk voornaamwoord van de eerste
persoon enkelvoud en komt daarin overeen met het Latijnse ego en med, terwijl de
Sabellische talen weliswaar een nominativus */egō/ kunnen hebben gehad, maar in
ieder geval een anders gevormde accusativus /mēom/ hadden (thans geattesteerd in het
Vroegumbrische (?) míom).
Een bijzonder problematisch punt is het persoonlijk voornaamwoord van de
tweede persoon meervoud, dat in het Vroegfaliscisch verschijnt als ues in inscriptie 4.
Het e-vocalisme is moeilijk te verklaren naast het Latijnse uos (en het bijbehorende
bezittelijk voornaamwoord uoster, dat zich later ontwikkelde tot uester). Dit lijkt het
enige grote verschil tussen Faliscisch en Latijn te zijn op het punt van de morfologie
van de voornaamwoorden, maar eenzelfde eigenaardigheid doet zich ook voor binnen
de Sabellische talen, waar het Paelignische uus staat naast het Umbrische bezittelijk
voornaamwoord uestra (waar het e-vocalisme niet terug kan gaan op een ouder o-
vocalisme). Het probleem van deze vormen lijkt onoplosbaar: gesuggereerd wordt dat
de Proto-Italische vorm mogelijk */ s/ en */estros/ zijn geweest, waarnaast vormen
met een o-vocalisme ontstonden die gevormd zijn naar analogie van de
voornaamwoorden van de eerste persoon meervoud, */n s/ en */nostros/.
618
In hoofdstuk 5 (The verb, pp.155-176) wordt de morfologie van de werkwoorden

besproken. Hierbij zijn geen belangrijke methodologische problemen (§5.1).
Een bespreking van de morfologie van het werkwoord (§5.2) toont enerzijds
duidelijk dat het Faliscisch tot de Italische talen behoort en dat het op cruciale
verschilpunten overeenkomsten vertoont met het Latijn en niet met de Sabellische talen,
maar dat het anderzijds ook een aantal verschilpunten had met het Latijn. Allereerst
toont het Faliscisch een systeem van vier conjugaties en verschilt daarmee niet van de
overige Italische talen. In de manier waarop het futurum gevormd wordt toont het
Faliscisch overeenkomsten met het Latijn waar dit van de Sabellische talen verschilt
(Fal. f-futurum = Lat. b-futurum, tegenover Sab. s-futurum). In de vorming van de
wijzen valt geen verschil aan te wijzen met de andere Italische talen.
In de werkwoordsuitgangen komt het Faliscisch ook weer overeen met het Latijn
waar dit verschilt van de Sabellische talen, m.n. op het gebied van de uitgang van de
eerste persoon enkelvoud van het perfectum, geattesteerd in de Vroegfaliscische vorm
pepara[i in inscriptie 1, en van de secundaire persoonsuitgangen van de derde persoon
enkelvoud, aanvankelijk -ed, later vervangen door -et, een vervanging die ook in het
Latijn plaats vond maar niet in de Sabellische talen, en van de derde persoon meervoud,
-ond, die in de Sabellische talen vervangen was door -ns. Problematisch is het feit dat
deze uitgang in het Vroegfaliscisch opduikt in een perfectumform: indien het Faliscisch
daadwerkelijk een Latijns dialect was zou eerder een vorm als *-eri verwacht worden.
Het is echter niet duidelijk in hoeverre deze Vroegfaliscische vorm representatief is
voor de perfecta in het Vroegfaliscisch. Ook de uitgangen van de gebiedende wijs II
tonen anomalieën, die echter tot op zekere hoogte verklaarbaar zijn.
Vervolgens (§5.3) worden de geattesteerde werkwoordsvormen afzonderlijk besproken.
Belangrijke deeldiscussies betreffen hier de volgende vormen:
(1) esú(m) ‘ik ben’ in inscripties 389, 404, en 465 (alle uit de ager Capenas afkomstig),
een vorm die reeds bekend was uit de Sabellische talen maar ook voor het Latijn
geattesteerd is, zowel in epigrafische als literaire bronnen, en wellicht tot de dialecten
aan de grenzen van het Latijnse gebied behoorde;
(2) de perfectumvormen facet in inscriptie 470* en faced in inscriptie 471*, die pas
sinds enkele jaren bekend zijn: in een uitgebreide bespreking van de Italische perfecta
van de werkwoordsstam fac- wordt gesteld dat hoewel deze vorm verschilt van het
Latijnse perfectum feci, er desondanks overeenkomsten zijn in de zin dat zowel
Faliscisch als Latijn een oude aoristusstam gebruiken als perfectumstam, en dat deze
vernieuwing tamelijk recent moet zijn geweest aangezien het oorspronkelijke
reduplicatieve perfectum voor het Latijn nog geattesteerd is in vhevhaked CIL I2.3. Een
zelfde vernieuwing is te vinden in het Umbrisch; in het Oscisch is echter het oude
reduplicatieve perfectum bewaard.
619
(3) de Vroegfaliscische perfectumvormen f[.f]ịqod in inscriptie 1 en fifiked in inscriptie

9: hier wordt stelling genomen tegen het idee dat deze vormen perfectumvormen zijn
van fac-: ze worden beschouwd als perfectumvormen van fi(n)g- (de traditionele
interpretatie van deze vormen). Het schijnbare verschil tussen het reduplicatieve
perfectum van het Faliscisch en het Latijnse finxi wordt gesuggereerd dat net als in het
geval van de werkwoordsstam fac- eeen ouder reduplicatief perfectum, dat nog
geattesteerd is in het Vroegfaliscisch, later vervangen is door een aoristusstam.
(4) de vermeende werkwoordsvorm i*ice (ipice?) in inscripties 309 en 315: hier wordt
gesteld dat deze vorm waarschijnlijk geen werkwoordsvorm is;
(5) de futurumvormen pipafo in inscriptie 59 en ‹pi›pafo in inscriptie 60: hier worden
drie mogelijke manieren besproken waarop deze vormen afgeleid zouden kunnen zijn
van de praessenstam pip- of pipa-.
(6) de gebiedende wijzen tulate en urate in inscriptie 385, die tot nog toe nauwelijks
enige duiding hadden gekregen.
De deelconclusie van dit hoofdstuk (§5.4) is dat veel van de verschillen tussen het
Faliscisch en het Latijn op het gebied van de werkwoorden zich goed laten verklaren en
in een aantal gevallen slechts schijnverschillen zijn.
In hoofdstuk 6 (The lexicon, pp.177-212) wordt gekeken naar de lexicale elementen in

de Faliscische inscripties. Methodologische punten die hierbij besproken worden (§6.1)
zijn (1) de wenselijkheid om bij op vergelijking gebaseerd onderzoek naar de
taalkundige positie van een taal of dialect, prioriteit te geven aan synchrone vergelijking
boven diachrone woordafleiding, (2) het probleem van welke inscripties het Faliscisch
weerspiegelen en welke die van het taalkundig zeer nabijstaande Latijn, en (3) het
gebruik van het onomasticon als gegevensbron voor lexicale elementen.
Vervolgens (§6.2) worden alle in de inscripties geattesteerde lexicale elementen
afzonderlijk geëvalueerd. Interessante deeldiscussies hierbij betreffen de woorden efiles
(al dan niet een calque op het Latijnse aediles), cela in de betekenis ‘grafkamer’ (al dan
niet een ontlening uit het Etruscisch), het hypothetische woord *gutto/guttom/guttor
‘schenkkan’ (vorm en ontleningstraject), het hypothetische woord *putellios/putelliom
‘(jong) kind’ (mogelijke afleiding), het ghostword †sorex, en het mogelijk nog als
bijvoeglijk naamwoord fungerende titos ‘voorspoedig, welvarend’.
Evaluatie van deze lexicale elementen (§6.3) laat zien dat, hoewel de gegevens
lacuneus zijn, er valt vast te stellen dat het geattesteerde Faliscische lexicon vrijwel
geheel overeenkomt met het Latijnse, ook waar het Latijnse lexicon duidelijk verschilt
van het Sabellische, zoals in de woorden filius/filia voor ‘zoon’/‘dochter’, tegenover
het Sabellische *puclom/*futer. Woorden van andere herkomst kunnen worden
toegeschreven aan recente ontlening of interferentie, zoals in de gevallen van
postigna en pescum. Tevens worden lexicale subsets besproken, waarbij blijkt dat de
620
subset die betrekking heeft op de grafvorm het lexicon weliswaar vergelijkbaar is met
het Latijnse, doch een aantal eigen betekenissen toont die in Latium, waar de
grafvorm amders was dan in de ager Faliscus, niet of nauwelijks voorkomen, zoals
cella met de betekenis ‘grafkamer’ en lectus met de betekenis ‘ligplaats voor de
dode’. De subset die betrekking heeft op het politieke bestuur tenslotte is vrijwel
geheel aan het Latijn ontleend en weerspiegelt de door de Romeinen na de oorlog van
241-240 voor de christelijke jaartelling ingestelde bestuursstructuur.
Aparte aandacht wordt verder besteed aan de epigrafisch geattesteerde theonymen
(§6.4), toponymen, potamonymen, en ethnonymen (§6.5). De theonymen komen
overeen met die in het Latijn, en hebben vrijwel alle betrekking op Italische (en niet op
Etruscische) goden. In de toponymen, potamonymen, en ethnonymen komen diverse
geografische namen uit de directe omgeving van de ager Faliscus terug. Hierbij dient
gewezen te worden op de naam calitenes in inscriptie 265, een naam die misschien is
afgeleid van de oorspronkelijke Etruscische naam van het huidige Gallese, en de namen
ortecese in inscriptie 339 en urtcsnas in inscriptie XXXV, die zouden kunnen zijn
afgeleid van de oorspronkelijke Etruscische naam van het huidige Corchiano. Ten slotte
(§6.6) worden de weinige glossen besproken die door Latijnse of Griekse auteurs aan de
Falisken werden toegeschreven.
Hoofdstuk 7 (The onomasticon, pp.213-290) bestaat uit een evaluatie van het
onomasticon van het gebied. Methodologische problemen hier zijn (1) het feit dat
namen in zekere zin slechts zijdelings deel uitmaken van een taal, en zich niet in alle
opzichten gedragen als lexicale elementen, waardoor bijvoorbeeld de ontlening van
namen aan andere talen duidelijk gemakkelijker verloopt dan de ontlening van lexicale
elementen, (2) het feit dat namen primair niet zozeer een lexicale betekenis als wel een
symbolische functie hebben en derhalve als belangrijke markeerders van verschillende
soorten identiteit kunnen fungeren, (3) en dat derhalve het aanpassen van een naam aan
een andere taal (of, omgekeerd, het onveranderd laten van de eigen naam bij het gebruik
van een andere taal) van groot belang kunnen zijn bij het vaststellen van identiteit.
Gewaarschuwd wordt tenslotte voor het te gemakkelijk trekken van conclusies op grond
van een achternaam die uit een bepaalde taal afkomstig is: zo is het hebben van een (in
oorsprong) Etruscische achternaam zeker geen automatisch bewijs voor het feit dat een
persoon met een dergelijke achternaam ook (nog) Etruscisch was of zich als Etrusk
beschouwd, of zelfs maar Etruscisch sprak.
Vervolgens worden in diverse paragrafen de namen en de naamsformules uit het gebied
geanalyseerd: allereerst de namen in de Vroegfaliscische inscripties (§7.2), die
voornamelijk uit enkele namen bestaan (hoewel vroege voorbeelden van familienamen
ook al lijken op te duiken), en uit enkele lastig analyseerbare naamsgroepen, vervolgens
de Middel- en Laatfaliscische naamsformules van mannen (§7.3) en van vrouwen
621
(§7.4). Bij zowel mannen als vrouwen bestaat de officiële naamsformule uit voornaam
+ achternaam, eventueel gevolgd door een filiatie (vermelding van de vadersnaam). De
officiële naamsformule wordt vooral gebruikt in grafinscripties; in bezitsinscripties op
aardewerk wordt ook wel een kortere vorm van de naamsformule gebruikt, die bij
mannen bestaat uit de voornaam, maar bij vrouwen eerder uit de achternaam.
Hiernaast worden nog enkele uitbreidingen en varianten besproken.
Vrouwennamen konden worden uitgebreid met vermelding van de naam van de
echtgenoot als ‘echtgenote van …’ (§7.4.2), hetgeen een toevoeging is die alleen in
grafinscripties wordt gevonden, aangezien vrouwen die in het familiegraf van de
echtgenoot werden begraven een andere achternaam hadden dan de rest van de
overledenen aldaar. De filiatie (§7.5) kan twee vormen aannemen, nl. ‘zoon/dochter van
…’ of door middel van een van de vadersnaam afgeleid bijvoeglijk naamwoord, een
zgn. patronymicum. Het gebruik van patronymica moet zeer verbreid zijn geweest in
het oude Italië, aangezien patronymica de basis vormen van veel van de Italische en
Etruscische familienamen, maar alleen in de ager Faliscus blijft het patronymicum als
zodanig tot ver in de historische periode in gebruik (tot in de tweede eeuw voor de
christelijke jaartelling). Tenslotte wordt een kort woord gewijd aan de namen van
vrijgelatenen (§7.6). De naamsformules van vrijgelatenen lijken weinig af te wijken van
die van vrijgeborenen, hoewel er aanwijzing zijn dat in de ager Faliscus het Etruscische
systeem van de dubbele familienaam voor vrijgelatenen ook in gebruik was.
Vervolgens volgen uitgebreide analyses van alle uit de epigrafische bronnen bekende
voornamen (§7.7) en familienamen (§7.8) in het gebied, waarbij vooral gekeken is naar
de herkomst van deze namen (locaal, Etruscisch, Latijns, of Sabellisch), naar de
frequentie, en naar de distributie. De belangrijkste conclusies hier zijn:
(1) dat de ager Faliscus eigen voornamen kende die nergens anders voorkwamen, zoals
Iuna, Volta, en Voltius, terwijl sommige andere namen in de ager Faliscus duidelijk
vaker voorkwamen dan elders, zoals Gavius, Aufilus, en Laevius en Laevilius;
(2) dat in de ager Faliscus veel Etruscische voornamen werden gebruikt, maar dat deze
stuk voor stuk niet frequent voorkomen en niet gebonden blijken te zijn aan families
met (in oorsprong) Etruscische achternamen (zie hieronder over §9.2);
(3) dat in de ager Faliscus veel Etruscische achternamen werden gebruikt, maar dat deze
in veel gevallen zijn aangepast aan de morfologie van het Faliscisch;
(4) en dat veel van de Faliscische achternamen en (in mindere mate) voornamen
oorspronkelijk patronymica waren.
Vervolgens wordt een kort woord gewijd aan de cognomina of toenamen (§7.9) : deze
waren slechts weinig frequent, vermoedelijk nog niet erfelijk, en nog geen vast
onderdeel van de officiële naamsformule, hetgeen in overeenstemming is met het
gebruik van dergelijke namen in Latium in dezelfde periode.
622
Tenslotte (§7.10) worden de bevindingen met betrekking tot de namen geplaatst in het
kader van de vraagstelling met betrekking tot de (ethnische) identiteit. Anders dan vaak
wordt aangenomen blijken de namen slechts een tamelijk diffuus beeld te geven van de
ethnische samenstelling van de bevolking van de ager Faliscus. Dit blijkt vooral uit de
grote mengelmoes van namen van verschillende origine in het gebied, waarbij
voornamen en achternamen van verschillende herkomst zonder meer aan elkaar
gekoppeld worden. Wel zijn er duidelijke aanwijzingen dat bepaalde voornamen
functioneerden als markeerders van een ethnische identiteit: zo verdwijnen de specifiek
Faliscische voornamen spoedig na de overname van het gebied door de Romeinen na de
oorlog van 241-240 voor de christelijke jaartelling, terwijl tezelfdertijd tot dan toe
afwezige of niet frequente voornamen die typerend zijn voor het Latijn zoals Gaius en
Marcus hun intrede doen. Dit beeld wordt versterkt door duidelijke aanwijzingen dat
bepaalde voornamen een rol speelden binnen bepaalde families (zgn. ‘erfelijke
voornamen’).
In hoofdstuk 8 (Syntax and text structure, pp.291-318) wordt ingegaan op de manier

waarop in het Faliscisch zinnen en teksttypes geconstrueerd werden. Een
methodologisch punt hierbij (§8.1) is de uitsluiting van de Latijnse inscripties, daar deze
een zodanig andere repertoir aan teksten omvatten dat een vergelijking met de
Faliscische inscripties onmogelijk is en een scheef beeld op zou leveren van het
materiaal.
Een analyse van de morfosyntaxis van de naamwoorden (§8.2) en de werkwoorden
(§8.3), de volgorde van de zinsdelen in de zin (§8.4) en in woordgroepen (§8.5), en
coördinatie en subordinatie (§8.6) toont weinig verrassends: het Faliscisch toont op
deze punten geen verschillen met de andere Italische talen. Het weinige dat te zeggen
valt met betrekking tot de taaltypologie van het Faliscisch (§8.7) zou mogelijkerwijs
kunnen wijzen op iets meer overeenkomsten met het Latijn dan met de Sabellische
talen, maar het materiaal hiervoor is te schaars om tot duidelijke uitspraken te komen.
Veel interessanter blijkt de analyse van de verschillende teksttypes. Onder de
bezitsinscripties (§8.8) worden de gebruikelijke basistypes aangetroffen (naam in de
nominativus, naam in de genitivus), maar ook voorbeelden van ‘sprekende inscripties’
met teksten als eco gevolgd door een naam in de genitivus (‘ik [ben] van ...’), eco met
een vaasnaam in de nominativus en een persoonsnaam in de genitivus (‘ik ben de ... van
..’), en eco gecombineerd met een naam in de nominativus (‘ik [ben] ...’), en een naam
in de genitivus gevolgd door esú(m) ‘van ... ben ik’. Vergelijking met sprekende
inscripties uit andere talen van het oude Italië toont dat de eerste drie types van
Etruscische origine zijn, en het laatste type vermoedelijk ontstaan is bij vervanging van
deze voornaamwoordelijke formules door Latijnse werkwoordelijke formules.
Een dergelijk beeld komt ook naar voren bij de ‘signaturen’ (§8.9) van
pottenbakkers. Ook hier komen voorbeelden voor van een naam in de nominativus of de
623
genitivus, maar daarnaast worden ook voorbeelden gevonden van sprekende inscripties
bestaande uit een naam gevolgd door med fifiked ‘... kneedde mij’ (Vroegfaliscisch) of
met facet ‘... maakte mij’ (Middelfaliscisch). Interessant is dat deze formule weliswaar
teruggaat op een Etruscisch voorbeeld, maar aangepast is aan de woordvolgorde van het
Faliscisch. Uniek is het gebruikt, in de Vroegfaliscische vorm van deze formule, van het
specifieke werkwoord fi(n)g- ‘kneden’, terwijl de vergelijkbare formules in de andere
talen van het oude Italië een algemeen werkwoord ‘maken’ gebruiken.
De grafinscripties (§8.10) tonen een formule bestaand uit een of meer namen in
de nominativus gevolgd door de woorden hec cupat/cupant ‘... ligt/liggen hier’.
Vergelijkbare formules komen voor in andere talen van Centraal-Italië, maar de variatie
binnen deze formules is groot, en de frequentie is aanmerkelijk lager dan in het
Faliscisch, waar dit de standaardformule voor grafinscripties lijkt te zijn (met
incidentele variaties zoals cupat ifra ‘ligt hieronder’ in inscriptie 40 en lecet hec in
inscriptie 88, met een ander werkwoord voor ‘liggen’). Het Etruscisch had een
vergelijkbare formule met de woorden θui cesu, maar ook van deze formule komt de
frequentie niet in de buurt van die van de Faliscische. Afgezien van deze formule wordt
voorts aandacht besteed aan het verschijnen van cursus honorum in Latinofaliscische en
Latijnse grafinscripties, hetgeen duidelijk toe te schrijven is aan Romeinse invloed, aan
de problematische gevallen van grafinscripties bestaande uit twee namen, één in de
nominativus en één in de dativus) en misschien te lezen als ‘... [maakte dit graf] voor
...’, en de inscripties die refereren aan het grafrecht (ius sepulcrale).
De wijdingsinscripties (§8.11) bestaan uit godennamen in de nominativus of de
genitivus en zijn derhalve vergelijkbaar met de bezitsinscripties. Daarnaast komt ook
het meer gebruikelijke type voor waarin de godennaam in de dativus staat (‘voor ...’),
soms vergezeld van een naam van de schenker in de nominativus (‘..., voor ...’). Het is
echter mogelijk dat het relatief grote aantal inscripties waarin de godennaam in de
genitivus staat verklaard moet worden door invloed vanuit het Etruscisch aan te nemen,
waar een werkelijke dativus wellicht ontbrak en de functies die de dativus had in de
Italische talen deels werden vervuld door de genitivus. Daarnaast wordt er minstens een
voorbeeld gevonden van een wijdingsinscriptie met het woord sacer ‘gewijd’, hetgeen
eerder een Italisch dan een Etruscisch gebruik lijkt te zijn: het Faliscische voorbeeld
(inscriptie 127, sacra) is echter uniek in de zin dat de naam van de godheid niet
genoemd wordt.
Op het gebied van de officiële inscripties (§8.12) tenslotte is er een duidelijk
verschil tussen de eenvoudige manier waarop deze in het Faliscisch geformuleerd
werden, met alleen de namen van de magistraten, en de manier waarop deze in Latijnse
inscripties uit het gebied geformuleerd werden, met aanmerkelijk uitgebreider formules.
Dit weerspiegelt het andere tekstrepertoir dat zijn intrede deed in het gebied met de
overname door de Romeinen na de oorlog van 241-240 voor de christelijke jaartelling.
624
Hoofdstuk 9 (Language contact, pp.319-340) is gewijd aan de taalcontacten die

moeten hebben bestaan tussen het Faliscisch enerzijds en het Etruscisch, de Sabellische
talen, en het Latijn anderzijds. Eerst (§9.1) wordt hierbij ingegaan op het
methodologische probleem van het analyseren van taalcontacten uit een beperkt
epigrafisch corpus: de aard van dit materiaal maakt het niet alleen bijzonder moeilijk
om onderscheid te maken tussen interferentie en ontlening, maar verschijnselen die aan
interferentie of ontlening kunnen worden toegeschreven veronderstellen ook dat dit
‘vreemde element’ werd meegenomen in de geschreven versie van de taal waar het in
terecht kwam, hetgeen een extra stap vereist ten opzichte van vergelijkbare
verschijnselen in gesproken taal.
Vervolgens worden de contacten met het Etruscisch besproken (§9.2) moeten vooral in
de periode tot de vierde eeuw van groot belang zijn geweest, maar het lijkt er op dat in
de ager Faliscus in elke periode een deel van de bevolking Etruscisch in ieder geval als
tweede taal maar wellicht ook als eerste taal beheerste. De contacten zullen daarom
zowel op formeel (officieel) als op informeel (persoonlijk) niveau hebben
plaatsgevonden. Opvallend zijn in dit opzicht de Etruscische inscripties XXXVIII en
XXXIX: dit lijken officiële inscripties te zijn, hetgeen er op zou wijzen dat Etruscisch in
voorkomende gevallen gebruikt kon worden door (locale) magistraten, zelfs al was
Faliscisch vermoedelijk de eerste taal van de meerderheid van de bewoners van het
gebied. Desondanks blijkt het moeilijk om daadwerkelijk Etruscische kenmerken aan te
wijzen in de Faliscische inscripties. Als zodanig worden besproken :
(1) fonologische kenmerken, waarbij als enige plausibel punt wordt genoemd de
mogelijkheid van een ‘sterkere’ Etruscische realisatie van /s/ aan het woordeinde na een
korte klinker;
(2) morfologische kenmerken, die grotendeels in twijfel worden getrokken óf kunnen
worden toegeschreven aan onomastische ontlening, waarbij het morfeem van de
Etruscische nominativus mee ontleend wordt als onderdeel van de naam: hieronder
vallen wellicht ook de ‘Etruscoide’ vormen op -ies;
(3) lexicale ontleningen, die grotendeels in twijfel worden getrokken;
(4) een syntactisch kenmerk, namelijk de verwarring van verschillende naamvallen die
in sommige inscripties waarneembaar is, lijkt wél aan het Etruscisch toe te schrijven;
(5) de reeds in hoofdstuk 8 besproken tekstformules zijn ook met zekerheid aan het
Etruscisch toe te schrijven.
Een aparte paragraaf (§9.2.3) is gewijd aan de mogelijke aanwezigheid van
Etruscischtalige groepen of families in de ager Faliscus, die al eerder het onderwerp zijn
geweest van aparte studies. Het merendeel van de grafinscripties uit Corchiano
(inscripties 257-258, 265-266, en 269-272), alsmede de inscripties uit een tombe bij de
Ponte Terrano in Civita Castellana (inscripties 41-43) bevatten een aantal epigrafische,
orthografische, onomastische, en taalkundige kenmerken die alleen aan het Etruscisch
625
toegeschreven kunnen worden. De concentratie van deze kenmerken in deze inscripties

is hoog, en de conclusie is daarom dan ook dat deze inscripties duidelijk aan personen
toe te schrijven zijn van wie de eerste taal eerder Etruscisch was dan Faliscisch.
De contacten met de Sabellische talen (§9.3) moeten frequent zijn geweest. Gesteld
wordt echter dat in ieder geval op taalkundig gebied er nauwelijks aanwijzingen zijn te
vinden voor vermeende grootschalige invasies vanuit het Sabellische gebied ten oosten
van de Tiber, hoewel migraties op kleine schaal niet uitgesloten kunnen worden : in dit
kader wordt tevens besproken de implicaties van de Hirpi Sorani op de Soracte zoals
beschreven door Servius en Plinius de Oudere. Epigrafische en taalkundige aspecten
van de inscripties van het studiegebied die op Sabellische talen zijn terug te voeren zijn
deze vrijwel geheel beperkt tot de ager Capenas, zijn deze goed verklaarbaar als
interferentie, toe te schrijven aan individuen van wie de eerste taal een Sabellische taal
was eerder dan Faliscisch. Bijzondere aandacht is er voor inscriptie 474* van
onbekende herkomst (misschien afkomstig uit Falerii Novi) en inscriptie 431 uit het
heiligdom van Lucus Feroniae in de ager Capenas: in beide inscripties wordt het
voorwerp van de inscriptie zelf aangeduid met een woord dat alleeen parallellen heeft in
Sabellische talen, respectievelijk posticnu en pesco(m), terwijl de taal van de inscripties
verder volledig Faliscisch of Latijn lijkt te zijn.
De contacten tussen Faliscisch en Latijn (§9.4) blijken veel moeilijker analyseerbaar,
juist vanwege de grote mate van overeenkomst tussen beide. Zeker in de periode na de
oorlog van 241-240 voor de christelijke jaartelling moeten deze contacten frequent en
intensief zijn geweest, en op alle niveau’s van taal en taalgebruik hun invloed hebben
uitgeoefend. Voor zover deze contacten echter terugkeren in de Faliscische inscripties
lijkt er niet zozeer sprake te zijn van kenmerken van het Latijn die in het Faliscisch
doordringen, maar een volledige vervanging van de Faliscische manier waarop
inscripties vorm werd gegeven door een Latijnse manier: het ‘schema’ of ‘model’ van
een geschreven tekst wordt na de Romeinse overname van het gebied volgend op de
oorlog van 241-240 voor de christelijke jaartelling vervangen door een nieuw schema.
Dit uit zich in epigrafische, orthografische, morfofonologische, en tekstuele kenmerken
die radicaal anders waren dan in de Faliscische inscripties. Hoe zich dit verhield tot de
manier waarop gesproken Faliscisch beïnvloed werd door het Latijn is derhalve niet
echt waar te nemen.
In hoofdstuk 10 (Conclusion: Faliscan as a Latin dialect, pp.341-360) worden de

analyses en deelconclusies van hoofdstukken 2-9 bijeen gevoegd en geintegreerd.
Terugkerend naar de verschillende definities van dialect zoals die gegeven werden in
hoofdstuk 1 (§10.2) wordt allereerst geconcludeerd dat de aanwijzingen voor een
‘Faliscische identiteit’ en de rol van het Faliscisch zelf daarin dermate sterk zijn dat
626
vanuit een strict sociolinguistisch perspectief het welhaast noodzakelijk is om het

