Secondary Verbs in Old Irish

Provided by the author(s) and NUI Galway in accordance with publisher policies.
Please cite the published

version when available.
Title
Author(s)
Secondary Verbs in Old Irish: A comparative-historical study

of patterns of verbal derivation in the Old Irish Glosses
Le Mair, Esther
Publication
Date
2011-09-30
Item record
http://hdl.handle.net/10379/3113
Downloaded 2016-11-18T17:52:32Z
Some rights reserved. For more information, please see the item record link above.
Secondary Verbs in
Old Irish
A comparative-historical study of patterns of verbal derivation in the Old
Irish Glosses
Esther Le Mair
September 2011
PhD thesis, supervised by:

Dr. G.R. Isaac
Roinn na Gaeilge
School of Languages, Literatures and Cultures
National University of Ireland, Galway
To my father,
who steadfastly believes in me,
and to Dr. Kicki Ingridsdotter,
teacher, mentor, best friend.
Table of Contents
Table of Contents ........................................................................................... 5
Abstract .......................................................................................................... 9
Acknowledgments ........................................................................................ 11
Abbreviations ............................................................................................... 15
Primary and secondary sources ................................................................ 15
Language .................................................................................................. 15
Grammar .................................................................................................. 16
Others ....................................................................................................... 16
Symbols .................................................................................................... 16
1
Preliminaries ........................................................................................ 17
1.1
Aim, material, scope, and method ................................................. 17
1.1.1
Aim......................................................................................... 17
1.1.2
Material .................................................................................. 19
1.1.3
Scope ...................................................................................... 20
1.1.4
Method ................................................................................... 21
1.2
Structure of the thesis .................................................................... 22
1.2.1
Use of terminology................................................................. 22
1.2.1.1
The terms Common Celtic, Brittonic, Goidelic;
Insular Celtic vs. Gallo-Brittonic ..................................................... 23

2
The Old Irish Verbal System................................................................ 27

2.1
Word order .................................................................................... 27
2.2
Primary versus secondary verbs .................................................... 27
2.3
Primary versus secondary endings ................................................ 28
2.4
Weak, strong and hiatus verbs ....................................................... 28
2.4.1
Strong verbs ........................................................................... 28
2.4.2
Hiatus verbs ............................................................................ 30
2.4.3
Weak verbs ............................................................................. 30
2.5
Simple and compound verbs ......................................................... 31
2.6
Absolute and conjunct ................................................................... 31
2.6.1
2.7
Historical background ............................................................ 32
Stem, tense and mood .................................................................... 34
2.7.1
Present indicative ................................................................... 35
2.7.2
Imperfect ................................................................................ 36
2.7.3
Imperative .............................................................................. 36
2.7.4
Subjunctive............................................................................. 36
2.7.4.1
2.7.5
Future ..................................................................................... 38
2.7.5.1
2.7.6
Origin of the future formations ....................................... 38
Preterite active........................................................................ 40
2.7.6.1
2.7.7
Origin of the subjunctive ................................................ 37
Origins of the preterite .................................................... 41
Preterite passive ..................................................................... 43
2.8
Voice ............................................................................................. 43
2.9
Number and person ....................................................................... 44
Analysis ................................................................................................ 45
3.1
Basic statistics ............................................................................... 45
3.2
Latin loans ..................................................................................... 48
3.3
Morphology of AI and AII verbs................................................... 52
3.3.1
3.4
Semantic motivation for formation as AI or AII verbs .......... 55
3.3.1.1
Denominatives ................................................................ 55
3.3.1.2
Deadjectivals ................................................................... 62
3.3.1.3
Semantic motivation for formation as deponent verbs ... 63
Morphology of and motivation for formation as -igidir verbs ...... 65
3.4.1
Derivations from nouns and adjectives in -ch ........................ 67
3.4.2
The appearence of unexplained -n- between base and suffix 68
3.4.3
Calques on Latin .................................................................... 69
3.5
Simple vs. compound verbs........................................................... 74
3.6
Arguments for age ......................................................................... 75
3.7
Chronology .................................................................................... 77
3.8
Primary verbs taking on weak flexion ........................................... 83
3.9
Semantics of derivative verbs ........................................................ 86
3.9.1
Introduction ............................................................................ 86
3.9.2
Semantic categories ................................................................ 87
3.9.2.1
Causative......................................................................... 87
3.9.2.2
Iterative ........................................................................... 88
3.9.2.3
Fientive ........................................................................... 89
3.9.2.4
Inchoative........................................................................ 89
3.9.2.5
Essive .............................................................................. 90
3.9.2.6
Semantic relationship between subject or object of a verb
and its base ....................................................................................... 90

3.9.2.7
Subject causes [base] ...................................................... 96
3.9.2.8
Gerative ........................................................................... 97
Conclusion ........................................................................................... 99
Corpus of secondary verbs ................................................................. 107

5.1
Structure and notational conventions .......................................... 107
5.2
AI verbs ....................................................................................... 109
5.3
AII verbs ...................................................................................... 157
Appendix 1: Weak primary verbs ...................................................... 259
Appendix 2: Primary verbs ................................................................ 275

7.1
BI ................................................................................................. 275
7.2
BII ................................................................................................ 314
7.3
BIII .............................................................................................. 324
7.4
BIV .............................................................................................. 327

7
7.5
BV ............................................................................................... 332
7.6
Hiatus verbs ................................................................................. 334
Bibliography....................................................................................... 349
Abstract
Abstract
This thesis concerns the word formation of secondary verbs in Old Irish.
Although extensive work has been done on primary verbs, the secondary
adjectives and the nouns in Old Irish, and the formation of causatives and
iteratives and that of the verbal nouns in Welsh, the secondary verbs in Old
Irish have been almost entirely ignored (with the exception of the deverbal
verbs in -igidir), while they provide fascinating insights into the process of
word formation in Celtic and Early Irish. Their importance lies especially,
but not exclusively, in the obvious productivity of this morphology in Old
Irish and in the visible development of the morphology from Proto-IndoEuropean through Old Irish. The formation of secondary verbs in any
language and indeed in any stage of that language shows the creativity of
the users of that language and the secondary verbs in Old Irish show the
creativity of the speakers of Old Irish and its antecedents.
The thesis consists of five chapters and two appendices. The first
chapter
contains
the
preliminaries,
the
theoretical,
material
and
methodological basis of the thesis. The second chapter is an introduction

into the Old Irish verbal system and its origins to set the stage for the
remaining chapters. The third chapter is the analysis, morphological,
semantic and statistical, drawn from the corpus. The fourth chapter is the
conclusion. The fifth chapter contains all the secondary verbs found in the
Wrzburg and Milan glosses with cognates, discussion and notes. The first
appendix contains those primary verbs that have taken on weak flexion and
the second all the other primary verbs, for comparative purposes.
Acknowledgments
Acknowledgments
Beside me lie several notebooks full with notes taken during my meetings
with my supervisor, Dr. Graham Isaac. Without these meetings, of course,
there would never have been a thesis. But without these meetings, I would
not have been the scholar I am today either. Grahams wealth of knowledge,
constant support and unfailing belief in me and my academic capabilities
have brought me to where I am now and enabled me to finish a thesis that I
can be proud of. Thank you.
Compiling a corpus of all the verbal forms in the Wrzburg and
Milan Glosses is, frankly, boring work. It was at this point during my
research that Dr. Kicki Ingridsdotter entered my life with her firm belief in
love and joy and sparkle and colour as didactic and motivational tools. We
have had many an interesting discussion about Old Irish grammar. Without
the breaks for coffee and (home-made) cake this thesis would never have
gotten done. Thanks for never giving up on me and for never letting me
believe too little of myself.
All staff at Roinn na Gaeilge have been incredibly supportive of me
over the years. Many thanks to an Dr. Nollaig Murale and an Dr.
Donncha hAodha who were on my Graduate Research Committee.
Thanks to an tOllamh Gearid Denvir for cheering me on every step of the
way, an Dr. Lillis Laoire for a listening ear, an Dr. John Walsh for
checking up on me regularly, an tOllamh Michel Mac Craith for a much
needed spare copy of Thurneysens Grammar of Old Irish and thanks to an
tOllamh Nollaig Mac Congil for a ready smile. Fiona de Paor has by now
printed a million copies of my thesis and still her door is always open to me.
Many thanks.
My love for Old Irish was instilled in me from the first time I was
ever taught it at Utrecht University, by Dr. Leni van Strien. She and Dr. Bart
Jaski nourished this love for Old Irish and encouraged me to continue in the
field. In Utrecht, too, I was taught by Dr. Frank Brandsma and Dr.
11
Acknowledgments
Jacqueline Borsje. All are excellent teachers to whom sincere thanks are due
for giving me a strong foundation on which build my studies.
Throughout my PhD, I have received help and encouragement from
many esteemed scholars in the field. I would like to thank especially Dr.
Dagmar Wodtko for her genuine enthusiasm about my project when I was
only just starting, for forwarding me her article on Old Irish compound
verbs and for giving me a reference when I applied for funding. Dr. Aaron
Griffith has provided me with the entire digitalised corpus of the Milan
Glosses ahead of online publication, which has saved me a lot of time.
Many thanks. Thanks also to Prof. Peter Schrijver who took the time to
advise me on the best course for my future career when I had to decide on
where to do a PhD and for sending me his article on Old British ahead of
publication. Thanks to Dr. Jacopo Bisagni for sending me several of his
papers, explaining tricky bits of phonology, tracing some particularly pesky
references and for always being willing to help. Jacopo already had high
hopes for both me and my thesis when I had barely started. I can only hope I
have lived up to his expectations. Thanks to Dr. Paul Russell for sending me
his article on British uocridem ahead of publication.
My family and friends have been a constant source of support
throughout the years. I want to thank my father, my sister Eline, my brother
Ruben and my grandparents for being there for me, for putting up with the
fact that I never had time to come visit and for supporting me. My
grandfather unfortunately will never see the finishing of the thesis since he
passed away this summer, but his constant silent support meant the world to
me.
Here in Galway, Mona Jakob, Trish N Mhaoileoin, David Burke,
Hanne-Mette Alsos-Raae, Eibhln N Fhallamhin, Marta Rydlinska, Liam
hAisibil, Val Nolan, Frances McCormack, Steve Daly, Adam Murphy
and Kevin Jennings have provided much needed breaks and general support.
There was always someone for tea, coffee, lunch or more stringent
emergencies.
Further afield many friends have kept close track of how I was doing
and the knowledge of their virtual presence has kept me going on many a
long night. Thanks are due to Daphne Oosterhout, Karianne Lemmen,
12
Acknowledgments
Liselore van der Zweth, Christel Franken, Sean van der Meulen, Krista den
Uijl, John van Dam, Adinda Bremer, Camilla Hansen, Jessica Lehto, Piia
Anneli Huittinen, Stefanie Kelly, Rachel Miller, Daniel, Sara and Moa
Visn, Johan and Linna Anglemark, Terry Walker, Susanna Lyne, Autumn
Barnard-Tyron, Gretchen Kern, Katie Louise Mathis, Heather Key, Jimmy
Miller, Cormac MacAindir, Peadar Muicheartaigh and Madhbh Sullivan.
My colleagues at Lidl have had to deal with an increasingly stressed
colleague and have done so with admirable patience and good humour. I
really appreciate the flexibility in giving me time off and the regular
checking up whether I had gone crazy yet. Thank you, Martina, Shane,
Barbara, Natasha, Shaun, Ricardo, Imre, Kamil, Kris, Magda, Elizabeth,
Liam, Pauline, Trish, Trisha, Liz, Andrzej, Garret, Lorraine, Jakub, Martin,
Barbara, Marek and Piotr.
I hope I have not forgotten anyone, but it is, after all, the day before
submission. If I have forgotten you, I hope you will forgive me.
Esther Le Mair
Galway, 29 September 2011
13
Preliminaries
Abbreviations
Primary and secondary sources
For full details refer to bibliography.
ACC
AED
DIL
GOI
GPC
IEW
Amrae Coluimb Chille

Orel. Albanian Etymological Dictionary
RIA. Dictionary of the Irish language
Thurneysen. A Grammar of Old Irish
Geiriadur Prifsygol Cymru
Pokorny.
Etymologisches
Indogermanisches
Wrterbuch
Rix: Lexikon der indogermanischen Verben
The Milan Glosses on the Psalms
Pedersen. Vergleichende Grammatik der keltischen
Sprachen
Revue Celtique
The St Gall Priscian Glosses
Saltair na Rann
Thesaurus Palaeo-Hibernicus, Volume 1
The Wrzburg Glosses on the Pauline Epistles
Kavanagh. A Lexicon of the Old Irish Glosses in the
Wrzburg Manuscript of the Epistles of Saint Paul
Zeitschrift fr Celtische Philologie
LIV2
Ml.
Ped.
RC
Sg.
SR
Thes.
Wb.
Wb. Lex.
ZCP
Language
Alb.
Br.
Britt.
CC
Celt.
Corn.
G
Gk.
Goth.
Lat.
MB
MC
Mid. Ir.
ModB
ModC
ModI
MW
Albanian
Breton
Brittonic
Common Celtic
Celtic
Cornish
Gaulish
Greek
Gothic
Latin
Middle Breton
Middle Cornish
Middle Irish
Modern Breton
Modern Cornish
Modern Irish
Middle Welsh
OAv.
OB
OCorn.
OCS
OHG
OI
OIsl.
OW
PA
PG
PIE
Vann.
Ved.
W
YAv.
15
Old Avestan
Old Breton
Old Cornish
Old
Church
Slavonic
Old High German
Old Irish
Old Icelandic
Old Welsh
Proto-Albanian
Proto-Germanic
Proto-IndoEuropean
Vannetais
Vedic
Welsh
Young Avestan
Preliminaries
Grammar
abs.
acc.
act.
adj.
caus.
cmpd.
conj.
cons.
dat.
dep.
emp.
impf.
impv.
infix.
pron.
f.
fut.
gen.
ind.
intrans.
absolute
accusative
active
adjective
causative
compound
conjunct
consonant
dative
deponent
emphatic
imperfect
imperative
infixed
pronoun
feminine
future
genitive
indicative
intransitive
it.
m.
n.
nas.
nom.
pass.
perf.
pl.
pref.
pres.
pret.
prim.
prot.
red.
sec.
sg.
subj.
suff.
trans.
VN
iterative
masculine
neuter
nasal
nominative
passive
perfect
plural
prefix
present
preterite
primary
prototonic
reduplicated
secondary
singular
subjunctive
suffix
transitive
verbal noun
Others
cf.
e.g.
gl.
i.e.
n.
s.v.
confer, compare
exempli gratia, for example
glossing
id est, that is
note
sub voce, under the word
Symbols
This verb is not attested in Wb. or Ml.
This verb is not attested in OI, or, this is a reconstructed root.
16
The Old Irish Verbal System
1 Preliminaries
1.1 Aim, material, scope, and method
1.1.1 Aim
The aim of this research is to study the word formation of secondary verbs
in Old Irish. Whereas primary verbs in Proto-Indo-European and the Celtic
languages have been extensively studied (Schumacher 2004, Rix et. al.
2001), as have the secondary adjectives (Wodtko 1995) and the nouns in
Old Irish (De Bernardo Stempel 1999) and the formation of causatives and
iteratives (Schulze-Thulin 2001) and that of verbal nouns in Welsh
(Schumacher 2000), the formation of secondary verbs has mostly been
neglected. The -igidir verbs have received some attention with Joseph
studying the origin of their formation (Joseph 1987) and Crualaoich the
syncope patterns arising in these formations ( Crualaoich 1997). A study
of all the formations of secondary verbs in Old Irish is a desideratum. The
importance of the secondary verbs lies both in the productivity of this
morphology and in the visible development of this morphology from ProtoIndo-European through Old (and indeed Modern) Irish. Whereas the
primary verbs show the Indo-European heritage and conserve the old IndoEuropean patterns, the secondary verbs express the creative aspect of Irish
and its prehistoric precursors. It shows how speakers used their language
and its lexicon and morphological possibilities it contained. It shows
cultural influences (for example that of Christianity and Latin learning) on
the language. Examples of Old Irish secondary verbs are marbaid kills
(page 142), derived from the adjective marb dead, rmid counts (page
235), derived from the noun rm act of counting and oenaigidir makes
one (page 233) derived from the adjective oen one. It will be argued that
of these verbs, for example, rmid is the oldest and oenaigidir the youngest
(cf. page 77). Whereas age is not always apparent, certain layers in the
derivation can be ascertained. The relationship between morphology, origin
and semantics will also be discussed. What is the reason that a given
17

derivative verb becomes an AI verb, whereas others become AII? This
research shows that there are clear reasons for a denominative verb to
become either AI or AII (section 3.3.1, pages 55ff.). It becomes clear that
the deadjectivals become AI verbs with very few exceptions, and the
exceptions are all strongly motivated (section 3.3.1.2, pages 62ff.). In the
case of the -igidir verbs, the motives can mostly not be determined, since
that morphology is so prolific in Old Irish that most verbs coined in that
period will become -igidir verbs by default. Prior to Old Irish, on the other
hand, there are clear reasons for a derivative verb to become an -igidir verb
and these will be discussed (section 3.4, pages 65ff.). In the case of the
-igidir verbs, it is sometimes problematic to determine whether the verb has
been derived from a noun or adjective in -ach/-ech with haplology, or from
its base noun itself (fograigidir emits a sound, page 211, could be derived
either from fograch noisy, resounding or from its base noun fogur
sound). The methodology for this issue has been discussed in section
3.4.1, page 67. Verbs may have been derived from either a compound noun
or an uncompounded noun, as either a simple verb or a compound noun.
This issue has been discussed in section 3.4.1, pages 74f.
The primary verbs that have taken on weak flexion can be found in
appendix 1, pages 259ff. The handbooks have been followed silently, except
where there is significant unsolved controversy as to the origin of these
verbs, in which case these verbs are discussed in section 3.8, pages 83ff.
The semantic relationship between a verb and its base is discussed in
section 3.9, pages 86ff. There is a wide range of possibilities for the
relationship between the two. A verb can express causing the base, having
the base, being the base, to name but a few. There are however also
constraints. A noun meaning tree will not give rise to a verb meaning
sings in the first instance, for example. Some verbs show more creativity
on the part of the speaker than others. The verb ardraigidir appears from
airdrech face, for example, has a less direct relationship to its base than
the verb bocaid softens from boc soft (page 115).
The corpus itself (Chapter 5, pages 107ff.) informs the entire
analysis chapter. It contains all the forms of the simple secondary verbs and
any compound secondary verbs found in the Wrzburg and Milan Glosses,
18

the Proto-Indo-European and Common Celtic root where appropriate,
cognates, derivation and discussion. The discussion will be referred to
throughout the thesis, however the corpus chapter also includes many notes
on problems associated with individual verbs or verbal forms which are not
necessarily discussed in greater detail in the analytical chapter.
1.1.2 Material
The corpus for this research consists of all the verbs in the Wrzburg
and Milan glosses as edited in the Thesaurus palaeohibernicus (Stokes and
Strachan 1901). These collections were chosen since they are among the
oldest sources of Old Irish written in the Old Irish period (this contrary to
much Old Irish material that has come down in Middle Irish manuscripts).
The Wrzburg Glosses date from around 750 AD, although the prima
manus is decidedly earlier and its Old Irish more archaic (Thes. xxiii, GOI
5). The Milan Glosses date from about 800 AD (Thes. xviii, GOI 6).
These glosses are notes on the Latin texts of the Pauline epistles and the
Letter to the Hebrews and a Latin commentary on the Psalms respectively.
They are written interlinearly and in the margins of these texts. The glosses
translate, explain or expand on the Latin text. Both manuscripts have
survived on the continent, rather than in Ireland itself. A lexicon of the
Wrzburg Glosses has been prepared by Samus Kavanagh and was edited
by Dagmar Wodtko (2001). A lexicon of the Milan Glosses is currently
being undertaken by Aaron Griffith (for more information on the project and
full digitalised text of the Milan Glosses, see http://www.univie.ac.at/
indogermanistik/milan_glosses. htm).1
Whereas these Glosses contain generally speaking pure Old Irish
material, it is important to note the extent of Middle Irish forms already
found even here. It has been argued by McCone (1985) that what is termed
Middle Irish was already the spoken language long before it became the
written language (900 AD) and has occasionally influenced the writing of
the Glossators. Of course, this is not to say that the Glosses themselves are
1
I would like to thank Aaron Griffith for sending me his files of the Milan Glosses prior to
online publication. These have been immensely helpful to me.
19

Middle Irish. Occasional lapses from the conservative educated literary style
show the direction in which the language is changing, not the point at which
the language is at that time. Most of the language in Wrzburg and Milan is
classical Old Irish and any conclusions drawn from these forms can
therefore be said to be drawn about Old Irish. The fact that the manuscripts
in which these Glosses have been preserved are contemporary adds to that
argument.
Using the Glosses as basis for the corpus has one noteworthy
consequence and that is the influence of Latin. The issue of bilingualism in
Irish monasteries is still the subject of a heated debate (Genee 2005,
Bronner 2005, Bisagni and Warntjes 2007). Whereas this is not directly of
relevance to this thesis, the Wrzburg and Milan Glosses were written in a
Latin context. Considering that foreigners came to Irish monasteries to
study, and considering that Latin was the language of learning, it seems
logical to expect that Latin was not just read and studied, but also spoken
throughout the monastery. Moreover, the author of these glosses was
writing Old Irish glosses on a Latin text: his entire context was bilingual.
Whatever the extent of bilingualism was, from the language in the Glosses it
becomes clear that it was not zero. The formation of the compounds conto
and immsi (both meaning converts) for example, is directly influenced
by the Latin conuertere convert. The Latin is a compound of uertere
change, turn. The Old Irish verbs are compounds of sod also meaning
changes, turns. Other examples are the Old Irish verbs asgaib and
etargaib, both hapax legomena. Asgaib glosses Lat. excipere and etargaib
glosses Latin intercipere. gaib translates the stem capere, but the preverbs
have also been directly translated from Latin: as means ex, and etar means
inter. These are but two examples of the influence Latin exerted over the
language of the Glosses. The issue of calquing has been further discussed in
section 3.4.3, pages 69ff.
1.1.3 Scope
The focus of this thesis lies firmly on Old Irish and its predecessors.
Middle Irish and later developments will not be discussed.
20

In this thesis the focus lies on the derivation of a verbal stem from
other parts of speech or other verbs. That is, derivation by means of
suffixing, rather than prefixing. The morphology of compounding with one
or more preverbs lies outside the scope of the research.
The only primary verbs that will be discussed are those that have
taken on weak flexion but about whose origin there is still significant
controversy (see below, section 3.8, pages 83ff.); the great majority of
primary verbs will be left aside. However, a collection of all the primary
verbs in Wrzburg and Milan has been appended to this thesis for
comparative purposes (Appendix 2, pages 275ff.).
1.1.4 Method
For the collection of verbal forms from the Wrzburg Glosses,
Kavanaghs Lexicon of the Old Irish Glosses in the Wrzburg Manuscript
of the Epistles of Saint Paul (2001) was used. Each verbal form in that
lexicon was entered into a database. The forms from the Milan Glosses were
taken
from
Griffiths
online
edition
of
the
Milan
Glosses
(http://www.univie.ac.at/indogermanistik/milan_glosses.htm).2 Each verbal

form was analysed according to tense, mood, person and number and
entered into the database.
Through this method all the verbs occurring in the Wrzburg and
Milan Glosses have been collected. These were then divided according to
verb class. The strong verbs, being all primary, were kept aside. For each
secondary verb an origin has been proposed where possible. Where the
origin is debatable, it has been further discussed.
The weak verbs were then further subdivided according to
morphology and origin to determine the relationship between the two. The
relationship of this are discussed in the analysis chapter, pages 45ff. Details
My database has been checked against the Thesaurus Palaeohibernicus (Stokes and
Strachan 1901) and Griffiths online files when these became available. Most corrections
made by Griffith (revisions based on re-readings of the manuscript) became available at a
time when my corpus had already been compiled and have therefore unfortunately not been
taken into consideration.
21

on the amount of verbs and basic statistical analysis can be found in the
introduction to the the analysis chapter.
1.2 Structure of the thesis

The thesis has been divided into five chapters and two appendices.
The first chapter contains the preliminaries, the theoretical, material and
methodological basis of the thesis. The second chapter is an introduction
into the Old Irish verbal system. The aim of this chapter is to set the context
for my research. I will give a brief overview of the Old Irish verbal system
and its comparative-historical origins. For issues that are still under debate,
such as the development of VSO word order in Celtic, the origin of the
absolute and conjunct flexion, the origin of various tenses, etc., references
will be given to recent literature on the issue. No new analysis will be
attempted on these issues. The third chapter is the analysis, morphological,
semantic and statistical, drawn from the corpus. The fourth chapter is the
conclusion. The fifth chapter contains all the secondary verbs found in
Wrzburg and Milan with cognates, discussion and notes. The first
appendix contains those primary verbs that have taken on weak flexion and
the second all the other primary verbs, for comparative purposes.
1.2.1 Use of terminology

For verbal classification the conventions used in GOI (pp. 352-7)
will be followed throughout, since this terminology is still the most familiar
to scholars of Old Irish. Reference to other terminologies is given in the
introduction to the Old Irish verbal system below (strong verbs section
2.4.1, page 28; hiatus verbs section 2.4.2, page 30 and weak verbs section
2.4.3, page 30). To facilitate ease of reading, AI is used to refer both to
active and deponent verbs and AII to refer to both active and non-igidir
deponent verbs. Only where the morphological difference between active
and deponent is significant for the argument (e.g. below, section 3.3.1.3,
22

pages 63f., where the semantic motivation for formation as a deponent verb
is discussed), will this distinction be made and clearly marked.
Within the AII deponents there is one group of verbs that does show
a particular morphology. Whereas all AI verbs were derived using a suffix
*-- and the other AII verbs were formed using the suffix *--, this group of
deponents was formed using the suffix *-sag--. In Old Irish, these are still
unmistakable. The *-sag- suffix has developed into -ig- followed by
deponent endings. In order to distinguish these verbs from the other AII
verbs, they are named after their suffix with 3sg. ending, namely -igidir
verbs.
Where standard linguistic terminology is deviated from, this is
clearly noted in the relevant section. An example is my use of categories
such as causative and fientive not only for verbs that have Indo-European
causative and fientive morphology respectively, but also for verbs that are
semantically causative or fientive. When speaking of morphology, of how
verbs are formed, the terms causative and iterative refer to the verbal
formation. In this case, a causative or an iterative verb is always deverbal.
When speaking of semantics (section 3.9) and of why verbs are formed the
way they are (section 3.3.1) the terms causative and iterative refer to the
semantics of these verbs and the verb is always denominative.
1.2.1.1 The terms Common Celtic, Brittonic, Goidelic; Insular

Celtic vs. Gallo-Brittonic
In this thesis, the term Common Celtic is used for the reconstructed
language between Proto-Indo-European and Proto-Irish. The term
Brittonic is used in such cases where a distinction between the separate
Brittonic languages (Welsh, Breton, Cornish) cannot, need not or should not
be made. The term Goidelic (Irish, Scottish Gaelic and Manx languages) is
used to contrast with Brittonic but will in practice always refer to Old Irish.
The relationship between the Goidelic, Brittonic and continental
Celtic languages has long been the subject of debate with some scholars
suggesting a closer genetic relationship between the Gaulish and Brittonic
languages (the Gallo-Brittonic theory, for which see Schmidt 1986 and
23

Koch 1992a) and others suggesting a closer genetic relationship between the
Goidelic and Brittonic languages (the Insular Celtic theory, for which see
McCone 1996: 98-104), separating them from the continental Celtic
languages.
It is to be noted that the Insular Celtic vs. Gallo-Brittonic debate is
based on the Stammbaum theory of languages (cf. Isaac 2005: 194), i.e. the
idea that within a language, dialects develop, which develop into separate
languages that develop independently from each other. I take issue with this
theory, as it ignores the linguistic reality that dialects, and indeed languages,
that are in contact with each other continue to interact with and exert
influence upon each other (cf. Isaac 2005, Sims-Williams 2007, Bailey
1996, Duran 1995). Moreover, it requires a subjective weighing of
arguments. If one is predisposed to the Gallo-Brittonic theory, for example,
the developments of the labiovelar will weigh heavy in the arguments.
Conversely, predisposition to the Insular Celtic theory will see these
developments as a minor issue to be explained by other means, and will
instead consider the distinction between absolute and conjunct flexion of
importance in the argument. This lack of objectivity or, indeed, the lack of
possibility for objectivity is a problem for either theory.
An improvement on the Stammbaum theory is the Wellen theory or
wave theory proposed by Johannes Schmidt and Hugo Schuchardt. The
wave theory states that language change occurs in waves from an epicentre
of linguistic innovation. This certainly suits the situation of the Celtic
languages better. In practical terms, this means for example that the
development of the labiovelar that is pointed out by the Gallo-Brittonic
theorists on the one hand, and the distinction between absolute and conjunct
flexion that is noted by the Insular-Celtic theorists on the other, could both
have had their own epicentre of change from which they spread among the
Gaulish and Brittonic and the Brittonic and Goidelic dialects respectively
(cf. Sims-Williams 2007: 326-7). It also makes room for isoglosses of other
linguistic changes that have taken place within some of the Celtic
languages, but not all. The result shows a picture that is far more
complicated that the traditional tree diagram, but takes into account every
24

linguistic change, putting them on the map without having to argue for
taking some changes into account and ignoring the others.
Isaac (2005) proposed a convergence theory in which dialects are
seen to be in fluid relationship to each other, depending on tribal relations.
The dialects and their innovations influenced each other based on the
relationships between the tribes that spoke them. Thus, these relations will
have been one of the driving forces behind the spread of linguistic change,
leading to the isoglosses mentioned above. Isaacs theory is modelled on
Maurers (1952) work on the Germanic languages, which show similar
convergence. The wave theory and the convergence theory are not mutually
exclusive and it could be argued that both show part of the picture of
linguistic reality at the time.
Since the Insular Celtic or Gallo-Brittonic theories will not be
followed in this thesis, verbs will not be dated to these periods. If arguments
can be made for a more precise dating than merely an unspecified Celtic, a
verb may be dated to early Celtic or pre-Old Irish. Ciallaithir thinks about,
for example has been dated to early Celtic, whereas dnaid closes (page
132) has been dated to pre-Old Irish (see Chronology, section 3.7, pages
77ff.). The term Insular Celtic will only be used in direct quotes and
paraphrases from other scholars. Further discussion on dating criteria and
the reasoning behind dating to an unspecified Celtic category can be found
in the Chronology, sections 3.6 and 3.7, pages 75ff.
25
2 The Old Irish Verbal System

2.1 Word order
The basic word order of Old Irish is VSO, i.e. verb, subject, object.
The verb can only be preceded by preverbs, conjunctions and conjunct
particles. Old Irish word order was consistent with the correlates put
forward by Greenberg (1963, see also McCone 2006: 26): nouns follow
prepositions, adjectives and attributive genitives follow the noun they
qualify etc. For recent implications of the significance of Celtic word order,
see Isaac 2007b (28-9, 56-9, see also Isaac 1993: 8-13).
2.2 Primary versus secondary verbs

This thesis discusses the origin and morphology of secondary verbs
in Old Irish. A full discussion of the secondary verbs and their origin is
therefore given in Chapter 3.
The essential difference is that primary verbs are derived from a
verbal root, whereas secondary verbs are derived from an existing verb,
noun or adjective.3 Therefore, Old Irish primary verbs will have been verbs
already in Proto-Indo-European, whereas Old Irish secondary verbs may
have been derived within any of the language phases between Proto-IndoEuropean and Old Irish. One of the aims of this thesis is to ascertain to what
extent such derivation can be dated to a specific phase of the language for a
given secondary verb.
Although it is possible that a verb could be derived from any part of speech, in Celtic and
Old Irish only denominatives, deadjectivals and deverbals are found. For a possible
derivative from a pronominal, see ailigidir changes, page 162.
27
2.3 Primary versus secondary endings

The terms primary and secondary will also be used in the current
chapter with regards to endings. Considering that, with few modifications,
the same basic ending set is used for the present indicative, present
subjunctive, future and preterite active, it is customary to denote this ending
set as the primary endings. The ending set that is used for the imperfect,
past subjunctive and secondary future is then called secondary endings
(also often called imperfect endings).
2.4 Weak, strong and hiatus verbs

Old Irish verbs fall into three broad classes depending on their
morphology and behaviour in the different tenses. These classes are the
strong verbs (all primary), weak verbs (mostly secondary) and the hiatus
verbs (all primary). Strong verbs are characterised by a stem-final consonant
that is clearly seen in the 3sg. conjunct, which for strong verbs always ends
in a consonant. Pre-Old Irish, weak verbs were characterised by a stem-final
vowel *-- or *-- which is reflected by the 3sg. conjunct in Old Irish.
Hiatus verbs are characterised by a root-final vowel.
2.4.1 Strong verbs

Strong verbs are by definition primary, that is to say, they are direct
descendants of a Proto-Indo-European verbal root. Within the Old Irish
primary verbs, the following Proto-Indo-European present stem classes are
recognised (see Schumacher 2004: 36-474):
Schumacher also mentions the essive presents in this list, however, these are secondary
rather than primary, and only two, ruidid is red, becomes red and scibid moves were
extant in Old Irish (note that scibid is not attested in Old Irish itself, but the existence of the
root and its attestation in Middle Irish are evidence for its existence in Old Irish). These
verbs do not occur in the Glosses but will be briefly discussed as the only examples of
essive verbs in Old Irish in the analysis chapter, page 54.
28

Full-grade root with *-e/o-suffix
Zero-grade root with *-e/o-suffix
Zero-grade root with *-ske/o-suffix
Zero-grade root with *-i e/o-suffix
Full-grade root with *-i e/o-suffix
Nasal presents
> BI verbs
> BI verbs
> BI verbs
> BII verbs
> BII verbs
> BIII-V verbs
The above classification follows GOI (pp. 353-7) and Schumacher

(2004: 36-47). It can be seen that several Proto-Indo-European present stem
classes fall together in the Old Irish BI and BII verbs.
In Proto-Indo-European, the nasal presents were formed by infixing
full grade *-n- or zero grade *-n- before the last consonant of the root. This
last consonant could be an occlusive (leading eventually to the Old Irish
BIII verbs), a laryngeal (leading eventually to the Old Irish BIV verbs) or a
*-u- (leading eventually to the Old Irish BV verbs). By the nasal presents in
those roots ending in a laryngeal, the vocalising effect of this laryngeal
caused a distinction in vowel length and quality to develop in the 3sg. and
3pl. Thus, PIE *-nh1-/*-nh1- gave CC *-n-/-na/e- > *-n-/-ni-, PIE *-nh2/*-nh2- gave CC *-n-/-na- and PIE *-nh3-/-nh3- gave CC *-n-/-na/o- >
*-n-/-na-. As a rule, the short vowel variant of the suffix was subsequently
generalised.
In those nasal presents with Proto-Indo-European root ending in
*-u-, the resulting *-nu- became *-ni- in Celtic. In the nasal presents in roots
ending in an occlusive, the zero-grade suffix was generalised and
subsequently thematicised.5
McCone (1997: 29-31) divides the Old Irish strong verbs into three
classes, namely S1 (= BI and BIII), S2 (= BII) and S3 (= BIV and BV), of
which the S1 is further subdivided. Strachan (1949: 34) also divides the
strong verbs into three classes, namely A1 (= BI and BIII), A2 (= BIV and
BV) and A3 (= BII). Considering that GOIs subclassification remains the
best known, I have followed Thurneysens conventions in my thesis.
Strong verbs are characterised by extensive changes to the verbal
stem in the other tenses. There is considerable variation in their future and
Nasal presents have been discussed in McCone 1991b and Schulze-Thulin 2001.
29

preterite formation. Only a few form f-future or s-preterite (as weak verbs
do), but none have both formations.
2.4.2 Hiatus verbs

Like the strong verbs, hiatus verbs are by definition primary.6
Although in Old Irish the hiatus verbs are similar enough to treat them as a
single group, they have come into existence as a result of a variety
phonological changes, most often the loss of an intervocalic consonant,
through which *kis-e/o- for example is represented by Old Irish cid cries,
weeps. The consonant in question is most often -s-, -w- or -y-.
Hiatus verbs are phonologically distinct from other primary verbs,
because their root ends in a vowel rather than a consonant. They can be
morphologically distinct because some hiatus verbs have weak features in
the other tenses. However, since they are formed from verbal roots, they are
primary verbs.
GOI (547) classifies all hiatus verbs together as weak verbs and
names them AIII. McCone (1997: 28-9) divides them based on their root
vowel into H1 (root vowel a), H2 (root vowel i) and H3 (root vowel o, u, or
e; these are the ones that show weak features outside the present stem, cf.
McCone 1997: 24).
Hiatus appears when the verbal ending consists of vowel +
consonant; when the ending is zero, roots in -a, -i and -e lengthen their rootfinal vowel and roots in -o and -u add -i to form a diphthong. The hiatus is
only retained when the root is stressed, when it is unstressed, the hiatus
disappears.
2.4.3 Weak verbs

The weak verbs contain all the derived verbs within Old Irish and 25
primary verbs that have taken on weak flexion. The secondary verbs include
However note dd kindles, burns (page 340) which could be a causative. For details,
see there.
30

both causatives and iteratives on the one hand (deverbals), and
denominatives and deadjectivals on the other. Thurneysen (GOI 546)
divides the weak verbs into class AI and AII. The AI class only contains
denominatives and deadjectivals with denominative suffix *--, and the AII
class contains deverbals and denominatives with suffix *--. Some
causatives have the root vowel -u- in their present stem and -o- in the other
stems.7 For this reason, McCone (1997: 27) divides the AIIs (his W2, AI are
his W1) into category a and b, where W2b contains those causatives with
-u- in their present stem. Strachan (1949: 34) divides the weak verbs into the
same two classes as Thurneysen, however calling them B1 (= GOI AI) and
B2 (= AII). Weak verbs have a-subjunctive, f-future and s-preterite.
2.5 Simple and compound verbs

Compounding is a process whereby the meaning of a simple verb is
modified by adding one or more preverbs to it. Compounding is extremely
productive in Old Irish, and the distinction between absolute and conjunct
flexion discussed below has far-reaching consequences for this process.
Verbs may be compounded with up to five preverbs, although compounds
with one or two preverbs are most common.
2.6 Absolute and conjunct

Old Irish verbs have two sets of primary endings, the absolute being
used for forms of simple verbs not preceded by negative, interrogative or
conjunct particles, the conjunct endings being used for those that are.
Secondary endings, occurring only in tenses that are by definition preceded
by a conjunct particle, are always conjunct. Compound verbs, being
7
An example of this is suidid puts, places (page 247) of which the compound adsuidi
stops, holds back is attested in Wrzburg and Milan. The stem vowel in the present is -u-,
raised due to the original high vowel *-- in the following syllable, but -o- in other tenses,
where raising has not occurred. Therefore, pres. ind. 3sg. adsudi Ml. 114c6, but perf. 3sg.
atrosoid 39a16.
31

preceded by a preverb, always have conjunct flexion. However, due to
differences in the position of the stress, a distinction arises between
deuterotonic and prototonic forms, in usage corresponding to absolute and
conjunct flexion respectively.
The situation can be illustrated using a simple example. The simple
verb beirid carries carries the stress on the first syllable, beirid. When
negated, the stress remains on the verbal stem: nbeir. When a compound
with one preverb is made out of beirid, this still carries the stress on the
verbal stem: dobeir. When this verb is negated, however, the stress shifts
from the verbal stem to the preverb: ntabair. Shift of stress can have
several phonological consequences, acting as a catalyst for further changes,
these have been summarised by McCone (1997: 4-7).
Simple
cmpd.
Absolute
Conjunct
beirid
beir
deut.
dobeir
prot.
tabair
Old Irish simple verbs in sentence-initial position carry stress on the

verbal stem. When the verbal stem is preceded by one preverb, the stem still
carries stress, meaning that in practice, the stress falls on the second syllable
of the verbal form, hence the term deuterotonic for compound verbs in
sentence-initial position. However, when the verbal stem is preceded by two
or more preverbs, the stress is carried by the second preverb. This is why a
negated compound verb, for example, will carry the stress on its first
syllable, which will be a preverb, rather than the second. Hence the term
prototonic.
2.6.1 Historical background
There have been five main theoretical explanations of the

background of the distinction between absolute and conjunct flexion.
32

Pedersen (1913: 340-1) and Dillon (1947) derive the absolute
endings from a verbal form with a subject pronoun, e.g. OI. beirid < PIE
*beret-is he carries.
Meid (1963) and Watkins (1963b) see the absolute and conjunct
endings as reflexes of the Proto-Indo-European primary and secondary
endings respectively. Thus, Old Irish abs. beirid < Proto-Indo-European
primary *bhereti and Old Irish conj. beir < Proto-Indo-European secondary
*bheret. The problem with this is that the primary - secondary dichotomy is
one of tense and remains one of tense in all the other Indo-European
languages.
The particle theory, which appears in many variants, is one of the
currently popular ones and is defended by among others Boling (1972),
Cowgill (1975), Kortlandt (1979), Schrijver (1994, 1997) and Schumacher
(1999). The absolute forms are derived from a verbal form + particle,
although there is no consensus over the form of this particle (*-es-, *-edand *-s- have been proposed, among others) and it needs to be noted that no
one particle will give the right Old Irish forms in all contexts. Detailed
critiques on the particle theory can be found in McCone (1979 and 2006)
and Isaac (2000).
Koch (1987) explains the difference between absolute and conjunct
flexion in Old Irish as a result of prosody and accent shifts in Proto-IndoEuropean and Celtic.
McCone (1979) derives the absolute flexion from an original verbal
form followed by an enclitic which would have blocked apocope. What
became the conjunct forms would have been preceded by a preverbal
element to which the enclitic would have been attached, rather than to the
end of the verb. These forms would therefore have undergone apocope. The
form resulting from form + enclitic has then spread to the form without
enclitic: (table taken from Isaac 2007a)
33

I. Proto-Celt.
II. apocope
III.
*bereti ...
> *beret ...
*bereti-E ...
> *bereti-E ...
>
*P bereti ...
> *P beret ...
>
*PE bereti ...
> *PE beret ... >
IV. Old Irish

>
beirid
*bereti-E ...
>
e.g.beirthi
*P beret ...
>
e.g. dobeir
> *bereti ...
*PE beret ... >
e.g. do-tbeir
This last theory has been modified slightly by Isaac (2007a: 3 and
2009), who notes that if the apocope that gave rise to the conjunct forms
only affected atonic forms, but not the tonic forms, there is no need to
postulate a spread from form + enclitic to enclitic-less form.
2.7 Stem, tense and mood

Old Irish has five different verbal stems from which nine tense/mood
categories are formed with different ending sets. The present stem is used
for the present indicative (primary endings), the imperfect (secondary
endings) and the imperative (specific endings in the singular, conjunct
primary endings in the plural). The subjunctive stem is used for the present
and past subjunctive (primary and secondary endings respectively), the
future stem for the future and secondary future (primary, secondary), the
preterite active stem for the preterite active (specific sets of endings) and the
preterite passive stem for the preterite passive.
The relationship between stem and tense can therefore be shown as follows:
Stem
Present stem
Subjunctive stem
future stem
preterite active stem
preterite passive stem
Primary endings
pres. indicative
pres. subjunctive
future
preterite
preterite passive
Secondary endings
imperfect
past subjunctive
secondary future
imperative
As can be seen, Old Irish makes use of five different stems from which to
form its tenses and moods. Except for the preterite, each stem can take
either primary endings to make the primary (present) tenses or secondary
34

endings to make the secondary (past) tenses. For example, taking the present
stem and adding the secondary endings will form an imperfect. The
imperative is formed from the present stem using its own set of endings.
The perfect tense is formed by adding ro to the preterite forms which then
always occur in the conjunct.
2.7.1 Present indicative

For the origins of the present stem formation, see Weak, strong and
hiatus verbs, pages 28ff. A weak verb is AI if its stem-final consonant is
non-palatal, and AII if its stem-final consonant is palatal. In the present stem
of the BI verbs the endings were originally preceded by the thematic vowel
*e or *o depending on person, reflecting the Proto-Indo-European paradigm,
e.g.:
*bher-o-h2e *bher-o-mosi
*bher-e-si
*bher-e-ti
*bher-e-ti
*bher-o-nti
In Old Irish, the result of this is that the stem-final consonant of the 2sg.,
3sg. and 2pl. of the beirid type are palatal and that of the 1sg., 1pl. and 3pl.
are non-palatal. BI verbs of the canaid type on the other hand, where the
root vowel has blocked palatalisation, do not show this pattern of alternating
consonant quality (refer to McCone 1997: 29-30 for subclassification of the
BI verbs and McCone 1996: 116 for the sound law).
There are no deponents in this class (GOI 548); the original *-e/odeponents developed *-i e/o- flexion in Proto-Irish (Schumacher 2004: 481,
McCone 1986: 239-40).
The root-final consonant of the BII verbs was originally palatalised
throughout, the thematic vowel originally being preceded by *-i -. The BIII
verbs are inflected like the BI verbs but have a nasal infixed before their
root-final consonant. This nasal occurs only in the present stem and the rootfinal consonant is always d or g in Old Irish. In the BIV and BV verbs, the
present stem ends in a non-radical nasal, which is non-palatal in BIV and
originally palatal in BV. BV verbs are rare.
35
2.7.2 Imperfect
The imperfect is formed from the present stem using secondary
endings. The origin of these secondary endings is still very much obscure,
but it seems unlikely that that the usage of these forms as secondary endings
goes much further back than the late prehistory of Goidelic and Brittonic
(cf. McCone 1986: 240-1). The imperfect and the other secondary tenses are
always preceded by a verbal particle even when affirmative (the verbal
particle no is used when no other particle is required, but can be omitted in
poetry, GOI 580) and therefore only has a conjunct set of endings. Because
compound verbs are by their very nature always preceded by a preverb,
these do not use the particle no and the usual interplay between deuterotonic
and prototonic occurs. There are no deponent endings; the deponent verbs
use active endings. The imperfect is used for the habitual past.
2.7.3 Imperative
The imperative is formed from the present stem using distinct
endings for the 2sg. and 3sg. and primary conjunct endings for the plural.
The 1sg. is rarely attested, but seems to have been identical to the conjunct
ending of the present indicative. There is no distinction between absolute
and conjunct and the imperative of compound verbs is always prototonic
unless an infixed pronoun is used.
2.7.4 Subjunctive
Old Irish has two subjunctive formations: the a-subjunctive (weak
verbs and most strong verbs) and the s-subjunctive (strong verbs whose root
ends in a dental or guttural stop or spirant or with a present stem in -nn, but
not agaid drives, which takes a-subjunctive). McCone (2005a: 122) adds
to this a separate e-subjunctive of his class H2 of hiatus verbs, which
Thurneysen registers under the a-subjunctive (GOI 608).
The adding of the suffix -s- to the relevant roots leads to assimilation
of the root-final consonant and suffix -s-. Moreover, due to the phonological
36

processes in the prehistory of the language and analogical pressures, the
vowel of the stem can change between present and subjunctive (cf. McCone
2005a: 128-9, Watkins 1962b: 131). Thus the subjunctive stem of guidid
prays is gess- (GOI 387) and the subjunctive stem of bongaid is bss(GOI 756).
The flexion of the s-subjunctive is identical with that of the
s-preterite and the s-future, with the exception of the 1sg. absolute of the
s-future.
The past subjunctive is formed by adding the secondary endings to
the subjunctive stem. As with all verbal forms with secondary endings, these
forms are always conjunct and if not preceded by any other particle, they are
preceded by the verbal particle no.
2.7.4.1 Origin of the subjunctive

Pedersen (ii 354-5) and Thurneysen (GOI 596) derive the
subjunctive from the Proto-Indo-European aorist. Thurneysen specifies
(623) that the thematic forms of the s-subjunctive come from old aorist
subjunctives and the non-thematic forms are explained as due to influence
of the s-preterite. Even here, though, he states that it is possible that these
were originally aorist indicative forms used modally.
According to Watkins (1962b: 118-9, 133), the a-subjunctive
continues the Proto-Indo-European optative and was originally formed by
all roots. In the merger of aorist indicative and perfect that would produce
the preterite (see page 41 below), roots of a certain shape kept the aorist
formation and others kept the perfect. According to Watkins, the aorist
formations in the latter cases did not all disappear. Roots ending in a stop or
fricative kept their aorist indicative formation, used now no longer as aorist,
but as a subjunctive (p. 125).
Rix (1977: 152-4) derives the entire Old Irish subjunctive formation
from the Proto-Indo-European unreduplicated desiderative. Between rootfinal liquids and the initial *-s- of the suffix, a laryngeal was generalised,
which in Celtic appeared as *-a-. Therefore, the suffix *-Hse/o- appeared in
Celtic as *-ase/o-. This *-a- was then generalised in all roots with final
37

resonant. After lenition the suffix became *-ahe/a-. Finally, the *-a- is
lengthened and the *-h- disappears, giving the a-subjunctive of strong verbs.
McCone (1986: 244) points out several problems with Watkins
theory and suggests that the collapse of the aorist indicative was general:
none appeared as modal forms (cf. Cowgill 1983: 86). McCone agrees with
Rixs derivation of the subjunctive from forms with full-grade root and
thematic *-se/o- suffix, however of the two possible Proto-Indo-European
categories that would match those forms, he prefers the s-aorist subjunctive
to Rixs desiderative. He considers deriving the subjunctive from old
desideratives to be problematic, because it would involve a partial shift
from desiderative to subjunctive, whereas a derivation from IE reduplicated
desiderative [to Old Irish future] and s-aorist subjunctive [to Old Irish
subjunctive] respectively is completely straightforward (p. 245). It does not
require a shift from indicative to subjunctive and there are no clashes with
the t-preterite. The only anomalous form is the 3sg. which looks like it
should have come from athematic forms. McCone explains this by analogy.
The s-subjunctive and the s-preterite would have already conjugated
identically outside the 3sg., since the s-preterite was secondarily
thematisised. The formal identity was made complete by introducing the
3sg. forms of the s-preterite into the s-subjunctive.
2.7.5 Future
All weak verbs and the hiatus verbs in -o, -u and -e (McCones H3)
take f-future (except for scaraid separates, parts, gataid takes away,
removes and caraid loves, which have strong formations). Strong verbs
take sigmatic future if they have s-subjunctive and asigmatic future if they
have an a-subjunctive (except for compounds of icc, moinethar and some
of em, which take f-future).
2.7.5.1 Origin of the future formations

The entire Old Irish future formation can be traced back to the ProtoIndo-European reduplicated desiderative with *-se/o- suffix. The root is in
38

the zero grade and unaccented, the reduplication vowel is -i- and unaccented
and the thematic vowel of the suffix is accented (LIV2 23-4)8. The origin of
the Old Irish unreduplicated forms (as discussed in GOI 661-2) is disputed.
For bibliographical details see McCone (1991b: 165-6).
Historically, scholars have treated the weak and strong future
formations as reflexes of different formations (cf. GOI 637, 669, Pedersen
ii 364-5 and see Watkins 1966: 68-74 for a historical overview of both
sides) or as reflexes of the same formation (deriving the entire future
formation from one Proto-Indo-European formation, cf. Watkins 1966, who
derives the entire Old Irish future formation from a Proto-Indo-European
desiderative, though from different formations within that, and McCone
1991b: 180). The latter unified theory is the younger and will therefore be
discussed second.
Pedersen (ii 363) states that the f-future has been inherited from
Celtic and calls it a b-future of which the suffix occurs as -f- in certain
environments. He connects this b-future formation to the Latin b-future
where the suffix is derived from the Proto-Indo-European root *bheubecome (ii 364). Thurneysen (GOI 637) and Watkins (1966: 68) discard
this theory on the basis of the phonological evidence and state that not *-bbut *-f- was the original consonant and that the suffix was originally nonpalatalised, based on forms such as 3sg. conj lic-fea, where the
palatalisation can be shown to be due only to the syncopated high front
vowel before -f-. The palatalisation occurring in AI verbs is not explained,
although Watkins suggests analogy. As for the strong future formations,
Pedersen states that the s-future does not derive from a desiderative, but is
originally a subjunctive of the perfect and he relates it to the Latin futurum
exactum. He suggests that the asigmatic future could be related to the Latin
/-future. Thurneysen, on the other hand (GOI 669) does make the
connection with the Sanskrit desiderative and derives all the strong
formations ultimately from an s-formation. In roots with final sonant, a
8
The formation of the desiderative in PIE has been extensively debated, the problem being
that both the morphology and grammatical function of such a category in Proto-IndoEuropean are unclear, because the daughter languages point towards different prototypes.
Cf. McCone 1991b: 137-8 for fourteen proposed theories on the structure of the Proto-IndoEuropean desiderative.
39

laryngeal was inserted between root-final consonant and suffix, giving *-ain Celtic. The *-s- then disappeared in certain environments due to the
normal phonological processes leading up to Old Irish.
Watkins (1966: 67-81) derives the entire Old Irish future formation
from an originally unreduplicated Proto-Indo-European desiderative, with
thematic suffix *-s-e/o- and full grade or zero grade depending on the shape
of the root. Due to formal dependence of the future on the subjunctive, the
suffix was reinterpreted as if it were the same suffix as that which formed
the subjunctive. Watkins thus gives a morphological, rather than a
phonological reason for the disappearance of the *-s- in the asigmatic
future, unlike Thurneysen above. For the -f- of the weak futures Watkins
suggests that it is cognate to the Skt adjectival suffix -su- based on the
desiderative stem of verbs. To this was added the *-- of the Irish
a-subjunctive to give *-sw- giving lenited unstressed -fa-.
McCone (1986: 248-9, cf. 1991b: 180, 2005b: 202) derives the
strong futures from an Indo-European desiderative with i-reduplication,
zero grade root ..., thematic *-se/o- suffix and primary endings and expects
this formation to have provided the source for the f-futures first in those
primary verbs that took on weak flexion (McCone 1991b: 180). He finds it
problematic to derive an entire finite verbal category from a participle and
instead sees analogy within Old Irish itself for the spread of the weak future.
For further details of this development, see McCone (1991b: 180-1).
However, explaining the f-future by internal analogy within Old Irish
ignores the Brittonic h-future formations (cf. Isaac 1996: 368-71).
2.7.6 Preterite active

The preterite active stem is used to form the preterite and the perfect;
both are made with the same ending sets. The perfect is formed by using the
verbal particle ro and conjunct endings.
All weak verbs and the strong verbs gaibid takes and ibid drinks
have s-preterite. Already in the Milan Glosses and even occasionally in
40

Wrzburg, s-preterite forms of other strong verbs are found.9 The strong
verbs in -l and -r and some in -m and -g have t-preterite. All other verbs
have suffixless preterite.
In Old Irish, the s-preterite is formed by adding -s to the verbal root.
Historically, the flexion contains a mixture of thematic and athematic forms
(Schumacher 2004: 66-7). The t-preterite is formed by adding -t to the stemfinal consonant. Where the stem ends in -m, this disappears before -t, with
compensatory lengthening of the preceding vowel; a root-final dorsal
becomes -ch- before -t, but disappears between -r and -t. Thus the preterite
of em is t, that of aig is acht and that of oirg is ort. In the suffixless
preterite, the verbal endings are added directly to the verbal root which is
modified internally in some way. Most suffixless preterites are reduplicated,
but some are formed by changing the stem vowel to -- or --.
2.7.6.1 Origins of the preterite

The Old Irish preterite is the result of a merger of the Proto-IndoEuropean aorist and perfect. The distinction between aorist and perfect was
preserved in Common Celtic (Schumacher 2004: 59-60).
Most Proto-Indo-European aorist formations except the s-aorist
disappeared from the language (McCone 1986: 230). The only remnants of
other aorist formations in Old Irish are luid (suppletive preterite of tit
goes) and docer (suppletive preterite of dotuit falls). The Proto-IndoEuropean s-aorist was a late development in the proto-language (cf. Meillet
1908: 79-106), but a highly productive one, and it led to the Old Irish s- and
t-preterites (McCone 1986: 232, Schumacher 2004: 61, 66), the latter
coming into existence due to regular loss of -s- between a continuant and a
following consonant (Zimmer 1890: 198-217, see McCone 1996: 99 and
Isaac 2007c: 62, 64 and 71 for the sound law). E.g. *axst > *axt > OI acht.
In all other verbs in which the s-aorist was retained as s-preterite (i.e.
all weak verbs), the *-s- came between the stem-final vowel (-i or -a) and
The original preterite of frisacci looks forward to, expects is reduplicated,

ni-rufrescachtar Ml. 26b25, 34d17, but s-preterite forms are already found in Ml. 34c11
and 72c13: ni-rufrescisset. Arfom accepts, receives, should take t-preterite, ar-roet Ml.
17c3, but already in Wb. 26a23 is found with s-preterite -arromsat (McCone 1985: 92-3).
41

the ending. In the 3sg., this led to *-Vst-, which became *-Vss- (cf. also GOI
672, Watkins 1962b: 177). The double *-ss- then spread to other persons.
There is consensus about this origin of the s-preterite, but for the origin of
the t-preterite, several other theories have been put forward. The detailed
discussion of these by Watkins (1962b: 156-9, see note 3 for bibliographical
details) needs not be repeated here.
McCone (1986) picks up on Zimmers original idea and derives the
t-preterite from the s-aorist. However, his statement (1986: 233) that
whether the original aorist form or the original perfect form prevailed
depended on which one best preserved the shape of the root is arbitrary at
best and certainly rather subjective. The same can be said for Watkins
argument for the lack of t-preterites in roots ending in -n (1962b: 169-70)
that the shape of the preterite in those would be so different from the present
that that did not happen. Old Irish tolerated a subjunctive stem gess- for a
verb guidid (GOI 387), a future stem ebra- for a verb ernaid (GOI 403)
and a perfect focemallag to the verb foloing (GOI 688): arguments that
certain potential forms in Common Celtic were too unlike the present stem
to be tolerated are impressionistic and unconvincing.
The reduplicated preterite and most of the long vowel preterites can
be derived directly from the Proto-Indo-European reduplicated perfect
which had o-grade in the singular and zero grade in the plural. Late in the
prehistory of Old Irish, the zero grade of the plural stem was generalised
(McCone 1986: 233, 235). The - and -preterites are ultimately the result of
lenition and/or the loss of reduplication (Watkins 1962b: 108, McCone
1986: 235). The Old Irish -preterite on the other hand, goes back to the
Proto-Indo-European reduplicated perfect. The long vowel in the -perfect
is related to that in the Germanic -preterites and unless Latin has innovated,
to the Latin -perfects (for an alternative explanation of both the Latin and
the Germanic long vowel perfects, see Sihler 1995: 582). The long vowel
came into existence due to compensated loss of the second of two identical
consonants between *-e- and a stop or *-m- (*CeC-T/m- > *C-T/m-, -ebeing the reduplication vowel). This only occurred in the plural, since in the
singular, the root would be in the o grade, rather than the zero grade, and
therefore the second identical consonant would not stand between *-e- and a
42

stop, but between *-e- and *-o- (*CeCoT/m-). Long *-- changes to *-- by
analogy in Celtic and the split of *-- to *-- in final and *-- in non-final
syllables gives OI CT/m-: the -preterite (McCone 1986: 236-7).
2.7.7 Preterite passive

The preterite passive is formed by adding a dental suffix to the
general verbal stem in the case of weak verbs (with syncope of the stemfinal vowel) and to the verbal root in the case of strong verbs. The preterite
passive stem corresponds to the Proto-Indo-European verbal adjective in
*-to-, which was used to supply this category. However, in Old Irish the
forms are entirely verbal (GOI 705). There is no satisfactory explanation
for the absolute forms (McCone 1994: 172).
2.8 Voice
Old Irish has a fully functional active-passive dichotomy. In the
active voice, the agent is the subject of the verb, a possible patient being
expressed by the object in the accusative case. In the passive voice, the
patient is the subject of the verb, a possible agent being expressed by the
preposition la. The active voice is fully inflected, for the passive voice the
3sg. form is used for all persons except the 3pl., which has its own form.
Outside the 3sg. and 3pl., the subject is expressed by infixed pronouns.
Old Irish retains the deponent inflection, which, when the deponent
verb is not one of the -igidir verbs, retains the semantics of the Proto-IndoEuropean middle voice.10 Examples of Old Irish deponent verbs in my
corpus are labraithir speaks, talks and cobraithir helps. Despite the
obvious productivity of the -igidir formation, the deponent flexion is dying
10
The middle voice appears when the verbal action affects the subject directly or
indirectly, or, when the verbal action does not have an affect (sic, effect meant) beyond
the subject. When the subject is plural or dual, the middle voice also expresses reciprocity.
(Meier-Brgger 2003: 259)
43

out in Old Irish. Active by-forms are already found in Wrzburg and are
more widespread in Milan (Cowgill 1983: 73, cf. McCone 1985: 95).
Concerning the origin of the Old Irish deponent and passive flexions,
Cowgill (1983) has argued convincingly that the Indo-European
mediopassive split up in (Insular) Celtic, with the old 1st and 2nd persons
functioning as deponents and the old 3rd persons as passives, while for 3rd
person deponent Irish uses new forms of enigmatic origin, and for 1st and
2nd person passive all of Insular Celtic uses forms in which the 3rd singular
passive behaves like an impersonal active verb, with object pronouns
marking person and number of the patient (p. 104). For further details on
the origin of the deponent and passive flexion, see Cowgill (1983) and
McCone (1994: 143-5).
2.9 5umber and person

Old Irish verbs retain from the Proto-Indo-European verbal system
singular and plural inflection, however losing the dual (as opposed to the
nouns, which also retain the dual) and have first, second and third person in
both numbers. With a few exceptions, dual subjects take a plural verb.
Singular collectives also take plural verbs.
44
Analysis
3 Analysis
Morphologically and semantically, active and non-igidir deponent
verbs behave the same. Both are formed by adding the denominative suffix
*-- or *-- to a noun or adjective and both use either *-- or *-- depending
on the semantics of the resulting verb and how it relates to its base as
described below, section 3.3.1, pages 55ff. Although deponent verbs inflect
differently than active verbs (cf. section 2.8, page 43 above for a description
of voice and the semantics of the deponent in Old Irish), for present
purposes they can be treated together. Therefore, in the current chapter, AI
and AII will refer to both active and deponent verbs, though not to -igidir
verbs.
3.1 Basic statistics

The corpus consists of 365 secondary verbs overall. There are 106 AI
verbs and 259 AII verbs. Of the AI verbs, 97 are active and 9 are deponent.
Of the AII verbs, 64 are active, 194 are deponent and one, cuicsedar taxes
(page 187), could be either active or deponent. Of the deponent AII verbs,
187 are -igidir verbs and 7 are non-igidir verbs. In the diagram below,
percentages are given of the category the verbs belong to (i.e. 187 -igidir
verbs is 96.4% of the deponent verbs).
365 Verbs
106 AI verbs
29.0%
97 Active
91.5%
9 Deponent
8.5%
259 AII Verbs

71.0%
64 Active
24.7%
194 Deponent
74.9%
187 -igidir
96.4%
45
1 unknown
7 non-igidir
3.6%
Analysis
The 106 AI verbs can be further subdivided into 47 active

denominatives, 24 active deadjectivals, 19 Latin loans, 5 deponent
denominatives and 4 deponent deadjectivals. 7 Verbs cannot be counted,
since it cannot be determined whether they are deadjectival or primary
(caraid loves, page 117), denominative or deadjectival (dota clothes,
page 129) or whether they are denominative/deadjectival or a direct loan
from Latin (armaid arms, page 111, clannaid plants, page 120, draid
hardens page 132, notaid marks, page 145 and seccaid stiffens, page
152).
106 AI verbs
47 active denominatives
44.3%
24 active deadjectivals
22.6%
19 Latin loans
17.9%
5 deponent denominatives 4.7%

4 deponent deadjectivals
3.8%
7 unknown
Of the 259 AII verbs, the following 21 verbs present various

difficulties that prevent their being counted:
igid cries out (page 204): derivation unknown
foniti mocks (page 214): etymology unknown
glasid moves (page 218): etymology unknown
maithid forgives (page 230): etymology unknown
minid teaches (page 233): etymology unknown
sndaid protects (page 244): etymology unknown
tibid touches (page 250): etymology unknown
dobidci shoots (page 198): etymology unknown
sluindid expresses (page 244): denom./deverb.
bgaid declares (page 172): denom./deverb.
baitsid baptises (page 173): denom./loan
cinid is born (page 181): prim./denom.
adbligedar flatters (page 158): denom./deadj.
adchotadaigidir reconciles (page 158): etymology uncertain
tailngigidir deigns (page 254): denom./deadj.
atigidir becomes few (page 257): denom./deadj.
cuimnigidir remembers (page 188): denom./deadj.
follaigidir neglects (page 213): denom./deadj.
46
Analysis
idigidir lives as a Jew (page 225): denom./loan
adgldathar addresses (page 159): onomatopoeic
cuicsedar taxes (page 187): act./depon.
The 259 AII verbs can be further subdivided into 106 -igidir
denominatives, 73 -igidir deadjectivals, 1 -igidir depronominal, 26 active
denominatives, 25 active deverbals, 4 deponent denominatives, 2 deponent
deverbals and 1 Latin loan:
259 AII verbs

106 -igidir denominatives
40.9%
73 -igidir deadjectivals
28.2%
1 -igidir depronominal
0.4%
26 active denominatives
10.0%
25 active deverbals
9.7%
4 deponent denominatives 1.5%

2 deponent deverbals
0.8%
1 Latin loan
0.4%
21 unknown
Leaving out the 20 Latin loans and the total of unknown verbs from
both the AI category and the AII category (28 verbs), the 317 remaining
verbs are made up of 188 denominatives, 101 deadjectivals, 27 deverbals
and 1 depronominal:
365 verbs
188 denominatives 51.5%
101 deadjectivals
27.7%
27 deverbals
7.4%
1 depronominal
0.3%
20 Latin loans
28 unknown
47
Analysis
It becomes clear that the most derivative verbs in Old Irish, 51.5%,
are denominatives, totalling 188 verbs. Of these, the -igidir verbs are with
106 out of 188 (56.4%) in the majority, followed by the AI active category
with 47 verbs (25%), 26 AII active verbs (13.8%), 5 AI deponent verbs
(2.7%) and 4 AII deponent verbs (2.1%).
188 denominative verbs

106 -igidir verbs
56.4%
47 AI active verbs
25.0%
26 AII active verbs
13.8%
5 AI deponent verbs
2.7%
4 AII deponent verbs 2.1%
3.2 Latin loans

The influence of Latin on Old Irish vocabulary is clearly felt in the
verbs Old Irish borrowed from Latin. Some are direct borrowings, where
Old Irish took the Latin verb and used it to form an Old Irish verb. Most of
these are Latin a-verbs and become Old Irish a-verbs; some are from other
conjugations in Latin and become Old Irish a-verbs. One is a Latin 3rd
conjugation verb, ungere anoint, which is borrowed as an Old Irish i-verb,
oingid anoints (page 233). Indirect borrowings, where Old Irish has
borrowed the Latin noun and then derives a verb from that within Old Irish,
are more diverse, with both a-verbs and -igidir verbs occurring relatively
frequently. Exact figures will be given below. In a couple of cases it cannot
be ascertained whether a given verb is directly borrowed from Latin, or
indirectly. It can be questioned whether this distinction is, in fact, entirely
meaningful, because the ultimate question is always about how an Old Irish
speaker created a new verb. Whether they did that by taking a Latin verb
and making that into an Old Irish verb, or by taking an Old Irish noun and
making that into an Old Irish verb, is moot. Moreover, in the monastic
context in which these Glosses were written, Latin was in daily use (cf.
pages 20f. in section 1.1.2). The Old Irish noun, the Latin noun and the
48
Analysis
Latin verb would have been part of the speakers lexicon. The answer to the
question whether a speaker took a Latin verb or an Old Irish noun to create a
new verb may be both and indeed the question may again be meaningless.
Any verbs that are in this either/or category do still fit the scheme for
denominatives and deadjectivals that will be argued below (sections 3.3.1.1
and 3.3.1.2, pages 55ff.).
The Latin verbal borrowings into Old Irish in the Wrzburg and Milan
Glosses are the following:
Borrowed from Latin verbs (Lat. verbs 1st conjugation unless otherwise
stated):
Old Irish AI verbs:
adraid adores (page 110): Lat. adorare.
bennachaid blesses (page 112): Lat. benedicere (3rd conj.).
celebraid celebrates (page 118): Lat. celebrare.
consecra consecrates (page 123): Lat. consecrare.
damnaid condemns (page 126): Lat. damnare.
lgaid reads (page 140): Lat. legere (3rd conj.).
maldachaid curses (page 141): Lat. maledicere (3rd conj.).
mescaid mixes (page 142): Lat. miscere (2nd conj.).11
ordnaid appoints (page 145): Lat. ordinare.
panaid punishes (page 146): Lat. punire (4th conj.).
pretaid plunders (page 146): Lat. praedare.
pridchaid preaches (page 146): Lat. praedicare.
promaid puts to the test (page 148): Lat. probare.
rlaid reveal (page 149): Lat. reuelare.
scrbaid writes (page 151): Lat. scribere (3rd conj.).
scrtaid examines (page 151): Lat. scrutari.
slchtaid kneels (page 153): Lat. flectere (3rd conj.).
trchtaid comments (page 154): Lat. tractare.
11
OI mesc is borrowed from Lat mixtum and remodelled in analogy to the verb.
49
Analysis
Old Irish AII verbs:
oingid anoints (page 233): Lat. ungere (3rd conj.).
There are 19 direct borrowings in total. 18 (94.7%) are A1 verbs, one is an

AII verb. Six verbs are borrowed from 3rd conjugation Latin verbs, one from
a 2nd conjugation verb and one from a 4th. Note that the one AII verb is
borrowed from a 3rd conjugation verb. There is therefore no obvious
correlation between Latin and Old Irish conjugations.
Derived from Old Irish nouns borrowed from Latin:

Old Irish AI verbs:
certaid disposes (page 118): deadj. < cert, correct, right, Lat. certus.12
crochaid crucifies (page 124): denom. < croch cross, Lat. crux.
figraid prefigures (page 135): denom. < figair figure, type, Lat. figura.
onaid fasts (page 145): denom. < in fast, period of fasting, Lat.
ieiunium.
Old Irish AII verbs:

sroiglid scourges, lashes (page 247): denom. < sroigell scourge, Lat.
flagellum.
Old Irish -igidir verbs:
cestaigidir inquires (page 181): denom. < ceist question, problem, Lat.
quaestio question.
mesraigidir
moderates
(page
231):
denom.
<
mesar
measure,
measurement, Lat. mensura.

mindchigidir begs (page 231): deadj. < mindech poor, wretched, Lat.
mendicus.
ordaigidir orders (page 234): denom. < ordd order, sequence, Lat. ordo.
pecthaigidir sins, commits sin (page 234): denom. < peccad sin, Lat.
peccatum.
12
The Latin verb certare vie with, contest has been deemed unsuitable as origin on
semantic grounds.
50
Analysis
adfaligedar unveals, reveals (page 255): denom. < fal veil, Lat.
uelum, with preverb ad.
uimligidir humbles (page 257): deadj. < umal humble, obedient, Lat.
humilis.
There are 12 indirect borrowings in total, four are AI, one is AII and seven
are -igidir verbs. Three verbs are deadjectival, one in the AI category and
two in the -igidir category. Nine verbs are denominative, three in the AI
verbs, one in the AII verbs and five in the -igidir verbs.
Unclear:
Old Irish AI verbs:
armaid arms (page 111): denom. < arm armour, battle equipment, from
Lat. arma, or Lat. armare.
clannaid plants (page 120): denom. < clann plant, planting, Lat. planta,
or Lat. plantare.
draid hardens (page 132): deadj. < dr, hard, rigid, Lat. durus, or Lat.
durare.
notaid marks (page 145): denom. < not mark, sign, Lat. nota, or Lat.
notare.
seccaid stiffens (page 152): deadj. < secc stiff, rigid, Lat. siccus, or Lat.
siccare.
Either AI or AII:
baitsid baptises (page 173): denom. < baithis baptism, the order of
baptism, Lat. baptisma, or Lat. baptizare. If baitsid were a direct loan, it
would be an AI verb that looks like an AII verb due to syncope patterns (this
is discussed in its entry in the corpus, page 173). If it is derived within Old
Irish, it is AII.
There are five unclear verbs, of which four are AI verbs, and one that could
be either AI or AII. If these would have been derived within Old Irish, four
would be denominative, one deadjectival.
51
Analysis
These figures can be shown schematically as follows:
1 2nd conj.
18 AI
1 4th conj.
94.7%
19 direct
51.4%
1 AII
6 3rd conj.
5.3%
37 direct and
4 A1
3 denominative
indirect loans
33.3%
1 deadjectival
12 indirect
1 AII
32.4%
8.3%
1 denominative
7 -igidir
5 denominative
58.3%
2 deadjectival
6 unclear
5 AI
16.2%
1 AII
Others:
crsagaid corrects (page 125) is borrowed from the Latin expression curas
agere take care of.
idigidir lives as a Jew (page 225) is either derived from Id(a)ei, which
was adopted from the Latin but of which it is often unclear whether the
writers regarded it as Latin or Old Irish (DIL) or the verbal form iudigte
could have been directly adapted from the Latin.
3.3 Morphology of AI and AII verbs

In Celtic, secondary verbs are formed by adding a suffix *--,
*-- or *-sag-- to a noun or adjective to make a denominative or
deadjectival verb. Verbs formed with suffix *-- give the Old Irish AI active
and deponent verbs, verbs with suffix *-- the AII active and deponent
verbs. Verbs with suffix *-sag- are always deponents of the AII class. The
52
Analysis
other secondary verbs found in Old Irish are deverbals, namely causatives
and iteratives.13 Causatives and iteratives were formed in Proto-IndoEuropean by adding the suffix *-i e- (> CC *--) to the o-grade of the root.
Besides the normal hysterotonic causative with CoC-ie/o- structure, i.e.
where the accent remains on the suffix throughout the paradigm, there exists
a proterotonic lengthened-grade14 type *CC-ie/o- where the accent remains
on the root throughout the paradigm. Of this type, only sidid thrusts, fixes
(page 238) is attested in Old Irish (besides original full-grade hysterotonic
suidid puts, places) (Schulze-Thulin 2001: 4-6).15
Secondary verbs are by definition weak in Old Irish. The AI verbs
contain
denominatives
and
deadjectivals;
the
AII
verbs
contain
denominatives and deverbals; the -igidir verbs are denominative or

deadjectival. Verbs such as asrocholi defines, determines (page 120) and
sobid makes, becomes crooked, contorts (page 245) at first sight seem to
be AII deadjectivals, but are not. These will be further discussed below,
page 62.
The denominative suffix *-- has at least two sources in Proto-IndoEuropean (cf. Schumacher 2000: 75), the first of which is still partly
recognisable in Celtic and Old Irish. It continues an old athematic factitive
with suffix containing *e or *o plus laryngeal *h2 or *h3. Most of the AI
deadjectivals in my corpus are factitives, the exceptions are discussed
below, section 3.3.1.2, page 62. However, considering that deadjectivals
become AI verbs by default in Celtic and Old Irish, it can be questioned
whether this morphology has been maintained. Inherited denominatives
would also have had suffix *-- (< PIE *-eh2-i e/o-). The type was originally
used for denominatives of all -stem nouns and some o-stem nouns (Sihler
1995: 514). The majority of secondary AI verbs in my corpus are derived
from an or an o-stem. However, this does not seem to be the underlying
reason for derivation as an AI verb, see below, pages 55ff. It is no longer
13
The essive verbs ruidid and scibid have been mentioned above, page 54.
Roots in PIE appear in full grade, lengthened grade and zero grade. In the full grade, the
root contains a vowel, in the lengthened grade, this vowel is made long and in the zero
grade no vowel appears.
15
For alternative etymologies of sidid and other prototonic lengthened-grade roots, see
Schulze-Thulin (2001: 4-6).
14
53
Analysis
possible to ascertain whether an old denominative was inherited from ProtoIndo-European or coined in Celtic, however, and in no case has a substantial
amount of cognates been found to justify positing a Proto-Indo-European
derivation. Therefore, inherited denominatives have been placed in Celtic.
An example of an old denominative is forcenna puts an end to (page 135),
cognate to W gorffen to finish, end, which Schumacher sees as an
inherited denominative from Insular Celtic (2000: 75, 142).
The suffix *--, when not the causative / iterative suffix, is either the
essive suffix or perhaps goes back to the denominative suffix *-i e/o- added
to i-stem nouns (Schrijver forthcoming).16 Note that in this particular use of
a denominative of i-stems no longer applies in Old Irish: the only two bases
that are certainly i-stems (rind point and samail likeness) give a-verbs
rindaid cuts, incises (page 149) and samlaithir likens, compares (page
149) respectively. Already within Common Celtic the patterns underlying
verbal derivation point towards semantics (transitivity and iterativity), rather
than morphology (see further below, page 55).
Common Celtic had four essive presents, *klus--, *ksib--, *rud-*tum-- and *t-i e/o- (Schumacher 2004: 41-2), of which two are attested in
Old Irish.17 The other two are only attested in Brittonic18 and need not
concern us here. OI ruidid turns red, flushes and Mid. Ir. scibid moves
are not attested in the Glosses, but they are secondary verbs. For ruidid, see
Watkins (1969: 169-70), McCone (1991b: 44), LIV2 (508-9) and
Schumacher (2004: 552-3). The origin of scibid as an essive has been
proposed by Schrijver (2003, see also Schumacher 2004: 422-4, IEW 10412).
16
I would like to thank Prof. Peter Schrijver for sending me a copy of his article prior to
publication.
17
t is already in Celtic no longer felt as an essive, since the first vowel of PIE *sth2eh1i e/o- had become through colouring by and loss of the laryngeal(s).
18
CC *klus-- > MW clywet hear, feel, MB clevet hear, MC klywes hear (Schumacher
2004: 415-6) and CC *tum-- > MW tyfu grow, increase, ModB tevi grow, increase,
MC tevi grow (Schumacher 2004: 646-7).
54
Analysis
3.3.1 Semantic motivation for formation as AI or AII verbs

The semantic range of derivative verbs will be discussed extensively
below, section 3.9, pages 86ff. In the current section, the motivation behind
formation as either an a-verb or an i-verb will be discussed.
3.3.1.1 Denominatives
Whether a denominative verb becomes an AI or an AII verb depends
on its transitivity, iterativity and causativity. The terms causative and
iterative are here not used in their morphological meaning, these are not
deverbal verbs. The terms are here used semantically only.
The table below shows these three categories (transitivity T, with T
for transitive, A for ambitransitive and I for intransitive,19 causativity C
with Y for yes and N for no and iterativity It with Y for yes and N for
no), plus the stem class of the base noun. As has been pointed out above
(page 54), the stem class of the base noun is no longer of importance in the
formation of the verb. As is shown in the table, most base nouns are o- or stems, of the AI verbs, three are u-stems and two are i-stems. Of the AII
verbs, three base nouns are i-stems, one is an io-stem and one is a u-stem.
A question mark denotes that no stem is given in DIL and none can be
deduced from extant sources. The verbs are roughly subdivided into old,
young and unknown age. This will be referred to in the Chronology chapter,
pages 75ff. For this purpose, too, the presence or absence of Brittonic
cognates (Bc) is noted. Finally, the base noun and its meaning are given.
19
Note: verbs are considered transitive if the action itself must be done to an object. For
present purposes, it does not matter whether this object happens to be expressed as a direct
object in the accusative, or an indirect object following a preposition. Verbs are considered
ambitransitive if the action, depending on meaning and context, can require an object
(direct or indirect) or not.
55
Analysis
AI denominatives vs. AII denominatives

AI
stem Bc
It
origin
AII
stem
Bc
It
origin
Old
berbaid boils
bertaid brandishes
callaithir thinks about
crothaid shakes
donaid consoles
dnaid shuts
feraid grants
fobotha is alarmed
forcenna puts an end to
nertaid strengthens
todid shines
o/
?
o
?
?
o
o
?
5
5
Y
5
5
5
5
5
Y
Y
5
A
A
A
A
T
A
A
A
A
T
A
5
5
5
Y
5
5
5
5
5
5
Y
5
5
5
Y
5
5
5
5
5
5
5
? bubbling
bert load
call sense
CC *krotos shaking
*diden care
dn fort
PIE *h2uer- water
*butt
forcenn end
nert strength
*to-id-
laid requests
caithid consumes
ceisid complains
dilid portions out
durnid smites
rmid counts
?
?
?
Y
5
5
Y
Y
Y
T
T
T
T
T
T
Y
5
5
Y
5
5
5
Y
Y
Y
Y
Y
il act of asking
*katces debility
dl dispensing
dorn fist
rm act of counting
Young
bathaid feeds
crochaid crucifies
delbaid shapes
figraid prefigures
onaid fasts
rannaid divides
5
5/A
5
5/A
5/A
5
A
T
A
?
A
A
5
5
5
5
5
5
Y
5
5
5
Y
5
biad food
croch cross
delb shape
figair figure
an fast
rann part
baitsid baptises
cinnid defines
sroiglid scourges
?
o
o?
5/A
Y
5/A
T
T
T
Y
5
5
5
Y
Y
baithis baptism
cenn end
sroigell scourge
Unknown age
adcobra desires
araid satirises
armaid arms
said grows
brchtaid bursts
o
?
?
5
5
5/A
5
5
T
T
T
I
I
5
5
5
5
5
5
5
5
5
5
accobar desire
ar cutting
arm armour
s growth
brcht burst
airid guards
i
airlithir advises
i
athmuilnethar says again o
attobi adheres to
o
bgaid declares
5
5
5
5
5
T
T
T
T
T
Y
Y
5
5
5
Y
Y
Y
Y
Y
ire guarding
airle advising
muilend mill
taeb side
bg boast
56
Analysis
clannaid plants
cobraithir helps
condelca compares
crechaid plunders
crachaid piles up
dechraid marks out
dlomaid announces
doecrathar covers
dolbaid fashions
domolta goads
drbaid lingers
emnaid makes double
gataid takes away
gellaid pledges oneself
gallaid gives hostages
achtaid cries out
ccaid pays
lthraid arranges
lnaid fills
lobaid decays
logaid concedes
lommraid shears off
malartaid spoils
messaid examines
molaithir praises
mchaid covers over
rindaid cuts
samlaithir compares
ssaid satisfies
selbaid possesses
techtaid has
trebaid cultivates
?
o
o
?
o
o
u
o
o
?
o
u
?
s
?
u
o
i
i
o/
5/A
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
?
5
5
5
5
5
T
T
T
A
A
T
A
T
T
T
I/A
A
T
A
A
A
A
T
T
A
A
T
T
A
T
T
T
T
T
T
T
T
Y?
5
5
5
5
5
5
5
5
5
5
Y
5
5
5
5
5
5
5
5
5
5
Y
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Y
5
5
5
5
5
Y?
5
5
Y?
clann plant
cobair help
coindelc comparison
crech plunder
crach stack of corn
dechor difference
dlom proclamation
tcar shelter
dolb sorcery
molad praising
drb delay
emon twin
gait taking away
gell pledge
gall hostage
acht loud cry
cc paying
lthar arrangement
ln full number
luib herb
lg value
*lommar
malart damage
mess judging
mol reiterated refrain
mch smoke
rind point
samail likeness
ss snare
selb property
techt possessions
treb house
57
brissid breaks
badraid disturbs
i
canid laments
?
coillid damages
o
conruidethar intends io
conterchomraic(i) collects o
crdaid torments
?
cuicsedar taxes
u
dosc(u)lai finds out o
foirrgid presses hard
?
forcomai keeps
?
gledid cries out
scalid bursts
?
5
5
5/A
5
5
5
5
5
5
5
5
5
?
T
T
T
T
T
T
T
T
T
T
T
I
A
Y
Y
5
Y
Y
Y
Y
Y
Y
Y
5
5
Y
Y
Y
Y
Y
5
5
Y
5
Y
5
Y
Y
Y
bres fight
baidre confusion
cane act of weeping
coll destruction
suide act of sitting
comrac meeting
crd torment
cs tax
scl story
forrach oppressing
coim keeping
gled shout
scal scattering
Analysis
Based on these findings, the following rule can be formulated:
A verb is AII if it is transitive and iterative and/or causative. Otherwise it
is AI.
Exceptions:
gledid cries out (page 218): intransitive, AII. Iterative.
scalid bursts (page 240): ambitransitive, AII. Causative and iterative.
malartaid spoils (page 141): transitive, AI. Causative and iterative.
emnaid doubles (page 133): transitive, AI. Causative.
Gledid and scalid can be explained. Both have synonyms which
are themselves also denominative active verbs. These synonyms are
brchtaid bursts (page 116) and achtaid cries out (page 138)
respectively. Like scalid, brchtaid is also ambitransitive, and like gledid,
achtaid is also intransitive. Yet both brchtaid and achtaid are AI verbs,
whereas gledid and scalid are AII verbs. However, whereas the base for
brchtaid is brcht burst, the base for scalid is scal scattering. A burst
is a sudden occurrence; the derived verb therefore describes a sudden
occurrence. Something bursts. A bursting is caused to something. A
scattering, on the other hand, implies a repeated action. Scalid is iterative,
as opposed to brchtaid. From a basic meaning scatters a meaning bursts
has developed straightforwardly: drop a stone jar and it bursts; pieces scatter
across the floor.
A similar case can be made for gledid and achtaid. Whereas
achtaid is used for lamentations, groans and wailings (cf. DIL s.v. achtaid
12: 12), gledid is used for inarticulate animal noises (cf. DIL s.v. gledid
89: 32, the transitive meaning is late and likely to be a secondary
development). Animal noises are by their very nature iterative, they are
repeated several times. Therefore, it can be argued that gledid is more
iterative than achtaid. Note that canid laments has a similar meaning to
achtaid, but is AII. Canid is however always transitive, so does not pose a
problem.
59
Analysis
If both scalid and gledid are more iterative than their synonyms,
there will have been a certain amount of pressure on those verbs to become
AII verbs rather than AI verbs. The different verbal classes could then
emphasise that these verbs are not exact synonyms.
The opposite is the case for malartaid spoils (page 141). It too, has
a denominative synonym, namely coillid damages (page 183), derived
from coll destruction. According to the scheme, both malartaid and coillid
should become AII, because both these verbs are transitive, causative and
iterative. Yet only coillid is AII. The reason for this is that coillid is more
iterative than malartaid. After all, malartaid only means spoils, or, when
looking at its base, causes injury, whereas coillid means damages or
causes destruction. In order to cause destruction, one must repeat the
damage a lot more often than to just cause injury.
Emnaid doubles (page 133), on the other hand, is superficially an
exception. It is causative, possibly even iterative and medieval attestations
are apparently transitive. One would expect it to be AII. It is AI. Its
meaning, however, lends itself to an ambitransitive reading, even if not
actually attested in extant medieval sources as such. And the modern reflex
eamhnaigh is intransitive. AI morphology is accounted for by inherent
ambitransitivity.
Crothaid shakes (page 124) is given in the table as ambitransitive.
All entries in DIL are transitive (s.v. crothaid 552: 19), but crothaid has
more in common with the ambitransitive verbs than with the transitive
verbs. When the subject shakes something (transitive), the object is by
definition shaking (intransitive). None of the transitive verbs have this
connotation. If you arrange something, that something will not arrange, if
you shear something, that something will not shear, and so on. On the other
hand, ambitransitive verbs do have this connotation: if you double
something, that something will also double and if you boil something, that
something will also boil, to name but two examples. Crothaid behaves like
an ambitransitive verb, even if intransitive usage is not attested. This is an
old derivative, as has been detailed in the corpus below (page 124), so it is
also not impossible that its intransitive meaning got lost over time.
Moreover, if the rules given above are accepted, the only way this verb
60
Analysis
could rightly have become an AI is if it were ambitransitive rather than
transitive.
The verb crachaid makes a heap of something, piles up (page
125) could be seen as inherently iterative, but in relation to its base noun, it
is not. It is derived from the word crach, pile, stack, so the verb is about
the end product, not the process. Making a pile is not iterative. A similar
issue occurs with rindaid cuts, incises (page 149), which in meaning
carves could be seen as iterative. However, the meaning originally simply
meant putting a point to something. Whereas carves implies doing that
repeatedly, the verb itself does not necessarily.
Certain verbs occupy a similar semantic range, but end up in
different morphological categories. These verbs illustrate the theory set out
above. Cinnid defines, fixes (page 181) and forcenna puts an end to,
brings to an end (page 135), for example, are both ultimately derived from
cenn head and both ultimately mean putting an end to something.
However, cinnid means defining something, i.e. putting ends all around it,
iteratively, and forcenna is a single action: putting an end to something.
Finished. Moreover, cinnid is transitive only, whereas forcenna is
ambitransitive. So, cinnid, iterative and transitive, is an AII verb, and
forcenna non-iterative and ambitransitive, is an AI.
The exceptions show that this scheme is not necessarily set in stone.
It works over a substantial portion of the corpus, but there are some verbs on
the edges of it that either show a certain amount of flexibility on the part of
the scheme, or that show how the verbal formation and the scheme interact.
The fact that there was a scheme underlying verbal derivation also allowed
for interplay between this scheme and the derivative verbs. The hypothesis
of this thesis is that there is no random variation. A verb is AI or AII for a
reason. The creator of a new verb knew what he or she was doing, be that
consciously or subconsciously. Sobid makes crooked (page 245) and
consoba deceives (page 245) for example fall into two different classes,
at first sight for no particular reason. Thus far, the the scholarly tradition
within the study of the Celtic languages has been to accept this seeming
inconsistency, rather than to look for underlying motivation. In this thesis, I
61
Analysis
have teased out a possible motivation. In the case of sobid and consoba
the underlying motivation is an interplay between the general scheme of
formation of new verbs in Celtic and Old Irish on the one hand and these
particular verbs on the other. The general scheme is that deadjectivals
become AI verbs. The fact that sobid is an AII would therefore show that it
was considered to be iterative (the possibility of causativity is precluded by
the fact that this is already a factitive) to such an extent that its iterativity
exerted enough force to make it an AII. The compound consoba on the
other hand, contains the preverb con- which indicates completion. This
excludes the possibility of iterativity and therefore this verb is an AI. A
more complete understanding of the effect of preverbs is a desideratum.
3.3.1.2 Deadjectivals
The deadjectivals in Old Irish are mostly factitives, i.e. they mean
making [object] [base]. Glanaid cleanses (page 137), for example, is
derived from glan clean, pure. The verb means makes [object] clean.
Soraid frees (page 153) is derived from sor free and the verb means
makes [object] free. The exceptions are brnaid is rotten (page 116)20
and fgaid looks at (page 133).
The deadjectivals are AI with only three clearly motivated
exceptions, namely asrocholi defines (page 120), doaissilbi refers,
ascribes (page 152) and sobid makes, becomes crooked (page 245).
Sobid has already been discussed above, page 61.
Asrocholi defines is derived from cel thin, slender (DIL s.v.
cel 10: 82). It is a compound of colaid makes thin (page 120, DIL s.v.
celaid 13: 3), which is an AI verb (asrocholi will therefore be found
among the other AI verbs). There is a very good reason that this compound
would become an AII verb, and that is that this is iterative. Asrocholi
means defines, determines. Since this is a compound of a verb meaning
makes thin, the meaning defines must have been formed through the
implication of making thinner, making narrower. Narrowing something
20
Brnaid later also comes to mean putrifies, although this does not seem to be the
original meaning.
62
Analysis
down is not a punctual action, but a repetitive one. To define something, the
making narrower has to be done repeatedly. Asrocholi therefore fits in the
wider framework of Old Irish denominatives and deadjectivals. It is
transitive and iterative and thus AII.
A similar process is seen in doaissilbi refers, ascribes, which is the
AII compound of an AI verb selbaid possesses, holds (page 152). In the
case of asrocholi above the reason behind the change of verb class was
acquired iterativity; in the case of doaissilbi it is acquired causativity:
doaissilbi literally means assigns to the possession of (DIL s.v. do-aissilbi
196: 42), which implies a strong element of causativity: causing someone to
possess.
3.3.1.3 Semantic motivation for formation as deponent verbs

As has been pointed out above, page 43 and cf. note 10, Old Irish
deponent verbs retain the Proto-Indo-European middle voice meaning, i.e.
they imply a stronger subject involvement. This is not just the case in the
primary verbs, but is also borne out in the secondary verbs. If a secondary
verb has a stronger subject involvement, such as labraithir speaks (page
139), it will become a deponent verb.
There are sixteen secondary deponents in Old Irish. Of these, two are
deverbal, the causative cuirethar throws (page 188) and the iterative
tuigithir covers (page 254). Four are deadjectival, aditchethar detests
(page 109), labraithir speaks (page 139), drgidir directs, aims (page
127) and comalnaithir fulfils (page 121). The rest are denominative:
airlithir advises
athmuilnethar says again
conruidethar intends
adtluichethar gives thanks
callaithir thinks about
cobraithir helps
molaithir praises
samlaithir compares
doecrathar covers
adgldathar addresses
63
Analysis
Each of these verbs has a clear motivation for being a deponent verb,
because each has a strong subject involvement. There are no exceptions.21
This implies that when these verbs were created, the morphology itself
retained the middle voice meaning to such an extent that it could be used
productively.
No derivative deponent verbs can be placed with certainty in Old
Irish and it can therefore not be claimed that the deponent morphology was
still productive in Old Irish. On the other hand, not all of these deponent
verbs can be placed with certainty before Old Irish either, and therefore it
cannot be stated with certainty that the middle voice meaning had died out
before Old Irish. After all, all other verbs have to fulfil several criteria
before greater age is considered (Brittonic cognates, verbal noun formation
etc.), while the non-igidir deponents are considered to be of greater age by
default, because they are. This is circular, especially considering that many
of these deponents, such as aditchethar detests, comalnaithir fulfils
and airlithir advises do not fulfil these other criteria.
Another point. The appearance of otherwise unexplained pres. ind.
3sg. deponent coinedar of the active verb canid laments (below, page
177) in Ml. 74b1 could be explained as an early example of the confusion of
active and deponent verbs that is so common in Middle Irish. However,
given the context (is fursundud 7 is foillsigud dund eregim 7 dund imniud
coinedar duaid insalmso this psalm is an illumination and a manifestaion of
the complaint and of the trouble that David laments), an actual middle
voice meaning to the flexion is certainly not out of place.
Note that adgldathar (page 159) is not, strictly speaking,
denominative. It is onomatopoeic and derived from the sound to imitate
babbling, gla, gla, gla. However, it does wonderfully argue the
classification scheme established in this thesis. Its inflection as AII is
21
Doecrathar always has the implication of protection (cf. DIL s.v. do-ecrathar 250: 37
and imm-tecrathar 155: 22) and although tuigithir has a wider range of meanings in Old
Irish, the original connotations of covering a house (i.e. for oneself, for ones own
protection) become clear from the nominal reflexes of the root (cf. IEW 1013-4 and OI tech
house). Moreover, meanings of the Old Irish verb include covers a house, thatches (DIL
s.v. tuigithir 358.55) and meanings of the Greek verb include protect.
64
Analysis
consistent with its transitivity and iterativity. Moreover, its subject
involvement (speaking involves the subject closely) is consistent with its
deponent flexion. See further below in the corpus, page 159.
3.4 Morphology
of
and
motivation
for
formation as -igidir verbs

The -igidir verbs are by far the most productive category within Old
Irish. Though this derivational pattern developed within Celtic (cf. Joseph
1987: 137-8 for Brittonic examples), it came into its own in Old Irish. When
a scribe wanted to translate a Latin verb for which he did not have an Old
Irish verb, the first way to do so that would come to mind would be to create
an -igidir verb. Furthermore, if he wanted to translate a participle, he would
most likely create a participle based on an -igidir verb, even if that -igidir
verb itself did not exist (cf. indemnichthe, recte indenmichthe, weakened,
page 70 above). It could be speculated that the very productivity of the
-igidir morphology made the entire process of creating new verbs more
productive than for example the creation of new adjectives.
The -igidir verbs were originally derived from nouns which were
themselves already compounded with a suffix *-sag-. This *-sag-i oelement is a reflex of the root *seh2- pursue, seek (LIV2 520, IEW 876-7)
and the nouns therefore meant [noun] seeker (Joseph 1987: 155, see also
Uhlich 2002: 417-8). The original -igidir verbs therefore will have meant is
a [noun] seeker with quite probably an instantaneous reinterpretation to
seeks [noun]. Cathaigidir fights, derived from cath battle, for example,
originally meant seeks battle. The semantic field of these verbs allowed
for some bleaching. Tessaigidir warms (< tess heat, warmth, page 250),
although originally no doubt seeks warmth would have been reinterpreted
as makes warm.
Due to the nature of the suffix, -igidir verbs would originally all
have been denominatives. However, several of the oldest ones, such as the
above mentioned tessaigidir warms and teimligidir grows dark (< teimel
65
Analysis
darkness, gloom, page 250) could easily have been reinterpreted as
deadjectival verbs due to their meaning, opening the morphology up for
deadjectival possibilities such as sligidir lowers, ridigidir levels and
sonartnaigidir strengthens. These can still be shown to be old since they
all have Brittonic cognates and are in meaning very close to the original
denominative -igidir verbs, but they have moved away from the initial
semantic range. From making, the next step was to verbs meaning
becomes [base].
The early meanings became bleached further as the morphology
gained in productivity and within Old Irish the seek meaning has
disappeared completely except in a those few verbs that are not only old
enough to have been created with that meaning, but that have also kept it.
The specialised meaning of seeking in this suffix must have disappeared
from the language before the earliest written attestations. Any verbs created
in Old Irish itself do not show this semantic field at all. In Old Irish, there is
a wide range of meanings possible for -igidir verbs and indeed this
morphology is used for those verbs that will not fit into the other categories,
since those are more restricted. As has been discussed above, there are
specific semantic restrictions on which derived verbs become AI and AII
verbs. In Old Irish, the -igidir category no longer has any restrictions and as
such becomes the default category containing such verbs as crottigidir
plays on a harp (page 186) and indraicigidir thinks fit (page 224). This
distinction can also be used in the other direction. The verb ceisid
grumbles, complains (page 180) stands in stark contrast to the other AI
and AII verbs in that it is more a response to its base noun ces debility,
inertia than having the direct relationship *is debilitated / makes
debilitated that one would expect in this category. This verb cannot have
been created within Old Irish, because in Old Irish it is the -igidir category
that caters for the indirect relationship between noun and verb. The AI and
AII category had strict boundaries for what was permissible. Prior to Old
66
Analysis
Irish, however, it was the reverse: the -igidir category was had strict
boundaries. It is therefore likely that this verb was created in that period.22
3.4.1 Derivations from nouns and adjectives in -ch

A problem for deciding on the base of an -igidir verb arises when a
verb can be derived from either a noun or from the denominative noun or
adjective in -ch. An example is fograigidir emits a sound, sounds, which
could be derived from fogur sound or from fograch noisy, resounding. If
this verb is derived from fogur, it is an -igidir verb, but if it is derived from
fograch, it could either be a straightforward deponent or an -igidir deponent,
with lost -ach- by haplology: *fograchaigidir > fograigidir. Lionel Joseph,
in his article about the origin of Celtic denominatives in *-sag- (1987)
argues that although confusion arose in the later language, in Old Irish, a
verb derived from a word ending in -ach/-ech contains that ending (with
syncope of the vowel), before the full verbal stem with its second guttural
(page 117). He gives several examples, such as grschaigidir makes
continuous (page 219), derived from grsach lasting, continuous and
mindchigidir begs (page 231), derived from mindech poor. Whereas this
certainly is a valid argument, an opposing case can be made. My corpus is
restricted to the Wrzburg and Milan Glosses, which, although it could be
argued that the Milan Glosses contain Middle Irish elements (cf. McCone
1985), are Old Irish documents. Any verb found in these documents can be
safely considered to be an Old Irish verb. Therefore, cases such as
ardraigidir appears (page 169 below) from airdrech face cannot be seen
as a later confusion. Nor can it be derived from an Old Irish *ardVr,
because there was no such word. Similar cases can be made for dtnaigidir
gnashes (page 195) and cngaidigir reviles, slanders (page 203), where
the only attested possible base is a derivative in -ach: a non-derived base
*dtne or cVnd is not attested.
Considering that most -igidir verbs are young, if Old Irish has only
one of the possible bases attested, it is likely that the verb is derived from
22
It will be discussed in the corpus below, page 180, that there are more reasons for ceisid
to become an AII verb specifically, to do with its iterativity.
67
Analysis
that, rather than a non-attested noun or adjective. Therefore, the verb
ardraigidir appears, must be derived from airdrech face, front.
Consequently, it is not possible to argue the case as straightforwardly as
Joseph has done. There are cases in which a verb may theoretically be
derived from a derived base, with loss of -ch-. This leads me to conclude
that it is in most cases ultimately impossible to state with certainty from
which base a given -igidir verb has been derived. For that reason, where
both possible bases are attested, the principle of economy has been applied,
and consequently, Josephs theory has been followed. Deriving a verb from
a derived base in -ach is more involved than deriving it directly from the
non-derived base. First of all, when deriving from a base in -ach, the
resulting verb becomes unnecessarily complex (*fograchaigidir above is
one of the simpler ones) and secondly, it would require haplology to have
taken place. Making a non-igidir deponent verb from the base in -ach would
cause confusion with the -igidir verbs, because it would look exactly like
those. Deriving from the non-derived base, on the other hand, only requires
the affixing of the suffix. Verbs have therefore been deemed to have been
derived from the non-derived base, unless there is clear evidence for the
contrary.
3.4.2 The appearence of unexplained -n- between base and

suffix
There are seven -igidir verbs in the corpus in which an -n- can
appear between the base and the suffix. In four of them, tairisnigidir trusts
in (page 249), airmitnigidir honours (page 166), ainmnigidir names
(page 163) and toltanaigidir pleases (page 252) this -n- is part of the stem,
showing that the verb was formed from the stem rather than from the
nominative singular, as noted by Joseph (1987: 113-4). This leaves
brthnaigidir, crithnaigidir and sonartnaigidir to explain. The bases of
these three verbs are vocalic stems. In sonartnaigidir the -n- is perhaps due
to dittology: if the word is repeated often enough, the -n- will creep in. For
the other two verbs, and perhaps for sonartnaigidir as well, although
68
Analysis
arguing analogy on the basis of only four other verbs is fraught with
difficulties, I would like to make a case for it even so. The analogical
pressure is arguably very weak, but three arguments can be made in favour
of analogy. First of all, brthnaigidir exists in a context of several other
verbs to do with judgement. Both brithemnaigidir and brethnaigidir show
the same -n-. In brithemnaigidir this is due to the base noun brithem (DIL
s.v. breithem 174: 69), which will have added to the analogical pressure for
both brethnaigidir and brthnaigidir. Moreover, the shape of brth and
crith is similar: both have initial occlusive, followed by -r-, a vowel, and
both end in -th. Lastly, it can be seen in the later language that the -n- spread
subsequently. OI cestaigidir is attested with -n- in the later language, as is
bertaigidir (DIL s.v. cestnaigid 150: 23 and bertaigidir 86: 85; for further
examples see Joseph 1987: 114).
3.4.3 Calques on Latin
The -igidir verbs are occasionally used for calques on Latin verbs.
This thesis deals first and foremost with the morphology of secondary verbs.
In matters of dating, however, the issue of calques has to be taken into
account. If a verb can be shown to be a calque on Latin, it is unlikely to be a
formation older than the fifth century AD (cf. Chronology, section 3.6 and
3.7, pages 75ff. below). Due to the nature of the thesis and the material,
however, a discussion on calques necessarily has to be brief and limited. In
suggesting which verbs could be calques on Latin, I have therefore
restricted myself to discussing only those verbs which occur infrequently in
the language and in deciding whether or not something is a calque, I have
looked only at what the verb glosses in the Wrzburg and Milan Glosses.
Plenty of suggestions could be made for Old Irish verbs calquing Latin
verbs simply because they so happen to be formed of the same elements, but
it is only where we see the verbs side by side in the two languages that we
can say with any degree of certainty that calquing might have occurred.
Calque is defined as follows, for instance:
69
Analysis
A term used in comparative and historical linguistics to refer to a type
of borrowing, where the morphemic constituents of the borrowed word
or phrase are translated item by item into equivalent morphemes in the
new language. (Crystal 1991 s.v. calque)
A remarkable example of calquing is found in Ml. 50a3. The Latin

debilitatum weakened is glossed with the OI indemnichthe (recte
indenmichthe) weakened. This indenmichthe is hapax legomenon.
Moreover, it is the participle of a verb *indenmigidir that is not attested, and
that quite likely did not even exist, because it is not attested at any stage in
the later language either. The Lat. debilitatum is also a participle, but in the
case of Latin, the verb underlying the participle, debilitare weaken, disable
is extant and widely attested.
The Old Irish adjective indenim weak is attested. A verb formed
from indenim would be *indenmigidir weakens (page 223), which, as said,
is not attested. The participle of that verb would be indenmichthe. The
formation of the participle followed the standard paths, but passing over,
either consciously or unconsciously, the verbal step. Indeed, one might
speculate that the creator of this participle did not even consciously think
about the verbal step, but went straight from the adjective to the participle.
The participle may have been all there ever was; a fully paradigmatic verb
may never have existed.
The exact origin of this participle is therefore a matter of
speculation. The forming of new verbs in Old Irish in this way was very
productive, so it will have been an integral part of the language, not
something a scribe had to consciously think about. He could just have seen
the Latin and translated it into Old Irish, making up a new participle in the
process.
On the other hand, this verbal particle is very clearly modelled on
the Latin, which implies that the scribe might, in fact, have consciously tried
to create a word that not only translated, but also modelled the Latin.
Calques are by their very nature hard to determine decisively, since
it is never quite certain whether a word or phrase was coined independently
or copied from another language. If a word in two languages contains the
same elements, one is not automatically a calque on the other. The situation
70
Analysis
with Old Irish and Latin has its own particular difficulties. First of all, the
Old Irish material that has come down to us is incomplete. Secondly, the
earliest written material is overwhelmingly religious, skewing our
perception of the language: a verb may seem to only occur in religious
material because it had a much greater chance to get written down in
religious material. Consequently, it may seem like a calque on Latin, purely
due to its accidental religious context.
The list below is therefore neither exhaustive nor definitive and most
certainly subjective. Calquing cannot be shown decisively, because there is
always a possibility that speakers of two languages came up with the same
idea independently. If the influence of Latin on the formation of the verb is
not clearly evident (if the verb does not have, for example, a clearly
Christian meaning, or a technical or metaphorical one), I have not
considered the verb to be a calque. Arsadaigidir grows old (page 169), for
example, is hapax legomenon and could be seen as a calque on Lat.
ueterescere age, which it glosses. However, the derivation becomes
[adjective] is direct and concrete. Old Irish has several of these verbs and it
did not need to look at Latin for an example. Calquing can be more easily
argued in cases such as ingoistigedar ensnares (page 219), which could
have been formed on the model of Latin illaqueare ensnare. Both verbs
are derived from a noun meaning snare and compounded with the preverb
in-. When a verb has more elements that can be matched, like
ingoistigedar, it is also easier to recognise or dismiss a possible calque.
However, since derivation with prefixes lies outside the scope of this thesis
(cf. section 1.1.3, page 20), compounds have been left aside, unless a
simplex is not attested.
The list below is based first and foremost on the definition given
above. If an Old Irish verb does not match the definition, it has not been
considered a calque. In some cases, an Old Irish verb will consistently gloss
a Latin verb to which it is similar, or of which it is an exact translation. This
in itself does not make it a calque if the individual elements that make up
the verb, the morphemic constituents, do not match. Secondly, as mentioned
above, the frequency with which the verb is attested and the variety of
71
Analysis
sources in which it occurs has been taken into account, although it must be
noted that as a this consequence of this, some calques may have been
missed. Thirdly, the apparent ease with which the verb could have been
formed independently has been taken into account to dismiss possible
calques. Lastly, the context of the attestations of the verb in the Wrzburg
and the Milan glosses has been taken into account. If the verb consistently
glosses a Latin verb on which it could have been calqued, it has been
accepted as a possible calque. Details on each formation can be found in the
corpus, to which reference has been given. Several verbs that are clearly not
calques will be mentioned subsequently, for comparative purposes.
addanaigedar (page 192): << remunerare reward (later non-deponent
usage of the deponent verb remunerari).
ailigidir changes (page 162): VN ailigud << alternatio alternation.
aininnracaigidir regards as unworthy (page 163): << indignari deem
unworthy.
asmecnaigedar roots out (page 171): << eradicare root out.
cairdinigidir unites in friendship (page 178): << foederare establish by
treaty.
cairdigidir makes a treaty (page 178): << foederare establish by treaty.
comaicsigidir approaches, comes near to (page 183): << adpropriare
appropriate.
conasarcaigedar delights in, is delighted with (page 242, the simplex
sercaigidir is only attested later) << condelectari delight in.
*crottigidir plays on a harp (page 186): << citharizare play on a cithara.
dochenlaigidir degenerates (page 200): << degenerare be unworthy of.
dumaigidir piles up, exaggerates (page 202): << exaggerare heap up.
nirtigidir weakens, enfeebles (page 205): << eneruare weaken.
erlissaigidir loathes (page 206): << fastidire disdain.
feochraigidir grows fierce, wild (page 209): << efferare make wild
foammmaigedar subjugates (page 210): << subiugare subjugate.
goistigidir puts a noose around, hangs (page 219): << laqueare hang,
(also compound ingoistigedar ensnare << illaqueare ensnare).
72
Analysis
grschaigidir makes continuous (page 219): << continuare make
continuous.
ilaigidir increases, multiplies (page 220): multiplicare multiply.
*ilchigidir howls, screams (page 220): VN ilchugud << ululatus shriek.
indenmigidir weakens (page 223):
part. indemnichthe << debilitatum
weakened (see also above).

ingraigidir renders impious, pollutes (page 224): impiare renders
impious.
inilligidir protects, defends (page 224): tutare guard.
inlnaigidir ensnares, entangles (page 225): << irretire entangle.
lochtaigidir finds fault with (page 226): criminare accuse.
mindchigidir begs, is a mendicant (page 231): << emendicare begs.
ollaigidir makes larger, expands (page 234): << ampliare enlarge.
suthchaigidir makes fruitful (page 249): << foetare breed.
adfaligedar unveals, reveals (page 255): << reuelare show.
Boaigidir makes alive, vivifies (page 173) is a widely attested
verb that has been placed by Joseph (1987: 153-4) in the earliest phase of
the *-sag- derivatives (cf. Chronology, sections 3.6 and 3.7, pages 75ff.).
This is however problematic, since bo is never attested as a noun, but
always as an adjective. Moreover, it could be a calque on Latin uiuificare,
which it glosses in both Wrzburg attestations.
Examples of verbs which could be influenced by the Latin, but are not
calques:
ailithrigidir goes into exile (page 162): the Old Irish is derived from the
pronominal aile and tr, whereas Latin exulare live in exileis formed from
the preverb ex out of and solum bottom, ground.
Although OI deugaigidir drinks (page 195) and Lat. potare are formed in
the same way (both derived from a noun meaning drink, Lat. potus), I
hesitate to see the Old Irish as a calque on the Latin. Deriving a verb
meaning drinks from a noun meaning drink is so straightforward that it
would not have required a Latin example.
73
Analysis
airgalaigidir (page 164): The only form attested of this verb is the dative
plural participle ergalaichthib. The Lat. propugnaculis which it glosses is
taken by the glossator as pro pugnaculis. Pugnaculis means fortress and is
not verbal at all.
doirmmaigidir becomes worthless, vile (page 201): The verb is formed
from dormm servile yoke, rather than dor, which would have translated
Latin uilis from which uilescere become worthless is formed.
foniti mocks, derides (page 214): The Old Irish derivation of this verb is
not known, which makes any suggestion of a calque meaningless.
frisdorchaigedar darkens, obscures (page 201): Latin obtenebrare
darken is derived from a noun, not an adjective.
gaimigidir passes the winter (page 217): Although this verb is hapax
legomenon and despite the existence of Latin hibernare, I hesitate to see this
as a calque, since for the concept of wintering, speakers would hardly
require the example of Latin to form a verb like this.
Out of 187 -igidir verbs, I have therefore counted 28 (15%) possible calques
on Latin verbs. This number is tentative and based on a preliminary study
only. Further research is needed to investigate the extent of Latin influence
on verb formation and semantics in Old Irish.
3.5 Simple vs. compound verbs

Old Irish derived both simple and compound verbs from nominal
compounds and both simple and compound verbs from uncompounded
nouns and adjectives. Thus, when deriving a verb from a compounded noun,
the resulting verb could be interpreted as preverb + stem and be treated as a
deuterotonic verb form or it could be interpreted as a simple verb and be
treated as an absolute verb form. The verb comdemnigidir rules (page
184), for example, is derived from comdemin and is treated as a simple verb.
Even in the verb, the preposition is always com-, rather than con- which it
would be if this were a compound verb. The verb focridigedar girds (page
210), on the other hand, is derived from a formation *okridi o, not attested
74
Analysis
in Old Irish. In Old Irish, this verb is clearly a compound verb. The pres.
ind. 3sg. focridigedar Ml. 35c32 has a conjunct ending, as does the pass.
3pl. focridigter Ml. 135b1. Moreover, the infixed pronoun in Ml. 101c3
fotchridigther is infixed after the preverb, rather than using the particle no.
Consecra (page 123), borrowed from Latin consecrare, is also always
treated as a compound verb. The fut. 1sg. cosacrub Ml. 45b12 is
deuterotonic and the 3pl. rel. secrabat Ml. 66a6 would have had a relative
ending if this verb were treated as a simplex.
When deriving a verb from an uncompounded noun, on the other
hand, a preverb may be added in the process of derivation, omitting the step
of forming a simple verb. There is thus no simplex of attobi adheres to
(page 172), derived from tob. The pres. ind. 3sg. rel. form atoibi Ml. 57d18
would have had a relative ending if this verb were treated as a simplex. The
impf. 1pl. / past subj. 1pl. (at)toibmis Ml. 18c4 would have been preceded
by no if this were a simplex. The forms are deuterotonic. This verb is clearly
treated as a compound verb in the Glosses.
Adamraigedar wonders at (page 157), derived from adamrae very
wonderful has both simplex and compound forms in the Glosses. pres. ind.
1sg. rel. noadamrugur Wb. 16c3 is a simplex form (preceded by no-),
whereas pres. subj. 2sg. adnamraigetar Ml. 39b10 is a compound form
(nasalisation infixed between preverb and stem). As is perf. 3pl.
adruamraigset Ml. 88a18 (ro infixed between preverb and stem).
In my research, I have unfortunately not found a reason for these
alternations.
3.6 Arguments for age

I have dated the verbs in my corpus based on a number of criteria
which I will set out below. These criteria are used in combination with each
other and their weight will differ from case to case depending on the
conjunction of criteria. For example, I will generally take Brittonic cognates
as a good indicator for greater age. If an Old Irish verb shows highly
specific similarities in morphology and semantics to a verb in one or more
75
Analysis
of the Brittonic languages, this verb could have been formed in a common
phase of the language. However, I have taken account of the fact that
especially the -igidir category is very productive in both Old Irish and the
Brittonic languages. An -igidir verb with a Brittonic cognate is therefore not
by definition old. Conversely, verbs which show a clearly Christian context,
such as mraid makes great (page 144) are unlikely to have been formed
prior to the coming of Christianity to Ireland and can be dated to the Old
Irish period.
The existence of synonyms can be taken into account. If a verb is the
only verb for a given concept within the language, it is more likely that it is
old. It can of course happen that a new word replaces an old. But the
question of the existence of synonyms may nevertheless be taken as a
signifcant criterion in conjunction with the others.
Hypotheses can also be formed from the basis of these verbs verbal
noun formation. If the verbal noun is created using an obsolete formation, or
if the base noun is used as verbal noun, the verb is more likely to be old.
An example is the Old Irish verb rmid counts, which has an exact
cognate in MW kyfrif to count. In both languages, this is the only verb
used for this concept. Moreover, it uses its base noun as verbal noun, rather
than a verbal noun formed with a productive suffix. That, and the lack of
synonyms, make it likely that it is old. On the other hand, the Old Irish verb
bathaid feeds and the Welsh bwydo have several synonyms in both
languages (OI mthaid, DIL 119: 27, ssaid, DIL 63: 68, W llithyaw, Isaac
1996: 340). They also form their verbal noun following productive patterns.
The hypothesis is therefore that these verbs have been independently coined
within each language.
The semantic field of the verbs should also be taken into account.
The OI delbaid shapes (page 126) and the W delwi, for example, are
formally exact cognates, but the semantics of the Welsh verb especially
show that these were independently coined. The Welsh noun delw that the
verb is derived from has developed a specific meaning statue. The verb
has been derived from that particular meaning. It means stop, freeze,
become motionless, i.e. become like a statue. The Old Irish verb, on the
76
Analysis
other hand, has the wider meanings shapes, forms that correspond to the
wider meanings of the noun delb shape, form that this is derived from.
The -igidir verbs have been discussed above, page 65, and a relative
chronology can be proposed. If a verb has a Brittonic cognate, is
denominative and preserves the original seek-meaning, it is likely to be
old. If it has a makes meaning, it is slightly less old. If it is deadjectival,
but could have been derived from the nominal meaning it is likely to be
again slightly less old, but still old. True deadjectivals with meaning makes
[base] (< makes [base]-ness) are still inherited, as are those with meaning
becomes [base]. Further bleaching seems to have begun before Old Irish,
but only within Old Irish itself did this morphology become so productive
that the original meaning of the morphology is no longer recognisable in the
resulting verbs.
The causative / iterative morphology was still productive within
Celtic, but no longer within Old Irish. It is not clear when it ceased to be
productive, but any morphologically causative or iterative verb predates Old
Irish. These verbs have not been placed in the Chronology, since with the
exception of immfolngi causes, produces (page 221) and loscaid burns,
consumes by fire (page 227), which must have been formed after ProtoIndo-European, they cannot be dated any more specifically.
The dating criteria given are suggestions and further research will
enable us to add to and refine these criteria. None of these arguments are
final, none give a definitive answer. But each of them supports the
understanding of the formation of verbs and each of them supports a
hypothesis about the age of these verbs. Each of these arguments gives more
insight into the past, opens a window of possibility for further research.
3.7 Chronology
A hypothesis of age can be given for those verbs to which the above
arguments apply. All dating criteria have been taken into account in
combination, that is to say, an -igidir verb with clearly Christian meaning
will not have been dated to an early stage of the language even if it can be
77
Analysis
interpreted with a seek meaning or has Brittonic cognates. In the list
below, verbs are presented for which arguments for and/or against
derivation in a specific phase of the (proto-) language can be given,
followed by their first meaning as given in the corpus in Chapter 5, a page
reference to the corpus and a short reference to the argumentation for age, as
follows:
Brittonic cognate: As has been discussed above, a case can be made
that a verb with a Brittonic cognate is an inherited formation. Verbs with
this description have a cognate in at least one of the Brittonic languages.
Cognates are listed in the corpus in Chapter 5 and the page reference will
direct the reader to the correct verb.
Uses base as VN: As discussed above, if a verb uses its base as VN,
it is likely to be of greater age.
-igidir verbs: As has been discussed above, pages 65ff.,
denominative -igidir verbs that retain their seek meaning are the oldest,
deadjectival -igidir verbs that fall within the same shade of meaning are
slightly younger and the meaning becomes bleached in time.
Synonyms and degree to which the denoted concept is basic: As has
been discussed above, a verb that has no synonyms or denotes a very basic
concept is more likely to be old, because otherwise this concept would have
needed expression in the proto-language too. Therefore mention will be
made of these issues.
Christian concepts were of course only introduced with the coming
of Christianity and those verbs therefore have to be Old Irish.
Verbs which are hapax legomenon or only occur twice or three
times, such as colaigidir makes thin (page 183), have not been listed,
since it is not clear to what extent these were actually part of the lexicon.
The exception for this is durnid. Argumentation can be found in the corpus,
page 202.
Consequently, all old verbs will be listed, since there must be

evidence to argue their age, as discussed above. Verbs derived from bases
that are no longer extant in the language will be listed, since they must have
been derived in a phase of the language when the base was still extant. On
78
Analysis
the other hand, Christian concepts will be listed, since these will not have
been derived before the coming of Christianity. Several verbs will be
discussed more extensively below, but for most the discussion can be found
in the corpus.
The list is divided into Pre-Celtic (one verb only), Common Celtic
(one verb only), unspecified Celtic (the bulk of old verbs), pre-Old Irish
(three verbs that are definitely older than Old Irish, but cannot be dated back
further with any certainty) and Old Irish (verbs for which specific arguments
for younger age can be given). The unspecified Celtic category is
subdivided into AI and AII verbs on the one hand and -igidir verbs on the
other. This says nothing about the age of these categories as a whole, but for
the -igidir verbs a further tentative relative chronology can be proposed and
for the AI and AII verbs that is not possible. The subdivision exists only to
avoid confusion.
It must here be noted that there are no Celtic cognates attested
outside the Brittonic languages. This is not inexplicable: verbs are by
definition rare in the material available for the Continental Celtic languages.
Secondary verbs will however have been created at any point between
Common Celtic and Old Irish (and earlier and later too), because the
morphologies have been productive throughout this time frame. Lack of
Continental cognates, however, makes it generally impossible to date verbs
with any degree of certainty to early Celtic or Common Celtic
(causatives/iteratives excluded). It is however undesirable to date all preOld Irish creations to a time frame that some scholars have called Insular
Celtic (see above, section 1.2.1.1, pages 23ff. for my position on Insular
Celtic), since that would not reflect the reality that these verbs were created
throughout time and not just in Insular Celtic. I have therefore dated any
verb for which I have no explicit reason to give it a more specific dating to
an unspecified Celtic period. These verbs are not all necessarily very
ancient. Some of them will be, but some of them will be relatively young.
Further research may be able to date these verbs more precisely. The
Chronology is of course tentative and subject to revision after further
research.
79
Analysis
Pre-Celtic:
adtluichedar gives thanks (page 160): Shared nominal vocabulary with
Slavic and Italic that has gotten lost so early that there is no trace in Celtic
and even the denominative in Latin is entirely different.
Common Celtic:
callaithir thinks about (page 119): Brittonic cognate, uses base for VN,
very basic concept.
Celtic - unspecified (AI, AII)

adgldathar addresses (page 159): see below.
laid requests (page 167): Brittonic cognate, uses its base as VN.
asscarta drives (page 112): Brittonic cognate, base not attested in Old
Irish.
berbaid boils (page 113): Brittonic cognate, base not attested in Old Irish.
bertaid brandishes (page 114): Brittonic cognate, shades of meanings of
base matched by shades of meanings of verb (see below), base serves as
VN.
crothaid shakes (page 124): see below.
dlaid portions out (page 191): Brittonic cognate, uses its base noun as
VN.
durnid smites (page 202): Brittonic cognate, basic concept.
forcenna puts an end to (page 135): Brittonic cognate, uses base noun as
VN.
nertaid strengthens (page 144): Brittonic cognate, no synonyms in either
language, meanings congruent.
rmid counts (page 235): Brittonic cognate, no synonyms in either
language, uses its base noun as VN.
80
Analysis
Celtic - unspecified (-igidir, since the -igidir verbs can be placed in a more
precise relative chronology, the order below is deliberately nonalphabetical)
cathaigidir fights (page 179): Brittonic cognate, seeks battle,
denominative.
sraigidir violates (page 239): Brittonic cognate, seeks outrage,
denominative.
slaigidir sows (page 243): Brittonic cognate, seeks seed, denominative.
tessaigidir warms (page 250): Brittonic cognate, makes warm,
denominative.
follaigidir neglects23 (page 213): Brittonic cognate, makes neglect
denominative
sligidir lowers (page 225): Brittonic cognate, makes low, deadjectival.
ridigidir levels (page 235): Brittonic cognate, makes smooth,
deadjectival.
sonartnaigidir strengthens (page 246): Brittonic cognate, makes strong,
deadjectival.
gnthaigidir is accustomed to (page 218): Brittonic cognate, makes
usual, deadjectival.
cuimnigidir remembers (page 188): Brittonic cognate, denominative or
deadjectival,
in either case likely to be an early example of bleached
meaning.24
Pre-Old Irish
dodona consoles (page 131): see below.
dnaid closes (page 132): goes back to the earlier meaning of dn,
enclosure.
todid shines (page 252): derived from a base which is no longer extant in
Old Irish.
Old Irish
23
See corpus for an alternative, Old Irish dating.

The verb is derived from cuman remembered, memorable or from cuimne faculty of
memory, remembrance and can therefore not be taken as an earlier seeks memory.
24
81
Analysis
aithirgigidir repents (page 166): Christian concept.
baitsid baptises (page 173): derived from a Latin loanword (baithis
baptism, from Latin baptisma), and a Christian concept.
bathaid feeds (page 114): Brittonic cognate, but several synonyms in
both languages and verbs use productive VN suffix.
cestaigidir inquires (page 181): derived from a Latin loanword and a
rather technical concept.
cinnid defines (page 181): Welsh cognate, but meaning influenced by
Latin.
delbaid shapes (page 126): see above, page 76.
dchsnigidir exists (page 196): synchronic syncope: see below.
rannaid divides (page 148): Brittonic cognate, but several cognates within
OI and VN formation follows productive pattern.
In the case of bertaid brandishes both in Welsh and in Old Irish the
meanings of the verbs correspond closely to the meanings of the respective
base nouns and these shades of meanings are so similar that a common
origin can be postulated.
The verb crothaid shakes comes from a root that was present in
Celtic (cf. Gallo-romance *crottiare and OI crith act of shaking,
trembling). The noun *krotos from which the verb has been derived,
disappeared from the language before the Old Irish period, but the verb has
been preserved. This means the verb itself must has been created before the
Old Irish period.
Adgldathar has no Brittonic cognates, but it has resisted the first
palatalisation that most AII verbs would have undergone: the of the Celtic
verb *gld--tor would have palatalised the consonant immediately
preceding it, if that consonant itself had not been preceded by long (cf.
McCone 1996: 116). Adgldathar has however undergone third
palatalisation (cf. McCone 1996: 119). This is diachronically entirely
regular, but synchronically irregular. If this verb had been created within
Old Irish, it would have been synchronically regular instead.
82
Analysis
Dodona is derived from *diden reinterpretation of di in *diden as a
preverb. Di- could only have been interpreted as a preverb before falling
together of the preverbs di and do in deuterotonic position.
3.8 Primary verbs taking on weak flexion

Those primary verbs that have taken on weak flexion can be found
in appendix 1 below, page 259. The handbooks have been followed silently
(details and references can be found per verb in appendix A) except where
there is significant unsolved controversy or where clear evidence to the
contrary can be given. Those verbs are discussed here below.
Pedersen (ii 526) gives adferta turns (page 261) as e-grade of the
root that gives Old Irish dofortai pours (page 200). This is tentatively
followed by IEW (1150-1) and LIV2 (691-2). Stokes and Strachan (1901)
emend instead to adfertaigedar, but there is no evidence for this. The verb
adfertaigedar would be a borrowing from Lat. aduersari be against,
oppose. A borrowing from Latin is not unusual, but this attestation would
require
the
scribe
to
have
changed
Latin
-s-
to
Old
Irish
-t-, for which there is no parallel, to have borrowed as an -igidir verb, for
which there is no parallel and to either not have written the full verb or to
have borrowed a deponent verb as an active verb. This is not impossible, but
it is unlikely.
One attestation is a very slim basis to go on. However, I would argue
that this is a primary verb that has taken on weak flexion, following
Pedersen, IEW and LIV2. The Proto-Indo-European root ert- means turn
around, which matches the meaning of the Latin aduersari this verb
glosses. It also matches the meaning of the causative of this verb, dofortai
pours (see below, page 200), i.e. causing liquid to turn around. It is not
the first or only verb to take on weak flexion.
Consecha corrects, reproves (page 271, DIL s.v. consecha 460:
29), doduschi awakens, brings back to life (DIL s.v. doduschi 241: 81)
83
Analysis
and dofaisgi announces (DIL s.v. dofsaig 263: 48) are compounds of the
primary verb sechid asserts, declares (DIL s.v. sichid 214: 20) from root
*sek- say (LIV2 526-7). Sechid has been discussed as a weak i-verb (GOI
768), only compounded (Ped. ii 619) and already taking on weak flection
in Old Irish, together with its compounds (Schumacher 2004: 265-8). These
statements are all incorrect. There is a simplex sechid which can be found in
DIL s.v. sichid 214: 20. Of the Old Irish forms given, nad sich (214: 22)
seems a preterite 3sg. rather than a present 3sg., but is in either case
strong25, the present 3pl. no seghait (ibid.) cannot be a weak i-verb form and
pret 3sg. sichi[s] (214: 24) is emended by Thurneysen based on his
interpretation of this verb as an i-verb and can therefore not be taken as
evidence. The form given by Thurneysen in GOI is a misreading from Sanas
Cormaic and can therefore not be taken as evidence.26 The other forms
given are either late or non-diagnostic. It is clear that in Old Irish, sechid
still inflects as a strong verb of the BI class. Note that the DIL citation form
seems based on sich as a present form. The correct present stem would be
sechid.
Whereas sechid still inflects as a strong verb, consecha has taken on
weak a-flexion and doduschi awakens, brings back to life weak i-flexion.
Dofaisgi has been given the strong citation form dofsaig by DIL (263:
48). The forms registered seem to belong to two different forms, one a
compound of sechid, one a derivation from the VN tsc (DIL s.v. tsc 87:
62). This is not unusual. In fact, the same thing has happened with the verb
consecha, from the same root. The VN of consecha is cosc, from which a
verb coiscid has been derived, which already in Old Irish has been confused
with consecha. The entry for consecha in DIL reflects this.
Taking out the forms with -c- under dofsaig leaves two forms, of
which tafaisig (263: 53) is from the long version of Tochmarc mire, which
is not Old Irish. This leaves the Wb. 7d11 form dofarsiged. This is an
25
I thank Prof. Liam Breatnach for pointing out to me, at the Tionl of 2008, that it has
been concluded before that sechid has a long preterite. See Carney 1958: 14, 35, Binchy
1960: 81, Watkins 1963a: 224, Hamp 1975: 173-4 and Breatnach 1986: 191-2.
26
Thurneysen reads roseched, a weak perfect passive singular. However, the actual form is
fo-so-sech, a folk-etymological interpretation of fsach that does not even make sense.
Arguments about the inflection of the verb cannot be based on this form.
84
Analysis
s-preterite. Although one verb form is too little evidence to be certain, it
certainly does not point into the direction of the reconstructed third singular
citation form dofsaig of DIL and Kavanaghs Lexicon, which assumes it is
strong. While DIL and Kavanagh are not necessarily wrong in positing this
citation form, based on the attestation there is no certainty that the citation
form should be dofsaig. It could have been dofaisgi. In Appendix 1,
where this verb is found, I have given it a weak citation form for the sake of
consistency. The other compounds of this verb have taken on weak
inflection. It is therefore likely that this compound did too, especially since
the one attested form is weak.
Pedersen (ii 613) sees doinscanna begins (page 266, DIL s.v. doinscanna 319: 39) as originally a nasal present *scan-n-. LEIA (S-32)
suggests a causative of root *skend- jump, but this verb is inflected as an
AI verb. A causative would have inflected as an Old Irish AII verb.
Whatever the precise origin of the verb, there are no nouns or adjectives or
even a hint of anything that it could possibly be derived from. I have
tentatively classified it as primary.
85
Analysis
3.9 Semantics of derivative verbs

3.9.1 Introduction
The possibilities of creating a verb out of a noun or adjective are
constrained by the semantic field of the noun. It is not possible make any
verb out of any noun. Taking the noun tree will not lead to a verb sings
in the first instance. The noun and the verb have to be connected to each
other in some way for the verb to have meaning, a meaning that makes
sense and can be understood by other speakers. One of the aims of this study
is to find out how far the semantic fields of the nouns and adjectives can be
stretched in the process of creating the new verb. What are the limitations?
Which ways of creating new verbs are so semantically and morphologically
common that the speaker no longer thinks about what he or she is doing, and
which verbs show true innovation and creativity on the part of the speaker?
As has already been discussed above, page 66, in the -igidir verbs,
there is a lot of diversity in meanings and possibilities, especially within Old
Irish. There are many verbs that have stretched the semantic field of their
base nouns or adjectives. In the earlier language and among the AI and AII
verbs, on the other hand, the possibilities appear to be more restricted. These
categories are therefore not exhaustive: not each verb will be found in a
category. After all, the categories below serve to classify verbs that can be
grouped together based on how they relate to their base. Not every verb can
be grouped together with other verbs. Conversely, some verbs relate to their
base in several ways and will therefore be found in several categories. The
current section only deals with semantics and not with the morphological
correlates of semantics as found in Proto-Indo-European.
Below, each category will be discussed with examples. Other verbs

in that category follow in a list. Their first meaning (following DIL, unless
stated otherwise in the corpus) will be given, unless the reason for grouping
in a specific category is due to another particular meaning and not
86
Analysis
immediately clear. In that case, several meanings will be given. For further
details of these verbs, I refer to the corpus.
3.9.2 Semantic categories
3.9.2.1 Causative
In my thesis, there are two categories of causatives. The first is
morphological: these are the verbs that have the Proto-Indo-European
causative/iterative morphology. These are mostly active verbs, such as
bidid submerges (page 173; PIE *goh2dh-i e-, LIV2 206) and dofortai
pours (page 200; PIE *ort-, LIV2 691). The morphology also occurs in
two deponents, namely cuirethar throws, moves (iterative, page 188, PIE
korH-i e-, LIV2 353-4) and tuigithir covers (page 254; PIE *(s)teg-, LIV2
589, Schulze-Thulin 2001: 133). The morphological causatives have been
discussed above, page 52.
Aside from using the term to describe a morphological category, it is
helpful to use the term in its semantic meaning only and find that Old Irish
has several denominative verbs that express causing someone to do a verbal
action and could therefore, semantically, be termed causatives. Danaigidir
(page 195), for example, means hurries, but also causes to hasten, the
latter meaning being causative, because it causes the object to hasten.
Another example is is crithnaigidir which means both trembles and
shakes, i.e. causes to tremble. Note that these verbs have both noncausative and causative meaning in the same verb. The other causative verbs
that dont use causative morphology are the non-igidir verbs discussed in
section 3.3.1.1, pages 55ff. These are:
airid guards
airlithir advises
laid requests
baitsid baptises
brissid breaks
badraid disturbs
clannaid plants
coillid damages
conruidethar intends
conterchomraic(i) collects
87
Analysis
crdaid torments
crothaid shakes
cuicsedar taxes
dilid portions out
dosc(u)lai finds out
emnaid makes double
malartaid spoils
scalid bursts
todid shines
There are plenty of verbs that mean causing a state, or causing
someone to have something or become something, but those will be
discussed individually below.
3.9.2.2 Iterative
Semantically, an iterative is a verb that expresses doing something
repeatedly. The Old Irish verb gnthaigidir is the ultimate semantic
iterative. Its very meaning, is accustomed to, practises, frequents implies
doing something over and over again. Another semantic iterative is
crithnaigidir shakes, trembles, which is also semantically causative. The
other iterative verbs that dont use iterative morphology are the non-igidir
verbs discussed in section 3.3.1.1, pages 55ff. These are:
airid guards
airlithir advises
athmuilnethar says again
attobi adheres to
bgaid declares
bathaid feeds
brissid breaks
badraid disturbs
canid laments
caithid consumes
ceisid complains
cinnid defines
coillid damages
crdaid torments
crothaid shakes
dilid portions out
dosc(u)lai finds out
durnid smites
forcomai keeps
gledid cries out
88
Analysis
malartaid spoils
onaid fasts
rmid counts
ssaid satisfies
scalid bursts
sroiglid scourges
trebaid cultivates
3.9.2.3 Fientive
Although Old Irish has no descendants of the Proto-Indo-European
morphological fientive (zero-grade unaccented root with suffix *-h1/h1-) as
given in LIV2 (p. 25), it is nevertheless helpful to use the term in its
semantic meaning only and find that Old Irish does have several verbs that
express become something and are therefore, semantically, fientives. In
the case of derived verbs, they express the becoming of the adjective (or in a
few cases the noun) that they are derived from. Examples are dorchaigidir
grows dark (page 201), deadj. < dorchae dark, gloomy or bocaid
softens (trans. / intrans.), deadj. < boc soft, gentle. In the latter case, the
verb is not just fientive, but also falls into other categories. It is not just
intransitive, but also transitive.
Other examples of fientives:

arsadaigidir grows old
scthaigidir becomes tired
celaigidir makes thin
atigidir becomes few
feochraigidir grows fierce
lobraigidir is sick, becomes sick
londaigidir is angry, becomes sick
mincigidir becomes frequent
doirmmaigidir becomes worthless
alchaigidir is proud, becomes proud
fergaigidir grows angry
teimligidir grows dark
seccaid stiffens
imdaigidir abounds? In meaning increases, i.e. becomes more?
3.9.2.4 Inchoative
The difference between the fientive and the inchoative is that
whereas the fientive focusses on becoming something, the inchoative
89
Analysis
focusses on beginning a process. The verb ailithrigidir goes into exile
(page 162), for example, expresses the beginning of the going rather than a
becoming. This is however the only example in my corpus.
3.9.2.5 Essive
The morphological essives have been discussed above, page 54.
Again it is helpful to use the term in its semantic meaning. The original
Proto-Indo-European essives implied permanent states. Semantically, only
very few secondary verbs could be called permanent essives. The inherent
meaning of verbs that could be described as essives is always wider than a
purely essive meaning, often incorporating temporary meanings within the
same semantic field. Cumachtaigidir (page 190), for example, means
rules, but there also is an attestation cia dia cumachtachtaigther (sic) for
whom thou art powerful (Sg. 209b30). Considering that the thou being
addressed here is God, it seems likely that his being powerful is a permanent
state; this is an essive proper, but only in this instance. Brnaid is rotten
(page 116), on the other hand, is a permanent state. Dchsnigidir (page 196)
in its meaning exists expresses the very essence of an essive. It is a late
coinage based on the VN of docoissin there is, there are (DIL s.v. docoissin 232: 60) and seems to have been coined to express a more
permanent state of being, of existing.
When looking at the relationships between a verb and its base noun
or adjective, it is also important to look at all cases in which the verb simply
means is the noun or is the adjective. These verbs are discussed in the
next paragraph.
3.9.2.6 Semantic relationship between subject or object of a verb and its

base
In the relationship between subject or object of verb and its base, it
becomes clear that the formula [subject] is [base] and [subject] has
[base] are very common, as are [subject] causes [object] to be [base] and
[subject] causes [object] to have [base]. On the other hand, [subject] is
90
Analysis
caused to be [base] is rare and [subject] is caused to have [base] is very
rare. Category D, when deadjectival, has a large overlap with the factitives.
A: Subject is [base]
B: Subject has [base]
C: Subject is caused to be [base]
D: Subject causes Object to be [base]
E: Subject is caused to have [base]
F: Subject causes Object to have [base]
It is impossible for a verb to be both C and E or both D and F. If a verb is

deadjectival, it will be C or D; if it is denominative, it will be E or F. After
all, it is not possible to have [cold], the subject must be caused to be [cold],
araid (page 156). It is not possible to have [stiff], the subject must be
caused to be [stiff], seccaid (page 152). With nouns, it is mostly the other
way around. Cairdinigidir unites in friendship (page 178) is derived from
cairdine friendship. One can have [friendship], but one cannot be it, so
cairdinigidir falls into category E and F (being used both transitively and
intransitively). Another example is fachaigidir incurs a debt (page 209),
derived from fach obligation, payment due. You can have an [obligation],
but you cannot be an [obligation], so this falls into category E and F (being
ambitransitive27).
A possible exception to the above rule is cruthaigidir creates (page
187), derived from cruth form, shape. It could be argued that this verb
could mean either to cause to have [form] or to cause to be [form].
Considering that all the other denominative factitives have the have
connotation, it is tempting to interpret this verb in this way, but we cannot
forget that these are my categories that I am superimposing on these verbs.
It is just as reasonable to say that this verb does not fit perfectly into the
category, as it is to say that this verb specifically meant to cause to have
shape. Moreover, considering that the morphology does not seem to allow
for a form of derivation that expresses both these concepts, it is certainly
27
See footnote 19 for the definition of ambitransitive that is being used in this thesis.
91
Analysis
possible that though it ended up in one category, it still encompassed both
meanings. The same goes for delbaigidir (page 194) and delbaid shapes
(page 126).
Those verbs that are derived from adjectives that can be used
substantively28 fall, almost without exception, into the categories C and D,
implying that the verbs are based on the adjectival use rather than the
substantival use. The three seeming exceptions are quite likely not even
actual exceptions to that rule, because they depend on the interpretation of
each of these verbs.
The first of these verbs is mescaigidir intoxicates, confuses (page
230) which is derived from the adjective mesc drunk, intoxicated. The
adjective is also used as noun, meaning confusion and technically
speaking, the verb mescaigidir could either mean causes [object] to be
intoxicated or causes [object] to have confusion. The same goes for
trnaigidir strengthens (page 253). The adjective this is derived from is
trn strong, which can also mean strength. It is not clear whether the
verb means causes [object] to be strong or causes [object] to have
strength.
Sommaigidir enriches is a slightly different case, because it could
be derived either from the noun sommae riches, wealth or from the
adjective sommae rich, wealthy. The morphology, semantics and
attestations give no clue as to what this is derived from.
The category [subject] is [base] has a strong overlap with the

essives, except for the fact that the essives are permanent states, and these
are temporary. An example is airladaigidir obeys The other meaning of
this verb is is obedient. It has neither the morphology of an essive, nor the
implication of permanence. The verb means obeying to a specific order at a
specific time and place. It is deadjectival from airlithe obedient,
submissive, so this verb expressly means [subject] is [base]. Another verb
that explicitly states that [subject] is [base] is cmailtigidir is insolent
(page 203), derived from cmailt unquiet, restless, but also insolent.
28
One could of course argue that every adjective can be used substantively, but it is those
adjectives of which there are nominal attestations that are here referred to.
92
Analysis
Other As are:
dlmainigidir is free
feidligidir remains
imdaigidir abounds
lobraigidir is sick
londaigidir is angry
sainigidir is different
sulbairigidir is eloquent (although note that in the glosses this seems
to mean speaks well.
alchaigidir is proud
Other Bs:
cosmailigidir likens
badaigidir vanquishes
cumachtaigidir rules
dechraigidir is scattered
deithidnigidir concerns
dummussaigidir is proud, which in verbal meaning of course is
more a temporary essive, but in relationship to its base falls into
category B.
dogailsigidir grieves
espanaigidir is idle
etargnaigidir makes known
filtigidir rejoices
indbadaigidir abounds
miscsigidir hates
mothaigidir is amazed
sercaigidir loves
subaigidir rejoices at
tairisnigidir trusts in
tochomrachtaigidir is weary
aibrigidir treats insolently
dlomaid announces
techtaid has
selbaid possesses
ilid requests
cestaigidir inquires
Other Cs:
ailigidir changes
comoicsigidir approaches
sligidir lowers
lethnaigidir spreads out
mallaigidir makes slow
trmaigidir dries, if not fientive
93
Analysis
Other Ds:
acarbaigidir roughens
ailigidir changes
lgenigidir soothes
asmecnaigedar roots out
boaigidir vivifies
daingnigidir makes fast
echtrannaigidir alienates
comadasaigidir adapts, if transitive
comoicsigidir approaches
ilaigidir increases
cumgaigidir constricts
ingraigidir renders impious
anfossaigidir makes unstable
deimnigidir confirms
lethnaigidir spreads out
danaigidir hurries
dlsigidir makes forfeit
celaigidir makes thin
slnaigidir makes whole
dochraidigidir defiles
sorchaigidir illuminates
trmaigidir dries
nirtigidir weakens
rassaigidir makes void
trummaigidir lightens
fairsingigidir widens
fsaigidir empties
cutrummaigidir equalises
foillsigidir shows
foirbthigidir finishes
grschaigidir makes continuous
imdaigidir abounds
inilligidir protects
sligidir lowers
lobraigidir is sick
londaigidir is angry
madaigidir makes vain
dlmigidir nucleates
maethaigidir softens
mallaigidir makes slow
oenaigidir makes one
ollaigidir makes larger
ridigidir levels
sainigidir is different
sechtaigidir simulates
simigidir thins
sonartnaigidir strengthens
suthchaigidir makes fruitful
trebrigidir continues
94
Analysis
uimligidir humbles
frinaigidir justifies
doraid debases
berraid shears
marbaid kills
nobaid makes righteous
soraid frees
cammaid bends
derbaid certifies
dotromma makes heavy
fraid makes true
glanaid cleanses
mthaid feeds up
emnaid makes double
The other E is fachaigidir incurs a debt.
Other Fs:
ainmmnigidir names
airmitnigidir honours
eltesaigidir makes warm
fachaigidir incurs a debt
findbadaigidir makes happy
fortachtaigidir helps
fothaigidir establishes
foraisligedar ennobles
indbadaigidir abounds
lnaigidir fattens
ruccaigidir puts to shame
sercaigidir loves
soinmigidir makes prosperous
tessaigidir warms
aisligidir exalts
cobraithir helps
doecrathar covers
bathaid feeds
ccaid pays
lnaid fills
nertaid strengthens
rannaid divides
coillid damages
crdaid torments
mesraigidir moderates
ordaigidir orders
95
Analysis
3.9.2.7 Subject causes [base]
Certain verbs express causing their [base], whether a condition, a
state,
quality or
something
else.
Whereas
these
are
neither
morphologically nor semantically actual causatives, the implication of

causation is certainly there. Examples are badaigidir vanquishes, which
is derived from baid victory, triumph and means causes triumph, i.e.
causes [base]. Caird(i)(ni)gidir unites in friendship, derived from
cairdine friendship, amity means causes friendship.
Considering that, as we saw above, deadjectival verbs seem to be
based on the adjectival usage of their base, rather than the nominal one, this
category excludes those verbs. It is not possible to cause firm, for
example, one would have to cause [firmness].
Other base causatives:

aithisigidir reviles
anfossaigidir makes unstable
brthnaigidir judges
cathaigidir fights
crithnaigidir shakes
cruthaigidir creates
cuimlengaigidir fights
cutrummaigidir equalises
debthaigidir dissents
deiligidir separates
delbaigidir shapes
dlmigidir nucleates
etargnaigidir makes known
fachaigidir incurs a debt
fograigidir emits a sound
follaigidir neglects
sraigidir violates
soilsigidir makes bright
teimligidir grows dark
tessaigidir warms
toirthigidir fructifies
aisligidir exalts
molaithir praises
doecrathar covers
said grows
brchtaid bursts
crothaid shakes
96
Analysis
delbaid shapes
dlomaid announces
drbaid lingers
emnaid makes double
achtaid cries out
ccaid pays
nertaid strengthens
rannaid divides
trebaid cultivates
airid guards
coillid damages
conterchomraici collects
crdaid torments
dlaid portions out
gledid cries out
rmid counts
scalid bursts
araid satirises
baitsid baptises
ordaigidir orders
3.9.2.8 Gerative
The geratives (from Lat. gerere bear, carry) are derived verbs
which denote the doing of the meaning of the nominal base.
Other geratives:
crithnaigidir shakes
cundradaigidir trades
dchsnigidir exists
saethraigidir labours
suidigidir establishes
airlithir advises
said grows
brchtaid bursts
ccaid pays
airid guards
dlaid portions out
rmid counts
araid satirises
ilid requests
97
Conclusion
4 Conclusion
The aim of this research was to study the patterns of verbal
derivation in secondary verbs in Old Irish. I have aspired to ascertain the
relationship between morphology, origin and semantics of these verbs and
to place their formation in a relative chronology. I have done this through
the creation of a corpus of all verbal forms in the Wrzburg and Milan
Glosses. These verbs have been classified according to verb class and all the
secondary verbs have been analysed. The conclusions discussed below and
indeed throughout the thesis are based on this analysis. The analysis itself is
based on the corpus of secondary verbs, which can be found below, from
page 107 onwards.
Secondary verbs are by definition weak in Old Irish. The causative /

iterative
morphology
is
no
longer
productive;
the
denominative
morphologies in *--, *-- and *-sag-- are. The *-sag- morphology gains in
productivity in Old Irish and its original meaning becomes bleached. Verbs
formed with suffix *-- become the Old Irish AI verbs (active and
deponent), verbs formed with suffix *-- become the Old Irish AII verbs
(active and deponent, but excluding the -igidir verbs) and verbs formed with
suffix *-sag-- become the Old Irish -igidir verbs.
AI verbs contain denominative verbs and deadjectivals, AII verbs
contain
denominatives
and
deverbals,
and
-igidir
verbs
contain
denominatives, deadjectivals and the only Old Irish depronominal. Although

the very great majority of Old Irish derivative verbs is therefore
denominative and deadjectival (the deverbal morphology was no longer
productive in Old Irish) this shows that it was at least possible to derive
verbs from other parts of speech, even though this was uncommon.
All deverbal verbs are AII verbs. All deadjectival verbs are AI verbs.
Exceptions are the AII verbs asrocholi defines (page 120), doaissilbi
refers, ascribes (page 152) and sobid makes, becomes crooked (page
245). Asrocholi has become AII, contrary to its AI simplex colaid makes
99
Conclusion
thin due to its acquired iterativity. In sobid, the opposite has happened: its
compound consoba is an AI as is expected for a deadjectival. The
iterativity of sobid has motivated its change of class. In the case of
doaissilbi the reason for change from AI to AII is acquired causativity.
Deadjectivals verbs are mostly factitives.
Through careful consideration of the denominative AI and AII verbs,
the underlying motivation for a verb to become either AI or AII has been
worked out. Proto-Indo-European sources for both the denominative suffix
*-- and the denominative suffix *-- have been proposed (athematic
factitive with suffix containing *e or *o plus laryngeal *h2 or *h3 or
denominative suffix *-eh2-i e/o- used for all denominatives derived from all
-stem and some o-stem nouns for *-- and essive suffix or perhaps
denominative suffix *-i e/o- added to i-stem nouns for *--). However, the
original morphological reasons, insofar as those can be ascertained, for a
verb to be formed with either suffix *-- or *- are no longer maintained in
Celtic. A possible exception to this is the factitives, which are still formed
with suffix *--, however, considering that deadjectivals become AI verbs
by default, this cannot be claimed with certainty.
In Old Irish, the motivation for a given verb to become AI or AII is
semantic rather than morphological:
A verb becomes AII only if it is transitive and causative and/or
iterative. Otherwise it becomes AI.
Again, a couple of verbs could be seen as exceptions until they are

observed in their wider context. Gledid cries out and scalid bursts for
example are ambitransitive (i.e. they have both transitive and intransitive
usage) and yet AII. However, they exist in a wider context. Both have
synonyms that are also denominative verbs, namely achtaid and brchtaid
respectively. These verbs are AI, and fit in the scheme. It has been shown
that gledid and scalid are AII because these verbs are more iterative than
their synonyms. On the other hand, malartaid spoils should have been AII,
but is AI because it is less iterative than its synonym coillid damages.
These verbs show that the scheme is not a rigid cast into which every new
100
Conclusion
verb was forced, but that it existed as an underlying framework upon which
a new verb could find a fitting place. The speakers of Old Irish and its
precursors will have been aware of this underlying framework to a certain
extent and will have interacted with it to some degree when creating new
verbs.
The deponent verbs are morphologically derived from the ProtoIndo-European middle voice. Every single Old Irish derivative deponent
verb outside the -igidir class has been formed with that middle voice
meaning. This shows that the middle voice meaning was not only retained in
the primary verbs, but was still a productive part of the deponent
morphology in Celtic. No derivative deponent verbs can be dated with
certainty to the Old Irish period. It has however been speculated that the
criteria for dating these deponents to the Common Celtic period are flimsy
at best and that the hypothesis that the middle voice semantics were kept
into Old Irish should at least be considered.
The -igidir verbs are a reflex of the Celtic derivative in *-sag- which
is itself a reflex of the root *seh2- pursue, seek. Within Celtic, this
formation was at first restricted since it could only be used denominatively
and only in meaning seeks [base]. However, it soon spread to deadjectival
verbs and the meaning became bleached. In Old Irish, this is the most
productive morphology and has become the default category for creating
new verbs. Conversely, it could be argued that the AI and AII morphologies
were less constrained in Celtic and only became more so in Old Irish. Ceisid
grumbles, complains for example, does not have the expected direct
relationship to its base as the other AI and AII verbs have. Had this verb
been created within Old Irish, the expectation would be that it would have
become an -igidir verb. The fact that it is not is an argument for greater age.
-igidir verbs derived from adjectives in -ach/-ech commonly
preserve this in the verb. This is however not always the case. Certain verbs
must be argued to have been derived from a base in -ach/-ech and this -ach/ech was subsequently lost by haplology. A hard and fast rule cannot be
argued; therefore the principle of economy has been applied wherever
possible.
101
Conclusion
For -igidir verbs where an -n- appears between base and suffix, it
has been argued that this was originally the -n- of the stem of those verbs
derived from n-stem bases. This -n- spread by analogy.
There is no single underlying principle for deriving compound verbs.
These can be derived from compounded nouns and reinterpreted as
compound verbs, they can be derived from uncompounded nouns as simple
verbs or they can be derived from uncompounded nouns and in the process
of derivation a preverb can be added to make them compound verbs. If a
simple verb is derived from a compounded noun, this is liable to be
reinterpreted as a compound verb.
A relative chronology for the derivation of a number of verbs has

been proposed based on several factors. A common factor is the occurrence
of Brittonic cognates. If a verb has a cognate in one or more of the Brittonic
languages, it is more likely to be old. This can however not always be taken
on face value. The verb bathaid feeds, for example, has a cognate in W
bwydo. However, for this verb, other factors have to be taken into account.
Both bathaid and bwydo have synonyms in their respective languages (OI
mthaid, ssaid, W llithyaw). Moreover, both bathaid and bwydo form their
verbal noun with a productive suffix. The hypothesis is therefore that these
verbs were created independently in their respective languages.
If a verb on the other hand has a Brittonic cognate and also uses its
base as verbal noun, the argumentation for greater age is strong. Conversely,
a verb denoting a clearly Christian concept can be argued to be a young
formation.
For the -igidir verbs, outside the above concepts further evidence
can be taken into account. As the meaning of the underlying *-sag- suffix is
known to be seeks, verbs that retain this meaning can also be argued to be
old. Then, for those -igidir verbs that have been deemed to be of greater age,
a relative chronology has been proposed based on whether they are
denominative or deadjectival and based on the precision with which they are
based on the original seeks meaning. Verbs that are denominative and can
be argued to have been formed with that exact meaning, such as cathaigidir
fights, i.e. seeks battle have been assigned to the oldest chronological
102
Conclusion
layer among the -igidir verbs, whereas verbs that are deadjectival and have a
meaning becomes have been argued to belong to a younger chronological
layer. On the other hand, since the -igidir morphology is so productive in
Old Irish as well as in the Brittonic languages, case must be exercised when
dating these verbs. Several of the -igidir verbs are rare and some could be
calques on Latin, making it more likely that these are younger formations.
For individual verbs, further specific evidence has been taken into
consideration. OI delbaid and W delwi, for example, are considered to be
individual formations. Formally these are exact cognates, but the semantics
of the Welsh base noun have become restricted to statue and the Welsh
verb is also restricted to that meaning, namely become like a statue.
Another issue arises when the verb is derived from a base that is no longer
extant in Old Irish. In that case, the verb has most likely been derived before
that time.
On the basis of these arguments, several verbs have been placed in a
relative chronology, as follows:
Pre-Celtic: adtluichetdar gives thanks
Common Celtic: callaithir thinks about
Unspecified Celtic (non-igidir):
adgldathar addresses
laid requests
asscarta drives
berbaid boils
bertaid brandishes
crothaid shakes
dlaid portions out
durnid smites
forcenna puts an end to
nertaid strengthens
rmid counts
Unspecified Celtic (-igidir, since the -igidir verbs can be placed in a more
precise relative chronology, the order below is deliberately nonalphabetical)
cathaigidir fights
sraigidir violates
slaigidir sows
tessaigidir warms
follaigidir neglects
sligidir lowers
103
Conclusion
ridigidir levels
sonartnaigidir strengthens
gnthaigidir is accustomed to
cuimnigidir remembers
Pre-Old Irish:
dodona consoles
dnaid closes
todid shines
Old Irish:
aithirgigidir repents
baitsid baptises
bathaid feeds
cestaigidir inquires
cinnid defines
delbaid shapes
dchsnigidir exists
rannaid divides
From the semantic discussion, it has become clear that there is a
wide range of possibilities for the relationship between a verb and its base
noun or adjective. The verb can mean causing someone to do a verbal
action (semantic causatives), doing [base] repeatedly (semantic iteratives),
becoming [base] (semantic fientives), beginning [base] (inchoatives) or
being [base] (essives). The most common relationship is that where the
subject is caused to have [base] and that of the base causatives. Also
common are the verbs where the subject causes object to have [base] and
those where the subject has [base]. Furthermore, there are those verbs where
the subject is [base], subject is caused to be [base] and subject causes object
to be [base]. Lastly, the geratives are the verbs which denote the doing of
the meaning of the nominal base.
There is significant semantic overlap in the possible relationships
between verbs their bases: many verbs occur in several categories. This is
especially true of the -igidir verbs. It is also the -igidir verbs that are often
hard to classify at all. The relationship between the verb crottigidir plays
on a harp (page 186) and its base crott harp for example is immediately
clear. It is however quite unique as a relationship: there are not many verbs
in which the verb expresses using [base].
104
Conclusion
Those verbs that are derived from adjectives that are also used
substantively fall, almost without exception, in categories C and D, i.e. in
the categories Subject is caused to be [base] and Subject causes Object to
be [base], rather than in the categories Subject is caused to have [base] or
Subject causes Object to have [base]. Since one cannot have an adjective,
but one must be an adjective (one cannot have [cold] or [stiff], one must be
[cold] or [stiff]), it has become clear that these verbs are derived from
adjectival use only.
These results about the relationships between morphology, origin,

semantics and chronology are based on the clearly defined corpus of verbs
from the Wrzburg and Milan Glosses. The conclusions drawn can therefore
at this point not be taken as strict rules for the whole of Old Irish or indeed
the whole of Irish. The very size of the corpus and the large fraction of
contemporary Old Irish material that it forms nevertheless gives a valid
basis for a sound working hypothesis.
It has become clear that the speakers of Old Irish and its precursors
used all of the morphological means available to them to extend the
semantic fields their language as needed to provide for cultural
developments. This study of secondary verbal derivation has shown but a
fraction of the creativity of speakers when it comes to innovation and
originality in using their language and expanding it. Although the language
and the morphology itself provide not only opportunity but also constraint,
the speakers have used even the constraints to create even further
opportunities. The opportunities are clear: Any noun or adjective can be
turned into a verb by using the suffixes *--, *-- or *-sag-, depending on
how the verb relates to its base. A simple or compound verb can be made
out of an uncompounded or compounded noun. The constraints dictate in
most cases whether a verb shall become AI or AII. However, where the
constraints in the language mean that a deadjectival must become AI, but
other constraints dictate that transitive iterative become AII, speakers
interact with these constraints to turn the occasional AI deadjectival into an
AII. Moreover, where two synonyms, such as gledid and achtaid, both
105
Conclusion
meaning cries out have a subtle difference in meaning, speakers can
express this difference through the use of different verbal classes. In this
case, gledid is more iterative than achtaid and is therefore AII, even
though both these verbs are intransitive and both these verbs should
therefore have been AI. In this study, the ground rules have been established
and the possibilities have been shown, with fascinating implications and
promise for further research.
106
Corpus of secondary verbs
5 Corpus of secondary verbs

5.1 Structure and notational conventions
Verbs are organised under their simplex as headword, except where
no simplex is attested and none can be reconstructed, in which case the
headword is the verb with the least amount of preverbs. The headword is
printed bold. If compounds are attested in the corpus, the headword is
printed bold and underlined and each compound is printed bold. Headwords
are given in their classical Old Irish form, even if this form has not actually
been attested. This means that the conventions in DIL are sometimes
departed from. Where DIL gives Middle Irish active forms of originally
deponent verbs, these have been silently altered. The choice of convention
for form of headword should not be interpreted as a claim that the headword
necessarily existed and in the rare cases where an active form is the only
attested form of the verb, the hypothetical deponent form has still been
given as headword, for consistency (cf. McCone 1985 for a discussion of
Middle-Irishisms in the Old Irish Glosses). The diphthongs a~e and o~e
were already starting to fall together in Old Irish, and this is indeed already
the case with several verbs in the Milan Glosses, although this is rare. For
the sake of etymological consistency, in this thesis the diphthongs have been
normalised following their etymology. Palatalisation was blocked in several
environments (for details, see McCone 1996: 116), giving for example AII
folugai instead of **foluigi. However, confusion often arose in analogy to
the other verbs and in many verbs, a mixture of palatalised and nonpalatalised forms can be found already in Wrzburg. By Middle Irish, all
AII verbs have been synchronically palatalised. DIL usually gives the
Middle Irish forms for their headwords. In this thesis, the non-palatalised
forms are given for the verbs in question. Other deviations from DIL are
marked and discussed. Reference to the DIL headword is always given. If
the headword itself is not attested within the Wrzburg and Milan Glosses,
but is extant elsewhere, it is preceded by , if it is not attested at all, it is
preceded by *. The translations following the headword and the individual
107

compounds are the first two translations given in DIL, unless otherwise
stated.
Where a verb can be or has been directly derived from a Proto-IndoEuropean verbal root, either primary or secondary, this root is given, with
the development to Common Celtic. Where roots are quoted from prelaryngealistic sources, laryngeals are silently supplied. Cognates in other
languages follow.
Within the entries, first the Wrzburg, then the Milan forms are
given, in the following tense and mood order: present indicative, imperfect,
imperative, present subjunctive, past subjunctive, future, secondary future,
preterite active, preterite passive, participle, verbal of necessity, verbal
noun. Passives follow actives in each tense and mood. Following DIL,
suppletive forms are given under the verb they supply forms for. The one
exception to this are the cuirethar forms which supply forms of foceird.
Foceird is a primary BI verb and as such falls outside the discussion, but
cuirethar is a secondary AII verb. Moreover, it has several compounds that
do not stand in suppletive relationship to other verbs. Simplex and
compounds are given under headword cuirethar.
Nominal derivations are noted as denom. / deadj. < [base noun /
adj.] [meaning]. The verb arsadaigidir is therefore given as Deadj.
< arsaid old, ancient. If the base itself is derived, that derivation is given.
Verbal derivations are given as caus. / iterative (of [OI verb] if applicable).
Verbal Latin loans are given as Lat. [verb]. Where applicable, discussion
follows.
Full references are given at the end of each lemma.
108
5.2 AI verbs
adcobra desires, wishes
Wrzburg: pres. ind. 2sg. with interrog. inaccobri 6a10, 3sg.
adcobra 28b2, with ma and infix. pron. 3sg. n mtchobra 28b1, rel. with
emp. pron. 3sg. m. adchobrasom 28b1 with rel. n (and elision of infix pron.
3sg. n) asinchobra 10b18, pres. subj. 3sg. rel. ndaccobra 6c1, past subj. 1
sg. with cia ciadcobrinn 17d17, Milan: pres. ind. 2sg. rel. adchobrai 56b31,
65b2, 3sg. adcobra 73a14, 3pl. adcobratsidi 27b15, rel. adchobrat 87c3,
87c3a, 89b5, pass. sg. rel. adchoburthar 88b4, pres. subj. 3sg. adcobra
35a1, 1pl. adcobram 94a7, 3pl. cuadcobrat 89d16, with infix. 3sg.
ciatchobrat 57c5, perf. 1sg. ni con ruaccobrus 136b7, 1pl. rel.
adrochobursam 56b24, 3pl. rel. adrochobairset 67b10, VN (n o) accobar,
accobur 41d11-12, 54a9, 59b10, accubur 80a9, 90d11, acc. sg. accubur
53c16, 68b12, acc./dat. sg. accubur 69a6, gen. sg. acubuir 67c8 dat. sg.
accubur 68b9, gen. pl. accubur 54a34.
Denom. < accobar desire, wish, also used as VN. Cf. milchobur
honey-desirer, bear (Thurneysen, Watkins).
Thurneysen 1921a: 106, LEIA C-135, DIL s.v. ad-cobra 44: 38,
milchobur 136: 58, Watkins 1962a: 115.
*adcranna casts lots (translation mine)

Milan: part. accruinte gl. sortita 29c1.
Denom. < crann tree, wood, also wooden piece used in casting
lots (DIL 509: 33). LEIA does not register the verb. Only the participle is
attested, and this is hapax legomenon.
IEW 633, LEIA C-222-3, DIL s.v. accruinte 18: 82, crann 508: 17,
509: 33.
aditchethar detests
Milan: pres. ind. 3sg. adeitchethar gl. detestatur 50d9, 129c10, pass.
sg. adetchethar gl. detestatur 122b12, pres. subj. 2sg. aditchide gl. detestare
103a7a, part. (Note DIL: aditig? part. of aditchethar?; io, i)
109

aditchide, aideitchide gl. detestanda 36a16, aditchidi gl. exacrabilia 74c2,
nom. pl. adeitchidi gl. profanatis 100c16-17, VN (m u) aditchet, acc. sg.
aditched 74c3, gl. execrationem 74c4, adeitched gl. detestationem 103a7,
dat. sg. aditched gl. detestatione 50b4.
Deadj. < tig unnatural, unseemly. Although this looks at first sight
like an AII deadjectival, it is not. OI tig is a reflex of *an-tekis. An a-verb
formed from that would be *antektor and an i-verb *antektor (both with
regular syncope). They would be completely indistinguishable in Old Irish
in all their forms. The verb is synchronically AII, but was formed with
suffix *-- and is therefore originally AI. This also shows that this verb
must have been created before Old Irish. If it were, it would have been
formed as an -igidir verb (**itgigedar or possibly **itigedar), because
AII deadjectivals were not allowed (cf. p. 62). The verb only occurs in
Milan.
Deponents are discussed in section 3.3.1.3, pages 63f.
DIL s.v. ad-itchethar 52: 1, tig 239: 53.
adraid adores, worships

Wrzburg: pres. ind. 3sg. rel. adras 9c33, perf. 3sg. rel. adrorsat
1b19, impv. 3pl. adrat 6d7, Milan: pres. 1sg. nadnadraim 132c1, impf. 3pl.
(rel.) adordais 36d16, past subj. 3sg. (rel.) adrodar 14b4 (Note Thes.: For
adrodar, Ascoli suggests adradar, J.S. adrorad; DIL follows the adrorad
reading. A pass. is impossible in this context, and adraid would not take
dep. endings in the pret., so I accept DILs reading.), fut. 3sg. rel. adnorfa
81d6, perf. 3sg. rel. atror 69d3 (Note Thes.: MS. at ror), VN (m u) adrad
93a1, acc. sg. atrab 94b25, gen. sg. adrada 49a23, 73d2, 3 dat. sg. adrad
81d5, 104c4, nom. pl. adradai 46b30, acc. pl. adradu 62d1, dat. pl.
adradaib 46b26, 47d10.
Lat. adorare honour, adore (cf. section 3.2, page 48 for a
discussion of Latin loans). In Wb. Lex. as ad-ora. The only certain simplex
form in Wrzburg and Milan is the 3sg. rel. adras (Wb. 9c33), since the
relative endings are confined to the absolute forms of simplex verbs.
Against the one simplex form, we find four compound forms, three with roinfixed after ad- and one with a nasalisation infixed after ad-. That implies
110

that the scribe saw adas a preverb. The other attestations in Wrzburg and
Milan have no diagnostic features. Considering that the earliest attestations
already show a mixture, it is impossible to determine whether this was
originally a simplex or a compound verb and DIL has therefore been
followed. When a verb could be treated either as a compound or a simple
verb, it seems that Old Irish had no one way of dealing with these. Some
verbs became simple verbs, some verbs became compounds (cf. consecra
consecrates, page 123, borrowed from Lat. consecrare), and some, like
adraid became a mixture. The issue has been briefly dealt with in section
3.5, pages 74f.
Ped. ii 450, LEIA A-18-9, DIL s.v. adraid 66: 42.
araid satirises, lampoons

araid: Wrzburg: pres. subj. 2pl. with infix. pron. 1pl.: airid 16a1,
doathara (censures, finds fault with): Wrzburg: pres. ind. 3sg.
donathira 3d3.
Denom. < ar, cutting, incising, but also act of satirising and
defaming. Root uncertain. According to LEIA we can establish a
connection with Lat. aerumna toil, task, but aerumna is a loan from Greek
(Walde & Hofman), which precludes a connection with the Irish.
Note that the glide is not written in the Wb. form.
LEIA A-20, D-119-20, DIL s.v. eraid 79: 42, er 78: 48, doathara 201: 3, Walde & Hofman 1982 s.v. aerumna.
armaid arms, equips

Milan: pres. ind. pass. pl. narmtar (Note Thes.: MS. narintar: corr.
Ascoli) 16b6, pres. subj. 2sg. with infix. pron. notarmaesiu 44a21.
It is unclear whether armaid is denom. < arm armour, battle
equipment, from Lat. arma weapons, armour or directly borrowed from
Lat. armare equip. See above, page 48, for further discussion.
LEIA A-89, DIL s.v. armaid 403: 84, arm 402: 39.
111

said grows; increases
said: Milan: pres. ind. 3pl. rel. nastae 87b12, perf. 3pl. rosaiset
2a6, forsa (grows, increases): Wrzburg: impf. 1sg. forssin 18c15,
Milan: pres. ind. 3sg. forsa 29a10, pres. subj. 3pl. forsat 40b4, 46d7, perf.
2sg. forrassissiu (in DIL spelled forrasais-siu; Thes. has another
interpretation: leg. forrarsissiu (= fo-n-ror-asis-siu)? J.S.; in the MS. Nigra
and Ascoli read forrarsissiu) 38a9, forrassais 43d17, 3sg. forrs 115d11.
Denom. < s act of growing, growth.
LEIA A-92-3, DIL s.v. s 421: 61, said 422: 33, for-sa 309: 14.
asscarta drives
Cf. W ysgarthu, purge out, dysgarthu, cleanse, Br. skarza
vacate, clean, OB iscartholion gl. stuppa (Ped.).
Milan: past subj. 3pl. with infix. pron. 1sg. atamscartis 59a21;
dofoscarta (removes, puts aside): Milan: impf. pass. with infix. pron.
1sg. domfoscartae 127c20, part. diuscartae 19c15, diuscarte 32d16.
This verb is derived from a Celtic verbal adjective *skar-to- of the
verb that becomes OI scaraid separates, parts. This participle itself is not
attested in Old Irish. The Brittonic cognate and the fact that the base is not
attested in Old Irish points towards this being an inherited formation, see
Chronology, sections 3.6 and 3.7 pages 75ff. Asscarta is found in the
chronology on p. 80.
Ped. ii 616, LEIA S-34-5, DIL s.v. as-(s)carta 438: 27, do-foscarta
274: 48, scaraid 74: 62
bennachaid blesses
Wrzburg: pres. ind. 3sg. with infix. pron. 1pl. and emp. pron. 1pl.
nonbendachani 11b7, 3pl. with infix. pron. 2pl. nobbendachat 7b21, fut.
1sg. with interrog. part. and infix. pron. 2sg. indatbendachub 33d11, pres.
subj. 2sg. manibendachae 5d23, Milan: pres. ind. pass. sg. nobendachthar
33d23, verb. nec. bendachdai 138a1.
Lat. benedicere bless, praise. Latin loans have been discussed in
section 3.2, pages 48ff.
LEIA B-36, DIL s.v. bennachaid 77: 1.
112
berbaid boils, cooks

Cf. W berwi boil, Br. bervi, birvi boil (LEIA).
Milan: pres. subj. 3sg. maniberba 46c15.
Also beirbid. Based on the Proto-Indo-European root *bher-, which
gives the above Welsh and Breton verbs. The Old Irish verb is secondary
and an inherited denominative, because the base noun is not extant in Old
Irish. It has been derived in Celtic from *beros. This noun is attested in
Welsh, berw stock, bubbling (also as adj. boiling), MB beru, ModB bero
stock (LEIA). Berbaid is found in the Chronology on p. 80. It is not
impossible that berbaid was originally a primary verb that has taken on
weak flection, but note that in most cases where primary verbs have taken
on weak flection, reasons can be given (most often nasal presents which
become weak through regular phonological changes or a verb in a different
tense or mood that has been reinterpreted or re-formed as a present form of
a weak verb). There seems to be no underlying motivation for a primary
verb going weak in this case. (See chapter 6, pages 259ff. for an overview of
the primary verbs gone weak in the Wrzburg and Milan Glosses.)
IEW 143-5, LEIA B-40-1, LIV2 81, DIL s.v. berbaid 82: 22.
berraid shears, clips

Cf. W byrr, f. berr, Corn. ber, Br. berr short, MB berrhat
shortens, cuts (LEIA).
berraid: Wrzburg: impv. 3sg. berrthar 11c13, perfect. past subj.
3sg. roberrthe 11c13; doberra (shears): Wrzburg: past subj. pass. sg.
doberrthe 11c12.
Deadj. < berr short. DIL suggests that the adjective could also
mean short-haired. However, they base this on the meaning of the verb.
Considering the cognates in Welsh etc., it is clear that the adjective just
means short. The verb then originally meant shortens, and took on the
secondary meaning shears. The meaning of the adjective seems not to
have been affected. Doberra is hapax legomenon. The Middle Breton verb
berrhat is not a direct cognate of this, because the -ha- shows that this is a
*-sag-- verb.
113

LEIA B-42, DIL s.v. do-berra 214: 5, berraid 84: 74, berr 84: 21.
bertaid brandishes, flourishes

bertaid: Milan: pres. ind. pass. sg. bertaigther 75b13, arberta
(proposes, intends): Wrzburg: perf. 1sg. with emp. pron. 1sg.
arromertussa 31d14, 3sg. rel. with emp. pron. 3sg. m. arrobertsom 29d23,
Milan: pres. ind. 3sg. rel. araberta 102c5, impf. 3sg. rel. arambertad 53d4,
perf. 1sg. arromertus 51a12, 58c9, arrumertus 58d17, 3pl. rel. ararobert
66c19, VN airmert 98a7, dat. sg. airmiurt 40d12, airmbiurt 65a8.
Denom. < bert, burden, load, also deed, exploit. The forms with
-m- instead of -b- (such as arromertus Wb. 31d14 and the VN airmert Ml.
98a7 above) are not fundamental to the Old Irish verb but a contamination
based on a root *smer-, cf. W armerthu, d-armerthu, prepare, Vann.
armerhein, prepare (Isaac).
The different meanings of the verb correspond closely to the
different meanings of its base. Note the existence of a young verb
bertaigidir (see below, page 174). Bertaigidir is not found in Wrzburg and
only has the meaning shakes. Bertaid on the other hand occurs with wider
meanings that seem to cover the width of the meanings of the adjective,
suggesting greater age (see also Chronology, pp. 80 and 82). It is likely that
bertaigidir was formed for the more specific meaning of shakes. DILs
statement bertaigidir = bertaid (s.v. bertaigidir) is therefore incorrect.
Although there is certainly an element of iterativity in bertaid, this
has nothing to do with the relationship between the noun and the verb; the
fact that bertaid is an AI verb is thus consistent with the scheme proposed in
this thesis (see section 3.3.1.1, pp. 55ff. for the importance of iterativity, as
well as causativity and transitivity in the scheme underlying formation as AI
or AII verbs).
Ped. ii 475-6, LEIA B-4-3, DIL s.v. bertaid 86: 69, bertaigidir 86:
85, bert 85: 53, ar-berta 380: 8, Isaac 1996: 321.
bathaid feeds, nourishes

Cf. W bwydo feed (Ped., note that he cites a rare late form bwyda).
114

arbiatha (feeds, nourishes): Milan: pres. ind. 3pl. rel. arabithfet
46c8.
Denom. < biad food. The W bwydo is denominative from bwyd.
Whereas it could be argued that the Welsh and the Old Irish are cognates,
there are two things that speak against this. First of all, both Old Irish and
Welsh have synonyms for the concept of feeding (such as OI mthaid,
ssaid, cf. DIL, MW llithyaw, cf. Isaac), which makes it less likely that it
was this particular verb that was used in Celtic. Secondly, both verbs form
their verbal noun using a productive suffix within the individual language. I
argue therefore that these verbs are not cognates, but have been individually
coined within the languages, see also pp. 76 and 82.
Ped. i 58, LEIA B-47-8, DIL s.v. bathaid 96: 23, biad 93: 2, arbiatha 380: 28, mthaid 119: 27, ssaid 63: 68, Isaac 1996: 340.
bocaid softens, makes soft

Wrzburg: past subj. 2sg. with infix. pron. 2sg. notbocctha 5b32.
Deadj. < boc soft, gentle.
Thes. translates the form in Wrzburg above as boasts and DIL
follows that (s.v. bocaid 129: 75-78). They suggest that since the verb
modid boasts, which they derive from moth soft, has a similar semantic
development from soft to boasts, the verb bocaid could also be used in
this way. However, modid is a causative (see below, page 232) and
therefore DILs justification for translating bocaid as boasts falls. No
justification is given in Thes., but the context requires no translation
boasts. Softens will suffice just as well. The gloss reads ... nptha labar
et npadro (mr) notbocctha et armtairismech inhiris, ... thou shouldst not
be talkative (?), and thou shouldst not boast / soften yourself overmuch, and
that thou shouldst be constant in faith. The verbal form itself has an infixed
pronoun 2sg. In this context, you should not soften yourself, perhaps
meaning you should not be too soft on yourself fits just as well as you
should not boast. There is no reason to add an otherwise unattested
meaning to bocaid.
LEIA B-62, DIL s.v. bocaid 129: 50, boc 127: 80, modid 158: 35,
maeth 23: 61.
115
brnaid is rotten, putrid, putrifies

Milan: perf. 3pl. robrnsat 58a15.
Deadj. < brn stinking, fetid. This is an unusual deadjectival,
because most deadjectivals are factitives, see section 3.3.1.2, page 62. Its
base brn is not attested in Old Irish, but its existence is clear from
attestations in the later language. Note that the entry in LEIA is inconclusive
about the origin of the noun. The corresponding Brittonic forms (W braen,
MB breyn) are given, but the inconsistency in vocalism between OI -- and
MW -ae- is not addressed.
LEIA B-84-5, DIL s.v. brnaid 177: 14, brn 176: 69.
brchtaid bursts, belches forth

dobrchta (gushes forth, breaks forth): Wrzburg: pres. ind. 3sg.
dubrcti 9d1, Milan: VN tobrchtad, gen. sg. tobrchto 64c15.
Denom. < brcht burst, bursting.
Thes. notes: Three letters, cti (cti vel ctr Zimmer), seem to come
between and , leg. dubr a triassom? J.S. In the Addenda and
Corrigenda, page 726, the following note is added: Nach dubr lese ich
auch cti oder ctr; ersteres wahrscheinlicher, Chroust.
Wb. Lex. has the compound as dubrcti, seemingly based on the
emendation, and refers to the note in the addenda and corrigenda of Thes.
DIL registers the verb as do-brcta, gives the form in Wrzburg with a
question mark and states reading doubtful, see note.
Brchtaid would have resisted palatalisation even if it were AII, so a
conjunct form is needed to decide whether this verb is AI or AII. Two such
forms are attested. One is the form dobrcti in the Wrzburg gloss above,
the other the pres. ind. 3sg. conj brchta in the Old Irish Flire Oengusso
Cli D (Stokes 1905b) for June 21st. The reading in Wrzburg is doubtful,
but the reading in the Flire is clear. Moreover, taking brchtaid as an AI
verb is consistent with the subclassification scheme argued in this thesis as
brchtaid is intransitive (see section 3.3.1.1, pages 55ff. for the scheme;
brchtaid has been discussed on page 59.).
116