Faliscisch te beschouwen als een taal.
De conclusie wordt echter anders als de vraagstelling benaderd wordt vanuit de in
hoofdstuk 1 besproken structurele benadering. Achtereenvolgens wordt het materiaal
daarom besproken vanuit een zo synchronisch mogelijk perspectief en vanuit een
diachronisch perspectief. Een probleem bij het synchronisch perspectief is dat de
mogelijkheid tot vergelijking van het Vroegfaliscische materiaal uitgesproken beperkt is
en derhalve veel vragen onbeantwoord laat: een strict synchronische vergelijking van
het Faliscische materiaal met het contemporaine Latijnse en Sabellische materiaal kan
eigenlijk alleen zinnig worden uitgevoerd voor de periode tussen ong. 300 en 250 voor
de christelijke jaartelling, d.w.z. voor het einde van de Middelfaliscische periode. Bij
een vergelijking van het materiaal uit deze periode zouden echter weer de belangrijke
gegevens van het Vroegfaliscische materiaal buiten beschouwing blijven. Tevens doet
dit geen recht aan de diachronische ontwikkelingen die tot deze synchronie geleid
hebben.
Derhalve wordt verder uitgegaan van een meer diachrone dan synchrone
benadering, die het mogelijk maakt om gegevens uit alle periodes te gebruiken en deze
in een groter perspectief te plaatsen. Er wordt betoogd dat bij een dergelijke
beschouwing de verschillen tussen Faliscisch en contemporain Latijn gering zijn, en dat
de meeste van deze verschillen toe te schrijven zijn aan relatief recente ontwikkelingen.
De verschillen tussen Faliscisch en Latijn die aparte discussie vereisen zijn de volgende:
(1) de verschillende ontwikkelingen van */bh dh/ in woord-interne positie: er lijkt
binnen een strict structurele versie van de diachronische benadering geen verklaring te
vinden waarom het Faliscisch in één vroege fonologische ontwikkeling afwijkt van het
Latijn en een ontwikkeling toont zoals die plaats moet hebben gevonden in de
Sabellische talen, terwijl het Faliscisch in latere ontwikkelingen zich weer wél
ontwikkelt zoals het Latijn;
(2) de vorming van het Vroegfaliscische perfectum fifiked in inscriptie 9 en f[.]ịqod in
inscriptie 1 tegenover het Latijnse finxi, en van het Middelfaliscische perfectum facet in
inscriptie 470* en faced in inscriptie 471* tegenover het Latijnse feci (ouder vhevhaked
CIL I2.3): dit verschil valt te verklaren door aan te nemen dat in deze werkwoorden een
oorspronkelijk reduplicerend perfectum in een recente ontwikkeling werd vervangen
door een oorspronkelijke aoristusvorm, hetgeen overigens niet verklaard waarom het
Faliscisch in het geval van de werkwoordsstam fac- koos voor de aoristusstam fac-
/fak-/ (zoals ook het Umbrisch deed) en het Latijn voor de aoristusstam fec- /fēk-/;
(3) het Vroegfaliscische persoonlijk voornaamwoord ues in inscriptie 4, dat niet in
overeenstemming is met de corresponderende Latijnse vorm uos en uester: eerder werd
er echter al op gewezen dat een dergelijk probleem zich ook voordoet binnen de
Sabellische talen (Paelignisch uus tegenover Umbrisch uestra).
627
Deze problemen worden opgevat als de enige daadwerkelijk significante verschillen

tussen Faliscisch en Latijn, en daarmee als de enige sterke argumenten tégen een
Faliscisch-Latijnse eenheid.
Vervolgens (§10.2) worden de visies van een aantal auteurs besproken die zich in de
laatste 50 jaar hebben uitgesproken over de taalkundige positie van het Faliscisch:
(1) Campanile, Studi sulla posizione dialettale del latino (1969), pp.85-92, die het
Faliscisch zag als grotendeels een onafhankelijke Italische taal;
(2) Solta, Zur Stellung der lateinischen Sprache (1974), pp.45-47, die een convergentie
zag van het Faliscisch met de Sabellische talen
(3) G. Giacomelli, La lingua falisca (1963), p.21, die zich opmerkelijk vaag uitliet
over deze kwestie, maar later in ‘Il falisco’ in Lingue e dialetti dell’Italia antica
(1978), pp.509-535, uitging van een sterke beïnvloeding door of convergentie met de
Sabellische talen;
(4) Joseph & Wallace, ‘Is Faliscan a local Latin patois?’ (Diachronica 8 (1991),
pp.159-186, die ontkenden dat het Faliscisch een Latijns dialect kon worden genoemd;
(5) R. Giacomelli, Ricerche falische (1978), p.67 en Nuove ricerche falische (2006),
passim, die de verschillen tussen Faliscisch en Latijn vooral presenteert in het kader van
niveauverschillen binnen wat kennelijk in wezen dezelfde taal is.
In veel gevallen blijken de argumenten van deze auteurs niet gebaseerd op een weging
van de verschillende punten van overeenkomst of verschil, of gebaseerd op een onjuiste
vergelijking, bv. van Vroegfaliscisch materiaal met veel later Latijns of Sabellisch
materiaal. Gesteld wordt dat de taalkundige afstand tussen Faliscisch en Latijn in
weerwil van de genoemde verschillen te klein is om de benaming ‘taal’ voor het
Faliscisch op structureel taalkundige gronden te rechtvaardigen. Aan de andere kant
worden de verschillen te groot geacht om als ‘niveauverschil’ te betitelen.
Deel 2 (hoofdstukken 11-19) bevat de editie van het epigraphische materiaal uit de ager
Faliscus en Capenas. Hierin worden alle relevante inscripties (535 in totaal, inclusief de
Etruscische inscripties en de inscripties van onbekende of andere herkomst die
beschouwd worden als Faliscisch) opnieuw gepubliceerd. Elke inscriptie is voorzien
van een complete bibliographie, inclusief verwijzingen naar alle foto’s en tekeningen.
Hoofdstuk 11 (The epigraphic material, pp.361-392) bestaat uit een bespreking van de
epigrafische aspecten van het Faliscisch materiaal. Eerst (§11.1) worden de inscripties
in de editie uitgesplitst op vier verschillende manieren: (1) naar herkomst (Civita
Castellana (Falerii), S. Maria di Falleri, Corchiano en de noordelijke ager Faliscus, de
628
zuidelijke ager Faliscus en de ager Capenas, de inscripties van onbekende herkomst);

(2) naar periode/alphabet-groep (Vroegfaliscisch, Middelfaliscisch, Middel- of
Laatfaliscisch, Laatfaliscisch, Latinofaliscisch, Capenatisch, Latijn, Etruscisch); (3)
naar type (grafinscripties, inscripties op losse voorwerpen, wijdingsinscripties,
inscripties op openbare werken); (4) naar alfabet (Faliscisch, Etruscisch, Latijn).
Vervolgens (§11.2) wordt kort het ontstaan van het Faliscische alfabet behandeld. Een
belangrijk punt daarbij is of het Faliscische en het Latijnse alfabet direct uit een
Westgrieks alfabet ontstaan zijn, of uit indirect via een Zuidetruscisch alfabet. Een
belangrijke rol wordt daarbij gespeeld door twee uitzonderlijke zevende-eeuwse
alfabetaria uit Narce (ET Fa 9.1 = inscriptie I) en uit Capena (ET Fa 9.2 = inscriptie
XLIV). Mogelijkerwijs vertegenwoordigen deze een vroeg Etruscisch alfabet dat
ontwikkeld werd langs de benedenloop van de Tiber (‘Lower-Tiber alphabet’) en de
basis vormde voor niet alleen het Faliscische en het Latijnse alfabet, maar ook voor het
alfabet dat in diverse Vroegsabellische inscripties wordt gevonden.
Vervolgens wordt het alfabet besproken zoals het wordt aangetroffen in de
Vroegfaliscische inscripties. Interessante punten hierbij zijn de schrijfrichting, die
rechtsgericht is in de allervroegste inscripties maar al tijdens de Vroegfaliscische
periode verandert in linksgericht, en de zgn. ‘C/K/Q-conventie’ (het gebruik van c voor
e en i, van k voor a, en van q voor o en u). Daarna worden alfabet en orthografie van
de Middel- en Laatfaliscische inscripties besproken, met speciale aandacht voor de
lettervormen en hun (cursieve) varianten, en orthografische eigenaardigheden als het
gebruik van k voor /g/ (het Faliscische alfabet had geen aparte letter voor /g/), het
gebruik van θ en z in plaats van t en s (vermoedelijk toe te schrijven aan Etruscische
invloed), en omgekeerde initialen in het geval van vrouwennamen.
Tenslotte wordt in het algemeen de implicaties van het gebruik van het Etruscisch
en het Latijnse alfabet in Faliscische inscripties besproken (§11.3). Het Etruscische
alfabet is van zeer vroege tijd af aanwezig in het gebied, maar het komt slechts zeer
zelden voor dat Faliscische inscripties in Etruscisch alfabet of Etruscische inscripties in
Faliscisch alfabet worden geschreven. Het Latijnse alfabet is kennelijk niet van zeer
vroege tijd af in het gebied aanwezig: de inscripties in Latijns alfabet lijken vooral te
stammen uit de periode ná 240 voor de christelijke jaartelling, wanneer de Romeinen
zich blijvend in de ager Faliscus hebben gevestigd. In verband met de
dateringsproblemen van de inscripties is dit echter bijzonder moeilijk hard te maken.
Hoofdstukken 12-19 bevatten de inscripties zelf, allereerst de Vroegfaliscische

inscripties (hoofdstuk 12, pp.393-416), vervolgens de andere inscripties ingedeeld naar
herkomst (hoofdstukken 13-14, pp.417-492: Civita Castellana = Falerii Veteres;
hoofdstuk 15, pp.493-518: S. Maria di Falleri = Falerii Novi; hoofdstuk 16, pp.519-554:
Corchiano en de noordelijke ager Faliscus; hoofdstuk 17, pp.555-576: de zuidelijke ager
Faliscus en de ager Capenas; hoofdstuk 18, pp.577-593: inscripties van onbekende of
629
andere herkomst die als Faliscisch of Capenatisch zijn beschouwd; hoofdstuk 19,
pp.594-608: de Etruscische inscripties uit de ager Faliscus en de ager Capenas). Alle
inscripties zijn voorzien van een complete bibliografie, inclusief verwijzingen naar alle
bekende foto’s en tekeningen. Van belang zijn m.n. de volgende inscripties:
(1) inscriptie 214 (pp.494-497): deze inscriptie, hoogstwaarschijnlijk de jongste
inscriptie in Faliscisch alphabet, is de laatste jaren het onderwerp geweest van diverse
publicaties, m.n. uit het oogpunt van locale variatie binnen het Latijn;
(2) inscriptie 231 (pp.506-508): hier wordt op basis van autopsie een substantieel
nieuwe lezing voorgesteld;
(3) inscripties 242-249 (pp.512-515) en 361 (pp.553-554): de lezing en interpretatie van
deze inscripties, gepubliceerd in 1990, wordt op een aantal punten herzien;
(4) inscripties 470* (pp.579-580) en 471* (p.580) - Deze twee inscripties, onafhankelijk
van elkaar gepubliceerd in resp. 2005 en 2003 bevatten beide de perfectumvorm
faced/facet en zijn daardoor het onderwerp geworden van een hernieuwde discussie
over de perfectumvormen van de werkwoordstam fac- in de Italische talen;
(5) aan het einde van hoofdstuk 18 (§18.2) worden drie inscripties uit Ardea besproken
die beschouwd worden als Faliscisch. Hoewel er epigrafische overeenkomsten zijn, zijn
er geen doorslaggevende taalkundige argumenten om deze inscripties als Faliscisch te
beschouwen, of om een speciale band tussen Faliscisch en Ardeatisch aan te nemen.
Hetzelfde geldt voor het recentelijk geopperde idee dat de Satricum-inscriptie (CIL
I2.2832a) Faliscisch zou zijn (§18.3).
630
Concordances
The numbers of my edition are presented alongside those used in Giacomelli’s La lingua falisca
(1963), Vetter’s Handbuch der italischen Dialekte (1953), Herbig’s Corpus inscriptionum
Etruscarum II.2.1 (1912), Rix’s Etruskische Texte (1991), Lommatzsch’s, Degrassi &
Krummrey’s Corpus inscriptionum Latinarum I2 (1918, 1931, 1943, 1986), and Bormann’s
Corpus inscriptionum Latinarum XI (1888, 1901, 1925), and vice versa. Following these
concordances is a list of the first publications of the inscriptions that have been published since the
appearance of Giacomelli’s La lingua falisca in 1963.
1. GENERAL
Bakkum Giacomelli Vetter CIE ET CIL I2 CIL XI
1 1 241 8079 - - -
2 2a 242A - - - -
3 2b 242B - - - -
4 3 243 - - - -
5 22 250 8001 Fa 0.3 - -
6 4a 245b 8163 - - -
7 4b 245a 8163 - - -
8 11b 257 - - - -
9 11a 257 - - - -
10 31 260 8030 - 2912 -
11 68,I 271b 8077 - - -
12 68,IIa-b 271a 8076 - - -
13 69 272 8078 - - -
14 67 270 8075 - - 7516
15 66,I 269a 8070 - - -
16 66,II sub 269a 8071 - - -
17 66,III 269b 8072 - - -
18 66,V 269c 8074 - - -
19 66,IV sub 269c 8073 - - -
20 23 251 8002 - - -
21 29,III p. 294 8020 - - -
22 24,I 252 8003 - - -
23 24,II 252 8004 - - -
24 24,III 252 8005 - - -
25 24,IV 252 8006 - - -
26 24,V 252 8007 - - -
27 24,VI 252 8008 - - -
28 27 254b 8013 - - -
29 26 254a 8012 - - -
30 38,III - 8294 - - -
31 25,I 253 8010 - - -
32 25,II 253 8011 - - -
33 30 256b 8018 - - -
34 28,I 255 8014 - - -
35 28,II 255 8015 - - -
36 28,III 255 8016 - - -
37 29,I 256a 8017 Fa 2.18 - -
631
CONCORDANCES
38 29,II 256c 8019 - - -

39 87 288 8566 - - -
40 88 - - - - -
41 73,I 276a 8190 - - -
42 73,II 276b 8191 - - -
43 73,III 276c 8192 - - -
44 8 247 8193 - - -
45 9 248 8194 - - -
46 10 249 8195 - - -
47 70 273 8164 - - -
48 71,I 274a 8167 - - -
49 71,II 274b 8168 - - -
50 71,III 274c 8169 - - -
51 71,IV 274d 8170 - - -
52 71,VI 274f 8172 - - -
53 71,VII 274g 8173 - - -
54 71,V 274e 8171 - - -
55 - - 8175 - - -
56 72,I 275a 8176 - - -
57 72,II 275b 8177 - - -
58 - - - - - -
59 5,I 244a 8179 - - -
60 5,II 244b 8180 - - -
61 IX p. 294 8178 - - 6707,7
62 XI sub 244 - - 454 6708,13
63 6b pp. 293-4 8181 Fa 2.20b - -
64 6a pp. 293-4 8181 Fa 2.20a - -
65 7 246 8182 - - -
66 14,I p. 294 8567 Fa 2.26 - -
67 14,II p. 294 8577 - - -
68 - - 8569 - - -
69 12,I 258 - - - -
70 12,II 258 - - - -
71 12,IIIa 259a - - - -
72 12,IIIb 259b - - - -
73 13,II - - - - -
74 13,I - - - - -
75 13,III - - - - -
76 13,VI - - - - -
77 13,IV - - - - -
78 13,V - - - - -
79 77 279 8205 - - 3160,I.6
80 79,I 281a 8207 - - 3160,II.1
81 79,II 281b 8208 - - 3160,II.2
82 78 280 8206 - - 3160,I.5
83 81 282 8209 - - 3160,I.2
84 82 283 8210 - - 3160,I.1
85 83 284 8211 - - 3160,I.3
86 84 285 8212 - - 3160,I.4
632
CONCORDANCES
87 80 - - - - -
88 85a 286A 8213a - - -
89 85b 286B 8213b - - -
90 86,Ia 287a,A 8214a - - 3162c,4a
91 86,Ib 287a,B 8214b - - 3162c,4b
92 86,III 287c 8216 - - -
93 86,IV 287d 8217 - - -
94 86,II 287b 8215 - - 3162c,3
95 86,V 287e 8218 - - 3162c,5
96 86,VIII 287g 8221 - - 3162c,2
97 86,VII - 8220 - - -
98 86,XIII 287l 8226 - - -
99 86,IX 287h 8222 - - 3162c,1
100 86,XI - 8224 - - -
101 86,X 287i 8223 - - 3162c,6
102 86,XIV 287m 8227 - - -
103 86,XII 287k 8225 - - -
104 86,XV - 8228 - - -
105 86,VI 287f 8219 - - -
106 86,XVI - 8229 - - -
107 86,XVII - 8230 - - -
108 86,XVIII - 8231 - - -
109 32 261 8031 Fa 2.19 - -
110 34 263 8032 Fa 9.3 - -
111 35 sub 261 8033 - - -
112 33 262 8548 - - -
113 15,I 264a 8036 - - -
114 15,II 264b 8037 - - -
115 15,III 264c 8038 - - -
116 15,IV 264d 8039 - - -
117 15,XII 264m 8047 - - -
118 15,V 264e 8040 - - -
119 15,VI 264f 8041 - - -
120 15,VII 264g 8042 - - -
121 15,VIII 264h 8043 - - -
122 15,X 264k 8044 - - -
123 15,XI 264l 8046 - - -
124 15,IX 264i 8045 - - -
125 16,I 265a 8048 - - -
126 16,II 265b 8049 - - -
127 17 266a 8050 - - -
128 19 266c 8052 - - -
129 20a-b 267 8053 - - -
130 21 268 8054 - - -
131 18 266b 8051 - - -
132 - - - - - -
133 - - 8564 - - -
134 - - 8565 - - -
135 - - - - - -
633
CONCORDANCES
136 94a 293A 8237a - - -

137 94b 293B 8237b - - -
138 115,Ib 311B 8281b - - -
139 115,Ia 311A 8281a - - -
140 112 309 8256 - - -
141 113,IV - 8260 - - -
142 113,V - 8261 - - -
143 113,I - 8257 - - -
144 100a 299A 8243a, 8280 - - -
145 100b 299B 8243b - - -
146 97 296 8240 - - -
147 98 297 8241 - - -
148 106 305 8249 - - -
149 89 289 8232 - - -
150 90 290 8233 - - -
151 92 292 8235 - - -
152 107a 306 8250 - - -
153 110 308 8253 - - -
154 109 307 8252 - - -
155 105 304 8248 - - -
156 - - - - - -
157 - - - - - -
158 99 298 8242 - - -
159 91 291 8234 - - -
160 95 294 8238 - - -
161 96 295 8239 - - -
162 101 300 8244 - - -
163 102 303 8245 - - -
164 103 301 8246 - - -
165 104 302 8247 - - -
166 108 - 8251 - - -
167 113,IIa - 8258a - - -
168 113,IIb - 8258b - - -
169 113,VIII 310 8267 - - -
170 113,IX - 8268 - - -
171 93 - 8236 - - -
172 114 - 8266 - - -
173 111a 312a 8254 - - 7517a
174 111b 312b 8255 - - 7517b
175 - - - - - -
176 113,III - 8259 - - -
177 - - 8262 - - -
178 113,VI - 8263 - - -
179 - - 8263a - - -
180 113,VII - 8264 - - -
181 - - 8265 - - -
182 113,X - 8269 - - -
183 113,XI - 8270 - - -
184 113,XII - 8271 - - -
634
CONCORDANCES
185 - - 8272 - - -
186 - - 8273 - - -
187 - - 8274 - - -
188 - - 8275 - - -
189 - - 8276 - - -
190 - - 8277 - - -
191 - - 8278 - - -
192 - - 8279 - - -
193 115,II - 8282 - - -
194 115,III - 8283 - - -
195 116 313 8285 - - -
196 118,I 315 8287 - - -
197 117 314 8286 - - -
198 118,II 316 8288 - - -
199 36,I 353a 8289 - - -
200 36,II 353b 8290 - - -
201 39 354 8585 Fa 2.27 - -
202 37 - 8291 - - -
203 38,Ia - 8292a - - -
204 38,II - 8295 - - -
205 62,I p.310 8333 - - 3161
206 62,II p.310 8333 - - sub 3161
207 62,III p.310 8333 - - sub 3161
208 62,IV p.310 8333 - - sub 3161
209 62,V p.310 8333 - - sub 3161
210 63 - - - - -
211 119 321 8332 - - 3162a
212 LII - 8428 Fa 0.9 - -
213 61 317 8343 - - 3156a
214 59 320 8340 - 365 3081
215 60 318 8342 - - sub 3081
216 XII - - - 473 6709,26
217 XIVa sub 320 (A) 8341b - 364a 3078a =
7483a
218 XIVb sub 320 (B) 8341a - 364b 3078b =
7483b
219 X - - - 1991 3073
220 121,IV 322d 8347 - - 3159,8
221 121,I 322a 8344 - - 3159,5
222 121,IIa 322b,A 8345a - - 3159,2a
223 121,IIb 322b,B 8345b - - 3159,2b
224 121,Va 322e,A 8348a - - 3159,7a
225 121,Vb 322e,B 8348b - - 3159,7b
226 121,III 322c 8346 - - 3159,1
227 121,VII 322g 8350 - - 3159,4
228 121,VIa 322f 8349a - - 3159,6b
229 121,VIb 322f 8349b - - 3159,6a
230 121,VIII 322h 8351 - - 3159,3
231 121,IX 322i 8352 - 1988 3159
635
CONCORDANCES
232 122a 323A 8353a - 1989 3158

233 122b 323B 8353b - 1989 3158
234 120,I p. 310 8335 - - 7500a
235 120,II p. 310 8336 - - 7500b
236 120,III p. 310 8337 - - 7500c
237 - p. 310 - - - 7501
238 - p. 310 - - - 7502
239 - p. 310 - - - 7503
240 - p. 310 - - - 7504
241 41,I - 8338 - - -
242 - - - - - -
243 - - - - - -
244 - - - - - -
245 - - - - - -
246 - - - - - -
247 - - - - - -
248 - - - - - -
249 - - - - - -
250 - - - - - -
251 XIII p. 310 8334 - 1990 3160
252 41,II - 8354 - - -
253 - - 8383c - - -
254 - - 8383d - - -
255 - - 8383g - - -
256 - - - - - -
257 138 332 8392 - - -
258 139 333 8393 - - -
259 49,I 346 8395 - 2657 8124,18
260 49,II 346 8396 - 2657 8124,18
261 48,I 345a 8394 - - -
262 48,II 345b 8587 - - -
263 42 348 8399 - - -
264 43 348a 8400 Fa 2.13 - -
265 132 334 8387 - - 7513
266 133 335 8388 - - -
267 - - - - - -
268 XIX p. 327 - - 2437 8130,1
269 129 329 8384 Fa 1.1 - -
270 131 331 8386 Fa 1.2 - -
271 130 330 8385 - - -
272 127 328 8378 - - 7515
273 50 349 8592 - - -
274 - - - - - -
275 140,I 337a 8397 - - -
276 140,II 337b 8398 - - -
277 46,Ia 347 8383i,a - - -
278 - - 8383k - - -
279 46,IIa - 8383l,a - - -
280 46,IIb - 8383l,b - - -
636
CONCORDANCES
281 - - 8383m - - -
282 46,III - 8383p - - -
283 - - 8383q,a - - -
284 - - 8383r,b - - -
285 136 342a,2 8391 - - -
286 44,II - - - - -
287 - - 8400a - - -
288 XX - - - - -
289 137 342a,3 8588 Fa 1.5 - -
290 - - - - - -
291 XXI - - - 1992 7505
292 - - 8381b - - -
293 47 344 8381e - - -
294 - - - - - -
295 - - - - 419d 6704,2d
296 - - - - 420a 6704,3a
297 128,Ia - - - - -
298 128,Ib - - - - -
299 128,II - - - - -
300 128,III - - - - -
301 128,IV - - - - -
302 143,I 338a - - - -
303 143,II 338b - - - -
304 51 352 - Fa 2.21 - -
305 144,I 339a - - - -
306 144,IIa 339b,A - - - -
307 144,IIb 339b,B - - - -
308 144,III 339c - - - -
309 144,IV 339d - - - -
310 144,V 339e - - - -
311 144,VI 339f - - - -
312 144,VII 339g - - - -
313 144,VIII 339h - - - -
314 144,IX 339i - - - -
315 144,X 339k - - - -
316 144,XI 340a - - - -
317 144,XIII 340d - - - -
318 144,XIV 340b - - - -
319 144,XII 340c - - - -
320 - - - - - -
321 - - - - - -
322 - - - - - -
323 - - - - - -
324 125 325 8370 - - 7514
325 XVI 327d 8374 - - 7519
326 XVIII 327e 8375 - - 7520
327 XV,i-ii 327a-b 8372-8273 - - 7518-7522
328 XVIIa-b 327c 8376-8377 - - 7521
329 123,I 324a 8357 - - 3162b,6
637
CONCORDANCES
330 123,II 324b 8358 - - 3162b,1