LEIA B-102, DIL s.v. brchtaid 209: 53, brcht 209: 16, do-brcta
217: 66, Stokes 1905b.
cammaid bends
Cf. OW cam, MW cam, camu cuts, twists, MC cam-hinsic gl.
iniustus, OB camm gl. obliquus, MB kamm cut, twisted, kamma cuts,
twists (LEIA).
Milan: VN cammad, acc. pl. cammaigthiu 134d5.
Deadj. < camm, crooked, bent. OI camm is a loan from MW camm,
because the Old Irish cannot be a direct descendant of *cambos; that would
give OI **camb. Since OI camm is consistently spelled camm29, it cannot be
camb. For Continental Celtic cognates of the noun, see Delamarre s.v.
*cambo-.
LEIA C-28-9, DIL s.v. cammaid 63: 68, camm 62: 57, Delamarre
2008 s.v. *cambo-.
caraid loves
Wrzburg: pres. ind. 1sg. with infix. pron. 3pl. and emp. pron. 1sg.
noscarimse 5c7, with emp. pron. 1sg. no(ca)rimse 14b1, with rel. n and
infix. pron. 2pl. and emp. pron. 1sg. nondubcairimse 23c12, 2sg. rel. nocari
6c8, 3sg. carid 25d5, with infix. pron. 2pl. nobcara 27d9, rel. caras 25c19,
3pl. with infix. pron. 1sg. and emp. pron. 1sg. nmcharatsa 5c6, impf. 3sg.
with infix. pron. 2pl. nobcarad 23d10, fut. 3pl. ni chechrat 30c4, perf. 3sg.
with infix. pron. 2pl. and emp. pron. 2pl. robcarsi 23d4, 1pl. with rel. n and
infix. pron. 2pl. amal rondobcarsamni 25a35, perfect. pres. subj. 3sg. rel.
conrochra 6d1, 3pl. rel. aracarat 11b6, past subj. 3sg. rel. nd carad som
4d17, impv. 3sg. carad 22c19, 23c12, 25a35, 2pl. carid 27d11, Milan: pres.
ind. 1sg. rel. nocaraim 79d1, 3sg. caraid 75c4, rel. charas 30c3, caras
91b17, 3pl. carait 124d13, with infix. 3pl. noscarat 123c8, pass. sg. carthar
75c4, pass. pl. rel. cartar 78a2, impv. 2pl. with infix. 3sg. nacarid 68a15,
pres. subj. 2sg. care 43a21, nocarae 56b39, neg with infix. 3sg. n charae
56b31, past subj. 3sg. cocarad 14a8, perf. 3sg. rel. rocar 65d5, part.
29
The only exception is in the Annals of Ulster, the manuscript of which is much later than
the Old Irish period. Caimb, gen. sg. there is therefore most likely a hypercorrection.
117

carthach, carthacha 130b3, carthaig 133d7, VN (GOI 721, but cf. p. 683
note 157) serc 56b28-29, dat. sg. seirc 68b9, 98c3.
Given in IEW and by Pedersen as cognate to Latin crus, Sanskrit
cru- etc., but the OI -a- is short, while the -a- in other Indo-European
languages is long (LEIA). It is not registered in LIV2 or Schumacher (2004).
No reasonable Old Irish base noun or adjective can be found, and two
realistic possibilities for its primary origins have been put forward by
Watkins and Hamp. Watkins considers this to be a formally gramaticalized
borrowing from emotive child language and primary. Hamp suggests either
PIE *keHar- or *kerHa-. In the first case, the long found in for example
Latin and Sanskrit would be regular and Celtic would have methathesized
and the in second case, the short a found in Celtic would be regular and the
other languages would have methathesized. (Cf. also Beekes and McCone).
However, Isaac points out that this verb could also be an old
derivative from an adjective PIE *kh2-r--/*kh2-r-h2- desired, desirable,
loved, giving a verb PIE *kh2-r-h2-i e-ti > CC *kar--t(-i) > OI caraid,
cara.
Ped. i 184, ii 481-2, IEW 515, LEIA C-36, DIL s.v. caraid 73: 17,
Watkins 1962b: 185, Hamp 1976: 5-6, Beekes 1988: 88, McCone 1991b:
110, Isaac 1996: 367.
celebraid celebrates
Wrzburg: perf. 1sg. arocelebrus 14d31, Milan: past subj. 3pl. rel.
aracelebartis 102d3, VN celebrad 80a4, dat. sg. celebrad 133c9.
Lat. celebrare celebrate, perform. Latin borrowings have been
discussed above, section 3.2, pages 48ff.
LEIA C-60, DIL s.v. celebraid 109: 12.
certaid disposes, arrays

concerta (emends (text), corrects (person)): Milan: perf. 1sg.
conacertus 2a1, with infix. pron. 3sg. m./n. conaicertus 2a13, 3pl.
aconcertat 129c11, perf. 3sg. with infix. pron. 3sg. m./n. conaicert 2a6,
etarcerta (interprets, decides): Wrzburg: pres. subj. 3sg. rel. co
118

etercerta 12c32; pass.: etarcertar (Note Wb. Lex.: Editors: pres. ind. 3sg.;
DIL E, col. 231: impv. 3sg.; Ped. II, 485: leg. etarcertad.) 13a6.
Deadj. < cert, correct, right < Lat. certus certain (the Latin verb
certare vie with, contest has been deemed unsuitable as origin on semantic
grounds). Direct and indirect borrowings from Latin have been discussed in
Ped. ii 485-6, LEIA C-75-6, DIL s.v. certaid 145: 58, con-certa 428:
53, eter-certa 231: 51, cert 143: 1.
*callaithir thinks about, cares (transl. mine)

Cf. W pwyllaf think about, care, gobwyllaf think, meditate,
consider and prepare.
arfoichledar (gets ready beforehand, prepares): Wrzburg: pass.
perf. 3pl. arforchelta 4c37, Milan: perf. 2sg. with infix. pron. 1sg. rel.
arndamroichlis 74d7, part. airche(l)tae, erchelltae 19a2, ercheltae 53b14,
nom. pl. aircheltai 27b10, acc. pl. ercheltu 32a18, VN airichill, erfoichell
73d2, 3; focallathar (pays heed to, takes care of): Wrzburg: perfect.
pres. ind. (or pres. subj. rel.) 3pl. rel. with infix. pron. 3pl. nacharochlat
19c15, fut. 3pl. rel. fochelfatar 30d15; pass.: pres. ind. rel. focaltar 29c4,
perf. 3sg. with infix. pron. 1pl. fonrochled 19c13, Milan: pres. ind. 3sg.
fuciallathar 114b3, impv. 2pl. foichlid 68a15, fut. 3sg. fucheillfea (cf.
McCone 1985: 98) 90c15.
Denom. < call sense. Although there are only two unclear
attestations of the simplex *callaithir,30 the Welsh cognate confirms that
the simplex existed at some point. In the Chronology above (section 3.7,
pages 77ff.) I have given callaithir as an early Celtic derivation (callaithir
is on page 80). Its Welsh cognates and the fact that it uses its base as VN are
arguments for it being an inherited formation. I argue that this is an early
formation because thinking, considering, taking care, are very basic
30
Pedersen (ii 482) refers to ciallathar (LU 91a4 = Best & Bergin 1929: 226, line 7373)
and ciallastar (LL 14231 = Best & Bergin 1956: 455, line 14231). DIL has ciallathar as
headword with a question mark and a tentative translation gathers? and a reference to
focallathar. They register ciallastar under arcallathar.
119

concepts. The motivation for formation as a deponent verb has been
discussed in section 3.3.1.3, pages 63f.
Note the confusion with roots in cel- in the compounds (Ped.).
Ped. ii 482, 484, LEIA C-93-4, DIL s.v. ar-foichlea 393: 73, call
174: 82, fo-callathar 203: 18, ciallathar 177: 61.
clannaid plants, sows

Wrzburg: pass. perf. 3sg. with infix. pron. 2pl. rob clandad 21d6,
Milan: pres. ind. pass. pl. rel. clantar 63b9, impf. 3pl. rel. noclantis 63b9,
VN clannad, clandad 121c12.
Denom. < clann plant, planting, < Lat. planta cutting, seedling or
a direct borrowing from Lat. plantare propagate from cuttings. Direct and
indirect borrowings have been discussed in section 3.2, pages 48ff.
LEIA C-112-3, DIL s.v. clannaid 218: 27, clann 217: 8.
clonaid bends
Milan: VN clenad, gen. sg. clointa 127c22.
Deadj. < clen uneven, crooked.
IEW 601, LEIA C-110, DIL s.v. clenaid 211: 20, clen 208: 14.
cobraithir helps, succours

Wrzburg: past subj. 3sg. with infix. pron. 1sg. manimchobrad 3d16,
impv. 3sg. cobrad 12b8.
Denom. < cobair help, assistance. Motivation for formation as a
deponent verb has been discussed on pages 63f.
LEIA C-135, DIL s.v. cobraithir 280: 39, cobair 277: 39.
colaid makes thin, grows thin

Cf. W cul narrow.
asrocholi (defines, determines): Wrzburg: pres. ind. 3sg. (with
pret. and perf. meaning) asrochoili 10b20, impf. ind. 3sg. rel. asrochoiled
27a17, Milan: pres. ind. 3sg. asrochoili 35d22, impv. with infix. pron. 1sg.
atamrochoilse 24a15, past subj. 3sg. asrochoiled 36b13, pret./perf. 3sg. rel.
diandrerchoil 46c7, 1pl. (rel.) asrochoilsem 22c3, 2pl. rel. asrochoilsid
120

95c3, 3pl. rel. asrochoilset 95c2, part. erchoilte 130b9, VN rcholiud,
rchoiliud 22c3, rchoiliud 136d6, erchoiliud 31d1, dat. sg. rchoiliud
22c3, erchoiliud 22c3, gen. sg. rchoiltea 26d13, nom. pl. erchoilti 74d7,
rchoilti 74d9, acc. pl. aerchiltiu 132c15; *remiercholi (predestinates,
predetermines): Wrzburg: perf. 3sg. rel. remirrchil (MS remirirchil)
4b8.
This is related to colaigidir makes thin, grows thin (see below,
page 183), but pace LEIA the verbs were coined independently of each
other. Both are derived from col thin, slender, but they are
morphologically distinct: colaid is an AI active verb and colaigidir is an
-igidir verb.
The compound asrocholi has not only become figurative in
meaning, but has also changed class. As discussed above, section 3.3.1.2,
page 62, this is due to its acquired iterativity: Asrocholi means defines,
determines, which implies the repeated making thinner, making narrower.
Note that DIL cites the base adjective in its Middle Irish form with
-e-. The Old Irish would have been -o-. Pedersens connection with W
coel is incorrect: W -oe- does not correspond to OI -o-.
Ped. ii 495, LEIA C-6, DIL s.v. as-rocholi 436: 46, remi-rcholi
43: 32, celaid 13: 3, celaigidir 13: 6, cel 10: 82.
comalnaithir fulfils, discharges

Wrzburg: pres. ind. 3sg. rel. comalnas 7b15, nd chomalnathar
27c14, with infix. pron. 3sg. n. nodchomalnadar 15b14, 2pl. hore
nocomalnid 15a9, 23a15, with infix. pron. 3sg. n. manidchomalnid 18b7,
3pl. with emp. pron. 3pl. n comalnatsom 20c22, rel. forsnah comalnatar
20d1, comalnatar 20d2, hore nadcomalnat 31b31, perf. 3sg. with infix.
pron. 3sg. rachomalnastar sede 24a37, with infix. pron. 3pl. roscomalnastar
21b9, 2pl. with emp. pron. 2pl. rocomalnisidsi 26b6, 3pl. rel. ind
rochomalnisset 2c14, pres. subj. 3sg. rel. with infix. pron. 3sg. n. (elided)
arinchomalnathar 31c14, 3pl. machomalnit 28c7, 3sg. with infix. pron. 3pl.
maniscomalnadar 29a16, perfect. 3sg. with infix. pron. 3sg. n.
actrachomalnathar
31b11;
1pl.
diacomalnammar
15d27,
rel.
cocomalnammar 31c11, 2pl. dus incomalnid 14d23, with infix. pron. 3sg.
121

n. nachomalnith 13a32, rel. aracomalnid 32a3, perf. 2pl. conrochomalnid
7c10, 23b40; 3pl. manicomolnatar 2c10, past subj. 1sg. with infix. pron. 3sg.
n. (elided) danchomalninn 3c28, 2pl. with infix. pron. 3sg. n.
cenachomalnithe 13a33; impv. 2sg. with infix. pron. 3pl. noscomalnithe
30a1, 2pl. comalnid 8d26, with infix. pron. 3sg. n. nachomalnid 25c24, and
with emp. pron. 2pl. nachomalnidsi 15a7, with infix. pron. 3pl. noscomalnid
25b1; pass.: pres. subj. 3sg. manicomollnither 2c17, rel. aracomalnither
15a19, perf. act rocomalnither 26a15, past subj. 3sg. rel. conocomalnithe
3d26 (= 3d25, Thes. Suppl. p. 38 ), conocomalnide 19b22, Milan: pres. ind.
2sg. comallaidi 94b1, 3pl. comallait 146a1, neg. nacomallatar 94b4, pass.
sg. comalnithir 128d11, impv. 2pl. comallaid 115c2, impf. 3sg.
nochomallad 36a21, pres. subj. 1sg. manucomallainn 131d19, 3sg. man
comalla 94b10, 1pl. with infix. pron. 3sg. masc. dian d comallammar
46c20, 2pl. chomallaide 95c3, past subj. 2sg. chomallada 70c13, 3pl.
nocomallaitis 95c2, pass. sg. nochomallad 35c12, fut. 3sg. comallaibther
117c3, rel. with infix. pron. 3sg. nudcomlnabadar 46c20, 3pl. acomallaibte
121b8, pass. sg. comallaibther 89b11, pl. archomallaibtir 109c09, sec. fut.
pass. sg. nocomallaibthe 105b14, perf. 1sg. rocomallus 74d5, 3sg. with
infix. pron. 3sg. rondchomallastar 122d7, 3pl. neg with infix. pron. 3pl.
nruscomallas(atar) 105a6, pass. sg. rocomallad 81d5, pl. rocomallada
74c20, rocomalnada 44d31, rel. roncomallada 38c9, verb. nec. comallaidi
62c5, VN comalnad, comall, comallad 14c14, comallada (?) 122d4, acc. sg.
chomallad 70c15, 76a15, comallad 94b3 (2x), dat. sg. chomallad 39a12,
46c6, 95c3, chomalnad 77a12.
Deadj. < comlann (equally) full, complete. Motivation for
formation as a deponent verb has been discussed in section 3.3.1.3, pages
63f.
LEIA C-165, 175, DIL s.v. comalnaithir 349: 9, comlann 387: 20,
ORahilly 1942: 173-6, McCone 1997: 41.
condelca compares
Cf. G pres. ind. 1sg. delgu hold, MW daly capture, retain, kynnal
hold, hold together, MB delcher keep, take (Schumacher).
122

Wrzburg: fut. 1pl. with infix. pron. 1pl. cotondelcfam 17b10,
Milan: pres. ind. pass. pl. delgatar 72b9, pres. subj. 1sg. rel. with infix.
pron. 1sg. (refl.) diandamchondelc 91d8, perf. 3sg. rudelc 55d3, with infix.
pron. 3pl. cotarodelc 55d8, verb. nec. condelcthai 88d13.
Denom. < coindelg in the meaning of comparison. This is the VN
of a Celtic verb *delg-e/o- < PIE *delh- hold together, not attested in Old
Irish. The primary verb is attested in Brittonic, the root cognates are given
above. It is not impossible that condelca was originally a primary verb that
has taken on weak flection, but note that in most cases where primary verbs
have taken on weak flection, reasons can be given (most often nasal presents
which become weak through regular phonological changes or a verb in a
different tense or mood that has been reinterpreted or re-formed as a present
form of a weak verb). There seems to be no underlying motivation for a
primary verb going weak in this case. (See chapter 6, pages 259ff. for an
overview of the primary verbs gone weak in the Wrzburg and Milan
Glosses.)
LEIA C-150, 190, LIV2 113, DIL s.v. con-delca 432: 35, coindelg
310: 33, Vendryes 1955-6: 13-4, Schumacher 2004: 271.
consecra consecrates, dedicates

Milan: fut. 1sg. cosacrub 45b12, 3pl. rel. secrabat 66a6, part.
consecarthae 36c8, 124b2, secarthe 49a11, cossecarthae 100b11, nom. pl.
cosecarthi 21d2, 100c16-17, cossecarthi 137a4, cossecarthae 74d6, gen. pl.
cossecarthae 56a7, 92d5, superl. cossacarthumem 50c16, VN (m u)
cosecrad, dat. sg. chosecrad 48b11, chossecrad 48d27.
Lat. consecrare consecrate, dedicate, treated as a compound in Old
Irish, cf. adraid worships (page 110) which is another Latin loan that is
treated as a compound in certain cases. For further discussion on the
formation as either compound or simple verbs, see section 3.5, pages 74f.
Latin loans are discussed in section 3.2, pages 48ff.
Ped. ii 496, LEIA C-199-200, DIL s.v. con-secra 461: 3.
123

crechaid plunders, spoils
docrecha (excogitates, devises): Milan: pres. ind. 1sg. docrechub
126c23, 3sg. duchrecha (Note Thes.: MS duchrech) 85b15, 3pl.
dochrechat 22b5, ducrechat 28c9, 30b6, duchrechat 85b3, perf. 3sg.
durochrech 68c11, 3pl. durochrechsat 47d15, VN (m u) tochrechad 55c13,
acc. sg. tochrechad 103d8, dat. sg. tochrechad 19a, acc. pl. tochrechthu
26d4.
Denom. < crech plunder, booty.
LEIA C-225-6, DIL s.v. crechaid 514: 62, do-crecha 233: 20, crech
513: 44.
crochaid crucifies, hangs

Wrzburg: perf. 3pl. rochrochsat 5c11; pass.: past subj. 3sg.
nocrochthe 19b6, Milan: pres. ind. 3sg. rel. with infix. pron. 1sg.
nudamcrocha 32d28, perf. 3pl. (rel.) rocrochsat 24d4, rochrochsat 25b5,
part. crochthae, gen. sg. crochta 16c5, VN (m u) crochad, dat. sg. crochad
86d19.
Denom. < croch cross, < Lat. crux cross. Direct and indirect
borrowings from Latin have been discussed in section 3.2, pages 48ff.
LEIA C-242, DIL s.v. crochaid 541: 30, croch 540: 17.
crothaid shakes, causes to tremble

crothaid: Milan: VN crothad, dat. sg. crothad 63d8, focrotha
(makes quiver, shakes): Milan: pres. ind. 2sg. rel. nafochrothai 64a4, pres.
subj. pass. focrotar 129a10, past subj. 3sg. focrothad 64a6, cofucrothad
23b14, pass. sg. focrothtae 68a5, part. (DIL gives this as part., but with a
question mark) fochratae, dat. pl. fochrataib 80c11.
DIL refers to crod, act of destroying, destruction, but this seems
unlikely. Under the PIE root *kret- shakes, Pokorny mentions both
crothaid and a Gallo-romance *crottiare. *crottiare has been reconstructed
by Von Wartburg on the basis of Catalan gronar to rock (a baby to sleep)
(cf. French bercer < *bertiare), Occitan crossar id., Bern dialect crouchier
id., French (regional?) grotter id.. *crottiare appears to be a Vulgar Latin
verb based on a Gaulish causative *krot--. However, pace Schumacher, OI
124

crothaid does not reflect an inherited causative/iterative, because it is AI. I
propose that this verb is derived from a nomen actionis created using the
o-acrostatic form of the root *kret- declined as an o- or -stem: *krtos
(/-on/-). This noun is itself not attested in Old Irish, but the existence of the
root in Celtic is corroborated by for example OI crith act of shaking,
trembling, W cryd trembling, ague. Moreover, a denominative from this
root would result in Old Irish as crothaid and it is certainly possible for a
verb to survive while its base gets lost (cf. focridigedar, page 210). This
implies that this verb is old: certainly old enough for the base noun to
disappear from the language. Crothaids relationship to the scheme argued
in this thesis is discussed on page 60 and can be found in the chronology on
page 80 with discussion on page 82.
IEW 620-1, LEIA C-247-8, LIV2 370-1, DIL s.v. crothaid 552: 19,
fo-crotha 208: 54, crod 542: 55, crith 534: 22, Von Wartburg vol. 2, 1366a,
Schumacher 2004: 129, 419-20.
crachaid piles up, stacks up (LEIA)

Milan: pres. subj. 3sg. arnacruacha (Note Thes.: MS. arnacrucha)
gl. turgescat 68c7.
Denom. < crach, stack of corn, rick. The verb is not translated in
DIL, but Thes. translates the gloss as that he may not swell. As has been
discussed on page 61, although crachaid expresses an iterative action
(piling up), it is not iterative in relationship to its base (making stack) and is
therefore not AII. The verb only occurs twice outside of the Milan Glosses.
LEIA C-249-50, DIL s.v. crachaid 555: 14, 1. crach 554: 46.
crsagaid corrects, chastises

Wrzburg: perf. 1sg. rel. with rel. n and infix. pron. 3sg. m. and emp.
pron. 1sg. rondcrsagusa (= -gussa) 19a6, 3sg. rel. with infix. pron. 3sg. n.
rodchrsach 14d18, perf. pres. subj. 3sg. rocrsacha 4a2, Milan: impf. 3pl.
with infix. pron. 1sg. nomchuarsachtis 54c15, VN crsagad, cursachad
55c10, acc. pl. chuarsachada 54c14.
125

Lat. curas agere take care of, take in hand, the -ch- in the Old Irish
forms is due to modelling on maldachaid, bendachaid etc. (Thurneysen).
Loans from Latin have been discussed above, section 3.2, pages 48ff.
LEIA C-296-7, DIL s.v. crsagaid 636: 14, Thurneysen 1918: 4123, GOI 128.
damnaid condemns, punishes

Milan: pres. pass. sg. rel. huandamnither 57d8, VN damnad, dat. sg.
with poss. 3pl. dianndamnad 140b11.
Lat. damnare pass judgement, find guilty. Latin borrowings have
been discussed above, section 3.2, pages 48ff.
LEIA D-21, DIL s.v. damnaid 66: 52.
dechraid marks out, distinguishes

Wrzburg: pass.: pres. subj. 3sg. man dechrther (Editors:
dechrigther) 12c46.
Denom. < dechor the condition of being different, difference. DIL
specifies that the verb is an -derivative; Pedersen and LEIA derive the
noun dechor from *di-cuir-, which is only attested as VN.
Ped. ii 500, LEIA D-32-3, DIL s.v. dechraid 199: 47, dechor 196:
19.
delbaid shapes, forms

Cf. W delwi stop, freeze, delw statue (LEIA).
Milan: perf. pass. sg. rudelbad 74a11.
Denom. < delb, form, figure. Whereas the W delwi and the OI
delbaid are formally exact cognates, it is clear from the semantic fields of
these verbs that each has been coined independently (see also page 76). The
W delw that delwi is derived from, means statue. The verb is formed from
that specific meaning and means stop, freeze, become motionless, i.e.
become like a statue. The Old Irish verb, on the other hand, has the wider
meanings shapes, forms that correspond to the wider meanings of the noun
delb shape, form that this is derived from. Moreover, Old Irish has several
synonyms for the concept of shaping and forming. These verbs have
126

therefore most likely been coined within the individual languages. In the
Chronology, delbaid has therefore been assigned to Old Irish (page 82). The
semantics of delbaid have been discussed on page 92.
LEIA D-47, DIL s.v. delbaid 18: 10, delb 17: 28.
derbaid certifies, proves

Wrzburg: pres. ind. 2pl. with rel. n: annonderbid gl. Probantes
22b24, Milan: pres. ind. 3sg. rel. nderbas gl. adprobare 35a2, pass. sg.
derbthair gl. adprobatur 14b7, VN derbad gl. probationem 19d17, dat. sg.
derbad gl. probationem 16a8.
Deadj. < derb sure, certain.
IEW 215, LEIA D-54, DIL s.v. derbaid 31: 42, derb 30: 19.
dnochtaid lays bare, uncovers

Cf. W dinothi lays bare, uncovers.
Milan: pres. ind. pass. sg. dnochtar gl. denudatur 54d23.
Deadj. < nocht, naked, bare. The verb is hapax legomenon. The
Welsh cognate cannot be taken as an argument for age, since besides being
hapax, this is too common a concept. It could have been formed in any stage
of the language. It has therefore not been assigned to a specific language
phase in the Chronology (section 3.7, pp. 77ff.). It could also be seen as a
calque on Lat. denudare strip, but the same constraint applies that it is a
very common concept. Calquing has been discussed in section 3.4.3, pages
69ff.
LEIA D-93, DIL s.v. dnochtaid 131: 6, 2. nocht 53: 70.
*drgidir directs, aims

condrgedar (reproves): Milan: pres. ind. 3sg. condrgedar
34a23, 46a12, fut. 3pl. condrgebadar 130c15, part. (io, i) cuindrichthe,
gen. sg. / nom. pl. chuindrichthi 57d5, VN cuindrigud, nom. pl.
cuindrigthea 41d1, condrig (checks, controls): Milan: VN cuindrech
22c8, dat. sg. cuindriug 49b7, tremidrgedar (refers, directs): Milan:
pres. ind. 3sg. rel. trmedirgedar 54a22.
127

Deadj. < driuch straight, direct. In Old Irish, drgedar has a
mixture of AI and AII forms, cf. the pres. ind. 3sg condrgedar and the fut.
3pl. condrgebadar above. Both CC *d-reg--tor and CC *d-reg--tor
would have given OI drgedar, leading to confusion in the Old Irish
inflection. This is historically an AI verb (it must be, because it is
deadjectival, cf. p. 62), but looks like an AII verb. In fact, this could also be
an -igidir verb with loss of -ch- by haplology, but since there are other
alternatives, I have applied the principle of economy (cf. section 3.4.1, page
67).
*drgidir is not attested in Old Irish, and only as an active verb in
Middle Irish. It is however clear that this was the simplex of the compound
verbs attested in the glosses. DILs first meanings straightens, makes
straight do not show the subject involvement expected in a deponent verb
(cf. section 3.3.1.3, page 43, footnote 10, and see page 63), but the meanings
directs, aims, guides (DIL s.v. drgid 135: 36) do.
Ped. ii 595-6, LEIA C-191, LIV2 304, DIL s.v. con-drgedar 434:
40, driuch 137: 10, tremi-drgedar 295: 65, drgid 135: 36.
dlomaid announces, proclaims

Milan: pres. ind. 3 sg. rel. dlomas 30b19, pass. sg. dlomthair 59d7.
Denom. < dlom proclamation. LEIA refers to ODav. 697 dlom .i.
foillsiugud. DIL has this reference under the verb dlomaid and does not
register the noun itself, but has all the occurences under the verb.
LEIA D-108, DIL s.v. dlomaid 164: 66.
dlthaid compresses, condenses

Milan: pres. ind. 3pl. rel. dlutair 116a4, pass. sg. dlutair 74c21, pres.
subj. 3pl. codluthit 69d10, perf. 3sg. rel. rundlth 33a17, VN dlthad, acc.
sg. dluthad 144d4.
Deadj. < dlth, close, compact.
LEIA D-110, DIL s.v. dlthaid 169: 62, dlth 168: 32.
128

doecrathar covers, protects
immtecrathar (covers, protects): Milan: pres. ind. 3sg. rel.
immethecrathar 65a1.
Denom. < tcar shelter, cover. No simplex attested. As has been
discussed above (page 63), the motivation for formation as a deponent verb
is greater subject involvement. Since doecrathar always has the
connotation of protects, the subject clearly has a direct interest in the
action.
Ped. ii 501, IEW 1068, LEIA D-134, DIL s.v. do-ecrathar 250: 37,
imm-tecrathar 155: 22, tcar 95: 16.
dota clothes
Wrzburg: past subj. 3sg. with rel. n: donetadsom (Note Thes.: leg.
do-n-ethad-som) gl. Oportet enim corruptibile hoc induere incorruptionem
13d20.
Pedersen and Wb. Lex. take the above form donetadsom as coming
from a verb *dota clothes. DIL registers the form under do-etha goes to,
visits and does not list a verb dota. There is no Old Irish noun or adjective
to derive a verb *dota from, but cf. tgud covering, clothing, tiud
raiment, covering and tach covering, raiment, which all contain this
element t-. tach is given by DIL as VN of intuigethar covers, clothes.
Either the VN was morphologically related to this verb, or it came to be
used as the VN of this verb. In the first case it is possible that the element t
was reinterpreted as a noun in its own right of which a verb could be
derived. In the second case tach is itself derived from a noun or adj. *t. In
both cases, there is certainly a possibility for a verb *dota as Wb. Lex.
proposes. Moreover, the later verb taigid also looks like it was derived
from *t. The etymology of *t itself, however, remains unknown.
Both the Thesaurus and DIL emend the Wrzburg gloss above to don-ethad-som, as if it came from the verb doetha goes to, visits.
Considering that this requires an emendation that does not even make sense,
that analysis does not hold and I have followed Pedersen and Wb. Lex.s
suggestion for the verb dota.
129

Ped. ii 514, DIL s.v. do-etha 261: 33, tgud 235: 61, tiud 244: 3,
tach 208: 29, in-tuigethar 293: 73, taigid 209: 34.
dogatha deceives, dupes

Wrzburg: pres. ind. 3sg. with infix. pron. 1sg. nimthogaitha 4b27,
perf. 1pl. niruthgaitsam 16a22, perfect. pres. subj. 3sg. nitorgaitha 25b5,
pass. perf. pres. subj. 3sg. with infix. pron. 1pl. conintorgitar 14d27.
Milan: pres. ind. 3sg. dogaitha 31a13, (rel.) dugitha 55c13, pass. pl.
dugaithatar 31c8, 31c25, honuntotogaitar 32a6, impf. 3pl. dungaitis 31c20,
pres. subj. pass. pl. dungaithatar 54a30, fut. 3pl. dugaithfiter 54a26, perf.
3sg. nimthorgaith 38a13, part. (io, i) togaethae, togaitae 40b11, VN (f )
toges, togis 28b17, acc. sg. togais 28c14, 30d18, 75d2, dat. sg. thogais
33d19, thogais 38a13, 74a7, togais 61a6.
Deadj. < geth, wise, intelligent, shrewd.
LEIA D-147, DIL s.v. do-gatha 281: 21, geth 13: 15.
doraid debases, takes captive

Milan: impf. 3pl. with infix. pron. 1sg. nondoirtis 34a7, past subj.
3pl. with infix. pron. nondoirtais 28c18, part. doirthe, nom. pl. doirthi
34d10.
Deadj. < dor, unfree, low-born.
LEIA D-158-9, DIL s.v. doraid 322: 59, dor 320: 56.
dolbaid fashions, shapes

Wrzburg: perf. 3sg. with infix. pron. 3sg. m. rodndolbi 4c26,
Milan: impf. 3pl. nodolbtais 54c12, part. doilbthe, dat. pl. doilbthib 31a8,
61b13, VN dolbud, dat. sg. dolbud 24d8.
Denom. < dolb sorcery, illusion. The verb has meanings that
include the fashioning of something magic or the fashioning of something
by magic (DIL 327: 37), but these meanings are not attested in the Glosses
and from the attestations given in DIL seem all later.
LEIA D-160, DIL s.v. dolb 327: 10, dolbaid 327: 32.
130

domolta goads
Wrzburg: pres. subj. 2pl. with infix. pron. 1sg. nmthomoldid 20d4.
Denom. < molad act of praising, praise, VN of molaithir (page
143) (LEIA). Hapax legomenon. Simplex not attested.
LEIA D-168, DIL s.v. do-molta 343: 76, molad 161: 47.
*donaid consoles
Cf. W diddan pleasant, nice, pleasure, diddanu comforts, cheers
up, Corn. dythane delights, is delighted (Ped., LEIA).
*donaid: Wrzburg: VN donad 25c3, Milan: VN donad, dat. sg.
donad 86d8; dodona (consoles, comforts): Wrzburg: pres. subj. 3pl. rel.
dodonat 26d21, impv. 2pl. with infix. pron. 3sg. m. dandonid 14d21; pass.:
perf. 3sg. with infix. pron. 1pl. and emp. pron. 1pl. dorondonadni 16b17,
Milan: pres. ind. 3sg. neg with infix. pron. 1sg. nachamdidna 86d3, impf.
3sg. rel. with infix. pron. 1sg. dudamdonad 86d10, pass. sg. with infix. pron.
1sg. dumdontae 62c7, 62c18, VN ddnad, didnad 59a15, ddnad 126c15;
*comddnea (comforts, consoles): Milan: VN comddnad, acc. sg.
comdidnad 87d1; immdtnathar (releases, exempts): Wrzburg: fut.
pass. sg. ni imdidnibther (MS imdidnibter) 15c25.
*donaid is back-formed from dodona. Dodona is a verb based on a
false interpretation of OI *diden as di-den. The Welsh cognate diddan
shows that it was in fact did-en. OI *diden is not attested but its existence
can be inferred from the compound deithiden care, concern. The
interpretation of di as preverb also shows that this verb was created before
the falling together of the preverbs di and do in deuterotonic position in Old
Irish, see also the Chronology, pages 81 and 83.
The deponent flexion of immdtnathar is unusual. No active forms
are attested for this compound.
Ped. ii 56, 508, LEIA C-170, D-132, 169, DIL s.v. donad 346: 54,
do-dona 242: 53, comddnaid 368: 82, imm-dtnathar 139: 14, deithiden 15:
57.
dotromma makes heavy, weighs down

Wrzburg: pres. ind. 3sg. with infix. pron. 2pl. dobtromma 8c8.
131

Deadj. < trom heavy. Hapax legomenon, simplex not attested.
IEW 1095, LEIA D-187, T-149-51, DIL s.v. do-tromma 384: 41,
trom 317: 17.
drbaid lingers, abides

*fodrba (delays, lingers): Milan: perf. 3sg. rel. forrudrb gl.
inmoratus 49b10.
Denom. < drb delay, staying.
LEIA D-202, DIL s.v. drbaid 408: 28, fo-drba 218: 6, drb 408:
9.
dnaid shuts, closes

dnaid: Milan: part. nom. pl. f. dntai 46a9, fordna (shuts,
closes): Milan: VN fordnta, dat. sg. (?) forduntu 34a22, frisdna (shuts
up, blocks): Milan: pres. ind. pass. pl. frisduntar 14c7, impv. 2sg. frithdn
54b7, perf. 3pl. frisrodnsat 22a2, 39d4, part. frithdntae 23b9.
Denom. < dn fort. Watkins argues convincingly that the noun
originally had the broader meaning of enclosure. The Celtic word *dnowas at an early period borrowed into Germanic with the meaning
enclosure, from which the meaning of town develops, e.g. OE and ON
tn (from which OE tnan close, enclose) and German Zaun (p. 451).
Watkins connects OI dn with Lat. fnus funeral and Hittite tuuta it is
finished (p. 452) and proposes an etymology *dheuh2-, *dhuh2- finish,
come to and end, come full circle (p. 453). This develops quite naturally
into a meaning enclosure for dn, which itself only retains the meaning
fort in Old Irish. The verb dnaid, then, will have been created prior to
Old Irish, when dn still retained its wider meaning of enclosure, see also
Chronology page 81.
IEW 260, 263, DIL s.v. dnaid 450: 82, dor-dna 332: 6, fris-dna
426: 8, dn 449: 23, Watkins 1991: 451-3.
draid hardens
Milan: pres. ind. pass. pl. duratar 132c5 (?DIL), part. drtha 84b8.
132

Deadj. < dr, hard, rigid. LEIA has dr as a loan from Lat. durus
hard, stern; DIL suggests that with a question mark, but I see no problem
with it. The verb could however also be a direct borrowing from Latin
durare harden. Direct and indirect borrowings have been discussed in
LEIA D-223-4, DIL s.v. dr 452: 49, draid 453: 74.
emnaid makes double, doubles

Milan: VN emnad 76d2.
Denom. < emon a pair or triplet. As discussed above, page 60,
although the medieval attestations of emnaid are all transitive, it lends itself
to ambitransitive reading. The intransitivity of the modern verb eamhnaigh
supports this argument.
IEW 505, DIL s.v. emnaid 121: 13, emon 121: 52.
fgaid looks at, observes

Milan: pres. ind. pass. sg. nadfegar 36a38.
Deadj. < fig of sight, seeing, keen-sighted. The etymology is
complicated. The origin for the base adjective is unknown and the
relationship between base adjective and verb is unclear. This is the only
adjective in Old Irish from which the verb could conceiveably have been
derived, but it is an awkward derivation. It does, however, follow the
pattern. It is a deadjectival, so it is in correspondence with the other verbs
that it is AI. It is however not a factitive, which is unusual (see section
3.3.1.2, page 62). The relationship to the primary verb doccai (page 334)
is obscure (DIL), but it is likely that fgaid and doccai became confused at
an early stage.
Ped. ii 490, 611, DIL s.v. fgaid 59: 34, fig 65: 55.
feraid grants, affords

feraid: Wrzburg: pres. ind. 3sg. ferid 4b19, perf. pres. ind. 3sg. rel.
rofera 29b13, rel. with infix. pron. 1sg. romfera 24b15, perf. pres. subj. 3g.
rel. with infix. pron. 1pl. \dinroirea 29b16, 3pl. rel. \riret 29a9, Milan:
pres. ind. 3pl. with ro: romferat 22d5; confora (provides; in Wb. Lex.
133

forea): Wrzburg: pres. ind. 1pl. with emp. pron. 1pl. aconfoiremni 8b13,
Milan: pres. 2sg. confoirsiu 114c4, perf. 3pl. conforoirisset 69a9, verb. nec.
comfoiride (leg. comfoiridi) 64c4; dofera (assails): Milan: pres. ind. pass.
doferthar 138a9; fofera (causes, produces): Wrzburg: pres. ind. 3sg.
fofera 2a17, rel. with petrified infix. pron. 3sg. n. fodera (= fo-d-[f]era)
3c33, 34, 5d5, 14c42, and with infix. pron. 3sg. n. as object fuddera (= fu-dd-[f]era) 33c12, 1pl. with infix. pron. 3sg. n. faeram 15d3, 3pl. nifiret
27d24, fut. 3sg. rel. foirfea (= fo-firfea) 11d3, perf. 3sg. furuar 2a18, rel.
foruar 8b5, 14c43, with infix. pron. 3sg. n. fodruar 15a15, past subj. 1sg.
rel. with emp. pron. 1sg. arnafirinse 14d3, 3sg. rel. fuerad 33b13, Milan:
pres. ind. 3sg. rel. with 3sg. infix. pron. n. fodera 32a5, 55d11, 93a13, past
subj. 3sg. foired 43c3, fut. pass. sg. with infix. pron. 1sg. fomfirfider 33b10,
perf. 3sg. furuar 44b10-11, 3sg. with infix. pron. n. fodruar 20b17, part. (io,
i) foiride 39b7, dat. sg. f foiridi 27b11, nom. pl. foiridi 43d13, VN fuar
81c2; *remifoira (not in DIL, see part.): Milan: part. remiride, gen. pl.
remfoiride 26c5, dat. pl. remfoiridib 72c3, VN remuar, remuar 104b3.
Celtic derivative from the Proto-Indo-European for flowing water
*h2uer-. The element appears in a series of ancient place names in Britain
and on the Continent and a reflex of the element appears in Welsh gwr
tallow (see Isaac 2004a and b and cf. Isaac 2009). This makes it clear that
the root was productive in Celtic and therefore likely to have been common.
The Brittonic languages do not have a cognate verb, but there are other
arguments for ascribing this verb nonetheless to a common stage of the
language: a noun to derive OI feraid from is not extant in Old Irish, nor is a
cognate of this hypothetical noun extant in any of the medieval Celtic
languages. If this element was still as productive in the later stages of the
language as it had been in the earlier stages, we would expect it to occur in
many forms, also as the base noun of the verb. It does not. Other than W
gwr and OI feraid, there are no traces of it outside of toponomy. Lastly, the
semantics have undergone quite a shift. An understandable shift from
flowing water to grants, affords (pouring wine or mead, i.e. granting
wine or mead), but this is certainly not a direct link between noun and verb
as would be expected if the verb were created in Old Irish itself.
134

Ped. ii 517-8, IEW 80-1, DIL s.v. feraid 85: 1, con-fora 437: 84, dofera 264: 32, fo-fera 221: 80, remiride 42: 65, Isaac 2004a s.v. uaro-,
uero-, Isaac 2004b: 200-1, Isaac 2009: 54 n. 25.
figraid prefigures, foretokens

Wrzburg: pass.: perf. 3sg. rofiugrad 13a36, 15a31, with dem. rel.
anrofiugrad 15a34.
Denom. < figair figure, type < Lat. figura figure. The verb is
most likely not directly borrowed from Lat. figurare form, since that has a
different meaning altogether. For further discussion on direct and indirect
loans, see section 3.2, pages 48ff.
DIL s.v. fgraid 132: 38, 2. figair 130: 46.
fraid Makes true

Wrzburg: pass.: perf. 3pl. rofrtha 14c32, fut. 3sg. frfidir 13d23.
Deadj. < fr true, veracious.
IEW 1166, DIL s.v. fraid 149: 45, fr 146: 11.
fobotha is alarmed, frightened

Milan: past subj. 3sg. cofotabothad 33b16, VN (m u) fubthad 40c2.
This is a derivative of *butt of a root *bheh1u- pound (IEW,
LEIA). The Old Saxon verb unter-badon, Old Icelandic verb bauta, Lat.
verbs confutare, refutare and Irish verb balaid were formed from the same
root. All are secondary and the Latin and Old Irish stems are formally exact
cognates. The noun is no longer extant in Old Irish or anywhere else in
Celtic. I speculate that this common in the Celtic, Germanic and Italic
vocabularies and open to derivation at that time. While this verb was not
formed in Proto-Indo-European (otherwise there would have been more
exact cognates), I would argue that it is at least as old as Celtic and possibly
older.
Ped. ii 477, IEW 112, LEIA B-74, DIL s.v. fo-botha 184: 25.
forcenna puts an end to; brings to an end

Cf. W gorffen, forffennaf finish, end (Schumacher).
135

Wrzburg: perf. 3sg. rel. with infix. pron. 3pl. fodarorcenn (for
fordarorcenn) 11a27, pres. subj. 3sg. rel. arnafoircnea 20b13, Milan: pres.
ind. 3sg. forcenna 67b9, rel. forchenna 118d6, rel. neg. airnifoircnea
102a15, pass. sg. forcentar (this form not in DIL) 26b10, pl. forcennatar
48a15, foircniter 73d7, past subj. pass. pl. foircnitis (this form not in DIL)
54a18, fut. 3sg. neg n foircnibea 69b9, 3pl. forceinfiter 56c19, perf. 1sg.
with infix. pron. 1sg. forrumchennad 127c10, 3pl. forruchennsat (Note
Thes.: MS. forruchnsat which is what DIL cites) 100d9, pass. pl. neg.
nirrufoircneda 94c18.
Denom. < forcenn end, limit. The resulting verb is treated as a
compound, cf. Wb. 11a27 fordarorcenn above. Schumacher derives both the
Old Irish verb and the Welsh verb gorffennaf as denominatives from a
*or-kenno-m termination, which he dates back to Insular Celtic.
Considering that this verb exists in both Welsh and Old Irish, and that both
languages use the base noun as VN, I agree that it is likely that this verb has
been inherited and I have therefore dated it to my unspecified Celtic
category. The verb can be found in the Chronology on page 80. The verb
has also been briefly discussed on page 61.
Ped. ii 525, DIL s.v. for-cenna 320: 2, forcenn 319: 48, Schumacher
2000: 75, 142.
gataid takes away, removes

gataid: Wrzburg: pres. ind. 3sg. ngatda 10b1, fut. 1sg. with
interrog. ingt 9d4, sec. fut. 2pl. ngette 9c8, impv. 3sg. nagatad 22b6,
Milan: pres. ind. 2sg. neg n gati 106c13, VN (f ) gait 70a5, trisgata
(pierces, transfixes): Milan: pres. ind. 3pl. trisngatat 75b9, part. tregtae,
gen. pl. tragdai 77a19.
Denom. < gait taking away, abstracting (PIE *hazdho- rod,
stick), which is also used as the VN of gataid. The nom. sg. of this fem
-stem is palatal, because in verbal nouns the dative singular is often used
for the nominative singular. Gat is cognate to Lat. hasta spear, lance
(Thurneysen). The noun, and possibly the verb, would originally have meant
something like spearing or piercing, from which the meaning taking
away, stealing has developed (ORahilly).
136

Ped. ii 536, IEW 412-3, DIL s.v. gataid 49: 84, tregtaid 290: 52, gait
34: 26, Thurneysen 1936: 114 n. 4, ORahilly 1942: 168-9.
gellaid pledges oneself, vows

forgella (gives superior testimony, testifies): Wrzburg: pres. ind.
1sg. with infix. pron. 3sg. n. fortgillim 4b27, perf. 1pl. rel. with emp. pron.
1pl. forrogelsamni 25d20, Milan: pres. ind. 3sg. forceilla 64a2, 3pl.
fortgellat (GOI 423) 23c15, forgellat 87b15, past subj. 3sg. with infix.
pron. (cf GOI423) faridgellad 131d12, perf. 3sg. forrugll 97a12, VN
forgell, forcell 24d14, 126c5, forcal 42d4, dooggla (purchases, promises):
Wrzburg: perf. past subj. 1pl. with infix. pron. 3sg. f. manisdeirclimmis
26b16, Milan: perf. 3pl. with infix. pron. 3sg. darucellsat 126d7.
Denom. < gell pledge, surety. DIL gives dooggla a strong citation
form do-oggell, but its s-preterite shows that it is weak.
Ped. ii 537, 677, IEW 426, DIL s.v. gellaid 62: 53, for-gella 338: 64,
gell 61: 8, do-oggell 351: 45.
gallaid gives hostages, obeys

Milan: fut. 3pl. geillfit 41d13.
Denom. < gall, human pledge, hostage.
Ped. ii 537, IEW 426, DIL s.v. gallaid 78: 76, gall 78: 16.
glanaid cleanses, purifies

Wrzburg: perf. pres. subj. 3sg. with infix. pron. 3sg. m. (reflex.)
ranglana 30b19, impv. 3sg. with infix. pron. 3sg. m. (reflex.) nanglanad
(with the first n written over the g) 11d8; pass.: perfect. past subj. 3sg. rel.
roglante 31c18; with infix. 3pl. pron. 3sg. n. cerudglanta (for cerudglantae)
4a6, Milan: pres. ind. 3sg. coglanaid 51a10, rel. nglanas31 28b4, pres. subj.
2sg. with infix. pron. 3sg. manimglanae 71c19, perf. 1sg. with infix. pron.
31
The nasal here remains unexplained. The form glosses quae purificare who purifies. It
could possibly be a double marking of the relative with implied an although this is rare. It
could not be a double marking of the relative with nasalising relative clause, because the
Latin quae is nominative. In Old Irish, a leniting relative clause would therefore be
obligatory.
137

2sg. rutglanus 103a5, with infix. pron. 3sg. raglanus 91b8, VN glanad
71c19, acc. sg. glanad 31d4, 51a19, 105d4.
Deadj. < glan, clean, pure.
IEW 429, DIL s.v. glanaid 93: 53, glan 92: 25.
achtaid cries out, makes lamentation

achtaid: Milan: pres. ind. 3pl. rel. aniachtaite 63d9, conchta
(groans together; transl. Wb. Lex.): Wrzburg: pres. ind. 3sg. chta 4a22.
Denom. < acht, a loud cry, a shout. achtaid is further discussed in
comparison with gledid cries out, on pages 59f.
Ped. ii 509, LEIA C-191, DIL s.v. achtaid 12:12, con-chta 435: 16,
acht 11: 59.
ccaid pays, attones for

Wrzburg: pres. ind. 3sg. rel. ccas 30d7, ccass 31b12, hiccas
31b34, 3pl. rel. ccte 30a2, fut. 2sg. ccfe 10a9, rel. 10a10, perf. 3sg. with
rel. hoarcc 5c16, with rel. n roncc 28b1, with infix. pron. 1sg. and emp.
pron. 1sg. romccsa 28a12, with infix. pron. 1pl. ronnhcc 31d6, and emp.
pron. 1pl. ronnccni 21b8, rel. with infix. pron. 3pl. \darcc 5c13, past subj.
3sg. rel. with infix. pron. 3sg. m. nod ncad 2b4, perf. past subj. 3sg. rel.
with infix. pron. 3pl. condarccad 4d19; pass.: pres. ind. 3pl. rel. hccatar
4a8, fut. 3sg. ccfidir 25a30, rel. hcfider 4d6, with infix. pron. 2pl. nib
ccfither 20a11, perf. 3sg. rel. with infix. pron. 1sg. conromccad 28a10,
with infix. pron. 2pl. rob hcad 5c17, pres. subj. 3pl. rel. with prep. and rel.
tresanccatar 27c20, perf. pres. subj. 3sg. inrcthar (MS inrctar) 10a3, 4,
3pl. \rctar 5c3, past subj. 3sg. rel. dinadrcthe 28b1. Milan: pres. ind. 3sg.
caid 55d11, rel. with infix. pron. 3sg. nundaca 61b29, pres. subj. 2sg.
neg. with infix. pron. 3pl. cenisncae 88b7, pass. sg. rel. nad chonrcthar
(but this form is not in DIL) 33d10, past subj. 2sg. with infix. pron. 1sg.
conumctha 74c16, fut. pass. sg. interrog. incfider 96b11, pl. cfaitir 54a1,
pret. pass. 3pl. rel. ronictha 50b16, perf. 2sg. with infix. pron. 1pl.
rondanncais 89a6, pass. sg. rocad 50d15, part. acc. sg. nem nephcthe
76a17, VN (f ) cc, c 48b2, 58a9, acc. sg. c 33b5, 34d5, 57d11, gen. sg.
ce 23c7, dat. sg. c 23c9, 48d28, 58a4, cc 84c21.
138

Denom. < cc paying, payment, which is also used as VN of this
verb.
DIL s.v. ccaid 44: 80, cc 44: 11.
labraithir speaks, talks

Cf. W llafaru speaks, Corn. leuerel, Br. lavaret (Ped.).
labraithir: Wrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. n.
aridlabrathar (= arindlabrathar) 7d2, inti labrathar 12c19, int labrathar
14c29, 3pl. nilabrtar 12b22, fut. 1pl. nilabrafammar 12c4, pres. subj. 1sg.
labrar 12c36, 3sg. rel. labrathar 13a4, 3pl. labritir 13a5, with infix. pron.
3pl. cenuslabratar 12d28, Milan: pres. ind. 3sg. labraid 115a2 (2x) rel.
labrathar 35d22, 1pl. labraimme 31b15, 31b16, pres. subj. 1pl. labramar
31b23, past subj. 3sg. labrad 105d8, sec. fut. 3pl. nolabraifitis 55a10, perf.
3sg. anrolabrastar 126c10, with infix. pron. 3sg. rodlabrastar 126c10, VN
labrad 31b10, 35d26, 64c16, gen. sg. labartha 59a11, 89d3, dat. sg. labrad
31d10, arlabrathar (speaks for, on behalf of): Milan: VN airlabrae, dat.
sg. erlabrai 26c6, frislabrathar (speaks against): Milan: pres. ind. 3pl.
frislabratar 55a10, impf. 3pl. frislabritis 58d12, 61b24.
Deadj. < labar talkative, arrogant (DIL, Ped.). The motivation for
formation as a deponent verb is discussed in section 3.3.1.3, pages 63f.
Ped. ii 559, DIL s.v. labraithir 13: 58, ar-labrathar 401: 8, frislabrathar 429: 4, labar 12: 22.
lthraid arranges, disposes

Wrzburg: pres. ind. 1pl. with emp. pron. 1pl.: .i. lathrimmini 8d19,
Milan: pres. ind. pass. sg. latharthir 44b16, part. (not cited as such in DIL)
lthardae, latharth 32c2, VN alathar 114a9.
Denom. < lthar arrangement, disposition. DIL gives explains,
expounds as the first meaning of this verb, but the base noun shows that it
originally meant arranges, disposes.
IEW 806, DIL s.v. lthar 59: 46, lthraid 62: 6.
139

lgaid reads
lgaid: Wrzburg: perf. 1sg. rel. with emp. pron. 1sg. rollegusa
19a6, 2sg. with dem. rel. arrolegais 28d7, 2pl. rel. rolgsid 7b23, Milan:
perf. 1pl. with infix. pron. 3sg. n rundalegsamni, 2pl. nadroilgisid 17b18,
3pl. rolegsat 24d24, VN (n o) ligenn, gen. sg. legind 35a11, arlga (reads
out, aloud): Wrzburg: pres. subj. 2pl. rel. with emp. pron. 2pl.
conarlgidsi 27d14, pass.: pres. ind. 3sg. rel. arallegthar 9b3, 18a24, pres.
subj. 3sg. rel. rlgthar 27d13, co arlgthar 25c30, perf. pres. subj. 3sg.
arroilgither 27d13, Milan: perf. 3sg. -arleg 43b14.
Lat. legere read. Latin loans have been discussed in section 3.2,
pages 48ff. above.
Ped. ii 561-2, DIL s.v. lgaid 72: 79, ar-lga 401: 34.
lnaid fills, floods

lnaid: Wrzburg: perf. 3sg. roln 22b13, with infix. pron. 1pl.
ronln 20d11; pass.: pres. ind. 3sg. rel. lntar 11b13, Milan: pres. ind. 3sg.
anundlina 19b5, pass. pl., lntair 25a9, with infix. pron. 1pl. nonlntarni
18c3, impf. pass. sg. nolntae 32b6 (the Latin is pres.), nulintae 32b15, pres.
subj. 3sg. rolina 43c7, fut. pass. nomlnfider 33b6, rel. linfider 103a10, sec.
fut. 3sg. rel. nolnfed 25a8, 25a11, pret. pass. sg. lntae 63d12, perf. 3sg.
roln 64d16, part. lnta, lintae 25c16, 68a14, VN lnad, allinad 133c11, dat.
sg. linath 129d17, folna (fills up, makes up): Wrzburg: pres. ind. 3sg.
rel. with infix. pron. 3sg. n. fodlna 14b7, fut. 3sg. folnfea 12d14, pres. subj.
3pl. rel. folnat 24d25; pass. perf. pres. subj. 3sg. conrufailnither 1a9, Milan:
pres. ind. 3sg. folina 122d5, pass. follintar 123a10, pres. subj. 3sg. fulina
45c10, perf. 1pl. furoillissem 100d8, VN (m o, u) fuilled, acc. sg. (?) filliud
138a2, dat. sg. fuillned 26c6, fuilled 69b6.
Denom. < ln full number, complement.
Ped. ii 566-7, IEW 800, DIL s.v. lnaid 157: 51, fo-lna 268: 76, ln
154: 72.
lobaid decays, rots

Milan: pres. subj. 3pl. arnalobat 65c3, VN lobad 103c16.
Denom. < luib herb, plant.
140

DIL s.v. lobaid 172: 82, luib 237: 31.
logaid concedes, remits, in glosses obtains, acquires

Milan: past subj. 3sg. nulogad 39c5, sec. fut. 3sg. noloichfed 127a8,
perf. 3sg. with infix. pron. 3sg. m. rondlogad 24c23, part. loigthe, dat. sg.
loigthiu 48c3, VN logad, gen. sg. loichtho 62a19, dat. sg. dulogud 118a16.
Deriving this from lg value, equivalent is problematic, because
the loss of vowel length between base noun and verb remains unexplained
and because the relationship between base and verb is not altogether direct.
An -igidir verb with long -- would be expected. However, there seems to
be no alternative.
DIL s.v. logaid 185: 11, lg 182: 51.
lommraid shears off, strips off

Milan: perf. 3sg. rolommar 14b2.
Denom. < *lommar, a collective noun based on lomm bare, naked.
*Lommar is not attested, but cf. luma(i)r unfledged bird (?), down of
unfledged bird (?).
DIL s.v. lommraid 196: 14, luma(i)r 245: 26, lomm 193: 27.
malartaid spoils, ruins

Milan: pres. subj. 2sg. nimalartae 74d13.
Denom. < malart damage, injury, which is not attested in Old Irish,
but extant in Middle Irish. Malartaid and its place in the scheme has been
further discussed on pages 59f.
LEIA M-14, DIL s.v. malartaid 49: 46, malart 48: 66.
maldachaid curses
Cf. W melltigaw curses.
Wrzburg: pres. subj. 2sg. diammaldachae 5d23.
LEIA derives this as denom. < maldacht curse, malediction < Lat.
maledicto, but there is no explanation for the loss of -t-. I suggest a straight
borrowing from Lat. maledicere speak ill of. Latin loans have been
discussed in section 3.2, pages 48ff.
141

LEIA M-14, DIL s.v. mallachaid 51: 27, mallacht 51: 35.
marbaid kills, slays

Cf. W marw dead, Corn. marow, MB marf, maru, Br. maro
(LEIA).
Wrzburg: pass. pres. subj. 3pl. mamarbitir 4a13, Milan: pres. subj.
2sg. with uile: dianuilemarbae 77a12, fut. 2sg. neg. with infix. pron. 3pl.
with uile: nisnulemairbfe 77a15, pret. 3sg. idmarb 23b10, perf. 3sg.
nacharomarbsom 23b5, 3pl. romarbsat 56b6, VN marbad, gen. sg. marbtha
52 line 1632, dat. sg. marbad 45a17.
Deadj. < marb dead (LEIA).
IEW 735, LEIA M-19, DIL s.v. marbaid 60: 81, marb 59: 37.
mescaid mixes, dips

Cf. W mysgu, B meski (Ped.).
conmesca (mixes together, commingles): Milan: VN cummasc
34b4, gen. sg. cumaisc 58b3.
LEIA states that it is not clear whether mescaid is a denominative
verb or mesc a postverbal adjective, but that the second alternative seems
more likely. That however poses a problem for the origin of mescaid, which
according to Schumacher cannot be primary. Though Proto-Indo-European
had an -sk- present *mik-ske/o- (root *mei k-, LIV2 428-9), this has no direct
continuation in Old Irish according to Schumacher, who sees the verbal root
as the origin of the Old Irish adj. mesc < PIE *misko/-, of which mescaid is
then a deadjectival derivative. LIV2 has this as a primary verb that has taken
on weak flexion (428-9). However, PIE *mik-ske/o- would give OI
**meccaid (*mik-ske- > *mikske- > *mixske- > *mixke- > *mikke- >
**meccaid, Isaac 2007), rather than mescaid.
The problems are solved if both the Old Irish verb and its Brittonic
cognates and the adjective are taken as borrowings from Lat. miscere mix,
mingle and mixtum respectively. Latin mixtum would regularly give OI
32
Ml. folio 52 is just a fragment with Latin text on one side and Old Irish on the other. Due
to its size, no folio divisions are given in Thes.: the line numbers given in my corpus refer
to the line numbers of Thes. Folio 52 is found in Thes. on p. 164.
142

**mess but this has been reshaped to mesc in analogy to the verb. Latin
loans have been discussed in section 3.2, pages 48ff.
Ped. i 76, ii 577, IEW 714, LEIA M-42, LIV2 428-9, DIL s.v.
mescaid 113: 52, con-mesca 449: 53, mesc 112: 61, Schumacher 2004: 38
(see also note 20), Isaac 2007: 62-4.
messaid examines, judges

Milan: pres. ind. pass. sg. rel. mestar.
Denom. < mess, act of judging, judgement, VN of midithir
weighs, measures.
Ped. ii 578, LEIA M-42-3, DIL s.v. messaid 111: 44, mess 109: 53.
mthaid feeds up, fattens

Milan: pres. subj. pass. sg. with infix nndammetar 80a3.
Deadj. < mth, plump, fat, Note that LEIA does not register the
verb.
IEW 712, LEIA M-44-5, DIL s.v. mthaid 119: 27, mth 118: 41.
molaithir praises, extols

Wrzburg: pres. ind. 1sg. with rel. n and infix. pron. 2pl. and emp.
pron. 1sg. nondobmolorsa 14c18, 3sg. molid 4a2, rel. with infix. pron. 3sg.
m. didmoladar cch 16d1, nodmoladar fesin 17b21, fut. 1sg. rel. nomolfar
9a22, pres. subj. 2pl. with infix. pron. 1pl. cenonmolid 16a1, Milan: pres.
ind. 3pl. nimolat 130b6, impv. 3sg. molad 68a13, 2pl. molaid 114c15, pres.
subj. 3pl. comolait 51c10, past subj. 3pl. rel. arammoltis 102d1, fut. 3pl. rel.
molfait 69b1, sec. fut. 3sg. nommolfath (this form not in DIL, but in GOI
636) 94a14, pret. 3sg. rommolastar 126b16, perf. pass. sg. rel. with infix.
pron. 1sg. rundammolad 88a17, VN molad 24a4, 51c2, 55b14, 24a4, 82d11,
acc. sg. molad 39d22, gen. sg. molto 51d10, dat. sg. molad 24a3, 26b10,
33b8, 53b8, 53b13, 81a1, dat. pl. moltaib 83b9.
Denom. < mol a constantly reiterated refrain or noise. The verb is
deponent in Old Irish, but active forms occur in Milan and one even
inWrzburg above (cf. Cuv). Semantic motivation for formation as a
deponent verb is discussed above, section 3.3.1.3, pages 63f.
143

LEIA M-61-2, DIL s.v. molaid 161: 80, mol 161: 4, Cuv 1958: 6.
mraid makes great, magnifies

Wrzburg: pres. ind. 3sg. mrid 6a16, rel. mras 17b22, 3pl. rel.
mrate 6a9, Milan: pres. ind. pass. sg. with infix. pron. nundanmrthar
63c4, pres. subj. 2sg. with infix. pron. 1sg. nundammorae 70c11, perf. 3sg.
romorais 37a2.
Deadj. < mr great.
IEW 704, LEIA M-18, DIL s.v. mraid 170: 61, mr 166: 57.
mchaid covers over, presses down

Cf. W mwg smoke, Corn. mok ibid, Bret. mog or moug hearth
(LEIA).
Wrzburg: perf. 3pl. rumgsat 5a24.
Denom. < mch smoke, stifling vapour.
IEW 971, LEIA M-69-70, DIL s.v. mchaid 179: 60, mch 179: 55.
nertaid strengthens, confirms

Cf. W nerth strength, Corn. nerth ibid, Br. nerz ibid. Cf. also
the verbs W nerthu33 and nerthogi strengthen, Corn. nertha ibid, MB
nerzhat enforce, Br. nerza enforce (LEIA).
connerta (strengthens): Wrzburg: impv. 2sg. with infix. pron.
2sg. and emp. pron. 2sg. cototnertsu 30a9, nertaid: Wrzburg: pres. ind.
3sg. nertid 5d20, with infix. pron. 3pl. nosnerta 10c3, rel. with infix. pron.
1pl. nodonnertani 6d11, 14c35, 3pl. nertit 1d7, perf. 3sg. with infix. pron.
3sg. m. arranert (MS arnert) 2d1, perf. pres. subj. 1pl. rel. corronertamni
14b13, impv. 3sg. nertad 5d11; pass.: pres. ind. 1pl. with infix. pron. 1pl.
nonnertarni 14b13, fut. 3sg. nertfidir 10c9, Milan: pres. ind. 3pl. with infix.
pron. 3pl. nosnertat 133d6, perf. 3sg. arrunert 130b2, VN nertad 35c11,
53a18, 79b1, 126b2, dat. sg. nertad 85d12, dat. pl. nertaidib 115a5.
Denom. < nert, strength, might. The existence of a denominative in
W nerthu with congruent meaning makes it likely that this is a Celtic
33
Note that LEIA has this as nerthi, but for a verb of this shape a VN in -u is correct. derthi
is a substantive.
144

derivative (Chronology page 80). Corn. nertha, MB nerzhat and Br. nerza
are *-sag- verbs.
IEW 765, LEIA N-10-1, DIL s.v. nertaid 36: 75, con-nerta 452: 7,
nert 35: 25.
nobaid makes righteous, sanctifies

Wrzburg: pres. ind. 3sg. rel. nibas 19b14, 24a7, 28, fut. 3sg. with
infix. pron. 2pl. nibnibfea 13b19, perf. 3sg. rel. ronib 19b12, with infix.
pron. 2pl. and emp. pron. 2pl. robnibsi 19b12; pass. perf. 3sg. ronibad
2b26, 2c4.
Deadj. < nob holy, sacred.
IEW 760, LEIA N-20, DIL s.v. nob 56: 68, nobaid 58: 1.
notaid marks, denotes

*reminota (signifies beforehand): Milan: pres. subj. 3sg.
reminota 18c12.
Lat. notare observe, record, and / or denom. < not mark, sign,
Lat. nota mark, sign. Direct and indirect Latin loans are discussed in
LEIA N-22, DIL s.v. notaid 65: 60, remi-nota 43: 58, not 65: 39.
onaid fasts, keeps a fast

Wrzburg: pres. ind. 3sg. rel. inas 6b19.
Denom. < in fast, period of fasting, Lat. ieiunium fasting, fast
day (GOI). Latin loans are discussed in section 3.2, pages 48ff.
LEIA A-35, DIL s.v. an 133: 71, anid 150: 36, GOI 924.
ordnaid appoints, ordains

Wrzburg: pres. ind. 3pl. rel. with infix. pron. 3sg. m. nodnoirdnet
33c1, perf. 3sg. rel. with infix. pron. 3sg. m. rudanordan (Note Thes.:
rectius rudnordan34) 33c5, Milan: pres. ind. 3sg. rel. oirdnes 39d30, perf.
34
Note that this Old Irish sentence makes perfect sense as it is, although not in the context
of the Latin. I therefore accept the emendation.
145

3sg. roirdned 14a3, rohortan 107b2, part. (io, i) oirdnide 40a1, 43c10,
dat. sg. oirdnidiu 48b8, VN (m) oirdned, dat. sg. oirdned 73a18.
Lat. ordinare order, arrange. AI verb, but due to syncope patterns,
certain forms look like AII forms. Latin loans have been discussed in
DIL s.v. oirdnid 125: 10.
panaid punishes, tortures

Milan: pres. ind. 3sg. rel. pianas 115b15, VN (m u) panad, acc./dat.
sg. pianad 26d12
Lat. punire punish, inflict punishment. It is unlikely that this is a
denominative from pan punishment, pain, as LEIA suggests, since the
verb is both transitive and causative (causing punishment to the object) and
should therefore have been AII. Latin loans have been discussed in section
3.2, pages 48ff.
LEIA P-8-9, DIL s.v. panaid 185: 50, pan 184: 67.
pretaid plunders, spoils

Milan: pres. ind. 3pl. rel. pretae 134b10.
Lat. praedare pillage, despoil. This is a hapax legomenon, glossing
Lat. depraedantium. It seems unlikely that this was part of the wider
lexicon. Note that OI -t- for Lat. -d- is irregular. Latin loans have been
discussed in section 3.2, pages 48ff.
LEIA P-14, DIL s.v. pretaid 202: 21.
pridchaid preaches, teaches

cetupridcha (first preaches; not in DIL): Wrzburg: perf. 3sg. rel.
ceturupridach 26c4, pridchaid: Wrzburg: pres. ind. 1sg. pridchim 5c6,
pridchim 24d1, predchim 5b16, with emp. pron. 1sg. predchimse 1d9,
nopridchim 21c9, nopridchim 27c22, with emp. pron. 1sg. hre nopridchim
se 23b11, rel. nopridchim 11a15, diapridchimm 27c36, with emp. pron. 1sg.
nopridchimse 28b7, 29b3, with rel. n and infix. pron. 3sg. n. (elided) and
emp. pron. 1sg. amal nonpridchimse 30a21, with dem. rel. anopridchim
19d17, annupridchim 8d26, 3sg. pridchaid 13a22, rel. hore pridchas 7b15,
146

pridchas 12c23, praidchas (a prima manu) 12c27, apridchas 11c8, 1pl.
pridchimmi 8b12, 14c8, 15b11, with emp. pron. 1pl. pridchimmini 15b18,
nipridcham 15b17, rel. pridchimme 14d39, 40, 15b17, and with emp. pron.
1pl. pridchimmeni 12c6, 24d5, with dem. rel. apridchimme 13b14, and emp.
pron. 1pl. apridchimmeni 15b14, 2pl. nopridchid 23a15, 3pl. mapridchit
12d37, pridchit 24a29, 27a11, rel. pridchite 5a6, 13a15, 16d5, 29a12,
prechite 5a5, impf. ind. 3sg. rel. nopridchad 13a20, fut. 1sg. rel. nopridchob
17b20, 3sg. pridchibid 26a8, 3pl. with interrog.: inpridchabat (MS
hipridchabat) 13a13, perf. 1sg. cin ropridchos 17c1, ropridchus 23d18, rel.
rophroidech 10c20, rel. with emp. pron. 1sg. diarupridchossa 7b7,
ropridchussa 13b12, ropridchussa 20b4, 3sg. ropridach 27d3, with infix.
pron. 3pl. rospridach 21b9, 1pl. ropridchissem 24d4, rel. ropridchissem
24c17, with infix. pron. 3sg. n. pridchaisem 5a7, marudpredchisem 10d9,
and with rel. n (infix. pron. 3sg. n. elided) and emp. pron. 1pl.
ronpridchissemni
13b10,
with
dem.
rel.
and
emp.
pron.
1pl.
anropridchissemni 26b6, naropridchissemni 17b31, 2pl. rel. ropridchissi (=

ropridchissid) doibsom 24c17; pres. subj. 1sg. with emp. pron. 1sg.
pridchasa 10d23, manipridag 10d24, manipridach 10d25, 3sg. rel. pridchas
18c11, 25c24, pridches 23b24, with rel. n elided and infix. pron. 3sg. n.
nodpridcha 13a6, and with infix. pron. 3sg. f. nodapridcha 14d28, 1pl. with
dem. rel. apridchimme 31c11, apridchimme 15d27, with emp. pron. 1pl.
apridchimmeni 18c11, 2pl. with infix. pron. 3sg. n. and emp. pron. 2pl. cena
pridchidsi 15d6, nipridched (for nipridchid) 33d1, perfect. pres. subj. 2pl.
ropridchid 23b40; past subj. 1sg. nopridchinn 18c13, rel. conopridchinn
28a7, impv. 2sg. pridach 30d6, 31c20, predech 30a12; pass.: pres. ind. 3sg.
rel. pridchider 14c26, with rel. elided ipridchither 15b15, fut. 3sg. rel.
pridchabthar 26d6, perf. 3sg. ropridchad 13a21, 33, 19b6, ropridched
23a16, 25a41, rel. ropridched 25a40, ropriched 23a3, ropricad 7b12,
ropridchad 13a32, ropridchad 7d4, 27a3, ro pridchad 28a18, rel.
diarupridchad 8c17, with dem. rel. arrupridchad 14d23, arropridchad 18b7;
pres. subj. 3sg. pridchidir 10d27, 14d34, with dem. rel. apridchither 7c10;
nopridch... (? leg. nopridchad perf. pass. 3sg. or rel. impf. ind. 3sg. act.?)
147

14c34, Milan: pres. 3sg. rel. pridches 94b1, pridchas 131b11, impf. 3sg.
(rel.) with infix. pron. 3sg. nodprithched 42b18, fut. 1sg. noprithchib35 45a8,
3sg. rel. prithchibes 107a16, 108b14, perf. 3sg. (rel.) roprithach 50d17, rel.
ropridach 69d3, VN (f ) precept, preceupt 25a8, gen. sg. preceuptae 68c9,
dat. sg. preciupt 60c4.
Lat. praedicare proclaim, preach. Latin loans have been discussed
in section 3.2, pages 48ff.
LEIA P-14, DIL s.v. pridchaid 202: 65, ceta- 153: 58, 62.
promaid puts to the test, tries

Wrzburg: perf. 3sg. with rel. n and infix. pron. 3sg. n. and emp.
pron. 3sg. m. rondpromsom 4b20; pass.: fut. 3sg. promfidir 17b7, Milan:
part. (io, i) promthae, gen. pl. m. promthe 56b8, VN (m u) promad, acc. sg.
promad 54a16.
Lat. probare approve of, esteem. OI -m- for Lat. -b- is irregular.
Latin loans have been discussed in section 3.2, pages 48ff. above.
LEIA P-15-6, DIL s.v. promaid 207: 23.
rannaid divides, parts

Cf. OW guorenneu36 fractions, subdivisions, W rhan part,
rhannu divides, shares, MC ran and Corn. radn part, OB rannou parts,
MB rann part, piece, rannaff divides, shares, B rann (IEW, LEIA).
Wrzburg: perf. pres. subj. 1pl. cororannam 4a16, Milan: pres. subj.
pass. sg. nochis rantar 37d9.
Denom. < rann part, division. Despite the Brittonic verbal forms, it
cannot be claimed that these verbs are inherited. There are several
synonyms within Old Irish (cf. dlaid and its compounds below, p. 191) and
the VN formation follows the productive pattern. This has been discussed in
the Chronology chapter above, sections 3.6 and 3.7 pages 75ff.
35
This is the MS reading. Ascoli emends to notpridchib I will preach thee (Thes. p. 130 n.
i, see Add & Corr), but McCone (1985: 96) reads this as an early case of a fronted
prepositional phrase followed by relative nasalisation (contrary to common Old Irish usage
in which a relative would only follow fronted subject or object) and does not supply the
infix.
36
Note that this form is mis-cited by LEIA (R-7) as guorenniou. That is the expected form,
but the form found in the MS is guorenneu as given above.
148

IEW 817, LEIA R-7, DIL s.v. rannaid 13: 13, 1. rann 10: 62,
Lambert 2003: 110.
rlaid makes clear, manifests

Wrzburg: perf. 1sg. rel. rorlus 13a35, pres. subj. 3sg. rel.
ari(n)rla (= arindrla vel aridrla) 27b27, perf. pres. subj. rel. 1pl.
cororlam 27c21; pass. pres. subj. 3sg. manirltar 12d15, Milan: pres. ind.
3sg. rel. relas 24d20, pass. pl. nreltar (Note Thes.: MS. nretaltar) 27d7,
pres. subj. 1sg. conrrl 41d11-12, perf. 2sg. rorelais 50c15 (2x), 3pl.
rorlsat 117d2, adj. formation ril 42d7, neut. pl. rli 20c11, comp. reiliu
70b11, VN rlad, gen. sg. relta 60a1, dat. sg. relad 25b6, 93a5, 126c1.
Lat. reuelare show, reveal with loss of -u- after -e- (GOI). Latin
loans have been discussed in section 3.2, pages 48ff. above.
LEIA R-17, DIL s.v. rlaid 40: 14, GOI 922.
rindaid cuts, incises

Cf. W ryn cape, peninsula (only extant in placenames).
dofoirndea (expresses, signifies): Milan: pres. ind. 3sg.
dofoirndea 2d2 (2x), pass. sg. dufoirndither 86a2, pl. dufornditer 65c16,
perf. 3sg. tororan 29b8, VN (n o) trann, torand 65c16, fristirndea
(marks off, traces a limit, hapax legomenon): Milan: pres. ind. 3sg.
fristoirndea
67b9,
tremitirndea
(transfers,
hapax
legomenon):
Wrzburg: perf. 1sg. with emp. pron. 1sg. trimirothorn 8d26.

Denom. < rind point, tip. Rindaid is one of only two verbs that are
derived from i-stem nouns (the other one being samlaithir compares,
directly below), which is discussed on page 54. The relationship between
base and verb has been discussed on page 61.
Ped. ii 603-4, IEW 329, LEIA R-31-2, DIL s.v. rindaid 73: 76, dofoirndea 271: 31, fris-tirndea 431: 68, tremi-tirndea 296: 8, rind 71: 33.
samlaithir compares, likens to

Cf. B hevelout copies, immitates (Ped.).
Wrzburg: pres. ind. 1sg. with infix. pron. 3sg. n. nasamlur 3c6, fut.
1pl. with infix. pron. 1pl. nonsamlafammar 17b12, pass.: pres. ind. 3sg.
149

samaltir 8d22, perf. pres. subj. 3sg. cerusamaltar 34a4, Milan: pres. ind. /
subj. pass. pl. with infix. pron. 1 pl. condansamailter 63d7, VN samail
138c6, deintamladar (dissembles): Milan: pres. ind. 3sg. de intamladar
27d12, fut. 3sg. ni deintamlafa 69d8, disamlathar (dissembles): Milan:
pres. ind. 3sg. disamlathar 21b2, past subj. 3sg. disamlad 114c3, VN (m u)
damlad, dat. sg. diamlad 52 line 19, insamlathar (imitates, emulates):
Wrzburg: pres. ind. 3sg. insamlatharside 9a15, 1pl. with interrog. and emp.
pron. 1pl. inintsamlammarni 11b16, 2pl. with nch and elision of infix.
pron. 3sg. n. nch intsamlid 14a28, fut. 2pl. with infix. pron. 3pl. and emp.
pron. 2pl. atasamlibid si 5a13, perf. past subj. 2pl. \ro intsamlithe 17a13,
past subj. 3pl. with interrog. in intamlitis 5b20, impv. 2pl. intamlid 9a14,
Milan: impv. 2sg. na intamaltae 56b26, pres. subj. 2sg. insamailter 56b38,
n intamlae (active flexion) 56c13, 1pl. con tintamlammar 97c1, verb. nec.
intamaltai 79c1, VN (declension combining different stems, see DIL)
intamail 22c1, dat. sg. intamail 37c1.
Denom. < samail likeness, similarity. This verb is a root cognate to
the Br. hevelout. Samlaithir is the other of only two verbs that are derived
from i-stem nouns (the other one being rindaid cuts, directly above),
which is discussed on page 54. The motivation for formation as deponent
verb is discussed in section 3.3.1.3, pages 63f.
Ped. ii 611-2, IEW 903, LEIA S-21-2, DIL s.v. samlaid 51: 47, deintamladar 11: 39, di-samlathar 139: 49, in-samlathar 282: 72, samail 47:
32, McCone 1997: 41.
ssaid satisfies, feeds

Milan: pres. ind. pass. pl. ssatar 40a10, VN ssad 121d3, acc. sg.
sassad 129d16.
DIL derives this verb as a denominative from ss, snare, trap, but
also implement, instrument, means, device and fit, competent. Under
ss, DIL refers to sth, sufficiency, as much as one requires. Greene
explains that ss could be derived from sth with a semantic development
from sufficiency to that which is capable of dealing with, hence trap.
LEIA on the other hand states that ss snare, trap cannot be connected to
sth because the meanings are too different. According to them, ssaid is
150

not derived from ss, but from a non-attested noun *ss connected to sth.
Irslinger follows LEIA and adds that an st-formation such as proposed by
DIL and Greene (sth > ss) is morphologically unlikely, because these are
most often pairs of nouns and adjectives, rather than a pair of nouns.
However, she gives the exception of bs death > bath death. If there is
one exception to the rule, there is no reason to exclude the possibility of
another exception. Irslinger derives ss snare, trap from *seh2(i ) tie, bind
(LIV), but this root has been dropped in LIV2, in favour of *seh1(i ) let go.
Deriving OI ss from PIE seh1(i ) does not work semantically. Moreover,
although *seh1- could give OI ss if it were in the o-grade, that o-grade
would be difficult to motivate. I therefore suggest following Greenes
semantic pathway of ss and deriving the verb ssaid from the earlier
meaning sufficiency.
IEW 876, LEIA S-15-16, 26-7, DIL s.v. ssaid 63: 68, ss 62: 48,
sith 37: 72, Greene 1962: 111, LIV 471, Irslinger 2002: 424, LIV2 518.
scrbaid writes, writes down

Wrzburg: perf. 1sg. with dem. rel. anroscrbus 20c18, 3sg. roscrib
31d19, perf. past subj. 3sg. rel. with emp. pron. 3sg. m. roscribad som
26b31, with infix. pron. 3sg. n. rodscribad 27d16; pass.: pres. ind. 3sg. rel.
scrbther 27d13, 3pl. scribatar 17b2, perf. 3sg. roscribad 6c28, rel.
roscrbad 2d2, with prep. and rel. part. diaroscribad 3b20, 23a10,
diarroscribad
25d2,
ndroscribad
27d13,
Milan:
pres.
ind.
1pl.
nondascribam 35b1, pres. subj. pass. pl. rel. scribtar 87a13, perf. 2sg.
roscribais 74d13.
Lat. scribere write, compose. Latin loans have been described
above, section 3.2, pages 48ff.
LEIA S-54, DIL s.v. scrbaid 109: 54.
scrtaid examines, scrutinizes

Wrzburg: pres. ind. 2sg. neg. n scrti 1d22, pres. subj. 3sg. rel.
arascrta 31b10, Milan: impf. 1sg. noscrutain 91c1, perf. 2sg. neg n
roscrutais 80c9.
151

Lat. scrutari (scruto in some Christian sources) search, probe.
Latin loans have been discussed above, section 3.2, pages 48ff.
LEIA S-56, DIL s.v. scrtaid 114: 34.
seccaid stiffens, hardens

Milan: perf. 3pl. rosecsat 46a22, part. sectha, nom. pl. secthi 18b3,
secthai 93b6.
Deadj. < secc stiff, rigid < Lat. siccus dry or Lat. siccare dry.
Direct and indirect loans have been discussed above, section 3.2, pages 48ff.
The semantics of seccaid have been discussed on page 91.
LEIA S-60, DIL s.v. secc 122: 3, seccaid 122: 41.
selbaid possesses, holds

Milan: VN selbad, acc. sg. selbad 30c3, adselba (testifies, takes
possession): Milan: VN asseilbiud, dat. sg. asseilbiud 102d3, doaissilbi
(refers, ascribes): Wrzburg: fut. 1sg. with rel. n donaisilbub 7a10, perf.
pres. subj. 2sg. with infix. pron. 3sg. n. didtarilb 29c2. Milan: pres. ind.
3sg. (rel.) duaisilbi 51b12, pass. pl. duaisilbter 2b5, past subj.
codunnaissilbthe 110c4, perf. 3sg. duairilbed 117a6, ntirilb 36a36, with
infix. pron. nitharilb 49b3, 3pl. (rel.) duairilbset 46d10, (rel.) doairilbset
53b11, rel. with infix. pron. daairilbset 53b11, pass. sg. doairilbed 2b6,
verb. nec. taisselbthi 15d8.
Denom. < selb property, appurtenance. Selbaid is transitive, but
neither iterative nor causative, and accordingly AI. Doaissilbi, on the other
hand, is AII. Since it literally means assigns to the possession of (DIL 196:
42), there is a strong element of causativity in it: causing someone to
possess. This will have motivated the change in class (this has been further
discussed above, page 63). A similar process is seen in colaid makes thin,
grows thin and asrocholi defines, determines (page 120 above), where
the iterativity of asrocholi motivated its change in class.
Ped. ii 623-4, LEIA S-79-80, DIL s.v. ad-selba 71: 39, selbaid 165:
84, selbad 165: 32, do-aissilbi 196: 28.
152

slchtaid kneels, prostrates
Milan: pres. 3sg. rel. slchtas 115a3, 2pl. annuslechtaid 115a1, past
subj. 3pl. corruslechtais 89d15, fut. 3pl. slechtfait 89d14, VN slchtan, acc.
sg. slechtan 138a2.
Lat. flectere bend, bow. The borrowing of Lat. f- as OI s- shows
that this verb was most likely borrowed between lenition and the
delabialisation of /sw/ (McManus). Latin loans have been discussed above,
section 3.2, pages 48ff. The Old Irish verb only occurs in religious usage.
LEIA S-129, DIL s.v. slchtaid 266: 65, McManus 1983: 51-56.
soraid frees, liberates

Wrzburg: fut. 3sg. sirfed 32d13, with infix. pron. 1pl. nonsorfea
24c18, infix. pron. 2pl. and emp. pron. 2pl. nobsirfasi 11b4, perf. 3sg. with
infix. pron. 1pl. ronsor 24c18, and emp. pron. 1pl. hre ronsirni 2d14,
with infix. pron. 1sg. and emp. pron. 1sg. romsirsa 3d20, with rel. n and
infix. pron. 3sg. m. ronsir (for ro-n-dn-sir) 32d13, Milan: pres. ind. 2sg.
with infix. pron. 1pl. nachansoirai 93d10, 3sg. rel. with infix. pron. 1sg.
nundamsoira 62c6b, 3pl. rel. soirdae 75a2-3, pass. pl. rel. soirtar 75a4,
impv. 2sg. with infix. pron. 1pl. nonsoerni 46b26, nonsirni 136c11, pres.
subj. 2sg. soira 61c15, with infix. pron. 1sg. diarumsoirae 142b2,
manimsoirae 142b3, neg. with infix. pron. 1pl. maninsoerae 77d6, past subj.
3sg. conid soirad 24a17, pass. sg. interrog. with infix. pron. 1sg. in
damsoirthae 91d4, neg. with infix. pron. 1pl. maninsoirthae 96a10, fut. 3sg.
rel. with infix. pron. 3pl. nondasoirfea 27a6, with infix. pron. 2pl.
nobsoirfea 45d10, 3pl. interrog. rel. insirfetar 96b2, pass. sg. soirfithir
68d14, sec fut. 3sg. with infix. pron. 3pl. nondasoirfed 131c9, rel. neg. with
infix. pron. 3pl. nachasoirbed (with b for f, see GOI 635) 53d6, interrog.
with infix. pron. 1sg. in damsoirfad 90c19, pass. pl. nosoirfitis 91a10, perf.
3sg. rel. rosoer 60b16, inrusoer 33b23, with infix. pron. 3sg. m. nirasoir
44c19, rel. with infix. pron. 1sg. rodamsoer 48a21, rel. with infix. pron. 3sg.
rondser 52 line 2037, pass. sg. rosoirad 61d2, pl. rusoirtha 102d17, 124b7,
37
153

rosoirtha 102d17 (2x), 131c9, VN (m u) saerad, soirad 20a12, 30b18, 53d1,
61a37, 81a7, dat./acc. sg. soirad 19b2, 26d14, 55a7, acc. sg. soirad 33b21,
gen. sg. soirtha 45b2, 62b6b, soertha 55d6, dat. sg. soirad 33a13, 52 line
20, 67d2.
Deadj. < sor free.
LEIA S-16-3, DIL s.v. saer 10: 5, saeraid 12: 31.
techtaid has, possesses

Cf. W teithi characteristic qualities (LEIA).
Wrzburg: pres. ind. 3sg. tectid 26a6, inntecta 4c28, 2pl. with infix.
pron. 3sg. n. manudtectid 23c11, 3pl. nitectat 12b21, rel. nadtectat 10b5,
techte 2c11, perf. 3pl. rotectsat 1a9, pres. subj. 3sg. rel. techtas 12d41,
aratechta 28d22, 2pl. nochtectaidsi 93a9, past subj. 3pl. with infix. pron.
3pl. nstectitis 10b5, Milan: pres. ind. 2sg. (rel.) nitechtai 56b31, 56b33
(2x), 3sg. rel. techtas 37b19, 85a4, neg. rel. nad techta 94c3, 1pl. rel.
techtmae 74d4, 2pl. notechtaid 115b12, 3pl. neg. natechtat 94c3, 3pl. rel.
techtae 37b16, 56b26, impv. 2pl. techtid 115b9, pres. subj. 3sg. ciatechtid
56b30, 56b31, aratechta 122b7, with infix. pron. sg. na(c)hrthechta 56b33,
1pl. rel. techtmae 18d9, fut. 1sg. notechtub 59b3, perf. 3sg. rel. rothecht
37b27, with infix. pron. 3pl. nisrothecht 44b10-11, 3pl. rotechtsat 84c10,
rel. neg. nad roithechtsat 97d7, VN techtad, acc. sg. techtad 68c8.
Denom. < techt possessions, property.
IEW 1057-8, LEIA T-41-2, LIV2 618-9, DIL s.v. techtaid 101: 80, 3.
techt 100: 4.
trchtaid comments, annotates

Milan: pres. ind. 3sg. trachtid 48a6, 53d14, 55c14, 57c14, 57d17,
60b14, 63d6, 65b12, 67d8, 126c11, trachtaid 53d16, 55c6, 56b40, 60a14,
61c1, 84a1, 85b8, 94a13, 94a15, 94c5, 95c8, 103b12, 103b14, 103d26,
108c12, 111b25, 118c6, 122c6, 123c8, 123c11, 123c16, 128c5, 133a13,
133d9, 138c13, rel. forsatrachta 46c14, trachtait 51a14, ntrachta 74c12,
118d8, rel. thrachtith 93a21, perf. 3sg. rotracht 121a8, VN trchtad,
trachtad 25a4, dat. sg. trachtad 25a6, 25a7,78b7, dat. pl. trachtadaib (note
Thes.: MS. innatrachtaib) 35b22.
154

Lat. tractare treat, discuss. Latin loans have been discussed in
section 3.2, pages op pagina 48ff. above.
Ped. i 229, LEIA T-121-2, DIL s.v. trchtaid 269: 45, Bergin 1938:
223.
trebaid cultivates, tills

Cf. OW treb habitation, cultivated plot, OB treb gl. tribus, ModC
trev town.
adtreba (inhabits, dwells): Wrzburg: pres. ind. 3sg. rel. atreba
16a17, atreba 21c7, inatreba 27d26, perf. 3sg. adrothreb 27a12, rel.
atrothreb 26d3, fut. 3sg. atrefea 30b18, perf. pres. subj. 3sg. rel.
corroaitreba 6b3, Milan: pres. ind. 3sg. attreba 51d28, pres. subj. 3sg. (rel.)
adtrebea 35b24, fut. 3sg. (rel.) trebfea 36a19, 36b14, atrefea 107a15, rel.
with infix. pron. adidtrefea 107a15, perf. 3sg. adrothreb 17b7, (rel.)
adruthreb 92c4, 113c2, VN attrab 92a9, atrab 17b5, 66d4, dat. sg. atrub
126c12a, 127d14, 133c8.
Denom. < treb house, farm, a word of native origin, but which
could have been influenced by the Latin tribus in the meaning of
household, tribe (DIL). Note that IEW does not give this as a
denominative. Moreover, IEW splits trebaid (< *terh1- drills, rubs, cf.
LIV2) from adtreba (< *treb- building, house), which is unnecessary.
IEW 1071, 1090, LEIA T-126-8, LIV2 632, DIL s.v. trebaid 282: 12,
ad-treba 74: 56, treb 280: 69.
trothaid subdues, abates

artrotha (checks, restrains): Milan: impv. 2sg. erthroid 92c11,
pres. subj. pass. mani erthroitar 134d3, past. subj. 3sg. with infix. pron. 3pl.
condaarthroithad 104a6, fut. 2sg. artroidfea 134d3, perf. 3sg. arrothroithad
121d9, part. erthroitae 49b8.
Deadj. < treth neuter (LEIA).
LEIA T-122, DIL s.v. traethaid 270: 44, ar-trotha 419: 65, traeth
270: 3.
155

araid grows or becomes cold or cool
etarara (cools, hapax legomenon): Milan: pres. ind. 3sg.
-itirnuara 71b5.
Deadj. < ar cold, cool. The semantics of araid have been
discussed on page 91.
LEIA U-8-9, DIL s.v. araid 37: 34, etar-ara 224: 52, ar 28: 73.
156
5.3 AII verbs

acarbaigidir makes rough
Cf. W garw, agarw.
Milan: pres. ind. 3pl. rel. acarbaigte gl. asperantia 87b14.
Deadj. < acarb (o/a) very rough, very fierce, < ad + garb. The verb
is hapax legomenon.
LEIA A-9, DIL s.v. 1. garb 44: 7, acarb 14: 5, acarbaigidir 14: 29.
adamraigedar wonders (at), marvels (at)

Wrzburg: pres. ind. 1sg. rel. (as a simple verb) noadamrugur 16c3,
3sg. rel. adamrigethar 5c16, pass. perf. pres. subj. 3 sg. rel.
conroadamrigther 12d29; Milan: pres. ind. 3sg. rel. adamrigedar 64c19,
3pl. adamraigetar 124a6, adnamraigetar 39b10, pres. subj. 2sg.
adnamraigther 56b39, 3pl. co adamraigetar 39b8, perf. 3pl. adruamraigset
88a18, verb. nec. adamraigthi 64c3, 69c1, VN adamrugud 26b10, 122c2,
128c3, acc. sg. adamrugud 61a15, dat. sg. adamrugud 90a5.
Deadj. < adamrae very wonderful < ad + amrae, but in the Old
Irish glosses treated as compound, except in Wb. 16c3 above (DIL). It is
likely that the one verb was simply treated in two ways. However, there is a
slight possibility that these are actually two verbs, one simplex derived from
amrae and one compounded with ad-. The issue has been discussed in
section 3.5, pages 74ff. Amrae is without certain etymology (LEIA). Note
that LEIA does not register the verb.
Ped. i 357, LEIA A-68, DIL s.v. adamrae 32: 6, 32: 20, amrae 311:
6, ad-amraigedar 32: 20.
adbartaigidir opposes
Cf. W gwerthyd spindle, OCorn gurhthit gl fusus, OB pl. guirtitou,
MB guersit gl fusis.
Milan: pres. ind. 2sg. adbartaigther gl. uersaris 44b31, 3sg.
adbartaigid gl. auersetur 36a11, n adbartaigedar gl. non paupertatem
eorum auersatur 36a18a, impf. 1sg. nadbartaiginn gl. auersabar 132c10,
157

3pl. with infix. pron. 1sg. nudadbartaigtis 19c5, fut. 1sg. adbartaigfersa gl.
auersabor 37c12, 2sg. annunadbartaigfe (Note Thes.: MS. annu nad
bartaigfesiu) gl. auersato 48a12-13, pret. 3pl. roadbartaigset 26b20, perf.
1sg. roadbartaigsiur gl. auersatus sum 115a3, gl. sed etiam ita auersatus
sum 117c5, 3sg. nroadbartaigestar gl. non fuerat auersatus 55d1, part.
adbartaigthe (with a question mark in DIL), gen. pl. adbartaichthe 44d13,
dat. pl. adbartaichthidib 24b9, VN adbartugud, dat. sg. adbartugud 26b20.
Denom. < adbairt adversary. The verb does not occur outside of
the Milan Glosses.
Ped. ii 526, DIL s.v. adbairt 37:1, adbartaigidir 39: 37.
*adbligedar flatters, adulates

Milan: VN aiplugud, apelugud 31a1, gen. sg. apelaigthe gl.
adulatorum 28a8, nom. pl. abelaichthi 74a6.
Note that this verb is not attested in Old Irish, but we have the VN
and the Middle Irish attestation aiplaigid flatters. Because in Old Irish
only the VN is attested, it cannot be seen whether this is a compound or a
simple verb, nor whether it is derived from aipile quickness, suddenness,
aipil quick, sudden or from bl lips, mouth, (by extension words,
subject of talk (DIL B-62: 61)) with the preverb ad-. This issue is discussed
in section 3.5, pages 74f. Note that the verb does not occur in LEIA.
LEIA A-80, DIL s.v. aiplugud 163: 5, aiplaigid 163: 3, aipile
162: 77, aipil 162: 62, bl 61: 77.
adchotadaigidir reconciles
Milan: sec. fut. pass. sg. neg. n adchtadaigfide gl. nullum
reconciliatio inueniet tempus 96a6, perf. 3sg. inradchotadaiged (Note Thes.
MS. inradchot adaiged) gl. reconciliatus est 32d24.
The verb only occurs in Milan. DIL states that this is apparently
based on a (reduplicated?) stem of adcota. The -da- could indeed be seen as
reduplication, but it seems more likely that this verb is based on a nonextant and probably non-existent adjective from adcota in -de or a verbal
adjective (the regular verbal adjective of adcota is ttae). This points in the
direction of the verb being a nonce formation. The irregular lenition after
158

ad- underscores this. This lention however does show that this is treated as a
simple verb, which is discussed in section 3.5, pages 74f.
DIL s.v. adchotadaigidir 41: 13.
adgldathar addresses, speaks to

Milan: pres. ind. 3sg. adgladathar 30d10, 74a8, 83a9, 115d1, 120b8,
145c14, 146a7, adngladar (Note Thes.: leg. adngldadar? but Phil. Soc.
Trans. 1891-4, p. 495; adgladar LL. 249a 52. ata-geglathar LU. 19b33.
Probably rather haplology or haplography) 53c17, impf. 1sg. with infix.
pron. atatgladainn 62c16, 3sg. n aclad 108b09, part. (though see below)
nom./acc. sg. nephacailse 56c8, nom. pl. acailsi 48a10, VN acallam,
acaldim 35c27, acaldam 103b19, acaldaim 120c3, dat. sg. acaldaim 84d3.
This verb is given in GOI as strong, but it is a weak i-verb (McCone,
Schumacher). The root *ldh- is registered in IEW, but this is incorrect:
neither the Indo-Iranian, nor the Greek are related, because these verbs are
onomatopoeic and made up on a base of gla, gla, gla... (see also page 64).
They could therefore have been created at any point, anywhere and cannot
be derived back to a common Proto-Indo-European root. Adgldathar is an
AII verb and has resisted first palatalisation in accordance with the regular
sound laws that a consonant does not palatalise if preceded by long -- (cf.
McCone 1996: 116). It has however undergone third palatalisation, which
shows that this verb was created before Old Irish itself. If the verb had been
created in Old Irish, this discrepancy would not have occurred (see also
Chronology, pages 80 and 82). Motivation for formation as a deponent verb
has been discussed in section 3.3.1.3, pages 63f.
Ped. ii 537-8, IEW 451, GOI 596, 715, DIL s.v. ad-gldathar 57:
35, McCone 1997: 46, 75, Schumacher 2004: 49-50 n. 40.
admuilnethar says again, repeats

Wrzburg: pres. ind. 1sg. with infix. pron. 3sg. n. atmuilniur gl.
iterum dico 18c12.
DIL has the form in Wrzburg above as headword and refers to
admuinethar remembers, calls with a question mark. Strachan considers a
cognate with molur. However, it is probably simpler than that. This is
159

derived from muilenn mill, with a sense of going over things again and
again, grinding things again and again. Motivation for formation as
deponent verb has been discussed in section 3.3.1.3, pages 63f.
LEIA M-75, DIL s.v. atmuilniur 463: 28, ad-muinethar 63: 9,
muilend 184: 86, Strachan 1893b: 462 n. 1.
adribaigedar reverberates (translation mine)

Milan: pres. ind. 3sg. adribaigedar gl. resultat 146b3.
Denom. < rib a snare, spring. Hapax legomenon. The verb has not
been translated in Thes., by DIL or by Joseph, nor has a derivation been
given in either of these sources. The form found glosses Lat. resultat
reverberates. Deriving the verb from rib gives a meaning doing what a
spring does (although note that as described above, section 3.4, page 65,
-igidir verbs can have a wide range of meanings and relationships to their
base and therefore basing the meaning of an -igidir verb on its base is
always an approximation). The preverb ad- adds the sense of repetition.
LEIA R-28, DIL s.v. adribaigedar 68: 9, Joseph 1987: 118.
adtluichedar gives thanks, rejoices at

adtluichedar: Wrzburg: pres. ind. 1sg. atluchur 3b19, fut. 1pl.
atluchfam 17a2, impv. 2pl. atligid 27b24; dotluchethar (desires, craves):
Wrzburg: pres. ind. 1pl. with rel. n and with elision of infix. pron. 3sg. n.
dontlucham 21d9, Milan: pres. ind. 1sg. duthluchim (Cf. McCone 1985: 98)
71c20, with nas. dunthlaichiur 44c20, rel. duntluchur 74d3, 2sg.
cidaratodlaigther 32a5, 3sg. dotluichethar 36a28, duthluchedar (Cf.
McCone 1985: 98) 38d1, with infix. pron. dathluchethar 30a10, 3pl. rel.
duthluchetar 85d5, impf. 3sg. duthluiched 124d9, pres. subj. 1sg.
coduthlucher 49d10, ciafiu todlaigersa 38c22, perf. 3sg. nadrotodlaigestar
124d9, part. tothlaigthe 21b3, VN tothlugud, todlugud 97d10, dat. sg.
todlugud 71a2, nom. dual dathodlugud 38c16.
IEW, followed by Sihler, connects tluichethar to the Latin
denominative loquitur speaks and with the OCS tlk, R tolk
explanation, which is conjectural, but plausible. It is certainly possible that
Celtic, Italic and Slavic would have some shared nominal vocabulary, from
160

which Celtic and Italic derived a verb. This derivation would have taken
place independently in each language; the Latin formation is completely
different from the Old Irish formation. These are root cognates. Since the
linguistic relationships are at such a deep level, and since there is no trace of
the base noun in Old Irish, tluichethar is a very old denominative, possibly
even
pre-Celtic,
though
not
Proto-Indo-European:
although
the
reconstruction of the root gives *teh3lku-, this would be an impossible root

in Proto-Indo-European. tluichethar can be found in the chronology on
page 80.
An alternative etymology for Latin with original -a- in the root (cf.
Walde & Hoffman s.v. loquitur, Schrijver) is not impossible. In that case,
the connection with OI tluichethar would fall.
LEIA gives W adolwyn prays as cognate, but adolwyn is derived
from root *lek- become bright (Schumacher).
Motivation for formation as a deponent verb has been discussed in
section 3.3.1.3, pages 63f.
Ped. ii 650, IEW 1088, LEIA T-79-80, DIL s.v. ad-tluichethar 73:
85, do-tluchethar 383: 54, Walde & Hoffman 1982 s.v. loquitur, Schrijver
1991: 476, Sihler 1995: 210, Schumacher 2000: 198.
adathaigidir dreads, abhors

Milan: impv. 3sg. na adhuadaiged gl. formidat 28b5
Denom. < adath great dread, horror, ad + ath fear, horror.
LEIA is non-committal on the reconstruction of the noun, saying that
the best etymology consists of a connection with Lat. pauere be frightened
of. IEW connects OI ath to *au- from with t-formation *aut(i)o-, i.e.
*h2eu- and *h2eut(i)o-. Isaac on the other hand suggests that it is simply
derived from the exclamation of pain ow!
IEW 73, 827, LEIA U-11, DIL s.v. adathaigid 75: 45, adath 75:
25, ath 40: 74, Isaac 2004 s.v. auto-.
aidilccnigidir needs, wants

Wrzburg: pres. ind. 1pl. adilgnigmar15a2. Milan: pres. ind./subj.
3sg. an nad naidlicnigedar 130a2.
161

Denom. < aidilccne what is needful, need < aidliccen needy,
poor.
DIL s.v. aidilccnigidir 106: 11, aidilccne 105:60, aidliccen 107:27.
ailigidir changes
PIE *h2eli os
CC *alios, * ali-, *al-noCf. MW eil, W ail, Corn. l, Br. eil (< *elis, comp *alii s) other,
second, also W all, Br. all, other.
Milan: VN ailigud, dat sg. ailigud gl. alternatione 114d7, acc. pl.
ailichthiusa gl. has alternationes 19b15, ailigthiu gl. alternationes 146a2-3.
Depronominal < aile, other, one of two. This is the only verb in
Old Irish that is derived from a pronominal, rather than a noun or an
adjective and it could be argued that it is derived from the adjectival or
nominal usage. This would however be unusual in itself, because when a
verb is derived from an adjective, it is always derived from adjectival usage,
and never from nominal usage (cf. page 92). It would be strange if this
depronominal verb was suddenly derived from nominal or adjectival usage.
This derivation shows that although within Old Irish the common way of
creating new verbs was to derive them from nouns and adjectives, this
constraint was not absolute. Verbs could also be derived from other parts of
speech. The -igidir category is the natural category for this verb to end up
in, being the default category in Old Irish, as discussed above, page 66.
Note that the verb is hapax legomenon and occurs only in Sg. 166a2.
The form found in Milan above is the VN. It always glosses Lat. alternatio
alternation and could be a calque. OI ailigud is the VN of ailigidir, derived
from aile, just as Lat. alternatio is derived from the verb alternare, derived
from the adj. alternus. Calquing has been discussed in section 3.4.3, pages
69ff.
LEIA A-31, DIL s.v. ailigid 124: 83, aile 116: 30, Schrijver 1995:
321-2.
ailithrigidir goes into exile

Cf. W all-fro or all-tud, Britt., W, Corn., Br. tir earth.
162

Milan: pres. ind. 1pl. elithrigmi gl. exulamus 46c22.
Denom. < ailithir stranger, pilgrim, < aile other, one of two and
tr territory, province. Hapax legomenon. It seems likely that the creator of
the verb was influenced by the Lat. exulare live in exile, which it glosses,
but this is not a calque, since it is not an exact translation of the Latin verb.
The Latin is made up of the preverb ex- out of and solum bottom,
ground, whereas the Old Irish is made up of the adverb aile and tr as given
above. Calquing and related issues have been discussed in section 3.4.3,
pages 69ff.
The semantics of the verb have been briefly discussed on page 90.
LEIA A-33, T-74, 75, DIL s.v. ailithrigid 125: 57, ailithir 125: 7, tr
187: 1, 27.
aininnracaigidir regards as unworthy

Milan: perf. 3sg. rel. anrunaninraccaigestar gl. indignante 62b21.
Deadj. < aininnraic unworthy, defective, < indraic worthy,
fitting. The verb is hapax legomenon. It glosses Lat. indignari deem
unworthy, scorn on which it is a calque. Calques have been discussed in
DIL s.v. aininnracaigidir 152: 77, aininnraic 152: 82, indraic 242:
3.
ainmnigidir names
Cf. OW anu, pl. enuein, Corn. hanow, pl. hanwyn, MB hanu name.
Wrzburg: pass. pres. ind. 3sg. rel. ainmnigther 21a14, Milan: perf.
3sg. with infix. pron. 3sg. m. rondnainmnigestar 17b9, with infix. pron. 3pl.
rosnainmnigestar 26b8, 37b22.
Denom. < ainm name.
LEIA A-36-7, DIL s.v. ainmmnigidir 157: 35, ainmm 156: 17.
airdircigidir celebrates, solemnizes

Milan: pres. ind. 3sg. erdaircigidir 28b15, fut. 1sg. with infix. pron.
2sg. noterdarcugub (this form not in DIL) 55a5, 3sg. rel. erdarcaigfes 89b4.
163

Deadj. < airdirc well-known, < derc eye, squinting. This
compound adjective is old and has a Welsh cognate ardderch. However, the
verb is late, as it is deadjectival and does not have the seek meaning of the
old *-sag-- verbs, cf. section 3.4, pages 65ff above.
LEIA A-41, DIL s.v. airdirc 186: 79, 2. derc 33: 69, airdircigidir
187: 58.
*airgalaigidir
Milan: part. dat. pl. ergalaichthib gl. propugnaculis 78c5.
The only form attested of this verb is the dative plural participle
above. The Lat. propugnaculis which it glosses is taken by the glossator as
pro pugnaculis (Note Thes.). Pugnaculis means fortress and is not verbal
at all. The Old Irish verb is denominative from airgal act of fighting, strife
and is formed under influence of the Latin. This has been discussed in
section 3.4.3, pages 69ff. above.
DIL s.v. airgalaichthib 207: 18, airgal 206: 70, 76.
airid guards, keeps watch over

frisairi (keeps watch): Milan: VN frithaire, frithare 111a5.
Denom. < aire (f i) act of guarding, watching over. This is a
specialist usage of the same root that gave the preposition per. It is
fundamentally a verbal root (LIV2s 1. *per- pass, come through), but in
Old Irish only the secondary verb derived from the secondary noun is
attested. The early attestations of the fris- compound are all religious.38
IEW 818, LEIA A-81-2, LIV2 472-3, DIL s.v. airid 212: 1, 1. aire
190: 23, frithaire 435: 37.
airigidir perceives, notices

Wrzburg: pres. ind. 1sg. with emp. pron. 1sg. ni airigursa 16a28,
perf. 1sg. (with elision of rel. n) ndrairigsiur 3c27, with rel. n nn
38
IEW and LEIA refer to an Old Breton cognate, namely the accusative plural form
areanos from Ammianus Marcellinus Book 28, Chapter 3, 28. This is a ghost form. The
correct reading is Lat. arcanos.
164

rairigsiur 3c26; Milan: pres. ind. 3pl. with infix. pron. 2sg. nodairiget
125d4, pass. sg. nadnairigther 30a3.
According to DIL this is related to aire act of guarding, watching
over, though LEIA derives it from airech heedful, attentive < aire. As has
been discussed above, page 67, where it is possible to derive a verb from a
derived base or from a non-derived base, the principle of economy has been
applied, and the verb has been derived from the non-derived base, rather
than the derived base. I have therefore derived airigidir from aire.
LEIA A-45, DIL s.v. airigidir 213: 58, 1. aire 190: 23, airech 193:
82.
airladaigidir obeys
Wrzburg: perf. 3pl. roirladigsetar 7c16, past subj. 2pl. rel.
conoairladigthe gl. oboediatis 3b9, VN dat. sg. irladugud gl. oboeditis
3b15, Milan: pres. ind. 3sg. erladaigidir gl. obsequitur 64d5, nerladaigedar
gl. tamquam39 64d3, impf. 3pl. noerladaigtis gl. parebant 16b14, pres. subj.
1sg. coerdalagear (Note Thes.: rectius coerladaiger) 106c6.
Deadj. < airlithe (io/i) obedient, submissive. The semantics of this
verb have been discussed above, page 92.
IEW 665, LEIA A-46, DIL s.v. airladaigidir 220: 26, airlithe 226:
23.
airlithir advises, counsels40
airlithir: Wrzburg: pres. ind. 3sg. rel. airlethar 17b22, pres. subj.
3sg. rel. airlethar 28b32; canairlethar (takes good heed): Wrzburg:
pres. subj. 2sg. mani cinairlither 5b38; conairlethar (consults, takes
council): Milan: pres. ind. 3sg. conairlethar 125d2, 1pl. ailemmar, impf.
1sg. airlin 54c27, perf. 3sg. rairlestar 125c1, VN comairle 23b7.
Denom. < airle act of advising, giving counsel, which is also used
as VN of airlithir. This verb has synonyms, but they are all either from this
39
The entire sentence reads: ac s diceret: lingua mea impar est tant adnuntiationis, sed
ministerium prbet Spiritui Sancto, et tamquam cuidam scrib docto calamus aptus
obsequitur.
40
Note that the Wb. Lex. gives a different meaning for this verb, cares for, gives attention
to.
165

stem, or derived from comairle, which has the same root again. Motivation
for formation as deponent verb has been discussed in section 3.3.1.3, pages
63f.
IEW 665, LEIA A-46-7, DIL s.v. airlithir 226: 39, airle 224: 1, can
33: 73, con-airlethar 424: 63.
airmitnigidir honours, reverences

Milan: verb. nec. airmitnigthi 51d20.
Denom. < airmitiu act of honouring, respecting, VN of
armuinethar.
LEIA A-48, LIV2 435-6, DIL s.v. airmitnigidir 230: 71, airmitiu
230: 5.
aithboaigidir revives, comes to live again

Cf. Gaul. biuo-, W byw, MB beu, Br. beo, Corn. byw, bew alive.
Milan: impf. 3sg. adbeuaiged 48b4 (here treated as a cmpd., DIL).
There are three possibilities for the origin of this verb: 1. It is a
compound of boaigidir that has in most cases been reinterpreted as a
simplex; 2. This verb is derived from bo living, alive with reiterative ad-;
3. It is derived from aithbo survivor(s). The noun aithbo is only found in
the 8th century text The Expulsion of the Dssi (Meyer), in the following
context: Beritt a n-aithbiu
a marbu ... This noun has been translated as
survivor, but these survivors have to be carried (together with the dead),
they are clearly wounded, they will need to be revived. Whereas option 1
and 2 are not fundamentally impossible, I suggest that the most
straightforward explanation of this verb is as derivation of aithbo revivee.
What one does with an aithbo is aithboaigidir: revive.
LEIA B-37, DIL s.v. aithbeaigid 254: 9, aithbe 253: 79, bo 78:
26, Meyer 1907: 141.
aithirgigidir repents
Cf. MW dir-rwyn torture, suffering.
Milan: pres. ind. 3pl. rel. naithirgigte gl. penitendo 105d11.
166

Denom. < aithirge act of doing penance, penance VN of adeirrig.
Since this is a Christian concept, it is most likely that this verb was created
in Old Irish, after the coming of Christianity. It has therefore been assigned
to Old Irish in the Chronology and can be found on page 82.
LEIA A-56, LIV2 503, DIL s.v. aithrige 274: 1, aithirgigidir 266:
22.
aithisigidir reviles, abuses

Milan: pres. ind. 2sg. aithisigther 22a10, impf. 3pl. with infix. pron.
1sg. nummaithisigtis 54c12.
Denom. < aithis insult, reproach, considered by Pedersen to be the
abstract noun of root *finn- with preverb aith-. This is followed by LEIA.
The OW atuis or atius metioned by LEIA, with reference to Stokes, is a
ghost-word, see McKee.
Ped. ii 676, LEIA A-55, DIL s.v. aithisigidir 267: 8, aithis 266: 36,
McKee 2000: 485.
laid requests, entreats

Cf. W ioli demand, praise, er-ioli implore, W iawl praise, OB
subj. 3pl. iolent gl. precentur.
laid: Milan: pres. ind. 3sg. rel. ailes 104b5; arlai (enjoins,
urgens an action on): VN eril, dat. sg. eril 95b6; frisilethar (looks
forward, expects): Milan: pres. ind. 1sg. -frithalim 49d3, 49d4, 3sg.
frisailethar 129a2a, 1pl. frissailemmar 63c7, impf. 3pl. rel. frisniltis 68a9,
fut. 1sg. frisailefarsa 38a10, perf. 1sg. rel. frisrailsiur 86d8.
Denom. < il act of asking, request (fem i and k). LEIA states that
the relationship between the noun and the verb is unclear. Besides, the
etymology of il is not uncontested. It seems logical to derive the verb ilid
from il, which is then also used as the VN of the same verb. The Welsh
yoli is an exact cognate, making it likely that this denominative is inherited
(Schumacher). It can be found in the Chronology on page 80.
IEW 501, LEIA A-30, 31, DIL s.v. ilid 124: 30, 4. aill 125: 65, arili 376: 70, fris-ilethar 422: 81, Schumacher 2000: 76.
167

lgenigidir soothes
Wrzburg: pres. ind. 3sg. lgenigid 4a2.
Deadj. < ilgen (o/a) soft, mild, without certain etymology. It
could go back to the verb legaid melts away, dissolves, but according to
LEIA it is better to see it as a compound of ail gentle, shy and gen smile,
laugh.
Note that the form in Wrzburg is already active. This is an example
of a very early Middle-Irishism, cf. McCone for other Wrzburg examples.
LEIA A-33, DIL s.v. ilgenaigid 122: 24, ilgen 121: 73, legaid 72:
34, 4. ail 114: 13, gen 66: 33, McCone 1985.
andaid kindles, lights

Milan: part. andithe 56d3, VN andud 131d14; forosna (kindles,
lights): Milan: VN fursundud 74b1.
Despite the appearance of this verb, it is an i-verb, rather than an averb, cf. its compound adannai and Schumacher. The etymology is
complicated. Pedersen tentatively suggests a compound arising from
generalised proclisis of the preverb that is found in Welsh as han- with the
primary root *do- scorches, burns. LEIA also considers this primary and
connects the root with MW ennynn lights, sets fire to, but Schumacher
correctly rejects this, deriving ennynn from CC *-i-n-d-e/o- lights (PIE
*h2ei dh-) and andaid as a secondary verb from a *anad-- or *anod--. Anlooks like the privative un-, which is not part of the verbal system. This
verb would therefore be denominative from a noun or adjective with the
Celtic descriptive reconstruction *anad- or *anod-. Since it is an i-verb, a
denominative would be more likely than a deadjectival (cf. section 3.3, page
52). More cannot be said about it at this time.
Ped. ii 457, 507, LEIA A-75, Schumacher 2004: 375, DIL s.v.
andaid 333: 23, for-osna 355: 37, ad-annai 33: 36.
anfossaigidir makes unstable?, makes uncertain?

Wrzburg: pres. ind. 2pl. with the interrog. inanfossigid gl. nescitis
8d5.
168

Denom. < anfoss restlessness, wandering, < foss rest, remaining
quiet or stationary. The verb is hapax legomenon. Inanfossigid above
glosses the Lat. nescitis do you not know? and is translated by the editors
of Thes. as are ye ignorant? I have given the meaning makes unstable,
makes uncertain, which is from DIL and based on the meaning of the
adjective. Note that Stokes, in Thes. supplement (vol. 2, p. 470) sees another
verbal form, anfissigid. This is not followed by DIL or Wb. Lex. The verb
anfissigidir would be another hapax legomenon derived from ainfis,
ignorance, lack of knowledge or information. Although this may suit the
Latin better in our eyes, there is no valid reason for emending an
unproblematic form in the manuscript for another hapax based on what a
modern scholar would say translates the Latin better.
IEW 1171, DIL s.v. anfossaigidir 343: 53, anfoss 343: 20, foss 379:
52.
ardraigidir appears
Wrzburg: perf. 3sg. roardrigestar gl. apparuit 28c12.
DIL has no certain translation, giving is distinguished or dignified
or is apparent as possibilities. Note that the Wrzburg form is not in DIL.
Thes. and Wb. Lex. translate the verb as appears, which has been followed
here.
As has been discussed above, section 3.4.1, page 67, where a verb
could be derived from a non-derived base noun or from a derived base noun,
the principle of economy is usually applied and the verb is usually derived
from the non-derived base noun. However, in this case that is not possible.
There is no Old Irish word *ardVr and there is nothing in this verb that calls
for a pre-Old Irish dating. This verb must therefore have been derived from
airdrech face, front.
DIL s.v. ardraigidir 391: 5, airdrech 189: 28.
arsadaigidir grows old

Milan: fut. 3sg. arsadaigfith gl. ueterescet 69b4.
Deadj. < arsaid old, ancient. The verb is a hapax legomenon.
LEIA states that the connection with Gk. (with dialectal variant
169

) gives grounds to believe that arsaid is an old compound of which
the first element was an adverb of the type *peres- or *pres.-, connected
with *per- before. The second element is derived from root *steh2stands. However, Irish has also a word arsan old and arsanta old
Therefore the etymology is uncertain. Pokorny has arsaid under *peres,
*peros, going back to *per- pass, come through.
IEW 812-3, LEIA A-91, LIV2 472-3, DIL s.v. arsadaigidir 412: 71,
arsaid 412: 75.
artaisi keeps silence, listens

PIE *th2esCC *tausCf. W taw, silence, VN tewi (th. taw-) is silent, tawel silent, OB
taguel ibid. The W andaw listen has the same preverb as intaisi below
(Ped. ii 652, GOI 520).
artaisi: Milan: impv. 2sg. nierthuaisse 126d11, fut. 3pl. artaisbet
126b12; contaisi (hears, is silent, DIL has this as com-, but that seems a
spelling mistake, as it is alphabetised under con-): Wrzburg: pres. subj.
3sg. diacoitsea (MS dicoitsea) 13a10, 2pl. nicoitsi (=nicoitsid) 17b32,
Milan: VN (f ) coitsecht, acc. sg. coitsecht 48a11; foitsi (understands
(what is implied but not expressed)): Milan: pres. ind. pass. sg. fotsider
34d5, VN foitsecht, dat. sg. foitsecht 23a7; intaisi (listens (to)):
Wrzburg: impv. 3pl. itset 13a11. Milan: part. tside, itside 136a9, VN
tsecht, itsecht 20a2, acc. sg. itsecht 21b11, dat. sg. eitsecht 24a18, 53c18.
Iterative of *th2es- be silent. According to LIV2, this root gives
the Old Irish adj. t, tuae silent, which leads to the deadjectival Old Irish
verb tuaisi. See also Schumacher, who gives a Common Celtic nominal
root *taus-tV-. Although this is not impossible, an iterative is more likely,
since if it were deadjectival, it would be the only verb inconsistent with the
scheme proposed in this thesis (cf. section 3.3.1 and subsections on pages
55ff.). The Brittonic cognates could also be iterative. In both suggestions,
the -s- remains unexplained.
170

Ped. ii 651-2, IEW 1056-7, LEIA T-99, LIV2 642, DIL s.v. ar-taisi
419: 71, con-taisi 467: 8, fo-itsi 220: 5, in-taisi 293: 61, te 211: 79,
Schumacher 2000: 179.
asmecnaigedar roots out (transl. Thes.)

Milan: pres. ind. 1sg. with emp. 1sg. coasmecnugursa gl. eradicem
2a8.
Denom. < mecon root. Hapax legomenon. Note that LEIA does not
register the verb. This could be a calque on Lat. eradicare root out with OI
as- corresponding to Lat. e- and mecon translating radix. Calquing has been
discussed in section 3.4.3, pages 69ff.
LEIA M-27, DIL s.v. as-mecnaigedar 434: 11, mecon 76: 53.
asrocholi defines, determines

See colaid AI, page 120.
astoasci expresses
Cf. MW gwascu, Br. gwaska press (IEW).
Milan: pres. subj. 3sg. coastoaiscea gl. expremat 73a3, verb. nec.
esto(a)iscthi, estoiscthi gl. inpremendam 134b2, VN (o) est(a)sc, estoasc
114a9, acc. sg. estosc 24d9, gen. estoisc 102c7, dat. sg. estoasc gl.
interpretatione 97a15, gl. inpresione 144c8.
The simplex is not attested in Old Irish, but cf. ModI fisgium, I
squeeze and the Middle Irish attestations in DIL. Pedersen and IEW derive
this from PIE root *edhh1- push, pound with -sk- suffix. Pedersen gives
MW gwascu, Br. gwaska and OHG wascan as cognates and states that the
alternation within Celtic between Brittonic a and OI is a reflex of ProtoIndo-European ablaut a : .
The root *edhh1- is in LIV2, but not with the -sk- suffix, since that
can only occur with zero-grade root (LIV2 19). The derivation of *fiscid as
primary verb from this root therefore falls. Moreover, the alternation
between long in Old Irish and short a in Brittonic is consistent as it stands.
Celt. *sk- becomes *gwsk- in British, which becomes shortened in the
171

late
6th,
early
7th
century
to
*gwsk-,
and
dissimilated
to
gwask-.
This verb is derived from the noun that gave OI fsc. In DIL the first
meanings given are report, tidings, because these are most numerous. The
original meaning will perhaps have been notice in the legal sense, which is
not inconceivably distant from pressure. Cf. DIL s.v. fsc 44: 58 and the
Welsh noun gwasc squeezing, pressure, weight. Outside of Milan, the verb
occurs once in the St. Gall Glosses.
Ped. ii 515, IEW 1115, LIV2 19, 660, DIL s.v. as-toasci 440: 5,
fiscid 29: 83, fsc 44: 78, fasc 44: 58.
attobi adheres to
Milan: pres. ind. 3sg. rel. atoibi 57d18, impf. 1pl. / past subj. 1pl.
(at)toibmis 18c4, fut. 3sg. niatoifea 114c2, VN (m u) attobed, dat. sg.
atoibiud 54d3
Denom. < tob side with preverb ad-.
LEIA A-101, 102, T-91-2, DIL s.v. at(t)obi 477: 14, taeb 12: 25.
bgaid declares, asserts

PIE *bheh1h- get vexed, quarrel, nomen actionis *bhoh1h-eh2- (cf. IEW)
or iterative *bhoh1h-i e- (LIV2)
CC *bg-a- or iterative *bg-Cf. G bgaudae rebellious guerillas, MW kymwy (= Mid. Ir.
combg) fight, -boawc = Mid. Ir. bgach warlike. It is not clear if W bai
mistake, fault and beio blames can be connected to this.
Wrzburg: pres. ind. 1pl. bigmi 2d15.
Bgaid could either be denom. < bg boast, threat or iterative
< *bhoh1h-i e- from PIE *bheh1h- get vexed, quarrel. Both would
phonologically regularly become OI bgaid.
The derivation of verbs from W kymwy fight (< *bhoh1gh-ios)
happened within Welsh itself and are therefore not cognates of the Old Irish.
The simplex bwyo reported by IEW does not exist.
Bgaids more colourful meanings glories, prides oneself; fights
are more likely to be original than the bleached utters, declares, so if this is
172

denominative, in specific connection with the relationship with the base
noun there is a strong element of iterativity involved.
IEW 115, LEIA B-4, 6, LIV2 68, DIL s.v. bigid 14: 35, bg 10: 38.
bidid submerges, drowns

PIE *geh2dh-, caus. *goh2dh-i eCC *bd-Cf. MW bawd drowns, VN bodi, W boddi drowns, Corn. buthy,
bedhy, MB beuziff, drowns, Br. beuzi, W diffoddi (< *di-spad- < *d-eksbad-).
bidid: Milan: past subj. pass. pl. conubadaitis 96c14, part. bitte,
nom. pl. bati 84d5; dobdi (drowns, extuingishes): Wrzburg: pass. perf.
3sg. rel. conderbadad 27a21.
Caus. The verbal root is only extant in Celtic and this only as
causative (LIV2).
Ped. ii 459, IEW 465, LEIA B-6, LIV2 206, DIL s.v. bidid 13: 55,
do-bdi 201: 72, Schulze-Thulin 2001: 160-1.
baitsid baptises
Wrzburg: pres. ind. 1sg. with infix. pron. 2pl. and emp. pron. 1sg.
manubbaitsimse 8a1, perf. 1sg. and infix. pron. 3sg. n. marud baitsius 8a3.
Denom. < baithis baptism, the order of baptism, Lat. baptisma
baptism. The verb could also be a direct borrowing from baptizare. In that
case, it would be an AI verb that due to syncope patterns looks like an AII
verb in most cases. Since there are no diagnostic forms attested in medieval
Irish, I have treated this verb as an AII verb, but this is purely speculative.
Direct and indirect loans have been discussed in section 3.2, pages 48ff. The
verb can be found in the Chronology on page 82.
Ped. i 237, LEIA B-11, DIL s.v. baitsid 21: 62, baithis 20: 83.
boaigidir makes alive, vivifies

Cf. W bywhau, Br. bewaad vivify.
173

Wrzburg: pres. ind. 3sg. beoigidir gl. uiuificantem 13d7, 3sg. rel.
with infix. pron. 1sg. nombeoigedar gl. Quod autem nunc uiuo in carne, in
fide uiuo Filii Dei 19a20.
Deadj. < bo living, quick; alive. This verb is difficult to date.
Joseph (p. 153-4) interprets bo as originally a noun, rather than an
adjective, interprets the verb as seeks life and dates it to the earliest stage
of the *-sag- derivatives. The interpretation rests however on speculation.
Even as a deadjectival, this verb could still have been formed in a common
stage of the language and inherited into Old Irish, Welsh and Breton. On the
other hand, all the early Welsh attestations are religious. While in Old Irish
material, the verb occurs both in religious and learned material, as well as in
saga material (cf. DIL), it is not inconceivable that the verbs were
independently calqued on Latin uiuificare, which it glosses in both
Wrzburg occurrences. I have left it out of the Chronology, since a
definitive dating cannot be given at this point. Calquing has been discussed
in section 3.4.3, pages 69ff.
LEIA B-37, DIL s.v. bo 78: 26, boaigidir 80: 7, Joseph 1987: 146,
153-4.
bertaigidir shakes
Milan: pres. ind. pass. sg. bertaigther 75b13, perf. 3pl. robertaigset
26d5.
DIL states bertaigidir = bertaid. As has been discussed above,
page 114, this is incorrect. Bertaid is an inherited denominative with a wider
range of meanings than bertaigidir. Bertaigidir is younger and its meaning
is restricted to shakes. Both verbs are derived from bert burden, load.
LEIA B-42-3, DIL s.v. bertaigidir 86: 85, bertaid 86: 69, bert 85:
53.
boltaigidir smells
Wrzburg: pres. ind. 3pl. with anaph. pron. 3pl. niboltigetarside
14d34.
Denom. < bolad smell, scent. LEIA does not register the verb.
Outside of the Wrzburg occurrence, this verb occurs once in Thes. II 5.28.
174

LEIA B-66, DIL s.v. boltaigidir 141: 78, bolad 137: 80.
*brthnaigidir judges (transl. mine)

Cf. OW braut, W brawd judgement.
Milan: part. brathnigthe gl. examinatum 31c28, dat. sg. brathnigthiu
gl. iudiciali 26c9-10.
Denom. < brth judgement, the last judgement. Although DIL
refers to brethnaigidir under the headword of the participle, the vowel
shows that the participle actually implies the virtual existence of another
verb, *brthnaigidir, probably derived from brth judgement, esp. the Last
Judgement. The participle is only attested twice, both times in Milan, both
times referring to God as judge, and was possibly created specifically for
this religious context. The reason for the -n- appearing between root and
suffix has been discussed on page 68.
LEIA B-86, 7, DIL s.v. brathnigte 164: 57, brth 161: 64,
brethnaigidir 182: 73.
brissid breaks, smashes

forbrissi (breaks down, crushes): Milan: impf. 3pl. with infix.
pron. 1pl. fortanbristis 135d4, fut. pass. sg. forbrisfither 56d10,
niforbrisbedar 51b1, past subj. pass. pl. forbrissitis 85d10, perf. 3sg.
forrobris 34b16, part. (io, i) forbriste, nom. pl. forbristi 85d9, VN (m u)
forbrissiud, forbrisiud 37b23, acc. sg. forbrisiud 46d10, dat. sg. forbrissiud
36a32.
Possibly denom. < bres fight, blow (DIL, LEIA).
Ped. ii 478, LEIA B-91, DIL s.v. brissid 192: 33, bres 179: 8, forbrissi 317: 57.
badaigidir vanquishes, triumphs

Cf. W budd profit, benefit, OB bud gl bradium, G boud-, bud
(LEIA).
Milan: VN badugud, buaduguth 33c13.
Denom. < baid victory, triumph.
LEIA B-107, DIL s.v. badaigid 220: 6, baid 221: 43.
175
badraid disturbs, troubles

Cf. W budr filthy.
Milan: part. badarthae, dat. sg. buadarthu 2b3.
Denom. < baidre confusion, distraction, < baidir, confusion.
DIL states that the verb is also found as baidrigid, but the only attestation
that actually has -ig- in it is doubtful. It is from the 10th century Tochmarc
Ailbe (ed. Thurneysen), and Thurneysen himself reads the form buaidrigthir
as equative, rather than a verbal form. The non-palatal form in Milan shows
that this form was created when the diphthong -a- was still --, because a
consonant (cluster) would not have been historically palatalised when
preceded by long . Due to analogy with the other AII verbs this verb was
synchronically palatalised in the later language.
LEIA B-108, DIL s.v. baidrid 223: 46, baidre 223: 7, Thurneysen
1921c.
banaigidir makes lasting (transl. mine)

Milan: pres. ind. / pres. subj. 3pl. cofinbuanaigit 102a12.
Deadj. < ban (o/a) lasting, enduring.
The formation of the form found in Milan above only becomes clear
in context with the Latin. It glosses uindimiant, which is a form of the verb
uindemiare gather / harvest grapes. The Old Irish verbs shows OI fn
wine (Lat. uinum) and OI ban lasting, enduring. There never was a
compound fnban so it is only in the verb that these have been put together
to translate the Lat. uindimiant. The verb thus created is at first sight not a
perfect translation of the Latin verb, because the Old Irish verb means
makes wine lasting. However, it certainly falls within the same semantic
field: making wine is a way to conserve grapes, and gathering the vintage is
in preparation of making wine. Calques and other Latin influences have
been discussed in section 3.4.3, pages 69ff.
LEIA B-110, DIL s.v. banaigid 228: 50, ban 227: 38, fn 137: 49.
176

canid laments, bewails
canid: Wrzburg: pres. ind. 1pl. cinmi 4a23; Milan: pres. ind. 3sg.
conid 30d4, rel. coines 73b12, coineas (Note Thes.: regularly cines)
102a23, dep. coinedar41 74b1, 3pl. rel. chonte 29b4, coinde 86d1, 126d5,
impf. 3sg. nucoined 32b13, VN (m u) coiniud 100a3; dorochoni (despairs
of): Wrzburg: pres. ind. 3pl. rel. derochinet 21b1, Milan: pres. ind. 3pl.
arnaderchoinet 117b5, impf. 3pl. rel. andurochointis 98a2, past subj. pass.
sg. dorochinte 75b18, fut. 3pl. n derchinfet 57d11, pret. 1pl. rel.
dorochinsem 89a6, 3pl. dorochoinset 46a17, dorochinset 131c9 (DIL lists
this as dorochonti), perf. 3sg. nruderchoin 44a1, part. nom. pl. derchointi
46a18, VN derchoniud, derchoiniud 89a6, 90b19, acc. sg. derchoiniud
50d8, 96a6.
Denom. < cane act of weeping, lamenting, borrowed from W cwyn
complaint. This interpretation follows LEIA C-18. While it is certainly the
case that the OI cane and W cwyn cannot be simple cognates, and while a
borrowing would be a straightforward explanation, this ignores the fact that
we have the exact same problem with OI man riches and W mwyn. Rather
than a borrowing, the identical nature of the problem shows that there may
be a sound law that is not yet understood.
Ped. i 125, ii 495-6, LEIA C-18, DIL s.v. canid 40: 64, do-rochoni
362: 16, cane 37: 76, man 34: 83.
cinid reviles, rails at

Milan: VN (m u) cined, dat. sg. chaned 58c6.
Denom. < cin, law, regulation (DIL), with meaning development
punish by lampooning > lampoons (OBrien). I have followed this
tentatively, although note that the origin of cin itself is disputed, see
Pedersen and OBrien for suggestions. If this is indeed the correct
derivation, the verb must have been created prior to Old Irish, because in
41
The form is clearly deponent, as listed by Strachan (1893: 595), although the reasons for
this are not clearly understood. Above, pages 63f., I have tentatively suggested that this is
perhaps an actual occurrence of the middle voice in Old Irish. Note that Strachans footnote
on the syntax of Ml. 74b1 (p. 595) has been silently rejected by Thes., as they translate the
Gloss differently.
177

Old Irish itself, the indirect relationship between verb and base would have
lent itself to an -igidir verb, rather than an AII verb.
Ped. ii 496 (note that DIL refers to 497), see also i 125, 378 and 381,
LEIA C-15-6, 18, DIL s.v. cinid 40: 51, cin 30: 82, OBrien 1956: 172.
cairdigidir makes a treaty (translation mine)

Milan: pres. ind. 3pl. (rel.) cairdigter gl. federari 126c21.
Demom. < cairde pact, covenant. Hapax legomenon. This could be
a calque on Lat. foederare, which it glosses. Both OI cairdigidir and Lat.
foederare establish by treaty are derived from nouns meaning treaty,
cairde and foedus (gen. foederis) respectively. Calquing has been discussed
DIL s.v. cairde 44: 44, cairdigidir 46: 73.
cairdinigidir unites in friendship

Milan: past subj. 3pl. nocairdnigtis (Note Thes.: the first letter in
the MS. may be r, but no- is required) gl. amicarentur 90d4, pret. pass. pl.
rocairdnigthea gl. foderatae (leg. foederatae) sunt 137c10, part.
cairdinigthe gl. federatarum 51d16.
Denom. < cairdine friendship, amity, < cara friend. Note that
LEIA does not register the verb. The semantics of cairdinigidir have been
briefly discussed on page 91. This could be a a calque on Lat. foederare
establish by treaty. Calques have been discussed in section 3.4.3, pages
69ff. Outside of the Milan occurrences, there is once attestation of this verb
in Sg. 188a32.
LEIA C-20, DIL s.v. cairdinigidir 47: 18, cairdine 47: 5, cara 72:
11.
cairigidir rebukes, accuses

Wrzburg: pres. ind. 1sg. rel. nochairigur 11d1, 2sg. rel.
nocairigther 1c10, 3sg. rel. with emp. pron. 3sg. m. cairigedarsom 25b9,
pres. subj. 2sg. diacairigther 1c10; impv. 2pl. cairigid 26b26, verb. nec.
cairigthi 28d22, carichthi (MS arichthi), Milan: pres. ind. 3sg. cairigidir
36a39, rel. n chairigedar 36a20, pass. sg. cairigthir 16c16.
178

Denom. < caire crime, fault.
LEIA C-21, DIL s.v. cairigidir 48: 42, caire 47: 23.
caithid consumes, spends

caithid: Wrzburg: pres. ind. 3sg. rel.: nad chaithi 6c11. docaithi
(spends, consumes): Wrzburg: impv. 3sg. with infix. pron. 2sg.
nachitochthad (= nachi-t-tochaithed) 31d11, Milan: pres. ind. pass. pl. rel.
duchaiter (Note Thes.: this might be read duchaster) 82a9, perf. 3sg.
durochthaisset (DIL durachthaisset) 98b13.
For caithid consumes, spends different origins have been proposed,
most of which LEIA correctly dismisses. Loth has suggested a connection
with W peidio abstain, in the primitive sense of supports, which could be
borrowed from Lat. pati. Pedersen suggested a connection with Lat.
quatere, but this is not convincing. Pokornys suggestion of a comparison
with G cateia projectile on the other hand, is valid. Caithid will have been
derived in Celtic from a nominal base *kat- that is no longer extant in Old
Irish but that was clearly extant in Celtic. This base has given G cateia.
From the same base was derived for example OI cath battle (< *katu). The
verb *katti originally meant casts, hurls, which is still one of the
meanings of caithid in Old Irish. (DIL s.v. 56: 7).
Ped. ii 479, LEIA C-24, DIL s.v. caithid 55: 5, do-caithi 219: 31,
cath 86: 24, Loth 1892, Pokorny 1936: 96.
cathaigidir fights, gives battle

Cf. W cattau fight (Joseph).
cathaigidir:
Milan: impf.
3pl. no
cathaigtis
16b20, part.
innephchathaigthe 126c22, VN cathugud 74b9, nom. sg. comchadugud

112b8, gen. sg. cathaigtho, 128a11 nom. pl. frithchathaigthi 36d4, acc. pl.
cathigthiu 55d17; friscathaigedar (fights against, opposes): Milan: pres.
ind. / pres. subj. 3sg. friscathaigedar 90a6, VN frithchathugud 73a12, gen.
sg. frithchathaigtho 16a12, acc. pl. frithcathaigthiu 134c2.
Denom. < cath battle, fight. Note that DIL registers the simple verb
as active and compound as deponent. Both verbs are originally deponent.
This is an old denominative. There is an exact cognate in W cattau.
179

Moreover, the meaning retains the meaning of the old *-sag- suffix seeks.
This verb originally meant seeks battle. It can be found in the Chronology
on page 81.
LEIA C-47-8, DIL s.v. cath 86: 24, cathaigid 88: 52, friscathaigedar 425: 3, Joseph 1987: 138, 146.
ceisid complains, grumbles

arceissi (pities, has compassion): Wrzburg: pres. ind. 3sg.
arcessi 4c19, rel. dianairchissi 4c20, pres. subj. 3sg. rel. coarcessea 5b35,
ndnairchissa 5b35, Milan: pres. subj. 3sg. erchissea 46c15, past subj. 3sg.
conna erchissed 61a5, VN (f ) airchissecht, gen. sg. airchissechtae 22c14,
erchissecht 120a5, dat. sg. erchissecht 32d23.
According to Pedersen this is derived from a nasal-less variant of
*kent-, which gives OI csaid suffers, endures. However, this root is not
*kent-, but *kendh- (LIV2), which could not give OI ceissid. Moreover, it
is not possible to take out a nasal that is part of the root. The connection
between csaid and ceissid falls.
Ceisid is instead derived from ces debility, inertia (LEIA). This
verb is strongly iterative, as complaining about a debility is not a single
event, and although it stretches the boundaries of transitivity, it is not
inconceivable to see this verb as transitive. Although complaining about
debility requires a preposition to express it, the complaining itself still needs
an intrinsic object - the debility. It is therefore not surprising that this verb
became an AII (cf. section 3.3.1.1, pages 55ff.). What is however surprising
is that this did not become an -igidir verb. The relationship between base
noun and verb is not direct. The verb does not mean is debilitated, makes
debilitated as is customary with the non-igidir verbs. The verb is a response
to the noun: it means complains about a debility. If this verb had been
created within Old Irish, this more indirect relationship would have called
for an -igidir verb (see above, page 66, where this verb is also discussed). I
therefore propose that this verb was created before Old Irish, quite possibly
at a point when the *-sag- suffix was still restricted to the meaning seeks,
so that that possibility was excluded. The compound arceissi was then
180

formed as another response to the noun, and possibly to the verb, with the
meaning pities, has compassion.
Ped. ii 486, IEW 641, LEIA C-77-78, 79, LIV2 390, DIL s.v. arceissi 381: 61, ceisid 104:45, ces 146: 79, csaid 149: 3.
cestaigidir inquires, investigates

Milan: perf. 2 sg. arrucestaigser 2d3.
Denom. < ceist question, problem < Lat. quaestio question (DIL,
LEIA). Considering that this is rather technical, I would argue that this is a
formation from the Christian period. It has therefore been dated to the Old
Irish period and can be found in the Chronology on page 82. Direct and
indirect loans have been discussed in section 3.2, pages 48ff.
LEIA C-57, DIL s.v. cestnaigid 150: 23, ceist 105: 26.
cinid is born, descends (from)

Wrzburg: perf. 3pl. rel. rochinset 4c6, Milan: perf. 3pl. rocinset
26b17.
IEW derives the verb directly from the primary root *ken- arise,
begin, but LIV2 states that that root is verbally only attested in Slavic and in
the Mid. Ir. denominative cin- arises. LIV2 is clearly not entirely correct,
as there are two Old Irish attestations, however treating it as a denominative
is a valid possibility. There is no Old Irish word that this verb could have
been derived from, but cf. OI cenl kindred, race, OW cenetl kindred, W
bachgen boy and *kni- in OI can fine, good (LIV2). There is moreover
an element *cen- attested in a large amount of Continental Celtic names
(Evans). The Gaulish, Welsh and Old Irish nominal attestations are
consistent with the verb either being an an older derivative (because the
base noun itself is not attested in Old Irish), or with it being primary in the
first place. No definitive answer can be given at this time.
IEW 563-4, LEIA C-103, LIV2 351, DIL s.v. cinid 193: 73, cenl
116: 17, can 32: 40, Evans 1967: 175-7.
cinnid defines, fixes

Cf. W pennu defines, specifies.
181

Milan: pres. 3sg. rel. n cinni 128a4, 128a8, pres. subj. 3sg. rel.
nocinned 17d4.
Denom. < cenn head (DIL, LEIA). Although Welsh has a formal
cognate within the same shade of meaning, this is a young verb (see
Chronology, page 82). Whereas the meaning defines can be seen as a
direct derivation of the word cenn and penn especially when taking into
account that these words can also mean end, termination, the other
meanings of the verb show the influence of the Latin word terminare mark
the boundaries of, derived from terminus end, termination. The semantic
field of the verb, where it exceeds that of the base noun, is too common
between these three languages to be coincidental. The Old Irish and Welsh
verbs were formed under influence of the Latin. The intellectual unity of
medieval Ireland and Wales caused similarity in intellectual discourse and
when one language used the denominative from cenn in this way, it is likely
that this will have influenced the use in the other language. Cf. also the
Welsh cognate dibennu determine, settle (outcome, debate, etc.) that
corresponds to the semantic field of Lat. determinare.
The relationship between cinnid and forcenna, both having been
derived from the same base, has been briefly discussed on page 61.
LEIA C-104, DIL s.v. cinnid 195: 41, cenn 120: 3.
cluichigidir sports, plays

Milan:
pres.
3sg.
rel.
acluichigedar
(Note
Thes.
MS.
acluichigegedar) gl. eluctatus 122a10, dond chluchigedar gl. ludenti

122a12, 3pl. cluichegatar gl. arietes 129c20.
Denom. < cluiche game, play (DIL, LEIA). Outside of these Milan
occurrences, there is one attestation in OMulconrys Glossary.
LEIA C-127, DIL s.v. cluichigidir 260: 66, cluiche 259: 23.
*cobrigidir foams? (Thes.)

Milan: pres. ind. 3sg. acobrigedar gl. eluctatus 122a10.
Hapax legomenon. No translation given in DIL, though they refer to
cubar foam, froth. Thes. translates as foams (?). Note that DIL cites the
182

3sg. conjunct form as found in Milan, whereas I have given the hypothetical
citation form.
LEIA C-137, 264-5, DIL s.v. cobrigedar 280: 61, cubar 579: 34.
colaigidir makes thin, grows thin

Cf. MW culhau make thin (Joseph).
Milan: impf. pass. nocoilaigthe gl. attinuabatur 127c15.
DIL refers to colaid makes thin, grows thin (see above, page 120),
deadj. < col thin, slender, and just as colaid, colaigidir can be derived
from col. Both verbs have only two attestations.
LEIA C-6, DIL s.v. celaigidir 13: 6, celaid 13: 3, cel 10: 82,
Joseph 1987: 146.
coillid damages, violates

Milan: perf. 3pl. nadrochoilset 48d28.
Denom. < coll destruction, spoiling. Coillid has been discussed in
its relationship to malartaid on page 60 above.
LEIA C-146, 158, DIL s.v. coillid 297: 78, coll 326: 22.
comadasaigidir adapts, fits

Milan:
pres.
ind.
3sg.
comadasaigidir
gl.
aptat
53c8,
comadassaigidir gl. accomodat 80d6, perf. 3sg. rocomdasaiged gl. aptata

esse 86d17.
Deadj. < comadas (o/a) suitable, fitting, < adas according to,
proportionate to. No translation in DIL, but this verb is used to gloss Lat.
aptare and accomodare, both meaning adapts, fits, which is also the
translation Thes. gives for Ml. 53c8. The only other attestation of this verb,
according to DIL, is in the glosses on the Carlsruhe Beda.
DIL s.v. adas 35: 63, comadasaigidir 334: 83, comadas 334: 25.
comaicsigidir approaches, comes near to

Milan: pres. ind. 3sg. rel. comaicsigedar gl. adpropiare 107b9, fut.
3sg. comaicsigfid 53b17.
183

DIL refers to the adjective comocus near, proximate in their entry
for this verb, but it is more likely that this is derived from the noun
comoicse proximity, nearness. In the base adjective, the expected syncope
is missing (**comcus42) and in the noun, the syncope is in the wrong place
(**comcaise), probably in analogy to ocse nearness, proximity. If the
verb had been derived from comocus, the missing syncope in the adjective
would almost certainly have occurred in the right place in the verb,
**comcaisigidir, rather than in the third syllable as found in comaicsigidir.
When deriving the verb from the noun, on the other hand, it is likely that the
already existing syncope would have been kept, as found in comaicsigidir.
This could be a calque on Latin adpropriare, since it translates the
elements exactly. Latin adpropriare is derived from propius near with the
preverb ad-. OI comaicsigidir is derived from comocus near, which is
formed from com- and ocus near. Calquing has been discussed in section
3.4.3, pages 69ff.
LEIA C-172, DIL s.v. com()oicsigidir 377: 25, com()ocus 376: 29,
ocus 95: 29, com()oicse 377: 16, ocse 94: 61.
comdemnigidir rules?
Milan: pres. ind. 3sg. rel. comdemnigedar gl. dominari 17b1.
Deadj. < comdemin very certain, < demin sure, certain. Hapax
legomenon. DIL has no translation of this verb, but refers to coimdemnacht
lordship, rule < coimdiu lord, master. It translates Lat. dominare be
master, rule over. However, neither coimdiu nor coimdemnacht quite fits as
origin. Coimdemnacht would lead to **coimdemnachtaigidir, with possibly
a syncope, though the resulting consonant clusters would make that
difficult. Coimdiu is a dental stem, so that would leave the -m- unexplained.
Deriving this verb from comdemin has none of these problems. This verb is
treated as a simple verb in Old Irish, since the preverb is always com-. The
treatment of denominative verbs as simple or compound verbs has been
discussed in section 3.5, pages 74f.
42
Note the form comhguis in Keatings Dnta, amhrin is caointe Sheathrin Citinn 133
(DIL s.v. comocus 376: 62).
184

Ped. ii 102, LEIA C-147-8, DIL s.v. coimdiu 301: 10,
comdemnigedar 367: 80, coimdemnacht 300: 60, condenim 367: 78, deimin
8: 77.
conruidethar intends (Wb. Lex.)

Wrzburg: pret. 3sg. rel. rodastar 7a14.
Denom. < suide act of sitting. DIL splits conruidethar and its pret
conrodastar, but there is no need for this. Whereas suidethar has -u- in the
present stem, its root is in -o-, so it is phonologically regular that this -owould appear in the s-pret (cf. GOI). Since the verb is transitive and
causative (deciding is causing settlement on an idea) it is also in
accordance with the scheme suggested in this thesis that this verb is AII (cf.
section 3.3.1.1, pages 55ff.). Moreover, intention implies subject
involvement, which would require a deponent verb (cf. section 3.3.1.3,
pages 63f.).
Ped. ii 502, LEIA C-198, DIL s.v. con-ruidethar 459: 1, conrodastar 458: 68, suide 413: 73, GOI 677.
conterchomraici collects, convenes

conterchomraici:
Milan:
impf.
1sg.
conterchomraicinn
gl.
conueniendo 119a12, pret. 3sg. chontarchomraic (Note Thes. leg.

contarchomraic Ascoli; but for the aspiration of the initial consonant cf. KZ.
xxxv. 367, Celt. Zeitscr. IV. 61) gl. congregata 57a14; doairchomraicci
(heaps together, amasses): Milan: pres. ind. 3sg. duerchomraici gl.
adgregat 110d1, 3pl. duerchomracat gl. adglomerant 88c5, past subj. 3pl.
duerchomraictis 61b16.
Denom. < comrac, act of meeting, encounter, VN of conricc with
*kom-to-exs-ro- (Ped., DIL, com-to-are, proposed by LEIA, is incorrect).
*comraicid would mean calls a meeting, making this verb inherently
causative. The verb only occurs in the Milan Glosses.
Ped. ii 556-8, LEIA C-201, D-114, DIL s.v. con-terchomraici 465: 5,
do-airchomraicci 185: 62, comrac 405: 25.
185

cosmailigidir likens, compares
Milan: perf. 3sg. with infix. pron. 3pl. roscosmailigestar 55d8.
Deadj. < cosmail like, similar < com + samail likeness, similarity.
LEIA C-216, DIL s.v. cosmailigidir 494: 39, cosmail 493: 29, samail
47: 32.
crdaid torments, persecutes

doaccrdi (provokes, exasperates): Milan: pres. ind. 3sg.
doaccradi 18d13, 3pl. duacradat 74c1, impv. 3sg. tacradad (in DIL as
tacrdad) 87a1, impf. 3pl. duacratis 74c7, perf. 3sg. doracrid 28a17.
Denom. < crd torment, anguish. Formally, this could also be an
iterative, following the same line of argumentation as for bgaid above
(page 172). However, an added problem for crdaid is that there is no
known etymology for its stem.
Ped. ii 496, LEIA C-221, DIL s.v. cridid 507: 29, do-accrdi 178:
82, crd 503: 38.
crithnaigidir shakes, trembles

Milan: part. crithnaighte 115c4.
Denom. < crith act of shaking, trembling. The -n- appearing
between base and suffix has been discussed in section 3.4.2, page 68. The
semantics of crithnaigidir have been discussed on page 87.
LEIA C-239-40, DIL s.v. crithnaigid 536: 29, crith 534: 22.
*crottigidir plays on a harp

Wrzburg: pres. ind. pass. sg. rel. crottichther gl. citharizatur 12c45.
Denom. < crott harp, lute. Hapax legomenon. The semantic
relationship between this verb and its base have been discussed on pp. 66
and 104. This could be a calque on the Lat. citharizare play on a cithara,
lyre, which it glosses. Lat. citharizare is derived from cithara meaning
lyre and although the instruments are perhaps not the same, the actions are
certainly similar. Calquing has been discussed in section 3.4.3, pages 69ff.
LEIA C-248, DIL s.v. crottichther 553: 39, crott 552: 67.
186

cruthaigidir creates, shapes
Milan: pres. ind. 3sg. rel. cruthaigedar 140b5, pass. pl. rel.
cruthaigter 138c1.
Denom. < cruth form, shape. The semantics of cruthaigidir have
been discussed on page 91.
LEIA C-256, DIL s.v. cruthaigid 564: 82, cruth 563: 23.
cuicsedar taxes (transl. Thes.)

Milan: pres. subj. pass. sg. nadcuicsedar gl. taxari 36a38.
No translation in DIL. Hapax legomenon. Thurneysen takes this as a
denominative of com + cs census (< Lat. census).43 This seems correct, as
it translates the Lat. taxari, which it glosses, exactly. I see no basis for
DILs com-in-c. From the form, it cannot be seen whether this is an active
or deponent verb. I have therefore cited the form as found, rather than
reconstructed a 3sg. citation form.
LEIA C-107, DIL s.v. -cuicsedar 586: 19, cs 202: 19.
*cuimlengaigidir fights (Ped.) or engages (Thes.)

Milan: verb. nec. cuimlengaigthi (Note Thes.: The second g is
interlined) gl. congrediendum 16a11.
Denom. < cuimleng act of contesting, contest, VN of conling. DIL
cites this verb with -ig- in brackets because the only early attestation of this
verb is in the Milan gloss cited above. However, there the g is interlineair:
i.e. added later (it is not clear whether that is by the same scribe).
This verb seems to be a coinage of the learned tradition. It may well
first have taken place in this sort of Latinate context, but it cannot only have
existed in the context of the Milan glosses, since it appears in other contexts
too (cf. DIL).
Ped. ii 567, LEIA C-271, DIL s.v. cuimlenga(ig)id 590: 52, cuimleng
590: 24, conblicht 427: 10, conflechtaigthi 437: 54.
43
The reference is taken from Thes., but cannot be traced to its source.
187

cuimnigidir remembers, recalls
Cf. MW coffau, MB couffhat, ModB kounaat (Joseph).
Milan: pres. ind. 3sg. rel. cuimnigedar 16b24.
This verb could either be derived from cuman remembered,
memorable (DIL) or from cuimne faculty of memory, remembrance
(LEIA). It is a *-sag- derivative with Brittonic cognates, either
denominative or deadjectival, and widely attested in the later language. It is
to be an early example of bleached meaning (cf. section 3.4, pages 65ff.)
and can be found in the Chronology on page 81.
LEIA C-271, 272, 287-8, DIL s.v. cuimnigidir 591: 51, cuimne 590:
68, cuman 617: 45, Joseph 1987: 146.
cuirethar throws, moves

PIE *kerH- it *korH-i eCC *kor-Cf. W escor give birth, be delivered of, hebcor dispense with, do
without, ymatcor return, OB etcer returns, MB dazcor giving back,
digueriff open, hepcor renounce (Schumacher).
cuirethar: Wrzburg: impv. 2pl. cuirid 27b11, pres. subj. 2sg. with
infix. pron. 3sg. f. niscoirther 10a21, Milan: pres. subj. 3sg. arnach
corathar 68b9, pret. 3sg. dia corastar 39a3, verb. nec. cuirthi, nom. pl.
coirthi 72b13, part. cuirthe 56c5, VN (m o) cor, dat. sg. chor 72b12, cor
99d7, adcuirethar (restores, returns): Milan: impf. 3pl. aidchuirtis44
118c3; arcuirethar (increases, extends): Milan: VN (m) archor, dat. pl.
erchoraib 75b8; doaithchuiredar (returns): Milan: pres. ind. 3sg.
duadchuridar 34d7, fut. 3sg. donaithfoicherr 34d8, 3pl. dunathfoichret
72d1, part (io, i) tathchoirthe, gen. pl. taidchoirthe 82c6, VN (m o)
tathchor, tadchur 34d20, acc. sg. taidchor 47b3, 119c2, gen. sg. aithchuir
44
The reading of the form is unclear. DIL discusses this verb s.v. fristacuirethar (430: 65):
Cf. co fristacor aidchuirtis du aithis form (gl. obiicerent) Ml. 118c3, where Thurneysen
[Thes. - ELM] would read: co fritacuirtis du aithis, and Pedersen: fristacortais co naidchuirtis (VG ii 501-502), Strachan, Depon. Vb. p. 9 n. 5 renders that I may cast back
upon them. It is to be noted that the form is deuterotonic and therefore co renders ut in the
Latin text. The glossator evidently went astray in searching for the verb depending on ut.
Cf. further frithtacuirsimem Ml. 106b15.
DIL s.v. frithtacuirsimem (446: 27): adj. superl. (apparently corresp. to fris-tacuir-, see
Thurn., Gr. 443 ). gl. infestissimum Ml. 106b15.
188

42c26, taidchuir 46a26, 62c7, dat. sg. taidchur 62b4, 70c5; docuirethar
(puts, invites): Wrzburg: with forms from *to-ro-la-: pres. subj. 3sg. with
infix.pron. 3sg. m. didtar(la) 24a17; pass. perf. 3sg. doralad 13d8, Milan:
pres. ind. 3sg. with infix. pron. 3sg. dundchuirethar 35d22, pass. sg.
dofocuirther (Ped. ii 501) 18c10, tarsatochuirther 22c1, rel. diatochuirther
80b7, impf. 3sg. (rel.) tochorad 44a19, fut. 1sg. docuirifar 3a1, 3pl.
condichret 44a16, sec. fut. 3sg. nitochuiribthe 128a2, pret. 3sg. ducorastar
52 line 1845, perf. 3sg. do-rochurestar 16c6, nruthochurestar 18d6,
dorochuiristar 25c13, forms from *to-ro-la-: pres. subj. 3sg. torala 43b15,
perf. 3sg. dusrale 23c16, pass. sg. duralad 67d9, VN (m o) tochor, dat. sg.
tochur 28c12, 29a8, doetarcuirethar (intercedes): Wrzburg: pres. ind.
3sg. rel.: doetarcuirethar 4b19, friscuirethar (puts forward): Milan: pres.
ind. 3sg. (rel.) with infix. pron. 3sg. fritcurethar 41d16, pres. subj. pass. pl.
friscoirter 21d3, past subj. pass. sg. friscoirthe 95a1, fut. 1sg. frisfoichiurr
78c8, VN frecor, frecor 131a8, frecur 22a4, 137c1, acc. sg. frecur 43a2,
79b2, acc. pl. frecru 30d16; immcuirethar (carries, brings, bears):
Wrzburg: pres. ind. 3pl. rel. immechuretar 5a5.
Iterative, PIE *korH-i e- of root *kerH- disseminate, throw (LIV2,
for suggestion of alternative root, see Isaac), root cognate to OI foceird
sets, puts, places (301). Absolute forms of the simplex and deuterotonic
forms of the compounds with one preverb supplied by foceird in present
and subjunctive, all forms supplied by foceird in the future and the perfect
is supplied by ro-la- (Ped.).
GOI derives this as denominative from the VN cor act of putting,
placing, which is incorrect, considering its inflection as a causative: it
consistently shows the alternation between -u- in the present stem and -o- in
the other stems (cf. above, page 30 and GOI 607).
LEIA separates two roots, one strong (with meaning places, puts)
and one weak (with meaning invites), the weak root being further
subdivided between an originally denominative present and causative
subjunctive. There is no need for this separation. The verb inflects weak
45
189

throughout and the vowel alternation between present and subjunctive stem
is explained by regular phonological processes, through which an original
-o- followed by -- (< PIE *-i e-) is raised to u, but followed by remains o.
Ped. ii 498-502, IEW 933-5, LEIA C-276-7, LIV2 353-5, DIL s.v. foceird 186: 33, ad-cuirethar 47: 53, ar-cuirethar 385: 23, do-aithcuiredar
197: 42, do-cuirethar 233: 35, do-etarcuirethar 260: 84, fris-cuirethar 425:
52, imm-curethar 136: 34, cor 470: 16, Pokorny 1920: 79, Sjoestedt 1925:
327, GOI 607, 677, 762, McCone 1991b: 178, Isaac 1996: 308,
Schumacher 2000: 160-1, Schulze-Thulin 2001: 148.
cumachtaigidir rules, is powerful

Milan: fut. 3pl. cumachtaigfet gl. potituri 28a12.
Denom. < cumachtae power, strength. The semantics of
cumachtaigidir have been briefly discussed on page 90 above.
LEIA C-286, DIL s.v. cumachtaigidir 616: 2, cumachtae 615: 9.
cumgaigidir constricts, confines

Milan: pres. ind. pass. sg. rel. nicumgaigther 32d14, impf. / past
subj. pass. pl. nocumgaigtis 87c5, pret rel. racumgaigestar 133a9.
Deadj. < cumung (u) narrow, constricted.
LEIA C-291-2, DIL s.v. cumgaigid 619: 80, cumung 629: 29.
*cundradaigidir trades
Milan:
past
subj.
3sg.
nocundraiged
(Note
DIL:
Leg.
cundradaiged?) gl. mercari 39a6.

The form above is found under headword cundraiged in DIL. The
3sg. citation form might never have existed; this could be a nonceformation. DIL gives no translation, but Thes. translates the gloss as that he
should traffic. It glosses Lat. mercari to trade, to buy and it is denom.
< cundrad (act of) trading, buying and selling. I have therefore translated
it as trades. LEIA makes the same connection and has the same
translation.
LEIA C-292, DIL s.v. cundraiged 630: 50, cundrad 629: 83.
190

cutrummaigidir equalises, makes equal
Milan: pres. ind. 3sg. cochutrummaigidir 25c12, pass. sg. with infix.
pron. 1sg. nmchutrummaichther 44c3, perf. 3sg. rocutrummaigestar 55d3.
Denom. < cutrummae equal weight or amount, equivalent, < com +
trummae heaviness, weight.
LEIA C-298-9, DIL s.v. cutrummaigidir 639: 8, cutrummae 638: 16,
trummae 328: 37.
daingnigidir makes fast, fixes

Milan: pres. ind. 3sg. daingnigthir 134c7, pass. sg. daingnigther
51d8, pres. subj. 2sg. with infix. pron. 1sg. conumdaingnigese 49d13, fut.
2sg. with infix. pron. 1pl. rel. nundundaingnichfe 78c6, perf. 3sg.
rodaingnigestar 51d8, part. daingnigthe 47c8, VN (m u) daingniugud,
nom./acc. sg. daingnigud 39c4, acc. pl. daingnichthiu 26d9.
Deadj. < daingen firm, fast and stronghold, fastness.
LEIA D-11-2, DIL s.v. daingen 27: 38, daingnigidir 29: 12.
dlaid portions out, pours out

Cf. W gwaddol a portion or dowry (Ped. ii 503).
confodlai (shares jointly, divides): Wrzburg: pres. ind. 2pl.
aconfodlid (MS aconfodli) 24b19, Milan: fut. 2pl. confodlaibid 53b18;
dodili (pours out, dispenses): Milan: pres. ind. pass. sg. dudailter 76b4,
pres. subj. pass. pl. dundaleter 101d5, VN todail, dat. sg. todail 94c20,
134b4, dat. pl. todalib 30d1, 118c7; fodli (distributes, dispenses):
Wrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. n. fot dli (= fo-d-dli)
12a8, with rel. n and infix. pron. 3sg. n. fondrodil 10a11, pass. pres. ind. 3pl.
with rel. n fundailter 33b19, Milan: pres. ind. 2sg. fodali 39c2, 3sg. fundali
20c8, 30d26, fodali 42b27, fut. 1sg. fodalib 78a10, 3sg. fodailfea 90a8,
impf. 3sg. / past subj. 3sg. fodlad 82d2, perf. pass. pl. forodalta 99c4, verb.
nec. fodlaidi 105b13, part. fodlaide 26a9, VN fodail, dat. sg. fodail 22d16,
93d6.
Denom. < dl a dispensing of drink or food, a distributing, also
used as VN of dilid. Denominative in both Welsh and Old Irish, so this is
probably a Celtic formation. Schulze-Thulin sees a CC *dl-t,
191

denominative from PIE *deh2-lo- of PIE root *deh2 share. It can be found
in the Chronology on page 80.
Ped. ii 502, IEW 175-6, LEIA D-15, LIV2 103-4, DIL s.v. 4. dl 48:
34, dilid 22: 19, con-fodlai 437: 64, do-dili 239: 33, fo-dli 211: 20,
Schulze-Thulin 2001: 121 n. 192.
dnaigidir gives, bestows

dnaigidir: Wrzburg: perf. 3sg. with infix. pron. 3pl. rosdnigestar
21b9, Milan: pres. ind. 3sg. rel. with infix. pron. 3sg. n (rel.)
naddanaiged(ar), pass. sg. rel. danaigther 17c7, pres. subj. pass sg. rel.
danaigther 40b2, fut. 3sg. neg. with infix. 3sg. f. connachadnaigfea 96a7,
perf. 3sg. rel. rondnaigestar 96b9, neg. with infix. pron. 3sg. nach
radanaigestar 97d17, part. dnaigthe 135c4, VN (m u) dnaigiud, dat. sg.
danaigiud 96a8, *addanaigedar: Milan: fut. 3sg. addanigfea 112c4, part.
(io, i) atdnaigthe, atdanigthe 116b7-9, atdanaigthe (Note Thes.: at over
the line) 116b7-9, nephatdanaigthe 138a5, acc. pl. nephatdanigthi 56b10.
Denom. < dn gift, bestowal. Addanaigedar could be a compound
of dnaigidir or it could have been independently coined from dn. It only
occurs in Milan and was formed to gloss Lat. remunerare reward. It could
be a calque, since both the Old Irish and the Latin verb are derived from the
word for gift. Calquing has been discussed in section 3.4.3, pages 69ff.
above.
LEIA D-22-3, DIL s.v. dn 70: 54, dnaigidir 81: 27, addanigfea
51: 19.
debthaigidir dissents, quarrels

Milan: pres. ind. 1sg. nidephthigim gl. non desideo 21a2, 3pl. rel.
debthaigte gl. tumultuantibus 103d17, impf. 3 sg. nodebthaiged gl.
desedebat 19c13, 3pl. rel. nodebthaigtis gl. iurgantes 103b15, perf. 3 pl.
rodebthachsetar gl. desiderant 19c16.
Denom. < debuith strife, contention (DIL), cmpd of buith, VN of
att (LEIA).
LEIA D-30, DIL s.v. debuith 182: 2, debthaigidir 181: 54.
192

dechraigidir is scattered, dispersed
Wrzburg: pres. subj. 3sg. mani dechrigedar 12c46, Milan: pres. ind.
3sg. with infix. pron. 3pl. nosdechrigetar 115b2, rel. andechraigte 119d11,
1pl. dechrigmir 117b9, fut. 3pl. neg. nidechraigfet 90b6, VN (m u)
dechrugud, dat. sg. dechrugud 19d2, 62a5.
Denom. < dechor the condition of being different, difference. Note
that one of the meanings of the verb is is different. There is no reason to
assume that this verb is old. It is therefore likely that it developed multiple
meanings simultaneously. This is not uncommon for -igidir verbs and it is a
logical consequence of their varied relationship to their bases as described
above, page 66.
LEIA D-32-3, DIL s.v. dechraigidir 200: 17, dechor 196: 19.
deiligidir separates, scatters

Milan: VN deiliugad, acc. sg. deligud 37d11, 128d9.
Denom. < deil separation.
LEIA D-41, DIL s.v. deiligidir 6: 77, 1. deil 5: 38.
deimnigidir confirms, certifies

Wrzburg: perf. 3sg. with infix. pron. 3sg. n. radeimnigestar 32c20,
Milan: pres. ind. 1pl. demnigmi 35b1, verb. nec. demnigte 16a15, part. (io,
i) deimnigthe, demnigthe 39c18, VN deimnigud, demnigud 35b15,
demnigiud 70d7, dat. sg. demnigud 25b6, 35b17, 89b11, 89b12, demnigiud
50c19, 51d5, dat. sg. with intensive prefix imm imdemnigud 89d7.
Deadj. < deimin sure, certain.
LEIA D-43-4, DIL s.v. deimnigidir 9: 81, deimin 8: 77.
deithidnigidir concerns, troubles

Milan: pres. ind. pass. sg. with infix. pron. 1sg. nomdeithidnigther
gl. sollicitor 29d7.
Denom. < deithiden care, concern, generally bestowed on a person
or a thing, which would logically lead to a transitive verb. However, this
verb seems to have simultaneously developed an intransitive meaning.
LEIA D-46, DIL s.v. deithidnigidir 16: 35, deithiden 15: 57.
193
delbaigidir shapes, forms

Milan: pres. ind. rel. pass. sg. nephdelbaigther 138c3, nad
delbaigther 138c3, part. nephdelbaigthe 138c3 (Note Thes.: MS. neph
delbaigthe).
Denom. < delb form, figure. The three occurrences above are all in
the same gloss, glossing Lat. informem formless. There is only one
attestation of this verb outside Milan found in DIL. The semantics of
delbaigidir are briefly mentioned on page 92.
LEIA D-47, DIL s.v. delbaigid 18: 38, delb 17: 28.
drgid spreads (coverings, a couch), makes a bed

PIE *h3re- stretch, point, caus. * h3ro-i e- (LIV2)
CC *rog-ardrg(a) (proposes,
intends): Wrzburg: perf. 3sg. rel.
arrudrgestar 4c13; pass. perf. 3sg. arrudrged 2b10, Milan: perf. 1sg.
arrudergus 51a13; drgid: Milan: VN (m u) drgud, dat. sg. drgud 21a3,
21c7, dat. pl. dergudaib 20c1.
Pedersen states that although this verb inflects as a denominative, it
is originally a compound of *reg- binds, *di-ess-reg-, and therefore a
primary verb that took on weak flexion. In this case, reg- was reflexive
(deponent), whereas in its compound air-di-ess-reg- it was transitive.
DIL states that this verb in later texts sometimes occurs with nonpalatal -g-, thereby implying that that the -g- is palatal. It is however more
likely, as Thurneysen and Dillon have suggested, that drgid is originally a
compound of a causative rogid (page 236 below) of a root *h3re- and that
the -g- is therefore regularly non-palatal. LEIA suggests that rogid is an
iterative, rather than a causative, which certainly suits the meaning better,
but has no further consequences for the morphology of either rogid or
drgid. Occasional examples of drgid with palatal -g- are according to
Dillon explained by confusion with dorig, dirig abandons, but could
also be due synchronic palatalisation in analogy to other AII verbs. The
same confusion appears in the compound arderga.
194

Ped. ii 596, IEW 854-7, LEIA D-58, R-39, LIV2 304-5, DIL s.v.
drgid 38: 17, ar-drga 389: 71, DIL s.v. rogid 89: 45, Dillon 1971: 205-9,
Thurneysen 1913: 35.
dtnaigidir gnashes
Milan: impf. 3pl. noditnaigtis (in DIL without fada) gl. stridebant
54d20.
Denom. < dtnach gnashing of teeth. Although I usually derive
verbs from their non-derived base, as discussed above, section 3.4.1, page
67, that is problematic in this case, because a non-derived base is not
attested. It could be argued that a word *dtne might have existed, however
it is not extant in any of the later phases of the language either. I have
therefore chosen to derive the verb from dtnach instead, keeping in mind
that the non-attested *dtne may well have been a virtual intermediate in
this case.
The verb is hapax legomenon.
LEIA D-62-3, DIL s.v. dtnaigid 51: 22, dtnach 51:20.
deugaigidir drinks
Milan: fut. 3pl. deugaigfit gl. potabunt 30c18.
Denom. < deog drink, draught. Hapax legomenon. LEIA does not
register the verb.
LEIA D-51-2, DIL s.v. deugaigid 51: 32, deog 26: 10.
danaigidir causes to hasten, hurries, makes haste

Milan: pres. ind. 2pl. dianaigthe 49d9, impf. 3pl. nudianaigtis
54d15, sec. fut. 3sg. nundianaigfed 91a10, perf. pass. sg. rudianaiged
98b14, part. comp. dianaigthiu 91a13.
Deadj. < dan swift, rapid. The semantics of danaigidir have been
briefly discussed on page 87.
LEIA D-68, DIL s.v. danaigid 63: 65, dan 62: 12.
195

dchsnigidir exists, is extant
dchsnigidir: Wrzburg: pres. ind. 2sg. rel. dxnigedar 4c24, 3sg.
with interrog. indixnigedar 3c32, with the copula and nad, isnaddixnigedar
9c14, 3pl. rel. dxnigetar 12b7, Milan: pres. ind. 3sg. cate dixnigedar
80c2, 103d24, rel. ndixnigedar 23a1, neg. rel. naddixnigedar 55c9, nad
dxnigider 55c10, pres. subj. 2sg. rel. nundixnaigther 75c9; indxnigedar
(is in, exists in): Milan: pres. ind. 3sg. indxnigedar 19b14, indixnigedar
20c6 (2x), indixnigedar 37b17, 51c16.
Denom. < *dchsain, possibly VN of docoissin there is, there are.
Dchsnigidir shows double syncope: First in the VN of docoissin,
*dchsain. Then when the *-sag- suffix was added, it was synchronically
syncopated again. Therefore this form must have been created after
syncope, i.e. after the middle of the 6th cent and can be found in the
Chronology on page 82. The semantics of this verb have been briefly
Ped. ii 436, LEIA D-76, DIL s.v. dchsnigidir 81: 58, in-dxnigedar
231: 13, do-coissin 232: 61, GOI 782.
dlmainigidir is free, unoccupied

Milan: perf. 3pl. rondilmainaigset gl. uacasse 76a8.
Deadj. < dlmain free from restraint, without attachment, < di +
loman in meaning b) restraint (DIL). Outside of Milan, this verb occurs
twice in legal transcripts.
Ped. ii 561, LEIA D-86, DIL s.v. dlmainigidir 110: 48, dlmain 109:
18, loman 192: 49.
dlsigidir makes forfeit, renounces

Milan: VN (m u) dlsigud, dat. sg. dlsigud 74d13.
Deadj. < dles forfeited to, given over.
LEIA D-84, DIL s.v. dlsigid 112: 59, dles 103: 14.
dummussaigidir is proud
Milan: impf. nodiummussaigtis gl. superbientes 136b5, pres. subj.
3sg. arnadiummussaigedar gl. turgescat 68c7.
196

Denom. < dummus
lit overestimating,
hence vainglory,
arrogance, VN of *d-uts-med- (Ped.). The verb only occurs in Milan.