331 123,IV 324d 8360 - - 3162b,7
332 123,III 324c 8359 - - 3162b,5
333 123,V 324e 8361 - - sub 3162b
334 123,VII 324g 8363 - - 3162b,2
335 123,VIII 324h 8364 - - 3162b,3
336 123,VI 324f 8362 - - 3162b,4
337 123,IX 324i 8365 - - sub 3162b
338 141 341a 8401 - - -
339 142 341b 8402 - - -
340 XXIII 341e 8404 - - 7524
341 XXII,i-ii 341c-d 8403a-b - - 7523a-b
342 - - 8405 - - 7526c
343 - - 8406 - - 7526a
344 - - 8407 - - 7525
345 - - 8408 - - 7526b
346 - - - - - -
347 76,I 278a 8198 - - -
348 76,II 278b 8199 - - -
349 76,IV 278d 8201 - - -
350 76,VI 278f 8203 - - -
351 76,V 278e 8202 - - -
352 76,III 278c 8200 - - -
353 76,VII 278g 8204 - - -
354 74 277a 8196 - - -
355 75 277b 8197 - - -
356 65a - 8598 - - -
357 65b - 8599 - - -
358 - - - - - -
359 - - - - - -
360 126 326 8371 - - -
361 - - - - - -
362 124 342 8586 - - -
363 145,I 343b 8430 - - -
364 145,II 343a 8431 - - -
365 145,IVa 343d 8429a - - -
366 145,IVb 343d 8429b - - -
367 52,I 350 8435 - - -
368 52,II 350 8436 - - -
369 52,III 350 8437 - - -
370 52,IV 350 8438 - - -
371 53,II 351a 8440 Fa 2.23 - -
372 53,I 351b 8439 Fa 2.22 - -
373 54,I - 8441 - - -
374 54,II - 8442 - - -
375 54,III - 8443 - - -
376 - - - - - -
377 - 361 - - 2436 7762
378 - - - - - -
638
CONCORDANCES
379 - - - - - -
380 - - - - - -
381 - - - - - -
382 - - - - - -
383 - - - - - -
384 - - - Fa 2.17 - -
385 XXXVI sub 259 - - - -
386 - - 8449 - - -
387 XXIV 356a 8449 - 476,4 6706,4
388 XXV 356b 8450 - 476,5 6706,5
389 XXVI 356c 8451 - 476,6 6706,6
390 XXVIII 358a 8453 - 476,1 6706,1
391 XXIX 358b 8454 - 476,2 6706,2
392 XXX 358c 8455 - 476,3 6706,3
393 - - 8456 - 1987 3961a
394 - - - - 2903c -
395 XXXI,i - 8495 - 2496,1 8124,2
396 XXXI,ii 359a 8496 - 2496,5 8124,6
397 XXXII,i - 8497 - 2496,3 8124,4
398 XXXII,ii - 8498 - 2496,6 8124,7
399 XXXIII - 8500 - 2496,10a 8124,1a
400 - - 8508 - 2496,8 8124,15
401 - - 8511 - - -
402 - - 8514 - - 8124,11
403 XXXIV,i 359c 8515 - 2496,9 sub 8124
404 XXXIV,ii 359d 8516 - 2496,9 -
405 - - 8518 - - 8124,12-14
406 - - 8519 - - 8124,13
407 - - 8520 - - 8124,14
408 - - 8521 - - -
409 - - 8523 - 2496,2 8124,3
410 - - 8524 - - 8124,9
411 - - 8528 - - -
412 - - 8530 - - -
413 XXXV - 8534 - 2496,7 -
414 - - 8532 - - -
415 - - 8533 - - -
416 - - 8535 - - -
417 - - 8538 - - -
418 - - 8539 - - 8124,8
419 - - 8540 - 2496,4 8124,5
420 - - - - 2435 -
421 - - - - 2435 -
422 - - 8541 - - -
423 XXVII 357 8452 - 476,7 6706,7
424 - - 8466 - 476,10.3 6706,10.3
425 - - 8467 - 476,10.7 6706,10.7
426 - - 8471 - - -
427 - - 8478 - 476,10.1 6706,10.1
639
CONCORDANCES
428 - - 8479 - 476,10.6 6706,10.6

429 - - 8480 - 476,10.4 6706,10.4
430 - - - - 2496,10b 8124,1b
431 - - - - 2867 -
432 - - - - 2868 -
433 - - - - 2869 -
434 - - - - 2869b -
435 - - - - 2869a -
436 - - - - 2869c -
437 - - - - 2910 -
438 - - - - 2910a -
439 - - - - 2910b,1 -
440 - - - - 2910b,2 -
441 - - - - 2910b,3 -
442 - - - - 2910b,4 -
443 - - - - 2910b,5 -
444 - - - - 2910b,6 -
445 - - - - 2910b,7 -
446 - - - - 2910b,8 -
447 - - - - 2910b,9 -
448 - - - - 2910b,10 -
449 - - - - 2910b,11 -
450 - - - - 2910b,12 -
451 - - - - 2910b,13 -
452 - - - - 2910b,14 -
453 - - - - 2910b,15 -
454 - - - - 2910b,16 -
455 - - - - 2910b,17 -
456 - - - - 3338b -
457 XXXVII,ii 360e 8458 Fa 2.24 476,9 6706,9
458 XXXVIII,i - 8460 - 476,10.2 6706,10.2
459 XXXVIII,ii - 8461 - 476,8 6706,8
460 XXXVII,i 360f 8457 - 476,10.5 6706,10.5
461 XXXVII,iii - 8459 - 476,10.8 6706,10.8
462 XL 360b 8463 - - -
463 XLI,i 360c 8464 - - -
464 XLI,ii 360d 8465 - - -
465 - - - - - -
466 - - - - - -
467* 56 - - - - -
468* VI 513 - - 2917c -
469* 55 - - - - -
470* - - - - - -
471* - - - - - -
472* - - - - - -
473* - - - - - -
474* 58 319 8339 - - 3157
475* VII 359b - - - -
476* XXXIX 360a 8462 - 476,11 6706,11a
640
CONCORDANCES
477* VIII - - - 424 6704,6

478* - - - - 419e 6704,2e
479† I - - - - -
480† - - - - - -
481† - - - - - -
482† II 363 - - 474 -
483† III 364a - - - -
484† IV 364b - - 455 -
I XLV - 8414 Fa 9.1 - -
II XLV - 8414 Fa 0.1 - -
III XLVIa - 8415a Fa 2.1+6.2 - -
IV XLVIb - 8415b Fa 0.2 - -
V XLVII - 8416 Fa 2.2 - -
VI XLVIII,i - 8417 - - -
VII XLVIII,ii - 8418 - - -
VIII XLIV - 8413 Fa 3.1+6.1 - -
IX XLII - 8411 Fa 3.3 - -
X XLIII - 8412 Fa 0.4 - -
XI L - 8419 Fa 2.6 - -
XII L - 8420 Fa 2.7 - -
XIII L - 8421 Fa 2.8 - -
XIV L - 8422 Fa 2.9 - -
XV L - 8423 Fa 2.10 - -
XVI LIV - - Fa 2.12 - -
XVII XLIX - - Fa 2.14 - -
XVIII - - - - - -
XIX LI - 8426 Fa 3.2 - -
XX - - - Fa 1.6 - -
XXI - - - Fa 2.4 - -
XXII - - - Fa 2.11a - -
XXIII - - - Fa 2.11b - -
XXIV - - - Fa 0.5 - -
XXV - - - Fa G.1a-b - -
XXVI - - - - - -
XXVII - - - Fa S.1 - -
XXVIII - - - Fa S.2 - -
XXIX 57 - 8029 Fa 0.6 - -
XXX - - - - - -
XXXI - - - - - -
XXXII 45 sub 330 8382 Fa 2.15 - -
XXXIII - - - Fa S.3 - -
XXXIV 134 336 8389 Fa 1.3 - -
XXXV 135 342a,1 8390 Fa 1.4 - -
XXXVI - - - - - -
XXXVII - - - - - -
XXXVIII - - - Fa 0.8 - -
XXXIX 64 - 8379 Fa 0.7 - -
XL 44,I 355 - Fa 2.16 - -
XLI - - sub 8412 Fa S.4 - -
641
CONCORDANCES
XLII - - - Fa 3.4 - -
XLIII 145,III 343c 8432 Fa 1.7 - -
XLIV - p. 329 8547 Fa 9.2 - -
XLV - - - Fa 2.25 - -
XLVI - - - Fa 2.3 - -
XLVII - - - Fa 6.3 - -
XLVIII - - - - - -
XLIX LIII - - Fa 2.5 - -
L - - - Fa 7.1a - -
LI - - - Fa 7.1b - -
2. GIACOMELLI, LA LINGUA FALISCA
Giacomelli Bakkum Giacomelli Bakkum Giacomelli Bakkum
1 1 15,VII 120 33 112

2a 2 15,VIII 121 34 110
2b 3 15,IX 124 35 111
3 4 15,X 122 36,I 199
4a 6 15,XI 123 36,II 200
4b 7 15,XII 117 37 202
5,I 59 16,I 125 38,Ia 203
5,II 60 16,II 126 38,II 204
6a 64 17 127 38,III 30
6b 63 18 131 39 201
7 65 19 128 41,I 241
8 44 20a-b 129 41,II 252
9 45 21 130 42 263
10 46 22 5 43 264
11a 9 23 20 44,I XL
11b 8 24,I 22 44,II 286
12,I 69 24,II 23 45 XXXII
12,II 70 24,III 24 46,Ia 277
12,IIIa 71 24,IV 25 46,IIa 279
12,IIIb 72 24,V 26 46,IIb 280
13,I 74 24,VI 27 46,III 282
13,II 73 25,I 31 47 293
13,III 75 25,II 32 48,I 261
13,IV 77 26 29 48,II 262
13,V 78 27 28 49,I 259
13,VI 76 28,I 34 49,II 260
14,I 66 28,II 35 50 273
14,II 67 28,III 36 51 304
15,I 113 29,I 37 52,I 367
15,II 114 29,II 38 52,II 368
15,III 115 29,III 21 52,III 369
15,IV 116 30 33 52,IV 370
15,V 118 31 10 53,I 372
15,VI 119 32 109 53,II 371
642
CONCORDANCES
54,I 373 76,VII 353 103 164

54,II 374 77 79 104 165
54,III 375 78 82 105 155
55 469* 79,I 80 106 148
56 467* 79,II 81 107a 152
57 XXIX 80 87 108 166
58 474* 81 83 109 154
59 214 82 84 110 153
60 215 83 85 111a 173
61 213 84 86 111b 174
62,I 205 85a 88 112 140
62,II 206 85b 89 113,I 143
62,III 207 86,Ia 90 113,IIa 167
62,IV 208 86,Ib 91 113,IIb 168
62,V 209 86,II 94 113,III 176
63 210 86,III 92 113,IV 141
64 XXXIX 86,IV 93 113,V 142
65a 356 86,V 95 113,VI 178
65b 357 86,VI 105 113,VII 180
66,I 15 86,VII 97 113,VIII 169
66,II 16 86,VIII 96 113,IX 170
66,III 17 86,IX 99 113,X 182
66,IV 19 86,X 101 113,XI 183
66,V 18 86,XI 100 113,XII 184
67 14 86,XII 103 114 172
68,I 11 86,XIII 98 115,Ia 139
68,IIa-b 12 86,XIV 102 115,Ib 138
69 13 86,XV 104 115,II 193
70 47 86,XVI 106 115,III 194
71,I 48 86,XVII 107 116 195
71,II 49 86,XVIII 108 117 197
71,III 50 87 39 118,I 196
71,IV 51 88 40 118,II 198
71,V 54 89 149 119 211
71,VI 52 90 150 120,I 234
71,VII 53 91 159 120,II 235
72,I 56 92 151 120,III 236
72,II 57 93 171 121,I 221
73,I 41 94a 136 121,IIa 222
73,II 42 94b 137 121,IIb 223
73,III 43 95 160 121,III 226
74 354 96 161 121,IV 220
75 355 97 146 121,Va 224
76,I 347 98 147 121,Vb 225
76,II 348 99 158 121,VIa 228
76,III 352 100a 144 121,VIb 229
76,IV 349 100b 145 121,VII 227
76,V 351 101 162 121,VIII 230
76,VI 350 102 163 121,IX 231
643
CONCORDANCES
122a 232 144,IV 309 XXV 388

122b 233 144,V 310 XXVI 389
123,I 329 144,VI 311 XXVII 422
123,II 330 144,VII 312 XXVIII 390
123,III 332 144,VIII 313 XXX 392
123,IV 331 144,IX 314 XXXI,i 395
123,V 333 144,X 315 XXXI,ii 396
123,VI 336 144,XI 316 XXXII,i 397
123,VII 334 144,XII 319 XXXII,ii 398
123,VIII 335 144,XIII 317 XXXIII 399
123,IX 337 144,XIV 318 XXXIV,i 403
124 362 145,I 363 XXXIV,ii 404
125 324 145,II 364 XXXV 413
126 360 145,III XLIII XXXVI 385
127 272 145,IVa 365 XXXVII,i 460
128,Ia 297 145,IVb 366 XXXVII,ii 457
128,Ib 298 I 479† XXXVII,iii 461
128,II 299 II 482† XXXVIII,i 458
128,III 300 III 483† XXXVIII,ii 459
128,IV 301 IV 484† XXXIX 476*
129 269 VI 468* XL 462
130 271 VII 475* XLI,i 463
131 270 VIII 477* XLI,ii 464
132 265 IX 61 XLII IX
133 266 X 219 XLIII X
134 XXXIV XI 62 XLIV VIII
135 XXXV XII 216 XLIX XVII
136 285 XIII 251 XLV II
137 289 XIVa 217 XLV I
138 257 XIVb 218 XLVIa III
139 258 XV,i-ii 327 XLVIb IV
140,I 275 XVI 325 XLVII V
140,II 276 XVIIa-b 328 XLVIII,i VI
141 338 XVIII 326 XLVIII,ii VII
142 339 XIX 268 L XI-XV
143,I 302 XX 288 LI XIX
143,II 303 XXI 291 LII 212
144,I 305 XXII,i-ii 341 LIII XLIX
144,IIa 306 XXIII 340 LIV XVI
144,IIb 307 XXIV 387
144,III 308 XXIX 391
3. VETTER, HANDBUCH DER ITALISCHEN DIALEKTE
Vetter Bakkum Vetter Bakkum Vetter Bakkum
pp.293-4 63 p.294 61 p.310 234-240

pp.293-4 64 p.294 66 p.310 251
p.294 21 p.294 67 p.310 205-209
644
CONCORDANCES
p.327 268 265b 126 287b 94

p.329 XLIV 266a 127 287c 92
241 1 266b 131 287d 93
242A 2 266c 128 287e 95
242B 3 267 129 287f 105
243 4 268 130 287g 96
244a 59 269a 15 287h 99
244b 60 sub 269a 16 287i 101
sub 244 62 269b 17 287k 103
245a 7 269c 18 287l 98
245b 6 sub 269c 19 287m 102
246 65 270 14 288 39
247 44 271a 12 289 149
248 45 271b 11 290 150
249 46 272 13 291 159
250 5 273 47 292 151
251 20 274a 48 293A 136
252 22 274b 49 293B 137
252 23-27 274c 50 294 160
253 31-32 274d 51 295 161
254a 29 274e 54 296 146
254b 28 274f 52 297 147
255 34-36 274g 53 298 158
256a 37 275a 56 299A 144
256b 33 275b 57 299B 145
256c 38 276a 41 300 162
257 8-9 276b 42 301 164
258 69-70 276c 43 302 165
259a 71 277a 354 303 163
259b 72 277b 355 304 155
sub 259 385 278a 347 305 148
260 10 278b 348 306 152
261 109 278c 352 307 154
sub 261 111 278d 349 308 153
262 112 278e 351 309 140
263 110 278f 350 310 169
264a 113 278g 353 311A 139
264b 114 279 79 311B 138
264c 115 280 82 312a 173
264d 116 281a 80 312b 174
264e 118 281b 81 313 195
264f 119 282 83 314 197
264g 120 283 84 315 196
264h 121 284 85 316 198
264i 124 285 86 317 213
264k 122 286A 88 318 215
264l 123 286B 89 319 471*
264m 117 287a,A 90 320 214
265a 125 287a,B 91 sub 320 (A) 217
645
CONCORDANCES
sub 320 (B) 218 333 258 345a 261

321 211 334 265 345b 262
322a 221 335 266 346 259-260
322b,A 222 336 XXXIV 347 277
322b,B 223 337a 275 348 263
322c 226 337b 276 348a 264
322d 220 338a 302 349 273
322e,A 224 338b 303 350 367-370
322e,B 225 339a 305 351a 371
322f 228 339b,A 306 351b 372
322f 229 339b,B 307 352 304
322g 227 339c 308 353a 199
322h 230 339d 309 353b 200
322i 231 339e 310 354 201
323A 232 339f 311 355 XL
323B 233 339g 312 356a 387
324a 329 339h 313 356b 388
324b 330 339i 314 356c 389
324c 332 339k 315 357 423
324d 331 340a 316 358a 390
324e 333 340b 318 358b 391
324f 336 340c 319 358c 392
324g 334 340d 317 359a 396
324h 335 341a 338 359b 475*
324i 337 341b 339 359c 403
325 324 341c-d 341 359d 404
326 360 341e 340 360a 476*
327a-b 327 342 362 360b 462
327c 328 342a,1 XXXV 360c 463
327d 325 342a,2 285 360d 464
327e 326 342a,3 289 360e 457
328 272 343a 364 360f 460
329 269 343b 363 361 377
330 271 343c XLIII 363 482†
sub 330 XXXII 343d 365 364a 483†
331 270 343d 366 364b 484†
332 257 344 293 513 468*
4. HERBIG, CORPUS INSCRIPTIONUM ETRUSCARUM II.1.2
CIE Bakkum CIE Bakkum CIE Bakkum
8001 5 8007 26 8014 34

8002 20 8008 27 8015 35
8003 22 8010 31 8016 36
8004 23 8011 32 8017 37
8005 24 8012 29 8018 33
8006 25 8013 28 8019 38
646
CONCORDANCES
8020 21 8179 59 8232 149

8029 XXIX 8180 60 8233 150
8030 10 8181 63 8234 159
8031 109 8181 64 8235 151
8032 110 8182 65 8236 171
8033 111 8190 41 8237a 136
8036 113 8191 42 8237b 137
8037 114 8192 43 8238 160
8038 115 8193 44 8239 161
8039 116 8194 45 8240 146
8040 118 8195 46 8241 147
8041 119 8196 354 8242 158
8042 120 8197 355 8243b 145
8043 121 8198 347 8244 162
8044 122 8199 348 8245 163
8045 124 8200 352 8246 164
8046 123 8201 349 8247 165
8047 117 8202 351 8248 155
8048 125 8203 350 8249 148
8049 126 8204 353 8250 152
8050 127 8205 79 8251 166
8051 131 8206 82 8252 154
8052 128 8207 80 8253 153
8053 129 8208 81 8254 173
8054 130 8209 83 8255 174
8070 15 8210 84 8256 140
8071 16 8211 85 8257 143
8072 17 8212 86 8258a 167
8073 19 8213a 88 8258b 168
8074 18 8213b 89 8259 176
8075 14 8214a 90 8260 141
8076 12 8214b 91 8261 142
8077 11 8215 94 8262 177
8078 13 8216 92 8263 178
8079 1 8217 93 8263a 179
8163 6 8218 95 8264 180
8163 7 8219 105 8265 181
8164 47 8220 97 8266 172
8167 48 8221 96 8267 169
8168 49 8222 99 8268 170
8169 50 8223 101 8269 182
8170 51 8224 100 8270 183
8171 54 8225 103 8271 184
8172 52 8226 98 8272 185
8173 53 8227 102 8273 186
8175 55 8228 104 8274 187
8176 56 8229 106 8275 188
8177 57 8230 107 8276 189
8178 61 8231 108 8277 190
647
CONCORDANCES
8278 191 8357 329 8402 339

8279 192 8358 330 8403a-b 341
8281b 138 8359 332 8404 340
8281a 139 8360 331 8405 342
8282 193 8361 333 8406 343
8283 194 8362 336 8407 344
8285 195 8363 334 8408 345
8286 197 8364 335 8411 IX
8287 196 8365 337 8412 X
8288 198 8370 324 sub 8412 XLI
8289 199 8371 360 8413 VIII
8290 200 8374 325 8414 I
8291 202 8375 326 8414 II
8292a 203 8378 272 8415a III
8294 30 8379 XXXIX 8415b IV
8295 204 8381b 292 8416 V
8332 211 8381e 293 8417 VI
8333 205 8382 XXXII 8418 VII
8333 206 8383c 253 8419 XI
8333 207 8383d 254 8420 XII
8333 208 8383g 255 8421 XIII
8333 209 8383i,a 277 8422 XIV
8334 251 8383k 278 8423 XV
8335 234 8383l,a 279 8426 XIX
8336 235 8383l,b 280 8428 212
8337 236 8383m 281 8429a 365
8338 241 8383p 282 8429b 366
8339 471* 8383q,a 283 8430 363
8340 214 8383r,b 284 8431 364
8341b 217 8384 269 8432 XLIII
8341a 218 8385 271 8435 367
8342 215 8386 270 8436 368
8343 213 8387 265 8437 369
8344 221 8388 266 8438 370
8345a 222 8389 XXXIV 8439 372
8345b 223 8390 XXXV 8440 371
8346 226 8391 285 8441 373
8347 220 8392 257 8442 374
8348a 224 8393 258 8443 375
8348b 225 8394 261 8449 386
8349a 228 8395 259 8449 387
8349b 229 8396 260 8450 388
8350 227 8397 275 8451 389
8351 230 8398 276 8452 423
8352 231 8399 263 8453 390
8353a 232 8400 264 8454 391
8353b 233 8400a 287 8455 392
8354 252 8401 338 8456 393
648
CONCORDANCES
8457 460 8508 400 8541 422

8458 457 8511 401 8547 XLIV
8459 461 8514 402 8548 112
8460 458 8515 403 8564 133
8461 459 8516 404 8565 134
8462 476* 8518 405 8566 39
8463 462 8519 406 8567 66
8464 463 8520 407 8569 6
8465 464 8521 408 8577 67
8466 424 8523 409 8585 201
8467 425 8524 410 8586 362
8471 426 8528 411 8587 262
8478 427 8530 412 8588 289
8479 428 8532 414 8592 273
8480 429 8533 415 8598 356
8495 395 8534 413 8599 357
8496 396 8535 416 8243a, 8280 144
8497 397 8538 417 8372-8273 327
8498 398 8539 418 8376-8377 328
8500 399 8540 419
5. RIX, ETRUSKISCHE TEXTE
ET Bakkum ET Bakkum ET Bakkum
Fa 1.1 269 Fa 2.20a 64 Fa 6.3 XLVII

Fa 1.2 270 Fa 2.20b 63 Fa 7.1a L
Fa 1.3 XXXIV Fa 2.21 304 Fa 7.1b LI
Fa 1.4 XXXV Fa 2.22 372 Fa 9.1 I
Fa 1.5 289 Fa 2.23 371 Fa 9.2 XLIV
Fa 1.6 XX Fa 2.24 457 Fa 9.3 110
Fa 1.7 XLIII Fa 2.25 XLV Fa 0.1 II
Fa 2.1+6.2 III Fa 2.26 66 Fa 0.2 IV
Fa 2.10 XV Fa 2.27 201 Fa 0.3 5
Fa 2.11a XXII Fa 2.3 XLVI Fa 0.4 X
Fa 2.11b XXIII Fa 2.4 XXI Fa 0.5 XXIV
Fa 2.12 XVI Fa 2.5 XLIX Fa 0.6 XXIX
Fa 2.13 264 Fa 2.6 XI Fa 0.7 XXXIX
Fa 2.14 XVII Fa 2.7 XII Fa 0.8 XXXVIII
Fa 2.15 XXXII Fa 2.8 XIII Fa 0.9 212
Fa 2.16 XL Fa 2.9 XIV Fa G.1a-b XXV
Fa 2.17 384 Fa 3.1+6.1 VIII Fa S.1 XXVII
Fa 2.18 37 Fa 3.2 XIX Fa S.2 XXVIII
Fa 2.19 109 Fa 3.3 IX Fa S.3 XXXIII
Fa 2.2 V Fa 3.4 XLII Fa S.4 XLI
649
CONCORDANCES
6. CORPUS INSCRIPTIONUM LATINARUM I2
CIL I2 Bakkum CIL I2 Bakkum CIL I2 Bakkum
364a 217 1988 231 2903c 394

364b 218 1989 232 2910 437
365 214 1989 233 2910a 438
419d 295 1990 251 2910b,1 439
419e 478* 1991 219 2910b,10 448
420a 296 1992 291 2910b,11 449
424 477* 2435 421 2910b,12 450
454 62 2436 377 2910b,13 451
455 484† 2437 268 2910b,14 452
473 216 2496,1 395 2910b,15 453
474 482† 2496,10a 399 2910b,16 454
476,1 390 2496,10b 430 2910b,17 455
476,11 476* 2496,2 409 2910b,2 440
476,2 391 2496,3 397 2910b,3 441
476,3 392 2496,4 419 2910b,4 442
476,4 387 2496,5 396 2910b,5 443
476,5 388 2496,6 398 2910b,6 444
476,6 389 2496,7 413 2910b,7 445
476,7 423 2496,8 400 2910b,8 446
476,8 459 2496,9 403 2910b,9 447
476,9 457 2496,9 404 2912 10
476,10.1 427 2657 259 2917c 468*
476,10.2 458 2657 260 3338b 456
476,10.3 424 2867 431 2910b,14 453
476,10.4 429 2868 432 2910b,15 454
476,10.5 460 2869 433 2910b,16 455
476,10.6 428 2869a 435 2910b,17 456
476,10.7 425 2869b 434 2912 10
476,10.8 461 2869c 436 2917c 469*
1987 393 2903a 420 3338b 457
7. CORPUS INSCRIPTIONUM LATINARUM XI
CIL XI Bakkum CIL XI Bakkum CIL XI Bakkum
3073 220 3159,3 231 3160,I.4 86

3078a 218 3159,4 228 3160,I.5 82
3078b 219 3159,5 222 3160,I.6 79
3081 214 3159,6a 230 3160,II.1 78
sub 3081 216 3159,6b 229 3160,II.2 81
3156a 213 3159,7a 225 3161 205
3157 215 3159,7b 226 sub 3161 206-209
3158 233-234 3159,8 221 3162a 211
3159 232 3160 253 3162b,1 332
3159,1 227 3160,I.1 84 3162b,2 336
3159,2a 223 3160,I.2 83 3162b,3 337
3159,2b 224 3160,I.3 85 3162b,4 338
650
CONCORDANCES
CIL XI Bakkum CIL XI Bakkum CIL XI Bakkum
3162b,5 334 6706,10.4 430 7522 329