Ped. ii 577-9, LEIA D-105, DIL s.v. dummussaigidir 155: 22,
dummus 154: 59.
dlmigidir nucleates, masses together

Wrzburg: pass. perf. 3sg. with infix. pron. 1pl. and emp. pron. 1pl.
rondlmigedni gl. Etenim in uno Spiritu omnes nos in unum corpus babtizati
sumus 12a15.
Denom. < dlm, mass, nucleus. Hapax legomenon.
LEIA D-109-10, DIL s.v. dlmigid 168: 1, dlm 167: 42.
doirci causes, effects

PIE *h2nek- get, achieve (LIV2), caus. *h2nok-i eCC *onk-contirci: Milan: perf. 2sg. rtaircissiu (DIL as this as
ortaircissiu, which seems a misreading of the con-sign. They give no
translation, only refer to this gloss) 77d1, part. comthaircidi, comtaircidi
135a5. doirci: Wrzburg: pres. ind. 3sg. doirci 3d19, 15b28, doircci
12c26, nitirci 27a28, trci 25a4, an tairci 12c31, 3pl. nitircet 30b12,
ntarcat 31d10, tircet 3b8, perf. dorricc 3c29, dorrricc 3c15, doraricc
19b23, impv. 3sg. tarced 23c14, natirged 23c13. Milan: pres. ind. 3sg. rel.
duairci 76a16, 3pl. rel. dunirceat 86c14, impv. 2pl. taircid 48b15, impf.
3sg. dunairced 35a12, pres. subj. 1pl. coduaircem 35b3, 3sg. rel. dunirrcea
78b22, dunaircea 85c2, 3pl. dunaircet 46b15, pass. sg. ntaircither 17d5,
sec. fut. 3sg. duaircibed 29c10, rel. dunaircibed 29c11 (this form found
under another gloss in DIL), 100b7, perf. 3sg. durairric 33b20, verb. nec.
tircidi 19a4, taircithi 18a6, nom. pl. taircidi 34d14, part. tircide 27c20,
gerund tircidi 39d24, 50a4, VN trcud, acc. sg. trcud 74a1, dat. sg.
tairciud 47b4, 121c2; fotairci (substitutes, supplies): Milan: pres. ind.
pass. sg. fotairrcether 108c15, pl. fotairciter 122b2; intirci (inflicts,
brings (upon)): Milan: past subj. 3pl. intaircitis 33a3, pret. 3sg. indaricc
54b30.
197

Caus. If these were compounds of icc, they would have been the
only compounds of this root to inflect weak and they should have had a
preterite based on anaic rather than the attested s-preterite. Pedersen
suggests deriving from the VN tairecc providing, supplying, however this
would not explain the anomalous form in Wb. 31d10 above, where the pres.
ind. 3pl. is tarcat. If this verb were derived from tairecc, palatalisation
would have been regular. The lack of palatalisation makes sense if this verb
is a causative. The palatal forms can then by explained by analogy to other
compounds of icc and/or other AII verbs. The W deirwng46 is consistent
with a verb derived from *to-ad-ro-onk-. The meaning of doirci causes,
effects is also consistent with that of a causative.
Note that reflexes of a Proto-Indo-European causative formation
*h2nok-i e- are not attested. If a Proto-Indo-European causative formation
would have existed, it would have been *h2nok-i e-, but the only causative
formation attested is that in Celtic, which does not continue a Proto-IndoEuropean formation.
The original shape of icc itself is a matter of controversy. For
detailed discussion, cf. McCone 1991a 2-3, 1991b: 47-52, Schrijver 1993:
39-42, 1998: 468-71, LIV2 283 n. 7, Schumacher 2004: 200-4.
Contirci is not translated in DIL, Thes. translates gloss as thou
hast conferred.
Ped. ii 553, IEW 316-8, LEIA C-201, D-114-5, LIV2 282-3, DIL s.v.
con-tirci 464: 25, do-irci 185: 70, fo-tairci 390: 36, in-tirci 286: 69,
tairecc 36:52.
doaissilbi refers, ascribes

See selbaid AI, page 152.
dobidci shoots, casts

PIE *bherh- rise, *bhor-i e- (LIV2)
CC *borg--
46
Note that the citation form is hypothetical, since only two forms are attested, both too
obscure to draw any definite conclusions about its meaning. However, both these forms are
consistent with the derivation given for OI doirci.
198

Cf. MW bwrw, throw
Milan: pres. ind. 3pl. dubidcet 93d11, impf. 3sg. rel. andumbidced
53d7, past subj. 3pl. dubidctis 30b1, pret 3sg. with infix. pron. 3sg. dambidc
58c4, perf. 3sg. dorrubidc 40d9, dorobidc (darobidc in DIL) 58c3, VN
dbirciud, acc. sg. dibirciud 99d1, dat. sg. dibiercud (this form not in DIL)
26d6, dibirciud 58c6.
Pedersen (i 88) states that the the prototonic dbairg is a different
root from bidci and connects this to W bwrw. LEIA also separates the
roots, deriving dobidci from bidg, bedg start, spring, leap.
Thurneysen (GOI) connects dbairg to bidci with PIE *-zg- > OI
-rg-, while *-z- before *-d- disappeared. According to Schrijver, since the
-b- in dbairg is non-palatal, the verbal root can not have contained a front
vowel, contrary to that in bidci < *bisg- or *besg-. Moreover, OI cuit part,
sochuide multitude < *kwesd- shows that *-s- in unaccented syllables is
lost, rather than that it developed into -r- as Thurneysen claimed.
Geiriadur Prifsygol Cymru reconstructs a root *borg- rather than
*burg-, which is supported by the Old Irish VN dbirciud (see above Ml.
58c6, 99d1) which points to dberg with -e- due to palatal -rc-, but this is
hardly conclusive. If this reconstruction is correct, it points towards a
causative / iterative formation of Proto-Indo-European root *bherh-, which
would be in line with the weak i-flexion of OI dibairg. It is also possible
from the Welsh side of things, since there are a number of Welsh words
showing raising of *o to *u (Schrijver p. 67). This is the reconstruction
found in LIV2. Schulze-Thulin follows this analysis, and also points out that
the earlier considered *burg- cannot be connected with any known ProtoIndo-European root (suggestions would be ?**b(h)erg(h)- or ?**b(h)erg(h)-).
The meaning of the verb, shoots, casts is also in line with the causative
analysis: making high, raising.
Whereas the above analysis explains -dibairg, it does not explain
dobidci, which remains unexplained.
Ped. i 88, ii 476, IEW 140-1, LEIA D-121, LIV2 78-9, GOI 218,
Schrijver 1995: 55-6, 67, Schumacher 2000: 158, Schulze-Thulin 2001: 120,
DIL s.v. dobidci 214: 40, bedg 51: 5, cuit 604: 86, sochaide 317: 81,
dbirciud 70: 63.
199
dochenlaigidir degenerates
Cf. MW digenedylhau (Joseph).
Milan: perf. 1sg. rondoichenelaigsiur gl. degenerans 44b36.
Deadj. < dochenuil low-born, degenerate, cmpd. of cenl
kindred, race. Hapax legomenon. Despite the Middle Welsh cognate, it is
possible that this is a calque on the Latin degenerare which it glosses, since
the verbs are formed in the same way. Calquing has been discussed in
LEIA D-125, DIL s.v. cenl 116: 16, dochenlach 222: 41,
dochenlaigidir 222: 51, dochenuil 222: 53, Joseph 1987: 147.
dochraidigidir defiles
Milan: pres. ind. pass. sg. dochraidigthir gl. turpatur 28b2,
nadndochridichther gl. fedari 35d15.
Deadj. < dochraid unshapely, unsightly, cmpd. of cruth form,
shape. The verb only occurs in Milan.
LEIA D-127, DIL s.v. dochraidigidir 226: 35, dochraid 225: 54.
dofortai pours, pours out

PIE *ert- turns oneself, caus. *ort-i e- (LIV2)
CC *ort-Milan: pres. subj. pass. pl. dofortatar 124d12, past subj. 3sg.
dofortad 36d22.
Caus. The e-grade of this root is attested in OI adferta (page 261,
see discussion above, page 83). Adferta has a Welsh cognate gwerthu, but
the causative is not attested in Brittonic.
Ped. ii 526, IEW 1150-1, LEIA D-143-4, LIV2 691, DIL s.v. ad-ferta
53: 32, do-fortai 273: 44.
dogailsigidir grieves, is sad

Milan: pres. ind. 3pl. rel. dogailsegetar 87b17.
200

Denom. < dogailse grief, sadness < dogail anguish, affliction?.
The verb is attested once in Milan and once in the Vita tripartita Sancti
Patricii.
LEIA D-147, DIL s.v. dogailse 282: 25, dogail 282: 23,
dogailsigidir 282: 43.
*doirmmaigidir becomes worthless, vile (translation mine)

Milan:
perf.
3sg.
rondoirmmaigestar
(Note
Thes.:
MS.
rondoirammamaigestar) gl. uiluerit 101a8, 1pl. rondoirmmaigsem (Note

Thes.: MS. rondoirmaigsem: corr. Ascoli) gl. uiluisse 93d9.
Denom. < dormm servile yoke, servitude. This verb is found
under dor in DIL, as a denominative of the compound dormm. DIL has
no translation. The verb only occurs in Milan and glosses Lat. uilescere
become worthless, vile. Since dormm means servile yoke, servitude, it
seems likely that the verb was coined to translate the Latin. It is not a simple
calque, because although OI dor translates Latin uilis, it is OI dormm that
is used to form the verb, not dor itself. Calquing is discussed in section
3.4.3, pages 69ff.
LEIA D-158, DIL s.v. dor 320: 56, 321: 86 - 322: 3.
dorchaigidir grows dark, is eclipsed

frisdorchaigedar (darkens, obscures): Milan: pres. subj. pass. sg.
frisndorchaigther 73c3
Deadj. < dorchae dark, gloomy. Frisdorchaigedar is a hapax
legomenon formed to gloss Lat. obtenebrare darken, make dark (DIL).
The semantics of dorchaigidir have been briefly discussed on page 89.
LEIA D-175, DIL s.v. dorchaigid 356: 3, fris-dorchaigedar 425: 66,
dorchae 354: 84.
dosc(u)lai finds out, ascertains

Milan: pres. ind. 3sg. dusceulai 68d12, 102a22, pres. subj. 3sg.
dusceula 95c4, 3pl. dussceulat 83b8, fut. 1sg. dusceulub 59a2, 3pl. neg.
with infix. pron. 2 sg. toscelfat 107c11, VN (m u) taisclad, dat. sg.
thaiscelud 90c1.
201

Denom. < scl story, narration. Simplex not attested.
Ped. ii 617, IEW 898, LEIA D-183, DIL s.v. do-sc(u)lai 372: 33,
scl 83: 43.
dumaigidir piles up, exaggerates

Milan: perf. 3sg. rodumaigestar gl. exaggerauit 55d3, gl. exaggerat
80b3, rodumagestar gl. exaggerauit 83d1, VN (m u) dumugud, gen. sg.
dumichtho gl. exacerationis 44d4, dat. sg. dumugud gl. exaggerenter 77d2,
gl. emfasses 89d6, acc./dat. sg. dumugud gl. exageratione 76a3.
Denom. < duma mound, tumulus. DIL suggests deriving this from
dumach sandbank, also used for bank, mound in general. However, as
discussed above, section 3.4.1, page 67, if a verb could be from a derived
base in -ch- or a non-derived base, the principle of economy has been
applied and the verb has been taken from the non-derived base. I have
therefore followed LEIA and derived this verb from duma rather than
dumach. The verb is only attested in Milan and could be a calque on Lat.
exaggerare heap up, accumulate, which it glosses. Both the Latin and the
Old Irish are derived from a noun meaning heap, pile (Lat. agger) and the
verb deriving from it includes the metaphorical meaning of exaggerates in
both languages. Calquing has been discussed in section 3.4.3, pages 69ff.
LEIA D-221-2, DIL s.v. dumaigidir 449: 13, dumach 448: 79, duma
448: 33.
durnid smites, buffets (DIL)

Cf. MW dyrnu thresh, thumb, MB dournaff bash, thumb.
Wrzburg: pres. ind. 3sg. rel. with infix. pron. 1sg. nomdurni gl.
angelus Satanae ut me colaphizet 17d26.
Denom. < dorn hand, fist. Although this is hapax legomenon, I
would argue that this is a Celtic coinage. The Brittonic cognates are an
argument for this position. Although the i-verbs are productive in Old Irish,
the category is certainly not as productive as the -igidir verbs. Moreover, the
meaning is an elementary one, using ones fist to hit someone, whose
semantic field has broadened to include hitting someone by any means. This
indirectness is another argument for greater age, since in Old Irish, indirect
202

relationships between base noun and verb gives -igidir verbs, not a-verbs or
i-verbs. Durnid can be found in the Chronology on page 80.
IEW 203, LEIA D-177-8, DIL s.v. durnid 454: 33, dorn 360: 6.
echtrannaigidir alienates, is alienated

Milan:
perf.
3pl.
roechtrannaigsetar
66d2,
pass.
pl.
rel.
roechtrannaigthea 66d1.
Deadj. < echtrann strange, foreign. The verb only occurs in Milan.
DIL s.v. echtrann 37: 3, echtrannaigidir 37: 59.
cmailtigidir is insolent
Milan: pres. subj. 3sg. cmailtigidir gl. insolescit 28a9, 2pl.
nuncmailtigid (Note Thes.: MS. nunamailtigi) gl. insolescere 82a6, 3pl.
arna ecmailtigetar (Note Thes.: MS. ecmaittigetar: corr. Ascoli) gl. n
insolescant 27c6.
Deadj. < cmailt unquiet, restless. The semantics of cmailtigidir
have been briefly discussed on page 92. The verb is only attested in the
Milan Glosses.
DIL s.v. cmailtigidir 42: 56, cmailt 42: 38.
cndaigidir slanders, reviles

Wrzburg: pass. pres. ind. rel. with infix. pron. 1pl. and emp. pron.
1pl. eicndichtherni 2a11; Milan: pres. ind. 3pl. cndigitir 126d16.
Denom. < cndach reviling, slandering. As has been discussed
above, section 3.4.1, page 67, where possible a verb has been derived from
its non-derived base. However, for cndaigidir, a non-derived base is not
attested. Therefore, the verb has been derived from cndach with loss of
-ach- by haplology. The verb is not attested outside of the Milan Glosses.
DIL s.v. cndaigidir 48: 54, cndach 47: 61.
icnigidir compels, constrains

conicnigedar (forces, compels): Wrzburg: pass. pres. ind. 3sg.
nd comeicnigther 28b1, Milan: pres. ind. pass. sg. rel. cotammeicnigther
21b10.
203

Denom. < icen necessity, compulsion.
DIL s.v. icnigid 74: 64, icen 66: 57.
igid cries out, screams

arigi (cries out, laments): Milan: pres. ind. 3sg. rel. aranegi
18d1, argi 29d10, 30d5, aregi 31c4, 3pl. rel. araneiget 20b8, rel. araget
29b3, (rel.) arget (this is translated rel. by Thes., but this is unnecessary)
54a3, rel. araneget (Note Thes.: n is infixed because of the preceding cum
= Ir. lasse.47) 97d9, pres. subj. 3sg. rel. arangea 95c7, perf. 3sg.
ciarudreig 50d1, arruig 54b29, arroig (this form not in DIL) 58b14,
impf. / past subj. 3sg. araneged 24c11, VN (f ) airigem, eregem 35a20,
ergem 36b15, 90b19, acc. sg. erigim 55d11, gen. sg. erigme 27d5, dat. sg.
eregim 44b4, 74b1, 139a13, acc. pl. erigmea 26b22.
There is very little to base igid on. IEW has a root *eig-, *oigwail with Gk. the complaining and the Gothic denom. aihtrn
beg, but both the Greek and the Gothic are o-grades and both are
characterised by a dental suffix. It is not at all certain if the Old Irish verb
can be connected to these words or even if there is a basis for a Proto-IndoEuropean root at all.
In Old Irish itself, the nouns igem and igme both mean scream,
cry and both contain the element ig-, which is a good candidate as base for
this verb, but an actual base noun or adjective is not attested, and it cannot
be shown whether this element is itself primary or secondary.
Ped. ii 509, IEW 298, DIL s.v. igid 78: 29, ar-igi 391: 34, arneget406: 79, igem 77: 64, igme 78: 66.
eltesaigidir makes warm or lukewarm

Milan: pres. ind. 3sg. neg. rel. ndtesaigedar gl. tepefacit 94b21
(Note Thes.: MS. nadeltesaigedar: corr. Thurneysen, but note DIL: the
emendation in Thes. is shown by eltes, etc. to be unnecessary).
47
It is unusual for Thes. to explain relative nasalisation in a form, but in this case, this form
could belong to both arigi cries out, laments and to arneget prays. Thes. note shows
their opinion that it is arigi.
204

Denom. < eltes tepidity, < elte luke-warm. Both noun and verb
are hapax legomenon (although not occurring in the same text).
DIL s.v. eltesaigidir 117: 16, eltes 117: 8, elte 117: 1.
engraicigidir takes the place of

Milan: VN engracugud gl. pronominibus 114a4.
Denom. < engraic place, stead (DIL).
DIL s.v. engraicigidir 140: 53, engraic 140: 43.
nirtigidir weakens, enfeebles

Milan: past subj. 3pl. nonnartaigtis gl. eneruare 34d2.
Deadj. < nirt without strength, weak, < nert strength, might.
This could be a calque on Lat. eneruare weaken, which it glosses, if we
allow for a pseudo-etymological reading of the Latin verb as e-neru- > OI nert-. Calquing has been discussing in section 3.4.3, pages 69ff.
DIL s.v. nirtigidir 141: 54, nirt 141: 14, nert 35: 18.
rassaigidir makes void, invalidates

Wrzburg: impv. 3pl. naherassiget gl. inconfussibilem 30b10,
Milan: pres. ind. 3sg. rel. aerasaigthaer 131c9, with infix. pron. 3sg.
nosnerassaigedar gl. inritae 51b27, impv. 3sg. rasaiged gl. reprobat
51d17, perf. 3pl. rorasaigset 131c9, part. rassaigthe, nersaigthi gl. inrits
conatibus 127c25, rassaigthe gl. adprobatum (Note Thes.: rassaigthe
means reprobatum not adprobatum. The glossator seems to have
misunderstood his text) 24d1, nom. pl. rasaigthi gl. irrita 34d3.
Deadj. < raise useless, insignificant. DIL has the meaning of the
adjective with a question mark, but it is confirmed by the verb. It is only
attested once outside the Glosses.
DIL s.v. raise 161: 52, rassaigidir 161: 60.
erbirigithir charges (?); explains

Milan: pres. ind. 3sg. erbirigithir gl. causetur 35b6, erbirigid (Note
Thes.: MS. erbirigidid) gl. causatur 90a7, pass. sg. erbirigther gl.
causatur 54b10, perf. 1pl. roerbirigsem gl. causati enim sumus 35a5.
205

Denom. < airbire act of reproaching, reproach. DIL translates with
charges (?), explains. The only four attestations of this verb are found in
Ml. and gloss Lat. causari allege an excuse in all instances. Ml. 35a5 and
35b6 are one word glosses; Ml. 54b10 is followed by .i. connamanairi i.e.
that it may not succeed, which does not shed any more light on the meaning
of the verb. Ml. 90a7 is the only gloss that is followed by a more lengthy
explanation: .i. dobersom innaccuis aratartad indermitiu feid mrsa du
solmain... i.e. he gives the cause why this great reverence had been given to
Solomon... The verb erbirigithir is here explained by dobeir innaccuis
gives cause. Although DILs tentative translation charges, based on the
meaning of the base airbire reproach is not illogical, it is not borne out by
any of the attestations. In this case, the verb seems more a response to the
base noun, rather than being a direct extension of it, as is more common.
DIL s.v. erbirigithir 163: 64, airbire169: 75.
erlissaigidir loathes
Milan: pres. ind. pass. sg. n erlissaigther gl. numquam tamen
fastiditur 62a9.
Denom. < lius loathing, disgust. Hapax legomenon. This could be a
calque on Lat. fastidire, which it glosses. Both verbs are derived from a
noun meaning disgust (Lat. fastidium). Calquing has been discussed in
DIL s.v. erlissaigidir 171: 30, lius 171: 26.
espanaigidir is idle, vain

Milan:
past
subj.
3pl.
nonespanaigtis
(Note
Thes.:
MS.
nonepanaigtis) gl. otiari 93a3.

Denom. < esba uselessness, vanity, < be benefit, profit. Hapax
legomenon.
DIL s.v. espanaigid 202: 22, esba 181: 60, be 6: 75.
206

taigidir emulates, envies
Wrzburg: pres. ind. 3pl. with infix. pron. 2pl.: nobttigetar 19d27,
Milan: pres. subj. 2sg. -etaigther 56b31, nonetaigther 56b39, with infix.
pron. 1pl. nontagther 56b17.
Denom. < t emulation, jealousy.
DIL s.v. taigidir 209: 37, t 207: 40.
etargnaigidir makes known, denotes

Milan:
perf.
3sg.
roetarcnaigestar
gl.
notauit
32b5;
remietarcnaigedar (distuingishes, makes known beforehand): Milan:

pres. subj. 3sg. remnetarcnaigedar gl. prnotare 18c12.
Denom. < etargne knowledge (of), acquaintance (with), VN of
etargnn. The semantic development is from etargnn finds out, learns,
knows to the thing known, i.e. knowledge, etargne, to passing on that
knowledge etargnaigidir (which itself also has the meaning of knows).
The compound remietarcnaigedar could be a calque on Lat. prnotare
mark before, which it glosses: preverb and stem match each other exactly.
Calquing has been discussed in section 3.4.3, pages 69ff.
DIL s.v. etargnaigidir 218: 5, remietarcnaigedar 43: 37, etargne
218: 25, etargnn 219: 5, etargnaigidir 218: 5.
tendigidir loosens, relaxes

Milan: perf. 3sg. arrunetendiged 113c2.
Deadj. < *tend not strong, loose, relaxed, < tend strong, firm.
Hapax legomenon. The derivation is based on the fact that both ro and the
nasal are infixed before -. This shows this verb is treated as a simplex and
therefore must have been derived from a non-attested *tend.
DIL s.v. tendigidir 227: 17, tend 143: 39.
trummaigidir lightens, alleviates

Milan: fut. 3sg. trummaigfid gl. subleuabit 89c9, perf. pass. sg. with
infix. pron. 1sg. rommtrummaiged gl. releuatus 48b5, 3pl. rotrummaigset
gl. laxati sunt 113c8, part. gen. sg. trummaigthi (not in DIL) gl.
207

leuandorum 86d4, VN trummugud, dat. sg. trummugud gl. tenuitate
59a23.
Deadj. < tromm not heavy, light, neg. of trom heavy. The verb
only occurs in Milan.
DIL s.v. trummaigid 252: 13, tromm 251: 11, 3. trom 317: 17.
tailngigidir chafes (under), is out of patience (with)

Milan: pres. ind. 3sg. rel. etuailngigedar gl. indignatur 69d7, perf.
3sg. rel. arruntuailngigestar (Note Thes.: MS. arruntuailngistar) gl.
indignante 62b22, part. tailngigthe, etuailgigthe gl. indignans 136c6.
Deadj. < talaing intolerable, insufferable, neg. of talaing able
(to), capable (of). The verb is only attested in Milan.
DIL s.v. tailngigidir 255: 36, talaing 255: 44, talaing 339: 22.
filtigidir rejoices, is glad

Milan: pres. ind. 1pl. failtigmi 129d7, impf. 3pl. nufailtigtis 61a2,
pres. subj. 1sg. nufailtiger 46b16, 3pl. conifailtigetar 55b6, VN filtigud,
failtigiud 43c6.
Denom. < filte joy, happiness.
DIL s.v. filtigidir 25: 75, filte 24: 48.
fairsingigidir widens, enlarges

Milan: VN fairsingigud, gen. sg. forsingigthea 109c6.
Deadj. < fairsing ample, broad.
DIL s.v. fairsingid 29: 12, fairsing 28: 39, fairsingi(g)ud 29: 23.
fsaigidir empties, makes void

Wrzburg: perf. pass. 3sg. rofsiged 15a33.
Deadj. < fs empty, void.
DIL s.v. fsaigid 44: 36, 1. fs 42: 58.
feidligidir remains, abides

Wrzburg: pres. ind. 3sg. nfedligedar 2c3, pres. subj. 1pl. rel.
cofedligmer 6d11, Milan: pres. ind. pass. pl. feidligte 21d4, pres. subj. 3sg.
208

feidliged 136d2, fut. 3sg. feidligfid 128a7, sec. fut. 3pl. rofeidlifitis 108b05,
perf. 1pl. neg. rel. nrufeidligsemmar 105a4, 3pl. neg. rel. nadrofeidligset
105a4, VN (m u) feidligud 15a6, 63b15, gen. sg. feidligtheo 15a5;
tremifeidligedar (endures, remains): Milan: pres. ind. 3pl. n
tremfeidliget 21d4, 21d5, VN treimfeidligud 21d4.
Deadj. < feidil lasting, enduring.
DIL s.v. feidligidir 63: 3, tremifeidligedar 295: 82, feidil 62: 38.
feochraigidir grows fierce, wild

Milan: perf. 3pl. rofeuchraigset gl. efferati sunt 114b6, part.
feuchraigthi gl. seuitiosi 124c9.
Deadj. < feochair, stern, severe. Note that DIL has the meanings
fierce, indomitable, wild as later, although the verb seems to be derived
within that semantic range. It is likely that although those meanings are not
attested within Old Irish sources, they were present in the language. The
verb could be a calque on Latin efferare make wild, which it glosses.
DIL s.v. feochair 78: 50, feochraigid 79: 4.
fergaigidir grows angry

Milan: pres. ind. 3sg. n fercaigedar 24b18.
Denom. < ferg anger, wrath.
DIL s.v. fergaigid(ir) 91: 61, 1. ferg 90: 77.
fachaigidir incurs a debt, makes liable

Milan: pres. subj. 3sg. rel. diafiachaigedar 44b3.
Denom. < fach obligation, payment due. The semantics of this
verb have been briefly discussed on page 91.
DIL s.v. fachaigid(ir) 110: 4, 1. fach 108: 9.

Milan:
pres.
ind.
3pl.
(rel.)
with
infix.
pron.
1sg.
nodamfindbadaigetar gl. iecientes m 39d10, impf. 3pl. with infix. pron.

1sg. numfindbad(a)igtisse gl. beatificantes me 39d14.
209

Denom. < findbad bliss, felicity, possibly a compound of finn
white, bright and bith world (DIL, GOI). The verb only occurs in Milan.
DIL s.v. findbadaigidir 138: 32, findbad 138: 10, finn 141: 60, 1.
bith 106: 56, GOI 124.
frinaigidir justifies, makes righteous

Wrzburg: pres. ind. 3sg. with suff. pron. 3sg. m. firianichthi 2b28,
rel. firianigedar 19b1, with infix. pron. 1pl. nodonfirianigedar 19a14, perf.
3sg. with infix. pron. 3sg. m. rafirianigestar 19b13, pass.: perf. 3sg. rel.
rofrianiged 2b24, rofirianiged 2c8, Milan: perf. 3sg. with infix. pron. 3sg.
rel. rondfrianaigestar 19d16.
Deadj. < frn just, righteous, loan from W gwirion (GOI). The
verb is not attested outside of the Glosses.
GOI 917, DIL s.v. frinaigidir 151: 8, frn 150: 8.
foammmaigedar subjugates
Milan:
pres.
ind.
3sg.
foammamaigedar
113c6,
pass.
pl.
foammamaigter 26a8, perf. 3sg. with infix. pron. 3pl. fosroammamigestar

(Note Thes.: MS. fosro ammamigestar with the first s over the line) 67b24,
part. foammmaigthe, gen. pl. foammamaigthe 43c5, 105b1, VN
foammmugud, foammamugud 67c2, acc. sg. foammamugud 26a8, dat. sg.
foammamugud 28d5.
Denom. < mm yoke, could be formed on analogy to Lat.
subiugare subjugate, make subject (DIL), which has been discussed in
section 3.4.3 on pages 69ff. Simplex not attested. The verb only occurs in
Milan.
DIL s.v. fo-ammmaigedar 181: 76, mm 52: 25.
focridigedar girds
Milan: pres. ind. 3sg. focridigedar gl. accingat 35c32, pass. 3pl.
focridigter gl. accingi 135b1, impv. 2sg. fochridigthe gl. acingere 27c5,
pres. subj. 2sg. with infix. pron. 2sg. fotchridigther g. prcingere 101c3,
fut. pass. sg. with infix. pron. 1sg. fomchridichfidersa gl. accingar 31c7, VN
(m u) fochridigud, dat. sg. fochridigud gl. accinctum 37c9.
210

The simplex of this verb is not attested. Crualaoich derives this
from crde heart, which certainly works from a morphological point of
view, but is semantically problematic. Russell, in his forthcoming article on
British uocridem has connected OI focridigedar with uocridem and derived
it from a word *okridi o from the prepositional phrase *o krido under the
girdle, centrally.48 The exact etymological details of the root remain
inconclusive. Russell points out that whereas *kerd- bind is the most
attractive, the evidence is thin. However, the basic interpretation of
focridigedar and its implied nominal base is sound and I have followed it
here. The verb is treated as a compound in Old Irish, cf. section 3.5, pages
74f. It is not attested outside of the Milan Glosses.
IEW 579, LEIA C-236, DIL s.v. fo-cridigedar 208: 50, crde 527:
70, Crualaoich 1997: 247, 252, Russell forthcoming.
fograigidir emits a sound, sounds

fograigidir: Wrzburg: pres. ind. 3sg. rel. fogrigedar 12b27, 28,
Milan: pres. ind. 3pl. rel. fograigte 42c2, impf. 3pl. nofograigtis 54d20, VN
fograigiud, dat. sg. fograigiud 67a7; confograigedar (emits): Milan:
impf. 3sg. confograiged 50a9; *remifogrigethar (sings before): Milan:
pres. pass. remifogrigther 129b9.
Denom. < fogur sound.
DIL s.v. fograigid(ir) 239: 12, con-fograigedar 437: 76, fogur 240:
13, remi-fogrigethar 43: 40.
fodid sends
PIE *ei d- see, caus. *oi d-i e- (LIV2)
CC *oi d-Cf. Ved. -vedya- announce, OAv. -uuadaiia- allocate, OHG
weizen view, visit (LIV2).
fodid: Wrzburg: perf. 3sg. rel. with infix. pron. 3sg. n. rodfidi
7d2, pres. subj. 1sg. rel. nofid 23d1, pass.: pres. ind. 3sg. fitir 25a8, rel.
foiter 17a11, perf. 3sg. rofided 23a7, rofoided 32a25, 3pl. arrufoitea 9d14,
48
I am would like to thank Dr. Paul Russell for showing me his forthcoming article on
uocridem.
211

rofitea 27c35, n roitea 5a3; Milan: part. (io, i) fote 34c9, VN fot, dat.
sg. fit 34a6; asfodi (sends forth): Milan: pres. ind./subj. 3sg. with infix.
pron. 3pl. ardatuaisi 129b2, pass. sg. rel. assafoiter 48c8, part. (io, i)
esfote, nom. pl. esfoiti 48c8, acc. pl. esfoitiu 102b8, VN esfot, dat. sg.
esfit 44a10; dofodi (sends): Wrzburg: impv. 2pl. dofoidid (leg.
dafoidid) with infix. pron. 3sg. m. 14a17, pass. perf. 3sg. dofoided 32d14;
Milan: perf. 3sg. rel. with infix. pron. 1pl. donrid 53d9; infidi (sends,
sends down): Milan: part. infite 99b3; *remifodi (sends beforehand, in
front): Milan: pret. 3sg. arremiroid 31c9, part. remote, acc. sg. f. remfoiti
30c15, nom. pl. f. remfoiti 35a7, dat. pl. remfoitib 58a7.
Caus. Since this formation exists in several other IE languages, it is
likely that it is a Proto-Indo-European formation, though the possibility of
each language having formed it independently cannot be excluded. The
reinterpretation of this causative as causes to be seen rather than causes to
see is unusual.
Ped. i 359, ii 524-5, IEW 1125-7, LIV2 665-7, DIL s.v. fodid 243:
15, as-fodi 430: 23, do-fodi 270: 81, in-fidi 252: 53, *remi-fodi 43: 42,
tidid 25: 21, Sjoestedt 1925: 328.
foillsigidir shows, reveals

Wrzburg: pres. ind. 3sg. with suffix. pron. 3sg. n. foilsigthi 12a7,
perf. 3sg. with infix. pron. 3sg. f. and emp. pron. 3sg. m. rosfailsigestar som
31a9, 3pl. with infix. pron. 3sg. n. \idrofoilsigsetar 21c22, impv. 3sg.
foilsiged 12d41, pass.: perf. 3sg. rel. rofoilsiged 21c15, 13d26 (MS
rofollsiged); Milan: pres. ind. 3sg. foilsigidir 24c13, 53b27, nundfoilsigedar
42b18, 93a6, rel. foilsigedar 74d13, 85b11, pass. sg. foillsigthir 25c5, pres.
subj. 2pl. foilsigthe 56c2, past subj. 2sg. rofoilsigthea 61a3, fut. 3sg.
foilsigfid 69d8, perf. 3sg. rofailsigestar (this form not in DIL) 51d14-15,
rofoilsigestar 103d11, rafoilsigestar 109b02, rofoilsigestar 145b4, 3pl. rel.
arufilsigset 74a4, pass. sg. diarufoilsiged 136d9, part. (io, i) foillsigthe,
foilsichthe 72a9, failsichthe 91a5, VN foillsigud 74b1, foilsigud 70a5, dat.
sg. failsigud 36b16, 71c9, filsigud 78b7, foilsigud 131a1.
Deadj. < follus manifest, evident.
DIL s.v. foillsigid 247: 39, follus 273: 3.
212
foirbthigidir finishes, completes

Wrzburg: pres. ind. 3sg. rel. foirbthigedar 4d6, perfect. pres. subj.
1sg. rofoirbthiger 7a9, past subj. pass. sg. rel. nofoirbthichthe 32d5, Milan:
perf. 2sg. rofoirbthichser 43d17, rofoirbthichsir 50c13, VN foirbthigud
48b11.
Deadj. < foirbthe lit. finished, completed > complete, perfect,
part. of forfen (page 329).
DIL s.v. foirbthidid(ir) 254: 72, foirbthe 253: 47, for-fen 334: 35.
foirrgid presses hard, bears down

Wrzburg: pres. subj. 3pl. rel. ciforrgot 93b27.
Denom. < forrach act of oppressing, overpowering (Ped.). DIL
takes the one occurrence of subj. fooir as indication that the verb may
originally have been a strong compound forraig, but there is no other
evidence for this and it is more likely that this is analogical confusion.
Ped. ii 525-6, DIL s.v. foirrgid 258: 73, forrach 355: 84.
folcaid washes
PIE *elk- be or become damp or moist, *olk-i e- causes to become
damp: washes (Schulze-Thulin)
CC *olk-etarfolcai (washes between): Milan: pres. ind. 3sg. etirfolcai
81c1.
Caus. The compound etarfolcai imitates Lat. interluere flow
between, which it glosses. Interluere is a compound of luere wash, lave.
OI folcaid means washes and is compounded with the Old Irish equivalent
of the Lat. inter, eter. This has been discussed in section 3.4.3, pages 69ff.
The compound is hapax legomenon.
IEW 1145, LIV2 679, DIL s.v. folcaid 267: 11, etar-folcai 216: 30,
Schulze-Thulin 2001: 159.
follaigidir neglects, omits

Cf. MW gwallocau, Br. gwallekaat (Joseph).
213

Milan: pres. 3sg. rel. follaigedar 129a2a, past subj. 3 pl.
arnafollaigdis 113d5, part. follaigthe 130d12.
The principle of economy as usually applied in this thesis would
require this verb to be derived from faill neglect, negligence (cf. section
3.4.1., pages 67f.) However, the Brittonic forms above are both deadjectival,
*gwallog-ha-u, with gwallog being cognate to OI follach. If the Brittonic
and the Old Irish verbs are exact cognates, the Old Irish verb has undergone
dissimilatory haplology of -ch- and the verbs can be dated to Celtic (see
above, pages 75ff.). If, on the other hand, the principle of economy is
applied, these verbs cannot be exact cognates. They must then have been
formed in the individual languages and be ascribed to Old Irish and Middle
Welsh, Breton respectively. In the Chronology, this verb has been ascribed
to the Celtic phase of the language (page 81), although the alternative option
has been taken into account.
DIL s.v. follaigid 269: 60, follach 269: 34, 1. faill 23: 38, Joseph
1987: 147.
foniti mocks, derides

Milan: fut. 3sg. fonnitfea gl. subsannabit 17a4, VN fonitiud, dat. sg.
fonitiud gl. subsannatione 17a7.
Etymology unknown. The verb is a hapax legomenon. Pedersen
tentatively suggests a connection with the nessa verbs (adnessa, fonessa
etc), but neither vocalism nor consonantism are transparantly regular.
Simplex not attested. Note that both the attested verbal form and the verbal
noun in Milan gloss a form of subsannare mock, deride, which is a
derivation from the noun sanna mocking grimace with preverb sub-. Since
the Old Irish derivation is not known, however, this can not be claimed to be
a calque. Calquing has been discussed in section 3.4.3, pages 69ff.
Ped. ii 583-4, DIL s.v. fo-niti 290: 7.
forcomai keeps (in possession), preserves

Milan: pres. ind. 3sg. with infix. pron. 3sg. f. fortcomai 29a3, VN
forcomt, acc. sg. forcomet 49a11.
214

Denom. < coimm in meaning b keeping, protection, cf. coimge act
of protecting, safeguarding, cuimme protection.
DIL s.v. for-comai 323: 27, 1. coim 298: 40, coimge 302: 73, cuime
589: 64.
fortachtaigidir helps, succours

Milan: pres. ind. 3sg. with infix. pron. 3sg. n. nodafortachtaigedar
25c5, past subj. 3sg. mani fortachtaiged 88b16, part. (not in DIL)
fortachtaigthe, dat. pl. fortachtaigthib 63b13, verb. nec. fortachtigthi 64b2.
Denom. < fortacht act of aiding, succouring, VN of fortt.
DIL s.v. fortachtaigidir 364: 19, fortacht 363: 57, for-tt 367: 17.
foruimi sets, places

PIE *h1rem- becomes quiet, *h1rom-i e- (LIV2)
CC *rom-Wrzburg: pres. ind. 3sg. nifuirmi 8c1, Milan: pret 3sg. foruirim 2a6,
perf. pass. sg. foruirmed 74c20.
Caus. Simplex not attested.
Ped. ii 602, IEW 864, LIV2 252-3, DIL s.v. fo-ruimi 370: 61.
fothaigidir establishes, settles

Wrzburg: perf. pass. sg. with infix. pron. 2pl. robfothiged 21d6.
Denom. < fotha foundation, basis. Thurneysen has noted the
correspondence of the elements of fotha (*fo-suide) to Lat. subsidium. It is
likely that this noun was in fact a calque on Latin and if so, this verb will
not have been formed before contact with Latin. Calquing has been
DIL s.v. fothaigid 393: 77, fotha 392: 28, Thurneysen 1921: 300-3.
frepthanaigidir heals
Milan: perf. 1sg. rel. arrufrepthanaigisiur (Note Thes.: MS.
arrufrephanaigthiisiur)
gl.
medicatus
nephrepthanaigthe gl. inmedicabilem 76a17.
215
103a6,
part
acc.
sg.

Denom. < frepaid healing, remedy, VN of frisben. The verb only
occurs in Milan.
DIL s.v. frepthanaigidir 408: 74, frepaid 408: 55, fris-ben 424: 39.
frisbrudi rejects, refuses

PIE ?*bhredh- wade, caus. ?*bhrodh-i e- (LIV2)
CC *brod-Wrzburg: pres. ind. 1pl. frisbrdmor 15b22 (Note Thes.: Literal
and unidiomatic translation of the Latin deponent; Ped. II, 479), Milan:
pres. ind. 3sg. frisbrudi 28b8, 44b12, VN (m u) frithbruduth 76c4.
Etymology unknown and outside Pedersen not discussed, although
IEW (169) mentions the verb. I suggest this is a causative based on PIE
?*bhredh- wade (LIV2) with regular raising from -o- to -u- (cf. McCone
1996: 110). The attestations in IEW (164) make it clear that this wading
took place at a ford, which would have been a natural border. Causing
someone to wade, either through pursuit out of territory or by not allowing
someone into the territory in the first place, opens up the semantic
possibilities for this verb to come to mean reject, especially when
compounded with preverb fri- against.
Ped. ii 479, IEW 164, 169, LEIA B-102-3, LIV2 91, DIL s.v. frisbrudi 424: 61, McCone 1996: 110.
fuinid sets, sinks (of sun)

fuinid: Milan: VN (m o) fuined, gen. sg. fuinid 94b19; *dofuini
(meaning of VN descending, departing): Milan: VN (m? u) tiniud, acc.
sg. toined 42c30.
Dofuini is not in DIL, but it can be found under its VN tiniud.
There, the verb is cited as dofuin. This citation form implies that the verb is
strong. That is incorrect. Fuinid and dofuini are weak i-verbs. Fuinid has an
f-future and the other compound of this verb doairindi lets down, lowers,
also inflects weak. Pedersen states that fuinid itself is most likely already a
compounded verb of a primary root n- or ne-, and he connects it to Gk.
, meaning return home, escape. However, the root ne- does not
216

exist. Gk. comes from PIE *nes-, escape, return home unharmed
(LIV2 454-5).
Lewis and Pedersen reject Pedersens earlier analysis and register
fuinid under *ned-, compounded with fo-. This root *ned- consists,
according to them, of a preverb ni- plus the reduced grade of PIE *dh.
LEIA refers to both Pedersen and Lewis-Pedersen, but only cites the *nedreconstruction of Lewis-Pedersen. However, with this reconstruction there
are also problems. First of all, ni + reduced *dh would become **nid-, not
ned-, and secondly, there is neither nid nor ned in fuinid.
I argue that this is an iterative verb of a root *en-, possibly with
laryngeal. Two such roots are found in LIV2, *enh1- places, puts, lays
down and *enH- become fond of (LIV2 682-3). Although *enh1- at first
sight is the better fit, note that it is transitive. Therefore, an iterative formed
on this root would be transitive too, and fuinid is only attested as an
intransitive verb in Old Irish. Moreover, the root is speculative49 (this of
course also means that its transitivity is speculative). I do not dismiss the
root, but I prefer deriving fuinid from *enH- instead. Note that there are
Germanic forms meaning to live (OHG wonn) and the iterative to get
used to (ON venja). The sun sets every night in the place where it lives. This
semantic route is certainly not impossible, this is root securely attested and
it is ambitransitive so does not require the loss of transitivity.
Ped. ii 256-7, IEW 235-9, 766-7, LEIA T-97, LIV2 136-8, 454-5,
682-3, DIL s.v. fuinid 477: 54, toniud 223: 69, do-airindi 188: 85, Lewis &
Pedersen 1974: 383-4.
gaimigidir passes the winter

Cf. MB gouaffhat (Joseph).
Wrzburg: fut. 1sg. gaimigfer 14a9.
Denom. < gaim winter-storm, winter. Hapax legomenon. It is
difficult to ascertain the age of this verb. While it has a Middle Breton
cognate gouaffat, the *-sag- verbs were very productive both in the
49
Note that the AED has another etymology for the Albanian word that Klingenschmitts
root is based on: s.v. v ~ v to put: PA innovation *awena, a derivative in *-no- based on
*ae down, off.
217

Brittonic languages and in Old Irish. Moreover, gaimigidir does not have a
seek meaning. These verbs could have been independently formed at any
stage in the language. Despite the existence of a Latin verb hibernare, I
would hesitate to say this is a calque on Latin. If this verb were coined in the
Old Irish period, given the climate at the time, wintering would be a
common concept which a speaker would hardly need Latin to create a verb
for. Calquing has been discussed in section 3.4.3, pages 69ff.
DIL s.v. gaimigidir 24: 29, gaim 23: 71, Joseph 1987: 147.
gledid cries out (of inarticulate utterance of men and animals)

Wrzburg: pret. 3pl. rel. glidsete 93b14.
Denom. < gled shout, loud call. Gledid is further discussed in
comparison with achtaid cries out (page 138) and their respective places
in the scheme on page 59 above.
Ped. ii 538, DIL s.v. gledid 89: 32, gled 88: 64.
glasid moves, stirs

Milan: past subj. 3pl. neg. connachagluaistis 96c13.
The etymology of glasid is problematic. There are no Brittonic
cognates and there are no Old Irish nominal or adjectival bases that this
could conceivably be derived from. OI glas gloss, while fitting
morphologically, has no semantic relationship to this verb. The only
remotely possible Proto-Indo-European root would be *hleu- be jolly
(being moved with joy). It is only attested nominally and would leave the -sunexplained if this were a primary verb50. It would be possible to derive
glasid from a noun *gleu-sto-, although this is not attested in Old Irish or
indeed anywhere else.
IEW 451, DIL s.v. glasid 113: 38, 89: 32, gled 88: 64.
gnthaigidir is accustomed to, practises

Cf. W gnotaa is accustomed (Joseph).
Milan: perf. 3pl. rognaithaigsetar 34b2.
50
Although for other examples of unexplained -s- in verbs, cf. artaisi and asagsi.
218

Deadj. < gnth customary, usual. The verb can be found in the
chronology on page 81. Its semantics have been briefly discussed on page
88.
DIL s.v. gnthaigid 118: 23, gnth 116: 28, Joseph 1987: 138, 147.
goistigidir puts a noose around, hangs

Milan: impf. 3pl. with infix. pron. 1sg. nomgoistigtisse gl. laqueare
(Note Thes.: MS. lacerare: corr. Ascoli) 54b26; *ingoistigedar
(ensnares): Milan: pres. subj. pass. sg. ingoistigther 28c16.
Denom. < goiste halter, noose. The verb occurs only in Milan and
is a calque on Lat. laqueare, which it glosses. The Latin is denom.
< laqueus noose, snare. Both verbs are thus formed in the same way.
Moreover, the compound ingoistigedar is formed on the model of the Latin
compound illaqueare ensnare, which it glosses. Calquing has been
DIL s.v. goistigid 132: 48, ingoistigther 264: 78, goiste132: 7.
*gortaigidir is hungry
Milan: VN gortugud gl. esuriero, quod fieri non potest 70b12
Note: this is not DILs gortaigid salts, pickles, of which the VN is
also gortugud. This is verb not extant except for its VN. It is derived from
gortae hunger, famine.
DIL s.v. gortugud 140: 63, gortae 140: 13.
grschaigidir makes continuous

Milan: pres. ind. 3sg. grschaigidir gl. continuet 85a5, fut. 1sg.
grschaigfeasa gl. continuabo 122b10, part. greschaigte gl. diuturnorum
88b3.
Deadj. < grsach lasting, perpetual, < grs continuance, practice.
There is no translation of this verb in DIL. It is only attested in Milan and
glosses Lat. continuare make continuous, which fits the meaning of its
base, so I have used that as translation of this verb. Grschaigidir could be a
calque on continuare, since both verbs are derived from an adjective
219

meaning lasting, perpetual (Lat. continuus). Calquing has been discussed
DIL s.v. grschaigidir 156: 80, grsach 156: 58, 2. grs 156: 33.
gaigidir lies, tells a lie

Milan: pres. ind. 3sg. anguaigedar gl. mentiente 31b1, 3pl. guaigitir
gl. mentiente 31b1, pass. pl. niruguigter gl. mentiri 51c14.
Denom. < gu falsehood, misstatement. The verb only occurs in
Milan.
DIL s.v. gaigidir 168: 77, gu 51: 48.
guirid warms, burns

PIE *gher- becomes warm, caus. *ghor-i e- (LIV2)
CC *gor-- (McCone, Schulze-Thulin)
Milan: pres. ind. 3pl. guirit 39c24.
Caus. Cf. OI fogeir.
Ped. ii 537, IEW 493-5, LIV2 219-20, DIL s.v. guirid 174: 79,
fogeir 230: 64, Sjoestedt 1925: 327, McCone 1996: 38, Schulze-Thulin
2001: 146-7.
ilaigidir increases, multiplies

Milan: fut. 2sg. ilaigfe gl. multiplicasti 81c5, part. dat. pl. ilaigthib
gl. multiplicatis tamen benedictionibus 105a7.
Deadj. < il many, numerous. This could be a calque on Lat.
multiplicare multiply, which it glosses. Calquing has been discussed in
DIL s.v. ilaigidir 62: 8, il 55: 46.
*ilchigidir howls, screams

Milan: VN ilchugud 51c8, gen. sg. innailchaigthiu gl. ululats
76d13.
The verb itself is not attested but underlies the VN. It is a
denominative of ilach a cry of excultation, a howl of victory. The verbal
noun could be a calque on Latin, since the Latin, too, is a noun, ululatus
220

shriek, howl, derived from a verb ululare shriek, howl. Calquing has
DIL s.v.ilach 61:4, ilchugud 63: 21.
imdaigidir abounds, increases

imdaigidir: Milan: pres. ind. 3sg. imdaigidir 71c10, 121b14, 3pl.
imdaigitir 39b2, 70b5, imdaigetar 39d5, pres. subj. 2sg. nonimdaigi 88b6,
3sg. rel. imdaigedar 68b11, 3pl. imdaigitir 39d2, fut. 3sg. imdaigfid 81c11,
VN imdugud, gen. sg. imdaigthea 71c9; doimdaigedar (multiplies,
increases): Wrzburg: pres. subj. 2pl. donimdigid 23b1.
Deadj. < imdae many, numerous.
DIL s.v. imdaigid(ir) 83: 42, imdae 81: 49, do-im(m)daig- 309: 76.
immfolngi causes, produces

Cf. OW pres. ind. 3sg. rel. emmiguollig (computus fragment).
Wrzburg: pres. ind. 3sg. imfolngi 4d32, 33, 9d8, 16b10, 23c5,
imfolngi 10c13, rel. immefolngi 12b5, immafolngi 16b8, 3pl. rel.
immefolngat
5d1,
immefolnget
12a19,
13c4,
immafolnget
25c10,
immafolnget 27a13, perf. 1sg. rel. immoforling 10c18, 3sg. imforling 5b21,
16c2, rel. immeforling 15b8, with infix. pron. 1sg. immumforling 13b6, with
infix. pron. 3sg. n. immidforling 10c20, 21c20, 24a34, perf. pres. subj. 1sg.
rel. arnarimfolngar (Note Thes.: leg. arnarimfolga that I may not cause?
J.S.; Thurneysen KZ xxxvii 105 would take the word as deponent that I
may not cause) 10c14, pass. perf. 3sg. imforlinged 15d21, Milan: pres. ind.
1sg. rel. with infix. pron. 3sg. immid folngi 103b5, 2sg. rel. with infix. pron.
3sg. immidfolngi 92a17 (2x), 3sg. imfolngi 56b9, 126b16, (rel.) immefolngi
17d6, 34a27, 135a13, 43c6, 81c14, 126c15, 142d1, imfolngai 42c4, 126c16,
rel. immefolgi 77d4, imfolgi 89a6, 94b3, 3pl. (rel.) immefolngat 47b7,
pass. sg. imfolngar 31d10, immfolangar 145c4, rel. immefolngar 44a10,
88b15, impf. (entire impf. tense not in DIL) 3sg. rel. immefolnged 95a5, 3pl.
rel. immefolngitis
86d6, pres. subj. 2sg. imfolngae 43a20, with ro
conimforlainge 78a8, pass. sg. immefolngither 23c5, past subj. 3sg.

coimfolnged (this form not in DIL) 92a7, pass. sg. arna imfolangide 88b15,
fut. 3sg. airimfolnguba (Cf. McCone 1985: 99) 142b3, perf. 2sg. with infix.
221

pron. 1sg. imimforlaingisse 38d18, 3sg. immefolnga 27d20, imfoirlaing
61b4, (rel.) immeforlaing 38c10, 90b14, part. (io/i) imolngide, nom. pl.
imfolngidi 17a5, VN imolang, imfolang 30d22a, 80d2, acc. sg. folang
23c25, dat. sg. imfolung 24c17, 38d18; *remifolgni (bears beforehand,
anticipates): Milan: impv. 2sg. remfolaing 44c24, 100d12, pres. subj. 3sg.
coremifoil 23a8.
Pedersen says that this inflects like a regular i-verb, but is from a
stem lo-n-g-, to which it is related much like doirci is related to icc.
According to Pedersen, doirci is derived, via the VN tairec, from icc.
However, he is wrong about the relationship between icc and doirci (see
above doirci page 197). He is also incorrect about the derivation of
immfolngi. Although it is possible that a verb derived from the VN fulang
bearing, supporting would be reinterpreted as a compound verb *foluingi,
rather than a simplex *folngaid, such a verb is itself not attested. The only
attested verb is the (further) compounded immfolgni. If immfolgni were
derived from fulang, it would have been formed directly from it with the
extra preverb imm- and because of that, the elements of the original
compounded noun fulang should be inseparable. Therefore, in the perfect of
the verb we would expect to find moveable ro rather than fixed ro (cf. GOI
528). However, the perfect forms consistently have ro right before the
verbal root: this verb uses fixed ro throughout. This verb can therefore not
be derived from folung. It is certainly weak: it has s-subjunctive, f-future
and s-preterite and the nasal remains throughout. The use of fixed ro also
shows that it is most likely old.
This verb has further been discussed by Schrijver, Schulze-Thulin
and Schumacher. There is consensus that it is cognate to the loing verbs as
Pedersen noted, but the reason for it becoming AII, as opposed to the other
verbs of this root,51 has not been addressed.
I propose that immfolngi is a causative. Of course, in Proto-IndoEuropean the morphology of causatives and nasal presents was mutually
exclusive, but it is not at all certain that this was still the case in Celtic. In
Celtic, the AII class was still transparently used to form causatives. If a
51
The etymology of the loing verbs is unclear, but it is certain that these are nasal present
verbs.
222

Celtic speaker wanted to form a causative based on the verb *wo-lung-, the
most straightforward way to do that would be to take the present stem and
adapt it to the AII pattern, even though in Proto-Indo-European that would
not have been possible. Because this verb now inflects as a weak AII, it will
keep the nasal throughout the other tenses. This also works semantically:
causing someone to support isnt very far removed from causing,
producing and making. Moreover, this is not the only verb that formed
its causative from a base from which usually a causative would not have
been formed. OI loscaid (page 227) is a causative formed from an original
*-sk- present, on which a causative would also not have been formed in
Proto-Indo-European, and yet one was formed on this stem in Celtic.
Ped. ii 569, IEW 685-6, LIV2 416, DIL s.v. imm-folgni 141: 61,
*remi-folgni 43: 45, fulang 486: 77, GOI 528, Schrijver 1995: 434,
Schulze-Thulin 2001: 191-2, Schumacher 2004: 460-2.
inbothaigidir marries, weds

Wrzburg:
pres.
subj.
3pl.
dianinbothigetar
gl.
Habentes
dampnationem, quia primam fidem inritam fecerunt 29a2.

Denom. < inboth wedding, nuptials. Hapax legomenon.
DIL s.v. inbothaigidir 206: 36, inboth 206: 23.
indbadaigidir abounds, is rich

Milan: sec. fut. 3pl. no indbadaigfitis gl. lucrificarent 15c8, verb.
nec. indbadagthi (this is incorrectly analysed in DIL as a part.) gl. lucratiui
92a17.
Denom. < indbad riches. The verb is only attested in Milan.
DIL s.v. indbad 218: 72, indbadaigid 218: 76.
*indenmigidir weakens
Milan: part. nom. sg. indemnichthe (recte indenmichthe), gl.
debilitatum 50a3.
Deadj. < indenim weak. As has been discussed above, page 70, the
verb is unattested except for the participle, which is hapax legomenon and
seems to be a calque.
223

DIL s.v. indenmichthe 224: 29, indenim 224: 25.
indraicigidir thinks fit, deigns

Milan: pres. subj. 2sg. man inraiccaigther gl. s dedigneris 103a8.
Deadj. < indraic worthy, fitting, < in + reic act of selling or
bartering, VN of renaid. Hapax legomenon.
DIL s.v. indraic 242: 3, reic 30: 45, 3. in 190: 32, indraicigidir 243:
10.
ingraigidir renders impious, pollutes

Milan: pres. ind. pass. sg. ingraigther gl. impiatur 127a9, perf. pass.
pl. roingraigthea (DIL gives this as pres. ind. pass. pl., but the form is
perfect, and so is the Latin it glosses: impiata sunt) 100c18.
Deadj. < ingor undutiful, unfilial. The verb only occurs in Milan
and could be a calque on Lat. impiare render impious, which it glosses.
Lat. impiare is derived from impius wicked, impious, which is the
negative of pius conscientious, also patriotic, dutiful, just as OI
ingraigidir is derived from ingor, the negative of gor pious, dutiful.
DIL s.v. ingraigidir 265: 80, 2. ingor 265: 16.
inilligidir protects, defends

Milan: fut. 3sg. inilligfid gl. tutabitur 128c8, VN (m u) inilligud, acc.
sg. inilligud gl. tutionem 19a14, gl. tutum 47c18, inuilligiud gl. tutionem
57b8, dat. sg. inuilliugud gl. tutione 30b23, 35d1, inuilligiud gl. tutell
44c16.
Deadj. < inill safe, secure. The verb is hapax legomenon and could
be a calque on Lat. tutare guard, protect. Both protecting and defending
are of course common concepts in the Old Irish culture already, but any of
the available synonyms (e.g. adcomai, coneim) could have been used.
Instead, a new verb was created. Calquing has been discussed in section
3.4.3, pages 69ff.
DIL s.v. inilligid 269: 19, inill 268: 28.
224

inlnaigidir ensnares, entangles
Milan: pres. subj. 3pl. rel. inlinaigetar gl. inretiere 32a14, part. nom.
sg. inlnaigthe gl. inrettitus 39d19, nom. pl. inlnaigthi gl. inretiti 82b2.
Denom. < ln linen, cloth, but also hunting-net, snare. The verb is
only attested in Milan and is formed in analogy to Lat. irretire entangle,
which it glosses. This has been discussed in section 3.4.3, pages 69ff.
DIL s.v. inlnaigidir 272: 6, 2. ln 156: 12.
intledaigidir lies in wait for, ambushes

Milan: pres. ind. 3pl. intledaigtar gl. insidentur 58a1, rel. intledaigte
gl. insidientium 39c25.
Denom. < intled snare, ambush. The verb is not attested outside of
the Milan Glosses.
DIL s.v. intled 291: 12, intledaigidir 291: 40.
sligidir lowers, brings down

Cf. W. iselhau, B. izelaat (Joseph).
Wrzburg: pres. ind. 1sg. with infix. pron. 1sg. (reflex.) nomisligur
17d22, Milan: pres. ind. 3pl. rel. sligde 136a11, perf. pass. sg. with infix.
pron. 1sg. rommisliged 50a12, part. nom. pl. isligthi 27a2, VN (m u)
sliugud, sligud 94c2.
Deadj. < sel low, of low stature. This verb can be found in the
Chronology on page 81.
DIL s.v. sligidir 322: 60, sel 321: 37, Joseph 1987: 148.
*idigidir lives as a Jew

Wrzburg: pres. subj. 3pl. rel. iudigte gl. Si t, cum Iudus sis,
gentiliter et non iudaice uiuis, quomodo gentes cogis iudaizare? 19a13.
DIL registers this under the actual form in Wb. 19a13 above and
states: subj. pres. pl. 3 rel. of a vb. *idigid invented to translate the Latin.
iudigte (gl. (quomodo gentes cogis) iudaizare) that they live as Jews, Wb.
19a13. This verb is either derived from Id(a)ei, which was adopted from
the Latin but of which it is often unclear whether the writers regarded it as
Latin or Old Irish (DIL) or the verbal form itself could have been directly
225

adapted from the Latin. Latin loans and other influences have been
discussed in sections 3.2, pages 48ff. and 3.4.3, pages 69ff.
DIL s.v. idigte 332: 78, Id(a)ei 332: 21.
laisrigidir sets on fire, enflames (translation mine)

Milan: pres. ind. pass. sg. with infix. pron. 1pl. nundanlaisrigther gl.
inflammamur 43a12 (note Thes.: MS. nundan laisrigther).
Denom. < lasar, flame, fire. No translation given in DIL, hapax
legomenon. The form above glosses Lat. inflammamur, pres. ind. pass. 1pl.
of inflammare set on fire, enflame. My translation is based on the base
noun and the Latin.
DIL s.v. laisrigidir 33: 8, lasar 57: 40.
lethnaigidir spreads out, extends

Milan: perf. 2sg. rolethnaigser 50a14.
Deadj. < lethan broad, wide.
DIL s.v. lethnaigidir 137: 83, lethan 136: 38.
lobraigidir is sick, becomes sick

Cf. W llwfrhau become timid (Joseph).
Wrzburg: pass. pres. ind. 3sg. lobrigthir gl. infirmatur 6c15; Milan:
pres. ind. 3sg. rel. lobraigedar gl. egrotat 43d21, gl. grescentis 96b12, 3pl.
lobraigetar gl. egrescentium 131c6, pres. subj. 3sg. arna lobraigedar gl.
languescat 71b6, perf. pass. sg. inrulobraiged gl. infirmatus 71c8, VN
lobrugud, dat. sg. lobrugud gl. egrotatione 49a4.
Deadj. < lobur weak, infirm.
DIL s.v. lobraigidir 173: 67, lobur 174: 21, Joseph 1987: 148.
lochtaigidir finds fault with

Milan: impf. 3pl. nolochtaigtis gl. creminando 74c3, rel.
nollochtaigtis gl. creminando 74c3.
Denom. < locht fault, shortcoming. Since both the Latin criminare
accuse and OI lochtaigidir are derived from a noun fault (Lat. crimen),
226

this verb could be a calque. Calquing has been discussed in section 3.4.3,
pages 69ff. The verb occurs once outside of Milan.
DIL s.v. lochtaigid 180: 81, 1.locht 180: 15.
*lnaigidir fattens (translation mine)

Milan: part nom. pl. lnaichthi gl. adipati 20a24.
The verb itself is not attested and has to be inferred from its
participle as found in Ml. 20a24 above. The hypothetical verb is a
denominative from ln fat, provisions.
DIL s.v. lnaichthi 199: 6, 1. ln 197: 76.
londaigidir is angry, becomes angry

Milan: pres. ind. 3sg. rel. londaigedar gl. indignantis 64b5, gl.
indignantis 102b11, perf. 3sg. rolondaigestar gl. commouit 29a2.
Deadj. < lond fierce, strong.
DIL s.v. lonn 202: 35, lonnaigidir 203: 66.
loscaid burns, consumes by fire

PIE *leh2p- flash, gleam, light up (LIV2), *lh2p-ske- (Klingenschmitt)
CC *losk-Cf. MW llosci, MC lesky burn, OB lescsit burned, MB lesquiff
burn (Schulze-Thulin).
loscaid: Milan: pres. ind. 3sg. loscaid 48d1-2, 3pl. rel. loiscde
141b2, impv. 2sg. loisc 47a6, impf. pass. sg. nulloiscthe 32c13, pres. subj.
pass. pl. loscaitir 24c5, perf. 3sg. roloisc 123a15, part loiscthe 118d1, nom.
pl. huleloiscthi 60b13, VN (m u & o) loscud 102b9, acc. pl. loiscthiu 39a7,
dat. pl. loiscthib 133a3; forloisci (burns, sets on fire): Milan: part. (io, i)
forloiscthe 31c28.
Klingenschmitt suggests a causative formed from *lh2p-ske- with
analogical vocalism. CC *lask- is then falsely reinterpreted as the verbal
root. When a Celtic speaker wanted to form a causative based on this verb,
the easiest way to do this would be to use the present stem. Although in
Proto-Indo-European the morphology of *-sk- presents and causatives was
mutually exclusive, it is not clear whether this was still the case in Celtic. In
227

fact, it is likely that this was not so. Compare immfolgni above (page 221),
which was most likely a causative formed in Celtic from an original nasal
present stem. To make a causative of *lask-, the root vowel is analogically
changed to *-o- and the suffix *-- is added, giving *losk--. Since the *-owas accented and rounded, the dorsal *-k- resisted palatalisation, giving OI
loscaid (raising to -u- failed to take place over a group of consonants, cf.
McCone).
Ped. ii 570, IEW 652-3, LIV2 402, DIL s.v. loiscid 189: 85, for-loisci
347: 24, Klingenschmitt 1982, 194 n. 40, McCone 1996: 110, SchulzeThulin 2001: 151-2.
laidid moves across, traverses

laidid: Milan: fut. pass. sg. luaithfider 57c7; forladi (moves,
sways): Milan: pres. ind. 3sg. rel. forluadi 79b5; immladi (sets in
motion, moves): Milan: pres. ind. 3sg. imm-ladi 33d16, pass. pl.
immeluaiter 135b9, impf. 3sg. imluadad 33b25, pres. subj. 1pl. imluadam
93d13, VN (not in DIL as VN) imlad, dat. sg. imluad 15a17.
For this verb, different derivations have been given. DIL suggests it
is denom. < lad, act of moving, setting in motion, VN of lud moves.
GOI gives it as caus of luid went, suppletive preterite of tit goes. LIV2
(488) gives this as caus of root ?*pled- swim, flow. These three options
are each equally valid and it is not possible to ascertain which is ultimately
correct. It is also possible that the verb attested in Old Irish is the result of a
falling together of two or three originally distinct verbs. In fact, if a falling
together is not proposed, DILs option most likely falls, because it would
not be consistent with the scheme proposed in this thesis. Laidid is
ambitransitive: if it were denominative, it would have been an AI verb.
However, if there already was a causative verb with which the denominative
verb fell together, this is not an issue.
Ped. ii 572, IEW 306-7, 837, LIV2 248, 488, DIL s.v. lud 238: 70,
laidid 219: 79, forladi 347: 48, immladi 146: 67, 1. lad 217: 80,
Sjoestedt 1925: 328, GOI 523.
228

lugai
PIE *legh- lie down, *logh- (LIV2)
CC *lg-Cf. MW, MB golo, Goth. lagjan, OIsl. leggja, OHG leggen, OCS
loiti lays.
dulugai (forgives, remits): Wrzburg: pres. ind. 1sg. with emp.
pron. 1sg. doluigimse 14d25, 3sg. rel. with infix. pron. 3sg. n. dodlugi
14d25, 2pl. n dilgaid 9c22, with rel. and emp. pron. 2pl. diandlgidsi
14d24, fut. 1sg. with infix. pron. 3sg. n. dalugub 32a21, with emp. pron.
1sg. dalugubsa 14d24, impv. 2pl. dilgid (a prima manu) 18a11, dlgid
14d21, perf. pres. subj. 3sg. with infix. pron. 3sg. n. darolgea 31a2, 2pl.
with infix. pron. 3sg. n. darolged (for d-a-rolgid) 18a12, perf. past subj.
conderliged 32d12, pass.: fut. 3sg. nidlgibther 33b8, perf. 3sg. duroilged
2c16, durolged 3b12, 3pl. dorolgetha 26c11, ni derlaichtha (MS ni
derlaichta) 33b8, Milan: pres. 3sg. rel. dulugai 51a15, 3pl. (rel.) duluget
56a20, pass. sg. coduloigther 58d2, pl. duluigte(r) 32c15, pres. subj. 2sg.
conderlaige 21b7-8, cia dulogae 138b7, past subj. 3sg. manidilga 46c15,
pass. sg. aranderlaigthe 32c17, fut. 1sg. nadndilgub 31c15, 3sg. dulugfa
58c18, doluichfea 128c6, perf. 2sg. durolgis 124a9, dorolgis 125a12, 3sg.
dorolaig 49c9, 49c10, 50d15, 136b2, 3pl. durolged 124a2, 124b3, pass. pl.
dorolgida 32c15, VN dlgud, dilgud 59c8, acc. sg. dilgud 51a18, dlgud
59c3, gen. sg. dilgudo 32d15, dilguda 46b29, dat. sg. dilgud 32b20, 49d6;
*ascondlugai: Milan: perf. 3sg. aschndarlaig 83c6; folgi (lays low,
prostrates): Milan: perf. 3sg. foralaig 43d5, forlaig 123b9, VN (m u)
flgud, dat. sg. falgud 94b3.
lugai is the causative of OI laigid lies (down). Since this causative
is not only attested in Old Irish and Welsh, but also in Germanic, Slavic and
Hittite (see above), the formation should most likely be attributed to ProtoIndo-European, although the possibility of each of these languages forming
their own causatives cannot be discarded.
Ped. ii 572-3, IEW 658, LIV2 398, DIL s.v. do-luigi 333: 20,
aschndarlaig 428: 11, fo-lgi 181: 39, laigid 26: 59, Sjoestedt 1925: 327,
Schumacher 2000: 76, Schulze-Thulin 2001: 125-6.
229

madaigidir makes vain, frustrates
Milan: perf. 3pl. nirumadaigset gl. frustrata sunt 48a2, with infix.
pron. rusmadaigset 48a1, pass. pl. romadaichthea (Note Thes.: MS.
romadaichtea) gl. casata sunt 80d3.
Deadj. < madae vain, ineffectual. The verb only occurs in Milan.
LEIA M-6, DIL s.v. madaigid 17: 32, madae 16: 53.
maithid forgives
maithid: Wrzburg: pres. ind. 3sg. nimathi 6a5; domaithi
(threatens): Milan: pres. ind. 3sg. dommathi 18c7, domathi 31c24, VN (m
u, perhaps originally ) tomad, tomaith 31c26, gen. sg. thomtho (GOI 724)
26d2, dat. sg. tomaith 33b15, dat. pl. tomthaib 79a7.
LEIA suggests that both these verbs are derived from maith good.
Pedersen separates maithid from domaithi, stating that maithid is a loan
from W maddeu pardons, forgives and connecting domaithi to Alb. mas
(< *matj) measures. Domaithi then, according to Pedersen, had the
original meaning gesturing threateningly.
Ped. i 110, ii 575, LEIA M-12-3, DIL s.v. maithid 46: 37, do-maithi
335: 56, maith 43: 43.
mallaigidir makes slow, retards

Milan: part. mallaigthe (DIL has this as mallaighthe) gl. tardatus
122a14.
Deadj. < mall slow, sluggish.
LEIA M-15, DIL s.v. mall 50: 23, 2. mallaigid 51: 81.
mescaigidir intoxicates, confuses

Milan: fut. 2sg. mescaigfider 81c4.
Deadj. < mesc drunk, intoxicated. LEIA only registers the active
verb, not the deponent. The semantics of mescaigidir have been briefly
LEIA M-41-2, DIL s.v. mesc 112: 61, mescaigid 114: 4.
230

mesraigidir moderates, tempers
Milan: pres. subj. 3sg. -mesraigea 46c15, part. mesraigthe 22c13,
32d4, VN mesrugud 89d1.
Denom. < mesar measure, measurement, < Lat. mensura measure,
length (LEIA). Direct and indirect loans have been discussed in section 3.2,
pages 48ff.
LEIA M-41, DIL s.v. mesar 112: 14, mesraigid 115: 57.
mincigidir becomes frequent

Milan: perf. 3sg. romincigestar gl. increbruit 36a40
Deadj. < meinic frequent, recurring often. Hapax legomenon.
LEIA M-37, DIL s.v. meinic 84: 17, mincigidir 143: 83.
mindchigidir begs, is a mendicant

Milan: pres. ind. 3pl. mindchigitir gl. emendicant 36c18.
Deadj. < mindech poor, wretched, < Lat. mendicus poor as a
beggar (DIL). LEIA does not register the verb. Note that Thes. translates
this verb as attenuates (Ml. 36c18). They state that the glossator
misunderstood the Lat. emendicant which this glosses, because he was
thinking of the OI mindech. DIL correctly translates this verb as begs,
because there is no problem in taking OI mindech poor to make a verb
meaning begs. The -igidir category certainly allows for that. Moreover, the
Lat. emendicare begs, sollicits is derived from the adj. mendicus poor as a
beggar, so it is even likely that the Old Irish verb was calqued on the Latin.
Calquing has been discussed in section 3.4.3, pages 69ff. Direct and indirect
borrowings have been discussed in section 3.2, pages 48ff.
LEIA M-54, DIL s.v. mindech 145: 32, mindchigidir 145: 23.
miscsigidir hates
Wrzburg: perf. 3sg. romiscsigestar gl. gl. Iacob dilexi, Esau autem
hodio habui. 4c16.
Denom. < miscais hatred, dislike. Hapax legomenon.
LEIA M-55, DIL s.v. miscsigidir 149: 41, miscsech 149: 37, miscais
148: 54.
231
modid boasts, vaunts

PIE *smei (d)- laugh (LIV2), caus. *smoi d-i eCC *moid-Wrzburg: pres. ind. 1sg. with infix. pron. 1sg. (reflex.) nom midim
14c18, 3sg. with suffix. pron. 3sg. m. (reflex.) miti 23d29, 27a29, 1pl. with
infix. pron. 1pl. (reflex.) nnmidem 23d23 and emp. pron. 1pl. (reflex.)
nonmidemni 2d9, 3pl. with infix. pron. 3pl. (reflex.) nosmidet 17c5,
nosmoidet 31b14, rel. with infix. pron. 3pl. (reflex.) indammidet 24a30,
fut. 1pl. with infix. pron. 1pl. (reflex.) nnmidfem 17b10, perf. 3sg. with
rel. n and emp. pron. 3sg. m. rommoidisom 17a12, 1pl. with infix. pron. 1pl.
(reflex.) ronmoitsem 17a13, pres. subj. 2sg. with prep. and rel. and infix.
pron. 2sg. (reflex.) indit mide 2b12, 3sg. with infix. pron. 3sg. m. (reflex.)
connach moidea 2b4, 3pl. with infix. pron. 3pl. (reflex.) arnachamoidet
5a16, impv. 3sg. with infix. pron. 3sg. m. (reflex. with nas. elided) nach
moided 9d18, Milan: pres. ind. 3sg. inmoidi (Note Thes.: MS. inmodi: leg.
innidmoidi?) 49c13, rel. moides 41d4, VN modem, moidem 86b5.
DIL (s.v. bocaid 129: 75-78, see also page 115) suggests that this is
derived from moth soft. This is semantically problematic, because there is
no motivation for the verb derived from an adjective meaning soft to mean
boasts. I would rather follow Sjoestedts derivation as a causative and take
this from the root *smei - laugh, smile.
IEW 967, LEIA M-59, LIV2 568-9, DIL s.v. modid 158: 35, maeth
23: 61, bocaid 129: 75-78, Sjoestedt 1925: 328.
mothaigidir softens, migitates

Milan: pres. ind. / pres. subj. 3sg. with infix. pron. 1sg.
nommoithiged 74d12.
Deadj. < moth soft, tender.
LEIA M-61, DIL s.v. maeth 23: 61, maethigid 24: 79.
mothaigidir is amazed, wonders

Milan: pres. ind. 3sg. mothaigid 25c6, mothaigedar 26b9, impf. 3sg.
annumothaiged 25a12.
232

Denom. < moth amazement, stupor.
LEIA M-66, DIL s.v. mothaigid 175: 48, 1. moth 175: 13.
minid teaches, instructs

Wrzburg: perf. 1sg. with infix. pron. 1sg. (reflex.) rommunus
24b17, rommnus 24b18, 3pl. with infix. pron. 3sg. n. raminset 5b44.
Etymology unknown (Ped.). IEW connects this with *meH- push
away (LIV2s *mi eh1- move) on the basis of a note from E. Lewy that is
not referenced. The verb is not mentioned in LIV2. LEIA notes that it could
be connected to a root *men-, but it is not possible to go from PIE *men- to
OI --. There are no suitable bases in Old Irish to derive this verb from. The
etymology remains unknown.
Ped. ii 582, IEW 743, LEIA M-75, LIV2 445-6, DIL s.v. minid 190:
20.
nallaigidir cries aloud

Milan: pres. subj. 1pl. nuallaigem gl. ploremus 114d3.
Denom. < nall loud noise. The verb is only attested once outside
of the Milan Glosses.
LEIA N-24, DIL s.v. nall 69: 54, nallaigid 70: 66.
oenaigidir makes one, unites

Wrzburg: pres. ind. 3sg. with suffix. pron. reflex. 3sg. m. and emp.
pron. 3sg. m. oenichthisom 32d8, Milan: part. oinigthe gl. unitus 25a2.
Deadj. < oen one.
LEIA O-10-1, DIL s.v. oenaigidir 104: 17, oen 100: 13.
oingid anoints
Milan: past subj. pass. sg. rooingthe 47b16, part. (io, i) oingthe, dat.
sg. ointhu 48b8,VN ongad 43c11, gen. sg. ongaith 30b7.
Lat. ungere anoint, rub. Latin loans have been discussed in section
3.2, pages 48ff.
LEIA O-24, DIL s.v. oingid 122: 11.
233

ollaigidir makes larger, expands
Milan: impv. 2sg. ollaigthe gl. amplic 70c7, impf. 3sg. noollaiged
(Note Thes.: MS. no ollaiged) gl. ampliauit 61c6, verb. nec. ollaigthi gl.
ampliand 88a14.
Deadj. < oll great, ample. This could be a calque on Lat. ampliare
enlarge, which it glosses. The verb only occurs in Milan. Calquing has
LEIA O-21-1, DIL s.v. oll 136: 73, ollaigidir 138: 41.
ordaigidir orders, ordains

Wrzburg: perf. 3sg. rel. with infix. pron. 3sg. n. rodordigestar 6a3,
ronortigestar 6a4.
Denom. < ordd order, sequence, < Lat. ordo row, order. Direct
and indirect loans have been discussed in section 3.2, pages 48ff.
LEIA O-29, DIL s.v. ordaigidir 156: 16, 1. ordd 154: 15.
pecthaigidir sins, commits sin

Milan: impf. 3pl. nupectaigtis 74c3.
Denom. < peccad sin, < Lat. peccatum sin. Direct and indirect
LEIA P-5-6, DIL s.v. peccad 178: 60, pecthaigid 179: 54.
rdaid speaks, says

PIE *reh1dh- perform succesfully (LIV2), caus/it *roh1dh-i e- (SchulzeThulin)
CC *rd-- (Schulze-Thulin)
Cf. Goth. rdjan speak, Skt. rdhyati, brings about, satisfies,
OCS raditi, worry, OIsl. r a, speak, MW adraw tell, kyuadraw tell
about and ymadraw talk (Schulze-Thulin).
rdaid: Wrzburg: perf. 3sg. rel. rurdi 7d9, Milan: pres. ind. 1sg.
rel. noradim 47b13, 3sg. niradi 35d22, rel. radas 42c4, 42c10, 3pl. n radat
31a18, 31b11, pass. sg. rel. frissarater (Note Thes.: corrected from
frissather; this form not in DIL) 42c4, impf. 1sg. rel. nuradin 74c3, perf.
1sg. with emp. pron. ruradussa 50d7, VN (m o) rd, gen. sg. raid 39a11;
234

immrdi (thinks (of), reflects (on)): Wrzburg: pres. ind. 3sg. imrdi
17a13, rel. with infix. pron. 3sg. n. immidrdi 8b9, 3pl. imrdat 1d7, pres.
subj. 3sg. rel. immerada 23b24, 3pl. rel. immardat 28c1, past subj. 1sg. rel.
immeradin 14c23, pass.: pres. ind. 3sg. rel. immeraiter (MS immeraither)
14c22, perf. with infix. pron. 1pl. immunrordad 20d10, Milan: pres. ind.
3sg. immeradi 33d2, imradi 35d22, 1pl. imradam 93a5, 3pl. immeradat
54a25, imradat 105a3, n imradat 107c1, pres. subj. 3sg. arnachaimrada
51a1, past subj. 3sg. imradad 41a4, perf. 1sg. rel. immerordus 96a3, 2sg.
immeradi 43c4, 3sg. imraid (Note Thes.: leg. imruraid?) 90d14, 3pl.
immerordaisset 90d16, verb. nec. dat. pl. imratib 38a5, 96b16, VN imrd,
dat. sg. imradud 15a16, 55c19, VN (not in DIL as such) imradud, immrdud
15b3.
This verb has often been seen as an inherited denominative (cf. OI
rd act of saying, uttering) (GOI, GPC), but can on the basis of Goth.
rdjan speak, Skt. rdhyati, brings about, satisfies, OCS raditi, worry,
and OIsl. r a, speak be interpreted as the iterative of *reh1dh- perform
succesfully (Stokes, LIV2, Schumacher, Schulze-Thulin, cf. Isaac,
Schrijver). The Welsh cognate is only extant compounded.
Ped ii. 591-2, IEW 59-60, LEIA R-3, LIV2 499-500, DIL s.v. ridid
4: 20, imm-rdi 149: 17, rd 2: 19, Stokes 1876-8: 35, GOI 722, GPC 27,
Schrijver 1991: 307, Isaac 1996: 306, Schumacher 2000: 157, SchulzeThulin 2001: 134-5.
ridigidir levels, smooths

Cf. MW rhwyddhau facilitate, Br. rouzzt clear (Joseph).
Milan: pres. ind. 3sg. rel. nredigedar 24d22.
Deadj. < rid level, smooth (LEIA). Rididigir can be found in the
LEIA R-16-7, DIL s.v. rid 32: 5, ridigid 33: 70, Joseph 1987: 149.
rmid counts, reckons

Cf. W kyfrif to count (Schumacher).
adrmi (counts, numbers): Wrzburg: pres. ind. 3sg. rel. with
emp. 3sg. m. adrmisom 14d2, with rel. n and infix. pron. 3sg. n.
235

adindrmisom 13d17, n armi 24a17, with emp. pron. 3sg. m. nrmisom
6a1, ni irmisom 13d17, pres. subj. 1sg. rel. with emp. pron. 1sg.
conarimse 14d17, pass. perf. 3sg. rel. atruirmed (Cf. McCone 1985: 97)
2c6, with dem. rel. anadruirmed 2d7, Milan: pres. ind 3sg. adrimisn
24a13, adrimisiu 24b6, atrim (DIL: atrim[i]) 49a11, with infix. pron. 3pl.
atarimi, with infix. pron. 3sg. n n airmi 82a8, 3pl. adrimet 103b5, 129d19,
129d22, pass. sg. adrimther 46c24, pl. adrimter 99d9, adrimetar 111a10,
impf. 3sg. coniairmed 46b31, 3pl. adrimtis 91a6, past subj. 3sg. adrmed
21a4, fut. 1pl. adrifem 14d5, perf. 3sg. adruirim 28d5, with infix. pron. 3sg.
ninarraim 49c9, 3pl. adruirmisset 101b2, VN (f ) irem, acc. sg. arim
102d17, dat. sg. airim 17d6, dormi (counts, enumerates): Wrzburg:
pres. ind. 3sg. with infix. pron. 3pl. nistuirmi 30c17, perf. 1pl. with rel. n.
and infix. pron. 3sg. n. dondruirmissem 24d16; pass.: pres. ind. 3sg.
(depon.), rel. ituiremar 12d39; Milan: pres. ind. 3sg. dormi 44c28, dorimi
54a37, durmi 125b3, durimi 83d6, 96b14, torrimi 130c1, 1pl. dormen
14d6, pass. sg. dorimther 16c10, 23b8, rel. durimther 103d12, pres. subj.
3pl. durmet 19b13, fut. 1sg. dorimiub 26c7, durimiub 106c10, 1pl.
dormfem 44a24, perf. 3sg. doruirim 36b6.
Denom. < rm act of counting, enumerating, also used as VN of
rmid. The lenited -m of the Old Irish and the -f of the Welsh noun are
reflexes of the Proto-Indo-European nominal suffix *-mo-. Both verbs are
therefore denominative. Considering that both in Welsh and in Old Irish this
verb (and compounds) are the only ones used for counting, it is likely that
this verb was created within Celtic. Moreover, both the Old Irish and the
Welsh (in the compound MW kyfrif) use the base noun as VN, rather than
adding a verbal noun suffix still productive in both Old Irish and Welsh to
create a new verbal noun, which would have been likely if these verbs had
been created within the individual languages. The verb can be found in the
Ped. ii 602-3, IEW 60, LEIA R-30-1, DIL s.v. 1. rmid 70: 63,
adrmi 68: 62, dormi 359: 5, 1. rm 70: 25, Schumacher 2000: 141.
rogid stretches, extends

PIE *h3re- stretch, direct, caus. h3ro-i e- (LIV2)
236

CC *rog-Cf. YAv. -rzaiieite stretches, directs, Goth. -rakjan elongate
(LIV2).
Milan: pres. subj. 3sg. rel. rogas 79d4, pass. sg. roichther52 44a4,
138a2, roigthir 110a1, part. rogid, nom. pl. roigthi 33b1, VN rogad, dat. sg.
rogud 37d6.
Thurneysen suggests this is a causative; LEIA classifies it as an
iterative. Morphologically, there is no difference, but semantically, an
iterative is more suitable.
Ped. ii 593, IEW 854-7, LEIA R-39, LIV2 304-5, DIL s.v. rogid 89:
45, Sjoestedt 1925: 327, Thurneysen 1913: 35.
roisid wavers, vacillates

Wrzburg: perf. 3sg. nrorois 2c28; Milan: pres. ind. 3sg. roissid
21a6, rel. roisses 18d21, pres. subj. 3sg. coroissid 73c10, sec. fut. 3sg.
noroisfed 85d10, perf. 3sg. rorois 84c16, nirorois 44a1.
Etymology unknown. This could possibly be derived from OI ros
flax-seed, linseed, in its second meaning of some small aquatic plant
(DIL s.v. 2. ros 98: 30), in the sense of wavering like a plant in the water.
Note though that the meaning of the base noun itself is uncertain.
LEIA R-41, DIL s.v. roisid 94: 16, 2. ros 98: 20.
roithid sets in motion, makes run

PIE *ret- walk (LIV2), caus. *rot-i eCC *rot-Milan: pres. ind. 3sg. rel. roithes 42b27, pres. subj. pass. pl. rel.
rothetar 92a16, fut. 3pl. roithfiter 15c18, part. roitte 36d21, dat. sg. m roitiu
36b4.
Caus. of reithid (page 306; LEIA), morphologically regular, but
phonologically irregular. CC *rot--ti > *roii. Regularly, an *-o- before
52
McCones (1985: 100) reading as an early simplex of rosaig reaches, comes to is

illogical in the context, considering that it glosses the Lat exsero stretch forth. The gloss
reads .i. ascesar .i. roichther. Ascesar is pres. subj. pass. sg. of ascid which also means
extends, stretches. The glossator translated the Latin with two synonyms.
237

and *-i- should be raised to *-u-: *ruii > OI ruithid, which is also extant
(DIL s.v. ruithid 94: 71). The form roithid was probably formed in analogy
to roth, a wheel.
IEW 866, LEIA R-23, 41, LIV2 507, DIL s.v. roithid 94: 75, reithid
38: 72, 1. roth 103: 56.
ruccaigidir puts to shame, confutes

Milan: pres. ind. pass. sg. rucaigthir gl. confutatur 24d3, pres. subj.
pass. sg. rucaigther gl. frustrata 87c2, pl. ruccaigter gl. confundantur 54b8,
past subj. 3sg. comruiccaiged gl. confutaret 25b10, part. ruccaigthe gl.
condemna 27c11.
Denom. < ruccae shame, disgrace. The verb only occurs in Milan.
LEIA R-50, DIL s.v. ruccaigid 111: 75, ruccae 111: 53.
saethraigidir works, labours

Milan:
pres.
3pl.
pass.
saithraite
123c8,
perf.
1pl.
rel.
frissarusaithraigestar 92c5.
Denom. < saethar work, labour.
LEIA S-7, DIL s.v. saethar 14: 42, saethraigidir 15: 60.
sidid thrusts, fixes

PIE *sed- sit, caus. *sd-i e- (Schulze-Thulin)
CC *sd-- (Schulze-Thulin)
Cf. OCS saditi, bed, plant, MW gwahaw, treat, invite, MW sodi
sink, drown.
sidid: Milan: pret. 3sg. with suffix. pron. 3sg. n. saidsi 55c1;
considi (complains, stirs): Milan: VN chosait 133a1.
For sidid, GOI suggests a lengthening of the vowel in saidid sits,
if this verb is not formed like Lat. spre. Nagy sees it as a neologism to OI
saidid sits, sits down, but as Schulze-Thulin correctly points out, this
would be a unique development. Moreover, there is no motivation for it.
Klingenschmitt derives W gwahaw and OI sidid from a causative
formation from an aorist stem *sd- < *ssde/o-. However, there are in
Celtic no verbal *sd- forms from which this could have been derived. If
238

sidid were a causative formation from this root, it would have to have been
a pre-Celtic derivation. There is no noun or adjective attested in Old Irish
(or Welsh) from which this could be derived. This and its Welsh cognate
gwahaw are lengthened-grade causatives *sd-i e- of a Proto-IndoEuropean root *sed- sit, as proposed by Schulze-Thulin (followed by
Schumacher and LIV2). These lengthened-grade causatives occur not just in
the Celtic languages, cf. Lat. spire lull to sleep and OIsl. sfa kill
(< cause to sleep) < *s()p-i e/o-. They were first discussed by
Klingenschmitt in 1978 and are therefore often called Klingenschmitt
causatives. The type is however not uncontested. For alternative
explanations of the long vowel in this type of verbs, cf. Schulze-Thulin (p.
4-6). The only other causative of this type in the Celtic languages is W
digoni make, do, work (< CC *kn- to the Proto-Indo-European root
*kenh1- try, strive, Schulze-Thulin p. 138-9). Note that Klingenschmitt
himself did not recognise the Celtic verbs as long vowel causatives.
The Proto-Indo-European root *sed-, to sit also has -grade
causative PIE *sod-i e/o- > CC *sod-- > OI suidid (page 247).
Ped. ii 605-6, IEW 884-7, LEIA S-8-9, LIV2 513-5, DIL s.v. sidid
19: 40, con-sidi 459: 11, saidid 17: 80, GOI 523, Sjoestedt 1925: 324,
Nagy 1970: 22, Klingenschmitt 1978 (for sidid see specifically p. 13 n.
27), Schumacher 2000: 161-2, Schulze-Thulin 2001: 4-6, 136-9.
sainigidir is different, distinct

Milan: VN sainigud 133c1.
Deadj. < sain different, distinct.
LEIA S-14-15, DIL s.v. sainigidir 32: 53, sain 29: 25.
sraigidir violates, outrages

Cf. MW sarhau injure, compensate (Joseph).
Wrzburg perf. 3pl. rel. rsrichset 1d4, pass. perf. 3sg. rosariged
3c24, 25, Milan: perf. 3sg. rel. rusarigestar 71b14.
Denom. < sr outrage, insult. Since this verb has a Welsh cognate
and preserves the old seek meaning of the *-sag- suffix, it is most likely an
inherited denominative. It can be found in the Chronology on page 81.
239

LEIA S-25-6, DIL s.v. sraigid 59: 65, 1. sr 58: 47, Joseph 1987:
149.
scalid bursts, scatters

arscaili (looses, hurls): Milan: VN airscaliud, acc. sg. aerscailiud
96a8.
Denom. < scal scattering, dispersion. The noun has been
connected with the root *skelH- cut, split (as found in LIV2 553 and IEW
923-6) and W chwalu shatter, smash (for discussion with references, see
LEIA), but if this is correct, the vocalism of the Irish forms is irregular. I
prefer to follow LIV2s (354 n. 1.) suggestion of taking the noun as a -loformation *skoi-le- to the root *skei- (IEW 920). In that case, W chwalu is
unrelated.
Scalids place in the scheme has been discussed in comparison with
brchtaid bursts on page 59.
Ped. i 359, IEW 623-6, 920, LEIA S-28-9, LIV2 354 n. 1., 553, DIL
s.v. scalid 68: 51, ar-scaili 414: 39, scal 66: 86.
scthaigidir becomes tired, fatigued

Milan: impf. 3pl. noscithaigtis 102d14.
Deadj. < scth weary, tired.
LEIA S-46-7, DIL s.v. scthaigidir 97: 31, scth 96: 42.
scuchaid moves, starts

PIE *skek- move fast, jump, caus/it *skok-i e- (LIV2)
CC *skok-- (Schulze-Thulin)
Cf. OCS skoiti jump, MW kyscoc shake (Schulze-Thulin).
conoscaigi (moves, changes): Wrzburg: pass. perf. pres. ind. 3sg.
nirochumscigthersn 30b15, Milan: pres. ind. pass. sg. chonnoscaigther
38d15-16, oscaigther 118c12, pl. conoscaigter 94c3, impf. 3sg.
conidchumscaiged 109d05, pres. subj. 1sg. cotammoscaig 29d3, 2sg.
conoscaigesiu 32a3, 3pl. conoscaiget 54a19, pass. sg. with infix. pron. 2sg.
cotoscaighter 55b3, cotatoscaigther 58d14, pret. 3sg. oscaig 55c1 (2x),
perf. 2sg. conroscaigis 21d7, 3sg. conroscaig 45c1, part. (io, i)
240

cumscaigthe 15a2, 33b2, 40d23, 46b9, VN (m u) cumscugud 26d12, 35c17,
105d7, acc. sg. cumscugud 27b15, 100b13, 118c12, dat. sg. cumscugud
31d4, 38d15-16; dofoscaig (moves, comes to): Milan: pres. ind. 3pl. (rel.)
dufoscaiget 33b3, dufoschaiget (DIL: dufoscaiget) 90c26, pres. subj. pass.
condatoscaigther 23d21, perf. 3sg. doforscaig 37d12; dorscai (stands
forth, distinguishes oneself), Wrzburg: pres. ind. 1pl. n derscigem 2a14,
fut. 3sg. with infix. pron. 2sg. ditrscibea 1d21, Milan: pres. ind. 3sg.
dirsci 57a1, diroscai 133a4, duroscai 129b10, (rel.) durscai 113a9,
114d1, 115b16 (x2), 134c4, doroscai 138c12, pass. sg. doroscither 107b5,
impf. 3pl. duroscaitis 79c6, 124d2, dorscaitis 136a7, pres. subj. 3sg.
doroscea 56a4, 3pl. ciadudrscat 121a3, past subj. 1sg. duroscainn 41c6,
3sg. dirosced 58d15, durosced 129b13, rel. with infix. pron. 3sg.
dundrosced 33c5, 1pl. doroscaimis 32a21, 3pl. doroscat 37b14, fut. 3sg.
duroscaifea 139b3, durscaibea 89c12, 3pl. duroscibet 84b1, duroscaifet
84b4, pass. pl. duroscaibtar 87b4, pret./perf. 3pl. dorscaiset 119d3, pret
pass. sg. rel. with infix. pron. 1sg. dundamroscad (Note Thes.: MS.
dundanroscad) 100b14, verb. nec. remderscaigthi (Note Thes.: MS. rem
derscaigthi) 105b3, part. (io, i) derscaigthe 109c08, 123a7, nom. pl.
derscaigthi 100c26, comp. nom. sg. derscaigthiu 78b1, 92d14, VN
derscaigthe, dat. sg. derscaigthiu 119d4, VN derscugud, 129b12; foscuichi
(moves away, departs); Milan: pres. ind. 3pl. fosoichet 36a8, VN foscugud
42a5, 126b2, dat. sg. foscaigiu 21c3.
The simplex scuchaid has a mixture of strong and weak forms.
Thurneysen (GOI) has stated that the simplex inflects strong, but this is
incorrect; there are plenty of weak forms in the simplex. Schulze-Thulin
makes the distinction between strong forms for the meanings moves, starts,
goes, intransitive, non-causative, and weak forms for the meanings moves,
stirs, sets in motion, transitive, causative. She states that this is the result of
confusion between an original primary verb < *skek- and a causative
< *skok-i e-. LEIA states that in the present the causative stem has
superseded the stem of the primary verb. However, against this can be said
that in the present stem the distinction is not quite so clear. Some present
forms have intransitive meanings, although these are all late. More
importantly, as Schumacher has pointed out, strong inflection is confined to
241

the meanings comes to an end, is finished, exhausted, whereas the
meanings moves (trans.), stirs, moves (intrans.), starts (intrans.) have
weak i-inflection. Strong inflection is confined to the subjunctive and the
preterite, so in the present there never was any confusion. It seems clear that
to the Irish themselves, these were two distinct verbs, with two distinct
inflections and two distinct meanings.
The presence of preverb *uts- in doroscai shows that this is an old
compound, cf. Marm and Russell.
Ped. ii 617, IEW 922-3, LEIA S-56-7, LIV2 551-2, DIL s.v. scuchaid
116: 31, con-oscaigi 455: 28, do-foscaig 274: 33, do-rscai 365: 54, foscuichi 377: 15, GOI 522, Russell 1988: 126, Marm 1995: 317, 319-20,
Schumacher 2000: 163, Schulze-Thulin 2001: 132-3.
sechtaigidir simulates, pretends

Milan: impf. 3sg. rel. nosechtaiged (Note Thes.: MS. nosechaiged:
corr. Ascoli) gl. simulantis 133a12, part. dat. pl. sechtaigthib gl. simulatis
38c17, VN (m u) sechtugud gl. simulatio 103b13, gen. sg. sechtaigtho gl.
simulandi 30d6, gl. simulationis 31d13, dat. sg. sechtugud gl. non
disimulatione 32b4, nom. pl. sechtaigtha gl. simulationis 30d17.
Deadj. < secht feigned, pretended. The verb only occurs in Milan.
LEIA S-66-7, DIL s.v. 2. secht 133: 41, sechtaigidir 135: 38.
simigidir thins, attenuates

Milan: pres. ind. 3sg. rel. cosemigidir gl. adtenuet 44d9, 3pl. rel.
smigte gl. extinuata 86a10, impf. pass. sg. with infix. pron. 1sg.
nomsimigthe gl. me iecirer 88a11, pres. subj. pass. pl. semigter gl.
tenuentur 54b15, perf. pass. sg. rosemiged gl. adtinuatus sit 118b5, pl.
roseimigthea gl. adtenuat sunt 93d2
Deadj. < sim slender, thin.
LEIA S-72, DIL s.v. simigidir 148: 56, sim 146: 35.
sercaigidir loves
conasarcaigedar (delights in, is delighted with, transl. Wb. Lex.):
Wrzburg pres. ind. 1sg.: conasarcaigim gl. condilector 3d9.
242

Denom. < serc love. The simplex is only attested later. The
compound is hapax legomenon and in DIL only found under the form
conasarcaigim from Wb. 3d9 above. This is a *-sag- derivative with an
active ending. Although active endings for deponent verbs are rare in
Wrzburg, they do occur (cf. McCone). The alternative would be that this is
a *-sag- derivative formed as an active verb and for that it is perhaps still a
bit too early in the language. It is possible that this is a calque on the Latin
condelectari delight in, which it glosses. Certainly, the elements match
each other in meaning, and the Old Irish form is hapax legomenon. The
simplex could the later have been back-formed. Calquing has been
LEIA S-91-2, DIL s.v. serc 189: 5, sercaigid 190: 40, conasarcaigim
426: 12, McCone 1985: 95.
staigidir goes on a way, journeys

Milan: pres. ind. 3pl. rel. staigte (in DIL without length mark) gl.
uiantium 82d4; instaigedar (travels): Milan: impf. 3pl. inseitaigtis (note
Thes.: leg. nusitaigtis?) gl. uiantibus 124b9.
Denom. < st, path, way. The emendation for Ml. 124b9 in Thes.
as given above is unnecessary. This verb behaves perfectly regularly as a
compound of staigidir. I suggest a citation form instaigedar.
LEIA S-98-9, DIL s.v. staigid 203: 20, instaigid 285: 31, 1. st
200: 18.
slaigidir sows; causes

Cf. OB hilheiat sower (Joseph).
Milan: past subj. 1sg. nosilaiger gl. exserere 23c28, 2pl. nosilaigid
gl. adserendo 115b11.
Denom. < sl seed of plants and animals.
LEIA S-108-9, DIL s.v. sl 223: 63, slaid 225: 11, Joseph 1987:
149.
slnaigidir makes whole, heals

Milan: pres. ind. 2pl. slanaigthe 17b15, 25b2.
243

Deadj. < sln whole, sound.
LEIA S-126-7, DIL s.v. slnaigidir 263: 3, sln 260: 28.
sluindid expresses, signifies

PIE *splend- shine, become bright, caus *splond-i e- (Schulze-Thulin,
LIV2)
CC *slond-Cf. MW kystlwn explain (Schulze-Thulin).
sluindid: Wrzburg: fut. 1pl. with infix. pron. 1pl. and emp. pron.
1pl. runsluinfemni 15a4; Milan: pres. ind. 3sg. rel. sluindes 37a10, 3pl. rel.
sluinde 139a6; dosluindi (denies, refuses to believe): Wrzburg: pres.
ind. 3sg. dosluindi 6c14, 3pl. disluindet 13b18, dosluindet 30a20, perf. 3pl.
(with ri in place of ro) doriltiset 25b13, with emp. pron. 3pl. dorltiset som
5c11, perf. pres. subj. 3sg. arnaderlind 10c14, past subj. 3sg. arandltad
17d11, pass. fut. 3sg. dosluinfider 28c14, Milan: pres. ind. 3sg. disluindi
44b12, 50d1, perf. 3sg. rel. dorosloinn (this form not in DIL) 58a11, rel.
doroslund 58a11, rel. durusluind 93c8, 3pl. dorusluindset 90b17, VN
diltad, diltud 15d2, 55c10, 96a7, dltud 91a21, dat. sg. diltud 55c8.
The relationship between this verb and the noun slond act of
expressing, mentioning is unclear. Sjoestedt and Schulze-Thulin take the
verb as a genuine old causative of root *splend- shines, becomes clear, but
do not take the existence of the noun into account. Sluindid could also be
denominative from slond as LEIA has suggested. LIV2 has sluindid as
causative, but note that even the root is given with a question mark and
outside of Old Irish this is only attested in a Latin essive splendere shine,
gleam. Is sluindid denominative, or is slond derverbal?
Ped. i 83-4, ii 632, IEW 987, LEIA S-137-8, LIV2 582, DIL s.v.
sluindid 282: 25, dosluindi 374: 84, slond 280: 52, Sjoestedt 1925: 328,
Schulze-Thulin 2001: 157.
sndaid protects, convoys

insndai (defers, postpones): Milan: pres. ind. 2sg. adsnadi
93a12, impf. 3pl. rel. annadnessnatis 98b3, pres. subj. pass. sg. insnter
56c20, VN esndud, acc. sg. snadud 24c1.
244

Etymology unknown. LEIA refers to Sjoestedt and Pokorny, who
derive it from *snd- < *nedh- and *snd- <*sn- (laryngealised *h3neh2-)
respectively. LEIA dismisses both these options, but endorse Fowkes, who
connects it with with Skt n-tha- < *neh2. This has been dismissed by LIV2.
Ped. ii 633, LEIA S-146-7, LIV2 449, DIL s.v. 1. snidid 294: 73, insndi 285: 48, Sjoestedt 1925: 324-5, Fowkes 1947: 420-1.
sobid makes / becomes crooked, contorts

Wrzburg: pres. ind. 3sg. rel. saibes 27a9, Milan: perf. 3 pl. with
infix. pron. 3sg. n. rondasaibset 24d24; consoba (deceives): Wrzburg:
pres. subj. 3pl. rel. consoibat 30c13.
Deadj. < sob crooked, aslant. Note that Wb. Lex. gives consoibat
above under saebid.
As has been discussed above, page 61, the deadjectival sobid is AII
due to its iterativity. Consoba remains AI because the finality implied by
the preverb con- blocks iterativity. These verbs are both derived from the
adjective sob but they have different meanings and different syntactic
behaviour.
Ped. ii 637, LEIA S-163-4, DIL s.v. con-soba 463: 3, saebaid 7: 51,
saeb 6: 10.
soilsigidir makes bright, lights

Milan: pres. ind. pass. soilsigthir 116b7-9.
Denom. < soilse brightness, light, < solus bright, clear.
LEIA S-168-9, DIL s.v. solus 337: 1, soillsigidir 329: 3, soillse 328:
50.