3162b,6 331 6706,10.5 461 7523a 343
3162b,7 333 6706,10.6 429 7523b 343
sub 3162b 335 6706,10.7 426 7524 342
sub 3162b 339 6706,10.8 462 7525 346
3162c,1 99 6706,11a 472* 7526a 345
3162c,2 96 6707,7 61 7526b 347
3162c,3 94 6708,13 62 7526c 344
3162c,4a 90 6709,26 217 7762 378
3162c,4b 91 7483a 218 8124,1a 400
3162c,5 95 7483b 219 8124,1b 431
3162c,6 101 7500a 235 8124,2 396
3961a 394 7500b 236 8124,3 410
6704,2d 297 7500c 237 8124,4 398
6704,2e 363 7501 238 8124,5 421
6704,3a 298 7502 239 8124,6 397
6704,6 473* 7503 240 8124,7 399
6706,1 391 7504 241 8124,8 419
6706,2 392 7505 293 8124,9 411
6706,3 393 7513 266 8124,11 403
6706,4 387 7514 326 8124,12 406
6706,5 389 7515 273 8124,13 407
6706,6 390 7516 14 8124,14 408
6706,7 424 7517a 173 8124,15 401
6706,8 460 7517b 174 8124,18 260-261
6706,9 458 7518 329 sub 8124 404
6706,10.1 428 7519 327 8130,1 269
6706,10.2 459 7520 328
6706,10.3 425 7521 330
8. FIRST EDITIONS OF THE INSCRIPTIONS PUBLISHED SINCE 1963
The following list contains the first editions of the inscriptions first published after the
appearance of Giacomelli’s La lingua falisca (1963). Note that in some cases the inscribed
objects themselves had been published previously, either without mentioning the inscription or
without giving the text.
58 Schippa 1980:48, nr.50 247 Renzetti Marra 1990:332-4, T 20

132 Fortunati, Sant p.112 248 Renzetti Marra 1990:332-4, T VIb
135 Colonna 1972:446, sub nr.56 249 Renzetti Marra 1990:333-4, T 22
156 Colonna 1972:446-7, nr.57 250 Pulcini 1974:138
157 Colonna 1972:446-7, nr.58 256 FI II.2, p.254
175 unpublished 267 FI II.2, p.300
242 Renzetti Marra 1990:333-4, 290 FI II.1, p.45
339-40, P Iabcd 294 FI II.2, p.321
243 Renzetti Marra 1990:333, P 9 346 FI II.1, pp.67-8
244 Renzetti Marra 1990:338, P 12 358 R. Giacomelli 1977:63-9
245 Renzetti Marra 1990:338, P 15 359 R. Giacomelli 1977:63-9
246 Renzetti Marra 1990:339, P 19 361 Renzetti Marra 1990:336-7, B 1
651
CONCORDANCES
376 Colonna 1976b 450 Moretti 1975:159, nr.17

378 Torelli 1967:536, nr.1 451 Moretti 1975:159, nr.18
379 Torelli 1967:536-7, nr.2 452 Moretti 1975:159, nr.19
382 Torelli 1967:537-8, nr.5 455 Moretti 1975:160, nr.22
383 Torelli 1967:538, nr.1 456 Moretti 1975:104-5, nr.141
384 Torelli 1967:538, nr.2 470* Wallace 2004
394 Briquel 1972:825 471* Olmos Romera 2003
434 Moretti 1975:175, nr. 44 472* Stanco 2001
435 Moretti 1975:173-4, nr.40 473* Tajan 2002, p.59
436 Moretti 1975:152, nr.156 480† Colonna 1983
438 Moretti 1975:152, nr.154 481† Firmani 1977:116
439 Moretti 1975:158, nr.7
440 Moretti 1975:158, nr.8 XVIII CVA Louvre 23, p.21
441 Moretti 1975:158, nr.9 XX Polidori 1977
442 Moretti 1975:158, nr.10 XXIV Beranger & Fortini 1978
443 Moretti 1975:158, nr.11 XXVI Gulinelli 1995a:319-20
444 Moretti 1975:158, nr.12 XXX Michetti 2003
445 Moretti 1975:159, nr.13 XXXVI Colonna 1990:118-20
446 Moretti 1975:159, nr.14 XXXVII Gulinelli 1995a:320-1
447 Moretti 1975:159, nr.15 XXXVIII Dini e.a 1985:69
448 Degrassi & Krummrey CIL sub XXXIX Colonna 1997
I2.2910b,10 XLV Moretti 1975:145-7, nr.142
449 Moretti 1975:159, nr.16 XLVIII Naso 1993
652
Indices
1. LITERARY SOURCES
Ammianus Donatus Hieronymus

23.5.20 – 21 n.11 Vita 17 – (52), 134 n.68 Ep. 57.3 – 39
Andronicus Ennius Homer
2L – 132 41V – 120 X 317 – 134
3L – 132 126V – 134 d 782 – 212 n.113
5L – 132 147V – 119 n.61 q 53 – 212 n.113
9L - 212 236-7V – 132
Justin
14L – 132 273V – 118
20.1.13 – 35
16L - 132 338V – 132
39.3.3.2 – 531
27L – 120
Eutropius
29L – 173; 402 Liber Coloniarum
1.20 – 39
30L – 132 217.5 – 21, 43
1.20.1-2 – 21 n.11
34L - 132
2.28 – 21 n.11, 41, 42, Liber de Praenominibus
Ausonius 45, 47 1 – 241 n.124
Cent. 11 - 259
Festus Livius Andronicus, see
Ep. 6.42 Prete - 134
112.24-5L – 190 Andronicus
Avitus 132.3-7L – 190
fr.2 apud Prisc. CGL 304.33-306.2L – 210 Livy
2.427.2 – 20, 39 306.4-6L – 34 2.16 - 214
4.16 – 21 n.12
341.35-343.1L – 74
Cato 4.17-34 – 38
410.6-9L – 211-2
Agr. 2.1 - 120 4.17.1-11 – 38
410.12-5L – 34
Agr. 4.1 – 27, 45 4.17.11-18.8 – 38
420.26ff – 195
Agr. 14.1 – 27, 45 4.21.1-2 – 38
472.15-6L – 212
apud Plin. NH 3.5.51 – 4.21.6-22.6 – 38
476.11-2L – 244
35 4.21.8 – 38
Florus 4.23.4-24.2 – 38, 321
Cicero
Epit. 1.6/12.5-6 – 39 4.25.7-8 – 38, 321
Div. 1.36.79 – 33 n.20
4.32.3 – 38
Div. 1.47.105 – 507 FronTInus
4.30.5-34.7 – 38
Dom. 101 Str. 2.4.18 – 40
4.31.1-34.7 – 38
Leg. agr. 2.25.66 – 45, Str. 2.5.9 – 40 n.24
4.47.1-7 – 38 n.23
48 Str. 2.8.3 – 38
5.8.4-12 – 38
Or. 153 Str. 2.8.8 – 38 n.23
5.8.5 – 38
Str. 4.4.1 – 39
Diodorus 5.10.2 – 39
14.96.5 – 20, 24, 39 Gellius 5.12.5 – 39
14.98.5 – 39 4.16 – 144 5.13.9-13 – 38
16.31.7 – 39 2.21.7 – 210 n.112 5.14.7 – 39
16.36.4 – 39 10.3.5 – 212 5.16.2 – 39
5.17.6-10 – 39, 321
Dionysius Grattius
5.19.7-8 – 39
1.21.1 – 22, 27, 30, 35 40 – 45
5.21 – 39
1.21.1-2 – 20, 24, 43
Hesychius 5.24.3 – 39
1.21.2 – 28, 32, 35
k 4788 Latte – 186 5.26.3-10 – 39
3.43.3 – 238
r 58 Latte/Hansen – 272, 5.27 – 39
12.4.6 – 21 n.12
409 5.43.7 – 39
9.5.4 – 29
5.58 – 24
13.1-2 – 39
653
INDICES
Livy (continued) Ovid (continued) Pliny (continued)

6.2.2 – 39 Am. 3.13.31-5 – 211 NH 3.5.51 – 21, 22, 24,
6.3.1-10 – 39 Ep. 4.4.32 – 45, 47 35, 43, 211
6.4.4 – 39 Ep. 4.8.41 – 45 NH 3.59 – 82
6.7.4 – 39 Ep. 7.162 – 182 n.97 NH 7.2.19 – 20, 33, 208,
6.9.3-4 – 39 Fast. 1.84 – 45, 47 265
6.9.4 – 39 Fast. 3.87-90 – 34 NH 21.2.3 – 212
6.9.7-10.6 – 39 Fast. 3.383-4 – 204
Plutarch
7.12 – 39 Fast. 3.843-4 – 34
Cam. 2.9-11 – 21, 39
7.16.7-8 – 40 Fast. 4.73-4 – 35, 82,
Fab. Max. 2 – 44
7.17.2-5 – 40 210-1
Fab. Max. 2.3 – 34
7.17.6-9 – 39 Fast. 6.49 – 32
Quaest. 54 – 385
7.22.5-6 – 40 Fast. 3.843 – 32 n.19
7.38.1 – 40 Met. 14.328-30 – 206,
[Plutarch]
9.32.1-12 – 40 263
Vit. Min. 35
9.36 – 40, 322 (with Tr. 3.3.76 – 182 n.97
n.167) Polyaenetus
Pacuvius
9.36.1 – 19 Strat. 8.7.1 – 21, 39
42R – 172
` 9.36.6 – 19 n.7
228R – 172 Polybius
10.12.7 – 40
1.65.2 – 41
10.26.5 – 40 Paulus Diaconus
10.45.6 – 40 Fest. 8.14L – 163 Porphyrio
10.45.10-15 – 40 Fest. 56.26L – 182 in Hor. Carm. 1.9.1 –
22.1.11 – 34, 44 Fest. 58.17L – 89, 180 20, 196
39.8-19 – 236 n.119 Fest. 73.19-22L – 82
39.17.6 – 34, 45 Ptolemy
Fest. 74.1-2L – 82
Per. 20 – 41, 42 3.1.43 Cuntz – 20, 24
Fest. 74.14L – 82
Per. 120 – 218 Fest. 74.9-10L – 82 Quintilian
Fest. 91.6L – 82 Inst. 1.7.18 – 110
Lucilius
Fest. 91.14L – 82 Inst. 8.6.33 – 134
25M – 134
Fest. 91.15L – 118 n.60
581M – 46 Servius
Fest. 91.17-8L – 82
Macrobius Fest. 91.25L – 82 CGL 4.465.5 – 28
1.3.13 – 190, 212 n.114 Fest. 93.25-6L – 37 in Verg. A. 7.607 – 24,
1.9.13 – 34 n.22, 211 34
Fest. 108.5-6L – 105 in Verg. A. 7.695 – 21
Marius Victorinus n.11, 21 n.12, 28, 30, 34,
Fest. 109.4-7L – 190
CGL 6.17.20 – 126 35, 82, 208, 210-1
Fest. 133.10-2L – 190
CGL 6.32.4 – 110 in Verg. A. 11.785 – 33,
Fest. 151.6-7L - 190
Fest. 235.16-7L – 37 37, 37 n.22, 211
Martial
n.22 in Verg. A. 11.787 – 33,
4.46.8 – 45
Fest. 247.24L – 118 37 n.22
Naevius n.60 in Verg. E. 3.1 – 134
Poen. 25.1S – 118 Fest. 305.10-2L – 210 Silius
Orosius Fest. 342.13 – 74 4.223 – 45
3.3.4 – 39 Fest. 473.4-5L – 212 8.489 – 21
4.11.10 – 41, 47 Plautus Solinus
4.15.1 – 44 Epid. 450 – 88 2.7 – 20, 35
Ovid Men. 144 – 203 2.26 – 33
Am. 1.8.108 – 182 n.97 Men. 289-90 – 195
Rud. 478 – 134 Statius
Am. 3.13 – 28, 32, 43
Silv. 4.9.35 – 45
Am. 3.13.1 – 21 Pliny
Am. 3.13.5-6 – 1 NH 2.95.208 – 33 n.20 Stephanus
Am. 3.13.14 – 45, 47 NH 2.106.230 – 45, 47 Ethn. 656.23-4 Meineke
Am. 3.13.32 – 35 NH 2.142 – 244 - 35
654
INDICES
Stephanus (continued) Terentius Scaurus Varro (continued)

Ethn. 656.24-5 Meineke CGL 7.28.11 – 61, 138 L 7.26 – 61
– 20, 21 n.74 L 7.97 – 244, 249
Ethn. 656.12-3 Meineke CGL 7.13.8 – 82, 210 L 9.57 – 162
– 20, 21, 24 CGL 7.13.8ff – 210 R 2 praef. 4 – 531
CGL 7.16.5-6 – 110 R 3.16.10-11 – 45, 207,
Strabo
275; 490
5.2.9 – 14-15, 20, 21, Tertullian
24, 45 Apol. 24 - 32 Velius Longus
5.4.1 – 37 n.22 CGL 7.69.6-10 – 82, 163
Valerius Maximus
5.4.12 – 37 n.22, 211 CGL 7.69.10 - 210
2.8.1 – 42, 47
Suetonius 6.3.1 – 21 n.12 Vergil
Aug. 87-8 – 46 6.5.1 – 39, 41, 42 A. 7.716 – 206, 263
Vesp. 22 – 106 A. 7.695 – 21
Varro
A. 11.785-9 – 33
Scaurus, see Terentius L 5.8 - 240
E. 3.1 – 134 n.68
Scaurus L 5.79 – 82
L 5.22.111 – 45 De Viris Illustribus
Tabula Peutingeriana
L 5.74 – 163 5.4 – 241 n.124
segm. V – 21, 21 n.10
L 5.81 – 197-8, 251
Zonaras
Terence L 5.85 – 462
7.22 – 39
Eun. 321 – 134 L 5.157 – 21 n.12
8.7.4-8 - 43
Haut. 287 – 244 L 5.159 – 180
8.18.1 – 29, 41, 42
L 5.162 – 46 n.29, 210
Terentianus Maurus L 6.1 – 105
CGL 6.385.1992 – 28 L 6.2.4 – 190, 212 n.114
2. INSCRIPTIONS IN THIS EDITION