Milan: pres. ind. pass. pl. soinmigter gl. prosperari 59a4, sec. fut.
3sg. nosoinmigfed gl. prosperari 23c2, part. soinmichthe gl. prosperatus
36d26.
Denom. < soinme great wealth. The verb only occurs in Milan.
LEIA S-166, DIL s.v. soinmigidir 330: 81, soinme 330: 22.
245

sommaigidir enriches
Wrzburg: pres. ind. 3pl. with rel. n and infix. pron. 2pl.
nondobsommigetar gl. locupletati 17a1, Milan: pres. ind. pass. pl. rel.
sommaigter gl. ditari 79c2.
Denom. < sommae riches, wealth, or deadj. < sommae rich,
wealthy, both of which are themselves derived from soim rich, wealthy.
The verb only occurs in the Wrzburg and Milan Glosses.
LEIA S-170, DIL s.v. sommaigidir340: 69, 1. sommae 340: 46, 2.
sommae 340: 52, soim 329: 45.
sonartnaigidir strengthens, confirms

Cf. OB henertha (Joseph).
Milan: pres. ind. 3sg. arnasonartnaigedar 28d9, perf. 3sg.
rosonartnaigestar 49b4, VN (m u) sonartnugud, acc. sg. sonartnugud
68a15, 121c7, dat. sg. sonartnugud 67d2.
Deadj. < sonairt strong, firm, < so good, excellent + nert
strength, might. LEIA does not register the verb. The appearance of the -nbetween the base and the suffix has been discussed in section 3.4.2, pages
68f. Sonartnaigidir can be found in the Chronology on page 81.
LEIA S-172, DIL s.v. sonartnaigidir 344: 51, sonairt 344: 1, 2. so
308: 25, nert 35: 18, Joseph 1987: 149.
sorchaigidir illuminates, enlightens

insorchaigedar (illuminates, makes bright): Milan: pres. subj.
pass. sg. with infix. pron. 2pl. coatabsorchaigther gl. inluminamini 53b15.
Deadj. < sorchae bright, luminous or denom. < sorchae light,
brightness. The compound could be a calque on Latin inluminare
illuminate, give light to, which it glosses. The simplex might be a calque
too, but there is no evidence to base this on. Calquing has been discussed in
LEIA S-174, DIL s.v. sorchaigidir 347: 44, in-sorchaigedar 285: 78,
1 sorchae 347: 1, 2. sorchae 347: 15.
246

sroiglid scourges, lashes
Milan: pret pass. 3sg. nosroiglide 32c3.
Denom. < sroigell scourge (DIL), Lat. flagellum whip, lash
(GOI). S- for f- is regular: Lat. f- was borrowed in Proto-Irish as *sw-. For
-r- for -l- see GOI, McManus and cf. W ffrewyll. Direct and indirect Latin
LEIA S-186, DIL s.v. sroigell 375: 76, sroigell 375: 76, GOI 921,
McManus 1983: 51-6.
subaigidir rejoices at, is cheerful

Milan: pres. ind. 3sg. subaigidir 122a11, 3pl. subaigit 61c18, VN
subugud 51c8, acc. sg. subugud 67d5.
Denom. < subae joy, pleasure.
LEIA S-198, DIL s.v. subaigidir 407: 32, subae 406: 57.
suidid puts, places (Schulze-Thulin)

PIE *sed- sit, caus. *sod-i e/o- (Schulze-Thulin)
CC *sod-- (Schulze-Thulin)
Cf. Goth. satjan, OIsl. setja, OHG sezzen < PG *sat-eja. MW
paratoi is a cmpd. of adj. *parad (< Lat. paratus prepared, ready > late
Brit. *pard > shortened and de-rounded with -o- in following syllable
*parad-) + -hoi < *hoi < CC sod-- (Schulze-Thulin).
adsuidi (stops, holds back): Wrzburg: pres. subj. 2sg. with infix.
pron. 3sg. m ciadsode (= ci-ad-dn-sode) 10a9, with infix. pron. 3sg. f.
ciatasode 10a10, 3pl. maniastat 29al6, impv. 3sg. with infix. pron. 3sg. m.
nach nastad 10a7, Milan: pres. ind. adsudi 114c6, pres. subj. 2sg. n
astaesiu 55a19, perf. 3sg. atrosoid 39a16, adrosoid 97d16, part. astide
20a22, VN (m u) astud 15c4, acc. sg. astud 15a15.
Causative. Two causative formations remain in Common Celtic of
the Proto-Indo-European root *sed-, sit > OI saidid. There is -grade CC
*sd-- > OI sidid (page 238), and -grade PIE *sod-i e/o- > CC *sod-- >
OI suidid (Schulze-Thulin).
247

Ped. ii 637, IEW 884, LEIA S-7-8, LIV2 513-5, DIL s.v. suidid 417:
5, ad-suidi 72: 1, Sjoestedt 1925: 327-8, Seebold 1970: 396f., SchulzeThulin 2001: 130, 136.
suidigidir establishes, founds

suidigidir: Wrzburg: perf. 3sg. rosuidigestar 12a30, Milan: pres.
ind. 3pl. rel. suidigitir 94c3, pass. sg. suidigthir 120d9a-b, rel. suidigther
87d15, pl. suidigter 65c16, suidigtir 120d9a-b, fut. 2sg. suidigfe 22d10, 3sg.
fuidigfith 46c20, perf. 1sg. rel. rusuidigsiur 59b2, 2sg. rosudigser 121a12,
3sg. rosuidigestar 46c20, 63c10, pass. rusuidiged 23a18, part. suidigthe
18d20, 48d6, suidigthi, VN (m u) suidiugad, acc. sg. suidigud 60b16;
consuidigedar (compounds, arranges): Milan: VN comsuidigud 126b10;
etarsuidigedar (places between): Milan: perf. pass. etarrusuidiged
27d23,
part.
(substitutes):
etarsuidigthe,
Wrzburg:
etarsuidigthi
perf.
3pl.
rel.
138d3;
fosuidigedar
forrusuidigsetar
7b5;
*frissuidigedar (places over against, opposes): Milan: past subj. 3sg.

conifrithsuidiged 34a8, nadfrithsuidiged 39a21, part. (io, i) frithsuidigthe,
nom. pl. frithsuidigthi 39c12; armisuidigedar (places after): Milan: perf.
3sg. iarmurusudigestar 130a7; inuidigedar (institutes, establishes):
Milan: part. nom. pl. anat n-insuidigthi 146a4, VN inuidigud, acc. pl.
insudigthisiu 30b5.
Denom. < suide act of sitting, sitting down, VN of saidid.
LEIA S-200, DIL s.v. suidigidir 417: 22, con-suidigedar 463: 74,
etar-suidigedar 224: 3, fo-suidigedar 385: 33, *fris-suidigedar 430: 51,
armi-suidigedar 26: 80, inuidigidir 286: 38, 2. suide 415: 36, saidid 17:
80.
sulbairigidir is eloquent, fluent

Wrzburg: fut. 1sg. sulbirigfer 12d12, pres. subj. 1sg. cia sulbiriger
12d10, 2pl. rel. sulbairichthe 12d8, impv. 3sg. sulbairiged 12d9, sulbirigfer
12d12, sulbiriger 12d10, sulbairichthe 12d8, sulbairiged 12d9.
Deadj. < sulbair eloquent, well-spoken, < so good, excellent +
labar talkative, arrogant. The verb is not attested outside of the Wrzburg
Glosses.
248

DIL s.v. sulbairigidir 428: 68, sulbair 428: 15, 2. so 308: 25, 2.
labar 12: 22.
suthchaigidir makes fruitful

Milan: perf. 2sg. rosuthchaigser gl. foetasti 81b9.
Deadj. < suthach fruitful, fertile, < suth fruit, produce. Hapax
legomenon. The verb could be a calque on Lat. foetare breed, which it
glosses, since both these verbs are derived from adjectives meaning
fruitful (Lat. foetus). Calquing has been discussed in section 3.4.3, pages
69ff.
LEIA S-205-6, DIL s.v. suthchaigidir 434: 54, suthach 433: 48, 1.
suth 432: 84.
tairisnigidir trusts in, relies on

Milan: impf. 1pl. notorasnigmis 43d4, 3pl. rel. annutorasnigtis 85d4,
pres. subj. 3sg. rel. tairisnigte 39b9, past subj. 1sg. notorasnigin 44d25,
nutarasniginn 46b13, perf. 3sg. rotorasnaigestar 106b8, verb. nec.
nantorasnigthi 144a3.
Denom. < tairisiu faith, trust.
LEIA T-30, DIL s.v. tairisiu 39: 68, tairisnigidir 41: 3.
taitnigidir shines, is bright

Milan: perf. 2sg. rotaitnigser 105c7, VN totnugud (this seems the
only time that the VN is extant. Irregular: the consonant cluster should be
palatal) 43a11.
This verb is derived from a VN *taitin that is not attested in Old
Irish itself, but that is extant in Scottish Gaelic taitinn with the meaning
pleasant, nice. This was originally the VN of doaitni shines, is refulgent
(itself deadj. < tind bright, dazzling), of which the extended VN taitnem
shining, gleaming is attested in Old Irish. That taitin existed in Old Irish is
borne out by its continuation in Scottish Gaelic and by the existence of the
denominative verb taitnigidir that is derived from it.
LEIA T-21, DIL s.v. taitnigid 60: 6, do-aitni 198: 35, taitnem 59: 31.
249

teimligidir grows dark, dull
Cf. W tywyllau, MB teffalhat, ModB tevalaad.
fristemligedar (darkens, obscures): Milan: pres. ind. pass. sg.
fristemligther 73c2.
Denom. < teimel darkness, gloom, < teim darkness, death. The
simplex does not appear in the Glosses and the compound is hapax
legomenon and might be a calque on Lat. obtenebrare darken, which
occurs further down in the Latin sentence. Both the Old Irish and the Latin
verbs are derived from nouns meaning darkness (Lat. tenebra) and the Old
Irish preposition translates the Latin. Calquing has been discussed in section
3.4.3, pages 69ff.
LEIA T-48, DIL s.v. teimligid 115: 42, fris-temligid 431: 32, teim
114: 17, teimel 114: 41, Joseph 1987: 150.
tessaigidir warms, heats

Cf. OB adj. teshegetic (Joseph).
tessaigidir: Milan: VN tessugad, dat. sg. thessugud 42c28;
contessaigethar: Milan: perf. 3sg. conrutessaigestar 59a16.
Denom. < tess heat, warmth. Contessaigethar has not been
translated by Thes. or DIL. The form in Milan above glosses Lat. concaluit,
which is the perf. ind. 3sg. of either concalere be, become warm or
concalescere become, grow warm. Tessaigidir can be found in the
LEIA T- 54-5, DIL s.v. tessaigid 159: 3, contessaigethar 465: 10, 2.
tess 158: 7, Joseph 1987: 150.
tibid touches, breaks against, develops into laughs, smiles (Vendryes)

contibi (laughs at, mocks): Milan: fut. 3sg. contifea 17a3, VN
cuitbiud, dat. sg. cuitbiud 86a11.
Etymology unknown. LEIA reports a discussion based on trying to
tie this to other Indo-European verbs and the assumption of the verb being
primary. However, they also state that none of the solutions proposed can be
imposed with certainty. Vendryes discusses the semantics of the verb, but
leaves the etymology mostly aside.
250

Ped. ii 648, LEIA T-61-2, DIL s.v. tibid 165: 30, con-tibi 465: 44,
Vendryes 1938: 38-45.
trmaigidir dries, dries up

Milan: pres. ind. 3sg. rel. cothrmaigid 44d8.
Deadj. < trim, tirim dry, formal, < tr dry.
LEIA T-75, DIL s.v. trim 189: 16, 2. tr 188: 44, trmaigid 190: 7.
*tocaid destines
PIE *temk- become solid, become manifest, caus. *tomk-i e- (LIV2)
CC *tonk-- (Schumacher)
Cf. MW tynghaf I destine, MB tonquaff destine (Schumacher).
Milan: VN tocad 19d5, 35d22.
According to Pedersen and McCone this is originally a nasal present
that has taken on weak flection. LEIA mentions a causative morphology,
however, Schumacher (followed by Schulze-Thulin and LIV2) has
conclusively shown that this is in fact a causative of the same root that gave
OI tongaid swears. The distinction between nasal present on the one hand
and causative on the other is clearly seen in MB toeaff swear and tonquaff
destine respectively (Schumacher p. 51). OI tocaid, too, behaves perfectly
regularly like a causative, and it can be argued that early Middle Welsh
would have preserved a distinction between tynghaf I destine and tyngaf I
swear, which would have merged due to the loss of post-tonic -h- during
the Middle Welsh period (Schumacher p. 51). On Gaulish toncsionto, see:
Fleuriot, Schmidt, Koch and Lambert.
Ped. ii 650, IEW 1068, LEIA T-84-5, LIV2 625-6, DIL s.v. tocaid
200: 83, tongaid 246: 53, Fleuriot 1976-7: 184, Schmidt 1981: 266, Koch
1992b: 249-261, Schumacher 1995, Charles-Edwards 1995: 1-15, McCone
1996: 107, Schulze-Thulin 2001: 158, Lambert 2002: 278.
tochomrachtaigidir is weary
Milan: impf. 3sg. notochomrachtaiged gl. tedebatt 127c13.
Denom. < tochomracht weariness, vexation, VN of *to-com-reth-,
compound of reithid runs, hastens (page 306). The compound itself is only
251

attested once, in Wb. 14b23, as toncomra (prima manus), glossing tederet
nos, impf. subj. 3sg. of taedere be tired, weary. Pedersen reports a reading
toncomrit pret. 3sg. due to Zimmer, reproduced without attribution by Wb.
Lex. The denominative tochomrachtaigidir is hapax legomenon as well.
Ped. ii 597-8, LEIA T-87, DIL s.v. tochomracht 205: 24, do-comra
233:13, tochomrachtaigid 205: 37, Wb. Lex. s.v. to-comrit.
todid shines, illuminates

astodi (shines, appears): Wrzburg: pres. ind. 3sg. rel. stidi
12b28, Milan: pres. ind. 3sg. astoidi 40c15, 115d3, 3pl. (rel.) astoidet 2a14,
impf. 3pl. rel. astoitis 99a4 (2x), pres. subj. 3sg. co is tuaethe 131d15,
impf. / past subj. 3sg. with infix. pron. 2sg. attided 56c3, VN attodiud,
dat. sg. atoidiud 85c5, nom. pl. atoitecha 20d7.
This verb is derived from a noun *to-id-, a compound of the root
*ed- see.53 The noun is no longer extant in Old Irish and the verb
therefore probably predates Old Irish (page 81). Since this ultimately goes
back to the see root, it can tentatively be argued that the verb fits the
scheme, although the exact meaning of the underlying noun remains
unknown. shines means causes to be seen.
Ped. ii 650-1, LEIA T-95-6, DIL s.v. tadid 25: 21, as-todi 440: 13,
Zupitza 1901: 281, Stokes 1905a: 161.
toirthigidir fructifies, makes fertile

Wrzburg: pres. ind. 3sg. rel. toirthigedar 26c3.
Denom. < torad produce, increase.
LEIA T-112, DIL s.v. toirthigidir 231: 8, torad 253: 36.
toltanaigidir pleases, is pleased

Milan: VN toltanugad, toltanugud 74d9.
Denom. < toltu, desire.
LEIA T-101, DIL s.v. toltanaigid 242: 77, toltu 243: 5.
53
Earlier disagreements over the etymology of this verb (Ped. ii 650-1, LEIA T-95-6,
Zupitza 1901: 281, Stokes 1905a: 161) were based on imperfect understanding of the
relevant soundlaws.
252
tosngachtaigidir weighs
Milan: pres. ind. 1sg. tosgachtaigim gl. pendo 79a9, past subj. 2sg.
notosngachtaigthea (Note Thes.: the last h over the line; is 2 sg. past subj.
and a mistranslation, cf. Ml. 19b12, 79a9) gl. pendere 78c3,
tosngachtaigtae gl. inpendentia 19b12.
There are no nouns attested in Old Irish that this could be derived
from, but this is a *-sag- derivative and must therefore be secondary. LEIA
suggests a VN of *to-oss-snig (oss = uts), compound of snigid snows. The
verb is not attested outside of Milan.
LEIA T-118, DIL s.v. tosngachtaigidir 264: 28, snigid 302: 1.
trebrigidir continues
Milan: pres. ind. 1sg. notrebrigim 88d7, part. trebrithe 88d10.
This verb could be derived from trebrech continuous, with
haplology of -ech-. Certainly, trebrech seems closest in meaning to
trebrigidir and haplology of -ech- is not unheard of. However, as Joseph
points out, the meaning given in DIL for the adjective trebrech seems based
entirely on that of the verb. Deriving the verb from this adjective based on
the meaning therefore becomes a circular argument. Joseph derives
trebrigidir from trebar careful, prudent instead, although he correctly
points out that there are semantic issues with deriving a verb meaning
continues from an adjective meaning careful, prudent. It is however
noteworthy that careful and prudent are not the only meanings of this
adjective; DIL also gives secure, solvent, strong and in compounds it
comes to mean things such as strongly secure and steadfastness. With
that background, I think it is entirely possible to derive this verb directly
from trebar.
LEIA T-126-8, DIL s.v. trebrigidir 285: 63, trebar 283: 78, treb
280: 69, Joseph 1987: 135 n. 127.
trnaigidir strengthens
253

immtrnaigedar (enjoins, assures): Milan: pres. ind. 3sg. rel.
immetrenaigedar 62c5, perf. pass. sg. imruthrenaiged 102d15, part.
imthrnaigthe 42c21, VN (m u) imthrnugud, dat. sg. imthrenugud 89a2.
Deadj. < trn strong. The semantics of this verb have been briefly
LEIA T-135-6, DIL s.v. trn 296: 25, trnaigid 297: 72, immtrnaigedar 158: 30.
tailngigidir deigns, considers worthy

Milan: perf. ind. 3sg. n rothuailngigestar gl. non est dignatus
16b12, with infix. pron. 1pl. nuntuailngigedar gl. dignatur 146b2.
DIL derives this from tailnge ability, capability, < talaing able
to, capable of. LEIA derives the verb directly from talaing. Both are
possible. The verb only occurs in Milan.
LEIA T-160, DIL s.v. tailngigidir 335: 33, tailnge 335: 17,
talaing 339: 22.
tuigithir covers
PIE *(s)tegcover, caus. *(s)tog-i e- (LIV2)
CC *tog-Cf. MW toi cover (Schulze-Thulin)
fortuigethar covers: Milan: verb. nec. fortgaidi 90d2, part.
fortchide, fortgide 29d14, adv. infortgidiu 30a3; intuigethar (covers,
clothes): Wrzburg: pass. pres. subj. 3sg. rel. intoichther 13d20; Milan: fut.
3 pl. intuigfet 121c9.
DIL and LEIA refer to tugae, tuige act of covering, cover, also
used as VN of tuigithir, but this verb is an an iterative of PIE *(s)tegcover (Sjoestedt, Schulze-Thulin, LIV2). Motivation for formation as a
deponent verb has been discussed in section 3.3.1.3, pages 63f.
Ped ii. 654-5, IEW 1013, LEIA T-169-70, LIV2 589, DIL s.v.
tuigithir 358: 42, for-tuigethar 369: 43, in-tuigethar 293: 73, tugae 353: 79,
Sjoestedt 1925: 328, Schulze-Thulin 2001: 133.
254

tuilid sleeps, falls asleep
PIE *telH- become silent, caus/it *tolH-i e- (LIV2)
CC caus/it *tol-Cf. OCS u-tolj, -toliti stills (LIV2).
contuili (sleeps, falls asleep): Wrzburg: perf. 3sg. ma \atil
29d15, Milan: VN cotlud 95a6.
Causative.
Ped. ii 655-6, IEW 1061-2, LEIA T-170, LIV2 621, DIL s.v. tuilid
360: 33, con-tuili 464: 37, Strachan 1893a: 369.
*adfaligedar unveals, reveals (translation mine)

Wrzburg: part. adfialichthe, dat. sg. fem. adfialichthi, gl.
reuelata 15b4.
Derived from fal veil < Lat. uelum (cf. section 3.2, pages 48ff.),
with preverb ad. It could be a calque on Lat. reuelare show, reveal,
which it glosses. Latin uelare means veil, cover and combined with the
preverb re becomes unveil. The glossator has translated the components of
the participle individually and created an Old Irish participle out of them.
Cf. indenmichthe weakened above, page 70 and see section 3.4.3, pages
69ff. Note that since the verb itself is not attested, it is not certain if this
verb would have been treated as a simplex or a compound verb (cf. section
3.5, pages 74f.). The hypothetical citation form given is as if this were a
compound, but might have been *adfadligidir.
DIL s.v. adfalichthe 7: 29, 2. fal 116: 47, 4. ad 7: 14.
aibrigidir treats insolently, profanes

Milan: fut. 3sg. huaibrigfidir gl. frustrabitur 109b08, perf. 3sg. rel.
runuaibrigestar gl. profanauerat 73b10, part. aibrigthe, nom. pl. uaibrigthi
gl. profanatis 100c16-17, VN aibrigiud, uaibrigiud gl. frustratio 67d18,
dat. sg. huaibrigud gl. frustratione 63c2.
DIL derives this from aibrech proud, arrogant, < aibre pride,
but as discussed above, section 3.4.1, page 67, the principle of economy has
been applied and this verb has been derived directly from aibre so that
255

there is no loss of -ech- to be accounted for. aibre itself is derived from
abar pride, arrogance (LEIA). The verb is only attested in Milan.
LEIA U-1, 3, DIL s.v. aibre 8: 7, aibrech 8: 12, aibrigidir 8: 49,
abar 3: 23.
aigid stitches together, sews

PIE *peug- stick, prick caus. *poug-i e- (LEIA)
CC *g-- (LEIA)
Cf. Lat. pungere prick, pierce (LEIA, LIV2).
conigi (sews, knits together): Wrzburg: pret. 3pl. conogset
(Note Thes.: Here the glossator seems to have mistaken consuerunt (from
consuesco) for consuerunt (from consuo)) 19a1, Milan: perf. 3sg. conruaig
99a2.
Iterative from PIE *peug- stick, prick, *poug-i e-, CC *g-- (cf.
LEIA).
Pedersen takes conruaig Ml. 99a2 above as a compound of figid
weaves rather than a form of conigi. There are no etymological or
contextual grounds on which preference can be given to this or taking it as a
form of conigi. It glosses Lat. contexuit, perf. 3sg. of contexere weave,
entwine. The Old Irish continues .i. arrudibaid .i. rouc cechnrdataid as
conrubu chrn i.e. destroyed, i.e. brought all freshness out of it so that it
became withered. Both sewing and weaving fit the Latin verb, and the
Old Irish context adds nothing to our understanding. Both the perf. 3sg. of a
verb config, of which a pret. 3sg. is attested in ACC VIII 19 (Bisagni), and
that of conigi could be conruaig.
IEW 480, LEIA C-197, U-3, LIV2 480, DIL s.v. aigid 9: 3, conigi
454: 54, figid 131: 63, Bisagni 2008: 271.
aisligidir exalts, ennobles

Cf. MB vhelhat, ModB uhelaat (Joseph).
foraisligedar (ennobles, exalts): Wrzburg: pres. ind. 1pl. with
emp. pron. 1pl. niforuasligemni 17b17.
Deadj. < asal high, lofty.
256

LEIA U-4-5, DIL s.v. aisligid 18: 60, asal 39: 20, for-aisligid(ir)
370: 7, foraisle 370: 1, Joseph 1987: 150.
alchaigidir is proud, becomes proud

Milan: impf. 1sg. conuhualchaiginn gl. adrogarem 49b14.
Deadj. < allach proud, arrogant, < all, pride, arrogance. Hapax
legomenon.
LEIA U-6, DIL s.v. alchaigid 24: 37, 1. allach 23: 68, 1. all 22:
53.
atigidir becomes few

Milan: pres. ind. 3pl. uatigitir gl. rarescunt 33a15.
Denom. or deadj. < a(i)te oneness, a small number or few.
Hapax legomenon.
LEIA O-35, U-5, DIL s.v. atigidir 45: 20, 1. aite 19: 26, 2. aite
19: 61.
uimligidir humbles
Milan: pres. ind. 3pl. rel. huimligte 136a11.
Deadj. < umal humble, obedient, Lat. humilis low, humble.
Direct and indirect loans have been discussed in section 3.2, pages 48ff.
LEIA U-25, DIL s.v. umlaigid 85: 22, umal 80: 74.
raigidir grows green or fresh

Cf. MW irhau make or become fresh (Joseph).
Milan: pres. 3sg. rel. nuraigedar gl. uirere 15b14.
Deadj. < r fresh, new.
LEIA U-26-7, DIL s.v. raigidir 90: 70, 2. r 87: 69, Joseph 1987:
150.
257
Appendix 1: Weak primary verbs
6 Appendix 1: Weak primary verbs

adgathar fears, dreads
PIE *h2egh- become afraid, sad (LIV2)
CC *g- (Schumacher)
Wrzburg: pres. ind. 3sg. n gathar 1a3, with infix pron. 3pl. n
snagathar 6a7; Milan: pres. ind. 1sg. nadnagursa, 74b19, rel. adnagur
74d4, 3sg. (rel.) adagadar 53c14, rel. agathar 87d15, rel. agether 129a2,
nadagathar 129a2a, nadnagathar, 74b21, nitagathar 129a12, pass. sg.
with infix. 3sg. m adnaigther 27c1, impv. 3pl. agatar 51d11, impf. 1sg.
adagain 63d11, pres. subj. 3sg. rel. coruagathar 66a2, fut. 1sg. adaichfer
68c17, 3sg. (rel.) adaichfedar 46c20, 3pl. neg n aichfetar 80b13, perf. 3pl.
adraichsetar 80d4, conaruaigsetar 35c4, adraigset 124b6, adraigsetar
124b6; verb. nec. aigthi 104a5, VN (f n) igthiu, acc. sg. aigsin 51d12, gen.
sg. aichthen 82d8.
Common Celtic perfect re-formed to an AII (Schumacher).
Ped. ii 454-5, IEW 7-8, LEIA A-23, GOI 553, LIV2 257, DIL s.v.
ad-gathar 27: 11, LIV2 257, Hull 1947: 422-6, Watkins 1969: 122,
Schumacher 2004: 206.
adella visits, approaches

PIE *pelh2- approach, nas pres. *pl -n/n-h2- (LIV2)
CC *al-naadella: Wrzburg: pres. ind. 3sg. with infix pron. 3pl. ataella
19b15, fut. 1sg. adeilliub 14a7, with rel. n adnelliub 14a6, with infix pron.
2pl. atdubelliub 7a4, Milan: pres. ind. 3pl. rel. adnellat 102a15, VN ([n] o)
adall, nom. pl. adaill 102a15, doaidlea (comes to, approaches):
Wrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. f. dodaaidlea 9d5, fut.
3sg. with infix. pron. 3pl. dosnaidlibea 25d14, perfect. pres. subj. 2sg.
nitirle 30d20, Milan: fut. 1pl. doaidlibem 14d5, VN tadall, acc. sg. tadal
25a3. 1. doalla (takes away, steals): Wrzburg: perf. 3sg. with infix.
259

pron. 3sg. n.: .i. madudll 22b7, Milan: pres. subj. 1sg. tall 58c6, 2. doalla
(there is room for): Wrzburg: pres. ind. 3sg. nitella 5c13, fut. 3sg.
nitellfea 26a23, nitelfea 25b18, with infix. pron. 3sg. n. nthelfea 25c2,
Milan: pres. ind. 3sg. nitalla 65d6, 97a7, doecmalla (gathers, collects):
Wrzburg: pres. ind. 3sg. doecmalla 9d5, with anaph. pron. doecmallaside
9d5, (pret. &) perf. 3pl. tercmlsst (a prima manu; Thurn. Gr. p. 519 tue-r-c[h]omlassat) 7a7 (= 7a5a, Thes. Suppl. p. 42 ), Imperat. 2pl. tecmallid
14a1. 539, doella (turns aside, deviates): Milan: impf. 3sg. rel. dunellad
83b6, pres. subj. 1pl. neg rel. nad dillem106c4, perf. 3sg. dureill 54d8,
3pl. dorellsat 36c22, rel. durellsat 105b18, verb nec dillithi 29d6, VN
andiall 119b9, foindlea (wanders, roves): Milan: pres. subj. 3sg. rel.
fonindlea 25a5, VN fondel, foindel 121d8, sechmiella (passes by,
passes): Milan: pres. ind. 3sg. nach sechmalla 35d13, sechminella 61a5,
impf. 3sg. ni consechmallad 33c21, fut. 1pl. nsechmalfam 25a3, pass. sg.
nisechmalfaider 14d3, VN sechmall, dat. sg. sechmall 70b3.
Schumacher separates the root of doalla there is room for, fits
from that of doalla takes away, steals and doella turns aside, deviates.
He bases doalla there is room for on a hypothetical weak verb *tallaid of
which the conjunct form talla was reinterpreted as the prototonic form of a
compound verb. There is however no reason to separate two roots. These
verbs are, as Schumacher himself points out, formally indistinguishable
already in the Glosses. There is overlap in meaning between the three of
them. In Wrzburg, both the doalla verbs are consistently spelled as doella
and in Milan they are both consistently spelled as doalla. Finally, the
derivation is wrong. Schumacher derives *tallaid from the root *telp- make
room (cf. LIV2) > *tl > *tal > OI tall-, which would be possible if the
development p > happened before the development *l > *li and > *al,
since *l became *li before stops. This development is however contradicted
by the derivation of OI cr, body, cognate to W cryf, < PIE *krp-. The
development can only have gone: *krp > *krip > *kri-. If *p > *, this
would have been: *krp > **kr > **kar (Cf. Isaac 2007c: 62, 71 for
relative chronology). That is not what we have, so by implication the
etymology of tallaid is wrong and therewith the justification for two roots
falls.
260

Ped. ii 510, IEW 801-2, LEIA D-119, LIV2 470-1, 623, DIL s.v. adella 52: 22, do-aidlea 181: 38, 1. do-alla 199: 7, 2. do-alla 199: 49, doecmalla 248: 80, 1. do-ella 252: 81, fo-indlea 251: 33, sechmi-ella 129: 75,
cr 524: 20, GOI 523, Schumacher 2004: 503-7.
adferta
PIE *ert- turns oneself (LIV2)
CC *ertMilan: pres. 3sg. adferta (note Thes.: leg. adfertaigedar?) 123d8.
No translation in DIL, Thes. translates the gloss as he turns. Hapax
legomenon, glossing Lat. aduersatur. As has been discussed above, page 83,
this is probably the e-grade primary verb corresponding to the causative
dofortai pours (page 200) that has taken on weak flexion. Thes.
emendation to adfertadaigedar has no basis in evidence.
Ped. ii 526, IEW 1150-1, LIV2 691-2, DIL s.v. ad-ferta 53: 32.
anaid stays, remains

PIE *h2enh1- breathe (LIV2)
CC *ana- (Schumacher)
Wrzburg: fut. 1sg. ainfa 14a8, sec. fut. 2sg. rel. noainfeda 27d20,
3sg. niainfed 32a11, perf. 1sg. nruanus 14d29, 2sg. with rel. n and emp.
pron. 2sg. ronanissiu 29d9, 3sg. rel. nadran 14d30, 1pl. nirransam 19d6,
perfect. pres. subj. 1sg. rel. corran 7a17, impv. 1pl. naanam 20c14, Milan:
pres. ind. 3sg. cena nid (Note Thes.: leg. ce anaid?) 27b16, rel. nanas
71b5, impf. 3sg. nunanad 83a4, pres. subj. 3pl. corruanat 42a4, fut. 1sg. n
ainiubsa (Note Thes.: MS. ainib: corr. Ascoli; but cf. Celt. Zeitschr. IV.
49) 53b8, perf. 3sg. ronan 126b1, 126b2, VN andud, dat. sg. andud
131d14, conosna (ceases, stops): Milan: impf. 3pl. nosnaitis 46b3, pres.
subj. 2sg. conosnai (note Thes.: rather conosnaesiu (Ascoli)) 70c3, fut.
3sg. neg n cumsanfa 80d5, perf. 1sg. neg n ruchumsanus 94b14, 3sg.
inruchumsan 32d26, corosan 113c5, VN cumsanad 33d20, acc. sg.
cumsanad 21c3, gen. sg. dinchumsanad 112c11, dat. sg. duchumsanad
121d12; fofasna (agitates, disturbs): Milan: pres. ind. 3sg. with infix.
261

3sg. m fafuasna 94c8, pass. sg. fufuasnither 66d17, part. fuasnaide, nom. pl.
fuasnaidi 22d8, VN fasnad 2d5, gen. sg. fuasnada 16b12 (2x).
Primary verb that has taken on weak flexion.
Ped. ii 455, IEW 38-9, LEIA A-71, LIV2 267-8, DIL s.v. 1. anaid
320: 9, con-osna 455: 79, fo-fasna 224: 12, osnad 165: 75, GOI 523,
Schumacher 2004: 196-7.
arcela takes away, steals

PIE *kelh1- turn, nas pres. *kl -n/n-h1- (LIV2)
CC *kal-na- (Schumacher)
arcela: Wrzburg: pres. ind. 2pl. arcelith 9c23, Milan: pres. ind.
3sg. with infix. 3sg. n. arachela 31a13, 3pl. rel. arachelat 134b10, pass. pl.
ercheltar 21c12, impf. 3pl. with infix. 3pl. arascelatais 26b19, pres. subj.
3sg. rel. aracela 36b2, fut. 3sg. with infix. arachelfea 18c9, VN (m u)
airchellad 28c5, erchellad 102a10-11, acc. sg. erchellad 57a2, airchellad
90a11, dat. sg. airchellad 36b2, erchellad 38a13, gen. pl. erchelta (this form
not in DIL) 63c16, doairchella (hides, withdraws): Wrzburg: pres. ind.
3sg. tairchella 12b33; doimmchella (goes round, makes a circuit of):
Milan: pres. ind. 3sg. duimchella 67d8, pass. sg. duimcheltar 108a14, past
subj. 3pl. doimchelltis 90d5; immcella (surrounds): Milan: pres. ind. 3sg.
(rel.?) immechella 67d12; immtimchella (surrounds): Milan: pres. ind.
3sg. pass. imthimcheltar 127b17, rel. immetimcheltar 121a1, 3pl.
imtimchellat 116a2, rel. immethimchellat 68d3, impv. 2sg. imthimchellsu
28d10, 3sg. imtimchella 40c13, impf. pass. sg. imthimcheltae 65c2, pl.
immintimcheltis 32a19, pres. subj. pass. sg. arnachanimthimchelltar 69b7,
fut. 1pl. imtimclfamni 24a7, sec fut. 3sg. imthimchellfad 43b9, perf. 3pl.
imimthimmerchelsat 44c25, immumthimmerchellsat sa 130b12, VN
doimthimchiull 120d2, dat. sg. imthimchiull 123b9.
Nasal present that took on weak flexion due to *-ln- > -ll- (McCone).
The simplex ceilid also lost the nasal, but remained strong and inflects in
Old Irish as a BI verb. Pedersen states that the compounds of this root have
become mixed up with the roots cell- and ciall-. It is therefore not possible
to make a clear distinction between the three.
262

Ped. ii 482, IEW 639-40, LEIA C-53-4, LIV2 386-8, DIL s.v. ar-cela
382: 17, do-aircheil 185: 9, do-immchella 309: 8, imm-cella 134: 46, immtimchella 157: 49, ceilid 98: 66, GOI 760, McCone 1996: 49, Schumacher
2004: 427-9.
asfena attests, swears

asfena: Wrzburg: pres. ind. 1sg. asfenimm 22a20, Milan: pres. ind.
3sg. (rel.) aspena 39b6, perf. 3sg. with infix. 3sg. adruspn (DIL gives this
as pret, but it contains ro) 78a5-6, VN aspenad, dat. sg. aispenud 54d2,
aspenud 102d2; doaisfna (shows, exhibits): Wrzburg: perf. 1sg.
doairfenus 18d7.
There is no Old Irish base to derive this from. It is therefore
problematic to classify it as secondary. If this is primary, its origins are
unclear. Pedersen derives this from PIE *ei d- as a nas. pres. *i-n/n-d- >
CC id-na-, cf. also IEW.
Ped. ii 517, IEW 1125-7, LEIA D-116-7, LIV2 665-7, DIL s.v. doaisfna 195: 42, as-fena 430: 16.
aslena pollutes, defiles

PIE *h2lei H- besmear (LIV2)
CC *li-na- (Schumacher)
Milan: pres. ind. pass. sg. rel. asslentar 92d12, rel. aslentar 92d12,
pl. aslentar 129c17, impf. 3pl. aslentis 74c3, with infix. 3sg. atlntaissom
22a4, perf. 3sg. rel. arruneillestar (cf. McCone 1985: 95) 63a14, 1pl.
nadreildisem 63d15, 3pl. asrulensat 74a3, pass. sg. roheilled 127a13, pl.
asrulenta 28a22, 100c18, part. eilnithe, gen. sg. masc illidi 63a16, verb.
nec. ilnithi, eillnithi 128d2, VN ilned, eilled 22b1, ennachneilled 137c5,
dat. sg. elled 92d12; *drailnid (corrupts, pollutes): Milan: part.
trailnide, nom. pl. druilnidi 76a5, VN trailned, acc. sg. druailned 38c9.
Simplex not attested. Nasal present that took on weak flexion.
The verb *drailnid is not attested in Old Irish. The participle and
VN found in Milan are based on the participle eilnithe from aslena with
intensive prefix dru- (Ped., LEIA, cf. GOI). The verb *drailnid itself is
263

found in Middle Irish, with d- changed to t- and probably back-formed
based on the then already existing participle and VN.
Ped. i 495, ii 565-6, IEW 662, LEIA T-154-5, LIV2 277-8, DIL s.v.
as-lena 432: 39, trailnid 326: 72, GOI 366, Schumacher 2004: 453-4.
csaid suffers, endures

PIE *kendh- undergo an impression, experience, desiderative *kndh-s(LIV2)
CC *kanss-eCf. Lit. keia suffer, Gk. suffer (Watkins).
Wrzburg: pres. ind. 1sg. rel. with emp. pron. 1sg. nocssimse 14d8,
1pl. with dem. rel. acsme 13c7, perf. 1sg. rel. with emp. pron. 1sg.
rochssusa (-sa = -ssa) 17d12, perf. 3sg. roces 6b20, 6c27, rocss 10c10,
rocess 4b13, rel. rochess 6c8, pres. subj. 2pl. with cia, ciacheste 25a10,
Milan: perf. 3sg. rocs 17a13, roces 44b1, rel. roches 44d4, VN (m u)
csad, dat. sg. cssad 74b1, csad 86d15.
PIE desiderative *kndh-s- reinterpreted as pres. ind. of a weak verb
(LIV2).
Ped. ii 486, IEW 641, LEIA C-79-80, LIV2 390, DIL s.v. csaid 149:
3, Watkins 1964-7: 131-4.
creitid believes, trusts

PIE *dheh1- put (LIV2)
CC *kret-di-dcreitid: Wrzburg: pres. ind. 1sg. nocretim 1a2, 13c10, with infix.
pron. 3sg. n. (ma)nid chretim 13c10, 3sg. rel. cretes 2b11, 15b1, 23a8,
creites 2b27, crettes 2b19, with infix. pron. 3sg. n. nadid chreti 15b14, 1pl.
rel. with emp. pron. 1pl. cretmeni 13b21, 2pl. nicretid 20a13, rel. nadcreitid
1d14, with infix. pron. 3sg. n. manidchretid 13b19, with rel. n and emp.
pron. 2pl. nonchretidsi 23c7, 3pl. rel. cretite 3d18, fut. 3sg. cretfed (for
cretfid) 1a3, rel. cretfes 1a3, 32d13, creitfes 4d5, creifess 4d21, 23b7, with
infix. pron. 3sg. n. nodchreitfea 4d7, 3pl. rel. cretfite 14c4, 15b1, creitfite
5c12, sec. fut. 3sg. rel. nochreitfed 5a20, pret. 3pl. o chretsit 31c7, perf. 1sg.
rocretus 17a6, 2sg. rocretis 10a29, 30, 3sg. rel. rochretti 5a7, rochreti 5b21,
264

24b31, nadrochreit 5c2, 2pl. with infix. pron. 3sg. n. and emp. pron. 2pl.
rachretsidsi 13b10, with emp. pron. 2pl. rocreitsidsi 25d20, sechrocreitsidsi
26a23, rel. diaruchretsidsi 8c11, rocretsisi 1a3, rel. rochreti 13a34 (leg.
rochretsid), nadrochreitsid 5c10, 3pl. ro creitset 1a3, canirochretset 10c20,
rel. rochretset 7b11, rochreitset 24d23, 31a6, perfect. pres. subj. 3sg.
actrocretea 27b15, corrochraitea 12c33, 1pl. mani cretem 30b2, perf.
actrocretem 27a15, 2pl. ni cretid 18c11, with infix. pron. 3sg. n. and emp.
pron. 2pl. nichretidsi 27a27, perf. past subj. 2pl. rel. with emp. pron. 2pl.
rochretesi 17a13, n ruchrete si 17a13, 3pl. n manu cretitis huili 5b21a,
pass.: fut. 3sg. rel. nad creidfider 28c14, impf. ind. 3sg. rel. 527 ndcrette
15a31, Milan: pres. ind. 3pl. n cretet 96b1, rel. crete 17b4, nocrete 33b26,
126c10, impf. pass. sg. with infix. 2pl. nobcrete 46a13a, pres. subj. 3sg.
arnarochretea 127a7, perf. 3sg. rochreti 46a14, narochret 33b5, 3pl.
rochreitset 35c20, rocretset 96b5, 123c1, 131c9, rel. nadrochretset 131c9,
131d11, with infix. 3pl. nsrochretset (Note Thes.: MS. nsrochret) 39d3,
rel. rochreitset 60b16, nirochreitset 90c22, verb. nec. creiti, creti 96b5, VN
creitem, cretem 15b4; cetacreti (first believes; this verb not in DIL):
Wrzburg: perf. 3sg. rel. cetaruchreti 7b11, 3pl. rel. caturuchreitset 14a29.
The Celtic athematic i reduplicated present stem was replaced by the
stem -d- in Celtic (Schumacher). This -d- was the original imperative of
the aorist. It is only attested in two compounded forms, neither of which are
still felt as compounds in the attested Celtic languages. In Old Irish, only
creitid (< *kred-d-) is attested, which is itself compounded to cetacreti. In
MW also roi (< *ro-d-) is attested.
Ped. ii 497, IEW 235-9, 580, LEIA C-228-9, LIV2 136-8, DIL s.v.
creitid 519: 55, Schumacher 2004: 278-81.
dedaid melts away, decays

PIE *dhe- walk, rush
Milan: pres. ind. 3sg. rudeda 22d7, 3pl. dedait 79b10, cudedait
(Note Thes.: MS. seems cdaitt; Sarauw 138 conjectures combaat; 77b10,
impf. 1pl. nodedmais (Note Thes.: MS. notedmais) 131c4, 3pl. nodedais
(note Thes.: rectius nodedtais, cf. noreithis 129d4) 102d14, pres. subj. 3sg.
265

dedas 104d1, 3pl. rel. nundedais 83b3, perf. 3sg. rodedussa 44d10, 50b3,
roded 111b12 3pl. rudedsat 22d6, conrodeda 118b2.
If the opinions expressed in the literature are correct, this is a
primary verb that has taken on weak flexion, but the derivation is quite
unclear, see Pedersen, IEW and LEIA.
Ped. ii 504-5, IEW 239, 260, LEIA D-34-5, LIV2 147-8, DIL s.v.
dedaid 208: 24.
doinscanna begins
doinscanna: Wrzburg: pres. ind. 3sg. with emp. pron. 3sg. m.
doinscannsom 17c8, fut. 1pl. with interog. in intinscanfam 15a1; foscanna
(tosses, winnows): Milan: pres. ind. 3pl. with infix. 3sg. fascannat 63b17,
impf. 1sg. fusscannainn 96a4, intinnscana (begins, commences): Milan:
pres. ind. 3sg. intinscana 26b10, 1pl. intinscanam 23c3, VN intinnscetal,
dat. sg. intinscitul 15a3.
Origin disputed, see above, page 85.
Ped. ii 613, LEIA D-158, S-32, DIL s.v. do-inscanna 319: 39, foscanna 376: 31, in-tinnscana 290: 18, Sarauw 1900: 70.
erbaid entrusts, commits

PIE *h1erbh (Schumacher)
CC ?*erb-(i-)e/o- (Schumacher)
erbaid: Wrzburg: pres. ind. 1sg. with infix. pron. 1sg. nomrpimm
6c3, 2sg. with infix. pron. 2sg. nterpi 1d10, pres. subj. 2pl. with infix. pron.
2pl. nob irpaid 8b2, impv. 2pl. with infix. pron. 2pl. and emp. pron. 2pl.
nachiberpidsi 22d6, pass.: perf. 3sg. roerbad 10d18, 25d20, roerpad 10d27,
18d15, 31a10, 3pl. roairptha 8c12, pres. subj. 3sg. rel. with dem. pron.
anerpther 31c12; Milan: pres. ind. 3sg. erbaid 14d15, with suff. pron. 3sg.
m. eirbthi 51b12, rel. with infix. pron. nundnerbai 65b6, impf. 3pl. with
infix. pron. 3pl. nudaerbtais 46d12, co nisnerbtais (Note Thes.: MS. nis
nerbtais) 46d10, past subj. 1sg. indamerbainn 29d5, 3pl. rel. with infix.
pron. nosnerbtis 85d4, fut. 3sg. rel. with infix. pron. nodneirbea 51b10,
perf. 3pl. roeirpset 43c18, VN (m u) erbud 43d3; conerbai (trusts):
Milan: fut. 3sg. with infix. pron. cotnerba 112a3, perf. 3sg. with infix. pron.
266

3sg. m./n. connidrerb 33b5, with infix. pron. 3sg. m./n. conidreirb 54b1,
with infix. pron. conidrerp 106b8.
Schumachers reconstruction is based on McCone 1999 who
connects OI erbaid, erbai with Pokornys (IEW) etymology of *orbh-. He
states that the underlying root is *h1erbh- and that erbaid, erbai are the
reflexes of a primary verb based on this root. The basic meaning of the root
is according to McCone leave (after death) (information taken from
Schumacher). Erbaid and its compounds inflect weak in Old Irish, which
according to Schumacher can be explained because of its future, which
would have lost reduplication due to syncope and thus become
unrecognisable as strong future and replaced by a weak f-future.
Ped. ii 513-4, IEW 781-2, LEIA C-191-2, DIL s.v. 2. erbaid 162: 18,
con-erbai 436: 46, McCone 1999, Schumacher 2004: 293.
ethaid goes
PIE *h1ei - go (LIV)
CC perf. pass. *i-to- (Schumacher)
atetha (goes towards, attacks): Milan: pres. ind. 3sg. adetha 48d12.
Common Celtic perf. pass. reinterpreted as s-pret. 3sg. act. conj. of a
weak verb. Weak present stem with pres. ind. 3sg. conj. etha was then
backformed. Supplies pret. pass. of tit (Schumacher).
Ped. ii 514, IEW 293-6, GOI 769, LIV2 232-3, DIL s.v. ethaid 236:
32, at-etha 444: 85, Schumacher 2004: 375-6.
fillid bends, kneels

PIE *el- turn, roll, nas pres. *l -n/nu- (LIV2)
CC *elinfill: Milan: perf. 3sg. inrufill 33c11.
Nasal present that has taken on weak flexion, cf. IEW and Pedersen.
Note that the only form attested for infill is an s-preterite. If this were
strong as DILs headword suggests, it would have a t-preterite. The
headword should be infilli.
267

Ped. ii 522, IEW 1140-2, LIV2 675, DIL s.v. fillid 135: 18, in-fill
252: 30.
adaid closes, fastens

PIE *deh3- give (LIV2)
CC *-dWrzburg: pass. fut. 3sg. rel. intain neidfider 32c13, Milan: VN (m
u) adad, gen. sg. ithiata 46a8.
The Celtic reflection of PIE *deh3- was an athematic root with
i-reduplication. This has been replaced by the verbal stem *-d- in the same
way as *-d- replaced the original root in the case of creitid above (page
264). The basis of this *-d- is the root aorist imperative *d, which itself is
also not an inherited formation (Schumacher). The original formation was
compounded with *epi- (< PIE *h1epi-dideh3-), but has been reinterpreted as
a simple weak verb.
Ped. ii 551, IEW 323-5, LIV2 105-6, DIL s.v. adaid 13: 31,
Schumacher 2004: 265-7.
legaid melts away, dissolves

PIE ?*leg- or ?*le- drip, trickle (LIV2)
CC *leg-e/o- (Schumacher)
dolega (destroys, abolishes): Milan: pres. ind. 3sg. (rel.) dolega
46d6, 62b20, impv. pass. sg. dilegar 127b5, pres. subj. 2sg. n derlegae
74d13 (2x), n derlega 74d13, pass. sg. with infix. 1sg. dundamlegthar
126d10, sec. fut. 3pl. dusleichfitis 84c20, VN (n o) dlgend 112d5, d(il)gent
33c13, 48b1, 53d1, acc. sg. dlgent 83a12, gen. sg. dilgint 48b17, dat. sg.
dilgiunn 33c15; legaid: Milan: VN legad 103c16, acc. sg. legad 94b3, gen.
sg. legtha 38c9.
The continuants of this root inflect as a weak -verb in Old Irish
(Pedersen, Schumacher). This could possibly be under influence of lgaid
reads.
Ped. ii 562, IEW 657, LEIA D-161, LIV2 397, DIL s.v. 1. legaid 72:
34, do-lega 330: 70, Schumacher 2004: 449.
268

mogaid makes greater, increases
PIE *me-, *m-i e/oCC *mag-i e/o- (Watkins)
Milan: part. mogthae 110a2, 124c18.
Wodtko et. al. express doubt about whether this verb is in fact
primary at all. However, there are no arguments for deriving it as a
secondary verb.
IEW 708-9, DIL s.v. maigid 152: 52, Watkins 1963a: 235, Wodtko
et. al. 2008: 470.
loingid eats, consumes

PIE *(s)legswallow, gulp, nas. pres. *slu-n-g- (LIV2)
CC *lu-n-gMilan: VN longud, acc. sg. longud 56a13.
DIL derives this as denominative from long in the meaning of
container, cauldron. LIV2 gives it as a nasal present that took on weak
flexion. I see two problems with DILs derivation. First of all, the semantic
route from container to eats is not direct and in Old Irish this indirectness
would have led to an -igidir verb, rather than an active verb. Secondly,
although the relationship between verb and noun could be seen as iterative,
using a container repeatedly (thus motivating the flexion as AII), this
stretches both the category of iterativity and the link between base and verb
uncomfortably far. This is not impossible, but since there is another option
available, namely deriving this verb from the primary root *(s)legswallow, gulp as LIV2 has done, the derivation from long is not necessary.
Note that Pedersen gives strong subjunctive forms foll etc., which DIL
gives as a completely different verb lsaid.
Ped. ii 568, IEW 964, LIV2 567-8, DIL s.v. 2. loingid 188: 47, 1.
long 199: 21, lsaid 248: 33.
oslaicid opens, lets loose

Milan: VN oslucud, acc. sg. oslucud 46b5.
Etymology contested, for details, see LEIA.
269

Ped. ii 563-4, LEIA O-33, DIL s.v. oslaicid 165: 14, Sjoestedt 1925:
328, GOI 181.
scaraid (separates, parts)

PIE *(s)kerH- part, separate (LIV2)
CC *skara- (Schumacher)
conscara (destroys, breaks): Wrzburg: pres. ind. 1pl. with emp.
pron. 1pl. incoscramni 2b20, nconchoscram 2b21, fut. 3sg. scra 26a8,
past subj. 3sg. rel. arn coscrad 10c1; pass.: perf. 3sg. conascrad 21b15,
Milan: past subj. 3sg. chonscarad 23b14, fut. 3sg. conscera 56d6, perf. 2sg.
with infix. 3sg. cotascrais 91b12, 3sg. incomscar 91c9, 3pl. ascarsat
87b22, part. coscarthae, acc. pl. chosgarthai 128b5, VN (m u?) coscrad
48c33, 49a10, 49a11, acoscrad (Note Thes.: MS. -cosrad; corr. Ascoli)
102b9, dat. sg. occoscrad (Note Thes.: MS. -cosrad; corr. Ascoli) 102b7,
VN coscar, acc. sg. choscur 95a5; doscara (overthrows, knocks down):
Wrzburg: impv. 3sg. with infix. pron. 2pl.: [na]chibtascrad 22b1;
etarscara (separates, divides): Wrzburg: perf. 3pl. rel. itirroscar(sat) (=
facs.) 5b34, perfect. pres. subj. eterroscra 9d31; pass.: fut. 3sg. eterscrthar
8b3, 3pl. eterscrtar 8b3, Milan: pres. ind. 3sg. neg rel. with infix. 3sg. m.
nachnetarscara 79b2, 2pl. etarscarid 67b4, pres. subj. 3sg. neg with infix.
3pl. naruetarscara 54d5, pass. sg. with infix. 1sg. neg nacammetarscarthar
47c10, perf. 3sg. retarschar 91c1, inruetarscar 91c1, with infix. 1pl.
etardanroscar 120a3, part. (io, i) etarscarthae 107c2, etarscarthe 30c5,
47a11, dat. sg. masc nephetarscarthu 54d3, VN etarscarad 124b6, dat. sg.
etarscarad 89d18; immtascra (struggles together, knocks down): Milan:
pres. ind. 3pl. rel. immehascrat 118c11; scaraid: Wrzburg: pres. ind. 3sg.
with infix. pron. 2pl. nbscara 13b19, rel. with infix. pron. 3pl. nodascara
28c15, impf. 1sg. noscarinn 24a4, perf. 3sg. rel. frissaroscarsom 3a14, 1pl.
roscarsam 24d26, 2pl. with infix. pron. 3sg. n. marudscarsid 27a30, 3pl.
roscarsat 28a20, 29b22, past subj. 2sg. niscartha 29d8, impv. 3sg. na
scarad 10a3, 29, Milan: sec. fut. 1sg. noscairiub 43a23.
Primary verb that has taken on weak flection (Schumacher). What
Pedersen calls a Middle Irish innovation of the -future is already present in
several Wrzburg forms seen above, cf. McCone.
270

Ped. ii 613, IEW 938-40, LEIA S-33-4, LIV2 558, DIL s.v. scaraid
74: 62, con-scara 459: 56, do-scara 371: 24, etar-scara 221: 56, imm-tascra
155: 13, GOI 651, McCone 1985: 93-4, Schumacher 2004: 576-8.
sechid asserts, declares

PIE *sek- say (LIV2)
CC *sek- (Schumacher)
diuschi wakes, starts: Wrzburg: pres. subj. 3sg. rel. with infix.
pron. 3sg. m. diusgea 4a6, consecha (corrects, reproves): Wrzburg: pres.
ind. 3sg. rel. with emp. pron. 3sg. m. secha som 7d10, (con)secha 26c1,
3pl. rel. sechat 6a9, perf. 3pl. with infix. pron. 3sg. n. marudchoiscset
28c7, perfect. pres. subj. 3sg. with emp. pron. 3sg. manirochoscasom
28b28, impv. 2pl. with infix. pron. 3sg. m. nacoscid 26b29,pass.: pres. ind.
3pl. coscitir 22c10, coisctir 31b25, 3sg. or 3pl. sechtar (from sechthar)
22c20, Ped. II, 69; fut. 3sg. with infix. pron. 2pl. cotob sechfider 9a23,
Milan: pres. ind. pass. sg. sechthar 57d14, perf. pass. sg. rocoscad 49a5,
part. coiscthe, nom. pl. coiscthi 49a7, VN cosc 109c14, dat. pl. choscaib
41d11-12; doduschi (awakens, brings back to life): Wrzburg: pass.: fut.
3sg. dodusgibther 9d3, perf. 3sg. dorodiusgad 9d3, 21b6, impv. 3sg.
todiusgadar 5c18, Milan: past subj. 2sg. with infix. 1pl. dundiusgada 102c3,
perf. 3sg. with infix. 3pl. dandersaig 66c14; dofsaig (announces):
Wrzburg: pass. perf. 3sg. dofarsiged 7d11.
See discussion above, page 83.
Ped. ii 619-20, IEW 897, LEIA C-199, D-132, 141, S-62-4, LIV2
526-7, DIL s.v. do-fuschi 266: 21, sichid 214:20-26, con-secha 460: 29, doduschi 241: 81, do-fsaig 263: 48, GOI 768, Schumacher 2004: 565-8.
sitid blows, blows across

PIE *sei-zd- fizz, whistle (IEW, Ped.)
CC *seizdCf. W chwythu, blow, blast, Corn. whythe, whethe, blow, Br.
choueza, blow (Ped.).
Milan: pres. ind. pass. pl. setir 38d18.
271

Based on the Old Irish, *sweizd- would be presumed, based on the
Brittonic, *swizd-, cf. OCS zvizdati, svistati, (assimilated from
*sweizd-). Irish st is close in meaning with seth, to which it however
cannot be related (Pedersen, LEIA). The primary root is tentative (note that
it is not taken up in LIV2, nor are any of the cognates given in IEW) and a
secondary origin can not be excluded for this verb, there is however no base
to derive it from.
Ped. ii 627, IEW 1040, LEIA S-76-7, DIL s.v. sitid 158: 6,
Schrijver 1995: 158, Isaac 1996: 206.
silid drops, flows

CC *seldofuisli (slips, stumbles): Milan: pres. subj. 3sg. dufuisledar
30c10, fut. 3sg. nicontuislifea 27b18, VN tuisled 2d5; dofochsla (seizes,
carries off): Wrzburg: perf. 3sg. doforchossol 13d27, 1pl. doforchosalsam
21b4. Milan: past subj. 3pl. tofoxlaitis 47a13; focoisle(a) (takes,
abstracts): Wrzburg: perf. 3sg. forrxul 27a19, Milan: perf. 3pl.
forochsalsat 18d11, pass. sg. foroxlad 31a5, VN fochsal, foxal, dat. sg.
difoxul 134a3.
Silid itself is not attested in Old Irish. It is originally strong with
reduplicated future and preterite (see Pedersen & DIL). Silid is in IEW
under root *sei-, *soi- (= *seh1i-) trickles, runs, but this root cannot be
brought further back than Celtic. There is no reason for this to be secondary,
since it inflects mostly strong and there is no suitable base to derive it from,
but there are no appropriate Proto-Indo-European roots for it either. The
preverbs of fofuisli are given by Pedersen, LEIA and DIL as *to-uts- with
the -f- being inorganic, but there is no reason for this. Although it is not
impossible for an inorganic -f- to develop this early, *to-fo-uts-sel- would
also give OI dofuisli. For dofochsla, too, a root *sel- could be given. It is
not clear if any of these verbs are directly related to each other, or indirectly.
Dofochsla could also be derived from sel a turn.
Ped. ii 621-3, IEW 889, LEIA D-146, S-78-9, DIL s.v. silid 226: 69,
do-fuisli 277: 74, do-fochsla 270: 37, focoisle(a) 205: 60.
272

sirid ranges, traverses
PIE *ser- join things up in a row (LIV2)
CC *sertirid (seeks, searches): Milan: perf. 3pl. rutuirset 44d23.
Pedersen sees this as a deadj. < sr long, lasting, but the vowel is
originally short (Bergin) so that is impossible. IEW registers sirid under
*ser- flows, runs. This is followed by LEIA, but this root has been
deconstructed and the various forms have been added to other roots by LIV.
Since there is nothing in Old Irish to derive this from, I do suggest this is
primary, and would assign it to PIE *ser- join things up in a row (LIV2
2.*ser-).
Ped. ii 628, IEW 909-10, LEIA S-116-7, T-174-5, LIV2 534-5, DIL
s.v. sirid 245: 55, tirid 368: 26, Bergin 1916: 196.
slucaid swallows, engulfs

PIE *slek- swallow, nas pres. *slu-n/n-k- (LIV2)
CC *slu-n-k-e/o- (Schumacher)
Cf. W lluncu, Br. lonka, MC loncaff.
Wrzburg: perf. pres. subj. 3sg. with infix. pron. 3sg. m. arnch
rllca 14d21, pass. perf. 3sg. roslogeth (a prima manu) 13d24, Milan: pres.
ind. 3sg. sluicthi 76b5, rel. sluces 22b1, 3pl. slogait 123d3.
This was originally a nasal present verb, but since the nasal was lost
through regular phonological processes, it was no longer regular as such and
has been reinterpreted as a weak i-verb (Schumacher).
Ped. ii 632, IEW 964, LEIA S-139, LIV2 568, DIL sluicid 281: 76,
GOI 677, Schumacher 2004: 593-4.
273
Appendix 2: Primary verbs
7 Appendix 2: Primary verbs

7.1 BI
ad-eirrig repeats, reitterates
PIE *rek- (LIV2 506, IEW 863)
CC *-rek-e/o-54
Wrzburg: pres. ind. 3sg. rel. adeirrig 22c2, pres. subj. 2pl. with
infix. pron. 3sg. n. diandaithirsid 9a23, 3pl. with infix. pron. 3sg. n.
indaithirset 30b30, impv. 2pl. aithirgid 9a23, Milan: pres. ind. 3sg. (rel.)
aderrig 46a21, with infix. 1pl. atanneirrig 114d10, pres. subj. pass. sg. rel.
-aithirrestar 32d13, pl. cho adersetar 30d11, fut. 1sg. aderrius 89b3, VN
aitherrach, aitherrech 94a3, 131a8, dat. sg. aithirriuch 22d5, VN aithrige,
acc. sg. aithirgi 16c10, 23a2, 23a5.
ad-ft tells, relates

PIE *ei d- (LIV2 665-7, IEW 1125f.)
CC *eid-e/oadft: Wrzburg: pres. ind. 3pl. rel. adfiadat 28c23, perf. 3sg.
adcuaid 21d11, pass. pres. ind. 3sg. adfiadar 18c8, with dem. rel. anadiadar
(Thes. & Wb Lex emend this to anadfhiadar) 27a10, Milan: pres. ind. 3sg.
(rel.) (ad)fet 31b7, (rel.) adfet 31b19, 84d4, 89d6, 94b11, 99d9, 110c5,
115a14, 123d4, 125a11, adft 62a6, 87d1, 115d9, 118d10, 1pl. (rel.)
adfiadam 14d10, 93d14, rel. with infix. asdafiadam 93d14, 3pl.
ciasidfiadat 19b11 (Note Thes.: MS. ciardfiadat, Windisch proposes
ciaradfiadat), rel. adfiadat 51d2, pres. ind. pass. adfiadar 14b4, anadfiadar
14b6, pass. sg. rel. adfiadar 72d9, 109b02, 128a15, 128d14, 128d15, pass.
54
The Common Celtic reconstructions are as given by Schumacher 2004. For details, s.v.
the respective headwords there. Where Schumacher reconstructs * for Common Celtic,
this has been retained, although it is questionable whether this is appropriate at this time.
See Isaac (2007c: 11-4, 62-4, 70-1).
275

pl. (rel.) adfiadatar 23a13, 46c14, 123d6, perf. 3sg. adcuaid 65a1, 110c5,
126b2, anadchuaid 123d4; asindet (declares, tells): Wrzburg: pres. ind.
3sg. assindet 6a9, 11a18, with emph. pron. 3sg. m. assindetsom, 19b14, fut.
3pl. rel. asindisset 30d8; pass. pres. ind. 3sg. with rel. n asn.indedar 13a36,
Milan: pres. ind. 3sg. (rel.) asindet 23c12, 40c20, 64c19, 65b14, 87d1,
100b8, 101a3, 102d16, 104c5, 108b6, assindet 95d3, (rel.) -aisndet 54a12,
103a1, rel. dianaisndet 54a20, asnindet 41a3, 42b18, 94b3, 3pl. -aisndedat
31a19, rel. asnindedat 42b7, pass. sg. asindedar 17a9, niaisndedar 119c1,
assindethar 90b18, pl. asindetar 92c9, impf. 3sg. atindided 123b13, impv.
2sg. aisndid 70c8, pres. subj. 3sg. rel. asnind. 23d2, rel. aisnindised 31a22,
2pl. asnindidid 115b11, 3pl. asindiset 23a19, past subj. 3sg. asindissed
131b1, with infix. 3sg. asidindissed 42b18, fut. 1pl. asindisem 35a6, 3pl.
asindisset 45b19, perf. 3sg. asrindid 25a10, 58a8, 64c10, 83d5, 107a12,
pass. sg. asrindes 104c8, verb. nec. nom. pl. diasndisi 60b10, VN (f n, GOI
730) aisnds 14c19, 14d10, isnds 14d03, acc./dat. sg. aisndis 23c7, gen.
sg. aisndsen 93d12, dat. sg. aisnds 42d10, 64d2, 66c2, aisndis 65a1,
aisdis 76b7, nom. pl. aisndissi 16a13, 34b7, acc. pl. aisndisnea 24b5, dat.
pl. aisndisib (dat. pl. not in DIL) 23a14, 27b6; doadbat (shows,
displays): Wrzburg: pres. ind. 3sg. with rel. n donadbat 3c21, with infix.
pron. 3sg. n. and emph. pron. 3sg. m. daadbatsom 24a7, with emph. pron.
3sg. m. tadbat som 4d10, tadbat som 10b21, tadbatsom 13c26, 26c1, perf.
3sg. dorbith 19c11, dorbuid 19c11, impv. 2pl. taidbdid 14d22, pass.:
pres. ind. 3sg. doadbadar 3b20, 7a13, 7d10, 12a11, 12d20, 14d30, 23a13,
27d13, 29c4, (doadb)adar 13b1, 3pl. doaidbdetar 27a27, perf. dorbas
3d21, 15a18, Milan: pres. ind. 3sg. doadbat 19d17, 20a8, 39c22, 53b27,
94c10, duadbat 42b27, 47a17, 51c11, 62d9, 100c7, rel. donad(ba)t 15a2,
rel. dunadbat 22c1, 77b6, 101a5, rel. dundadbat 89d6, (rel.) -tadbat 40a15,
56b33, 74d13, 100c3, 138c1, with infix. 3sg. ciadusnadbat 135b5, 3pl. with
infix. 3sg. daadbat (Note Thes.: The infixed pronoun anticipates the
dependent clause) 98c10, rel. dunaidbet 42c2, 138d12, pass. sg. duadbadar
40d3, 42c17, (rel.) -tadbadar 24d25, 32b18, 40c17 (2x), 40c22 (2x),
110d16, pres. subj. 2sg. condrbais 101c6-7, pass. sg. condrbastar 95b6,
past subj. 3sg. donaidbset 20a9, pass. sg. rel. donadbastae 37b23, perf. 3sg.
duarbuid 46b28, duarbaid 46d15, 62c5, durbaid 129c15, pass. sg. nad
276

tarbas 64d13, n trbas 65d16, VN (f n) taidbsiu 42b9, acc. sg taidbsin
96b18, dat. sg. taidbsen 128c7, 128d11, taidbsin 39c22, 40b7, 51c22,
56c11, 62a2, 71c9, taidbse 90b14, 94c10; inft (tells, relates): Wrzburg:
perf. 2pl. rel. conecdid 24c17, Milan: pres. ind. 3sg. inf(e)t 31b25, impv.
1sg. indiad (Note DIL: cf. Ml. 35d11, but innadiad gl. proinde 63a7)
103b11, past subj. (GOI 615) 1pl. infesmais 17d8, perf. 3sg. incuaid
123d7; *remiaisndet (tells beforehand, previously): Milan: pres. ind.
pass. sg. remiaisdider (in DIL remiisdider) 64c9.
agaid drives, impels

PIE 1. *h2e- (LIV2 255-6, IEW 4f.)
CC *ag-e/oadaig (drives, impels (to)): Wrzburg: pres. ind. 3sg. with infix.
pron. 1sg. rel. atomaig 10d26, with infix. pron. 2sg. rel. attotaig 6c16, with
infix. pron. 2pl. rel. atobaich 9c20, atobaig 19d10a, Milan: pres. ind. 3sg.
rel. with infix. 1pl. atannaig 93d12; immaig (drives around, drives):
Milan: pres. ind. 3sg. with infix. 3pl. imdaaig 66d18.
aingid saves, protects

PIE ?*h1egH- or ?*h1eH- (LIV2 231, IEW 290)
CC *aneg-e/oaingid: Wrzburg: pres. ind. 3sg. aingid 1d15, with infix. pron. 1pl.
nonanich 16a4, 5, fut. 3sg. niain 25d14, with infix. pron. 3pl. nsnain 1d1,
pret. & perf. 3sg. with infix. pron. 3sg. m. rananacht 17d6; adanaig
(escorts, conveys): Wrzburg: past subj. pass. rel. with infix. pron. 1sg.
atomanaste 14c20, Milan: past subj. 3pl. coadanastais 100c23, verb. nec.
adnachti, dat. pl. adnachtib 33d7, VN adnacul 22a11, 22b1, dat. sg. adnacul
100c23; adindnaig (leads): Wrzburg: pass. pres. subj. 3sg. with infix.
pron. 1sg. atdomindnastar 7a5; doindnaig (gives, bestows): Wrzburg:
pres. ind. 3sg. with rel. and emph. pron. 3sg. m. donindnigsom 28a17, fut.
3sg. lasse donindin 13b29, perf. (with com-) 3sg. with rel. n donecomnacht
33d8, re doncomnacht 1a1, tecomnacht 26d23, past subj. 1sg. with rel. n
and infix. pron. 3sg. m. and emph. pron. 1sg. dondnindinsinse 9b7, (with
elision of infix. pron. 3sg. m.) manitindnised 4b13, 2pl. with elision of rel. n
277

and infix. pron. 3sg. m. and emph. pron. 2pl. dodnindnastise (=
dodnindnaste si Ped. II, 456, cf. Add. & Corr., Thes. Pal. I, 725 ; leg.
dondnindnaste J.S., Thes. Pal. I, 551 note c) 9b7; 551, pass.: pres. ind. 3sg.
doindnagar 16d14, with rel. n donindnagar 14b15 (bis), 16a11, 3pl. rel.
tindnagtar 15c2, perf. (with com-) 3sg. docomnacht 14c33, amal
donecomnacht 25c26, with prep. and rel. frisatecomnacht 19c8, with rel. n
(and -ro-) dorrindnacht 20d15, pres. subj. 3pl. doindnasatar 17a2, Milan:
pres. ind. 3sg. rel. duindnaig 59c11, 3pl. rel. tinnagat 93a20, pass. sg.
doindnagar 103b13, pres. subj. pass. sg. -duindnastar 56a13, past subj. 3sg.
rel. with infix. 1sg. duindainsed 78b18, fut. pass. sg. doindnastar 46c20, pl.
doinnasatar 30c17, perf. 2sg. rel. dunecomnacht 56a18, 3sg. (du)ecomnacht
54c18, ducomnacht (DIL has this as 2sg., but Thes. translates it as 3sg.)
77c5, rel. donecomnacht 54c23, doecomnacht 54c26, rel. dunecomnacht
55c1, 96b5, rel. with infix. 3sg. d(un)decomnacht 69a16, part. thinnachtae
126d12, VN (n o) tindnacol, tinnacul 96d6, 105d10, acc. sg. tinnacul 93d6,
gen. sg. tindnacuil 55c1, dat. sg. tindnacul 15d6, 22d16, tindnaccul 61b28,
thinnaccul 124c10.
alaid nourishes, rears

PIE *h2el- (LIV2 262, IEW 26f.)
CC *al-e/oalaid: Wrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. m.
nodnai(l) 5b28, with infix. pron. 2sg. not ail 5b28, Milan: past subj. 1pl.
nonalmais 104d6, perf. 3sg. rommalt 45c3, verb. nec. ilti, ailti 130c2, part.
naltai 121c10.
areim accepts, receives

PIE *h1em- (LIV2 236, IEW 310f.)
CC *-em-e/oareim: Milan: pres. subj. 3pl. coarfemat 15d4, impv. 2sg. erim
132c15, fut. 3sg. arfema 105a8, VN (f n) airitiu, eritiu 59a15, acc. sg. eriti
29a15, dat. sg. airite 24a14, (?) eritnichi 132c15; arfom (accepts,
receives): Wrzburg: pres. ind. 3sg. rel. arafom 13c24, 3pl. arfoemat 34a6,
fut. 3sg. with infix. pron. 1sg. aromfoimfea 31a3, perf. 1sg. with emph.
278

pron. 1sg. arroiitsa 6d14, 3sg. maarroit 28d28, arrit 32d10, rel. ararit
9c10, and with emph. pron. 3sg. m. ararit som 4b19, araroit 24a32, 1pl.
rel. araritmar 9c10, 2pl. arittid 13a30, 3pl. rel. nd arrimsat 26a23,
Perfect. pres. subj. 3pl. rel. with emph. pron. 3sg. m. arna arromatsom
14d37, past subj. 2sg. with emph. pron. 2sg. ciarfemtha so 8d28, perfect.
3pl. with infix. pron. 3sg. n. dianderimtis 26a22, impv. 3sg. roimed 6d4,
pass. pres. subj. 3sg. arfemthar (MS arfenithar) 28c19, Milan: pres. ind.
2sg. rel. arafoimi 68a8, 3sg. (rel.) arafoim 17c7, 21c4, 29a3, pass. sg.
honeroimer 17c7, 3pl. arafoimat 22b4, pres. subj. 3sg. (rel.) arafoima 17c3,
arfema 68d13, 3pl. eroimet 30d13, past subj. 3pl. arafoimtis 28c18, perf.
3sg. anarret 16a4, ararot 17a8, arrot 17c7 (2x), araroiat 24d28, (rel.)
arroet (Cf. McCone 1985: 93) 17c3, 25d11, 28d13, arrot 25d10, araroet
25d11, -arroit 36a34, withinf 1sg. anaramroet 131b8; coneim (protects,
preserves, this compound not in Primrverben): Milan: perf. 3sg. roitatar
55c1, VN coimt (see also con-o), gen. sg. chometa 55d6, dat. sg. comt
100c19, 100c21; doeim (covers, shelters): Wrzburg: pres. ind. 3sg.
doem 9d2, perf. 1sg. with rel. n and emph. pron. 1sg. dorretsa 31a1, pres.
subj. 3sg. with infix. pron. 2pl. dofema (= do-f-ema) 5d34, Milan: pres. ind.
3sg. withnas dunem 112a5, 3pl. with nas donemat 39c22, pass. sg. duemar
87d15, with nas dunemar 39c7, with infix. 1pl. donemar 53b18, donemar
51d8, pl. neg rel. nad dimter 100a5, impv. 2sg. with infix. 1sg. dunem
62d4, doneim 110d9, pres. subj. 2sg. with infix. 1sg. codumemae 55b2, rel.
dunemae 93c18, 3sg. donema 23d3, 128b4, with infix. 1sg. dundamema
(Note Thes.: MS. dumamema: corr. Ascoli; dunamema is sufficient, cf.
Celt. Zeitschr. IV. 56) 87d11, with infix. 3sg. arinderoima 39c22, neg with
infix. 1pl. manndimea 88c2, pass. sg. donemthar 112a2, withinf 1pl.
dunnemthar 127c7, with infix. 2pl. codobemthar 53b15, past subj. 3sg. rel.
donemthae 120c11, with infix. 3sg. idnderoimed 55d4, 1pl. dunemmais
110b5, pass. sg. dimthe 39c7, pass. sg. with infix. 1sg. innandimithe fut.
1sg. (rel.) doemsa 37c19, 3sg. duema 53c2, 67c5, doemfea (Cf. McCone
1985: 93) 128c8, pass. sg. with infix. 2pl. dobemthar 53b18, 3pl. with infix.
pron. 2sg. dotemfet 112c1, perf. 3sg. doret 16c8, with infix. 3sg. dundret
40b8, part. dite 103a4, 103c5, gen. sg. / nom. pl. nephdti 85d13, VN (f n)
dtiu, acc. sg. ditin 19a14, 62b21, 83a8, dtin 47c18, dat. sg. ditin 37c15,
279

67d9, 67d10-11, 76d9, 90d3, 101c14, dite 38b5, dtin 57c4, diten 85a4;
*addeim (uncovers): Milan: perf. 3sg. huaderet 51d14-15.
ascid extends, stretches

PIE *kei s- (IEW 538)
CC subj. *keis-se/oascid: Milan: impv. 2sg. escse 65a4, pres. subj. pass. sg. ascesar
44a4, perf. pass. sg. asrochess 39c11, VN (f n) eiscsiu, escsiu 65a5; cessair
(carries, brings forth): Milan: pres. ind. pass. sg. cochessair 51b25, pret.
pass. 3pl. rocessa 114a6; rem-ess-cid (stretches before): Milan: past subj.
3sg. remiescsed (under this form in DIL) 33d20.
asgleinn investigates, examines

PIE *ghlendh- (LIV2 200, IEW 431)
CC *-glan-n-d-e/oasgleinn: Milan: pres. ind. 2sg. neg rel. nadneclainni 64a4, 3sg.
atgleinn 61a8, asglinn 70a12, with rel. asgleinn 105d9, 3pl. inglennat
137c2, impf. 1sg. asgleinn (Note Thes.: leg. asglenninn) 96a4, pres. subj.
inglis 140c7, 3sg. rel. nad ecail 56c8, verb. nec. ecailsi 15d7, 114b4, VN
eclaimm, acc. sg. eclaim 114b15; doeclainn (picks out, selects): Milan:
VN teclaim, acc. sg. teclaim 35a13; doglenn (selects, collects): Milan:
pres. ind. pass. doglennar 133b7; fogleinn (studies, learns): Milan: pres.
ind. 3pl. (rel.) with infix. fodglein 37b12; VN (n n) foglaimm, foglaim
14c10, 14c11; fordiuclainn (devours, swallows): Milan: pres. ind. pass.
sg. rel. fordiuclannar (Note Thes.: MS. for diuclannar) 104b5, pres.
subj. 3sg. arnachtfordiucail 36a32, 3pl. with infix. 1sg. fortamdiucuilset
44c32, fut. pass. pl. fordiuguilsiter 84d2, part. fordiulcailse, nom. pl.
fordiucailsi 134b5, VN fordiuclaimm 102a10-11, fordiuclaim 104b5, dat.
sg. fordiuclaim 19d5, 75b4.
atbail(l) dies, vanishes, perishes

PIE *gelh1- LIV2 208 (*gel-, *gel-, *gl- IEW 471f.)
CC *bal-ni-
280

atbail(l): Wrzburg: pres. ind. 3sg. atbail 4d15, rel. adbaill 16b11,
ni epil 30d14, 3pl. atballat 9d5, fut. 1sg. atbl 10d24, 3pl. atblat 1d4, Sec.
fut. 1pl. atblmis 4d9, Perfect. pres. subj. 1pl. rel. with emph. pron. 1pl. and
elision of infix. pron. 3sg. n. arnachrbalamni 4b19, Milan: pres. ind. 3sg.
rel. asindbail 57a10, adbaill 108a3, rel. -epil 73d7, -apail 91d2, ipv. 3pl.
aipleat 104b2, neg na eiplet 73d7, impf. 1sg. atbelain 54c33, pres. subj. 1sg.
asindbelsa 48a14, 3sg. asinbela 38b6, 1 pl. dianrbalam 107d04, 3pl. neg
arna eplet 77a13, past subj. 1sg. asindbelainse 131b13, 3sg. rel. neg arnach
aipled 85d8, 3pl. atbeltais 36d16, neg coniepltais 99b2, neg arna epltis
121d16, fut. 2pl. atbelaid 29c4, 3pl. atbelat 57a3, 102b10, sec. fut. 3sg.
atbelad 16c10, 1pl. atbelmais 96a10, pret. 3sg. trubalt 144d3, perf. 3pl.
attrubaltar (Note Thes.: = atrubaltatar, cf. Ml. 67b8, 100c27, Tur. 143)
100a1, narrative tense supplied by suppletive verb. ad-bath: pret. 3pl.
asindbathatar 36d10, pass. pl. atbatha (note Thes.: MS. at batha;
commonly atbathatar, cf. as-in-d-bathatar Ml. 36d 10, but cf. conidaptha
Rev. Celt. XI 450, Zimmer KZ. XXX 148; see Strachan, 28 Rev. Celt. 205)
98b8, VN (f n) epeltu 98c6, VN (of ad-bath) apthu, nom./acc. sg. apthu
74c11, nom. pl. apthain 104b5.
beirid carries
PIE *bher- (LIV2 76f., IEW 128-32)
CC *ber-e/o
beirid: Wrzburg: pres. ind. 3sg. berid 6a13, 29b9, rel. with infix.
pron. 2sg. notbeir 6c9, with 3sg. n. nchbeir 6c18, 2pl. rel. noberid 9d15,
3pl. berit 18c10, with suffix. pron. 3sg. n. -i with -t of 3pl. ending added
bertit 13a16, rel. with infix. pron. 3sg. f. nodaberat 28d26, fut. 3sg. with
suffix. pron. 3sg. n. brthi (Editors note b, MS berthi = Ped. II, 464: pres.
ind. 3sg. with suffix. pron. 3sg. n.) 23a19, 1pl. with infix. pron. 3sg. n.
nibram 29b15, pres. subj. 3sg. rel. with infix. pron. 1sg. and emph. pron.
1sg. no.domberasa 1a8, 3pl. rel. berte 9c12, past subj. 1sg. rel. with infix.
pron. 3pl. conosberinn 10d36, impv. 3pl. berat 13a11, na berat 31c12,
pass.: pres. ind. 3sg. berir 7a16, 10b19, 12d38, 23c11, 29a28, 31a4, 31d18,
32a29, rel. berar 16d7, 3pl. rel. bertar 25c23, fut. 3sg. rel. brthar 12d27,
3pl. with interrog. part. in inbrtar 15a3, pres. subj. 3sg. berthir 27d13, past
281

subj. 3pl. nobertis 16d4, impv. 3sg. berar 7b4, with infix. pron. 2pl. nachib
berar 27a24, perf. forms with ro-: 1sg. rouiccius 9b6, 3sg. rouic 27a22, with
infix. pron. 3pl. rosnuicc 5b3, with infix. pron. 1pl. ninruc 21b3, 3pl. with
infix. pron. 3sg. n. raucsat 26b11, Perfect. pres. subj. 3sg. rel. conrucca
12c32; Milan: pres. ind. 3sg. berid 42b7, rel. beres 27c1, mberes 129c8,
with infix. 3sg. n namber 86d16, rel. with suff pron. 3sg. beirthi 42b7, 3pl.
nodberat 16a17, no berat 16a18, neg. n berat 103b16, pass. sg. berair 80c7,
pl. bertair 26c6, pres. subj. 2sg. neg. n berae 74d13, 94a1, 3sg. with infix.
3pl. nosbera 43a2, 3pl. comber(at) 20b11, -berat 54a28, pass. sg. (rel.)
berthair 37a8, 37a10, berthir 37a8, fut. 3sg. rel. beraid 37a9, rel. bras
94b7, 3pl. neberat 54a23-24, 54a26, pret. 3pl. rel. bertar 127d6, rel. with
infix. 3sg. nundabertatar 82d9, pass. sg. brethae 52 line 15, Perfective
forms: pres. ind. 3pl. neg. nadrgat 75b3, pass. sg. ruucthar 35b10, past
subj. pass. sg. ructhae 104a8, pl. innaructais 125b7, rel. neg. arnaructais
125c2, perf. 3sg. rouc 99a2, with infix. 3sg. ruc 45a1, rel. with infix. 3pl.
rodaucai 46a19, 3pl. nadrucsat 23b5, -rucsat 54d16, with infix. 3pl.
nisrucsat 44a15, pass. sg. rel. runucad 104a8, pl. ructha 102c7, verb. nec.
with d dbrethe, dibrithi 58a16, VN (f a) breth 36c21, acc. sg. brith 16b5,
113d2, gen. sg. brith 33c13, dat. sg. brith 36c20 (used as nom.), 66c14,
73d2, 73d3, 103c15, acc./dat. sg. brith 95a11; adopair (offers up,
sacrifices): Wrzburg: pres. ind. 3pl. rel. with con, conidbarat 1b20, pass.:
pres. ind. 3sg. rel. adobarar 11b12, adoparar 11b15, 20d13, with dem rel.
anadobarar 10c3, with prep for and rel. part. foranidparar 5b6, 3 pl. rel.
adopart.ar 10c1, imperf. 3sg. rel. adopart.he 15d20, perf. 3sg. with emph
pron. 3sg. m adropredsom 15d20; Milan: pres. ind. 3sg. rel. with infix. 3sg.
adidnopair 66b4, pass. sg. adopar 73a10, pl. rel. adobartar 60b17, perf.
3sg. adrobart 32b23, 3pl. adrobartar (Note Thes.: MS. adrobartat; DIL:
adrobartat (sic MS.)) 14a16, pass. sg. atamroipred 44c17, VN (f a) idbart,
edbart 87b8, acc. sg. idbairt 43c1, edbairt 92b9, dat. sg. edbairt 33a10,
73a10, nom. pl. edbarta 47c14, 70b4, gen. pl. edbart 60b17; arbeir (lives,
eats): Wrzburg: pres. ind. 3sg. arbeir 10c6, pres. subj. 2sg. rel. arambere
28c11, pres. subj. 2pl. rel. (with elision of infix. pron. 3sg. n.) airbirid
11b25, 3pl. n arbarat 10c1, past subj. 3pl. ni airbertis 10b8, impv. 2sg.
airbir 29a25, fut. 1sg. (with elision of infix. pron. -a-, 3sg. n.) niairber
282