(Bold page numbers indicate the main discussion of the inscription in the chapters 12-19.)
1 (EF) – 13 n.4, 16, 34, 36, 408, 418, 439, 578, 579, 11 with n.207, 590, 599
58, 59, 60, 61, 62, 64, 584, 590, 607
4 (EF) – 16, 36, 59, 60, 63,
70, 71, 72, 74, 75, 76,
2 (EF) – 16, 36, 60, 64, 77, 73, 74, 86, 89, 91
77, 85, 89, 90, 91
89, 91 fig.3.5, 92, 96, fig.3.5, 105, 112, 113,
fig.3.5, 92, 93, 96, 103
121, 127, 140, 157, 158, 124, 127, 128, 150, 156,
n.50, 105, 113, 118, 122,
170, 180, 193, 196, 198, 158, 159, 170-1, 172,
123, 124, 126, 127, 128,
200, 291, 294, 298, 302, 195, 196, 197, 199, 207,
142, 148, 155, 156, 157,
303 n.152, 323, 343; 217, 218, 219, 220, 241,
158, 159, 160, 162, 165,
362, 365, 373, 375, 376, 291, 293, 294, 298, 299,
167-8, 169, 170, 172,
381, 382 fig.11.1, 393, 304, 323, 343, 350, 351,
173, 174, 175, 176, 180,
401, 406-11 with n.207, 353; 362, 365, 373, 375,
181, 183, 184, 185, 186,
415, 584, 590 376, 381, 382 fig.11.1,
187-8, 188, 190-1, 192,
393, 399, 406-411 with
194, 198, 199, 199-200, 3 (EF) – 16, 36, 58, 62, 63,
fig.12.2, 522, 590
203, 216-7, 217, 218-9, 76, 88, 89 n.46, 90, 93,
219, 250, 291, 292, 293, 129, 137, 141, 142, 145, 5 (EF/Etr) – 36, 89, 91
294, 298, 299, 299-300, 148, 155, 156, 157, 159, fig.3.5, 123, 184, 323,
301, 301 n.151, 303, 170-1, 184, 186, 190-1, 343; 362, 365, 373
309, 323, 326, 343, 348, 195, 203, 216, 218, 220, n.182, 376, 381, 382
349, 350, 352, 354, 355; 251-2, 291, 294, 298, fig.11.1, 393, 414-5, 600
362, 365, 372, 373, 374, 301, 303 n.152, 306,
6 (EF) – 36, 58, 93, 129,
375, 376, 381, 382 323, 327, 343; 362, 365,
140, 143, 148, 184, 216,
fig.11.1, 387, 393, 393- 372, 373, 375, 376, 381,
406 with figs.12.1-3, 219, 245, 306, 323, 327,
382 fig.11.1, 393, 406-
343; 365, 372, 375, 381,
655
INDICES
382 fig.11.1, 393, 401, 234 fig.7.3, 241-3, 253- fig.8.1, 324; 366, 372,
415, 578, 590 4, 269, 324; 366, 370, 376, 425-6
375, 418, 421, 540, 541
7 (EF) – 36, 58, 83, 111 37 (MF/Etr) – 259, 323;
fig.3.9, 137, 148, 184, 16 (MF) – 225, 234, 235, 362, 366, 368, 372, 376,
216, 218, 240, 306, 323, 269; 366, 370, 375, 418, 426, 599
327, 343; 365, 372, 375, 421, 540, 541
38 (MF?) – 241-3, 274; 362,
381, 382 fig.11.1, 393,
17 (MF) – 90, 113, 122, 366, 372, 376, 423, 426
401, 415, 578, 590
141, 143, 188-9, 193,
39 (MF) – 87, 91 fig.3.5, 93,
8 (EF) – 36, 323, 343; 362, 198-9, 230-1, 314, 350;
118, 224, 243-4, 267;
365, 375, 381, 382 366, 369, 370, 375, 418,
366, 371, 376, 422, 426
fig.11.1, 393, 415-6, 421, 446, 516
437, 578, 590 40 (MF) – 60, 61, 70, 73,
18 (MF) – 81, 97, 98, 186,
78, 88, 91 fig.3.5, 92,
9 (EF) – 36, 58, 59, 61, 62, 228, 248, 264; 366, 370,
93, 96, 99, 109, 124,
71, 75, 93, 148, 155, 375, 418, 421-2, 446,
125, 126, 131, 155, 157,
156, 159, 160, 165, 167- 534
161, 182, 188, 233
8, 169, 172, 175, 184,
19 (MF) – 81, 93, 97, 98, n.118, 240, 261, 292,
185, 190-1, 216, 219,
186, 225, 234, 235, 243- 293, 298, 300, 310, 311,
293, 298, 299, 299-300,
4, 264; 366, 370, 375, 312-3, 324, 330; 366,
309, 323, 327, 343, 348,
418, 422 with fig.13.2, 370, 376, 422, 426-7,
350, 353, 354, 355; 365,
534 486, 496
372, 375, 381, 382
fig.11.1, 393, 400, 415- 20 (MF) – 84, 109, 111 41 (MF) – 29, 59, 72, 73,
6, 437, 578, 579, 590 fig.3.9, 122, 126, 131, 84, 90, 94, 97, 100, 103
137, 221, 240, 296, 305, n.49, 105, 107 fig.3.7,
10 (EF) – 33, 36, 93, 143,
307 fig.8.1; 362, 366, 109, 111 fig.3.9, 113,
202, 295, 307 fig.8.1,
372, 375, 422-3 130, 141, 189, 198-9,
314, 323, 327, 343; 365,
228, 230-1, 236, 240,
374, 375, 381, 382 21 (MF/LtF) – 229, 307
245, 263, 267, 302, 325,
fig.11.1, 393, 416, 425, fig.8.1, 337; 366; 367,
330; 366, 371, 376, 427-
457, 578, 590 376, 383, 423
8, 475, 523, 524
11 (MF) – 130, 224, 234, 22-27 (MF) – 228, 274, 307
42 (MF) – 72, 76, 86, 90,
235, 253-4, 273-4; 366, fig.8.1; 366, 372, 376,
94, 130, 141, 144, 198-
369, 375, 418-9, 420 423, 426, 430, 527
9, 207, 228, 230-1, 241-
12 (MF) – 77, 93, 94, 97, 28 (MF?) – 196-7, 197, 250, 3, 263-4, 272, 325, 330;
107 fig.3.7, 181, 224, 317; 366, 372, 376, 383, 366, 371, 376, 387, 428-
234 fig.7.3, 234, 235, 423, 424, 464 9, 524, 560
241-3, 245, 270, 282,
29 (MF?) – 196-7, 250, 276; 43 (MF) – 198-9, 225, 229,
292, 295, 302, 313, 325,
362, 366, 372, 376, 423, 230-1, 252, 252-3, 330;
326, 327; 366, 369, 375,
424, 464 366, 371, 376, 429, 527
418-9, 444, 445, 531
30 (MF?) – 239, 240; 362, 44 (MF?) – 274; 362, 366,
13 (MF) – 81, 107 fig.3.7,
362, 366, 372, 376, 424 372, 376, 430
130, 224, 239, 247, 260,
263; 366, 369, 375, 386, 31-32 (MF) – 59, 73, 97, 45 (MF) – 130, 221, 259,
419-20 with fig.13.1 100, 105, 109, 143, 189, 295, 307 fig.8.1; 366,
204, 295, 307 fig.8.1, 372, 376, 430
14 (MF) – 59, 60, 62, 72,
314, 327; 366, 374, 376,
80, 97, 111 fig.3.9, 130, 46 (MF?) – 81, 263; 362,
131, 184-5, 226, 228, 418, 424-5
366, 372, 376, 430
232, 234 fig.7.3, 241-3, 33 (MF?) – 107 fig.3.7, 208,
47 (MF) – 87, 228, 276;
253, 266, 324; 362, 366, 245, 265; 362, 366, 372,
366, 370, 376, 387, 430,
371, 375, 418, 420, 471, 376, 425, 469
441 n.229, 446, 486, 522
552
34-36 (MF) – 130, 131, 221,
48 (MF) – 74, 93, 107
15 (MF) – 118, 184-5, 224, 249-50, 273, 295, 307
fig.3.7, 118, 224, 227,
656
INDICES
228, 243-4, 248, 260, 366, 368, 376, 435-6 77 (MF/Etr) – 86, 323; 366,
287; 366, 370, 376, 431 with fig.13.3, 599, 600 368, 372, 376, 386, 387,
440
49 (MF) – 74, 78, 86, 93, 62 (MF) – 28, 59, 89 n.46,
107 fig.3.7, 141, 224, 90, 96, 141, 142, 147, 78 (MF?) – 229, 372, 376,
227, 228, 241-3, 251, 182, 183, 192, 203, 204, 383, 440
260, 287; 366, 370, 376, 204 n.106, 295; 366,
79 (MF) – 15, 59, 62, 93,
386, 387, 431 372, 373, 376, 383, 409,
94, 103, 111 fig.3.9,
436
50 (MF) – 74, 93, 107 126, 130, 224, 234
fig.3.7, 224, 227, 228, 63 (LtF) – 130, 202, 221, fig.7.3, 244, 253-4, 270,
241-3, 248, 260, 287; 259, 295, 337; 362; 367, 333; 366, 369, 376, 384,
366, 370, 376, 431 372, 376, 387, 437 440, 486, 489 n.255
51 (MF) – 93, 111 fig.3.9, 64 (MF/Etr) – 94, 221, 228, 80 (MF) – 59, 60, 61, 78,
224, 227, 240, 260, 287; 253-4, 277, 323, 325; 90, 91 fig.3.5, 93, 101,
366, 370, 376, 386, 432 362, 366, 368, 372, 376, 108 fig.3.8, 111 fig.3.9,
384, 388, 437, 521, 529, 118, 125, 157, 161, 182,
52 (MF) – 93, 107 fig.3.7,
599 190, 194, 224, 226, 228,
225, 260; 366, 370, 376,
229, 234 fig.7.3, 241-3,
432 65 (MF) – 33, 89 n.46, 142,
246, 252-3, 253-4, 270,
226, 307 fig.8.1; 362,
53 (MF) – 74, 107 fig.3.7, 273, 275, 275, 281, 282,
366, 374, 376, 409, 437
118, 224, 260, 263-4, 283 n.142, 293, 295,
287; 366, 370, 376, 432 66 (MF/Etr) – 226, 229, 297, 303, 310; 366, 370,
323; 362, 366, 368, 372, 376, 386, 443-4, 444
54 (MF) – 72, 81, 91 fig.3.5,
376, 437-8, 599
93, 97, 98, 186, 224, 81 (MF) – 86, 107, 108
248, 264; 366, 370, 376, 67 (MF/Etr) – 85, 94, 226, fig.3.8, 228, 241-3, 250,
432, 534 229, 237, 258, 323, 325; 262-3, 275, 282; 366,
366, 368, 372, 373, 376, 376, 386, 387, 388, 442,
55 (MF) – 362, 366, 371, 443, 471
387, 438 with fig.13.4
376, 433
68 (MF?) –239-40, 240; 82 (MF) – 29, 84, 103, 107
56 (MF) – 80, 85, 93, 97, fig.3.7, 118, 141, 191,
362, 366, 372, 376, 383,
99, 108 fig.3.8, 125, 224, 237, 260, 276, 284;
437, 438
141, 151, 187, 224, 273, 366, 370, 376, 433, 443-
282, 288, 293, 297, 310, 69-71 (MF) – 109, 122, 126, 4, 486, 546, 576
329; 366, 371, 376, 387, 131, 137, 138, 221, 251,
433, 486, 525, 602 262, 296, 305, 307 83 (MF) – 77, 86, 130, 181,
fig.8.1; 366, 372, 376, 224, 276, 282, 295, 302,
57 (MF) – 88, 224, 241-3, 313, 326; 366, 369, 369,
437, 438-9
258, 261, 263, 267; 366, 376, 386, 387, 444, 444-
376, 383, 433 72 (MF) – 89, 91 fig.3.5, 5, 445
102, 128, 172, 198, 221,
58 (MF) – 71, 74, 130, 221, 84 (MF) – 77, 87 n.45, 130,
252, 297; 366, 372, 376,
241, 260; 366, 372, 376, 181, 224, 241-3, 273,
408, 414, 437, 439
434, 564, 567 282, 295, 302, 313, 326;
73 (MF) – 118, 221, 243-4, 366, 369, 369, 376, 418,
59-60 (MF) – 59, 60, 63, 73,
307 fig.8.1; 366, 372, 444, 445
78, 80, 81, 84, 86, 90, 91
376, 439
fig.3.5, 92, 94, 96, 101, 85 (MF) – 88, 91 fig.3.5,
109, 127, 147, 151, 155, 74 (MF) – 121, 122, 221, 102, 226, 282, 324; 366,
156, 157, 161, 169-70, 243-4, 296, 305, 307 376, 387, 388, 418, 445,
174, 175, 180, 182, 183, fig.8.1; 366, 372, 376, 445, 599
187, 199, 292, 294, 298, 439
299, 301, 326, 350, 353, 86 (MF) – 226, 229, 276;
75 (MF) – 228, 266; 366, 366, 370, 376, 430, 445-
355; 362, 366, 373, 376,
372, 376, 440 6, 486, 522
383, 385, 432 n.223,
434-5, 453 76 (MF?) – 195, 316; 362, 87 (MF) – 118, 224, 246;
366, 372, 376, 383, 440 366, 370, 376, 441, 446
61 (MF/Etr) – 323; 362,
657
INDICES
88 (MF) – 58, 63, 72, 75, 225, 261, 371, 376, 447, 372, 376, 417, 457, 458-
80, 88, 93, 101, 107, 108 451, 558 9
fig.3.8, 125, 151, 156,
98 (MF) – 101, 102, 103, 113 (MF) – 30, 34, 58, 63,
167, 169, 187, 189, 190,
111 fig.3.9, 190, 225, 73, 94, 98, 109, 111
224, 234, 235, 253-4,
271-2, 281; 362, 366, fig.3.9, 114, 126, 131,
276, 293, 297, 298, 300,
371, 376, 447, 451, 452, 137, 138, 145, 146, 179,
301, 310, 311, 350, 356;
455 197-8, 204, 296, 300,
362, 366, 370, 376, 422,
315, 324, 339; 362, 366,
441, 445, 446-7, 481, 99 (MF) – 103, 111 fig.3.9,
374, 376, 457, 459, 459,
508 225, 228, 271-2; 366,
462-4
371, 376, 447, 451, 452,
89 (MF) – 81, 93, 101, 102,
452 114 (MF) – 30, 34, 58, 63,
111 fig.3.9, 118, 187,
73, 98, 109, 111 fig.3.9,
190, 218, 224, 238, 263, 100 (MF) – 93, 103, 111
114, 126-7, 114, 146,
281, 283 n.142, 300; fig.3.9, 141, 225, 271;
179, 197-8, 204, 296,
362, 366, 370, 370, 376, 362, 366, 371, 376, 447,
300, 315, 324, 339; 362,
441, 446-7, 578 452, 452
366, 374, 376, 457, 459,
90 (MF) – 30, 31 fig.2.1, 34, 101 (MF) – 78, 84, 90, 198- 460, 462-4
59, 93, 96, 101, 102, 111 9, 228, 230-1, 243, 251,
115 (MF) – 30, 34, 58, 63,
fig.3.9, 141, 195, 224, 301; 366, 371, 376, 383,
73, 94, 98, 109, 111
240 n.121, 244, 261, 386, 453
fig.3.9, 114, 126, 131,
281, 282; 362, 366, 370,
102 (MF) – 143, 198-9, 225, 145, 146, 179, 197-8,
371, 376, 447, 447-9,
229, 230-1, 248, 277, 204, 296, 300, 315, 324,
450, 515, 558
282; 362, 366, 371, 376, 339; 362, 366, 374, 376,
91 (MF) – 30, 31 fig.2.1, 34, 453 457, 459, 460, 462-4
59, 141, 148 n.78, 184,
103 (MF) – 104, 191; 362, 116 (MF) – 30, 34, 58, 73,
195, 281; 362, 366, 370,
366, 371, 376, 453-4 94, 98, 109, 111 fig.3.9,
370 n.179, 370 n.180,
114, 126, 131, 146, 179,
371, 376, 447, 447-9, 104 (MF) – 362, 366, 371, 197-8, 204, 296, 300,
558 376, 447, 454 315, 324, 339; 362, 366,
92 (MF) – 111 fig.3.9, 228, 105 (MF) – 93, 111 fig.3.9, 374, 376, 457, 459, 460,
261; 366, 371, 376, 447, 261; 362, 366, 371, 376, 462-4
449, 558 454 117 (MF) – 30, 34, 63, 94,
93 (MF) – 111 fig.3.9, 224, 106 (MF) – 362, 366, 371, 98, 114, 145, 296, 300,
229, 244, 266; 366, 371, 376, 455 315, 324, 366, 374, 376,
376, 447, 449, 558 457, 459, 460, 461, 462-
107 (MF) – 74, 121, 122, 4
94 (MF) – 93, 111 fig.3.9, 307 fig.8.1; 362, 366,
130, 151, 184-5, 187, 371, 376, 454, 455, 599 118 (MF) – 34, 98, 109,
225, 229, 240, 261, 293, 114, 126, 131, 146, 197-
297, 310; 362, 366, 371, 108 (MF) – 194; 362, 366, 8, 204, 296, 315, 324;
376, 447, 449-50, 558 371, 376, 455, 599, 600 362, 366, 374, 376, 457,
109 (MF) – 113, 122, 229, 459, 460, 462-4
95 (MF) – 80, 90, 93, 111
fig.3.9, 125, 157, 162, 296, 307 fig.8.1, 315, 119 (MF) – 34, 98, 109,
182, 225, 261, 293, 297, 324; 366, 372, 376, 457- 114, 126, 146, 197-8,
310; 362, 366, 371, 376, 8 204, 296, 315, 324; 362,
447, 450, 558 110 (MF/Etr MF) – 112, 366, 374, 376, 457, 459,
113, 229, 258; 366, 372, 461, 462-4
96 (MF) – 99, 100, 111
fig.3.9, 118, 224, 229, 373, 374, 376, 457, 458 120 (MF) – 34, 98, 109,
243-4, 260, 261; 362, ; 111 (MF?) – 366, 372, 372, 114, 126, 131, 146, 197-
366, 371, 376, 447, 450- 376, 384, 441, 458 8, 204, 296, 315, 324;
1, 451, 558 362, 366, 374, 376, 457,
112 (LF) – 113, 221, 243-4, 459, 461, 462-4
97 (MF) – 93, 111 fig.3.9, 296, 305, 307 fig.8.1,
658
INDICES
121 (MF) – 34, 98, 109, 135 (MF) – 91 fig.3.5, 208, 578, 587, 588
114, 126, 131, 146, 197- 228, 240, 261; 362, 366,
147 (MF) – 59, 62, 91
8, 204, 296, 315, 324; 371, 376, 426, 466, 467
fig.3.5, 93, 111 fig.3.9,
362, 366, 374, 376, 457,
136 (MF) – 71, 90, 206, 225, 228, 243, 265, 266,
459, 461, 462-4
228, 248, 263, 325; 362, 282, 366, 371, 376, 386,
122 (MF) – 34, 98, 109, 366, 371, 376, 466, 467- 425, 466, 469, 472
114, 126, 131, 146, 197- 7
148 (MF) – 84, 93, 111
8, 204, 296, 315, 324;
137 (MF) – 93, 277; 362, fig.3.9, 118, 224, 234
362, 366, 374, 376, 457,
366, 371, 376, 466, 467- fig.7.3, 243-4, 246, 266;
459, 461, 462-4
7 362, 366, 371, 376, 466,
123 (MF) – 34, 98, 114, 472, 476
138 (MF) – 126; 362, 366,
204, 296, 315; 362, 366,
371, 376, 466, 468, 599 149 (MF) – 59, 93, 98, 99,
374, 376, 457, 459, 460,
101, 118, 151, 187, 190,
461, 462-4 139 (MF) – 71, 90, 206,
224, 253-4, 271, 281;
225, 263; 362, 366, 371,
124 (MF) – 34, 98, 114, 362, 366, 371, 376, 442,
376, 466, 468
146, 197, 204, 296, 315; 466, 473, 474
362, 366, 374, 376, 457, 140 (LtF) – 43 n.26, 99, 111
150 (MF) – 80, 99, 208,
459, 461, 462-4 fig.3.9, 194, 224, 234
226, 229, 271, 293, 297,
fig.7.3, 238, 240, 264,
125 (MF) – 34, 98, 114, 310; 362, 366, 371, 376,
287, 337; 362; 367, 371,
146, 197, 204, 296, 315; 466, 473
376, 383, 389, 417, 468-
362, 366, 374, 376, 457,
9, 481, 542, 546 n.283 151 (MF) – 93, 97, 99, 113,
462-4
224, 232-3, 234 fig.7.3,
141 (MF) – 93, 97, 111
126 (MF) – 34, 98, 197, 243-4, 252, 271; 362,
fig.3.9, 113, 225, 232-3,
204, 296, 315; 362, 366, 366, 371, 376, 401, 439,
234 fig.7.3, 243-4, 265;
374, 376, 457, 462-4 466, 473-4 with fig.14.2
362, 366, 371, 376, 425,
127 (MF) – 59, 195, 295, 466, 469 152 (MF) – 87, 93, 100,
316, 317; 362, 366, 374, 185, 194, 224, 234, 235,
142 (MF) – 93, 111 fig.3.9,
376, 457, 464 253-4, 268, 282; 362,
224, 234 fig.7.3, 265;
366, 371, 376, 466, 474-
128 (MF?) – 196-7, 197, 362, 366, 371, 376, 425,
5
250, 317; 362, 366, 372, 466, 469
374, 376, 424, 457, 464, 153 (MF) – 85, 85 n.44, 93,
143 (MF) – 93, 224, 265;
465 97, 111 fig.3.9, 224,
362, 366, 371, 376, 425,
240, 272; 362, 366, 371,
129 (MF?) – 78, 194, 249; 466, 469
376, 387, 466, 475
362, 366, 372, 374, 375,
144 (MF) – 193-4, 208, 228,
383, 386, 457, 464-5, 154 (MF) – 85, 85 n.44,
230-1, 234 fig.7.3, 253-
129 113, 121, 122, 229, 272,
4, 266-7, 326; 362, 366,
297, 300, 312, 313; 362,
130 (MF?) – 108, 270; 362, 371, 376, 466, 470-1
366, 371, 376, 387, 466,
366, 372, 374, 376, 457, with fig.14.1
475
465, 481
145 (MF) – 208, 225, 253-4;
155 (MF) – 29, 97, 106, 107
131 (MF?) – 195, 316; 362, 362, 366, 371, 376, 466,
fig.3.7, 114, 184-5, 189,
366, 372, 374, 376, 383, 470-1
228, 234 fig.7.3, 236,
417, 457, 465
146 (MF) – 59, 60, 62, 72, 245, 251, 274, 284, 301,
132 (MF) – 362, 362, 366, 80, 87, 90, 91 fig.3.5, 302; 366, 371, 376, 466,
374, 375, 376, 465-6 93, 97, 103, 108 fig.3.8, 475-6
111 fig.3.9, 125, 151,
133 (MF?) – 241-3; 362, 156 (MF) – 101, 123, 190,
157, 161, 182, 208, 224,
366, 372, 374, 374, 376, 225, 229, 253-4, 277,
228, 234 fig.7.3, 241-3,
457, 466 281 n.141, 300, 312,
243, 266, 277, 293, 297,
366, 371, 376, 466, 476
134 (MF?) – 362, 366, 372, 310; 362, 366, 371, 376,
374, 374, 376, 457, 466 385, 458, 466, 471-2, 157 (MF) – 224, 243-4, 276;
659
INDICES
, 366, 371, 376, 466, 476 171 (LtF) – 43 n.26, 94, 187 (MF) – 362, 366, 371,
184-5, 226, 235 fig.7.4, 376, 467, 485
158 (MF) – 78, 80, 90, 91
239-40, 240, 337; 362,
fig.3.5, 108 fig.3.8, 118, 188 (MF) – 362, 366, 371,
367, 371, 376, 389, 417,
151, 157, 161, 182, 187, 376, 467, 485
465, 466, 467, 481
194, 225, 229, 234, 235,
189 (MF) – 362, 366, 371,
253-4, 266, 293, 297, 172 (LtF) – 43 n.26, 98,
376, 467, 485
303, 310; 362, 366, 371, 156, 172-3, 175, 184-5,
376, 386, 467, 477 197, 226, 235 fig.7.4, 190 (MF) – 362, 366, 371,
241-3, 337; 367, 371, 376, 467, 486
159 (MF) – 59, 93, 111
376, 389, 417, 466, 467,
fig.3.9, 125, 157, 161, 191 (MF) – 99, 141, 224,
481
182, 225, 244, 293, 297, 252, 273, 288, 366, 371,
310; 362, 366, 371, 376, 173 (LtF) – 43 n.26, 90, 376, 467, 486
467, 477 102, 125, 142, 191, 229,
192 (MF) – 362, 366, 371,
241-3, 246, 272, 337;
160 (MF) – 228, 248; 362, 376, 467, 486
367, 371, 376, 389, 417,
366, 371, 376, 467, 477-
466, 467, 482, 482 193 (MF) – 93; 362, 366,
8
371, 376, 467, 486
174 (LtF) – 43 n.26, 182,
161 (MF) – 93, 118, 161,
184-5, 226, 230-1, 235 194 (MF) – 362, 366, 371,
182, 229, 293, 297, 310;
fig.7.4, 239, 262, 337; 376, 467, 486
362, 366, 371, 376, 467,
362; 367, 371, 376, 389,
478, 552 195 (MF) – 91 fig.3.5, 93,
417, 466, 482, 482 with
n.252 99, 100, 103, 224, 260,
162 (MF) – 93, 190, 226,
366, 370, 376, 467, 487
234 fig.7.3, 253-4, 281,
175 (MF) – 93, 126; 362,
282, 366, 371, 376, 467, 196 (MF) – 87, 88, 112,
366, 371, 376, 467, 482-
478 228, 264, 270, 277, 366,
3
370, 376, 467, 487, 580
163 (MF) – 225, 234 fig.7.3,
176 (MF) – 362, 366, 371,
253-4; 362, 366, 371, 197 (MF) – 93, 224, 241-3,
376, 467, 483
376, 467, 478 366, 370, 376, 467, 487
177 (MF) – 362, 366, 371,
164 (MF) – 93, 226, 253-4; 198 (MF) – 118, 221, 243-4,
376, 467, 483
362, 366, 371, 376, 467, 366, 370, 376, 467, 487
479 178 (MF) – 93, 126; 362,
366, 371, 376, 467, 482, 199 (MF/Etr) – 108 fig.3.8,
165 (MF) – 29, 106, 107 118, 221, 270, 309, 323;
483
fig.3.7, 113, 121, 122, 366, 368, 372, 376, 467,
189, 236, 245, 284, 301, 179 (MF) – 362, 366, 371, 488
302, 366, 371, 376, 467, 376, 467, 484
479 200 (MF) – 98, 130, 225,
180 (MF) – 93; 362, 366, 241-3, 262, 295, 309;
166 (MF) – 181, 226, 234, 371, 376, 467, 484 366, 372, 376, 454, 467,
235, 243-4, 282, 371, 488
181 (MF) – 362, 366, 371,
376, 467, 479-80
376, 467, 484 201 (MF) – 121, 220, 227,
167 (MF) – 226, 253-4, 251-2, 295, 307 fig.8.1;
182 (MF) – 362, 366, 371,
253-4; 362, 362, 366, 362, 366, 372, 376, 467,
376, 467, 484
371, 376, 467, 480 488
183 (MF) – 362, 366, 371,
168 (MF) – 238; 362, 366, 202 (MF?) – 246-7, 366,
376, 467, 484-5
371, 376, 467, 480 372, 376, 387, 467, 488
184 (MF) – 362, 366, 371,
169 (MF) – 85, 91 fig.3.5, 203 (MF?) – 243-4; 362,
376, 467, 485
143, 226, 238; 362, 366, 366, 372, 376, 467, 488
371, 376, 467, 480 185 (MF) – 362, 366, 371,
376, 467, 485 204 (MF?) – 240; 362, 366,
170 (MF) – 78, 194, 303, 372, 376, 467, 489
337; 362, 366, 371, 376, 186 (MF) – 362, 366, 371,
386, 467, 480-1 205 (LtF) – 15, 30, 74, 108
376, 467, 485
660
INDICES
fig.3.8, 179, 207; 367, 339, 342, 355 n.175; 113, 118, 125, 130, 141,
375, 376, 489-90, 490, 362, 368, 374, 376, 377, 151, 157, 182, 187, 189,
500 n.262, 599 387, 387, 389, 493, 494- 190, 206-7, 229, 236,
7, 497, 498, 513 246, 253, 256, 257-8,
206 (MLF) – 15, 30, 84, 93,
293, 297, 301, 301
118, 225, 239, 252, 268, 215 (LtF) – 94, 102, 125,
n.151, 302, 310; 362;
277, 303, 317; 366, 374, 225, 235, 241-3, 273,
367, 371, 376, 387, 476,
376, 429, 441 n.229, 277, 317, 336, 337; 367,
493, 501-2, 507, 508-9
489, 490, 490, 491, 500 375, 376, 454 n.242,
n.262 493, 494, 497, 498 222 (LF) – 15, 44, 59, 73,
80, 90, 98, 108 fig.3.8,
207 (MLF) – 15, 30, 87, 216 (Lat) – 43, 102, 107
130, 198-9, 229, 230-1,
225, 241-3, 267, 317, fig.3.7, 125, 235 fig.7.4,
234 fig.7.3, 241-3, 246,
339; 366, 374, 376, 387, 251-2, 265, 309, 336;
253-4, 256, 257-8, 272,
441 n.229, 489, 490-1, 367, 372, 376, 389, 493,
276, 288, 339; 362; 367,
491, 500 n.262 497-8
371, 376, 493, 502, 508-
208 (MLF/Etr) – 15, 226, 217 (Lat) – 54, 95, 101, 9, 545 n.282
270, 317, 323; 366, 368, 102, 107 fig.3.7, 108
223 (LF) – 15, 44, 80, 90,
375, 376, 441 n.229, fig.3.8, 178, 201, 291,
91 fig.3.5, 94, 97, 108
491, 491, 500 n.262 303, 317, 336, 337
fig.3.8, 113, 118, 125,
n.173, 339; 367, 374,
209 (MLF/Etr) – 15, 204, 151, 157, 161, 182, 187,
376, 384, 389, 493, 494,
270, 323; 366, 368, 374, 206-7, 224, 229, 230-1,
497, 498-500
375, 376, 441 n.229, 241-3, 256, 276, 288,
489, 491, 491, 500 n.262 218 (Lat) – 25-6, 54, 101, 293, 298, 310; 362; 367,
102, 108 fig.3.8, 125, 370, 371, 376, 493, 502,
210 (MLF) – 15, 30, 87, 88, 508-9, 545 n.282
178, 201, 204 n.106,
93, 196, 225, 241-3,
208, 225, 235 fig.7.4,
264, 266, 339; 366, 374, 224 (LF) – 15, 59, 80, 93,
240, 241-3, 245, 253-4,
376, 441 n.229, 487, 94, 97, 108 fig.3.8, 113,
254, 266, 272, 291, 303,
491, 500 n.262, 580 125, 151, 157, 161, 182,
317, 337, 337 n.173,
187, 190, 224, 228, 232-
211 (MLF) – 93, 111 339; 367, 374, 376, 386,
3, 234 fig.7.3, 241-3,
fig.3.9, 224, 234, 235, 389, 493, 494, 497, 498-
253-4, 267, 276, 288,
240, 241-3, 261, 267; 500
293, 298, 310; 362; 367,
362, 366, 371, 376, 491 371, 376, 493, 503, 508-
219 (Lat) – 30, 31 fig.2.1,
212 (MLF) – 86, 91 fig.3.5, 33, 48, 54, 97, 125, 178, 9, 545 n.282
228, 272, 277; 362, 366, 201, 202, 209, 225, 235
225 (LF) – 15, 59, 94, 97,
371, 376, 386, 387, 492, fig.7.4, 241-3, 274, 283,
108 fig.3.8, 113, 125,
554, 565 291, 311, 317, 337, 339;
190, 224, 228, 229, 241-
367, 370, 374, 376, 384,
213 (LF) – 30, 31 fig.2.1, 3, 267, 276, 288; 362;
493, 500, 556
34, 76, 80, 94, 97, 98, 367, 371, 376, 388, 493,
113, 184-5, 186, 193, 220 (LF) – 15, 80, 93, 94, 503, 508-9, 545 n.282
225, 234 fig.7.3, 239-40, 97, 101, 107, 108
226 (LF) – 15, 93, 94, 99,
240, 241-3, 246, 264, fig.3.8, 113, 141, 151,
108 fig.3.8, 157, 162,
273, 303, 336; 367, 375, 157, 161, 182, 187, 190,
182, 187, 206-7, 224,
376, 420, 493, 493-4 224, 229, 232-3, 234
229, 234 fig.7.3, 239-40,
fig.7.3, 241-3, 243-4,
214 (LF/Lat) – 30, 31 240, 246, 248, 251-2,
253-4, 258, 276, 281,
fig.2.1, 34, 47, 54, 76, 257-8, 270, 293, 298,
282, 283, 288, 293, 297,
85, 92, 95, 100, 101, 310, 325; 362; 367, 371,
310, 339; 362; 367, 371,
102, 110, 111 fig.3.9, 376, 388, 493, 503-4,
376, 493, 501, 503, 508-
114, 123, 124, 142, 144, 508-9
9
146, 178, 183, 184-5, 227 (LF) – 15, 106, 107
193, 195, 225, 234 221 (LF) – 15, 29, 44, 59,
fig.3.7, 197, 198-9, 230-
fig.7.3, 244, 262, 288, 80, 85, 90, 93, 96, 97,
1, 234 fig.7.3, 246, 247-
291, 303, 317, 336, 337, 98, 106, 107 fig.3.7,
8, 268, 282; 362; 367,
661
INDICES
371, 376, 388, 493, 504, 234, 235, 239-40, 240, 244 (LF) – 17, 94, 103, 127,
505, 507, 508-9 277; 367, 371, 376, 493, 193, 200, 225, 271, 294;
511 367, 371, 376, 493, 512,
228 (LF) – 15, 94, 225, 234
513, 514
fig.7.3, 246, 268; 362; 236 (LF) – 131, 224, 234
367, 371, 376, 493, 505, fig.7.3, 241-3, 260; 367, 245 (LF) – 17, 30, 31
508-9 371, 376, 493, 511 fig.2.1, 78, 103, 111
fig.3.9, 127, 194, 200,
229 (LF) – 15, 111 fig.3.9, 237 (Lat) – 30, 31 fig.2.1,
294, 311, 339; 367, 370,
184-5, 206, 229, 234 54, 102, 125, 178, 235
371, 376, 386, 493, 512,
fig.7.3, 240, 251, 274; fig.7.4, 241-3, 246, 273,
514, 514
362; 367, 371, 376, 493, 287, 291, 311, 337, 339;
505, 508-9, 574 368, 370, 371, 376, 493, 246 (LF) – 17, 30, 262-3;
507, 508, 511, 515 362; 367, 371, 376, 386,
230 (LF) – 15, 88, 97, 100,
472, 493, 512, 514
141, 182, 224, 245, 262, 238 (Lat) – 30, 31 fig.2.1,
283, 290, 326; 362; 367, 54, 102, 125, 178, 235 247 (LF) – 17, 30, 31
371, 376, 493, 505-6, fig.7.4, 241-3, 246, 273, fig.2.1, 59, 78, 90, 111
508-9, 579 287, 291, 311, 337, 339; fig.3.9, 128, 183, 184-5,
368, 370, 371, 376, 493, 193, 194, 199, 226, 234
231 (LtF) – 15, 30, 31
507, 508, 511, 515 fig.7.3, 311, 339; 367,
fig.2.1, 33, 34, 59, 60,
370, 371, 376, 386, 493,
70, 72, 88, 90, 94, 97, 239 (LtF) –225, 246, 269,
507, 512, 514
100, 101, 102, 108 337, 339; 367, 370, 371,
fig.3.8, 141, 151, 161, 376, 493, 511-2 248 (LF) – 17, 30, 31
181, 182, 184-5, 187, fig.2.1, 59, 90, 128, 141,
240 (Lat) – 30, 31 fig.2.1,
194, 195, 196, 224, 228, 183, 193, 199, 224, 273,
54, 128, 178, 291; 368,
232, 235 fig.7.4, 239-40, 311, 339; 367, 370, 371,
376, 493, 512
240, 241-3, 246, 249, 376, 493, 512, 515
262, 270, 282, 283, 293, 241 (MLF/LtF) – 90, 337;
249 (LF) – 17, 30, 31
298, 310, 311, 325, 326, 362, 366, 367, 372, 376,
fig.2.1, 34, 59, 128, 183,
337, 339; 362; 367, 370, 493, 512
184-5, 195, 199, 226,
371, 376, 387-8, 389,
242 (LF) – 17, 30, 31 229, 234 fig.7.3, 241-3,
448, 493, 505, 506-9,
fig.2.1, 78, 84, 89, 90, 339; 367, 370, 371, 376,
510, 513, 514, 515, 516,
91 fig.3.5, 92, 93, 94, 448, 493, 507 n.267,
579
111 fig.3.9, 141, 142, 508, 512, 515
232 (part LF, part LtF) – 15, 147, 156, 157, 159, 160,
250 (Lat) – 54, 87, 102,
30, 31 fig.2.1, 33, 34, 169, 174, 180, 184-5,
125, 178, 225, 235
60, 70, 72, 90, 91 186, 189-90, 193, 194,
fig.7.4, 241-3, 248, 265,
fig.3.5, 101, 141, 181, 198-9, 224, 227, 228,
283, 291, 303, 337; 368,
187, 196, 226, 238, 246, 230-1, 232, 234 fig.7.3,
372, 376, 480, 493, 512,
311, 337, 339; 362, 367; 239, 241-3, 246, 271,
515, 590
367, 370, 371, 376, 389, 293, 298, 299, 311, 317,
493, 505, 507, 508, 509- 339, 344; 367, 370, 371, 251 (Lat) – 54, 59, 76, 94,
10 with fig.15.1 376, 386, 471, 472, 493, 102, 106, 107 fig.3.7,
494, 495, 512, 512-3, 108 fig.3.8, 125, 178,
233 (LtF) – 31 fig.2.1, 70,
513, 514, 523, 579 188-9, 193, 206, 225,
88, 97, 100, 102, 125,
228, 229, 235 fig.7.4,
141, 181, 182, 194, 225, 243 (LF) – 17, 30, 31
241-3, 245, 247-8, 253-
230-1, 246, 262, 326, fig.2.1, 59, 78, 84, 111
4, 257, 266, 270, 275,
337, 339; 362; 367, 370, fig.3.9, 128, 141, 156,
291, 314, 325, 337; 368,
371, 376, 389, 493, 509- 157, 159, 169, 174, 183,
369, 376, 421, 471, 493,
10 with fig.15.1, 579 186, 189-90, 193, 194,
514, 515-6, 553
199, 293, 298, 299, 311,
234 (LF) – 184-5, 226, 229,
317, 339; 367, 370, 371, 252 (MF/LtF) – 244; 362,
234 fig.7.3, 239; 367,
376, 386, 386, 387, 472, 366, 367, 372, 376, 517
371, 376, 387, 493, 510
493, 507, 512, 513, 523
253 (MF/LtF) – 362, 366,
235 (LF) – 111 fig.3.9, 229,
662
INDICES
367, 372, 376, 383, 519, 376, 519, 524 287, 327, 328, 329, 330;
520 362, 366, 371, 376, 386,
267 (MF/Etr) – 86, 99, 100,
387, 519, 527-8
254 (MF?) – 337; 362, 366, 225, 238, 309, 323, 366,
372, 376, 519, 520 368, 372, 376, 384, 386, 277 (MF/LtF) – 241, 337;
387, 519, 524, 604 362; 367; 367, 372, 376,
255 (MF?) – 362, 366, 372,
519, 528
376, 519, 520 268 (Lat) – 13, 54, 102, 107
fig.3.7, 125, 149, 178, 278 (MF/LtF) – 244, 337;
256 (MF/Etr) – 323; 362,
267, 291, 309, 335, 337; 362; 367; 367, 372, 376,
366, 368, 372, 376, 519,
367, 368, 372, 376, 389, 519, 528
520
498, 519, 525, 579, 602
279 (MF/Etr) – 94, 226,
257 (MF) – 76, 84, 85, 90,
269 (MF) – 70, 85, 90, 91 229, 237, 258, 323; 366,
91 fig.3.5, 94, 99, 100,
fig.3.5, 94, 99, 100, 118, 368, 372, 376, 519, 528
111 fig.3.9, 141, 143,
141, 145, 224, 238, 268,
224, 232, 234 fig.7.3, 280 (MF/Etr) – 94, 226,
268, 286, 289, 323, 325,
238, 261, 289, 325, 328, 229, 238, 259, 323; 366,
326, 329; 362, 366, 371,
366, 371, 376, 387, 519, 368, 372, 376, 519, 528
376, 387, 409, 519, 525,
521
526, 601 281 (MF?) – 362, 366, 372,
258 (MF) – 94, 226, 229, 376, 519, 528
270 (MF) – 70, 84, 90, 91
252-3, 272, 275, 325,
fig.3.5, 113, 118, 121, 282 (MF/Etr) – 94, 226,
328, 366, 371, 376, 384,
122, 142, 145, 224, 228, 229, 238, 259, 366, 368,
437, 519, 521, 529
245, 261, 268, 286, 289, 372, 376, 519, 528
259-260 (MF) – 93, 221, 297, 300, 312, 313 (with
n.161) , 324, 325, 326, 283 (MF?) – 362, 366, 372,
229, 267, 307 fig.8.1,
329; 362, 366, 371, 376, 376, 383, 519, 528
366, 372, 376, 519, 522
387, 388, 409, 519, 525, 284 (MF?) – 252; 362, 366,
261-262 (MF?) – 130, 221, 526, 557, 601 372, 376, 519, 529
276, 295, 307 fig.8.1,
366, 372, 376, 388, 519, 271 (MF) – 85, 97, 99, 228, 285 (MLF) – 74, 77, 85, 90,
522 248, 273, 286, 329; 362, 91 fig.3.5, 92, 98, 102,
366, 371, 376, 387, 433, 109, 113, 128, 144, 156,
263 (MF) – 111 fig.3.9, 226, 519, 526 172-3, 175, 181, 188-9,
234, 235, 240, 243, 265,
193, 197, 226, 277, 279,
366, 372, 376, 519, 522- 272 (MF) – 84, 85, 93, 118,
292 n.146, 293, 295,
3 with fig.16.1 224, 228, 241-3, 247,
298, 302, 313, 314, 326,
269, 273, 281, 283
264 (MF/Etr) – 12, 98, 118, n.142, 286, 329; 362,
344, 353; 366, 369, 369
307 fig.8.1, 323; 362, 366, 371, 376, 387, 519, n.178, 376, 384, 387,
366, 368, 372, 376, 519, 418, 421, 437, 444, 445,
526-7
523, 602 446, 516, 519, 521, 529,
273 (MF) – 130, 221, 241-3, 553
265 (MF) – 84, 85, 90, 91 274, 295, 307 fig.8.1,
fig.3.5, 93, 94, 99, 100, 286 (MLF) – 362, 366, 372,
366, 372, 376, 426, 430,
111 fig.3.9, 132, 141, 376, 519, 529, 604
519, 527
143, 207, 219, 221, 228,
287 (MF/Etr) – 362, 366,
230-1, 234 fig.7.3, 238, 274 (MF) – 274; 362, 366,
368, 372, 376, 519, 530
245, 248, 259, 261, 265, 372, 376, 519, 527
with fig.16.2
286, 289, 292, 325, 326,
275 (MF) – 74, 93, 97, 107
327, 328; 362, 366, 371, 288 (LtF) – 337; 362, 367,
fig.3.7, 113, 224, 232-3,
376, 418, 519, 521, 523- 369, 376, 519, 530
234 fig.7.3, 241-3, 253-
4, 601, 602
4, 260, 287, 329; 362, 289 (MLF/Etr) – 84, 323;
266 (MF) – 76, 85, 90, 91 366, 371, 376, 519, 527 366, 375, 376, 489, 519,
fig.3.5, 118, 141, 143, 530, 601
276 (MF) – 74, 86, 93, 97,
224, 232, 234 fig.7.3,
107 fig.3.7, 111 fig.3.9, 290 (LtF) – 94, 97, 102,
238, 252, 273-4, 286,
113, 224, 232-3, 234 125, 195, 225, 226, 245,
289, 328; 362, 366, 371,
fig.7.3, 240, 253-4, 260,
663
INDICES
258, 270, 277, 281, 282, 366, 370, 376, 420, 453, 2, 253-4, 275, 287; 366,
303, 317, 337, 339; 367, 519, 534, 534 371, 376, 519, 534, 537,
376, 418, 489, 501, 519, 569
303 (MLF) – 16, 228, 248,
531, 541, 548
262; 366, 370, 376, 519, 313 (MLF) – 16, 72, 78, 88,
291 (Lat) – 30, 54, 102, 534, 534 194, 224, 228, 253-4,
125, 178, 225, 235 265, 275, 287, 303; 366,
304 (MLF) – 16, 99, 121,
fig.7.4, 241-3, 250, 262- 371, 376, 386, 519, 534,
228, 248, 271, 307
3, 291, 317, 337, 339; 538, 569
fig.8.1; 366, 372, 376,
368, 374, 376, 489, 519,
519, 535, 601 314 (MLF) – 16, 107
531
fig.3.7, 113, 234 fig.7.3,
305 (MLF) – 16, 80, 88, 93,
292 (LtF) – 104, 189, 337; 241-3, 267; 366, 371,
109, 113, 125, 126, 131,
367, 372, 376, 519, 531- 376, 519, 534, 538, 559
157, 161, 184-5, 224,
2
233 n.118, 234 fig.7.3, 315 (MLF) – 16, 88, 93,
293 (MLF) – 107 fig.3.7, 251-2, 275, 287, 295, 113, 121, 122, 159, 168,
109, 111 fig.3.9, 131, 312-3; 366, 371, 376, 188, 224, 234 fig.7.3,
218, 227, 231-2, 233 433, 519, 534, 535, 569 243-4, 251-2, 281, 282,
n.118, 234 fig.7.3, 237- 287, 300, 312, 326; 366,
306 (MLF) – 16, 84, 229;
8, 245, 296, 300, 305; 371, 376, 387, 519, 534,
366, 371, 376, 519, 534,
366, 372, 376, 519, 532, 536, 538, 569
535-6
578
316 (MLF) – 16, 88, 93,
307 (MLF) – 16, 88, 93,
294 (LtF) – 239, 337; 362; 224, 248, 275, 287; 366,
113, 118, 224, 243-4,
367, 372, 376, 519, 532 371, 376, 387, 519, 534,
251-2, 275, 287, 293,
538, 569, 590
295 (Lat) – 99, 271, 295, 298, 310; 366, 371, 376,
309; 368, 372, 376, 519, 519, 534, 536, 569 317 (MLF) – 16, 125, 226,
532 229, 248, 275, 287; 366,
308 (MLF) – 16, 87, 88, 93,
371, 376, 519, 534, 535,
296 (Lat) – 54, 99, 178, 113, 113, 121, 122, 184-
538-9, 569, 590
271, 291, 295, 309; 368, 5, 224, 228, 241-3, 248,
372, 376, 519, 532 275, 287, 297, 300, 312, 318 (MLF) – 16, 84, 88, 93,
313; 366, 371, 376, 519, 195, 225, 230-1, 241-3,
297 (MLF) – 118, 221, 243-
534, 536, 537, 569 270, 281, 282; 366, 371,
4; 366, 371, 376, 519,
376, 519, 531, 534, 535,
532 309 (MLF) – 16, 88, 93,
539
113, 125, 159, 168, 188,
298 (MLF) – 118, 221, 243-
224, 234 fig.7.3, 234, 319 (MLF) – 16, 93; 362,
4; 366, 371, 376, 519,
235, 246-7, 251-2, 275, 366, 371, 376, 519, 534,
532 281, 282, 287, 300, 312, 535, 539, 539
299 (LtF) – 103, 225, 253-4, 326; 366, 371, 376, 454,
320 (MLF) – 16, 239, 240;
268, 337; 367, 371, 376, 519, 534, 536-7, 538,
362, 366, 372, 376, 519,
429, 519, 533 569
534, 539
300 (LtF) – 81, 103, 107 310 (MLF) – 16, 69, 70, 76,
321 (MLF) – 16, 239, 240;
fig.3.7, 141, 184-5, 198- 78, 88, 91 fig.3.5, 93,
362, 366, 372, 376, 519,
9, 226, 230-1, 235 97, 113, 121, 195, 224,
534, 539
fig.7.4, 239, 268; 367, 229, 249, 275, 287, 297,
371, 376, 429, 519, 533 300, 312, 313; 366, 371, 322 (MLF) – 16, 252; 362,
376, 386, 519, 534, 537, 366, 372, 376, 519, 534,
301 (LtF) – 97, 141, 198-9, 569 539
229, 230-1, 235 fig.7.4,
243, 246, 269, 277, 337; 311 (MLF) – 16, 196, 228; 323 (MLF) – 16, 241-3;
367, 371, 376, 504, 519, 366, 371, 376, 519, 534, 362, 366, 372, 376, 519,
533-4 537 534, 539
302 (MLF) – 16, 72, 81, 97, 312 (MLF) – 16, 59, 88, 93, 324 (MLF) – 93, 224, 241-
98, 186, 228, 234 111 fig.3.9, 113, 196, 3, 266; 362, 366, 371,
fig.7.3, 243, 251-2, 264; 224, 227, 228, 249, 251- 376, 519, 540, 540
664
INDICES
325 (LtF) – 81, 97, 184-5, 334 (LF) – 13 n.4, 229, 241- 348 (MLF) – 74, 93, 107
225, 228, 232, 235 3, 276; 362; 367, 371, fig.3.7, 224, 239, 259;
fig.7.4, 246, 264, 269, 376, 468, 519, 545 366, 370, 376, 427
303, 337; 367, 371, 376, n.217, 467 n.247, 507
335 (LF, falsum?) – 13 n.4,
420, 519, 540, 540, 541, n.267, 519, 533, 547,
94, 225; 362; 367, 371,
548 550
376, 468, 519, 545
326 (LtF) – 93, 157, 162, 349 (MLF) – 93, 224, 241-
336 (LF) – 76, 94, 225, 234
182, 337; 367, 371, 376, 3, 259; 366, 370, 376,
fig.7.3, 241-3, 251-2,
519, 540, 541 427 n.217, 467 n.247,
253-4, 268, 276, 339;
507 n.267, 519, 547, 550
327 (LtF) – 59, 111 fig.3.9, 362, 367, 371, 376, 388,
184-5, 225, 229, 235 468, 519, 546 350 (MLF) – 94, 224, 251-
fig.7.4, 244, 250, 251-2, 2, 259; 366, 370, 376,
337 (LF) – 248, 268; 367,
258, 266, 275, 337; 367, 427 n.217, 455, 467
371, 376, 468, 519, 546,
371, 376, 501, 519, 540, n.247, 507 n.267, 519,
590
541, 548 519, 547, 550
338 (MLF) – 228, 250, 266;
328 (LtF) – 225, 250, 269, 351 (MLF) – 94, 111
366, 371, 376, 519, 547,
337; 367, 371, 376, 420, fig.3.9, 224, 240, 274;
547
471, 519, 540, 541, 548 366, 370, 376, 427
339 (MLF) – 76, 98-9, 141, n.217, 455, 467 n.247,
329 (LF) – 81, 85, 92, 94,
142, 207, 234 fig.7.3, 507 n.267, 519, 547,
97, 99, 103, 196, 224,
252, 269; 366, 371, 376, 550, 553
250, 264, 287; 362; 367,
519, 547, 547, 558, 603
371, 376, 387, 388, 468, 352 (MLF) – 71, 88, 94, 111
519, 531, 542, 546 n.284 340 (LtF) – 229, 241-3, 269, fig.3.9, 195, 224, 241,
337; 362; 367, 371, 376, 258, 263; 366, 370, 376,
330 (LF) – 81, 85, 92, 94,
420, 519, 540, 541, 547, 427 n.217, 434, 467
97, 99, 103, 109, 126,
548 n.247, 519, 547, 551,
131, 196, 224, 233
551, 564, 567
n.118, 234 fig.7.3, 250, 341 (LtF) – 94, 125, 182,
253-4, 264, 287, 297, 226, 229, 230-1, 243, 353 (MLF) – 58, 74, 191,
300, 312; 362; 367, 371, 258, 277, 283, 337; 367, 224, 277; 362, 366, 370,
376, 387, 387, 468, 519, 371, 376, 453, 501, 519, 376, 427 n.217, 467
542, 543 541, 547, 548 n.247, 519, 547, 551
with fig.16.4
331 (LF) – 81, 84, 85, 94, 342 (LtF) – 235 fig.7.4,
99, 103, 111 fig.3.9, 337; 362; 367, 371, 376, 354 (MLF) – 87, 94, 99,
225, 240, 264, 287; 362; 519, 547, 548 131, 84, 224, 234, 235,
367, 371, 376, 387, 468, 251-2, 270-1; 362, 366,
519, 543 343 (LtF) , 337; 362; 367, 371, 376, 427 n.217, 467
371, 376, 519, 531, 547, n.247, 519, 547, 551-2
332 (LF) – 81, 85, 94, 97, 549
99, 103, 113, 130, 184- 355 (MLF) – 94, 97, 221,
344 (LtF) , 337; 362; 367, 224, 246, 259, 268; 362,
5, 225, 227, 228, 232,
371, 376, 519, 547, 549 366, 371, 376, 427
234 fig.7.3, 234 fig.7.3,
239-40, 241, 248, 252, 345 (LtF) , 337; 362; 367, n.217, 467 n.247, 519,
264, 282, 287, 297, 339; 371, 376, 519, 547, 549 547, 552
362; 367, 371, 376, 387,
346 (MLF) – 29, 81, 93, 356 (MLF/Etr) – 30, 323;
468, 519, 543-4, 564
118, 207, 224, 237, 246, 366, 375, 376, 519, 552
n.284
251-2, 253-4, 264, 274, 357 (MLF/Etr) – 30, 86,
333 (LF) – 81, 84, 85, 94, 284, 292; 366, 370, 376, 259, 323; 366, 375, 376,
97, 99, 103, 109, 126, 442 n.233, 519, 547, 549 519, 552
131, 225, 233 n.118, 234
fig.7.3, 264, 300, 312; 347 (MLF) – 78, 141, 228, 358 (MLF) – 94, 111
362; 367, 371, 376, 387, 251; 366, 370, 376, 386, fig.3.9, 239-40, 274;
468, 519, 544-5 with 427 n.217, 455, 467 362, 366, 371, 376, 519,
fig,16.3, 551 n.247, 519, 547, 550
552-3
665
INDICES
359 (MLF) – 78, 94, 225, 367, 374, 376, 555, 558, 176, 178, 197, 208, 211,
260; 362, 366, 371, 376, 563, 568 221, 239, 265, 282, 293,
384, 519, 552-3 298, 308, 332, 333, 334,
378 (LF) – 58, 94, 148, 184,
337 n.173, 344; 367,
360 (MLF) – 81, 94, 228, 221, 246, 258, 306, 308,
372, 376, 561, 563, 562-
241-3, 263, 298; 362, 327, 344; 367, 372, 376,
3, 566, 575
366, 371, 376, 420, 446, 555, 559
540, 553 390 (Cap) – 94, 112, 126,
379 (LF) – 59, 111 fig.3.9,
221, 232 n.117, 234,
361 (MLF) – 63, 69, 70, 76, 113, 121, 122, 196, 227,
235, 240, 247, 258, 307
78, 88, 91 fig.3.5, 92, 249, 296, 305, 307
fig.8.1, 309, 333; 367,
111 fig.3.9, 125, 152, fig.8.1; 367, 372, 376,
372, 376, 386, 561, 563,
171, 188-9, 197, 226, 537, 555, 559, 560
566
293, 294, 300; 366, 369,
380 (LF) – 228, 272, 307
371, 376, 386, 421, 516, 391 (Cap) – 71, 94, 98, 103,
fig.8.1; 367, 372, 376,
531, 540, 553-4 125, 221, 239, 241, 264,
555, 559
307 fig.8.1; 367, 372,
362 (MLF) – 86, 91 fig.3.5,
381 (LF?) – 243-4; 362; 376, 388, 561, 563
193, 228-9, 243, 271,
367, 372, 376, 555, 559
272; 366, 371, 376, 386, 392 (Cap) – 71, 94, 125,
387, 554, 565 382 (LF) – 94, 125, 221, 221, 241, 269, 307
241-3, 307 fig.8.1; 367, fig.8.1; 367, 372, 376,
363 (MLF) – 12, 91 fig.3.5,
372, 376, 555, 560, 580 551, 561, 563, 564, 567
94, 97, 118, 209, 224,
243-4, 274; 366, 376, 383 (LF) – 94, 114, 121, 393 (Lat) – 54, 106, 107
500, 555, 556, 605 122, 148, 184, 207, 221, fig.3.7, 108 fig.3.8, 178,
228, 252, 274, 306, 325, 228, 232, 235 fig.7.4,
364 (MLF) – 12, 91 fig.3.5,
327, 344; 367, 372, 376, 241-3, 262, 291, 337;
94, 97, 209, 226, 274;
555, 560 368, 372, 376, 561, 564
366, 371, 376, 500, 555,
556, 605 384 (LF) – 90, 91 fig.3.5, 394 (Cap) – 125, 225, 250,
102, 103 n.49, 144, 145, 262, 307 fig.8.1; 362;
365 (MLF) – 276; 366, 371,
221, 263-4, 295, 324, 367, 376, 561, 564, 565
376, 555, 556, 605
337; 367, 372, 376, 429,
395 (Cap) – 130, 241-3,
366 (MLF) ; 366, 371, 376, 439, 445, 496, 555, 560,
258, 295, 307 fig.8.1;
555, 556, 605 576, 605
362; 367, 372, 376, 561,
367-370 (MF) – 113, 121, 385 (EF/Etr) – 155, 157, 564, 576
122, 221, 253-4, 288, 158, 171-2, 172, 191,
396 (Cap) – 241-3, 260;
296, 305, 307 fig.8.1, 197, 298, 299, 323, 343;
362; 367, 372, 376, 561,
366, 372, 376, 555, 557 365, 373, 376, 382
565
fig.11.1, 393, 407, 439,
371-372 (MF) – 70, 90, 91
555, 561, 590, 605 397 (Cap) – 102, 240, 259,
fig.3.5, 118, 142, 145,
262; 362; 367, 372, 376,
221, 244, 245, 276, 307 386 (Cap) – 54, 178, 291;
561, 565
fig.8.1, 324, 325, 326, 362; 367, 372, 376, 561,
366, 372, 376, 555, 557, 561-2 398 (Cap) – 241-3; 362;
605 367, 372, 376, 561
387 (Cap) – 130, 221, 241-
373-375 (MF) – 244; 362, 3, 270, 295, 307 fig.8.1; 399 (Cap) – 91 fig.3.5, 130,
366, 372, 376, 383, 409, 367, 372, 376, 388, 561, 221, 250, 272, 295, 307
458, 555, 557, 563 561-2, 567, 575 fig.8.1; 367, 372, 376,
561, 565
376 (MF) – 93, 111 fig.3.9, 388 (Cap) – 94, 125, 206,
221, 250, 261, 307 221, 232 n.117, 234, 400 (Cap) – 208, 241-3,
fig.8.1; 366, 372, 376, 235, 240, 277, 307 272; 362; 367, 372, 376,
555, 558 fig.8.1; 367, 372, 376, 561, 565
386, 458, 561, 562, 563
377 (LtF) – 54, 84, 94, 102, 401 (Cap) – 362; 367, 372,
125, 178, 201, 225, 247, 389 (Cap) – 84, 94, 98, 102, 376, 561, 566
269, 283, 291, 317, 337; 125, 156, 157, 162-3,
666
INDICES
402 (Cap) – 362; 367, 372, 362; 367, 372, 376, 561, 376, 570, 571
376, 561, 566 568
434 (Lat) – 97, 102, 178,
403 (Cap) – 270; 367, 372, 420 (Cap) – 94, 99, 125, 201, 291, 317, 337; 368,
376, 386, 561, 566 178, 221, 246, 259; 367, 374, 376, 570, 572
372, 376, 558, 561, 568
404 (Cap) – 84, 102, 126, 435 (Cap) – 94, 102, 125,
156, 157, 162-3, 176, 421 (Cap) – 201, 225, 231, 178, 201, 225, 236, 246,
178, 197, 221, 240, 272, 235 fig.7.4, 246, 247, 249, 251-2, 252, 265,
282, 293, 294, 298, 308, 271, 291, 317; 367, 374, 277, 282, 291, 317; 367,
332, 333, 334, 337 376, 561, 568 374, 376, 549, 570, 572
n.173, 344; 367, 372,
422 (Cap) ; 362; 367, 376, 436 (Lat) – 125, 138, 201,
376, 386, 561, 563, 564,
561, 568 225, 236, 245, 291, 337;
566, 576
368, 374, 376, 388, 570,
423 (Cap) – 332, 334; 362;
405 (Cap) – 241-3; 362; 572
367, 372, 372, 376, 561,
367, 372, 376, 561, 566-
569 with fig.17.1 437 (Cap) – 97, 178, 201,
7
291; 367, 374, 376, 570,
424 (Cap) – 241-3, 275;
406 (Cap) – 241-3, 270; 572
362; 367, 372, 376, 561,
362; 367, 372, 376, 561,
569 438 (Lat) – 178, 201, 291;
566-7
368, 374, 376, 570, 573
425 (Cap) – 251-2, 269;
407 (Cap) – 241-3; 362;
362; 367, 372, 376, 561, 439-450 (Cap) ; 362; 367,
367, 372, 376, 561, 566-
569 372, 376, 573
7
426 (Cap) ; 362; 367, 372, 451 (Cap) – 267, 268; 362;
408 (Cap) – 241-3; 362;
376, 561, 569 367, 372, 376, 573
367, 372, 376, 561, 566-
7 427 (Cap) – 241-3, 260; 452 (Cap) – 267, 268; 362;
362; 367, 372, 376, 561, 367, 372, 376, 573
409 (Cap) – 248; 362; 367,
569
372, 376, 561, 566-7, 453 (Cap) – 362; 367, 372,
590 428 (Cap) – 245, 260; 362; 372, 376, 573
367, 372, 376, 561, 569
410 (Cap) ; 362; 367, 372, 454 (Cap) – 269; 362; 367,
376, 561, 567 429 (Cap) – 241-3, 276; 372, 376, 573
362; 367, 372, 376, 561,
411 (Cap) ; 362; 367, 372, 455 (Cap) – 362; 367, 372,
565, 569
376, 561, 567 376, 573
430 (Cap) – 91 fig.3.5, 102,
412 (Cap) ; 362; 367, 372, 456 (Lat) – 54, 108 fig.3.8,
130, 221, 250, 272, 295,
376, 561, 567 125, 178, 225, 235
307 fig.8.1; 367, 372,
fig.7.4, 241-3, 246, 251-
413 (Cap) ; 362; 367, 372, 376, 561, 570
2, 262, 276, 291, 317,
376, 561, 567
431 (Cap) – 97, 102, 114, 337; 368, 372, 375, 376,
414 (Cap) – 276; 362; 367, 123, 178, 192, 192 n.99, 570, 573
372, 376, 561, 567 200, 201, 236-7, 249,
457 (Cap) – 229, 239, 262,
251-2, 259, 270, 291,
415 (Cap) – 251-2, 271; 269, 307 fig.8.1; 367,
317, 334, 337 n.173,
362; 367, 372, 376, 561, 372, 376, 574
354; 367, 374, 376, 570,
567
570-1 458 (Cap) – 228, 241-3;
416 (Cap) ; 362; 367, 372, 367, 372, 376, 388, 574
432 (Lat) – 102, 125, 178,
376, 561, 567
201, 225, 245, 261, 291, 459 (MLF) – 239, 241-3,
417 (Cap) ; 362; 367, 372, 317, 337; 368, 374, 376, 269; 367, 372, 376, 505,
376, 561, 568 570, 571 574, 459
418 (Cap) ; 362; 367, 372, 433 (Cap) – 94, 97, 102, 460 (MLF) – 206, 274; 362,
376, 561, 568 114, 123, 125, 178, 201, 366, 372, 376, 574
291, 303, 317; 367, 374,
419 (Cap) – 241-3, 259; 461 (Cap) – 254; 362; 367,
667
INDICES
372, 376, 575 608 482† (Lat) – 140, 148, 306;

363, 368, 376, 378, 393,
462 (Cap) – 94, 221, 233, 471* (MF) – 13, 17, 59, 71,
471, 577, 578, 586, 588-
248, 266, 307 fig.8.1; 72, 78, 88, 93, 94, 96,
9
367, 376, 575, 590 97, 112, 113, 125, 156,
159, 160, 164-7, 167 483† (Lat) – 13, 132, 137;
463 (MLF) – 228, 275, 307
n.89, 175, 182, 184, 186, 363, 368, 376, 384, 577,
fig.8.1; 366, 372, 376,
225, 241-3, 262, 264, 586-7, 588-9
575
294, 299, 309, 344, 348,
484† (Lat) ; 363, 368, 376,
464 (MLF, falsum?) – 13 349, 352; 363, 366, 372,
577, 587-9
n.4, 94, 176, 221, 275, 376, 487, 525, 577, 579,
307 fig.8.1, 325, 337 580, 584, 585
n.137; 362, 366, 372,
472* (MF) – 130, 221, 246, Etruscan inscriptions
376, 575
295, 307 fig.8.1; 363,
465 (Cap) – 84, 87, 88, 90, 366, 372, 376, 577, 580 I (Fa 9.1) – 36, 323; 363,
91 fig.3.5, 102, 112, 368, 373, 376, 379, 393,
473* (MF) – 93, 97, 98, 99, 593, 593-4
128, 156, 157, 162-3,
221, 241-3, 270, 307
178, 197, 221, 258, 258- II (Fa 0.1) – 36, 323; 363,
fig.8.1; 363, 366, 372,
9, 293, 294, 298, 308, 368, 372, 376, 393, 593,
376, 577, 580
332, 334, 344; 367, 372, 593-4
372, 376, 576 474* (MLF/Cap) – 50, 74,
100, 118, 130, 131, 191, III (Fa 2.1+6.2) – 36, 323;
466 (Cap) – 118, 228, 250, 363, 368, 372, 372, 376,
200, 225, 241-3, 273,
272, 334; 367, 372, 376, 393, 409, 593, 594-5
282, 302, 324, 332, 334,
576 354; 363, 372, 373, 376, with fig.19.1
467* (EF) – 58, 60, 83, 111 383, 493, 494, 577, 580- IV (Fa 0.2) – 36, 323; 363,
fig.3.9, 129, 137, 140, 1 368, 373, 376, 393, 395,
148, 184, 216, 218, 237- 593, 594-5 with
475* (MLF/Cap) – 113,
8, 306, 327, 343; 363, fig.19.1, 607
121, 122, 221, 243-4,
365, 372, 375, 381, 382
296, 305, 307 fig.8.1; V (Fa 2.2) – 36, 323; 363,
fig.11.1, 393, 401, 415,
363, 372, 376, 555, 577, 368, 372, 376, 393, 593,
577, 577-8
581 595, 597
468* (Sab) – 78, 126, 214,
476* (MLF/Cap) – 90, 94, VI-VII – 36, 98, 265, 323;
247, 260, 332; 363, 368,
98, 221, 272-3, 307 363, 368, 372, 376, 393,
372, 376, 393, 555, 577,
fig.8.1; 363; 366, 376, 593, 595
578
577, 581-2
469* (MF) – 113, 130, 221, VIII (Fa 3.1+6.1) – 36, 239,
477* (Lat) – 271, 295, 309; 306 n.153, 317, 323;
253-4, 259, 261, 295,
363, 372, 376, 493, 577, 363, 368, 373, 376, 393,
307 fig.8.1, 324; 363,
578, 582 593, 595-6
366, 372, 376, 438, 577,
579 478* (Lat) – 99, 271, 295, IX (Fa 3.3) – 36, 317, 323;
309; 363, 372, 376, 532, 363, 368, 373, 376, 393,
470* (MF) – 13, 17, 59, 71,
577, 582 593, 596
72, 80, 87, 88, 93, 97,
100, 107 fig.3.7, 112, 479† (Lat) – 35, 120, 140, X (Fa 0.4) – 36, 205 n.107,
113, 125, 130, 148, 149, 148, 149, 165, 184, 199, 323; 363, 368, 373, 376,
156, 159, 160, 164-7, 306; 363, 376, 378, 401, 393, 593, 596
167 n.89, 175, 182, 184, 577, 583-4
184-5, 190-1, 221, 225, XI-XV (Fa 2.6-10) – 36,
480† (Sab, Um 4) – 150, 219, 245, 275, 323, 325;
231, 234 fig.7.3, 239,
190, 332; 363, 368, 376, 363, 368, 372, 376, 393,
262, 267, 290, 293, 299,
380, 393, 577, 584-5, 593, 595, 597 , 597
309, 326, 327, 344, 345,
605
348, 349, 352; 363, 366, XVI (Fa 2.12) – 214; 363,
372, 376, 505, 509, 525, 481† (Lat) ; 363, 368, 376, 368, 372, 376, 393, 593,
577, 579, 580, 584, 585, 378, 380, 577, 585 597
668
INDICES
XVII (Fa 2.14) – 219, 245; 368, 374, 376, 457, 593, XLI (Fa S.4) – 35, 205
363, 368, 376, 393, 593, 600, 601, 603, 605 n.107, 368, 376, 593,
595, 597, 597, 603 601, 603, 605
XXX – 323 , 368, 372, 376,
XVIII – 36, 323; 363, 368, 593, 599, 601 XLII (Fa 3.4) – 121, 122,
374, 376, 393, 593, 598 126, 131, 253-4, 288,
XXXI (falsum?) – 35, 202,
323, 328, 368, 372, 376,
XIX (Fa 3.2) – 36, 73, 105 205 n.107, 323; 362,
393, 579, 593, 605
n.51, 216, 217, 245, 252, 368, 373, 376, 525, 593,
306 n.153, 317, 323, 599, 601 XLIII (Fa 1.7) – 12, 209,
325, 328; 363, 368, 373, 274, 323, 368, 371, 376,
XXXII (Fa 2.15) – 99, 205
376, 393, 593, 597, 598 388, 449, 500, 555, 556,
n.107, 214, 273, 305,
593, 606
XX (Fa 1.6) – 36, 270, 323; 323, 329; , 368, 372,
593, 598-9 376, 393, 433, 593, 602 XLIV (Fa 9.2) – 323, 368,
373, 376, 379, 393, 411,
XXI (Fa 2.4) – 261, 323, XXXIII (Fa S.3) – 35, 368,
561, 593, 594, 606
368, 372, 376, 393, 593, 373, 376, 414, 522, 593,
599 602 XLV (Fa 2.25) – 92, 241-3,
242, 323, 325, 328, 368,
XXII (Fa 2.11a) , 368, 372, XXXIV (Fa 1.3) – 219, 245,
372, 376, 384, 570, 585,
376, 393, 593, 599 252, 261, 282, 325, 328;
593, 606 , 606
362, 368, 371, 376, 384,
XXIII (Fa 2.11b) , 368,
521, 524, 593, 601, 602, XLVI (Fa 2.3) – 368, 373,
372, 376, 393, 593, 599
602-3, 605 376, 393, 395, 593, 607,
XXIV (Fa 0.5) , 368, 372, 607
XXXV (Fa 1.4) – 99, 143,
376, 393, 593, 599
207, 219, 245, 269; 362, XLVII (Fa 6.3) – 368, 373,
XXV (Fa G.1a-b) – 35, 205 368, 371, 376, 547, 558, 373, 376, 393, 593, 600,
n.107, 323; , 368, 373, 593, 602, 603, 604 607, 607
376, 393, 593, 599, 600
XXXVI – 368, 373, 376, XLVIII – 113, 216, 267,
XXVI – 323, 325, 328; 368, 529, 593, 597, 603-4 290, 325, 328, 368, 373,
372, 376, 393, 593, 600, 376, 393, 579, 593, 607,
XXXVII – 368, 372, 376,
600 607-8
593, 603
XXVII (Fa S.1) – 35, 205 XLIX (Fa 2.5) – 362, 368,
XXXVIII (Fa 0.8) – 252,
n.107, 238, 323; , 368, 373, 376, 393, 593, 607,
322, 339, 368, 374, 376,
373, 376, 593, 599, 600 608
593, 602, 604
XXVIII (Fa S.2) – 35, 204, L (Fa 7.1a) – 35, 368, 373,
XXXIX (Fa 0.7) – 219, 245,
205 n.107, 317, 323; 376, 593, 607, 608
322, 325, 339, 368, 374,
368, 373, 376, 593, 599,
376, 593, 602, 604 LI (Fa 7.1b) – 368, 373,
600 376, 524, 593, 607, 608
XL (Fa 2.16) – 205 n.107,
XXIX (Fa 0.6) – 36, 63,
279 n.140, 368, 372,
105, 245, 259, 323, 328;
376, 593, 604-5
3. OTHER INSCRIPTIONS
Etruscan ET AH 1.80 – 261 ET Ar 1.13 – 246-7