32a20; Milan: pres. ind. 3sg. arbeir 43d14, 60b11, with infix. araber 46b7,
rel. aramber 69a18, 3pl. (rel.) araberat 50d13, 67c17, 129c1, with nas
aramberat 61d13, 69a19, pass. sg. coarbertar 121b3, pres. subj. 2sg. ma
arber 133d8, 3sg. arbera 72b26, rel. arabera 86d12, pass. sg. armberthar
53c9, nondaberthar 134c6, impf. 1pl. erbirmis 135d3, 3pl. arabertis
131d10, fut. 3sg. -erbera 48b12, naderbra 110c2, sec. fut. 3sg. -erberad
48a5, pret. 3sg. -arbart 36a4, perf. 1sg. arrubart 53a12, 2sg. mararubart
112b5, 3sg. arrubart 21a11, 35b11, 42b6, 60a3, 61d11, ararubartatar 91b1,
97d2, 125d5, 131a11, 136b3, arrubartatar 100c25, -arrubartatar 33c14,
34c4, 123a4, verb. nec. erbirthi 98c10, part. dat. sg. erberdu 97d2, nom. pl.
airbirthi 94d1, VN (f part. i, part. ) airbert 59a10, erbert 47c4, 68b14,
69a23, acc. sg. aerbert 23b5, erbirt 46b28, 56a13, dat. sg. erbeirt 20d1,
29b1, airbirt 31b24, eirbirt 56a19, erbirt 68c14, 128c2, gen. sg. / nom. pl.
erbertae 101d2, dat. pl. erbertaib 38d17; asbeir (says, speaks):
Wrzburg: pres. ind. 1sg. rel. asbiur 2d12, 8a9, 12d6, 14c30, with emph.
pron. 1sg. rel.asbiursa 10a4, 12c29, with ci, ciasbiursa 2b18, with rel. n
asm.biur 9b25, and infix. pron. 3sg. n. and emph. pron. 1sg. asindbiursa
13a25, (with elision of infix. pron.) asinbiursa 13a29, with ci and infix.
pron. 3sg. n. and emph. pron. 1sg. ciasidbiursa 3a2, with infix. pron. 3sg. n.
atbiur 9b20, with dem. rel. anasbiur 17d12, 19c1, and emph. pron. 1sg.
anasbiursa 5a10, 24a19, niepur 8d26, 24a38, ni epur 9b19, arniepur 16d7,
prep. and rel. and infix. pron. 3sg. n. indid epiur 4b26, arind. epur 5a31,
2sg. rel. with emph. pron. 2sg. asbirsiu 12d17, 3sg. asbeir 4d23, 12c22, rel.
28b18, 28d2, asbeir 17a12, 13, rel. 13c13, with emph. pron. 3sg. m.
asbeirsom 18d1, asbeirsom 4b3, 5a1, 12d21, 15a16, 31a7, asbeir som 4d17,
5a16, asbeirsom 11c10, (as)beir (som) 27b16, with rel. n asmbeir 3d10,
28b20, and emph. pron. 3sg. m. asmbeirsom 8d14, asm.beirsom 13a13,
(as)m.beir som 1a9, with infix. pron. 3sg. n. atbeir 10d5, adbeir (= ad-dbeir Thes. Pal. I, 525 note b) 5a11, rel. and with emph. pron. 3sg. m.
atbeirsom 27c18, rel. with infix. pron. 3sg. n. assidbeir 20a10, with dem.
rel. anasbeir 6d15, asber 21d11, rel. with emph. pron. 3sg. m. asbersom
23d23, asbersom 23d26, with rel. n asm.ber 10b13, (with elision of infix.
pron. 3sg. n.) cani epir 10d5, with emph. pron. 3sg. m. niepir som 25d4,
niepeirsom 17d11, and with cid and prep. and rel. ciddianepirsom 19d11,
283

rel. eiper 33b7, 2pl. with emph. pron. 2pl. asberidsi 5a31, with dem. rel.
anasberid 5a8, 3pl. asberat 30c12, rel. 14a36, 29b12, asberat 17c6, with rel.
n asmberat 24a28, and infix. pron. 2pl. asndonberat 2a12, with dem. rel.
anasberat 25c17, 28a2, imperf. ind. 3sg. with dem. rel. anasbered 15a18,
1pl. with infix. pron. 3sg. n. atbeirmis 26b19, 3pl. rel. asbeirtis 4c23,
asbertis 25b27, fut. 1sg. rel. asber 32a22, ni per 32a20, 3sg. asbera
9a17, 26a6, rel. asbra 12d6, 1pl. rel. asbram 17c23, 3pl. rel. asbrat
12d36, with dem. rel. anasbrat 12d12, perf. 3sg. rel. asrubart 4d16, 7c8,
18, 9c1, 10d2, 11b5, 25c20, 26a3, 27d27, 32c13, 32d2, asrbart side 10a26,
with dem. rel. anasrubart 13d23, with prep. and rel. dianrbart 13c12, 1pl.
rel. asrubartmar 8d26, 3pl. rel. asrubartatar 18d1, pres. subj. 1sg. with cia
and emph. pron. 1sg. ciasbersa 16a24, rel. coniepersa 32a20, 2sg. rel. with
emph. pron. 2sg. asbere siu 12d13, with ci and infix. pron. 3sg. n. ciatbere
5a28, with dem. rel. anasbere 30b10, 3sg. with ci ciasbera 22b23, and infix.
pron. 3sg. n. catbera 12a4, with dem. rel. anasbera 12c32, 1pl. with rel. n
and emph. pron. 1pl. asmberamni 2a13, 2pl. with dem. rel. anasberaid
7d10, anasberith 12d3, perf. nerbarid 13c13, rel. with emph. pron. 2pl.
rbaridsi 15d6, 3pl. with ci ciasberat 1d1, 26a1 27a27, with dem. rel.
anasberat 28c1, rel. arnaepret 7d8, 30b17, perf. arnarbarat 27c8, past
subj. 1sg. with dem. rel. anasberinn 17b1, anasberin 14c23, 3sg. with ma
ma asberad 13a13, rel. asberad 13a19, 1pl. with dem. rel. anasbermis
18a18; impv. 3sg. naepred 10a16. pass.: pres. ind. 3sg. asberar 9d5, 3c21,
6b7, 9b7, 22a6, rel. asberar 14c21, 16a17, with rel. n asm.berar 9d5, 10b21
(bis), with dem. rel. anasberar 23c11, 27c14, asberr 21c7, with rel. n
asmberr 33a18, with prep. and rel. dianeperr 21d4, 3pl. asbertar 28a20,
30a3, perf. 3sg. rel. asrobrad 3c31, 5a4, 10d8, 12d26, with dem. rel.
anasrobrad 33b16, nierbrad 32c8, pres. subj. 3sg. with ci ciasberthar
10a12, with dem. rel. anasberthar 5d14, rel. araneperthar (MS aranepertar)
5a5, perf. arn rbarthar 10d13, arnarbarthar 31c7, 33b16, arna
rbarthar 29b2, past subj. 3sg. rel. with ci ciasberthe 15d20, cini eperthe
19a5, rel. arnaeperthe 23d25; Milan: pres. ind. 3sg. (rel.) asbeir 17b9, 17c6,
31b17, 45b9, 45d8, 46a12, 48b9, 50c5, 50d1, 51c3, 51d25, 53c13, 54a7,
54d4, 58b5, 61a23, 62a21, 62d13, 65b5, 83b13, 89d6, 90d19, 91a14, 91d7,
94a11, 94b15, 94c5, 96b18, 100c2, 102b1, 102b5, 126c1, 139a6, (rel.) asber
284

22a8, 24d17, 27b9, 37b19, 40a15, 45b3-4, 46d3, 50d2, 51d25, 55d25,
56b33, 60a12, 61a23, 67c2, 67c2, 67d8, 74d7, 74d9, 89a8, 94a13, 101c13,
108c09, 114a2-3, 120d15, 126c10, 127d14 (x2), 145d4 with infix. 3sg. n
asindber 25d12, asinbeir 27d22, asmbeir 25d18, 30b2, 34d5, 46d10, 53c11,
55c20, 55d11, 58a11 (2x), 67b2, 89a8, 102a15, 102c6, 129a2a, 130a6,
145d4, with infix. atbeir 46a11, assidbeir 130c3, with infix. 3sg. atbeir
56b3, atber 124b3, asmber 51b10, 56b3, 77d11, 107a11, n eper 69a21, 1pl.
(rel.) asberam 25d11, 25d12, 26a8, 34b5, 37b16, 3pl. (rel.) asberat 17b4,
17c7, 20c5, 21c11, 24d23, 24d25, 24d29, 25b5, 26c1, 34d6, 44c9, 56b15,
93d14, 132a1 (x2), atberat 44c19, pass. sg. asberr 37a18, 110d16, 127d2,
127d9, 127d15, 128d14, asber 53c14, asberar 76a5, asmberar 30c3, nieperr
145c3, neg. neperr 76a6, coneperr 2b17, 21c3, 133a10, dianaiperr 14d13,
ineperr 24d14, -eperr 32c15, pl. asrobarad (this form is given in DIL as
asrobar preceded by a question mark) 17b23, asbertar 27b19, 123c8 rel.
asmbertar 23a12, impf. 1sg. asberin 54c26, asberinn 91b10, 3sg. (rel.)
asberad 32d5, 51b15, 53d6, 95b7, asbered 54c18, 54c23, 55c17, rel. with
infix. 3sg. atberad 62c2, 3pl. asbertis 62c13, 95a5, 106d3, pres. subj. 3sg.
(rel.) asbera 31d9, -airpred 33c17, past subj. 3sg. as-berad 59a7, asberad
67a8, 129b12, conepred 28b11, 36a32, 69a21, -rbarad 31b20, 31b24,
-epreid 70a6, 3pl. -epertais 28d8, conepertis 136b4, rel. -asndarobartis
31b22, pass. sg. asberthae 77a6, eperthe 35a8, coneperthae 65d13, fut.
1sg. asbr 91b10, 1pl. (rel.) asberam 14a9, 94a6, 131d1, 3sg. asbera
70c12, sec. fut. pass. nieperthae 127d14, pret. 3sg. asbert 16c10, with nas
asmbert 58c6, 124d9, 3pl. rel. withinf 3sg. asindbertatar 124d9, perf. 1sg.
(rel.) asruburt 50d7, -rburt 44b19, with rel. infix. pron. asidrubart 3a15,
with infix. 2sg. ciasidruburt 66c1, 3sg. (rel.) asrubart 15d2, 21d4, 32d5,
35a7, 35c26, 37b28, 43d1, 45a8, 48c10, 50b8, 51a19, 51d2, 53a17, 53b26,
53c16, 55c3, 62c1 (2x), 64d8, 65c6, 65c11, 66d15, 67b3, 73d4, 83b13,
89b2, 92a12, 94a13, 100c13, 101c4, 102d3, 112d2 (2x), 125c2, 131c14,
asrubbart 59a7, rel. with infix. 3sg. asinrobrad 50b8, 1pl. (rel.)
asrubartmar 43b8, 37a14 (2x), 54a32, 111c9, 136c11, 3pl. asrubartatar
16c5, 24d4, 97d4, asrobartatar 20b2, withinf asinrubartatar 131d12,
-rbartatar 29a4, pass. sg. (rel.) asrobrad 16a14, 31b24, 32c15, 33d12,
37b24 (2x), 45a9, 72b4a, verb. nec. eperthi 46a11, VN (f ) epert 15a2,
285

23a13, 31b12, 57c4, 72b3, etbert (Note Thes.: usually epert) 83a5, acc. sg.
epert 30c3, 31b24, 51a19, epirt 31b17, epirt 51d13, gen. sg. epertae 59a18,
dat. sg. ebert 35c27, epert 36a32 (2x), ebirt 40c6, epirt 46a24, 48b16, acc.
pl. eperta 26a2; conbeir (conceives, begets): VN compert, gen. sg.
combartae 71c6; doaidbir (displays, shows): Wrzburg: pres. subj. 2pl.
nitidbarid 3b11; Milan: pres. ind. pass. sg. dunedbarar 123c4, nadtedbardar
19c18, pl. duaidbetar 30b2, past subj. 3sg. donedbarad 23a4, part. (io, i)
tedbarthae, tedbarthe 47a5, nom. sg. with bas bastidbarthae 76a16, dat. pl.
tedbartaib 45b6, verb. nec. tedbarthi 126d3; doairbir (bends, bends
down): Wrzburg: pass.: pres. ind. (or pres. subj.) 3sg. with infix. pron.
1sg. nimtharberar 9c31, 3pl. doairbertar 22c10, with rel. n donairber(ta)r
25c23, impv. (or pres. ind.) 3pl. tairbertar 25c23; Milan: pres. ind. pass. pl.
with nas dunerbarar 99d1, pres. subj. pass. sg. with infix. 1sg.
dummerberthar 49c3, pret. pass. pl. duarbartha 99a1, part. tairberthae
130c7, VN (f a) tairbert, dat. sg. therbirt 46d2; dobeir (gives):
Wrzburg: pres. ind. 1sg. dobiur 14a20, 16c4, with emph. pron. 1sg.
dobiursa 27d8, 3sg. dobeir 1c15, 4d25, 6a11, 14, 12a7, 15a8, 21a13, 29a23,
29b5, 33c3, dobeir 15a6, 16a18, dober 14b15, with emph. pron. 3sg. m.
dobeirsom 7d8, 12a6, 30a17, dobeirsom 16d14, with rel. n dombeir som
5b42, lase dombeir 7c4, amal dombeir 11c14, and with infix. pron. 3sg. n.
dondbeir 5c24, with dem. rel. andobeir 10d28, nitabir 14b15, 16a1, with
infix. pron. 3pl. nstabir 16a30, 1pl. doberam 13b15, 25d21, nitaibrem
15c9, with conjn. and neg. anataibrem 15d25, 2pl. doberid 24b19, with
infix. pron. 3sg. n. and emph. pron. 2pl. daberidsi 24c16, 3pl. doberat 2a23,
doberat 15b2, fut. 1sg. dober 12c35, 3sg. with rel. n dombra 25d19, with
infix. pron. 3pl. dosbra 13b30, 3pl. dobrat 6b27, sec. fut. 2pl. with infix.
pron. 3pl. dosbrthe 19d24, pres. subj. 2sg. nitaibre 29a22, 28, nitoibre
29a20, with prep. and rel. forataibre 29a23, 3sg. with infix. pron. 3sg. n.
manithobrea 4c20, 2pl. with rel. n and emph. pron. 2pl. dom beraidsi 8d21,
with elision of infix. pron. 3sg. n. donberaidsi 16c9, nitaibrid 6c1, with
conjn. and elision of infix. pron. 3sg. n. arintaibrid 7b3, past subj. 3sg. rel.
doberad 18a3, 32a12, 2pl. ciado berthe 15d8a, impv. 3sg. taibred 5d16,
22b7, 27c28, nataibred 6b11, 27b12, with elision of infix. pron. 3sg. n.
(leg.) nch thaibred or elision of infix. pron. 3sg. f. nacha taibred (MS nch
286

taibred) 5d16, 1pl. taibrem 13d15, 2pl. taibrid 16c17, taibrith 6a20, pass.:
pres. ind. 3sg. doberr 3c1, 28c4, with prep. and rel. diatabarr 17a3,
aratobarr 12d29, 3pl. dobertar 10d20, imperf. ind. 3sg. doberthe 10d16,
fut. 3sg. dobrthar 10d21, 16a13, doberthar 27d13, with rel. n dombrthar
25d12, nitibrthar 18b11, pres. subj. 3sg. doberthar 8d18, 13a3, perfect.
forms from *to-rat-. perf. 1sg. doratus 8c6, 3sg. dorat 4b10, (do)rrat 14c38,
with infix. pron. 3sg. m. (reflex.) darrat fessin 28b4, dorat 4c35, 6d7,
31b19, with enclitic doratside 23c17, 1pl. doratsam 13d15, 2pl. with infix.
pron. 3sg. n. and emph. pron. 2pl. daratsidsi 24b21, 3pl. nitartsat 24b20, n
tartissed 1b17, with infix. pron. 3sg. n. and emph. pron. 3pl. nithartsatsom
24b20, perfect. pres. subj. 1sg. with conjn. and emph. pron. 1sg. aratartsa
14d17, 3sg. with conjn.s arna tarta 11b18, darta 28b31, 2pl. with emph.
pron. 2pl. ciadorattid si 16c11, pass.: perf. 3sg. doratad 21c3, rel. doratad
4a18, 19b15, 21c17, 22a19, doradad 23c16, with rel. n intain dorratad
28d9, (gem.) dorratad 19c6, 28b6, duratad 33b8, perfect. pres. subj. with
conjn.s aratartar 15a19, condartar4a18, 20c24, perfect. forms from *toucc-, perf. 1sg. with rel. n and emph. pron. 1sg. donuccussa 30a11, 3sg.
with rel. n donuic 10d37, with emph. pron. 3sg. m. tuicsom 21d1, ni tuicsom
23a10, nitucsam 29b14, with dem. rel. and encl. atucside 24b25, impv. 2sg.
natuic 10a30; pass.: pres. ind. 3sg. nitucthar 27d22, perf. 3sg. tuccad 24b26,
28a3, with prep. and rel. diatuiced 12b12, pres. subj. 3sg. with interrog. in
induccatar (d = nas. t) 9b19, Forms from do-ucci, understands pres. ind.
3sg. tuicci 12c26, 12d42, nitucci 12d6, with infix. pron. 3pl. dosnucci 22c1,
with infix. pron. 3sg. n. daucci 13a8, nthucci 12d11, nithucci 12c46, and
with encl. 3sg. m. dhucciside 12c21, 3pl. with infix. pron. 3sg. n. nthuccat
12c20, nithucat 15a34, and with rel. n amal donducet 8a14, fut. 3sg. tucfa
12d12, (tucfa) 15b1, nitucfa 12d3, with infix. pron. 3sg. n. nthucfa 13a7,
2pl. with infix. pron. 3sg. n. and emph. pron. 2pl. daucbaid si 21c12, perf.
2pl. with emph. pron. 2pl. nitucsidsi 12a3, 3pl. nitucsat 15a32, nadtuicset
15a29, with infix. pron. 3sg. n. and emph. pron. 3pl. nthucsatsom 8a10a,
with rel. n and infix. pron. 3sg. n. and encl. 3pl. amal (MS am) donducatsidi
21d7, pres. subj. 2sg. with conjn.s aratuicce 28d7, and with infix. pron. 3sg.
n. condidtucce 30a19, 3sg. aratucca 27b27, 1pl. with rel. n and infix. pron.
3sg. n. donduccam 24a22, 2pl. with conjn.s aratucid 32a3, conducaid (d =
287

nas. t) 21a8, and with infix. pron. 3sg. n. conidtuccid 21c11, past subj. 1sg.
with infix. pron. 3pl. nistuccin 12d25, 1pl. with conjn. and infix. pron. 3sg.
n. and emph. pron. 1pl. condidtucmisni 20d17, 3pl. with conjn. and infix.
pron. 3sg. n. condidtuctis 21c21, pass.: pres. ind. 3sg. ntucthar 12c46, 3pl.
ntucatar 12c43, fut. 3sg. ntuccfither 8a5; Milan: pres. ind. 1sg. dobiur
58d5, 2sg. rel. dubir 105d4, 111c13, airdubir 110d13, rel. dombir 87a7,
with infix. 1sg. dombir 45a14, 111c13, neg. ntabair 63c13, rel. dumbir
88a4, -tabair 91a17, 106c12, 3sg. (rel.) dobeir 51b8, 103c12, (rel.) dober
16b12, 46c17, 49b7, 51c2, 51d5, 51d10, 78b12, 90a7, 121c16, 126b4a,
127d10, (rel.) dubeir 65b10, 89d6, 90b13, 109a01 (2x), duber 77a2, 77b11,
77d3, 77d5, (rel.) with infix. 3sg. dodbeir 35c23, rel. domber 44b10-11,
101c6-7, 114d11, 126d4, dombeir 51c12, 48c34, 104c4, 107c16, 111d4,
116d3, rel. dumber 74d13, neg. nitabair 21b2, 51d2, nitaiber 51d5, rel. neg.
nad tabair 91a20, 93c10, (rel.) with infix. 3sg. dabeir 47a14, 51d10, with
infix. dosber 49b7, rel. dumbeir 62c5, 1pl. domberam 21a8, (rel.) doberam
37a10, -taibrem 37a10, 3pl. rel. duberat 69a11, 69a13, 85d5, -taibret 45a3,
with inf. 1sg. domberat 124c16, pass. sg. doberthar 19b11, doberr 51b23,
dober 56b11, with inf. 1sg. dombertar 45b5, (rel.) -tabarr 30a9, 35b16,
35b18, (rel.) duberr 42a15, duber (Note Thes.: rectius duberr) 77d3,
doberar 46b1, 60b19, 132a8, duberar 105d6, (rel.) -tabar 54a2, pl. tobertar
25b6, (rel.) dobertar 27b15, 37a12, 37a14, 55d19, with nas. dumbrtar,
69b1, rel. dubertar 81c4-6, neg. airnitabartar 101d6, impv. 2sg. tabair
27c12, 87d12, 93c15, 3sg. taibred 127b6, impf. 2sg. dubertha 130d15, 3sg.
(rel.) doberad 35c23, duberad 55d11, 118c6, ntaibred 118b6, 56a13, 3pl.
dobertis 54c30, dubertis 89a2, rel. dombeirtis 90d17, pass. sg. doberthe
83d9, 115a14, pl. dubertis 84c9, pres. subj. 2sg. doberae 106c1, rel.
domberae 94d4, with infix. 3sg. dundaberae 50c3, 3sg. dumbera 40c21,
-taibred 32d5, -taibrea 93a9, 3pl. duberat 73d12, dumberat 40c21, pass. sg.
with infix. domberthar 24d14, past subj. 3sg. doberad 35c26, -taibred
35c26, conitaibrea 109c14, with nas dumberad 40c17, pass. sg. tabarthe
36a1, -tabarthae 40d20, pl. dobertais 90a14, fut. 2sg. doberae 44a20, rel.
domberae 55b13, int intibr 77a17, 3sg. dobera 34d5, 90c14, rel. dombera
27a6, 30d25, 51b10, 57c4, rel. dumberae 100d5, pass. sg. doberthar 24c2,
28a10, 33a9, dobrthar 118d22, duberthar 28a10, 66b3, pl. rel. dubrthar
288

72d18, sec. fut. 3sg. domberad 108a5, neg. with infix. 3pl. nandatiberad
97d10, 3pl. with inf. 3pl. nondabrtais 124b6, pass. pl. nitibertais 15c7,
pret. 2sg. dobirt 56a13, 3sg. dobert 23b10, 58c4, rel. dombert 23b7,
Perfective forms: From *to-rat-: pres. ind. 3sg. -tarti 51b7, impf. 3sg.
dardad 77a7, 98b8, 3pl. neg n taircitis 98c2, rel. neg nadtardatis 85d5,
pres. subj. 2sg. diatar 89c5, 3sg. durata 57a13, 2pl. diatartaid 83b11, pass.
pl. arnachontartar 127a14, perf. 1sg. doratus 103a6, 2sg. doratais 43d18,
44a14, doratis 56a15, duratais 56b24, (rel.) duratis 80b2, 92c8, 3sg. (rel.)
dorat 22d19, 23c7, 25c11, 35d03, 37d3, 54c16, 118d19, 136c11, durat
40b8, 48a21, 91a21, 94c17, -tarat 36a1, 90c18, 91a21, 3pl. doratsat 115b8,
rel. duratsat 82c8, dorratsat 113d7, with infix. 3sg. daratsat 73b17, with
infix. 3pl. dosratsat 44a14, pass. sg. (rel.) doratad 24d31, 34a23, 37a16,
44b19, 44b29, 46b26, 47a18, 53d2, 138a6, duratad 102c7, 104b2, daratad
(Note Thes.: rectius doratad) 25a1, ntardad 63d5, rel. forsatardad 80d4,
rel. aratartad 90a7, pass. pl. dorata 54c17, 59a18, 82c8, 90c25, -tarta
40a13, From *to-ucc-: pres. ind. 3sg. tuic 30c5, duuic 40c22, pass. sg.
tuucthar 42c2, duucthar 51c24, impf. 3pl. tuctais 125d13, past subj. 3sg.
neg withinf 3pl. asnachatucad 125b7, perf. 1sg. neg. ntucus 91c1, 2sg.
tuicais 56a13, 3sg. donuic 16b12, dunuic 118b6, douic 18c5, duuc 136c1
(no glide), (rel.) duuic 25d18, 84c24, 111b15, duduccai (GOI 678) 27d23,
duic (this form not in DIL) 30b10, (rel.) tuic 48c19, 51c30, 62a2, 63c9,
67a8, ntuic 51d2, rel. with infix. 3sg. dudnuic 44d14, -tuc 40c17, with
infix. 3sg. dauc 38c1, 38c3, danuic 38c4, dauic 50b8, withinf 3pl. dodauic
131c14, rel. fuantuic 35a9, fuantuc 38c5, (rel.), rel. tunuic 51c26, neg n
tuicc 98c11, 3pl. with infix. 1pl. dunnucsat 92d1, pass. sg. rel. duucad
56c11, tuccad 71c9, Forms from do-ucci, understands: pres. ind. 3sg.
-tucai 42c7, with infix. 3pl. dusnucai 42c12, 3pl. -thuccai 42c8, citidtucat
125d4, pass. sg. duucthar 55a10, nadtuctar 112d7, pres. subj. pass. sg. rel.
dunucthar 79d2, past subj. pass. sg. conducthe 51d1, perf. 3pl. neg.
nadtucsat 75d10, ntucsat 75d10, verb. nec. thabarthi 22a4, VN tabairt
2b10, 42c13, 67c2, 76b7, tabart 35b17, 51d2, 54a4, acc. sg. tabairt 30a9,
30a10, 30a12, 34c11, 44d6, 54b22, 55a19, gen. sg. tabarthi 25b12, 112d1,
tabartae 73b8, dat. sg. tabairt 22c14, 23b12, 24c1, 26d14, 27b15, 42c33,
49b6, 53a23, 53d13, 54a4, 54b30, 57d16, 62a4, 72b25, 76d9, 103d16,
289

110c1; doopir (takes away, deprives): Wrzburg: pres. ind. 2pl. dioiprid
9c23, impv. 3sg. na tiubrad (t = dd, gem.) 9d20; Milan: pres. ind. pass. sg.
ndiubarar 112a4, part. dupart.hae, diupart.hae 68c8, dat. sg. diubarthu
54d18, VN dupart., dat. pl. diubartaib 89c14; forbeir (grows, increases):
Wrzburg: pres. ind. 3sg. with rel. n: amal formbeir et toirthigedar indibsi
as it (i.e. faith) increases and fructifies in you (gl. fructificat et crescit, sicut
in uobs) 26c3; Milan: pres. ind. 3pl. rel. formberat 71c1, impf. 1sg.
formberinn 88a10, 3sg. forberad 88a10, pres. subj. 3pl. -foirbret 27c6, perf.
3sg. forrubart 33c10, 1pl. forrubartmar 102a1, 3pl. forrubartatar 101a10,
103d6, VN forbart, dat. pl. forbartaib 89c13; fotabair (places under):
Milan: pres. ind. 3sg. futabair 40d2, impv. 2sg. fothabair 76d11, past subj.
3pl. rel. with infix. 1sg. fundamthabartisse 54c29, rel. with infix. 3pl.
fundatabartis 72a7; frisbeir (opposes, resists): Milan: pres. ind. 3pl. rel.
frismberat 14c21; fristabair (sets against, opposes), Milan: perf. 3sg.
fristarat 51d3; immbeir (carries around, moves, sets in action):
Wrzburg: pres. subj. 3sg. rel. immabera 131a2, pres. subj. 3pl. mitnimret
74b22, VN (f a) imbert 17c6, dat. sg. imbirt 32a8, 38a13, imbert 33d18, acc.
pl. imberta 81c4-6; immtabair (carries round, surrounds): Milan: pres.
subj. pass. sg. -immimthabarthar 41c2; intairbeir (bows down by force):
Wrzburg: impv. 2sg. intairbir 5b9; maipir (speaks badly): Milan: pres.
ind. 3sg. neg n maipir 56d16; remiepir (says beforehand): Wrzburg:
perf. 3sg. rel. tredeso 5c22; Milan: pret. 1sg. remirburt 23c24, 3sg.
remerbart 15b3, rel. remrbart 74d7, rel. remirbart 94b2, rel. remierbart
94c13, 1pl. remirbartmar 36c21, rel. remerbartmar 42c34, 73b2, perf.
1pl. remirbartmar 33a1, 97a7, merbartmar
56b26, part. (io, ia)
remeperthae, remeperthe 68d11 (note in Thes.: for the nominative cf. Sg.
188a 6, Laws IV 312 l. 1), acc./dat. sg. remeperthe 131c14, dat. sg.
remeperthiu 100b21, nom. pl. remeperthi 23d4, gen. pl. remeperthe 48a11,
dat. pl. remeperthib 23c27, Milan: part. (io, ia) remeperthae, acc. sg.
remeperthe 32d6, 126b14, acc. pl. remeperthi 27c15, meaning of part.:
aforesaid, above mentioned; tremibeir (transfers): Wrzburg: pass. pres.
ind. 3sg. tremiberar 8a5, Milan: pres. ind. pass. sg. trimiberar 31b22, perf.
pass. sg. (suppl.) trimirucad 2b17, trimiberar 21c3.
290

canaid sings, recites
PIE *kan- or *kan- (LIV2 342f., IEW 525f.)
CC *kan-e/ocanaid: Wrzburg: pres. ind. 3sg. rel.: acanas 27b27, Milan: pres.
ind. 3sg. rel. chanas 24d14, 3pl. rel. trissacantar 122c2, pass. sg.
cianudchanar 135a6, impf. pass. sg. nocainte 102c8, perf. 3sg. rocachain
48b11, pass. sg. ro-cet 2b6, rel. rochet 25b6, 35b19, 86d19a, roct 25b6,
25b8, 57d13, roichet 44b4, rel. (cf. McCone 1985: 96) rucht 64a13, pass.
pl. (rel.) roceta 30a9, nadrocheta 115b4, verb. nec. cheti 126c4, VN ctal
60a12, cetal 40d7, acc. sg. cetal 24d14, dat. sg. cetul 33b7, gen. pl. cetal
72a2; arcain (sings, recites):
Milan: pres. ind. 3sg. with infix. 3sg.
aracain 64c14, pass. pl. arcantar 66c7; concain (sings, utters in

harmony): Milan: pres. ind. 3pl. canat 32c17; doairchain (foretells,
prophesies): Wrzburg: pret. & perf. 3sg. tairchechuin 4c40, 4d8, 3pl.
nadtairchechnatar 5a1, pass.: pret. & perf. 3sg. doairchet 6b26, doarchet
4d4, 26a11, doarrchet 5a16, 6d8, with rel. n donairchet 13a36, donarchet
29c3, tarrchet 26d9, tairrchet 6d6, 7a2, with dem. rel. atairchet 15a34,
Milan: pres. ind. 3sg. andunerchain 15d10, doerchain 21a7, rel. duerchanar
111d4, 3pl. rel. donerchanat 19b11, impf. 3sg. rel. donerchanad 54c24,
pret. 3sg. rel. dunairchechainn 64c22, dunarchechainn 64c22, with infix.
3sg. dundarchechainn 66c12, pass. sg. (rel.) duaircet 24d5, rel. donairchet
35b9, perf. 3pl. (rel.) taircheta 38c9, part. terchantae, dat. sg. terchantu
24d6, VN (n o) tairchetal, terchital 64c22, gen. sg. tairchitil 89b10, dat. sg.
taircitul 54c30, acc. pl. tercitla 19b1l; doinchain (chants, utters): Milan:
pres. subj. 3pl. rel. duninchanat 76a18 ; forcain (teaches, instructs):
Wrzburg: pres. ind. 1sg. rel. forchanim 8c3, forchun 10a13, 3sg. with
emph. pron., 3sg. m. (for)cain som 27c8, rel. forchain 5d10, 8c2, with
emph. pron. 3sg. m. forchain som 27b27, with infix. pron. 1pl. fordoncain
31c16, ni forcain 11c19, 3pl. rel. forchanat 20c20, 28c16, 29a11, 12, 30c12,
31b17, fut. 3sg. with infix. pron. 2pl. fordub cechna 9a16, pres. subj. 2sg.
rel. forcane 28c11, 21, with infix. pron. 3sg. n. \didforcane 28d13, 3sg. rel.
forchana 31b10, 2pl. rel. forcanit (a prima manu; = forcanith) 12c18, impv.
3sg. forcanad 12d41, 22c8, 2pl. forcanid 22d1; pass. pres. ind. 3sg. with rel.
n and infix. pron. 2pl. forndobcanar 3b23, Milan: pres. ind. 3sg. (rel.)
291

forcain 128d9, 129c9, forchain 35d7, 67b3, 136c11, with infix. fordacain
30d12, 3pl. (rel.) forcanat 28a15, rel. forchanat 45b15-16, pass. sg.
forcanar 50d12, 82c8, pl. forcantar 34b12 impf. 1sg. forchanainn 54c31,
pres. subj. 1sg. -farc(a)n 68c15, 3pl. co farcanat (in DIL forcanat) 45b18,
past subj. 1sg. forcanainn 94b13, fut. 2sg. forcechnae 114b11, perf. 1sg.
forroichansa 17d1, 2sg. fortanroichan 22c3, 3sg. forrochain 68b8, 3pl. with
infix. 1pl. fortanroichechnatar 63b1, part. foircthe 23b12, 35d6, 111c19
nom. pl. foircthi 16b4, 132a4, VN (n o) forcetal, forcital 70d7, gen. sg.
forcitil 19b17, dat. sg. forcitul 35b12, 42d5, 61d3, forcital 49a6, gen. pl.
forcital 35b15, 47d11, dat. pl. foircthib 68c14; *remicain (sings
beforehand (prophetically)): Milan: pres. ind. pass. sg. remcanar 67b11.
ceilid hides, conceals (prim B1)

PIE 1. *kel- (LIV2 323f., IEW 553f.)
CC *kel-e/oceilid: Wrzburg: pres. ind. 1sg. with emph. pron. 3sg. m. nceilsom
4d16, with infix. pron. 3sg. n. nicheil 5b5, Milan: impf. 3pl. with infix.
nudachiltis 61a2, pres. subj. 3pl. with infix. 3pl. nudachelat 54c9, VN
cleth, dat. sg. chlithon 87d8, chlith 100a7, acc. pl. cletha 59a9; conceil
(hides, conceals): Wrzburg: pres. subj. 3sg. with infix. pron. 3sg. n
actmachotchela, Milan: pres. subj. 1sg. concela 101a1, 2sg. rel. chonchelae
(Note Thes.: chon over the line: rectius choncelae; if a relative form comes
after cia, chon-chelae may be sound, cf. Celt. Zeitschr. IV. 61) 106c14,
past subj. 3sg. rel. chelad 114c3 perf. 3sg. with inf.? conaicelt 49c9;
doaircheil (restricts, withholds): Wrzburg: pres. subj. 3sg. aratairchela
9d1; doceil (conceals, denies): Milan: pres. ind. 2sg. nadndichil 56c11,
pass. sg. (rel.) duceltar 111b11, VN (f ) dchleth, acc. sg. dichlid (this form
not in DIL) 59a9.
cingid steps, paces

PIE *(s)keng- or *(s)ken- (LIV2 555 IEW 930)
CC *keng-e/o
Milan: VN (n n) cim, cimmim 41d7 (this form not in DIL), acc. sg.
cim 19a3, nom./acc. pl. ceimmen 22a7.
292
cladaid digs, excavates

PIE **kelh2d(h)- (*kelh2- LIV2 350, IEW 546f.)
CC *klad-e/o
Wrzburg: perf. 3pl. rocechladatar 5a24, Milan: pres. ind. 3pl.
cladait 76d14, part. claisse, gen. sg. claissi 24c21; adclaid (hunts for):
Milan: pres. ind. 3pl. anadcladat 112b2, doclaid (digs up, uproots):
Milan: pres. ind. 3sg. with infix. 3sg. dusclaid 102a15; foclaid (digs
under, digs up): Milan: perf. 3sg. foroichlait 24c18.
clichid stirs?
PIE ?, CC *klek()CC *klik()-e/oarclich (wards off): Milan: pres. subj. 2pl. rel. araclessid 22d18.
conicc can, is able

PIE *h2nek-, *nek- (IEW 316-8)
CC *-an-n-k-e/oadcumaing (strikes, happens): Wrzburg: pret. 3sg. with infix.
pron. 2sg. attotchomnicc 6b13, 3pl. adcomcisset (Cf. McCone 1985: 93)
4d13, pres. subj. 3sg. rel. with con conecmi 5b35, Milan: pres. ind. 3sg.
adcumaing 19c10-11, with infix. pron. atchomaing 24c16, pres. subj. 3sg.
nadnecmai 22c8, 121c13, 122b5; aricc (comes upon, meets): Wrzburg:
pres. ind. 3sg. n airicc 9b24, n airic 19b2, Milan: pres. ind. pass. pl.
aranecatar 19d21, pres. subj. 3sg. -ar 30d24, -airi 30d24, pass. sg. -aristar
30d23, perf. 3sg. nadnairnettar 108c08; conicc (can, is able): Wrzburg:
pres. ind. 1sg. conicimm 20d6, 3sg. conicc 4b11, 6a17, with infix. pron. 3sg.
n. cotdicc 5b40, nicumaing 11b9, ni cumuing 9d16, with anaph. pron.
nicumuingside 9d18, 1pl. with infix. pron. 3sg. n. and emph. pron. 1pl.
nchumcamni 4a27, fut. 1sg. conicub 32a20, 2pl. with emph. pron. 2pl.
conicfidsi 13a10, conicfedsi 13a13, sec. fut. 1pl. icfimmis 17a10, pret. and
perf. 1sg. with infix. pron. 3sg. n. and emph. pron. 1sg. cotaneccarsa 14c40,
3sg. with infix. pron. 3sg. n. and emph. pron. 3sg. m. cotnic som 8a14, ni
choimnucuir 19c10, 1pl. with emph. pron. 1pl. n comtachtmarni 24b20,
293

2pl. rel. nd cimnacaid 9b10, 3pl. rel. nad comnactar 8a14, pres. subj. 2sg.
cechonis 10a21, past subj. 1sg. ni cuimsin 17b1, 1pl. with infix. pron. 3sg.
n. machotsmis 25a1, 3pl. nicuimsimmis 17c18, Milan: pres. ind. 1sg.
-cumcu 18b5, 3sg. nic 24d14, with infix. 3sg. ni chumgat 92c14, 3pl.
-cumcat 23a5, 41d15, 57c5, 60b6, -cumgat 54b12, 94b3, 107a8, with infix.
n chumgat 56c7, connacumgat 76a19, with infix. 3sg. nand chumgat 94b3,
annadcumget 112b16, rel. nadchumgat 121c22, impf. 3pl. rel. nacumgaitis
102a9, pres. subj. 3sg. -cumai 31c9, 32d15b, ndchum 87d13, 3pl.
nacumset 39c26, past subj. 3sg. -cuimsed 33b21, fut. 1sg. -cumgub 49c5,
3pl. -cumgubat 54a19, sec. fut. 3sg. icfed 14a6, -cumgaibed 42c32, pret.
3sg. conicoimnacuir 116c5, conanacuir 119d7, rel. nad coimnacuir 97d4,
97d5, 97d10, 1pl.-coimnacmar 53d9, 135d4, 3pl. nichoimnactar 19c5,
135d4, rel. nadchoimnactar 66d6, perf. 3pl. nad coimnactar 76a5, n
comnactar 76a7, VN cumang 60d3; conricc (meets, encounters):
Wrzburg: pres. ind. 3sg. ricc 7d9, fut. 3sg. rel. frissacomricfea 12c35,
pres. subj. 3sg. rel. (conco)muir 24a17, 1pl. rel. comairsem 33a9, Milan:
pres. ind. 3sg. rel. with fris frissacomraic 48d1-2, 69a18, 3pl. rel. with fris
frisacomracat 69a19, rel. with fris frisa comrucat 69a23, past subj. 3pl.
cocomairsitis 119d12, perf. 3sg. conranaic 52 line 17, past subj. 3sg. with
infix. 3sg. comairsed 39c15, VN (o) comrac, gen. sg. chomraic 75a1;
contecmaing (no translation in DIL): Milan: past subj. 3pl. chuntecmaistis
102a24; doairicc (comes): Wrzburg: pret. & perf. 3pl. donarnactar
7b13 ; doecmaing (hits, strikes): Wrzburg: pret. & perf. 3sg.
teccomnocuir 10a4, past subj. 3sg. doecmoised 5d26, Milan: pres. ind. 3sg.
rel. dondecmaing 54a7, 3pl. rel. doecmungat 63c8, pres. subj. 3sg.
nadnecmai 15d5, VN tecmang 35d22, tecmaing 54a31; dofuiric (finds,
comes upon): Wrzburg: fut. 3sg. dofuircifea 25b16; doicc (comes to,
approaches): Wrzburg: pres. ind. 2sg. with conjn. and infix. pron. 3sg. n.
condid ticci 24b5, fut. 3sg. with rel. n and infix. pron. 3sg. n. intain
dondiccfa 5c5, (with elision of rel. n) dodiccfa 29c4, with infix. pron. 3pl.
dosnicfa 5c5, with conjn. articfea 4d6, sec. fut. 3sg. with rel. n donicfad
21a3, pret. & perf. 3sg. tnicc 30d11, tanicc 3a7, nuie tanicc 7c7, with
anaph. pron. 3sg. m. taniccside 4d26, with conjn. tanic 3a1, tanicc 29b2,
danicc (d = nas. t) 3c27, and with infix. pron. 3sg. n. didtanicc 12b34a,
294

with infix. pron. 2pl. dubtanicc 5c10, with infix. pron. 1sg. domanicc 12c9,
with infix. pron. 1pl. nntnicc 1d1, and with infix. pron. 1pl. only hre
dunnnic 25a21, pres. subj. 3sg. with conjn.s cot 8d25, 26a5 (bis), t
26a11, 3pl. with infix. pron. 3pl. arnachatsat 17d10, past subj. 3sg. with
conjn.s resu tised 27c8, dsed (d = nas. t) 25a6, with infix. pron. 3sg. n.
condidtsed 21a1, with infix. pron. 3sg. n. manithised 4c15, (with only len.
remaining) nachthised 26a12, Milan: pres. ind. 3sg. withinf 3pl. doda-ic
123d3, with nas dunnic 69a20, impf. 3sg. -ticed 41d9, past subj. 3sg. -tissed
(this form not in DIL) 27c4, with infix. 3pl. dodassed 39c15, fut. 3pl.
dundicfet 19b11, sec fut. 3sg. neg nad ticfed 46a19, with infix. 2pl.
nachabticfed 46a14, 3pl. rel. dundicfitis 68a1, pret. 3sg. tanaic 35d1, 82d9,
with infix. 3pl. dudanac 123c3, pret./perf. 3sg. with infix. 1pl. nntanic
14d04, 37a15, VN (f n) tchtu, acc. sg. tchtain 34d12, tchtin 60a12, dat. sg.
tichtin 39d3, dcthin (Note Griffith: leg. dchtin) 90c22; forcumaing
(happens, comes into being): Wrzburg: pret. 3sg. forcomnucuir 22b8,
forcomnucuir 22b8, farcomnucuir 19c3, rel. forcomnucuir 28c14,
forchomnucuir 3d25, forchomnucuir 11c15, rel. hiforcomnucuir 28b6, past.
subj. 3sg. forcuimsed 4d8, Milan: pret. (GOI 688) 3sg. forcomnacair
131c14, forcomnacuir 16c5, 3pl. forcomnactar 97a5, 145d8, rel.
forchomnacuir 16c5, 67c18, forchumnaccuir 113d3, part. forcmachtae,
nom. pl. forcmachti 34a10; foricc (comes across, finds): Wrzburg: pret.
3sg. with rel. n hore fonairnicc 16b2, nifarnic 2a21; immairicc (concerns,
suits): Milan: pres. ind. 3sg. (rel.) immeairic 35a7, 73a11, 74b1, 74d13,
immeairc 72d9, 74d13, 103b7, immairc 74d13, 119d5, commaairic 133c4,
imminndairc 135a7, 3pl. rel. with infix. 3sg. immidaircet 2b5, 2d4,
immaircet 17b20, rel. with infix. 3sg. masc immindaircet 25c15, pres. subj.
3sg. rel. with infix. immandairi 14d16, rel. immenairi 27b12, rel.
connamanairi 54b10, pret. 3sg. immearnaic 24d5, 3pl. maimmidarnactar
17b19; immricc (happens, befalls): Wrzburg: (with petrified infix. pron.
3sg. n.) Sec. fut. 3sg. nmariccfed 4b13, pres. subj. 3sg. rel. commar 5d37;
roicc (comes, reaches): Wrzburg: pres. ind. 1sg. rel. roiccu 23d9, rel.
an ricu 32a4, niriccim 12a33, 2sg. corricci 9a9, conricci 9a9, ricci 31b13,
rici 33b18, 3sg. rel. ndricc 12a34, 1pl. ni recam 15d1, fut. 1sg. with rel. n
ronicub 18b14, with emph. pron. 1sg. riccubsa 9a19, mos riccubsa 28c9,
295

rel. ricubsa 32a27, 3sg. with rel. n ronicfea 14a25, pret. 3sg. with infix.
pron. 2pl. and emph. pron. 2pl. rob nicsi 16d10, 3pl. with emph. pron. 3pl.
nirncatar som 6c31, pres. subj. 1sg. with emph. pron. 1sg. diarrsa (s = ss)
9a20, resu rssa 14a17, 3sg. rel. roh 20c11, sechihed ri 7b3, conr 24a17,
1pl. act rsam 12c9, 2pl. rel. risid 24b2, 3pl. rsat 5b39, past subj. 1sg. with
emph. pron. 1sg. nirisinse 9a19, resu rsinse 18a23; pass.: pres. ind. 3sg.
nirecar 12c3, ni reccar 15a5, fut. 3sg. niricfider 13b28, Milan: pres. ind.
1sg. ruicim (DIL rucim) 22c14, 2sg. darici 54c34, 3sg. -ric 36c10, 3pl.
nrecat 3a14, neg nadrucat 87c3, rel. ronecat 58d5, pass. sg. rosegar 14c19,
rel. ronecar 87d14, pres. subj. 1sg. muriissi 34a4 3sg. r 22c1, arnachrsat
112b20, past subj. 1sg. arindrisinn 92a5, pret. 3sg. conricht 2a6, 3pl.
-rancatar 35b25, 90d20, neg rel. nadrancatar 97d7, VN (f n) rchtu, dat. sg.
rchtin 105a8.
conrig binds together, ties

PIE *rei g- or *rei - (LIV2 503, IEW 861f.)
CC 2. *-rig-e/oconrig (binds together, ties): Wrzburg: perfect. pres. ind. 3sg.
with infix. pron. 2pl. cotobrrig 9b19, Milan: pres. ind. 2sg. conrigi 119b8,
3sg. conrig 23c12, pres. subj. 1sg. condarias 21b7-8, fut. 2sg. riris 134d3,
pass. withinf 1pl. cotanrirastar 134a1, perf. 3sg. (with infix.) idnarraig
15c1, perf. pass. sg. conarracht 123b2, verb. nec. cuimrechtai, cuimrechta
(recte cuimrechtae) 46a27, cuimrechti 137c12; arfuirig (holds back,
restrains): Wrzburg: pres. ind. 3sg. with infix. pron. 3sg. n. aridfuirig
26a11, Milan: past subj. pass. sg. arfuirestae 47b6, fut. 2sg. -arafoiris
37c18; dorig (strips, lays bare): Milan: pres. ind. 2sg. durigisiu 108d08,
3sg. durig 28a19, impv. pass. pl. drgetar 136a2, pres. subj. 3sg. withinf
2sg. dundatre 133a14, perf. 3sg. docoimarraig 48d15, rel. dochoimarraig
14b1, 155b1, 3pl. ducoimrachtar 100c27, part. (io, ia) drechtae, direchtae
91a4, nom. pl. direchtai 48d12, VN (n o) drech, dat. sg. diriug 28c11;
forig (holds back, detains): Wrzburg: pres. ind. 3sg. rel. with infix. pron.
3sg. m. fodrig 26a12, Milan: fut. 3pl. with infix. 1sg. arndamfuirset 114c11.
296

contuit falls together
PIE 1. *(s)ted- (LIV2 601, IEW 1033f.)
CC *-tud-e/ocontuit: Milan: pres. subj. 3pl. contotsat 16a19, perf. 3pl.
contorchratar 48c28, VN cutaim, cutuim 91c19; dotuit (falls):
Wrzburg: pres. ind. 3sg. rel. dothuit 4d15, with prep. and rel. and emph.
pron. 3sg. m. foratuitsom 4d15, perfect. pres. subj. 1pl. with conjn.
natorthissem 32c16, past subj. 3pl. with conjn. condodsitis 5b11, Milan:
impf. 1sg. dotodsinn 131b7, pres. subj. 1sg. dorothuusa 23c23, 3pl. dotodsat
124d12, arnatodsat 118a12, perf. 3sg. -torchar 34c14, 3pl. dorochratar
36d13, durochratar 91c18, VN (n n) tuitim, gen. pl. totman 19d4, adothaim
131b4.
dligid is entitled to, has a right to

PIE *dhlegh- or *dhleh- (IEW 271f.)
CC *dlig-e/oMilan: pres. ind. pass. sg. dlegair 55a7, 62c6b, 90b13, airnidlegar
90d11, perf. pass. sg. rodligestar (Cf. McCone 1985: 93, 95) 36a29.
dobruinn flows, trickles

PIE ?*bhrend(h)- (LIV2 95, IEW 167f.)
CC *bran-n-d-e/o- or *brend-e/odobruinn: Milan: pres. ind. 3sg. dubrinn 81c14; doeprinn
(flows, trickles): Milan: pres. subj. 3pl. doeprannat 39d2, VN (f n)
teirpersiu, gen. sg. tepairsnea 123d1, nom. pl. tepairsnea 129d12, acc. pl.
tipirsnea 81c9.
doessim sheds, pours

PIE *semH- (LIV2 531, IEW 901f.)
CC *-sem-e/o- and *-sem-de/odoessim: Wrzburg: pres. ind. 3pl. rel. doesmet 31b21, pass. perf.
3sg. rel. doresset 20d13, Milan: pres. ind. pass. sg. duesemar 56a13, rel.
dunesmar 44d1, rel. donesemar 56a13, pass. 3pl. forsateismetar 141b2,
impv. 2sg. tessim 134a6, pres. subj. pass. sg. rel. donesmider 37c11, pass.
297

pl. rel. dunesmiter 54b6, past subj. pass. sg. with infix. 1sg. dummesmide
44c11a, part. (io, ia) teiste, rtheste 58d10, gen. sg. testi 43b4, acc. pl. testiu
55a16, gen. pl. teste 77a18, VN (f n) teistiu, acc. sg. testin 37c6, gen. sg.
testiu 22b1, dat. sg. teste 30b7, thestin 37c4; dofuissim (pours out, pours
forth): Wrzburg: pass. fut. 3sg. dofuismthar 4c7, Milan: pres. ind. 3sg.
dofuissim 111a13, pass. sg. dofuissemar 44a11, perf. 3 sg. duforsaisset
110a8 (Note Thes.: leg. duforsaisset (cf. rosaiset, Ml. 2ab), the glossator
mistaking
creauere
for
creuere?
Windisch,
leg.
dofors[attar
.i.
imforlang]isset? Thurneysen, but cf. McCone 1985: 93), (rel.) doforsat

17b2, 17b6, rel. with infix. dudaforsat 94b7, pass. pl. rel. aratorsata 42b13,
rel. hitorsata 99a8, part. (io, ia) tuiste, nom. pl. tuisti 40a8-9, VN (f n) 2.
tuistiu, gen. sg. tuisten 42c16, tusten 85b14, dat. sg. tuistin 19d1, 74a11,
85b16, 99a8, *fotessim (no transl. in DIL): Milan: past subj. pass. sg. neg
arna fothismide 85d8.
dofil is coming, is at hand

PIE 3. *el- (LIV2 675, IEW 1136f.)
CC *el-e/oWrzburg: pres. ind. 3sg. dofeil 26a1, dofil 4d29.
doformaig increases, amplifies

PIE *meh2k- (IEW 709)
CC *mak-e/o
Wrzburg: pres. ind. 3sg. ardofrmaich 1b5, Milan: pres. ind. 3sg.
doformaig 55c20, impv. 2sg. trmaig 105d4, pres. subj. pass. sg.
diatormastar 20a19, cenitormastar 20a20, fut. pass. sg. dufrmaster (DIL
gives this as dufrmastar) 105a8, sec. fut. 3sg. dofoirmsed 35a17, perf. pass.
sg. durormacht 90d1, verb. nec. tormachtai 88a14, VN tormach 2a15 (2x),
97c2, acc. sg. tormach 83b2, dat. sg. thormuch 16c10, 51d10, 54a14, 60a1,
61c3, 76a5, 94c5, 96b7, trmuch 43d27, tormuch 44d18.
doim(m)na commits, entrusts

PIE *ne- (LIV2 455f., IEW 767)
CC ?*-ne-e/o298

Milan: perf. pass. sg. doimmarnad 34a6, VN (n io) timne, acc. sg.
atimnae 115c2, acc. pl. thimnae 74d5, gen. pl. timnae 46c4, 51b20.
doinfet blows, breathes

PIE **sei sd- (IEW 1040f.)
CC *sisd-e/odoinfet: Wrzburg: pres. ind. 3sg. with rel. n doninfet 4b3, with
emph. pron. 3sg. m. tinfetsom 4b4, 1pl. with rel. n hore doninfedam 14d32,
pres. subj. 3sg. with infix. pron. 3sg. n. mani thinib 4a27; pass.: perf. 3sg.
dorinfess 30d1, Milan: pres. ind. 3sg. withnas dundinfet 41d17, perf. 3sg.
rel. durinfid 96c4; fristinfet (blows against): Milan: pres. ind. 3sg.
fristinfet 28b8.
doseinn pursues, drives

PIE *senh2- (LIV2 532f., IEW 906)
CC *san-naMilan: pres. ind. 3pl. with infix. 1sg. dumsennat 39c28, impv. 3sg.
toibned 44a13, impf. 3sg. (rel.) duseinned 41d10, pass. sg. dusente 142b4,
pres. subj. 1sg. dussa 61c16, past subj. 1sg. dusesinn 41c5, pret. 3sg. rel.
with infix. 3pl. andundasepfainn 36d17.
feidid brings, leads)

PIE edh- and *ed- (LIV2 659, 658f., IEW 1115-7, IEW 78-80)
CC *ed-e/ofeidid: Wrzburg: pres. ind. 1pl. rel. afedme 15b26, perf. 3pl. rel.
rofadatar (leg. rofdatar) 29c13, Milan: pres. ind. pass. sg. rel. nadfedar55
39d7, impf. 3pl. nofeidtis 54d12, VN (f a) fedan, gen. sg. fednae 76a9;
dodat (leads down, leads): Milan: pres. subj. 3sg. -dud 35c30, past subj.
3sg. rel. with infix. 1sg. dumdsed 78b18, pass. sg. dundiastae 45c4, perf.
3sg. with infix. 3pl. dusderaid 99b13, VN tuididen, tudidin 137c6, acc. sg.
55
Reading following DIL; Thes has this as nadfendar with note MS. nadfedar: corr.
Thurneysen, here f is for bh, and translates it as they are not stricken. It is not clear where
Thurneysen emends this, but he seems to take it as from benaid rather than feidid. There is
no reason for this emendation.
299

tudidin 96d1, gen. sg. tuididne 93b3, dat. sg. tudidin 50a1, tuididin 86a1,
tuididen 92c5; dodchid (leads down, leads): Milan: fut. pass. sg.
dudichestar 30d25, part. tuidchisse 72d1; dofed (brings, leads): Milan:
impf. 3sg. with infix. 1pl. dusfeded 97c5, with infix. 3pl. dudafeded 100b17,
past subj. 3sg. with infix. 3sg. dudfessed 78b14, perf. 3pl. with infix. 3pl.
dudaruid 63b12; forfeid (carries over): Milan: pres. pass. forfedar
129b14; *immfed- (brings around, carries around, found in DIL under
pres. ind. 3pl. imm-fedat): Milan: pres. ind. 3pl. rel. immefedat 47d7.
fethid watches, observes

PIE *et- (LIV2 694, IEW 346)
CC 1. *et-e/oarcoat (prevents, hinders): Wrzburg: fut. 3sg. with infix. pron.
1sg. nimirchi 7a11, past subj. 3sg. rel. with infix. pron. 3sg. m. and emph.
pron. 3sg. m. arnach naurchoissed som 8a4, Milan: pres. ind. 3sg. rel.
arachoat 31d10, impf. 3sg. rel. arachoided 83d2, pres. subj. 3sg. arci
46d11, pret. pass. sg. neg olnadnairchs 97d1, part. (io, i) airchoisse, nom.
pl. rel. nerchoissi 73c9, VN (n o) aircht, acc. sg. erchot 56a13 (2x), dat. sg.
ercht 39c20, erchot 121d4, erchoat 47c7; indfeith (pays heed to, is intent
on): Milan: VN indithem 65a5, acc. sg. indithem 21b2, indithim 93c15, gen.
sg. indithem 31a12, dat. sg. indithim 75d7.
fichid fights
PIE 2. *ei k- (LIV2 670f., IEW 1128f.)
CC *ik-e/oarfich (fights, does battle): Milan: VN (f a) airgal, dat. sg. ergail
34c12; 1. dofich (punishes, avenges): Wrzburg: pres. ind. 3sg. dofich
9d2, 15a16, dofeich 6a16, Milan: pres. ind. 1sg. with infix. 1sg.
n(o)mdichimse 38c21, 2sg. rel. dufichi 71b18, 3sg. (rel.) dofich 19d3,
115b13, dufich 123d8, neg with infix. 3pl. nsndg 24b17, pass. sg. diagar
101c16, impf. pass. pl. dofestais 29c7, pres. subj. 2sg. -dufess 44a9, pass. sg.
-dufessar 32c20, fut. 1sg. withinf 3pl. withsuff 3pl. dondafius 126c19 3sg.
duf 67c5, pass. sg. dufiastar 27c4, 129b4, pl. dufesatar 24b19, perf. 3sg.
with infix. 3sg. dafuich 43d19 pass. sg. duruacht 43d11, verb. nec. diachti
300

23d18, part. diachtae 111b6, part. toachtae, innephthoachtae 126c22, VN (f
a) dgal, digail 22b3, digal 22c13, 22c14, acc./dat. sg. digail (in this form,
prep, nor art, nor noun give any clue as to whether it is acc. or dat.), acc. sg.
digail 54b3, 91a21, gen. sg. diglae 24c1, 62a4, dat. sg. digail 26b23, 31c17,
dat. pl. diglaib 40a6; 2. dofich (attacks, fights): Milan: past subj. 3sg.
dufesed 33b12, -toisset 40d13, 3pl. arnachatoirsitis 48d27, VN (f ) togail
53d5, togal 40d14, dat. sg. togail 14a13, 54c17, 66c9, 68a4.
foceird throws, casts

PIE *//kerdh-// (3. *kerH- LIV2 353-5, IEW 934f.)
CC *kerd-e/oNote: Suppletive forms based on cuirethar are AII verbs and
recorded there.
Wrzburg: pres. ind. 2sg. rel. focheirt 13c21, 3sg. foceirt 32c11, 1pl.
focertam, 15b4, pres. subj. 2sg. rel. foceirr 13c24, impv. 1pl. with infix.
pron. 3sg. f. and emph. pron. 1pl. fuscerdam ni 34a2; pass.: pres. ind. 3sg.
fucertar 34a2, perf. 2pl. rolasid 15d1, canralsid 15a1, pres. subj. 1sg. rel.
\ral 7a4, 3sg. rel. arnarala 11b21; pass.: perf. 3sg. rel. rolaad 13d9, Milan:
pres. ind. 3sg. with infix. 3sg. m facheird 94c8, 3pl. rel. fuchertat 93a25,
pass. sg. forcertar 30c3, pl. focerdatar 96c11, fut. 3sg. focicherr 87d6, pret.
3pl. rel. with infix. 3sg. fachartatar 64a10, perf. 1sg. nachamralae 90c17,
3pl. nrolsat 16d2, pass. sg. rollaad 29c1, rolaad 44d2
gelid grazes, consumes

PIE ?*ghel- (LIV2 217f., IEW 365)
CC *guel-e/oMilan: pres. ind. 3pl. rel. gldae 80a11
gonaid pierces, wounds

PIE *ghen- (LIV2 218-9, IEW 491ff.)
CC *gan-e/ogonaid: Milan: part. gen. pl. goite 77a19; congoin (wounds,
pierces): Milan: part. congoite 58c17, VN conguin 23a5, acc. sg. chongain
90d10.
301
ibid drinks
PIE 2. *peh3(i )- (LIV2 462f., IEW 839f.)
CC *ib-e/oWrzburg: pres. ind. 3sg. ni ib 28b24, perf. 1pl. assibsem (MS assib
sem) 12a17, pres. subj. 3sg. [ni] roba 31b9, impv. 2pl. ibid 22c7, Milan:
pres. subj. pass. pl. with nas nebtar 101d5, fut. 3pl. ibait 30c18, VN (m o, u)
l, dat. sg. ul 94c12.
immcomairc questions, asks

PIE *prek- (LIV2 490f., IEW 821f.)
CC *ar-ske/oWrzburg: impv. 3sg. .i. na imchomarcad 28b12, Milan: pres. ind.
3sg. rel. immechomairc 27d4, impv. 1pl. imchomarcam 18a1, 18a3, past
subj. 3sg. immchomairset 20b18, rel. immechomairsed (Cf. McCone 1985:
96) 63c9, VN imchomarc 20b14, acc. sg. imchomarc 30b10, gen. sg.
inchomairc 107d07, dat. sg. imchumurc 77a15.
ingreinn persecutes, pursues with hostile intent

PIE ?*ghrei dh- (LIV2 203, IEW 456f.)
CC *gri-n-d-e/o
Milan: pres. ind. 2sg. with infix. 3pl. -indagreinn 36d2, 3sg. rel.
adgrinn 54b23, 73c1, adgreinn 140c2, 3pl. (rel.) ingrennat 26d12, 127c4,
with infix. 1sg. atamgrennat 127c8, withinf 3sg. m Cl C asidgrennat (as for
in) 18d2, with infix. 1pl. atangreinnat 45a16, with infix. 1sg. atamgrennat
39d13, impf. 3pl. rel. ingreintis 87c5, pres. subj. 3sg. arnaingre 111c6,
impv. 3sg. ingrainned 44a13, impf. 3sg. ingreinned 36d25, perf. 2sg.
-inroigrann 36d7, 3sg. inrograinn 26b24, inroigrainn 26d3, 37a4, 3pl.
adroigegrannatar 25b11, VN (n n) indgreimm, ingraim 26d12, acc. sg.
ingraim 56c7, 57c5, dat. sg. ingrammaim 33a9, ingraimmim (this form not
in DIL) 38c14, ingraimmm 87c1, gen. sg. ingraimme 18d23, ingraimmae
29d8, ingraimm 75a13, dat. sg. ingraimmim
ingrammanaib 38c13.
302
74a10, 74a12, dat. pl.

in-snaid inserts, ingrafts
PIE ?*sneHdh- (?*snedh- LIV2 571, IEW 972f.)
CC *snad-e/oWrzburg: pass. pres. subj. 3sg. with infix. pron. 1sg coatomsnassar
5b30, Milan: pres. ind. 3sg. insnadat 118d20, past subj. 3pl. insnastis 26c17
part. esnaisse 33c6, nom. pl. esnaissi 32c7.
ligid licks
PIE *lei h- (LIV2 404, IEW 668)
CC *lig-e/oMilan: fut. 3pl. lilsit 89d14.
lingid leaps, jumps

PIE *(s)prengh- (LIV2 583, IEW 660f., IEW 1000)
CC *leng-e/olingid: pres. subj. 3pl. rel. lengtae 129c20; conling (contests):
Milan: VN cuimleng, acc. pl. cuimlengu 112b8; foling (anticipates):
Milan: pret. 2sg. foroiblang 43d16, pass. pl. arforoiblachta 58d6, perf. 1sg.
foroiblang 95d11, 3pl. forruleblangatar 129c21; forling (leaps on/over,
(GOI 548)): Milan: VN (n n) forlaim(m) 23c21.
maraid lasts, persists

PIE ?*mer(H)- (IEW 969f.)
CC *mar-e/oWrzburg: pres. ind. 3sg. hore ndmair 3c15, Milan: fut. 3pl.
meraid 100b4.
meilid grinds, crushes

PIE *melh2- (LIV2 432f., IEW 716f.)
CC *mal-e/oMilan: VN mleith, dat. sg. mlith 23a20; domeil (spends,
consumes): Wrzburg: pres. ind. 3sg. rel. dommeil 6c7, 1pl. with dem. rel.
andumelam 13c12, 3pl. with infix. pron. 3sg. n. damelat 10c3, perf. 1sg.
ntormult 18a10, past subj. 3sg. with infix. pron. 3sg. n. ciadamelad 10d3,
303

1pl. cedumelmis 10c21, 3pl. domeltis 10c3, impv. 2sg. tomil 6c7, 3sg. with
infix. pron. 3sg. n. (len. only remaining) nach thoimled 11b18, Milan: pres.
ind. 1pl. with infix. dosmelmais 111b15, pres. subj. 1 sg. anatormal 119b6,
past subj. 3sg. doromlad 36a29.
mligid milks
PIE *h2mel- (LIV2 279, IEW 722f.)
CC *mlig-e/odoinmlig (promulgates, proclaims): Milan : pres. subj. 3sg.
coduinmail 50b1, perf. 3sg. durinmailc 31d3, VN (m o) tinmlegon, dat. sg.
tinmlegun 71c18.
naiscid binds
PIE *neHd- (*Hnedh- LIV2 227, IEW 758f.)
CC nad-ske/oarnaisc (guarantees, betroths): Wrzburg: perf. 1sg. with infix.
pron. 2pl.: arobrinasc 17b27; immnaisc (binds round, twists round):
Milan: part. immainse 62c9, nom. pl. immainsi 36d11.
orgaid kills, slays

PIE *h3erg- (LIV2 301, IEW 819)
CC *org-e/oorgaid: Wrzburg: pres. ind. 3sg. rel. nech orcas afini one who slays
his kindred (gl. homicidis) 28a4, Milan: pres. ind. 3sg. orcaid 19d6, fut. 2sg.
inn rr 77a10, 77a15, in n rr 77a13, int with infix. 3pl. indahierr 77a16,
perf. 3sg. roort 48c8, ruhort 63b12, VN orgun, orcuin 99b4, acc. sg. orcuin
23b4, 23b5, 85b9, orcain 73d6, gen. sg. oircne 22b1, dat. sg. orcuin 60d3;
as-oirc (cuts down, beats): Wrzburg: pres. ind. 3sg. rel. asircc 11a11,
Milan: perf. pass. sg. aschomart 34b18, ascomart 36b22, pl. aschomarta
26d11, resarta 34b13, part. esartae, dat. pl. esartaib 36b17, 37b29, VN
esorcon, nom. pl. esaircnea 51b21, dat. pl. essarcnaib 131b12; contairc
(strikes, pounds): Milan: pres. ind. pass. rel. contuarcar 34a27, past subj.
pass. pl. tuartis 54a18, pret. 2sg. contochmairt 17a2, 19c7, VN
comtharcon, contuarcan 34a27; 1. doessuirg (saves, excepts (?)):
304

Wrzburg: fut. 3sg. with infix. pron. 3pl. and emph. pron. 3pl.
dodaessarrsom 5c12, pres. subj. 2sg. with ma and infix. pron. 3sg. n.
madaessoir 17d2; 2. does(s)uirg (smites, slays): Milan: VN dat. sg.
dasarcuin, diasarcuin 84b11; dofairc (grinds, crushes) Wrzburg: pres.
ind. 3sg. rel. dofuircc 10d6, Milan: pres. ind. 3sg. dufuairc 84b12, 3pl.
dufuargat 85b6, impv. pass. pl. tuargatar 86d22, impf. 3sg. arnatuarrad
121d17, nachartuarad 121d18, 3pl. dufuarctis 33a6, pres. subj. 3sg. dufuarr
48c31, with perf. part. docomar 23d5, ducomarr 85c3, pret. 3sg. rel. with
infix. 3pl. annundacomart 36d9, perf. 3sg. docomart 45a11, 3pl.
docomartatar 22d4, verb. nec. tuarti 126c18, VN (f a) tarcon, gen. sg.
tuarcnae 29b9, dat. sg. tuarcain 56d4, 84c22, 129b11, acc./dat. sg. tuarcain
58a6; doimmoirg (presses, compress):Wrzburg: pres. ind. 3sg. with
infix. pron. 1sg. domimmuirc 23b36, fut. 1sg. doimmarr 9a20, perf. 3sg.
with infix. pron. 2pl. dobimchomartt 3b21, Milan: impv. 3sg. timmargad
136c8, impf. pass. sg. with infix. 1sg. dumimmaircthe 73c4, pl.
doimmairctis 36a26, pres. subj. 3sg. with perf. com daimchomarr 77a12,
past subj. pass. donimmarthae 130c21, fut. pass. sg. diummarthar 90a9,
perf. 3pl. with infix. 1sg. dommimchomartatar 39c32, part. timortae,
timmartae 44a2, timmartai 118d11, dat. pl. timmartaib 27a4, VN timorcon,
acc. pl. timmaircnea 55d18; frisoirg (molests, offends): Wrzburg: perf.
1sg. with emph. pron. 1sg. friscomurtsa 33a12, 3pl. rel. friscomartatar
5b11, past subj. 2pl. frisorthe 10c12, impv. 3sg. fridoirced 14a27, pass.:
pres. ind. 3sg. frissorcar 6c13, perf. 3sg. with infix. pron. 1sg. and emph.
pron. 1sg. fritumchomartsa 33a12, Milan: pres. ind. 2sg. frisorcai 44b31,
with infix. 1sg. frittamoirci (GOI 164) 44b26, 3sg. rel. frisoirc 63b17,
frisnoirc 28a20, 3pl. (rel.) frisorcat 15a8, 86c6, 104b2, frisorgat 39d30,
with infix. 1sg. fritammorcat 39c27, 62c21, pass. sg. frisnorgar (note Thes.:
n is inserted because of the preceding cum = Ir. an) 77d10, impf. 3sg.
friscorcad 118a1, 3pl. frisoirctis 67b14, rel. with infix. 3sg. frissidnoirctis
39a20, impv. 2sg. nafrithorcaid 114a9, pres. subj. 3sg. rel. frisnorr 15a10,
rel. with perf. com frischomarr 77a12, 3pl. rel. frisnorrat 80b9, past subj.
3sg. rel. frisnorrad 124d8, fut. 1sg. frissiurr 37c12, 3sg. (rel.) with infix.
1sg. fritammior 32d27, with infix. 2sg. fritatnarr 93a15, 3pl. with infix.
1sg. fritammiurat 33a1, rel. frisnerrat 100c9, perf. 1sg. -frithchomart 47a2,
305

3sg. friscomart 63b11, 3pl. friscomartar 111b1, part. (io, ia) frithortae 39a4,
39a14, 50a2, 59a5, 127b10, gen. sg. frithorti 58b8, nom. pl. frithortai 75b5,
gen. pl. frithortae 29b2, VN (f a) frithorcun, acc./dat. sg. frithorcuin 39c7,
acc. sg. frithorcuin 22d21, 32d7, 44a14, 55b11, 117b3 gen. sg. frithoircne
19c5, 38c24, 39c9, 90d18, 101d1, dat. sg. frithorcuin 36d1, 56b7, 61a10,
95b2, frithorcoin 130c6, acc. pl. frithoircnea 54d13, dat. pl. frithoircnib
55d11; fristuairc (thumps): Milan: pres. ind. 3sg. fristuairc 143c2;
immtairc (smites mutually): Milan: VN imtharcon, imthuarcain 16b7,
indsoirg (drives out, sets in motion): Milan: perf. 3pl. arrinsartatar 99c5.
reithid runs, hastens

PIE *ret- (LIV2 507, IEW 866)
CC *ret-e/oreithid: Wrzburg: pres. ind. 3sg. rel. rethess 24a17, 3pl. rethit
11a4, 13b13, Milan: pres. ind. 3sg. rethes 42c19, 1pl. a retham 120b2, 3pl.
rethait 138d6, rel. rethae 68b10, arretae 138d8, impf. 3pl. noreithis 129d4,
perf. 3pl. neg nadrorthatar 129c21, VN (m u) rith, gen. sg. retho 42c19,
retha 93b4, dat. sg. riuth 89c10, 129d4, rith 81b11; arreith (sets out for,
captures): Wrzburg: pres. ind. 2sg. rel. ararethi 6b22, Milan: pres. ind.
3pl. arrethat 56d12; contetarrat (comprises, contains): Milan: part. nom.
pl. / verb. nec. nephchomtetarrachti 55d11, VN comthetarracht, dat. sg.
chomtetracht 56b3; doairndret (runs about, roams): Milan: pres. ind. 3pl.
rel. duairndredat 81b2; doairret, doirret (brings in contention or
contest with?): Wrzburg: pres. ind. 3pl. rel. tairthet 27d23; doetarrat
(encompasses, comprehends): Wrzburg: pret. 3sg. doetarrid 19c11,
duetarrid 5c13, Milan: pres. ind. 3sg. duetarrat 30c6, pret. 3sg. duescmart
(Note Thes. : MS. duescmat: corr. Ascoli) 103d4, perf. pass. sg.
doretarracht (according to DIL this is pret., but the r shows it is perf.; this
form is peculiar: adding the pret. pass. suff -t to the stem ending -t should
give -s. See GOI p. 438-9 707, 8, where this peculiarity is noted, but not
explained. It could be a scribal error.) 33c20; dofarat (remains over,
remains): Milan: pres. ind. 3sg. rel. ni diurat 72b17, impf. 3sg.
dufuairthed 97d10, 97d12, past subj. 3p difuairsitis 57d3, fut. 3sg. n diuair
56d2, perf. 3sg. duruarid 44a20, -deruarid 31a6; doarmrat (follows,
306

comes or goes after): Wrzburg: pres. ind. 3sg. with infix. pron. 3sg. n.
dodirmo(rat) 9c1, Milan: pres. ind. 3sg. withinf 3sg. n dodiarmorat 21c3,
VN (f [a]) tarmracht, dat. sg. thiarmoracht 87d4; doim(m)thiret
(administers, serves): Wrzburg: perf. 3sg. with infix. pron. 3sg. n.
dodrimthirid
32c15,
and
with
ma
madudrimthirid
28d30,
3pl.
dorimthirthetar 32b5, Milan: VN timthirecht, thimthrecht 23a18, timthrecht

37b3, 73a10; fordiurat (no translation in DIL): Milan: pres. subj. 3sg.
fordiuair (found under for-duair) 23d7; indreith (invades, overruns):
Milan: pres. ind. 3sg. inreith 19d13, pres. subj. 3sg. co inr 134d1, past
subj. 3pl. inrestais 37d1, pret. 3sg. rel. aninrorad 84c2, 3pl. inrorthetar
104b8, perf. 3sg. inroraid 66d21, 3pl. inrorthatar 35a21, verb. nec. indrissi
127b15, part. indrisse 18c14, dat. pl. indirsib 35a21, VN (n o) indred 43d25,
102a21, gen. sg. indrid 48d20, 66c19, dat. sg. indriud 53a17, dat. pl.
indredaib 67c4.
rigid stretches, distends

PIE *rei- (LIV2 503, IEW 862)
CC 1. *rig-e/o- and *ri-n-g-e/origid: Milan: perf. 3sg. rorecht 20a23, 39c11, part. rechtae, nom. pl.
rechti 20a23; atreig (rises): Wrzburg: pres. ind. 3sg. adreig (for atreig)
13a12, fut. 1pl. with prep. In (and rel. which is elided) and emph. pron. 1pl.
ineresomni 93a7 (Taur. F. IV. 24), Thes. Pal. I, 713, Milan: pres. subj. 3pl.
ndarset 46a12, impv. 2sg. atr 126c3, fut. 1sg. ataimmresa 31c14, VN (n
io) irge, gen. sg. rge 83a5, dat. sg. rgiu 21c3; asirig (rises again):
Wrzburg: fut. 2pl. with rel. n asnersid 25b25, 3pl. rel. asserset 13b26,
25b16, with prep. and rel. inesirset 13c20, perf. 3sg. asreracht 4d27,
15d12, rel. asrracht 13b12, Milan: past subj. 3pl. esersitis 15c7, esersitis
15c8, VN (n io) es(s)irge, srge 38c9, gen. sg. esrgi 15d2, 45d7, esrgi
81d1; dodrig (incites, provokes): Milan: pres. ind. pass. sg. with infix.
2pl. dundabdurgathar 79c4, impf. pass. sg. dudurgaithe 80b5, dodiurgat
(Note Thes.: leg. dodiurgatha Ascoli; Sarauw emends to dodiurgatis, 3d
pl. impf.. passive) 105c11, past subj. 3sg. codudursed 33b14, verb. nec.
tudrachti 93a8, part. tudrachtae, tudrachtai 135b3; dorig (leaves,
abandons): Wrzburg: pres. ind. 3sg. with infix. pron. 3sg. f. and emph.
307

pron. 3sg. f. nsnderigsi 9d5, 1pl. (depon.) with emph. pron. 1pl. ni
dergemor ni (a prima manu) 15b23, perf. 2pl. 601, dorrachtid 18c6,
dorreractid 18c6, pres. subj. 2pl. nidrsid 20b10, with conjn. condirsid
25d27, Milan: pres. ind. 3pl. codurget 56b13, rel. with infix. 3sg.
dudnrget (DIL: -gat) 46c2, pass. sg. duragar 100c22, impf. 1sg.
codursinn 103b3, fut. 1sg. drus 137c7, 3sg. nsdr (in DIL as nisdr)
57a7, perf. 3sg. dureracht 74b4, 3pl. n derachtatar 57d12, part. drechta,
derachtae 18d18, drachtae 48d4, verb. nec. derachtae 110b1, derachtai
128d1, VN ndrge 111b3, modrgise 118b15.
saidid sits
PIE *sed- (LIV2 513-5, IEW 884-7)
CC *sed-e/osaidid: Wrzburg: fut. 3sg. seiss (MS isses? issis?) 26a8, perf. 3sg.
hodesid 3a7, amal donessid (MS donesfid) 26a8, Milan: pres. ind. 2sg. saidi
101c6-7 (2x), impf. 3sg. nosessed 135a13, perf. 3 sg. dunessid 121a11, VN
suide 15a1, 135a13, acc. sg. suide 15a13; adsaid (expects, awaits):
Milan: impf. 3pl. adsaitis 26c15; arnet (expects, awaits): Wrzburg:
pres. ind. 1sg. with emph. pron. 1sg. arneutsa 14a18, rel. araneutsa 23b27,
3pl. rel. anaranethem 31c17, Milan: pres. ind. 3pl. arneithet 39d25, past
subj. 2sg. arranethe 55a2, pass. sg. araneirnestar 118d10, perf. 1sg.
ararutneithius 46d14, with infix. 2sg. arrotneithius 46b20, 3sg. arroneith
50b8, arruneastar 50b8, arrneid 68a6, 3pl. arruneithset 50b9; immsaid
(surrounds, besieges): Milan: pret. 3sg. immasiassair 43b1, part. impesse,
gen. sg. impessi 20a25, gen. sg. / nom. pl. impessi 106d9, gen. pl. impesse
49b9, VN impuide, impsuide 43b10, dat. sg. impudiu 106b2, immsuidiu
123b9; indnat (waits, awaits): Wrzburg: perf. 3sg. (with ad- for ind- in
pretonic position) adroneestar 4c35, impv. 2sg. (depon.) indnite 10a21, 3sg.
indnadad 11d14, Milan: VN (n io) indnaide, acc. sg. indnaide 42c22,
nnaide 42c23; *insaid (No translation found, glossing insidere (ped ii:
605)): Milan: perf. 3sg. indessid 20a27, 3 pl. indestetar 58a2; remisaid
(sits before, presides): Milan: pres. ind. 3sg. remisaid 50d5.
308

sceirtid strips, peels off
PIE ?*skerdh (LIV2 558, IEW 940f.)
CC *serd-(i )e/oMilan: perf. 3sg. roscaird 14b2.
seinnid plays (an instrument), sounds

PIE *senh2- (LIV2 611, IEW 1046f.)
CC *san-naWrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. m. nodseinn
12c46, pass.: pres. ind. 3sg. rel. frissasennar 12c46, Milan: impf. 3sg.
nosenned 2b9.
sichid asserts, declars

PIE 2. *sek- (LIV2 526f., IEW 897f.)
CC 2. *sek-e/o
docoisig (supports, sustains): Wrzburg: impf. ind. 3sg. with
infix. pron. 3sg. m. dodcoisged 9a6; incosaig (signifies, indicates):
Wrzburg: past subj. 3sg. rel.: inchoissised 2c7, Milan: pres. ind. 3sg. rel.
inchoissig 59a7, inchoisig 137d5, inchosaig 116d3, pass. pl. incoissegar
48a11, 127d14, impf. 3sg. incoisged 24c14, with infix. 3sg. atcoisged
24c15, past subj. 3sg. incoississed 24c22, 28b10, rel. inchoissised 56a13,
pret. 3sg. rel. inchoisecht 16c10, perf. 3sg. rel. inrochoissecht 43d27.
sligid cuts, fells

PIE *slei - (LIV2 566f., IEW 663f.)
CC *slig-e/osligid: Milan: perf. pass. pl. roslechta 48d28, with infix. 3pl.
conrusleachta 53d11, VN (f ia and t) slige 2a6, 51c17, 67d2, acc. sg. sligi
49b2, 16c12, dat. sg. sligi 2a6; adslig (tempts, entreats): Wrzburg: pres.
ind. 3sg. adslig 14d27, pres. subj. 3sg. with cia, ciadsli 20b2, Milan: VN (n
o) aslach 95b6, acc. sg. aslach 141d1, gen. sg. aslaig 28b7, dat. sg. asluch
26c9-10, gen. pl. aslach 38a6; foslig (smears, anoints): Wrzburg: pass.
perf. 3pl. foruillecta 7d9, Milan: part. fuillectaese 127c17.
309

teichid (flees, runs away)
PIE *tek- (LIV2 620f., IEW 1059f.)
CC *tek-e/oteichid: Milan: pres. subj. 1sg. rel. notes 29d2, past subj. 3sg.
nutesed 29d9, pret. 3sg. tich 32b24, perf. 3pl. rotachatar 44a19, VN (m o)
teiched, acc. sg. teiched 54b12, 63c14, dat. sg. teichiud 44a18, hteichiud
109c02, acc./dat. sg. teiched 77a18, gen. sg. / nom. pl. teichid 77d13, 100b1;
adteich (flees to, resorts to): Milan: pres. ind. 3sg. ateich 39b6, perf. 3 pl.
attechtatar 90b16, VN (n o) attach, atach 40b13, 66d1, nom./acc. sg. atach
49d14, gen. sg. ataig 40b12, 54b21, dat. sg. atuch 121c21; tremiteich
(flees across): Milan: impf. 3sg. tremiteiched 54c21.
teinnid cuts, cracks

PIE 2. *tend- (LIV2 628, 1062f.)
CC *tan-n-d-e/o- or *tend-e/ofoteinn (nibbles): Milan: pres. subj. 1sg. cofotheasa 3a9.
tit goes, proceeds

PIE *stei gh- (LIV2 593f., IEW 1017f.)
CC *ti-n-g- & *teig-e/otit: Wrzburg: pres. ind. 1sg. tiagu 17b20, with emph. pron. 1sg.
tagussa 17b18, 3sg. ntit 9d31, rel. tte 9a3 (bis), 9d5, 20a12, hre dte
11d7, 1pl. with emph. pron. 1pl. tigmini 15b28, rel. intain diagmani 3a15,
tiagmeni 15b28, 3pl. tiagait 5c16, 11d12, Perfect. pres. ind. 2pl. with infix.
pron. 3sg. n. (elided) nachdigtith 9b19, fut. 1pl. rigmi 15c23, sec. fut. 3sg.
n rgad 11a22, pret. 3sg. luid 3c37, perf. 1sg. with emph. pron. 1sg.
docoadsa 18d6, rel. dochood 17d7, with emph. pron. 1sg. nidechudsa
14c40, 3sg. docoid 21a12, 31d19, docoith 11a22, rel. dochoid 14a19, 14d30,
28b30, with emph. pron. 3sg. dochoidsom 14c20, with prep. and rel.
diandechuith 16c7, rel. condechuith 22b11, 1pl. with infix. pron. 3sg. n.
ciadodchommar 23d23, 3pl. docotar 29a8, pres. subj. 1sg. with emph. pron.
1sg. ciathiasusa 23c31, 3sg. mathis 14a14, perfect. pres. subj. 3sg. doci
29a28, rel. arna dich 9d24, ar nadecha 28b30, past subj. 3sg. rel. notheised
32a17, 2pl. notsst (MS ntresst) 9b19, impv. 3sg. na tit 10a15, pass.:
310

fut. 3sg. rigthir 9a23, Milan: pres. ind. 3sg. (rel.) tit 103d27, tt 21c3, (rel.)
teit 24d30, 27c10 (2x), 37a10, 94c10, 101c4, 109a02 (2x), 114a4, (rel.) tit
47a17, 56b33, 67d24, 109a02, 123b13, rel. tete 22b1, 28c19 (2x), 62b20,
rel. dete 93b12-13, hitait 24d30, nitait 109d05, conatait 68d5, rel. diateit
100b9, with nas with infix. nntet 77a14, 3pl. tiagait, -tiagat 51d3, rel.
tiagtae 125a2, pass. sg. tiagair 15a2, 16c5, impv. 2sg. tit 34a1, 53c13, tait
53c11, 3pl. tiagat 54b12, impf. 3sg. noteged 54c18, pres. subj. 3sg. ciathes
23d23, arnate 36a23, arnat 126a4-5, pass. sg. cianutiastais 117d3, past
subj. 1sg. nothisinn (Cf. McCone 1985: 96) 41d9, 3sg. notsed 23b12,
54a21, notesad 34d6, rel. neg with infix. 1sg. nachamthsed 107b8, 3pl.
ducoistis 34a9, -dechsaitis 42a6, -dechsitis 42a7, fut. 1sg. tiagsa 58c6,
annurig 87b18, sec. fut. 3pl. diaregtais 9b10, pret. 3sg. luid 16c10, 52 line
14, 55c1 (3x), 74b5, luide 55c1, 127d3, perf. 3sg. (rel.) docoid 32d10,
43d27, ducoid 63c19, ducuaid 65c9, (rel.) -dechuid 54d3, 54d7, 65d12, rel.
duchoid 74a12, 84c9, 1pl. condechummar 63c14, 3pl. dachotar 38b2,
ducuatar 66c16, dochotar 124c26, perfect. pres. subj. 1pl. -dechsam 62d1,
VN (f a) techt 15a1, 15a2, 40a5, 72b3 (Note Thes.: For the construction cf.
Ml. 15a 2; a different construction in Ml. 83a 5), gen. sg. techtae 32c17,
54b2, dat. sg. techt 22d17, 58c4, 80d9, VN dul, acc. sg. dul 23d23; cetatat
(comes first; this is not a citation form in DIL): Wrzburg: perf. 3pl. rel.
cetathuidchetar 21c5; conitet (goes with, accompanies): Wrzburg:
pres. ind. 3sg. rel. ndchometig 22a13, 2pl. itgid 22a26, fut. 3sg. rel.
conit 30c27, pres. subj. 2sg. mani comitis 6c6, 3sg. conit 6c7, coneit 6c1,
impv. 3sg. na comitged 10a7, Milan: pres. subj. 3pl. eirsitis 87c4, VN
coimitecht, comaitecht 72b18, acc. sg. chomaitecht 15b5, acc./dat. sg.
coimtecht 82d1; conimthet (accompanies): Wrzburg: fut. 3sg. with
infix. pron. 3sg. m. cotnimth 12c4, Milan: impf. 3sg. -conimtised 42c31;
contat (accompanies): Milan: VN coimitecht, nom. pl. comthaitechta
37c7-8; dott (comes): Wrzburg: pres. ind. 3sg. with emph. pron. 3sg.
m. dotit som 2a3, amal dotitside 25b28; 503 rel. dothet 22a11, 3pl.
dotagat 5a5, dotiagat 27a27, fut. 1sg. doreg 7d15, 3sg. with rel. n dorriga
25b28, and with infix. pron. 3sg. n. dondriga 25a38, niterga 17c4, pres.
subj. 1sg. dutas 1a7, 3sg. rel. dothi 13a12, impv. 2pl. natat 25c14; pass.:
pres. ind. 3sg. dotiagar 6a29 (= 6a28, Thes. Suppl. p. 42 ). Perfect.
311

suppletive forms: 1sg. with prep. do and rel. diatuidched 10a28, 3sg.
dodechuid 17b29, 24c10, 27c8, (do)dech(uid) 31d2, with prep. do and rel.
diandechuith 16c7, 1pl. dodechommar 25a12, with rel. n dondechommar
24c17, past subj. 3sg. with conjn. tuidchissed 15c16, Milan: pres. ind. 3sg.
dott 21c3, dutt 22b1, dotet 53a8, rel. dotht 46a7, 46a12, dothet 53a8, 3pl.
(rel.) dutiagat 45d12, 103b16, 111c9, rel. diatt 56b1, pass. sg. dutiagar
101c6-7 (2x), 116b5, pl. dutiagtar 106c3, impf. 3sg. cotaised 121b3, 3pl.
dutigtis 108b01, pres. subj. 2pl. -thaisid 33b15, past subj. 1sg.
tuidchesmais (Note Thes.: MS. tuichesmais: corr. Ascoli) 93b5, 3pl. rel.
dondichsitis (note Thes.: regularly dodichsitis) 104c5, fut. 3sg. dorega
28a10, ntirga 121a17, 1pl. duregam 80d10, nitergam 107d11, 2pl. daregaid
33b15, 3pl. doregat 61a13, 111c14, rel. durigat 106a4, cond. 3sg. noregad
118b6, pret. 3sg. doluid 55c1, dialuid 58c4, perf. 3sg. (rel.) dodechuid 16c5,
46b6, 126b10, rel. dundechuid 44b1, ndechuid 98b8, 3pl. dundechutar
111c14, VN tuidecht, acc. sg. tuidecht 46a17, 53d9, 104c5, dat. sg. tuidecht
44b1, 60a10; fortt (helps, assists): Wrzburg: pres. ind. 3sg. fortit
4a27, rel. forthit 4a27, with infix. pron. 1pl. and emph. pron. 1pl.
ninfortitni 4a27, 2pl. with emph. pron. 2pl. afortitsi 14c3, afortegidsi
14c2, pres. subj. 2pl. rel. with infix. pron. 1sg. and emph. pron. 1sg.
cofardumthsidse 7a12, Milan: pres. ind. 2sg. fortig 44b32, 3sg. (rel.) fortt
30c3, 80d8, rel. forthet 35a8, with infix. 3sg. m foridtet 30c3, impf. 3sg.
fortattet 43b11, pres. subj. 2sg. fortis 78c1, 3pl. fortiassat 68d7, fut. 2sg.
forregae 19a1, perf. 2sg. rel. fordechud 78c9, 3sg. fordechuid 138a7, VN
(f a) fortacht 38b7, acc. sg. fortacht 27d19, fortachtain 83b11, gen. sg.
fortacte (this form not in DIL) 19c19, dat. sg. fortacht 22c3, 30b11, 30b22,
33a13, 37b25, 38c4, 46b20, 86b7, 88a6, dat. pl. fortachtaib 83b10; ?fott
(goes under, passes beneath): Milan: fut. 3sg. forriga (Note Thes.: the
first r over the line; leg. foriga?; rr may denote unaspirated r) 85b1a;
fristat (comes against, is at variance with): Wrzburg: pres. ind. 3sg.
fristait 14a13, rel. fristait 12d39, nifrittit 31a6, Milan: pres. ind. 3sg. (rel.)
fristait 23c11, 3pl. frithtaigat 17c5, pres. subj. 3sg. frestai 31d6, past subj.
1sg. fristaissin 132a5, perf. 3pl. fristuidchetar 21c2, 67b20, 67b22, VN
frithtuidecht, dat. sg. frithtuidecht 16b6, gen. sg. frithtuidechtae 25b13;
inotat (enters into, comes upon): Milan: pres. ind. 1pl. rel. inotgam 96b5,
312