(Etruscan inscriptions from ET AH 2.3 – 245 ET Ar 1.53 – 100
the ager Faliscus are listed ET AH 3.4 – 306 n.153 ET Ar 1.94 – 252
under I-LI in Index 2) ET AH 4.1 – 314 ET Ar 3.1 – 314
ET Ar 0.3 – 100 ET Ar 3.2 – 317
ET AH 1.8 – 253 ET Ar 1.1 – 252 ET Ar 4.2 – 317
ET AH 1.67 – 250 ET Ar 1.4 – 208, 243, 265 ET AS 1.9 – 313
ET AH 1.74 – 209 ET Ar 1.7 – 100 ET AS 1.11 – 260
ET AH 1.81 – 313 ET Ar 1.9 – 252 ET AS 1.40 – 36, 208
669
INDICES
ET AS 1.129 – 209 ET Cl 1.318 – 214 ET Cl 1.1669 – 208

ET AS 1.160 – 260 ET Cl 1.319 – 214 ET Cl 1.1682 – 263
ET AS 1.174 – 274 ET Cl 1.320 – 214 ET Cl 1.1686 – 208
ET AS 1.227 – 260; 487 ET Cl 1.352 – 208 ET Cl 1.1744 – 208
ET AS 1.232 – 263 ET Cl 1.388 – 263 ET Cl 1.1769 – 273
ET AS 1.236 – 100 ET Cl 1.394 – 267 ET Cl 1.1770 – 273
ET AS 1.316 – 263 ET Cl 1.395 – 267 ET Cl 1.1771 – 273
ET AS 1.393 – 311 ET Cl 1.454 – 218 ET Cl 1.1913 – 208
ET AS 1.395 – 208, 274 ET Cl 1.501 – 266 ET Cl 1.1955 – 267
ET AS 1.472 – 311 ET Cl 1.578 – 218 ET Cl 1.2008 – 268
ET AS 3.1 – 306 n.153 ET Cl 1.633 – 218 ET Cl 1.2009 – 268
ET AT 1.4 – 105 n.51 ET Cl 1.725 – 250 ET Cl 1.2010 – 268
ET AT 1.7 – 105 n.51 ET Cl 1.794 – 208 ET Cl 1.2026 – 269
ET AT 1.8 – 105 n.51 ET Cl 1.835 – 80 ET Cl 1.2027 – 269
ET AT 1.9 – 105 n.51 ET Cl 1.843 – 276 ET Cl 1.2028 – 249
ET AT 1.11, 313 ET Cl 1.844 – 276 ET Cl 1.2035 – 270
ET AT 1.12 – 105 n.51 ET Cl 1.845 – 276 ET Cl 1.2079 – 248
ET AT 1.28 – 253 ET Cl 1.846 – 276 ET Cl 1.2080 – 248
ET AT 1.30, 313 ET Cl 1.848 – 276 ET Cl 1.2109 – 248
ET AT 1.49 – 105 n.51 ET Cl 1.849 – 276 ET Cl 1.2173 – 273
ET AT 1.67 – 261 ET Cl 1.850 – 276 ET Cl 1.2179 – 248
ET AT 1.70 – 313 ET Cl 1.851 – 276 ET Cl 1.2214 – 218
ET AT 1.71 – 597 ET Cl 1.852 – 276 ET Cl 1.2206 – 267
ET AT 1.74 – 253 ET Cl 1.946 – 314 ET Cl 1.2207 – 267
ET AT 1.102 – 245 ET Cl 1.969 – 269 ET Cl 1.2344 – 248
ET AT 1.107 – 105 n.51 ET Cl 1.1045 – 249 ET Cl 1.2466 – 208
ET AT 1.108 – 105 n.51 ET Cl 1.1102 – 269 ET Cl 1.2467 – 208
ET AT 1.111 – 105 n.51 ET Cl 1.1103 – 269 ET Cl 1.2589 – 267
ET AT 1.138 –313 ET Cl 1.1136 – 267 ET Cl 1.2609 – 258; 508
ET AT 1.140 – 313 ET Cl 1.1241 – 260 ET Cl 1.2611 – 207, 258;
ET AT 1.141 – 313 ET Cl 1.1243 – 249 508
ET AT 1.145 – 261 ET Cl 1.1280 – 208, 265 ET Cl 1.2612 – 207, 258;
ET AT 1.146 – 313 ET Cl 1.1281 – 208 508
ET AT 1.148 – 313 ET Cl 1.1282 – 208 ET Cl 1.2613 – 207, 258;
ET AT 1.149 – 313 ET Cl 1.1294 – 209, 274 508
ET AT 1.159 – 313 ET Cl 1.1298 – 260 ET Cl 1.2620 – 209, 274
ET AT 1.163 – 313 ET Cl 1.1302 – 208 ET Cl 1.2621 – 209, 274
ET AT 1.177 – 313 ET Cl 1.1308 – 260 ET Cl 1.2673 – 207, 263
ET AT 1.178 – 313 ET Cl 1.1327 – 252 ET Cl 2.11 – 265
ET AT 1.188 – 313 ET Cl 1.1328 – 252 ET Cl 2.18 – 267
ET AT 1.192 – 313 ET Cl 1.1347 – 260 ET Cl 2.26 – 603
ET AT 1.200 – 218 ET Cl 1.1428 – 66 ET Cm 3.1 – 186
ET AT 3.1 – 242, 306 n.153 ET Cl 1.1429 – 66 ET Cm 2.8 – 246-7
ET AT 4.1 – 317 ET Cl 1.1455 – 263 ET Cm 2.48 – 246-7
ET AT 5.2 – 245, 264 ET Cl 1.1470 – 267 ET Cm 2.49 – 261
ET AV 1.5 – 313 ET Cl 1.1467 – 276 ET Cm 2.57 – 208, 266
ET AV 1.8 – 313 ET Cl 1.1510 – 208 ET Cm 2.83 – 414
ET AV 1.14 – 313 ET Cl 1.1511 – 208 ET Cm 2.84 – 414
ET AV 2.5 – 603 ET Cl 1.1512 – 208 ET Co 1.3 – 66
ET AV 2.11 – 251 ET Cl 1.1513 – 208 ET Co 3.1 – 603
ET AV 6.2 – 105 n.51 ET Cl 1.1514 – 208 ET Co 4.7 – 317
ET Cl 1.102 – 261 ET Cl 1.1515 – 208 ET Co 4.8 – 317
ET Cl 1.201 – 218 ET Cl 1.1524 – 262 ET Co 4.9 – 317
ET Cl 1.202 – 218 ET Cl 1.1568 – 263 ET Cr 1.5 – 311
ET Cl 1.203 – 218 ET Cl 1.1596 – 269 ET Co 1.28 – 66
ET Cl 1.229 – 263 ET Cl 1.1619 – 273 ET Cr 1.22 – 597
670
INDICES
ET Cr 1.59 – 269 ET Li 1.1 – 314 ET Pe 1.875 – 274

ET Cr 1.64 – 269 ET Li 4.1 – 314 ET Pe 1.880 – 274
ET Cr 1.66 – 269 ET OA 0.1 – 100 ET Pe 1.889 – 208
ET Cr 1.100 – 273 ET OA 2.11 – 242 ET Pe 1.891 – 606
ET Cr 1.149 – 208 ET OA 2.52 – 242 ET Pe 1.896 – 260
ET Cr 1.152 – 273 ET OA 2.58 – 219 ET Pe 1.897 – 260
ET Cr 1.155 – 273 ET OA 2.62 – 261 ET Pe 1.898 – 260
ET Cr 1.172 – 273 ET OA 2.63 – 251 ET Pe 1.904 – 263
ET Cr 2.5 – 603 ET OA 3.1 – 306 n.153 ET Pe 1.928 – 237
ET Cr 2.6 – 603 ET OA 3.9 – 306 n.153 ET Pe 1.943 – 260
ET Cr 2.18 – 186 ET OB 3.2 – 105 n.51 ET Pe 1.951 – 258
ET Cr 2.19 – 186 ET OI S.5 – 202 ET Pe 1.965 – 273
ET Cr 2.20 – 603 ET OI S.46 – 203 ET Pe 1.973 – 218
ET Cr 2.29 – 603 ET OI S.52 – 276 ET Pe 1.1031 – 254
ET Cr 2.56 – 242 ET Pa 3.1 – 265 ET Pe 1.1087 – 254
ET Cr 2.30 – 186 ET Pe 0.6 – 317 ET Pe 1.1091 – 266
ET Cr 2.31 – 273 ET Pe 1.1080 – 268 ET Pe 1.1126 – 268
ET Cr 2.42 – 251 ET Pe 1.53 – 261 ET Pe 1.1127 – 268
ET Cr 2.54 – 262 ET Pe 1.54 – 261 ET Pe 1.1132 – 260
ET Cr 2.55 – 262 ET Pe 1.71 – 250 ET Pe 1.1190 – 250, 273
ET Cr 2.57 – 262 ET Pe 1.86 – 237 ET Pe 1.1191 – 250, 263,
ET Cr 2.74 – 242, 269 ET Pe 1.93 – 260 273
ET Cr 2.139 – 245 ET Pe 1.115 – 606 ET Pe 1.1211 – 66, 249
ET Cr 3.4 – 306 n.153 ET Pe 1.135 – 250 ET Pe 1.1235 – 207, 263
ET Cr 3.5 – 306 n.153 ET Pe 1.142 – 275 ET Pe 1.1242 – 237
ET Cr 3.6 – 306 n.153 ET Pe 1.150 – 105 n.51 ET Pe 1.1264 – 606
ET Cr 3.7 – 306 n.153 ET Pe 1.168 – 313 ET Pe 1.1267 – 247; 606
ET Cr 3.8 – 306 n.153 ET Pe 1.197 – 269 ET Pe 1.1268 – 274
ET Cr 3.10 – 306 n.153 ET Pe 1.198 – 269 ET Pe 1.1297 – 260
ET Cr 3.12 – 306 n.153 ET Pe 1.201 – 66, 249 ET Pe 1.1441 – 261
ET Cr 3.13 – 267, 306 n.153 ET Pe 1.213 – 218 ET Pe 3.3 – 306 n.153
ET Cr 3.14 – 240 ET Pe 1.214 –218 ET Pe 5.3 – 311
ET Cr 3.15 – 306 n.153 ET Pe 1.277 – 254 ET Po 2.21 – 603
ET Cr 3.16 – 306 n.153 ET Pe 1.323 – 261 ET Po 4.4 – 251
ET Cr 3.17 – 168 n.91 ET Pe 1.325 – 261 ET Ru 0.13 – 263
ET Cr 3.18 – 267, 306 n.153 ET Pe 1.326 – 261 ET Ru 2.4 – 260
ET Cr 3.20 – 306 n.153 ET Pe 1.327 – 261 ET Ru 2.5 – 260
ET Cr 3.23 – 306 n.153 ET Pe 1.336 – 276 ET Ru 3.1 – 244
ET Cr 4.1 – 204, 315 ET Pe 1.436 – 218 ET Ru 4.3 – 315
ET Cr 4.4 – 66 ET Pe 1.485 – 207, 263 ET Sp 2.71 – 250
ET Cr 4.9 – 315 ET Pe 1.505 – 261 ET Sp 2.76 – 260
ET Cr 4.12 – 317 ET Pe 1.564 – 254 ET Ta 1.9 – 262, 270
ET Cr 4.17 – 315 ET Pe 1.565 – 254 ET Ta 1.13 – 270
ET Cr 4.18 – 315 ET Pe 1.630 – 66, 249 ET Ta 1.14 – 262
ET Cr 5.1 – 242 ET Pe 1.638 – 606 ET Ta 1.15 – 270
ET Cr 6.2 – 100 ET Pe 1.639 – 606 ET Ta 1.31 – 262, 313
ET Cs 2.3 – 240 ET Pe 1.686 – 263 ET Ta 1.39 – 253
ET Fa, see I-LI in Index 2 ET Pe 1.688 – 313 ET Ta 1.45 – 313
ET Fe 1.2 – 314 ET Pe 1.748 – 261 ET Ta 1.50 – 269
ET Fe 1.7 – 314 ET Pe 1.756 – 606 ET Ta 1.51 – 269
ET Fe 1.9 – 314 ET Pe 1.813 – 268 ET Ta 1.66 – 181, 311, 326
ET Fe 2.7 – 208 ET Pe 1.817 – 247; 606 ET Ta 1.93 – 253
ET Fe 2.15 – 242 ET Pe 1.846 – 100 ET Ta 1.95 – 262, 263
ET Fe 3.3 – 260 ET Pe 1.865 – 274 ET Ta 1.96 – 262
ET La 2.3 – 413 ET Pe 1.869 – 218 ET Ta 1.97 – 262
ET La 3.1 – 306 n.153 ET Pe 1.871 – 237 ET Ta 1.113 – 270
671
INDICES
ET Ta 1.116 – 261 ET Vc 1.53 – 276 ET Vs 1.151 – 262

ET Ta 1.166 – 261 ET Vc 1.59 – 311 ET Vs 1.152 – 265
ET Ta 1.182 – 29 n.16 ET Vc 1.69 – 313 ET Vs 1.153 – 262
ET Ta 1.185 – 269 ET Vc 1.77 – 313 ET Vs 1.154 – 262
ET Ta 1.216 – 105 n.51 ET Vc 1.78 – 314 ET Vs 1.170 – 263
ET Ta 1.237 – 253 ET Vc 1.87 – 312 ET Vs 1.190 – 251
ET Ta 1.250 – 269 ET Vc 1.92 – 276 ET Vs 1.202 – 263
ET Ta 1.1222 – 267 ET Vc 1.102 – 313 ET Vs 1.203 – 276
ET Ta 3.1 – 306 n.153 ET Vc 2.10 – 175 ET Vs 1.208 – 262
ET Ta 5.2 – 264 ET Vc 2.34 – 208 ET Vs 1.282 – 245
ET Ta 5.4 – 264, 311 ET Vc 2.52 – 603 ET Vs 1.287 – 251
ET Ta 5.5 – 264 ET Vc 3.2 – 306 n.153 ET Vs 1.299 – 263
ET Ta 6.12 – 105 n.51 ET Vc 3.6 – 306 n.153 ET Vs 1.92 – 218
ET Ta 6.15 – 208 ET Vc 4.1 – 315 ET Vs 1.99 – 242
ET Ta 1.17 – 208 ET Vc 4.2 – 315 ET Vs 1.133 – 253
ET Ta 1.51 – 311 ET Vc 4.3 – 315 ET Vs 1.159 – 242
ET Ta 1.52 – 311 ET Vc 4.4 – 315 ET Vs 1.183 – 264
ET Ta 1.53 – 311 ET Vc 6.6 – 105 n.51 ET Vs 1.231 – 253
ET Ta 1.149 – 245 ET Vc 6.12 – 245 ET Vs 1.233 – 253
ET Ta 1.175 – 245 ET Vc 7.24 – 239 ET Vs 1.244 – 260
ET Ta 1.205 – 311 ET Vc 7.30 – 36 ET Vs 1.268 – 246-7
ET Ta 1.220 – 245 ET Vc S.1 – 202 ET Vs 1.307 – 263
ET Ta 1.217 – 208 ET Ve 2.1 – 186 ET Vs 2.1 – 240
ET Ta 1.256 – 263 ET Ve 2.4 – 208, 266 ET Vs 2.7 – 603
ET Ta 2.1 – 251 ET Ve 3.1 – 306 n.153 ET Vs 2.35 – 264
ET Ta 3.1 – 246-7 ET Ve 3.4 – 218 ET Vs 4.7 – 314
ET Ta 4.2 – 314 ET Ve 3.10 – 204 ET Vs 4.10 – 317
ET Ta 4.8 – 315 ET Ve 3.19 – 113, 265, 275 ET Vs 4.11 – 317
ET Ta 4.9 – 315 (with n.138) ET Vs 4.13 – 317
ET Ta 4.11 – 315 ET Ve 3.29 – 204 ET Vs 4.14 – 317
ET Ta 4.13 – 315 ET Ve 3.32 – 252 ET Vs 7.34 – 264
ET Ta 4.12 – 314 ET Ve 3.33 – 204 ET Vs 7.38 – 264
ET Ta 5.3 – 311 ET Ve 4.1 – 315 ET Vs 7.35 – 264
ET Ta 7.29 – 276 ET Ve 6.2 – 309 n.156 ET Vs 7.36 – 264
ET Ta 7.31 – 245, 251 ET Vn 1.1 – 26, 36, 208, ET Vs S.4 – 239
ET Ta 7.40 – 263 239 ET Vt 1.73 – 100
ET Ta 7.3 – 251 ET Vn 2.7 – 243 ET Vt 1.110 – 263
ET Ta S.8 – 605 ET Vs 0.23 – 264 ET Vt 1.124 – 242
ET Ta S.12 – 203 ET Vs 1.9 – 276 ET Vt 1.137 – 270
ET TC 2 – 274 ET Vs 1.28 – 264 ET Vt 4.1 – 270
ET TC 16 – 274 ET Vs 1.29 – 248 ET Vt 4.6 – 243
ET TC 26 – 274 ET Vs 1.38 – 267
CIE 8380 (falsum) – 604
ET TC 28 – 274 ET Vs 1.43 – 314
NRIE 991 – 575
ET TC 37 – 274 ET Vs 1.48 – 262
TLE 495 – 261
ET Um 4.1 – 314 ET Vs 1.54 – 314
ET Um 4.2 – 314 ET Vs 1.58 – 244
ET Vc 0.40 – 181, 326 ET Vs 1.73 – 314 Greek
ET Vc 1.1 – 313 ET Vs 1.86 – 314 IG 13.1658 add – 435
ET Vc 1.2 – 313 ET Vs 1.98 – 314 SIG 558 – 270
ET Vc 1.3 – 313 ET Vs 1.88 – 259
ET Vc 1.4 – 313 ET Vs 1.99 – 264 Hernician
ET Vc 1.5 – 208, 313 ET Vs 1.116 – 186
ET Vc 1.10 – 313 ET Vs 1.120 – 186 ST He 2 – 77, 198, 343; 408
ET Vc 1.16 – 313 ET Vs 1.126 – 276 ST He 3 – 162, 197, 251-2,
ET Vc 1.31 – 276 ET Vs 1.136 – 273, 314 333
ET Vc 1.45 – 313 ET Vs 1.138 – 258
672
INDICES
Latin
2
CIL I .347 – 90 CIL I2.560 – 90, 192
CIL I2.350 – 110, 241 CIL I2.561 – 160 , 164, 300
(Latin inscriptions in my CIL I2.360 – 143 fig.4.1 n.150, 309
edition are listed in Index 2) CIL I2.361 – 143 fig.4.1 CIL I2.563 – 103, 126, 202,
CIL I2.1 – 61, (63), 93, 126, CIL I2.362 – 316 333
158, 195, 354; 408, 588, CIL I2.364 – 204 n.106 CIL I2.564 – 82, 203
589 CIL I2.366 – 110, 132, 159 CIL I2.566 – 82
CIL I2.2 – 90 CIL I2.368 – 316 CIL I2.568 – 203
CIL I2.3 – 158, 160, 164-7, CIL I2.369 – 316 CIL I2.580 – 114, 123, 316
175, 184, 343 CIL I2.370 – 316 CIL I2.581 – 88, 93, 110,
CIL I2.4 – 126, 149, 158, CIL I2.371 – 106 n.54 143 fig.4.1, 144, 146,
160, 164, 180, 188, 304, CIL I2.373 – 316 149, 150 n.80
343, 354; 411, 588, 589 CIL I2.375 – 314 CIL I2.582 – 149
CIL I2.5 – 109 CIL I2.376 – 110 CIL I2.583 – 90
CIL I2.7 – 133 CIL I2.377 – 93 CIL I2.584 – 90
CIL I2.8 – 143 CIL I2.378 – 90, 106 n.54 CIL I2.586 – 146
CIL I2.9 – 88, 104, 109, 128 CIL I2.379 – 93, 106 CIL I2.589 – 143 fig.4.1
CIL I2.14 – 160 CIL I2.380 – 93 CIL I2.969 – 120 n.63
CIL I2.15 – 160 CIL I2.381 – 316 CIL I2.970 – 314
CIL I2.22 – 93 CIL I2.388 – 112, 123 CIL I2.971 – 316
CIL I2.25 – 160 CIL I2.397 – 316 CIL I2.973 – 143 fig.4.1,
CIL I2.30 – 93 n.47, 292; CIL I2.401 – 98 203
549 n.285 CIL I2.404 – 316 CIL I2.975 – 120 n.63, 316
CIL I2.39 – 316 CIL I2.412a – 149 CIL I2.991 – 316
CIL I2.45 – 191, 315 CIL I2.412c – 149 CIL I2.995 – 316
CIL I2.47 – 160 CIL I2.416 – 160 CIL I2.1028 – 103, 126, 333
CIL I2.47b – 93 CIL I2.439 – 120 CIL I2.1220 – 230
CIL I2.48 – 143, 160 CIL I2.440 – 132 CIL I2.1259 – 182, 311
CIL I2.49 – 143, 160 CIL I2.441 – 120 CIL I2.1289 – 230
CIL I2.52 – 97 CIL I2.443 – 120 CIL I2.1294 – 230
CIL I2.59 – 93 n.47 CIL I2.444 – 143 fig.4.1 CIL I2.1312 – 182 n.97, 311
CIL I2.60 – 143 fig.4.1 CIL I2.445 – 132 n.158
CIL I2.61 – 93, 292; 549 CIL I2.446 – 120 CIL I2.1328 – 230
n.285 CIL I2.447 – 120 CIL I2.1340 –90
CIL I2.62 – 143 fig.4.1 CIL I2.449 – 132 CIL I2.1349 – 230
CIL I2.77 – 90 CIL I2.450 – 143 fig.4.1 CIL I2.1352 – 230
CIL I2.79 – 90 CIL I2.451 – 143 fig.4.1 CIL I2.1413 – 110
CIL I2.126 – 90, 143 fig.4.1 CIL I2.452 – 120 CIL I2.1424 – 230
CIL I2.139 – 90 CIL I2.453 – 132 CIL I2.1430 – 316
CIL I2.161 – 90 CIL I2.454 – 204 n.106 CIL I2.1432 – 230
CIL I2.166 – 82 CIL I2.455 – 61, 62 CIL I2.1435 – 314
CIL I2.171 – 230 CIL I2.462 – 306 CIL I2.1439 – 316
CIL I2.184 – 230 CIL I2.477 – 93 CIL I2.1446 – 82
CIL I2.194 – 90 CIL I2.479 – 120, 148, 306; CIL I2.1447 – 499
CIL I2.197 – 90 588 CIL I2.1480 – 315-6
CIL I2.198 – 90 CIL I2.500 – 149 CIL I2.1493 – 252; 427
CIL I2.233 – 90 CIL I2.501 – 149 CIL I2.1497 – 252; 427
CIL I2.270 – 90 CIL I2.547 – 92; 495 CIL I2.1509 – 316
CIL I2.271 – 90 CIL I2.548 – 63, 63 n.38 CIL I2.1513 – 90
CIL I2.288 – 230 CIL I2.550 – 92 CIL I2.1529 – 172, 175, 197
CIL I2.300 – 230 CIL I2.552 – 103, 126, 333 CIL I2.1536 – 230
CIL I2.312 – 99 CIL I2.553 – 98, 297, 326 CIL I2.1579 – 316
CIL I2.329 – 90 CIL I2.555 – 132, 133 CIL I2.1581 – 316
CIL I2.330 – 90 CIL I2.558 – 90, 297, 326 CIL I2.1582 – 316
CIL I2.344 – 90 CIL I2.559 – 112, 240, 243 CIL I2.1595 – 230
CIL I2.346 – 143 fig.4.1 n.126 CIL I2.1616 – 93 n.47
673
INDICES
2
CIL I2.1618 – 93 n.47 CIL I .2444 – 314 n.135)
CIL I2.1626 – 316 CIL I2.2460 – 230 CIL XI.2977 – 250
CIL I2.1626a – 315 CIL I2.2471 – 90 CIL XI.2980 – 265
CIL I2.1635 – 93 CIL I2.2482 – 90 CIL XI.2990 – 273
CIL I2.1638 – 182 , 311 CIL I2.2486 – 99 CIL XI.3038 – 274
CIL I2.1824 – 90, 230 CIL I2.2531 – 316 CIL XI.3064 – 274
CIL I2.1829 – 90, 230 CIL I2.2540c – 93 n.47 CIL XI.3080 – 259 (with
CIL I2.1838 – 316 CIL I2.2542 – 93 n.133)
CIL I2.1843 – 230 CIL I2.2628 – 316 CIL XI.3083 – 34, 262-3, 43
CIL I2.1886 – 230 CIL I2.2630 – 316 CIL XI.3089 – 43, 43 n.27
CIL I2.1907 – 230 CIL I2.2631 – 122, 316 CIL XI.3090 – 43 n.27
CIL I2.1990 – 188-9 CIL I2.2636 – 230 CIL XI.3091 – 43 n.27
CIL I2.1993 – 316 CIL I2.2658 – 97, 149, 303; CIL XI.3092 – 43 n.27
CIL I2.2081 – 90 413, 588 CIL XI.3093 – 43 n.27
CIL I2.2111 – 123 CIL I2.2830 – 195; 588 CIL XI.3094 – 43, 43 n.27
CIL I2.2117 – 315 CIL I2.2832a – 2, 98, 129, CIL XI.3100 – 32, 44
CIL I2.2135 – 182, 311 133, 136, 140, 161, 343, CIL XI.3101 – 43
CIL I2.2171 – 316 349, 352, 354; 577, 588, CIL XI.3109 – 267
CIL I2.2194 – 316 589 CIL XI.3110 – 269 (with
CIL I2.2197 – 146 CIL I2.2833 – 63, 316; 588 n.135)
CIL I2.2219 – 315 2833a – 588 CIL XI.3112 – 43, 44
CIL I2.2220 – 316 CIL I2.2845 – 61, (63); 589 CIL XI.3115 – 258
CIL I2.2221 – 316 CIL I2.2846 – 61, (63); 589 CIL XI.3116 – 43, 258
CIL I2.2232 – 499 CIL I2.2848a – 316 CIL XI.3117 – 258
CIL I2.2233 – 316; 499 CIL I2.2872 – 147 CIL XI.3118 – 258
CIL I2.2234 – 499 CIL I2.2876 – 316 CIL XI.3119 – 258
CIL I2.2235 – 499 CIL I2.2877 – 118 n.60 CIL XI.3121 – 43
CIL I2.2236 – 499 CIL I2.2884b – 143 fig.4.1 CIL XI.3123 – 44
CIL I2.2237 – 499 CIL I2.2885 – 143 fig.4.1 CIL XI.3125 – 32, 43, 44
CIL I2.2238 – 499 CIL I2.2887b – 316 CIL XI.3127 – 43
CIL I2.2239 – 499 CIL I2.2889 – 316 CIL XI.3132 – 274
CIL I2.2240 – 499 CIL I2.2909 – 114, 123 CIL XI.3136 – 44
CIL I2.2241 – 499 CIL I2.2916g – 588 CIL XI.3147 – 43
CIL I2.2242 – 499 CIL I2.2916k – 588 CIL XI.3155a,1 – 43
CIL I2.2243 – 316; 499 CIL I2.2917a – 258, 259; CIL XI.3156 – 265
CIL I2.2244 – 499 588 CIL XI.3166 – 269; 574
CIL I2.2245 – 499 CIL I2.2917b – 120; 588 CIL XI.3170 – 265
CIL I2.2246 – 316; 499 CIL I2.2917c – 148 CIL XI.3174 – 266
CIL I2.2247 – 499 CIL I2.3171 – 316 CIL XI.3179 – 275
CIL I2.2248 – 499 CIL I2.3375 – 316 CIL XI.3181 – 267, 272;
CIL I2.2249 – 499 CIL I2.3472a – 316 582
CIL I2.2250 – 499 CIL I2.3472b – 316 CIL XI.3189 – 238
CIL I2.2251 – 499 CIL I2.3474 – 316 CIL XI.3206 – 271
CIL I2.2252 – 499 CIL III.10444 – 238 CIL XI.3207 – 270
CIL I2.2253 – 499 CIL V.4251 – 258 CIL XI.3208 – 271
CIL I2.2254 – 499 CIL VI.1302 – 241 n.124 CIL XI.3223 – 273
CIL I2.2255 – 499 CIL VI.4925 – 262 CIL XI.3254 – 267, 281
CIL I2.2256 – 499 CIL VI.6945 – 260 CIL XI.3257 – 262-3
CIL I2.2257 – 499 CIL VI.21470 – 275 CIL XI.3264 – 265
CIL I2.2258 – 499 CIL VI.25097 – 271 (with CIL XI.3327 – 261
CIL I2.2259 – 499 n.137) CIL XI.3370 – 442
CIL I2.2284 – 230 CIL VIII.15474 – 257 n.132 CIL XI.3447 – 266
CIL I2.2296 – 316 CIL X.6098 – 251 CIL XI.3782 – 246
CIL I2.2438 – 93, 160 CIL X.8230 – 273 CIL XI.3805 – 207, 275;
CIL I2.2442 – 292; 549 CIL XI.2004 – 262 490
n.285 CIL XI.2958 – 269 (with CIL XI.3843 – 275
674
INDICES
CIL XI.3855 – 490 ST VM 5 – 192, 314, 334; ST Cp 30 – 173

CIL XI.3864 – 267 571 ST Cp 31 – 210
CIL XI.3887 – 267 ST VM 6 – 202 ST Cp 32 – 195
CIL XI.3930 – 44 ST VM 7 – 202 ST Cp 33 – 180, 200
CIL XI.3934 – 266 ST Cp 34 – 180, 200
NDI 223 – 166 n.88
CIL XI.3943 – 571 ST Cp 36 – 62, 191, 245
CIL XI.3941 – 268 ST Cp 37 – 76, 148, 160,
CIL XI.6704,6 – 582 Oscan 165, 168, 172, 175, 184,
CIL XI.7487 – 266 (Hernician and Samnitic 185, 191, 194, 198-9,
CIL XI.7494 – 44 inscriptions are listed 203, 269, 354
CIL XI.7531 – 258 separately) ST Cp 38 – 259
CIL XI.7768 – 258 ST CA, see Cm 1 ST Cp 42 – 248
CIL XI, falsae 350* – 15 ST Cm 1 (CA) – 156, 159, ST Fr 1 – 181, 239, 240,
n.5 169, 173, 194, 195, 197, 242, 253
CIL XII.5728 – 271 206, 257 ST Fr 4 – 134 n.70
CIL XIV.1016 – 270 ST Cm 2 – 170 ST Fr 5 – 106, 315
CIL XIV.3110 – 251 ST Cm 4 – 170 ST Fr 10 – 242
ST Cm 6 – 76, 180, 246-7, ST Fr 16 – 62
AÉ 1982, no.292 – 44, 512
268, 271 ST Hi 3 – 66, 252, 316
n.268
ST Cm 7 – 64, 180, 247 ST Hi 4 – 66, 227 n.115,
Colonna 1994 – 162, 197,
ST Cm 9 – 170 316
333
ST Cm 11 – 316 ST Hi 6 – 315
Elogia Tarquiniensia – 40
ST Cm 12 – 146, 204, 316; ST Hi 7 – 227 n.115
Fasti Triumphales AUC
462 ST Hi 10 – 242
DXII – 41, 42
ST Cm 13 – 462 ST Lu 1, see TB
Garigliano inscription –
ST Cm 14 – 58, 76, 118 ST Lu 2 – 76, 170, 268
162, 197, 333
n.60, 210, 242, 246-7, ST Lu 5 – 66, 83, 119 n.62,
ILLRP 54 – 202
247, 260, 268 181, 195
ILLRP 303 – 166 n.88
ST Cm 18 – 196 ST Lu 6 – 160, 194
ILLRP 1204 – 203
ST Cm 22 – 252 ST Lu 7 – 194
NDI 223 – 166 n.88
ST Cm 24 – 34 ST Lu 8 – 194
J.-L. Zimmermann (1986) –
ST Cm 25 – 227 n.115, 261 ST Lu 10 – 194
41
ST Cm 27 – 297, 327 ST Lu 13 – 160 n.85
ST Cm 28 – 69, 195, 268 ST Lu 15 – 251-2
Lepontic ST Lu 16 – 66
ST Cm 30 – 253
PID 255 – 135 ST Cm 35 – 219 ST Lu 18 – 160, 164-6
PID 274 – 135 ST Cm 38 – 196 ST Lu 19 – 160
PID 302 – 135 ST Cm 39 – 196 ST Lu 23 – 247
ST Cm 40 – 118 ST Lu 26 , 317
Lejeune 1989 – 135 ST Lu 27 – 251-2, 316
ST Cm 42 – 134
ST Cm 47 – 219 ST Lu 31 – 83, 316
Marrucinian ST nCm 3a – 66 ST Lu 32 – 66, 316
ST MV 1 – 141 n.75, 195, ST nCm 3b – 66 ST Lu 33 – 316
271 ST nCm 3c – 66 ST Lu 34 – 66, 316
ST MV 3 – 166, 248, 270 ST nCm 3d – 66 ST Lu 35 – 316
ST MV 4 – 259 ST nCm 3e – 66 ST Lu 36 – 191, 316
ST MV 6 – 162, 196, 227 ST nCm 3f – 66 ST Lu 39 – n.85
n.115 ST nCm 3g – 66 ST Lu 40 – 196, 247; 578
ST MV 7 – 196, 315 ST Cp 2 – 192, 239 ST Lu 45 – 242
ST MV 8 – 182, 247, 311 ST Cp 3 – 239 ST Lu 46 – 227 n.115, 242,
ST MV 9 – 247 ST Cp 4 – 308 n.155 246-7, 253, 268
ST Cp 9 – 333 ST Lu 47 – 242, 246, 246-7
Marsian ST Cp 24 – 66, 169, 173 ST Lu 51 – 240
ST Cp 25 – 169, 240 ST Lu 62 – 62
ST VM 3 (Volscian?) – 99 ST Cp 28 – 69 ST Lu 63 – 242, 247
ST VM 4 – 109, 185, 194 ST Cp 29 – 173 ST aLu 15 – 66
675
INDICES
ST aLu 33 – 66 399 n.195, 414, 462, 585 ST Ps 17 – 297, 327

ST tLu 1 – 100, 238 ST ZO 1 – 263 ST Ps 20 – 165, 184
ST Me 1 – 69, 195, 202,
249, 261, 317 Paelignian Samnitic
ST Me 2 – 195, 202, 261,
317 ST Pg 4 – 204, 247, 269 ST Sa 1 (TA) – 62, 169,
ST Me 3 – 68, 195, 202, ST Pg 5 – 69, 185, 194, 195, 173, 185, 190, 203
261, 317 249 ST Sa 4 – 66, 106, 123, 168,
ST Me 5 – 202 ST Pg 8 – 204, 316 181, 208
ST Po 1 – 62, 69, 179, 195, ST Pg 9 – 46, 99, 106, 141 ST Sa 5 – 244
249 n.75, 150, 189, 198-9, ST Sa 10 – 77
ST Po 2 – 98 n.48, 179 199, 230, 343, 351; 406 ST Sa 11 – 77
ST Po 3 – 76, 123, 184, 194, ST Pg 10 – 162, 182, 259, ST Sa 14 – 179, 181, 242
276; 414 311 ST Sa 15 – 181
ST Po 4 – 194 ST Pg 11 – 106 ST Sa 17 – 69
ST Po 8 – 194 ST Pg 12 – 227 n.115 ST Sa 24 – 170
ST Po 9 – 194 ST Pg 13 – 227 n.115, 271 ST Sa 27 – 118 n.60, 316
ST Po 10 – 194 ST Pg 14 – 196, 227 n.115, ST Sa 30 – 185, 190
ST Po 11 – 179 249 ST Sa 31 – 134, 148, 179
ST Po 14 – 194, 202, 375 ST Pg 15 – 227 n.115, 251- n.96, 184; 585
ST Po 15 – 179 2, 259 ST Sa 36 – 250
ST Po 16 – 179 ST Pg 16 – 227 n.115 ST Sa 51 – 269
ST Po 19 – 196 ST Pg 17 – 227 n.115 ST nSa1b – 66
ST Po 22 – 192, 203, 314 ST Pg 26 – 62, 99 ST nSa1c – 66
ST Po 27 – 159 ST Pg 27 – 240 ST tSa 9 – 195
ST Po 29 – 159 ST Pg 28 – 227 n.115 ST tSa 11 – 195
ST Po 31 – 159 ST Pg 29 – 227 n.115 ST tSa 15 – 238
ST Po 32 – 173, 197 ST Pg 30 – 227 n.115 ST tSa 21 – 249
ST Po 36 – 196, 250 ST Pg 31 – 227 n.115 ST tSa 32 – 249
ST Po 38 – 204 ST Pg 32 – 227 n.115 ST TA, see Sa 1
ST Po 47 – 238 ST Pg 33 – 227 n.115, 253,
ST Po 51 – 227 n.115 259 Sicilian
ST Po 52 – 118 ST Pg 38 – 259
ST Pg 39 – 259 Paino 1958 (Sicel?) – 136
ST Po 55 – 219 Ve 186 (Elymian?) – 136
ST Po 58 – 100, 238 ST Pg 40 – 208, 259
ST Po 65 – 118 ST Pg 51 – 227 n.115, 253
ST Pg 52 – 99, 227 n.115, South Picene
ST Po 66 – 118 n.60, 227
n.115 248, 270 AP.1 – 197-8, 251-2, 259;
ST Po 68 – 62 ST Pg 53 – 99, 227 n.115 463
ST Po 87 – 247 ST Pg 54 – 227 n.115 AP.2 – 180, 190
ST Po 89 – 258 ST nPg2 – 66, 208 AP.3 – 120, 162, 182, 311,
ST tPo 13 – 251-2 ST nPg 8 – 266 414
ST tPo 30 – 62 AP.4 – 233
ST tPo 32 – 62 Praesamnitic AP.5 – 233, 248
ST tPo 34 – 62 AQ.1 – 185, 194, 414
ST Ps 1 – 315
ST Si 4 – 173 AQ.2 – 123, 150 n.79, 180,
ST Ps 3 – 233, 253, 276
ST Si 5 – 173 248, 312; 397 n.192
ST Ps 4 – 134, 135, 140,
ST Si 6 – 173 AQ.3 – 120, 242
162, 197, 333
ST Si 8 – 185 CH.1 – 120, 123, 148, 184,
ST Ps 5 – 62, 134, 135, 140,
ST Si 9 – 185 413
162, 197, 219, 333
ST Si 20 – 202 CH.2 – 123, 165, 192, 197-
ST Ps 11 – 134
ST nSi 2a – 66 8, 209, 282, 334; 463,
ST Ps 12 – 253
ST TB – 58, 69, 83, 146, 581
ST Ps 13 – 134, 245
147, 150, 165, 172, 180, MC.1 – 75, 162, 182, 189,
ST Ps 14 – 134
181, 183, 184, 187, 190, 311, 351; 446
ST Ps 16 – 297, 327
191, 193, 198-9, 303; TE.2 – 123, 134, 140, 251,
676
INDICES
259 TI III.16 – 98 n.48 343, 351

TE.3 – 197-8, 251-2, 274; TI III.22 – 203 TI VIb.63 – 180
463 TI IV.15 – 194 TI VIIa.3 – 105 n.51
TE.4 – 134, 135, 140, 162, TI IV.31 – 164, 184; 585 TI VIIa.6 – 105 n.51
197, 333 TI Va.2 – 198 TI VIIa.11 – 146
TE.5 – 105, 120, 150, 197- TI Va.3 – 195 TI VIIa.39 – 151
8, 208, 251-2; 463, 585 TI Va.5 – 147 TI VIIa.43 – 164, 184; 554,
TE.6 – 208 TI Va.6 – 171, 197 585
TE.7 – 105, 120, 144 n.77, TI Va.10 – 195 TI VIIa.47 – 147, 170
208 TI Va.13 – 58, 180 TI VIIb.2 – 147
RI.1 – 105, 120 TI Va.15 – 198 TI VIIb.4 – 171, 197
TI Va.23 –164, 194
ST Um 2 – 165
Umbrian TI Va.24 – 171, 197
ST Um 3 – 227 n.115
TI Va.27 – 171, 197
TI in general – 45 ST Um 4 – 150, 160 n.84,
TI Vb.2 – 194
TI Ia.1 – 192; 571 164, 174
TI Vb.3 – 164, 171, 197
TI Ia.3 – 164 ST Um 5 – 253
TI Vb.7 – 171, 197
TI Ia.4 – 180, 199 ST Um 6 – 195
TI Vb.9 – 184, 210
TI Ia.14 – 185 ST Um 7 – 180, 190
TI Vb.10 – 76, 184
TI Ia.20 – 106 n.53, 198 ST Um 8 – 239, 243, 260
TI Vb.14 – 184
TI Ia.22 – 199 ST Um 9 – 184, 194
TI Vb.15 – 76
TI Ib.6 – 199 ST Um 10 – 269, 317; 464
TI Via.4 – 148
TI Ib.11 – 151 ST Um 12 – 195
TI Via.5 – 157 n.81, 195
TI Ib.12 – 198 ST Um 13 – 195
TI Via.8 – 554
TI Ib.32 – 184 ST Um 16 –, 317
TI VIa.9 – 62
TI Ib.33 – 554 ST Um 17 – 180, 190, 317
TI VIa.10 – 62
TI Ib.34 – 170; 585 ST Um 18 – 162, 190, 197,
TI Via.11 – 194
TI Ib.45 – 194, 233, 245, 317, 333
TI Via.15 – 173, 197
251-2 ST Um 19 – 190, 195, 317
TI VIa.22 – 164
TI IIa.5 – 203 ST Um 20 – 180
TI Via.23 – 171, 197
TI IIa.10 – 146 ST Um 21 – 269
TI Via.27 – 173
TI IIa.11 – 146 ST Um 23 – 242
TI Via.30 – 180
TI IIa.12 – 184 ST Um 24 – , 317
TI Via.36 – 173
TI Iia.21 – 62 ST Um 27 – 245
TI Via.42 – 199
TI IIa.16 – 164 ST Um 29 – 185
TI Via.43 – 150
TI IIa.17 – 164 ST Um 30 – 227 n.115
TI Via.46 – 173
TI IIa.23 – 198 ST Um 37 – 253
TI Via.50 – 199
TI IIa.24 – 183, 343 ST Um 38 – 227 n.115, 246-
TI Via.51 – 196
TI IIa.25 – 585 7
TI VIb.2 – 103 n.50, 184
TI IIa.31 – 198 ST Um 39 – 252
TI Vib.3 – 185
TI IIa.44 – 194, 245 TI Vib.7 – 171
TI IIb.15 – 436 TI VIb.9 – 58 Venetic
TI Ib.17 – 25 TI VIb.10 – 180 Lejeune 1989 – 135
TI Ib.32 – 164 TI Vib.11 – 69, 186; 402 Le 26 – 218, 238; 578
TI Iib.2 – 69 TI VIb.13 – 200 Le 60 – 135
TI IIb.7 – 203 TI VIb.19 – 106 n.53 Le 63 – 312 n.160
TI IIb.17 – 203 TI VIb.24 – 195 Le 65 – 135
TI IIb.22 –164, 203 TI VIb.25 – 197 Le 75ter – 312 n.160
TI IIb.23 – 146 TI VIb.26 – 171, 197 Le 77 – 243, 312 n.160
TI IIb.24 – 203 TI VIb.29 – 173 Le 78 – 312 n.160
TI IIb.26 – 203, 245 TI VIb.44 – 184 Le 79 – 312 n.160
TI III.2 – 196 TI VIb.50 – 554 Le 102 – 312 n.160
TI III.3 – 198 TI VIb.54 – 146 Le 109 – 312 n.160
TI III.4 – 69 TI VIb.58 – 147 Le 110bis – 135, 312 n.160
TI III.9 – 69 TI Vib.59 – 58 Le 124 – 243
TI III.10 – 69 TI VIb.61 – 150, 151, 199, Le 127 – 309
677
INDICES
Le 128 – 309 Vestinian Volscian

Le 148 – 135
Le 150A-B – 135 ST MV 2 – 99, 179 ST VM 1 – 109, 140, 179
Le 226 – 312 n.160 ST MV 7 – 182, 227 n.115, n.96
Le 236 – 135 311 ST VM 2 – 62, 133 n.67,
ST MV 10 – 202 199; 464
ST MV 11 – 259 ST VM 3 (Marsian?) – 99
Mattiocco 1986 – 164
678

The Latin Dialect of The Ager Faliscus (Faliscan)

Uploaded by

Copyright:

Available Formats

The Latin Dialect of The Ager Faliscus (Faliscan)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Latin Dialect of The Ager Faliscus (Faliscan)

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What languages are discussed in the text?

What languages are discussed in the text?

The Latin Dialect of the Ager Faliscus

150 Years of Scholarship

Lay out: Gabriël C.L.M. Bakkum, Amsterdam

ISBN 978 90 5629 562 2

© G.C.L.M. Bakkum / Vossiuspers UvA – Amsterdam University Press, 2009

ter verkrijging van de graad van doctor

Gabriël Cornelis Leonides Maria Bakkum

Promotor: prof. em. dr. H. Pinkster

Overige leden: prof. dr. Ph. Baldi

Faculteit der Geesteswetenschappen

INÆ VAN CAMERIJK

CONIVGIO · NOSTRO · SPATIVM · BREVE · CONTIGIT · ÆVI

3.7.1. General 104

4.6.4. The fifth declension 147

8.11.2. Other types of dedicatory inscriptions 316

11.4. A note on the presentation of the inscriptions 398

16.5. Carbognano-Vallerano [nrs. 329-337] 542

I also wish to thank:

● dr P. Pelagatti (then Soprintendente per l’Etruria Meridionale), dr F. Buranelli (then Museo

1. CORPORA, CATALOGUES, AND OTHER PUBLICATIONS REFERRED TO BY SIGLA

2. PUBLICATIONS REFERRED TO BY AUTHOR

ADAMS, John Norman, & Simon Swain

ASCOLI, Graziadio Isaia

BORMANN, Eugen (continued)

CAMPOREALE, Giovannangelo (continued)

COLONNA, Giovanni (continued)

CORNELL, Tim (continued)

DIEHL, Ernst (continued)

ERNOUT, Alfred (continued)

GIACOMELLI, Gabriella (continued)

GRAN-AYMERICH, Jean, & Dominique BRIQUEL

HELBIG, Wolfgang / Hermine SPEIER

HOLLAND, Louise (also ADAMS HOLLAND, Louise)

KAJANTO, Iiro (continued)

LEJEUNE, Michel (continued)

LOMMATZSCH, Ernst, & H. DESSAU,

MARINETTI, Anna (continued)

VON MERCKLIN, Eugen

MOSCATI, Paola (continued)

OLMOS ROMERA, Ricardo

PASQUI, Angelo (Angiolo)

PFISTER, Raimund (& Ferdinand SOMMER)

POCCETTI, Paolo (continued)

RIBEZZO, Francesco (continued)

RIX, Helmut (continued)

SOLIN, Heikki (continued)

TAYLOR, Lily Ross (continued)

UNTERMANN, Jürgen (continued)

ZMIGRYDER-KONOPKA, Zdzisław, & Kazimierz ROZENBERG

3. A NOTE ON SOURCES AND EDITIONS

4. CONVENTIONS IN THE REPRESENTATION OF EPIGRAPHIC TEXTS

1.1. Introductory remarks

Part II is a presentation of the epigraphic data on which the discussions in part I

1.2. Dialect and language

As the main problem of this study is whether Faliscan is a dialect or an independent

LF 242 cụestod pi pretod pis

LF 248 pre]tor ii duu[muiru

Lat 237 bis q duouir

/dhħ1ilios/ / īlos/ fileo MF 471*, hileo MF 146