3pl. (rel.) inotgat 44c22, pres. subj. 1pl. inotsam 16a16, pret. 3sg. inolaid
25a21, perf. 1sg. inrualad 142a2, 2sg. rel. huarindualad 93c14, 3sg. rel.
inrualaid 71c7, 3pl. nad rindualdatar 24b11, inrualdatar 24b11,
inraldatar 62b15; immtit (goes around, goes about): Wrzburg: pres.
ind. 1pl. imtiagam 6a30, pres. subj. 2pl. ni imthesid 22c4, Milan: pres. ind.
3pl. rel. immetiagat 45c9, rel. imm(t)t 45c9, pres. subj. 1pl. nimthiasam
46c20, VN (f a) imthecht, dat. sg. imthecht 46a14; remitt (goes before):
Wrzburg: perf. 3pl. n remdechutar 5a30, pres. subj. 3pl. with infix. pron.
3sg. n. diandremthiasat 5a32, Milan: pret. 3sg. remelluid 132c13; tarmitit
(traverses, goes across): Milan: VN (f a) tairmthecht, gen. sg. tairmthechta
74b7, dat. sg. tairmthecht 39a17; tremitat (penetrates): Wrzburg: pres.
ind. 3pl. rel. with infix. pron. 3sg. n. tremitagat 25d14, Milan: pres. ind.
3sg. rel. tremethait (Note Thes.: MS. tremietthai: corr. Thurneysen; should
probably be emended to tremietha, cf. Ml. 127b 11, sg. 3 tremaetha Laws
IV, 344, 346) 43c14.
313
7.2 BII
daimid endures, suffers
PIE 2. *demh2- LIV2 116f. (IEW 199f.)
CC *dam-i e/oaddaim (acknowledges, admits): Milan: pres. ind. 3sg. ataim
42b26, 3pl. ataimet 131d16, pres. subj. 3pl. conrdamat 131d16, datardat
131c16; confodaim (co-endures): Milan: pres. subj. 1pl. mconfodmam
4a17, fodaim (suffers, endures): Wrzburg: pres. ind. 1sg. rel. fodaimim
se 23b12, fodaimimse 23b17, 2pl. fodaimid 17b24, 23c7, 3pl. rel. fodaimet
2c10, n fodmat 20c21, perf. 1sg. rel. forrodamar 19d20, pres. subj. 1sg.
with infix. pron. 3sg. n. fadam 29d27, Milan: pres. ind. 2sg.
cidaradodmaisiu 55d11, (rel.) fodaimi 55d11, 55d14, 3sg. rel. fodaim 99b1,
cidarafodaim 55d11, 1pl. fodaimem 111c13, 3pl. (rel.) fodaimet 26d12,
77d2, 87b17, 87b20, impf. 1sg. rel. neg. with infix. nadfordamainn 107b8,
pres. subj. 3sg. fudama 68d14, pass. pl. fondamtar 46d6, past subj. 1sg.
confodmain 73c7, 3pl. afondamtis 34b11, fut. 2sg. rel. fondidmae 35c33,
3sg. n fuidema 56c9, 3pl. (with infix. 3pl.) fosdidmat 15c10, perf. 1sg. (rel.)
forodamar 22d5, 58d13, 95d13, 95d14, 132c12, with infix. 3pl. fosrodamar
39a13, 3sg. (rel.) forodamair 54b28, furodamair 131b12, 62d9, rel.
forudmair 54a35, 3pl. rel. forodamnatar 90c13, furodamnatar 96b8, with
infix. 3sg. fondrodamnatar 105b9, VN foditiu, dat. sg. fodaitin 24b13.
figid weaves, plaits

PIE ?*eg- LIV2 662 (IEW 1117)
CC *eg-i e/oMilan: VN (f i) fige, gen. sg. fige 59d2, figi 111b16.
gaibid takes
PIE ?*gheHb- or ?*gheHb- LIV2 195 (IEW 407-9)
CC *gab-i e/ogaibid: Wrzburg: pres. ind. 1sg. nngabimmse 16d4, 2sg. with
infix. pron. 1sg. manumgaibi 32a16, 3sg. gaibidside 9a3, nigaib 3d8, nigaib
314

16d4, with prep. and rel. ingaib 29b19, rel. gaibes 3d15, 11a4, 26a13, with
dem. rel. angaibes 12d42, fut. 3sg. gebaid 8a7, rel. gebas 11a6, 3pl. nigebat
4c8, with suffix. pron. 3sg. m. as object gebtit (for gebti) 26a8, perf. with
rel. n 1sg. rongabus 23b18, amal rongabusa (-sa = -ssa) 9d25, amal
rongabussa 23c11a, 3sg. rongab 6d12, amal rongab 12b1, 27a11, 15, 3sg.
rogab 2c21 (bis), 4b18, 12a25, 24c14, 1pl. nirogabsam 24d2, with emph.
pron. 1pl. nirogabsamni 24b20, 2pl. rel. rogabsid 26a25, 3pl. rel. rogabsat
12b13, pres. subj. 3sg. rel. arnagaba 28a20, arangaba 32a2, congaba 31c8,
perf. rel. rogba 28d6, 11, with infix. pron. 1pl. nachin rogba 15d40,
nachinrogba 15d40, rel. lasse gabas 13d22, perf. rel. 2pl. rogbaid 26b7,
3pl. rel. rogbat 30a24, past subj. 2pl. with rel. n nongabthe 26b13, perf. rel.
with emph. pron. 2pl. rogabthesi 24c8, 26d7, impv. 3sg. gaibed 12d41,
2pl. gaibid 7b14, 9a14, 22d11, 27b16; pass.: pres. ind. 3sg. gaibthir 16d7,
with dem. rel. angaibther 12c44, rel. gaibther 28c20, perf. hrogaibther
13b13, perf. 3sg. rogabad 7a7, pres. subj. 3sg. rel. cia gabthar 29b5, past
subj. 3pl. rel. congabtis 19a4, Milan: pres. ind. 3sg. gaibid 50a5, 64c12,
nadngaib 34b6, with infix. 1pl. nongaib 43a14, with infix. 3pl. nisgaib
34b6, rel. gaibes 36a1, 101c6-7, 1pl. gabmai 32a4, 3pl. gabit 38a8, rel.
gaibde 76a16, pass. huangaibter 122c2, impf. 2pl. with infix. 3sg.
nosgabth 68b2, nusgabtis (note Thes.: MS. nus gabtis, leg. nusgabthe?
Thurneysen, which has been translated. Can the -s- be an affixed pronoun,
as in ro-fiti-s Wb. 6a 18, 14c 12? W. S.) 68b3, 3sg. nogaibed 138d1, 3pl.
nogaibtis 83a4, with infix. 3pl. nosgaibtis 2b7, condagaibtis 31c6, pass. sg.
nogabthe 35b18, nogabthe 48d27 nogaibthe 138d1 pass. pl. nogabtis 35b16
pres. subj. 3sg. irogbad 39d22, compounded with lm: nondalmgaba
43a2, pass. sg. arnarogabthar 30d13, diangabthar 118a7, past subj. 3sg.
(rel.) nogabad 33a5, 113c7, neg. with infix. 3sg. nachgabad 69a17, 3pl.
nugabtis 39c15, rogabtis 123a1, 139a9, rel. nongabtis 53c13, rel.
annungebtais (this form not in DIL) 95a9, sec. fut. 3pl. rel. with infix. pron.
3sg. nundagebtis 34c8, pret. 3sg. with infix. 3sg. nogabad 35a9, pass. sg.
gabthe 55c1, perf. 3sg. (rel.) rogab 25a16, 35b24 (2x), 44b1, 48d28, 52 line
315

2056, 55c1, 90b15 (2x), 127a12, 137b7, (rel.) rogab 36a32 (2x), 38c3 (2x),
45a3, 50a10 (2x), 74b12, 111c11, with infix. 3sg rundgab 32d05, 67d14,
with infix. 3sg. fuandragab 38c5, 42b7, 50a8, rel. with infix. 3 sg. ragab
44b2, 50d18, rel. with infix. 3sg. masc. rondgab 30b11, rel. rongab 38c7,
82d11 (2x), 131c12, rel. with infix. 3sg. m. arinrogab 35a8, rel. with infix.
rodgab 49a2, rel. with infix. 3sg. rodngab 61a1, with infix. 3pl. rosgab
57c13, 3pl. rogabsat 139a15, rundgabsat 67d14, pass. sg. (rel.) rogabad
14a4, 17b18, 24d10 (2x), 24d12, 24d14, 24d26, 24d29, 35a8 (2x), 35a10,
35b10, 38c4, 45d7 (2x), 56b16, 67c18, 86d13, 98c10, 110d7, 112c11 (x2),
113d3, 116c2, 133b1, rogabthe 16a10, rogabath 24d13, rel. ararogbad
74b11, rel. diarogbad 103b7, pass. sg. rel. rogabad 74b1 pl. ragbtha 35b24,
rogabtha 133b2, 139a6, 139a8, 139a10, 139a11, part. rongabthe 24d7, verb.
nec. gabthi 76d4, VN (f ) gabl, acc. sg. gabail 22b1, gen. sg. gabalae
83a3, dat. sg. gabail 29d4 (this gloss not in DIL), 31b3, 47b16, 60a12 (2x),
dat. pl. gabalaib 54b25; adgaib (gets back, recovers): Milan: impv. 3sg.
atabgabed 20d11; argaib (seizes, captures): Wrzburg: perf. past subj.
3sg. rel. with infix. pron. 1sg. and emph. pron. 1sg.: condomarrgabadsa
17d14, Milan: pret. 3pl. arangabsat (note Thes: cf. aritngaib Laws IV. 336,
YBL. 91b 29; n is the infixed relative after cum = Ir. lasse or intain. If it
were to gloss cum tenuerunt, it would have been arrungabsat) 74b2, VN (f
) airgabl, gen. sg. ergabale 29c18; aringaib (avoids, flees from):
Milan: pres. ind. 3sg. nerngaib (Note Thes.: MS. ner n gaib; cf. erngabthi
Ml. 16a5) 32d19, verb. nec. airngabthi, erngabthi 16a5, VN airngabl, dat.
sg. airngabail 29c14; asgaib (no translation given in DIL, hapax
legomenon, glossing excipere, take out, remove): pres. ind. pass. pl. ni
esgaibter 145c1; asingaib (surpasses, exceeds): Milan: pres. ind. 3sg.
(rel.) asingaib 23a13, 57d16, 89d8, 125d8, 131a6, 144d1, pres. subj. 3sg.
arna esngaba 22c8, perf. 1sg. asringbus 130d4 (x2), 3sg. (rel.) asringaib
(cf. McCone 1985: 94) 22d9, 32d8, 32d10, 3pl. asringabsat 113d8,
nadresngabsat 122d8, part. esngabthe, esngabthi (Note Thes.: rectius
-esngabthe For other instances of final i for e see 4 CZ. 52) 22d9; congaib
56
316

(contains, includes): Wrzburg: pass.: pres. ind. 3sg. congaibther 21c6,
Milan: pres. ind. 3sg. congaib 131c12, impf. 3pl. with infix. 3pl.
nscongbaitis 55a18, pass. sg. conumgabthae 27b7; conocaib (lifts up,
raises): Milan: pres. ind. pass. sg. conocabar 57d8, impv. with infix. 2pl.
cotabucabar 46a7, impf. 3sg. ucbad 35c6, cotnucbad 53d8, 3pl.
-conucbaitis 58c2, with infix. 1sg. condammucbaitisse 39d11, pass. sg.
conucbad (Note Thes.: leg. conucbada, unless conucbaim be here
intransitive, which is contrary to the general usage in O. Ir.; but cf. Hy. II
66, v 107) 83a4, pres. subj. 3sg conocba 79c5, 3pl. arndacumcabat 46a12,
past subj. 3sg. conucbad 39a22, fut. 3sg. conocaeba 20b5, pass. sg.
conuicgebthar 64b8, perf. 3sg. nadruchumgab 20a7, with infix. 3pl.
connuargab 37b15, pass. sg. conuargabad 32c1, impf. / past subj. 3pl.
conucabtis 68c6, part. cumgabthae, chumgabtha (Note Thes.: rectius
chumgabthae) 84c3, nom. pl. cumgabthi 43d7, VN (f ) cumgabl 57d4,
79d5, chumgabal 94c2, cumgabal 138a1a, acc. sg. chumgabail 39d22;
dofcaib (lifts up, raises): Wrzburg (note that Wb. Lex. puts these under
a separate headword, docaib): perf. 1sg. rel. with infix. pron. 3sg. f.
condaturgabusa 26d10, tuargab 26d11, ndturgab (MS ndturgabar)
14b22; pass.: perf. 3sg. turgabad 14b22, Milan: pres. ind. 1sg. dorogbaim
43a5, dofurgaib (raises, lifts up): Milan: pres. ind. 3sg. dufurgaib 138b1,
rel. dufurgabat 75a2-3, pret. 3sg. dufurcbad (DIL dufurcbat) 72b7, perf.
3sg. dururgaib (cf. McCone 1985: 94) 63a15, 138d11, dururgab 72c16, 3pl.
dufuargabsat 96c1, part. (io, i) turcabthae, turgabthae 86d14, nom. pl.
turcabthi 32b1, gen. pl. turcabthae 75a4, VN turcbl, turcbal 75b19, gen.
sg. turcbalae 85c13, dat. sg. turcbail 94b18; dogaib (lessens,
diminishes): Wrzburg: perf. pres. subj. 3sg. arnadergaba 10d13, Milan:
fut. 3pl. neg. n digebtar 73d13, perf. pass. sg. dorogbad 17a13, durogab
34d18, 111b28, part. digabthe 45c12, digabthi 107c07; dorogaib
(commits, transgresses): Wrzburg: pres. subj. 3sg. with dem. rel.
andorogba 22b9, 2pl. andorogbid 27b18, Milan: pres. ind. 3sg. dorogaib
16c13, 3pl. dorogbat 28d11, past subj. 1sg. dorogbinn 39a18, pret./perf.
1sg. durogbusa 23c13, 3sg. dorogbad 58d1, dorogab 71b14, hotorgab
32a23, 3pl. dorogabsat 54a36, 98c6, codurogabsat 108a06, pass. sg.
durogab 32c9, part. (io, i) torgabthe 27c17, f acc. sg. torgabthi 23c14, VN
317

(f ) targabl, torgabal 15c11, acc. sg. torgabail 99d6, gen. sg. torgabale
58a19, targabal 71b3, dat. sg. torgabail 22c7, 22d24, targabail 59a6, acc.
pl. torgabala 44b10-11, 138b7, dat. pl. targabalaib 70d6; focaib
(leaves): Wrzburg: perf. 1sg. with infix. pron. 2sg. fotrcbussa 31b1,
pass.: perf. past subj. 3pl. rel. arnafarcabtis 31d13, Milan: pres. ind. 3sg.
rel. neg. with infix. 3sg. nadfacaib 102a20, 3pl. rel. fucbat 80a10, fut. 3pl.
ni confoigebat 69a8, perf. 3sg. foragab 30a9, foracab 37d10, 53c14, 3pl. rel.
foragabsat 95a12; fogaib (finds, gets): Wrzburg: sec. fut. 3pl. with infix.
pron. 3sg. n. 8a14, Milan: pret. 1sg. n cosfuarsa 57d3, nachfuar 57d3,
fresgaib (rises, ascends): Milan: VN (f ) fresgabl, acc. sg. frisngabail
42c30, dat. sg. fresngabail 56b1, 56b3; frisgaib (checks, restrains):
Milan: pres. ind. 3sg. with infix. 3sg. neg. nachidfrithgaib 55c11, impf. 3sg.
frisgaibed 49a24, perf. 3sg. neg. nrufrithab 124c11; immimgaib (goes
around; avoids): Wrzburg: impv. 2sg. imcaib 28c24, with infix. pron. 3sg.
m. immanimcab 30d20, 2pl. imgabaid 9d6, Milan: pres. ind. 3sg. imimgaib
22c11, immeimgaib 35d29, with infix. imdaingaib (Thes: leg. imdaimgaib;
this form not in DIL) 38a3, impv. 2pl. imgabaid 116b5, pres. subj. 1sg.
immenimgaib 29d2, 3sg. imnimgaba 15b12, -nimgaba 36b11, con imgeba
52a13, 1pl. -imgabamsn 35d1, imdaimgabam (this form not in DIL) 35d19,
comimmimgabat 114b8, past subj. 3sg. immenimgabad 41a6, verb. nec.
imcabthi, imgabthi 40b9, 51b8, 51b10, 58a20, dat. pl. imgabthib 29d6, VN
imgabil, imgabail 14c12, 35d14, indocaib (exalts, glorifies): Milan:
part. inducbaide 64d9, 75c5, 139c2, VN (f ) indocbl, inducbl 75c4, gen.
sg. inducble 51d10; ingaib (reproaches, censures): Wrzburg: fut. 1sg.
with emph. pron. 1sg. ingeb sa 8d12, pres. subj. 3sg. rel. with infix. pron.
3sg. m. indid ningaba 11d8; pass.: fut. 3sg. with infix. pron. 1pl.
ninincbthar 15d27, Milan: pres. subj. pass. sg. ingabthar 22c10, pl.
ingabtar 54a7, impf. ingaibthe 133b5, 133b7, past subj. pass. sg. -ingabthe
32c19, VN ingabal 54a10, acc. sg. ingabail 74c3.
gairid calls, cries out

PIE *ar- (LIV2 161, IEW 352)
CC *gar-i e/o-
318

adgair (sues, prosecutes): Wrzburg: pres. ind. 2pl. with infix. pron. 3sg.
n atgairith 9c22, perf. 3sg. w infix. pron. 2pl. adobragart (a prima manus)
19b5, pass. perf. 3sg. adrograd 3c22; ? adtogair (no transl.): Milan: pres.
subj. 3sg. adintogra 22d23; arfcair (proclaims, announces): Wrzburg:
pres. ind. 3sg. rel. arafocair 16d7; pass. pres. ind. 3sg. arfocarar 16d7;
argair (forbids): Wrzburg: pres. ind. 3pl. with infix. pron. 3sg. m.
arangairet 1d7, pret. 3sg. arniargart 31c25, perf. 3sg. rel. with emph. pron.
3sg. m. ararogartsom 5c23, ararogartsom 5c23, pres. subj. 3sg. mani
airgara 2c18; pass. perf. 3sg. with rel. n arrograd 9b13, rel. ararograd
3c25, Milan: pres. ind. 3sg. argair 27c21, with infix. 3sg. masc aringair
27c21, fut. 3pl. argerat 121b8, pret. 3sg. argart 55c1, neg. with infix. 3sg.
nandargart 53d9, part. ergarthai 121d6, VN (n io) airgaire, gen. sg. ergairi
35c11, 35d18, dat. sg. ergairi 48c5, anergairiu 125a7; ascongair
(proclaims, gives notice): Milan: pres. ind. pass. sg. ascongarar 14c20,
fut. ascongr 126c8, pret. pass. sg. asrochongrad 121d19, verb. nec.
asngairthi, dat. pl. esngarthib 104d7, part. esngairthe, gen. sg. / nom. pl.
esngarthi 94b5, VN esgaire, esgaire 105c6; doaccair (pleads, sues):
Wrzburg: fut. 3sg. taiccra 6b28, perf. 1pl. doracartmar 2a16, Milan: pres.
ind. 3sg. duacair 36b3; doairngir (promises, foretells): Wrzburg: perf.
3sg. dorairngert 14c32, 30b2, 31a9, intan durairngert 33d10, with dem. rel.
andorairn.gert 25c28, andorairngert 31a7, andurairngert 5c9, 30b3; pass.:
perf. 3sg. dorairngred 2c12, 19c5, with dem. rel. andurairn.gred 19b22,
andurairngred (MS anduraingred) 33b3, past subj. duairngerthe 2c17,
Milan: pres. ind. 2sg. rel. duargir 106a2, 3sg. duarngir 94a4, rel. tairngir
50d10, 2pl dunairngarid 75d6a, 3pl. duairgerat 87b15, perf. 1sg.
dundarairgiurt 109c09, 2sg. rel. durairngirt 74c20, 3sg. dorairngert 46c20,
136c12, (rel.) durairngert 105b11, 108b08, 108c02, 123c1, 126c10, 130c16,
3pl.
dundrairngertar
(Thes:
MS.
dundraingertar:
leg.
perhaps
dundrairgertar, 4 CZ. 60, DIL: dund- rai[r]ngertar) 67b8, rel.

dorairngerad 113d5, VN tairngire, gen. sg. tairngiri 68b4, tairngeri 78c11,
dat. sg. tairgiriu 33d22; dogair (calls, summons): Wrzburg: perf. 3sg.
dorogart 21d2, with infix. pron. 3pl. dodarogart 22c1; pass.: perf. 3sg.
dorograd 10a22, amal dorograd 10a12, with rel. n donrograd 20d9, with
infix. pron. 2pl. dobrograd 24c4, Milan: pres. ind. 3sg. (rel.) dogair 29c6,
319

30c17, 1pl. dogairemni 37b16, 3pl. rel. dugairet 62b7, pres. subj. 1sg. with
infix. 2sg. indattogar 72c4, pass. sg. torogarthar 43b15, perf. 3sg. rel.
dorogart 69d14, with infix. 3sg. m darogartsom 20b2, VN (n n) togairm,
acc. sg. thogairm 102d17; doingair (calls): Milan: part. dingarth 93b7;
forcongair (commands, ordains): Wrzburg: pres. ind. 1sg. lasse
forcongur 19d25, rel. forchongrimm 9d30, 3sg. with emph. pron. 3sg. m.
Farcongair som 13a27, rel. forchongair 15a26, with dem. rel. aforchongair
5c23, impf. ind. 3sg. rel. farchograd 32d9, perf. 3sg. forrochongart 20c9,
pres. subj. 1pl. rel. with emph. pron. 1pl. (with irreg. len.) coforchongramni
11b16a, Milan: pres. ind. 3sg. forcongair 53a23, 94b3, 129b7, 146a1, rel.
forchongair 94b10, rel. with infix. pl. fortacongair 59c11, rel. with infix.
3pl. forndacongair 94b3, pass. forcongarar 129b8, perf. 3sg. forrochongart
145d7, forruchongart 145d8, perf. pass. sg. forruchongrad 34d4, 102d15,
VN forngaire, forgaire 62c5, dat. sg. forgairiu 100b24, forngairiu 120c8,
dat. pl. forgairib 66c3, 103d27; fordingair (signifies, expresses): Milan:
pres. ind. 3pl. (rel.) fordingrat 37a14; foacair (proclaims, announces):
Wrzburg: pres. ind. 3sg. rel. with infix. pron. 3sg. n. fodacair 11b24, perf.
1sg. with infix. pron. 3pl. fosrocurt 24a26; pass. perf. 3sg. forcrad 19b6;
frisgair (answers, replies): Milan: pres. ind. 3sg. rel. frisgair 77a15, VN
frecra(e), frecrae 35c24, 56a13, 62c13, acc. sg. frecrae 30b10, 129d6;
immfreccair (answers, corresponds): Milan: pres. ind. 3pl. immefrecrat
58a11 (2x), VN (n io) imrecrae, imfrecrae 58a11.
guidid (prays, entreats)

PIE *ghedh- LIV2 217 ( IEW 488)
CC *ged-i e/oguidid: Wrzburg: pres. ind. 1sg. guidim 30d3, guidimm 22a20,
29a19, rel. noguidimm 21a8, with rel. n and emph. pron. 1sg. nonguidimse
27d19, with infix. pron. 2pl. and emph. pron. 1sg. nobguidimse 25c29, 3sg.
guidid 27d7, with infix. pron. 3pl. and emph. pron. 3sg. m. nosn.guidsom
25b9, rel. amal n. guidess 24d19, 1pl. with emph. pron. 1pl. guidmini 25d21,
26b8, rel. intain guidmeni 4a27, ndguidemni 21d10, with suffix. pron. 3sg.
n. as object guidmit 15d18, 3pl. with infix. pron. 2pl. nobguidet 7b21, fut.
2pl. with emph. pron. 2pl. gigestesi 14c2a, perf. 1sg. with infix. pron. 2sg.
320

and emph. pron. 1sg. rotgdsa 27d19, 1pl. with infix. pron. 2pl. and emph.
pron. 1pl. robgadammarni 24d20, with dem. rel. anrogadammar
(arrogadammar?) 15c22, pres. subj. 2sg. rel. annongeiss 30b4, 1pl. with
emph. pron. 1pl. nigessamni 11a24, rel. gesme 4a27, 2pl. nigessid 26a34,
with dem. rel. annogessid 24b3, perf. 3pl. rel. conroigset 16c23; pass.: pres.
ind. 3sg. rel. guiter 30d5, pres. subj. 3sg. ciagessir 17d27, Milan: pres. ind.
1sg. rel. nonguidim 21b7-8, (rel.) -guidim 38c18, 3sg. gudid 88b15, -guid
42a4, neg. with infix. ? nandguid 59c5, rel. guides 32d5, gudes 39b6, 127a9,
1pl. (rel.) with infix. 3sg. masc nundguidem 53b19, 3pl. gute 134d8, impv.
2pl. gudid 68a15, pres. subj. 1sg. rel. nondages 21b7-8, 21b9 with infix.
2sg. nundatges 21b5, 3sg. arang 51a16, 53b27, rel. ges 39b3, 2pl. -gessid
53b19, pass. sg. rel. gesar 51a17, past subj. 3sg. rogigsed 32d5, 1pl.
nongesmais 21b1, 3pl. nogestais 125a7, congestais 131d13, araroigsitis
131d14, fut. 1sg. nogigius 46b12, with emph pron. gigsesa 47d4-5, 3sg. rel.
giges 53c3, perf. 1sg. rel. rongad 43d18, 3sg. (rel.) rogaid 43d20,
guidid46b28, 55d4, 124c10, rel. rongaid 53b26, neg. rel. narogaid 55d4,
3pl. (rel.) rogad(a)tar 46b28, withinfix. rondgadatar 131d14, verb. nec.
gessi 26b3, VN (f i) guide 32c18 (this gloss not in DIL), gude 39b2, gen.
sg. guide 62c11, dat. sg. guidi 22a5, 23d19, 95d1, dat. sg. gudi 73b1,
nom./acc. pl. gudi 24b5, 26b21 (this gloss not in DIL), gen. pl. gude 33a11;
arneget (prays): Wrzburg: pres. ind. 3pl. arneigdet 25c23, impv. 2pl.
irnigdid 22c8, Milan: pres. ind. 3pl. rel. aranneget (DIL has this as 3sg.)
61b1, VN airnigde, ernaigde 38c11, ernigde 54d7, gen. pl. ernaigde 50d7.
indaim washes
PIE *h2emH- LIV2 265 (IEW 35)
CC *-am-(i )e/oMilan: press pass. pl. innindmatar 126c17, VN indmat 126c15.
maidid breaks, bursts

PIE *mad- LIV2 421 (IEW 694f.)
CC *mad-i e/omaidid: Wrzburg: impv. 3sg. maided 7c4, Milan: pres. ind. 3sg.
(rel.) forsamoith 110d10, forsamaid 115d12, fut. 3sg. roima 89c11, neg. n
321

mema 89c11, perf. 3sg. rommemaid 127d6, 3pl. rumaith 51c9, VN maidm
33b11, gen. sg. madmae 84c6, dat. sg. maidm 85c6, domaid (bursts forth,
breaks forth): Wrzburg: perf. 3sg. rel. asatorimed 11a19.
nigid washes
PIE*nei g- LIV2 450 (IEW 761)
CC *nig-i e/odofonaig (washes, washes away): Milan: pres. ind. 3pl. rel.
dufonget 78b5, fut. 1sg. dofonu 47a19.
saigid goes towards, approaches57
PIE *seh2g- or *seh2- LIV2 520 (IEW 876f.)
CC*sag-(i )e/osaigid: Wrzburg: pres. ind. 3sg. saigid 23c11, rel. with emph. pron.
3sg. m. saigessom 22a3, pass.: pres. ind. 3sg. rel. segar 11a5: Milan: pres.
ind. 3sg. ni saig 130a4, 3pl. segait 66b5, part. dat. sg. saigthe 19a11, verb.
nec. saigthi, VN (f i) saigid, acc. sg. saigid 66b5; condeig (asks, seeks):
Wrzburg: pres. ind. 1sg. rel. nd cuintgim 14c12, 3sg. rel. condaig 8d20,
condieig 23d5, 1pl. nicuingem 14d38, impf. 3sg. rel. nd _tagad 4b2, perf.
1pl. with emph. pron. 1pl. n comtachtmarni 24b20, 3pl. rel. an oitechtatar
8a14, past subj. 1sg. ciachondesin 19d24, impv. 2sg. nacuindig 10a20, 3sg.
nacuinged 10a15, 11b18, pass. pres. ind. 3sg. rel. degar 8d20, condegar
14c11, Milan: pres. ind. 3sg. condaig 30b8, 35c21, 1pl. -chondegam 107c8,
3pl. neg. nicuinget, pass. sg. (rel.) degar 48a7, pl. cuinchetar 123c4, impv.
2sg cuindig 102c5, pres. subj. 3sg. -cuintea 51a18, 3pl. condesat 46c13,
pret. 1sg. conaitecht 59c3, 2sg. -comtacht 60b20, 3sg. comtacht 35c26,
(-comt)acht 54b30, -comtacht 59c3, conaitecht 36b5, 132d15, 3pl.
conaitechtatar 44d27, pret. pass. nicomtacht 123c3, verb. nec. cuintechtai,
chuintechti 92a17, 93a8, part. (io, i) cuintechtae, dat. sg. cuintechtai
59b13, VN (i) cuingid, acc./dat. sg. chuinged 39a6, dat. sg. chuingid 56b15,
56c15, 62d12, 98c1, acc. / dat. sg. imchuinchid 59d3; doosaig (supports,
sustains; DIL has the verbal form here under fo-foscai): perf. 3pl. with
57
saigid is traditionally categorised as a BII, but since there are no 1sg., 1pl. or 3pl.
attested, it strictly speaking is not possible to tell whether this was BI or BII.
322

infix. pron. 1sg.: domroisechtatar 17c1; doroich (reaches, comes to):
Milan: pret. / perf. 1sg. doroacht 98d9, part. (DIL gives this with a question
mark) torachta, nom. pl. thorachti 18a11, VN toracht, dat. sg. toracht
51b30, 54c32, 56b19-20, 68c3, 90c7, 92c15, gen. pl. toracht 29a14, 81b10,
dat. pl. torachtib 34c3, 55b4, dosaig (seeks, searches): Wrzburg: pres.
ind. 3sg. dosaig 15a22, 23c23, Milan: VN degaid, digaid 138c18; indsaig
(approaches, goes to): Wrzburg: pres. ind. 3pl. rel. with infix. pron. 2pl.
atobsegatsi 14d37; rosaig (reaches, comes to): Wrzburg: past subj. 1sg.
rel. roissinn 26d17, Milan: pres. ind. 3sg. with infix. 3sg. rasaig 102a8,
connachroig (This form not in DIL) 39c7, fut. 3pl. neg. niroisset 74a11, sec.
fut. 3sg. with infix. 3pl. nisroissed 39c34, pret. 3sg. rosiacht 55d2.
slaidid strikes, slays

PIE *sleHd(h)- LIV2 - (IEW -)
CC *slad-i e/oslaidid: Milan: VN slaide 16b7, dat. sg. slaidi 91b9; doslaid
(strikes): Milan: pres. ind. 3sg. duslaid 140b5.
snigid pours down, flows

PIE *snei gh- LIV2 573 (*sneigh- IEW 974)
CC *snig-i e/o- or *snig-e/oMilan: pres. ind. 3sg. rel. sniges 89c15.
323
7.3 BIII
asboind refuses
PIE *bhedh- (LIV2 82f. IEW 150-2)
CC *-bu-n-d-e/oasboind: Milan: subj. 3sg. with infix. 3sg. conimop 20b6; opaid
(refuses, declines): Milan: pres. subj. 3sg. -oip 42a2.
boingid breaks, smites

PIE 2. *bheg-,*bhegh- (LIV2 84f., LIV2 85f. IEW 152f.)
CC *bu-n-g-e/oconboing (smashes, breaks): Wrzburg: pres. ind. 3sg boing
4d15, Milan: past subj. 3sg. with infix. 2pl cochotabosad 18a7, VN
combach, dat. sg. diacombuch 126a7.
dingid presses, thrusts

PIE *dhei h- (LIV2 140f, IEW 244f.)
CC *di-n-g-e/oarutaing (builds up, restores): Milan: pres. ind. 3sg. rel. arutaing
64c20, fut. 2sg. arutaissiu 56a11, verb. nec. airdachtaib, erdachtaib 130b10,
VN airtach, ertaig 118c7, anertach 121b7; conutainc (builds,
constructs): Wrzburg: pres. ind. 3sg. conutuinc 10b28, rel. int conutuing
(MS conutunig) 8d2, with infix. pron. 2pl. and emph. pron. 2pl.
cotofutaincsi (MS cotofutaircsi) 8c16, perf. 3sg. rel. rotaig 33a5, and with
infix. pron. 3sg. n. idrotig 33a2, impv.. 3sg. comtangad 31c15, Milan: pres.
ind. 3sg. with infix. cotautaing 36a18a, perf. 3sg. rotaig 40d5, pass. sg.
conrotacht 48d27, part. cumtachtae 57c2, VN (n o) cumtach, acc. sg.
cumtach 72a5, dat. sg. cumtuch 94b11, 131c8, dat. pl. chumtachtaib
102d10; fording (crushes, oppresses): Milan: pres. ind. 3pl. fordengat
29a13, pass. sg. fordengar 57d7, impf. 1sg. forndinginn 115a16, pres. subj.
pass. pl. forndiassatar 39b12, perf. 2sg. forrudedach 96c17, 3pl.
forrudedgatar 63c3.
324

dlongid splits, cuts
PIE *d(h)leg(h) or *d(h)le(h)- (*delgh- LIV2 113f., IEW 196)
CC *dlu-n-g-e/oin-dloing (cleaves, parts): Milan: pres. ind. 3sg. rel. asdloing
48c32, fut. 3sg. neg n indail 96a8, VN (n o) indlach, acc. sg. indlach 32a1,
40d16, 64a1, dat. sg. indluch 129d8.
foloing bears, supports

PIE 2. *leg- (LIV2 416, IEW 685f.)
CC *-lu-n-g-e/oWrzburg: pres. ind. 3sg. rel. foloing 29d17, 1pl.. rel. follongam
14b15, 2pl. with emph. pron. 2pl. nifulngidsi 8c3, fut. 1sg. with emph. pron.
1sg. folilussa 23b25, 3pl. rel. folilsat 25d15, sec. fut. 3pl. nifolsitis (MS
nifositis) 15a20, pres. subj. 1pl.. arafulsam 14c2a, perfect. pres. subj. 1pl.
rel. ndfochomolsam 14b15, 2pl. rel. nadfochomalsid 11b2, impv. 2pl.
fulget 20c5, Milan: pres. ind. 3sg. annadfulngat 122a8, nachful 129a5, rel.
nad fulaing 77d3, 77d6, 1pl. n fulgam 77d6, 77d7, rel. neg nad fulgam
77d7. impv. 3pl. fulgat 60d4, fulngat 60d7, pres. subj. 1sg. fuls 33a2,
-fullos 58c12, folls 62b12, 3sg. follo 32d18, foll 100d10, (rel.) -ful 32d5,
57d15, 3pl. follosat 69a7, 118a12, past subj. 3pl. folostais 104c5, fut. 3pl.
folilsat 80a13, pass. sg. fulilastar 109b02, sec. fut. 1sg. fulilsain 73d1, pret.
1sg. forroiblang 107d13, perf. 3pl. folongat 35a3, focoimlachtar 47c6, part.
fulachtae 88b12, VN fulach 22d9, 32d4, fulang (m o, in DIL given as later,
but already found in Wb); inloing (joins, unites): Milan: part. ellachtae,
nom. pl. ellachtai 84a6.
rondid colours, dyes

PIE 1. *(h1)redh- (LIV2 508f., IEW 872f.)
CC *ru-n-d-e/oforoind (darkens, stains): Milan: pres. ind. pass. sg. -forondar
35d8, pres. subj. pass. pl. furastar 15b11, pret. 3sg. fororaid 51a23.
tongaid (swears, takes an oath)

PIE *h2legh- or *h2leh- (IEW 687)
325

CC *tu-n-g-e/o- 2. *lu-n-g-e/otongaid: Wrzburg: perf. 3sg. docuitig 33d10, past subj. 3sg. rel.
taratoissed
33d10,
Milan:
impf.
3sg.
with
infix.
dodfongad
(Hypercorrection, cf. McCone 1985: 101) 36a21, rel. nothongad 36a20, past
subj. 3sg. corresp. to perf. daradochtaised 78a4, perf. 3sg. ducuitich 78a5-6,
VN luige, lugae 118c5; fristoing (forswears, abjures): Milan: VN (n o)
fretech, acc. sg. fretech 65d3, dat. pl. frit(ch)ib 44c14.
326
7.4 BIV
benaid beats, strikes
PIE *bhei H- (LIV2 72, IEW 117f.)
CC *bi-naWrzburg: perf. 2sg. rel. diaruba 13d25, impv. 3sg. benad 20c8,
pass. pres. ind. 3sg. benir 4d15, Milan: pres. ind. 3sg. benaid 46d4, pass. sg.
rel. benar 93d16, rel. nadfedar58 39d7 pl. ni bentar 91b3, pres. subj. pass.
pl. ciabetir 54a17, part. (io, i) bthe, nom. sg. bithe 45d6, VN bth, acc. /
dat. sg. bithin 59b9, VN (n n) bim, nom./acc./gen. pl. bemmen 39c17;
adcuimben (strikes, cuts): Milan: pres. ind. 3pl. adcumnet 77a1;
ardben (cuts off, slays): Milan: part. airdbide 135d6, acc. sg. fem.
airdbidi 49a25, gen. pl. airtbide 33d6, airdbidi 55c15, VN (n io) airdbe
41d5, gen. sg. airdbi 41d2, dat. sg. airdbiu 14a2, 55b17, 80c13, 100a6;
dofuiben (cuts, cuts off): Wrzburg: pass.: perf. 3sg. with infix. pron.
2pl. and emph. pron. 2pl. doforbadsi (for do-b-forbadsi) 20a15, 3pl. with
prep. and rel. asatorbatha 5b39, Milan: pres. ind. pass. pl. dufuibniter 24c6,
impf. 3sg. dufailced 35c1-2, pres. subj. pass. sg. codufobither 2a10, past
subj. 3sg. codufubath 35c1-2, 3pl. rel. dufubaitis 92d6, fut. 3sg. dufobi 96a7,
perf. pass. sg. duforbad 92d4, part. (io, i) tbaide, dat. pl. tobaidib 48d13,
VN tbae, nom./acc. sg. tobae 26c2 (2x), dat. sg. robu 37d8, nom. pl. tobai
92d6, doim(m)diben (cuts away, shortens): Wrzburg: pres. ind. 1sg.:
doimdibnim 17d22, Milan: impf. pass. sg. dunindbithe (Note Thes.: the
second d is over the line; leg. duimdbithe, Sarauw, citing timdhibhe .i.
laghdhughadh OCl., and doimdibnim gl. parco (=nomisligur), Wb. 17d
22) 88a11; dororban (comes, arrives)59: Wrzburg: pres. ind. 3sg.
ntorban 12b32, nitorban 12b33, n torban 17d18, Milan: pres. ind. 3sg.
dororban 62a20, dufrban 61a22, 3pl. durorbanat 43b5, impf. pass. sg.
58
Reading following DIL; Thes has this as nadfendar with note MS. nadfedar: corr.
Thurneysen, here f is for bh, and translates it as they are not stricken. The reference from
Thurneysen cannot be traced, but he seems to take it as from benaid rather than feidid.
There seems no reason for this emendation.
59
DIL: verb of uncertain form and analysis. Ped ii 445 postulates a compound *to-ro-ben(b-) influenced in meaning by a hypothetical *to-for-fen-, Thurneysen a verb *to-for-b- in
meaning arrives (GOI 551) and for the meaning profits a back-formation from torbae (toro-withbae profit) (852), both influenced in form by benaid and its compounds.
327

duforbaithe 31c11, pres. subj. 1pl. with infix. dundrbiam 105b6, 3pl.
dofrbiat 27a10, perf. 1sg. -ruthrba 44b29, 3sg. dororbai 123d5, 3pl. rel.
hrothorbatar 44b29, VN (n) torbae 42b13, 102d4, 120c7, acc. sg. torbae
41a10, eterdiben (destroys): pres. ind. 3pl. etirdib(n)et 15b6, impf. 3pl.
rel. etirdibnitis 99a7, pres. subj. 3pl. with infix. 1sg. etardamdibet 44c31,
past subj. 3sg. rel. with infix. 3pl. itirndadibed 45c6, sec. fut. 3pl. with infix.
1sg. coetardamdibitis 54d14, perf. 3sg. etirrudib 123b10, foben (attacks,
impairs): Wrzburg: pres. ind. 3pl. fobenat 13b13, perf. 2pl. rel. forubid
27c27, Milan: pres. subj. 3sg. rel. fombia 15a11; foindarban (subjects
(?)): Milan: sec. fut. pass. sg. -foindarpaide 25a1; frisben (heals, pays):
Milan: pres. ind. 3sg. frisben 125c4, VN frepaid 123c2, dat. sg. frebaid
58a4;imm-dben (excises, circumcises): Wrzburg: impv. 1pl. imdibnem
15b9; pass.: perf. 3sg. with infix. pron. 1sg. immumruidbed 23d30, 3sg.
niroimdibed 18d9, 23d25, with interrog. inroimdibed 2c8, rel. immeruidbed
18d9, past subj. 3sg. n imidibthe 1d17, rel. nd imdibthe 1d15; indrban
(drives out, expels): Wrzburg: perf. 3sg. rel. with infix. pron. 1sg. and
emph. pron. 1sg. nachimrindarpai se 5a18; pass.: perfect. pres. subj. 3sg.
with infix. pron. 2sg. arnachitrindarpither 5b33, past subj. 3pl. with emph.
pron. 3pl. inrpitissom 5c1, Milan: pres. ind. 3sg. with infix. 3pl. atarban
65a14, pass. sg. innarbanar 14c16, inrbanar 73a20, pl. innrbantar 15c5,
impv. 3sg. atatirbined 86c10, pass. sg. with infix. 1sg. nachamindarbanar
56a22, pret. 1sg. inrarba 46b10, 3pl. inrarpatar 23d8, VN (n io) indarba,
nom./acc. sg. indarbae 23c8, 41c10, acc. sg. indarbae 30d13, gen. sg.
indarpai 67a9, dat. sg. indarbu 48d27, 127b9. ocuben (touches)60: Milan:
pres. ind. 3sg. with infix. 3pl. n conrocmi 76a12, pass. pl. rel. ocubendar
54a12, nad ocmanatar 54a12, fut. 3pl. ocubiat 126b12 pass. sg. ocubether
53b17, perf. 3sg. occuroae 98d8, VN (f a) ocmad, gen. sg. ocmaide 39a10.
adfen (repays, requites)

PIE *i eh1- (LIV2 695 IEW 1120-2)
CC *-i-na-
60
Schumacher has this as a compound of benaid while DIL has it as a compound of the
substantive verb.
328

adfen: Wrzburg: past subj. 3pl. rel. with infix. pron. 3sg. n. and
emph pron. 3pl. arandathidisom 21d11, pass. fut. 3sg. adfether 20b7;
forfen (finishes, completes): Wrzburg: pass. pres. ind. 3sg. \forbanar
14d27, Milan: pres. ind. 3sg. forfen 64c2, pres. subj. 3sg. -farfia 55c16, 3pl.
forfiatsom 23a19, pass. sg. (perfective) burorbaither 15a6, pret. 3sg.
forruchui 121c24, part. (io, i) foirbthe 124c4, dat. sg. fem. foirbthi 39a2,
part. (io, i) forbaide 42c18, dat. pl. forbathib 105a10, VN forba, dat. sg.
forbu 15a4, 15a5, 50c14, 55c21; immfen (hedges round): Milan: part
imbide, imbithe 128a9.
renaid sells, barters

PIE *h2rei H- (IEW 60f.)
PIE *ri-narenaid: Wrzburg: perf. 3sg. with infix. pron. 1sg. romrir 3c38,
pres. subj. 3pl. niriat 28c2, Milan: pres. 3pl. rel. rend 120d1, verb. nec.
rthi 36a37, ar-ren (pays on behalf of another): Milan: pres. ind. 2pl
errenaid 20c2, as-ren (pays away, expends): Wrzburg: fut. 1sg. with
emph. pron. 1sg. asririusa 18a14, 3sg. asriri 25b6; pass. fut. 3sg. asrirther
1c3, Milan: pres. ind. 2sg. asrenai 44a6, 3sg. asren 36a30, 127a18, impf.
3sg. asrenad 32b10, pres. subj. 3sg. asria 127a18, past subj. 3sg. asriad
36a29, fut. 3sg. asriri 30c13, VN (f i) ric, raic 59a15, dat. sg. ric 127a1.
sernaid (arrays, disposes)

PIE *sterh3- (LIV2 599f., IEW 1029f.)
CC *star-nasernaid: Wrzburg: pass. past subj. 3sg. rel. sernte 18c8, Milan:
pres. ind. 3pl. -sernat 31a19, impv. 2sg. srnn 56c12, part. dat. sg. srithiu
31c18, VN (f ) sreth, dat. pl. srethaib 20d9; consern (spreads out, strews):
Milan: pres. subj. 1pl. consernam 35c36, impv. 2pl cosrid 68a15, 3sg.
cosrad 124a5, verb. nec. comsrithi (punctum delens over s) 16a13; dosern
(brings up?): Milan: part dirathe 33c4.
crenaid buys, obtains

PIE *kreih2- (LIV2 395f., IEW 648)
329

CC *kri-nacrenaid: Wrzburg: pres. ind. 3sg. rel. crenas 29d23, 3pl. rel.
nadchrenat 10b7, doaithchren (buys back, redeems): Wrzburg: perf.
3sg. doradchiir 2b9, with rel. n dorraidchiir 32d10, impv. 2pl.
(tai)dchrenid 27c27, Milan: impf. 3pl. pass. doaithchretis 123c10, perf. 3sg.
duarchiuir 73b5, pass. pl. dorathchratha 125b9; fo-cren (buys,
purchases): Milan: VN fochraic, acc. sg. fochraic 91b5, 91b6, gen. sg.
fochraice 15d6, 44d18, fochricce 54a14, acc. pl. fochraicci 90c14, dat. pl.
fochraiccib 18c5.
ernaid bestows, grants

PIE *perh3- (LIV2 474f., IEW 817)
CC *ar-naWrzburg: perf. 3sg. rel. ror 17b13, with infix. pron. 1pl. ronnr
20d11, perfect. pres. subj. 3sg. with infix. pron. 3sg. n. ra ra 25a31;
pass.: fut. 3sg. ebarthir 32a27, perfect. pres. subj. 3sg. ro erthar 18b23,
perfect. past subj. 3sg. roerthe 27a16, Milan: fut. 1sg. with suff 3sg. ebarthi
46b12, VN rath 89c15.
glenaid (adheres, cleaves)

PIE *glei H- (LIV2 190, IEW 362f.)
CC *gli-naMilan: pres. subj. 1sg. coniglu 86b8, 3pl. rel. glete 127b19, fut. 3pl.
giulait 65b7, perf. 3sg. roguil (DIL rogul) 98b8.
lenaid remains, continues

PIE *h2lei H- (LIV2 277f., IEW 662)
CC *li-naWrzburg: pres. ind. 3sg. lenid 5c16, 3pl. lenit 29a23, fut. 3sg. with
dem. rel. alliles 10a5, Milan: perf. 3sg. rolil (note Thes.: MS. rolin, but
this form is interpreted by McCone (1985: 93) as an innovative s-pret. that
still retains the vocalism of the suffixless form.) 54d7, 3pl. roleldar 96c13,
VN len(a)main, dat. sg. lenamain 101b1.
330

mairnid betrays; plays false
PIE *merh2- (LIV2 440, IEW 735f.)
CC *mar-naWrzburg: fut. 3pl. .i. nimmerat 30c20, Milan: past subj. 3sg.
conumerad 24c20, perf. 3pl. rel. romertatar 75d5.
tlenaid takes away, steals

PIE *telh2- (LIV2 622f., IEW 1060f.)
CC *tli-naMilan: pret. 3sg. with infix. pron. ciasidrogat 132a10.
331
7.5 BV
adgnin knows, recognises
PIE neh3- (LIV2 168-70, IEW 376-8)
CC *-gni-naadgnin: Wrzburg: pret. and perf. (with pres. ind. meaning) 1sg.
rel. and with emph pron. 1sg. adgnsa 12c13, 3sg. rel. adgeun 12c13, and
with infix. pron. 3sg. n adidgein 12c13, 1 pl. adgenammar 14d28, Milan:
fut. pass. sg. atatgentar 121d22, pret. 2sg. asgen 140b3, 3sg. with infix. 3sg.
ni naithgeuin 52 line 17, VN (n io) aithne, dat. sg. aithgniu 42b13,
etargnn (finds out, learns): pres. ind. 3sg. (rel.) -etarguin 42c10, 42c15,
fut. 3pl. etirgnat 68c20, etirgenat 73a1, pret. (often with pres. meaning)
1sg. etiringn 94b12, 3sg. (rel.) etirgein 24a19, etirgn 79a11, VN (n io)
etargne, etarcnae 58a20, 103a9, acc. sg. etarcnae 42b27, gen. sg. etarccnai
14c9, etarcni 53c5, 59a20, dat. sg. etarcnu 9a13, 27d5, 42b13, 42c11, 42d2,
46c20, 63a8, dat. pl. etarcnaib 72c6, ingnin (recognises, knows): Milan:
pret. 3sg. ningeuin 52 line 19, n inguin 69a15, verb. nec. ingnaidi 63a9,
105c10, VN engnae, ingnae, 26b8, dat. sg. engnu 14c19, 14d7, ingnu
140b2, gen. sg. / nom. pl. ingnai 89b8, *remiasgnin (knows beforehand):
Milan: past subj. 3pl. remiergnaitis 19b8.
arachrin decays, fails

PIE *kreh1(i )- (LIV2 366f., IEW 945f.)
CC *kri-niWrzburg: pres. ind. 3sg. with rel. n (and elision of infix. pron. 3sg.
n.) arinchrin 32c10, 3pl. arinchrinat 27b1, Milan: pres. ind. 3sg. arachrin
57c12, with infix. pron. 3sg arindchrin 57a10, 3pl. aracrnat 73c2, past
subj. 3pl. arindrochrietin 85d1, fut. 3pl. arachiurat 59b9, perf. 3pl.
arrceoratar 26d6 (Thes gives translation of this gloss under 7), part.
airchraide, erchride 26d12, VN (n io) airchra, acc. sg. erchrae 39c34,
119b9, 122a17, dat. sg. erchru 30d27, 31a4, 58b10.
332

dolin flows; swarms
PIE *pleh1- (LIV2 482f., IEW 798-800)
CC li-niMilan: pres. ind. 3sg. dolin 2a9, 3pl. (rel.) dulinat 56a14, 83c7, impf.
3sg. dolinad 133c7, pres. subj. 3sg. dulin (Note Thes.: rather the subj.
dulina to translate the infinitive) 68b11, fut. 3sg. dolili 30c13, perf. 3sg.
dorulin 64c18.
333
7.6 Hiatus verbs

Note that in a compound verb with two or more preverbs, it is meaningless
to speak of a hiatus verb, since the root is never stressed and therefore the
hiatus will never show. I classify a verb as a hiatus verb if its simplex is, or
if a hypothetical simplex would be such.
adc sees
PIE *kei s- (LIV2 381f., IEW 637)
CC *kis-e/oadc: Wrzburg: pres. ind. 3sg. adci 12c11, rel. adch 4a25, adchi
2a2, rel. with infix. pron. 3sg. n atch 6c17, rel. with anaph. pron. 3sg. m
atchside 24c3, atchiside 25a37, with infix. pron. 2pl. and anaph. pron. 3sg.
m atobcside 25a26, 1pl. adciam 6a30, with emp. pron. 1pl. adciamni
12c11, pass. sg. adcither 1b15, with dem. rel. anadchither 12c12, with nad
and rel. n ndnacastar 25b28, with cini and infix. pron. 3sg. n cinidaccastar
26a12, sec. fut. 3pl. rel. adcichitis 7a2, pret. 3pl. (with elision of infix. pron.
3sg. n) ni accatar 26b11, pres. subj. 3sg. rel. with infix. pron. 3sg. n
didaccadar 16d6, 2pl. maadced 11b22, past subj. 2pl. adcethe 19b6, pret.
3sg. rel. adchess 23c11, Milan: pres. ind. 2sg. innadnaccai 17b17, caniaccai
25b14, 3sg. n aicci 94c3, 1pl. with infix. 3sg. asidciamni 2b4, 1pl. niaccam
111d1, rel. anadciam 16c5, anadchiam 112b13, rel. adciam 42b22, 112b2
(x2), 3pl. adciat 103b13, rel. anadciat 69b2, rel. nadnacat 93d14, pres. subj.
3sg. coniaccadar 53a6, rel. arnacathar 68b9, 3pl. adceter 3a4, pass. sg.
-accastar 50a5, past subj. 3sg. adcheth (Note Thes.: The aspiration is
irregular, cf. McCone 1985: 96) 38c9, 3pl. -naccaitis 32d12, pass. sg.
coadcethe 77d8, fut. pass. sg. with infix. 2sg. atatchigestar 59c12, pret.
pass. sg. adchess (Cf. McCone 1985: 98) 96d1, perf. 1sg. atchondarc
113b6; addci (regards, looks at): Milan: pres. subj. 2sg. addicider
43a19; doccai (looks at, beholds): Wrzburg: pres. ind. 1sg. n dccu
24a13, 3sg. with infix. pron. 1pl. donnicci 9a4, impv. 2sg. dcce 10c20,
with emp. pron. 2sg. diccesiu 31b19, 3sg. dcad 18b6, nadecad 23c16, 2pl.
dcith 20c18, Milan: pres. ind. 3sg. docai 61a8, n dci 69a16, impv. 2pl.
334

dcaid 55c1, pres. subj. 2sg. denecaither 73c11, with infix. 3pl.
diandadercaither 102b13, fut. 3sg. (rel.) ducigi 111c13, perf. 3pl. (rel.)
dorecachtar 53b11, dorecatar 53d17, rel. with infix. 3sg. indorecatar
53b11,VN (f n) dicsiu, acc. sg. deicsin 56d15, gen. sg. dicsen 50d11, dat.
sg. deicsin 47d7, 55a8, 56a17, 82a7, 88b4; doincai (looks, gazes (at)):
Milan: perf. pass. sg. dorencanas 33c18; eterdccai (introspects): Milan:
pres. ind. 3sg. etirdcai 61a8; frisacci (looks forward to, expects):
Wrzburg: pres. subj. 2pl mafrisaiccid 25d24, Milan: pres. ind. 2sg. rel.
frisaiccai 68a8, 3sg. rel. frisnaccai 53b23, frisaccai 144a2, -frescai 62d7,
1pl. frisaccam 134a5, 3pl. neg nadfrescat 69b3, impf. 1sg. frisaccinn
133b11, 3sg. confresced 136d1, 3pl. rel. frisnaiccitis 68a9, pres. subj. 1sg.
frisaccar 49d3, 3pl. frissaccat 39d27, 124a2, rel. frisnaccat 45d17,
frisnaccatar 124a3, perf. 1sg. frisracacha 47a8, 3sg. rel. hirufrescachae
44c9, rel. inrufrescechae 44c19, with infix. frisracachae 68a7, 3pl.
-rufrescachtar 26b25, -rurescesset (Cf. McCone 1985: 22) 34c11,
-rufrescechtar (but this form not in DIL) 34d17, nirufrescisset (Cf. McCone
1985: 92) 72c13, fritracatar 131c9, part. frescastae 68a8, dat. sg. frescestu
36b9, dat. sg./acc. pl. nephrescastu 56d1, dat. pl. frescastaib 40b10, VN
frescisiu, frescissiu 38a13, 97d10, acc. sg. frescsin 23c7, frescissin 53b22,
69b3, 90a11, gen. sg. frescsen 58c11, dat. sg. frescsen 62c17; immacci
(looks after, regards): Wrzburg: pret. 1pl. with infix. pron. 1pl.
nmunaccammar 18d3, Milan: pres. ind. 3sg. with infix. 3sg. immaaccai
114a15, rel. immanaccai 17b6, nadnimcai 114a15, verb. nec. imcasti 18d22,
VN imcaisiu, acc. sg. imcaissin 105a1, dat. sg. imcaisin 28b3, 81d3, *remad-ci (looking before, providence): Milan: VN remcaisiu, remcaissiu
91c1, acc. sg. remcaisin 19c7, remcaissin 89a4, gen. sg. remcaissen 27d10,
40b15, remidci (foresees, provides (for)): Milan: pres. ind. 2sg.
remideci 55d26, 3sg. remideci 55d25, pret. 1pl. rel. annadremdercachmar
80b14.
adcomla joins, unites

PIE *leh1- (LIV2 399, IEW 666, IEW 682f.)
CC *-la-i e/o-
335

adcomla: Wrzburg: pres. ind. 3sg. adcomla 3c13, with infix. pron.
3sg. m. atcomla 16a19, with emp. pron. 3sg. atcomlasom 4a10, atcomlasom
25b16, Milan: pres. ind. pass. pl. (rel.) adchomlatar 35b2, perf. 3sg.
adrochomul 58b12, part. (io, i) accomoltae, nom. pl. acomaltai 44c1, VN
(n o) accomol, gen. sg. acumail 58b3, dat. sg. (?) accumuil 15b8, dat. sg.
accomul 35d26, 121c6; doinla (gathers, collects): Milan: pres. ind. 3sg.
doinola 25b3, pres. subj. pass. sg. doninoltar 24a8, pl. -tinoltar 55b5, perf.
3sg. dorinl 51a21; foaccomla (subjoins): Milan: pres. ind. 3sg.
fo(a)ccomla 30b10; immtecmalla
(assembles, collects together):
Wrzburg: perfect. pres. subj. rel. commaterchomla 14a2.
adcota gets, obtains

PIE *steh2- (LIV2 590-2, IEW 1004f.)
CC 1. *t-i e/oadcota: Wrzburg: pres. ind. 3pl. adcotat 6a6, pret. 1sg. with dem.
rel. and emp. pron. 1 sg. anadchodadossa 7a16, pres. subj. 1sg. in ds
intim techt? 1a6, 2pl. rel. contid 22d20, conetid 22d24, 3pl. rel. \etat
31b18, pass. pres. ind. 3sg. nietar 17d27, rel. with nad nd etar 16b11,
Milan: pres. ind. 3sg. adcota 20a13, 3pl. adcotat 87c3a, pass. sg. rel.
adchotar 56b15, 92b9, 128d18, trisnetar 126b19, trisantar 126c1, pres.
subj. 1sg. (rel.) adcot 60c6, 92a17 (2x), 3sg. coadcota 77b12, conneta
32d15, past subj. 3sg. coadcotad 39c5, 3pl. conetaitis 130b6, condetaitis
125a4, pass. sg. adchotate (DIL: adchotatae) 100c23, fut. 3sg. nietada
129b5, pass. sg. aracotar (Cf. McCone 1985: 99) 118d10, sec. fut. pass. sg.
taste 43d20, pl. neg nadtatais 124d9, pret. 1sg. adcotadus 44c18, netade
111b20, 3sg. adcotade 89c1, 1pl. adchotadsam 92c5, 3pl. adchotatsat 57a2,
adcotatsat 67b10, with infix. 3sg. atchotatsat 54a9 rel. trissantatsat 57a3,
adchotatsat 123a9, pret. pass. adcotad 43d24, part (io, i) ttae, tae, 129a3,
tite 118d11, dat. sg. ttu, iarneit 119a5, 59b13, dat. pl. tib 69a5, VN t, it
127a9 dat. sg. dut 118a16, duit 118d5; adnessa (lampoons, censures):
Milan: VN ainsem, gen. sg. ainsemo 34a18; arossa (awaits): Milan: pres.
ind. 3sg. arossa 96a11, with infix. pron. araossa 134d7, rel. with infix.
pron. aranossa 15b16; art (is in store for, impends): Wrzburg: pres.
ind. 3sg. rel. arath 10b3, arth 30d13; with infix. pron. 1pl. arunta
336

25d25, perf. 3sg. arrobe 27d13, with infix. pron. 1pl. hre arinrobe 29d22,
Milan: pres. ind. 3sg. arabitis 43d15; connessa (tramples under foot):
Wrzburg: pres. ind. 3pl. connessat 1d7, fut. 3sg. nesfea 4b15, Milan: pres.
ind. 1sg. with infix. pron. 3sg. cotneissiu 126c17, 3sg. rel. -chomainsea
42b28, pass. conestar 126c17, impv. 2sg. comainse 22b2, perf. 3sg. corunes
(Note Thes.: leg. conrunes? cf. 45c 2) 102d5, VN comainsem 74b17, dat.
sg. chomainsem 36a34; doesta (is absent, wantin): Wrzburg: pres. ind.
3sg. rel. infix. pron. 3sg. n. an dodesta 26d8, with infix. pron. 3sg. n.
andudesta 1a9, 14a33, 23d17, andudsta 12b5, desta 26d8, with infix. pron.
3sg. n. madudesta 25a30, perfect. pres. ind. 3sg. with infix. pron. 3sg. n.
manidtesarbi 28d31, 3pl. tesbanat 11d11, perf. 3sg. tesarb 17d2, Milan:
pres. ind. 3sg. doesta 35d20, duesta 71c19, -testa 65d6, rel. nad desta
94c10, perf. 3sg. tesarbae 34c16; donessa (spurns, despises): Milan: pres.
ind. 3sg. dunessa 36a12, dernessa 129a14, fut. 2sg. donesbe 112c3, perf.
1sg. -darunesus 36c1.
adno entrusts, commends to

PIE *ne- (LIV2 455f., IEW 767)
CC ?*-ne-e/oWrzburg: perf. 3sg. rel. adroni 29d29 (2x), pres. subj. 3 sg. rel.?
adnoodur 93b21.
asag(si) desires, wishes

PIE *es- (LIV2 166-7, IEW 399-400)
CC *-gus-e/oasag(si): Milan: pres. ind. 3sg. rel. asagusi 61b17, 3pl. rel. adguset
56b21, rel. assaguiset 61b17, impf. 3pl. rel. adgustis 69a2, pres. subj. 1sg.
asag 58b9, 3sg. assagusea 129b5
perf. 3pl. rel. anasruguset 48a17, part. ecguiste, nom. pl. ecguisti 65b2,
comp. ecguistiu 90c7; *remiuccu (chooses beforehand, pre-elects):
Milan: past subj. 3sg. remenuicsed 47c12; do-goa (chooses, elects):
Wrzburg: fut. 1sg. dogega 23b33, 3pl. with rel. n dongegat 30d8, pret. &
perf. 3sg. doroigu 5b12, hre doroigu 4c16, int orroigu 5b1, dorigu 4b31,
with infix. pron. 2pl. dobrigu 26a24, 2pl. doroigaid 20a4, 3pl. doroigatar
337

5b12, Milan: perf. 1sg. with infix. 2pl. dobroiga 103c15, 2sg. duroiga
138b8, perf. 3sg. dorogad 124c13, pass. sg. doroigad 123a14, part tuicse,
dat. sg. superl thuicsimem 71b21, VN toga, dat. sg. togu 123a13, 131c13.
aslu escapes, absconds

PIE *leH- (LIV2 417, IEW 681f.)
CC *-lu-e/oaslu: Wrzburg: pres. ind. 3pl. rel. asluat 29a3, perf. 1sg. rel.
asrulus 17d16,
perfect. pres. subj. 3sg. n rohla 30a10, pres. subj. 2pl. with ci and infix.
pron. 3sg. n. ciatloid 26a2, Milan: pres. ind. 3pl. asluat 44a17, impf. 3sg.
asluad 54c21, past subj. 1sg. coaslinse 59a1, VN (m u) ld, elud 86b11,
acc. sg. lud 92b4, gen. sg. luda 86c1, dat. sg. lud 44d30; ascomlai
(departs; escapes): Milan: perf. 3 sg. asruchumle 17b2. VN escomld,
acc. sg. escomld 61a14, gen. sg. escumluda 62b1 dat. sg. escumlud 15a3,
escumld 126b1.
bad dies
PIE *geh2- (LIV2 205, IEW 463f.)
CC *bbad: Wrzburg: fut. 3pl. rel. bebte 25b16, perf. 3sg. with rel. n
rombebe 10c11, rombebe 3b3, Milan: impf. 3sg. nombaad 23d13, VN bs,
bas 54a20, acc. sg. bas 16d2, 52 line 20; *remiba (dies prematurely):
Milan: pres. subj. 3sg. remimbaat 125d9; ardbdai (extinguishes,
quenches): Wrzburg: perf. 3sg. rel. arrodibaid 11a19, Milan: pres. ind.
3sg. (rel.) ardibdai 48c33, perf. 3sg. arrudibaid 99a2, verb. nec. dat. pl.
airtbidib 116d4.
brud breaks in pieces, smashes

PIE *bhres- (LIV2 97, IEW 171)
CC *brus-i e/o- (note: not in Schumacher)
Milan: VN brd, acc. sg. bruud 34a27, gen. sg. bruithea 34a26;
conbru (pounds, smashes): Wrzburg: pass. perf. 3pl. nabri(thea)
5b29, Milan: past subj. 3sg. conap bruithea 36d27.
338
cetab feels, perceives

PIE *bheh2- (LIV2 98-101, IEW 146-50)
CC *b-ii e/ocetab: Wrzburg: impf. 1sg. cetabiinn 2c8, Milan: cons. pres. 3sg.
citamb 36b1, pres. subj. 3sg. citamb 36b1, citab 68d15, perf. 1sg.
citaroba 44b22, VN ctfaid, acc. sg. ctbaid 68c14, dat. sg. chtbaid 61a7,
chetbaid 61c3; dodoichb (comes): Milan: fut. 3sg. dundoichfea (DIL
states -doichfia (sic MS)) 39d26, 3pl. -todoichfet 28b6.
cid cries, weeps

PIE ?2. *kei - (LIV2 321, IEW 542)
CC *ki-i e/oWrzburg: pres. ind. 3pl. rel. ndchiat 10b6, past subj. 3pl. ncetis
10b6.
clod turns, turns back

PIE *kleu- (LEIA C-121-2)
CC *klou-ie/oclod: Wrzburg: Pass. perf. 3sg. rocload 3b1: Milan: pres. ind. pass.
sg. cloithir 16b3, clither 30c1, past subj. 3pl. with infix. 3sg. tidcloitis
112b20, fut. pass. sg. with infix. 2pl. nobcloifether 67a11, perf. 2sg. roclois
43d18, 3sg. rel. rochli 37a5, part. cloithe 32b2, nom. pl. nephchloithi 83b7,
dat. pl. clothib 67b13, 131d11, 135c2, VN cld, cloud 70a7, dat. sg. chloud
56b16; conerchlo (leads, drives): Wrzburg: pass.: pres. ind. 3pl.
conirchloiter 4a14, pres. subj. 3pl. cochonerchloatar 14c6, Milan: pres. ind.
3sg. conerchli 33d16, pass. sg. erchloither 18d19, part (io, ia)
comerchloithe, gen. sg. comerchloithi 54b16, 80b4; conimchlo (changes,
alters): Milan: VN (m u and o) comchloud, choimchlod 109d04, gen. sg.
coimchloda 63d3.
concna chews, masticates

PIE *kneh2- (LIV2 365, IEW 560)
CC *kn-i e/o339

Milan: VN (m u) cocnam 75b7.
cono protects, preserves

PIE *h1e- (LIV2 243-4, IEW 78)
CC *e-i e/o- (not in Schumacher)
Wrzburg: pres. ind. 3sg. rel. i 29d29, with infix. pron. 3sg. f. and
anaph. pron. 3sg. f. cota eiade 29d29, impv. 2pl. comid 27a3, with infix.
pron. 3sg. n. and emp. pron. 2pl. cotdith si 7d4, Milan: VN coimt (see also
con-eim), gen. sg. chometa 55d6.
dd kindles, burns61
PIE *deh2u- (LIV2 104-5)
adda (kindles, lights): Milan: pres. ind. pass. sg. astoither 38d18.
dosl deserves, merits

PIE *selh1- (LIV2 529, IEW 899)
CC *-sl-ii e/odosl: Wrzburg: perf. 1pl. nirothuillissem 24d6, Milan: VN (m u &
o) tuillem, tuilem 72c15, dat. pl. tullemaib 39d18, adroilli (deserves, is
entitled to): Wrzburg: pres. ind. 3sg. with proleptic infix. pron. 3sg. n.
attroilli 2d13, pret. 1pl. with emp. pron. 1pl. niarilsemni 20d14, 2pl.
narilsid 21a17, 3pl. rel. atroillisset (Cf. McCone 1985: 97) 4c15, adroilliset
4c35, rel. with infix. pron. 3sg. n. assidroilliset 17a9, with infix. pron. 3sg.
n. cinid arillset 4c39, narilset 4d10, pres. subj. 3pl. with infix. pron. 3sg. n.
ciatroillet 31c23, ciniarillet 31c23, Milan: pres. ind. 1sg. adroilliu 75a11,
2sg. rel. nad nairilli 55d13, 3sg. atroilli (Cf. McCone 1985: 97) 51d12,
(rel.) asroilli 51d12, 111b28 withinf (?) 130b8, rel. adroilli 61a20, rel.
nachid irilli 84c13, 2pl. with infix. 3sg. arnachaairillid 115b11, 3pl.
asroillet 40a4, nadnairillet 40c12, rel. with infix. 3sg. assidroillet 54d6,
61
The derivation is uncertain. If this is primary, it could be either PIE *deh2u-e-ti >
*dah2u-e-ti > *dw-e-ti > OI dd or *deh2u-ie-ti > *dah2u-ie-ti > *dau-ie-ti > *daw-i-ti >
dd. However, the possibility that this is causative, *doh2u-ie-ti > *doh2u--ti > *dw--ti
> CC *dw--ti > OI *dd should not be discarded. Note that kindles is semantically
causative, although that meaning could also have developed directly from the meaning
catch fire.
340

pass. sg. asroilled 122b13, pres. subj. 3pl. cen airillet 90c2, fut. 3pl. rel.
with infix. 3sg. adidroillifet 61a20, pret. 3sg. asroilli 111b28, with infix.
3sg. asndaroilli 124d7, 1pl. asindroillissem 119d8, 3pl. nairillset 114c9,
atroillisset 46d9, rel. -airilset 54d9, rel. with infix. 3sg. assidroilisset 61b17,
reladidroillisset 61b17, perf. 3pl. with infix. 3sg. ciasidroilliset 77a15, verb.
nec. airillti 22d22, part. airillte 61a20, airilte 113b3, 122b13, acc. sg. airilti
56c6, nom. pl. airlti 61b17, acc. pl. ariltnea (GOI 733) 87c4, dat. pl.
airiltib 61a19, VN irilliud, airilliud 30d2, acc. sg. airilliud 61d2, nom. pl.
degairltin 91a10, drochairiltin 103b5, acc. pl. airilltinni 131a6; asroilli
(deserves, is entitled to): Milan: pres. subj. 3sg. coasroillea 107b6.
gnid acts, does

PIE *enh1- (LIV2 163-5, IEW 373-5)
CC *gn-ii e/oWrzburg: pres. ind. 1sg. with infix. pron. 3sg. n. and emp. pron.
1sg. nagnusa 3c30, 3sg. gniith 21b2, rel. gns 21b2, 3pl. rel. gnite 12b15,
fut. 3pl. rel. gende 93b16 (Taur. F. IV. 24), Thes. Pal. I, 714; perf. 3sg. rel.
rogni 3c25, with infix. pron. 3sg. n. ragni 3c25, Past subj. 1pl. rel.
congnemmis 4a9, impv. 2pl. gnid, pass.: fut. 3sg. rel. gntar 12b30, perf.
3sg. rel. rognith 12b30, Milan: pres. ind. 3sg. gnthi 30b20, rel. gns 29b8,
29b11, pass. sg. gnaither 38a5, pass. pl. rel. gniter 21c3, gnitir 38a5, pres.
subj. 3pl. rel. gnete 125c3, past subj. 3pl. nongnetis (this form not in DIL)
30a4, rel. -ngnetis 23a2, fut. 3sg. rel. genas 51b10, pass. sg. gentar 69d6,
perf. 3sg. rogni 48c6, rel. rongeni 22d19, 27a6, rel. rongni 31b24, rugeni
72b20, pass. sg. rongnth 14a18, 17d12, pl. rognatha 39a24, pret. 3pl. rel.
-gensat 80b11, pass. sg. rongnith 31b20, pl. nadrognatha 115b4, rogntha
115b4, perf. 3pl. rogenset 80c6, pass. pl. rel. rongnitha 97a3, part. gnthi
115b2, VN (m u) gnm 14c15, 15a6, 15b1, 15b3, acc. sg. gnmo 15a2, 15a3,
gen. sg. gnmo 15a5, dat. sg. gnm 15a16, acc. pl. gnimu 14c16, 14c19,
15a2, gen. pl. gnmae 19c4, dat. pl. gnimaib 19b9; dogn (does, makes):
Wrzburg: pres. ind. 1sg. dognu 17c11, 24a12, with emp. pron. 1sg.
dognusa 18a19, dogniusa 11a11, and with dem. rel. andognusa 24a23,
with infix. pron. 3sg. n. dagnu 13c11, and emp. pron. 1sg. dagnusa 14d26,
n denim 12c9, 2sg. dogn 6c16, with rel. n lasse dongn 1c10, 3sg. dogn
341

6a8, 8c9, 10b11, 27a7, 29b11, 24d11, with infix. pron. 3sg. n. dagn 12b34,
26a12, with rel. n dongn 4c33, amal dongn 24d10, and with emp. pron.
3sg. dongnsom 21d9, nd deni 31d12, 1pl. dognam 15d9, with emp. pron.
1pl. dogniamni 16a25, and with dem. rel. andogniam ni 16a24, and with
nad, annad dnamni 16a24, Thes. Suppl. p. 51 ; nd denam 17b10, 2pl.
dogniith 25c14, with emp. pron. 2pl. dognithsi 9c15, and with rel. n hore
dongniithsi 16a27, with infix. pron. 3sg. n. dagnith 11d6, with prep. and
rel. and emp. pron. 2pl. diandenid si 8c11, 3pl. dognat 30c14, dogniat
28c16, with infix. pron. 3sg. n. dagniat 23d21, 26a12, dgniat 20c21, with
rel. n dongnat 30c12, amal dongniat 30b24, with dem. rel. andugnat
14b17, andugniat 10c21, and emp. pron. 3pl. andogniat som 17b9, with
dem. rel. and nat anatdenat 15d6, ni dnat 12b20, nitnat 24a25. perfect.
pres. ind. 3pl. with anaph. pron. 3pl. nidernat sidi 22d3. fut. 1sg. with infix.
pron. 3sg. n. dagna 3d16, with interrog. in, indign 9d4, 2sg. with rel. n
and emp. pron. 2sg. dongenaesiu 32a25, 3sg. dogna 26a20, with dem. rel.
andogena 20c10, 1pl. ni dignem 17b9, and emp. pron. 1pl. nidignemni
15d6, 2pl. with rel. n dongnid 17a6, 3pl. with anaph. pron. 3pl. dognat
sidi 13a13. sec. fut. 3sg. with conjn. and dem. rel. and emp. pron. 3sg. m.
arandogenadsom 26a12, 2pl. ni dignte 9d9. perfect. sec. fut. with infix.
pron. 3sg. n. dargente 11d5, darigente 13b3, perf. 1sg. with emp. pron. 1sg.
dorigniussa 24b12, nddeirgenus 24a14, 2sg. with infix. pron. 1sg.
domrignis 4c27, 3sg. dorigni 4c32, dorgeni 11a28, 12a29, dorigni 13d30,
dorigeni 11a30, 27d10, (with gem.) dorrigeni 30d22, with rel. n hre
dorrigeni 15d13, and infix. pron. 3sg. n. amal dondrigni 6d2, with dem.
rel. andorigeni 21d11, 30a8, and with anaph. pron. 3sg. andorigeniside
22a12, and infix. pron. 3sg. n. ciadudrigni 11a20, nd deirgni 13b17, 1pl.
dorignsam 14b26, dorignsam 14b26, with dem. rel. andorgensam 24d3,
2pl. with emp. pron. 2pl. durigensidsi 33d5, with cia and infix. pron. 3sg. n.
ciadodrignsid 20d3, with infix. pron. 3sg. n. and emp. pron. 2pl. darigensi
(= -sid, Ped. II, 542) 9c29, 3pl. dorignsat 7d10, dorigensat 5a24, with
emp. pron. 3pl. dorignsatsom 11a30, with rel. n and infix. pron. 3pl. amal
dondrignsat 26a20, with anaph. pron. 3pl. amal dorgensat sidi 28d19,
pres. subj. 1sg. with cia, ciadogno 17c20, and with infix. pron. 3sg. n.
ciadagno 3d2, perfect. with rel. n and infix. pron. 3sg. n. dondrn 5b18,
342

2sg. dogn 32a24, with emp. pron. 2sg. dognsu 5d39, mani dne 10a17, n
dene 5d38, perfect. (gem.) dorronai 32a24, 3sg. dogn 12c46, with infix.
pron. 3sg. n. dodgn 27c14, perfect. with conjn.s arnaderna 4a5, derna
12b6, 17a13 (= Facs.), 1pl. with conjn. and
infix. pron. 3pl. codos gnem 21b9, perfect. with conjn. condernam 22a12,
dernam 25c10, ndernam 17b6, 2pl. dogneid 5d30, 11d16, 13a29,
dogneith 27c9, ciadugneid 22b23, with rel. n dongneid 5c20, dongneith
5d27, with emp. pron. 2pl. dogneithsi 12d7, dognedsi 25d22, with infix.
pron. 3sg. f. nsndenaith 6b29, perfect. with conjn.s dernaid 13d30,
arandernaid 16a24, arnadernaid 16a24, past subj. 1sg. dognein 10a27,
cedugnn 10c16, with ma and infix. pron. 3sg. n. madagnenn 10d31,
perfect. with nach and infix. pron. 3sg. n. and emp. pron. 1sg.
nachderninnse 8a5, 3sg. dogned 17a13, perfect. duronad 8a4, 1pl.
cedugnemmis 10c21, 2pl. dognethe 9d25, with prep. and rel. frisandnte
9c24, perfect. dornt 16a23, impv. 3sg. with infix. pron. 3sg. m. (reflex.)
dangnid (MS dangid) 8d9, dnad 6b18, 28b19, with emp. pron. 3sg. f.
dnadsi 28d19, 2pl. dnid 23c11, 25c17, denid 24a23, 27a7, with infix.
pron. 3pl. and emp. pron. 2pl. dosngniithsi 24b12, pass.: pres. ind. 3sg. with
prep. and rel. arandntar 2a10, dognther 9c22, 10c11, 14c21, 20a7, 22b21,
23c11, dognither 3c36, 9c14, 10c11, 23b20, with prep. and rel.
frissandntar 11a9, impf. 3sg. dognthe 15a18, fut. 3sg. dogntar 4d1, 26a8,
perf. 3sg. doronad 15a33, (with gem.) dorronad 26d5, with ce and infix.
pron. 3sg. n. cedudrnath 33a15, pres. subj. dogneither 13a29, dognether
29a21, madugnether 5c23, manidnatar 8a11, perfect. with rel. n dorrntar
13d21, past subj. 3sg. dognethe 9c20, Milan: pres. ind. 1sg. -denim 37c3,
37c5, 40a17, neg with infix. 3sg. nsdnaim 75a9, 2sg. dugn 106c12, rel.
dongni 44a23, arandeni 55d21, neg cenid deni 56b33, 3sg. (rel.) dogn
16c10, 20a19, 23c20, 26c2, 37a10, 51b12, 145b6, (rel.) dogni 59a7, 81a5,
(rel.) dugn 39b2, 128c3, dugni 93a13, with infix. 3sg. n dusgn 29a3, rel.
dungn 29a3, 90d3, 96a9, 102a15, 138a4, rel. dungni 90d12, 92b10, 114d12,
127a5, rel. dong(n) 51b27, rel. dugn 81a6, rel. with infix. 3sg. dudgn
44d21, with ro nachderni 128c3, 2pl. with nas dundgniid 103b13, 3pl. (rel.)
dogniat 3a5, 42c2, dugniat 120c7, 139a8, (rel.) -denat 18b4, 37b26, with
infix. dundgniat 26b10, with infix. 3sg. n dagniat 26c2 (translated as pass.
343

in Thesaurus), withinf 3pl. dudagni 130a8 28c8, rel. dongniat 51c9, 55a8,
rel. dungniat 56b3, 56d11, 86c9, 121d11, 128d3, pass. sg. (rel.) dugnither
27d2, 30a3, 32d6, 92b9, 96c10, 126d9, (rel.) dognither 32a10, 38c19,
51b28, 111c3, rel. in dentar 44c12, with nas dungnither 67b19, pl. with nas
dongniter 26c1, dungnter 138c3, rel. -ndenatar (in DIL ndnatar) 84a11,
impv. 2sg. -dene 55a1, 56c15, 3sg. denad (this form not in DIL) 39a6, 2pl.
denid (this form not in DIL) 54c26, impf. 1sg. dugninn 91b7, dognin
136a8, 3sg. (rel.) dugnth 30a3, 3pl. dognitis 22a4, 54c12, 132c4, with infix.
dagnitis 28d8, rel. dugntis 62c13, rel. dugnitis 102d13, pass. sg. dognithe
38d20, pres. subj. 1sg. dugneu 23c24, 23d4, rel. dungneu 92a17, 2sg. -ndene
23c20, -dene 37c15, with infix. 1sg. dumgne 42a8, 3sg. dungn
129b1,condena 20a14, nadeni 23c20, -dena 35d14, 51a16,
1pl. nad
ndenamni 23c6, -dugnem (this form not in DIL) 23c6, 3pl. dugnet 120c1,
with infix. 3sg. dagnet 83b14, rel. -gnetis 54c18, pass. pl. dugnetar 112d8,
-dentar 30d13, neg arna derntar 93a1, past subj. 1sg. durininn 46b13, con
deninn (this form not in DIL) 60a10, 3sg. rel. with infix. 3sg. dagneth 51d2,
3pl. -dentis 35c18, 124c22, mani dendis 75d2, pass. sg. -dente 29a6, rel.
dognethe 49a11, fut. 1sg. dugnsa 30b9, 63a6, -digensa 37c2, dugn 69a21,
n digen 69a21, 2sg. dugene 41b4, 3sg. dugena 50d10, rel. dungena 56d5,
dungenae 129b4, neg ndignea 96a8, 1pl. -digenam 30c9, dungenam 111d3,
3pl. rel. dungenat 56b15, rel. neg nad ndignet 56b15, neg ndergenat 80a9,
sec. fut. 3sg. digned 14b04, rel. dungendad 123c1, pret. 3sg. nadndigni
23b10, 3pl. rel. dugnetis 74a1, rel. with infix. 3pl. dudagnetis 73d1 (2x),
pass. pl. dugnitha 73a19, perf. 1sg. (rel.) dorignius 47a20, dudrignius
23c27, -dernus 39a11, with emp. 1sg. (GOI 681) dorigenuassa, 2sg. (rel.)
dorignis 46b24, 46b26, 63c5, 91b12, dundrignis 128a12, 3sg. (rel.) dorigni
19c19, 23b11, 42b24, 46b30, 48d28, 50d15, 51a16, 98c6, 103d7, 116d5,
124c23, (rel.) durigni 50d10, 54d16, 62c19, 67d2, 91c9, 96d3 (2x), 124b5,
-ndergini 23c15, rel. andurigni 24c8, (rel.) dorigeni 24c13, 27b13, 55d4,
98c6, 129d5, (rel.) durigeni 67b17, 74b7, 85c9, 129d5, (rel.) -durigni (this
form not in DIL) 31b24, 39b2, 128d5, (rel.) -dergeni 36a1, 39a15, 69c7,
-dergni 114b12, with infix. 3sg. dundrigni 39a3, 51a16, 53b27, rel. with
infix. dundarigni 93d14, (rel.) with infix. darigni 51d2, 55c3, 55c4, with
infix. 3pl. dusrigeni 54a34, 1pl. (rel.) dorigensam 46b26, 3pl. doringensat
344

16d6, 23b5, 28d7, 29d8, (rel.) dorigensat 43b13, 46d10, 48b18, 53b11,
54a34, 80b10, 90c12, 97b4, 124d4, 136b4, (rel.) durigensat 50c7, 80b4,
84c1, 91a21 (2x), 105a2 (2x), rel. durignsat 62d6, with infix. 3sg. n
dondarigensat 31b17, rel. with infix. 3sg. dundrigensat 87a8, with infix.
3pl. rondageinset 29a4, with nas dundrigensat 68b4, pass. sg. doronad
90c11, duronad 136c1, pl. doronta 133c5, verb. nec. dnti, denti (this form
not in DIL) 51b8, 82a7, deinti 58a20, part. margnethi 126c6 VN (m u)
dnum, denum 2d11, 14c12 (2x), 22b1, 23c20, 35d14, 55c18, 114b9, dat. sg.
denum 14a8, 18c6, 23a5, 23c5, 40c5, 54c12, 60b6, gen. sg. denma 34b10,
35b25, 38d20, 47c13; *condni (this verb not in DIL, but referred to under
VN): Milan: VN comdnam, dat. sg. comdenum 34a23; congn (helps,
assists): Wrzburg: pres. ind. 3sg. with emp. pron. 3sg. m. gn som 26d17,
1pl. congniam 14c42, fut. 3sg. congna 22d9; fogn (serves): Wrzburg:
pres. ind. 3sg. with rel. n amal fongn 12a12, perf. 3sg. foruigni 13b7, 1pl.
with prep. and rel. pron. diaforgnsam 3c15, 2pl. amal foruigensid 3b28,
3pl. rel. foruigensat 1b22, pres. subj. 3sg. rel. arafogna 12a12, 1pl. rel.
fognem 5d19, 2pl. fogneith 20a11, impv. fognad 20b11, 2pl. fognid 3b28;
pass.: pres. ind. 3sg. fognther 27b9, 3pl. amal fongnter 27b9, Milan: pres.
ind. 3pl. fongniat 112b4, fut. 1sg. rel. fungen 78d2, perf. 3sg. rel. diaforgeni
44c9, VN fognam, dufognam 121b13, gen. sg. fognama 84c12, dat. sg.
fognam 18b8, 28b9; frisgn (exercises, practises): Wrzburg: pres. ind.
3sg. rel. frisgn 27b3, Milan: pres. ind. 3pl. (rel.) frisgnat 37b12, VN (m u)
frithgnam, nom. pl. frithgnamai 39d15, acc. pl. frithgnamu 56b4;
immfogni (is construed): Milan: pres. ind. 3sg. (rel.) -imfogni 27c10 (2x);
*remign: Milan: part. remgnethe, gen. sg. remgnethi 103d15.
ld charges, accuses
PIE *slei H- (IEW 650f.)
CC 2. *lii -e/oWrzburg: pres. ind. 1sg. with emp. pron. 1sg. liimse 13b18, ni liim
10a1, 1pl. lmmi 13b17, 3pl. lit 2a13.
raid rows, sails

PIE 2. *h1reh1- (LIV2 251f., IEW 338)
345

CC ?*r-i e/oimmr (rows): Milan: VN imram 126a1.
snd swims, floats

PIE *(s)neh2- (LIV2 572f., IEW 971f.)
CC *snMilan: pres. ind. 3sg. snaid 93c1.
snid spins
PIE *sneh1- (LIV2 571f., IEW 973)
CC *snii -e/osnid: Milan: part snithe 24b7, VN snm 89b7; immfresnai
(contends, disputes): Wrzburg: pres. ind. 3sg. imfresna 30c16, 3pl.
imfresnat 31b30, ni imbresnat 29d2, Milan: pres. ind. 3pl. immafresnat
20d6, impv. 3sg. naimresnad 139a12, VN imresan 46b5, dat. sg. imresun
132c3; adcosnai (strives, seeks): Milan: pres. subj. 2sg. adcosnae 56b39,
-ascnae 56b31, 3sg. adcosna 51b6, perf. 3sg. adruchoissni 69d4
part (io, i) ascnaide, dat. sg. f ascnaidi 31c21, dat. sg. / acc. pl. ascnaidiu
53b25, VN (m u) ascnam 56b28-29, acc. sg. ascnam 48d2162, dat. sg.
ascnam 72b23, acc. pl. ascnamu 53a7; asroinni (escapes): Milan: pres.
subj. 3sg. asroinnea 31a2, part. rnaide, nephrnaidi 134b9; consn
(contends, contests): Milan: VN (m u) cosnam, dat. sg. cosnam 73a12;
*to-di-ess-ro-sn: Milan: VN todrnide, todernide 24b7, VN todrnam, acc.
sg. todernam 77d4.
sd turns, turns round

PIE *seh1- (LIV2 538f., IEW 914)
CC *su-e/osd: Milan: pres. ind. 3sg. rel. soas 64b4, 123d8, past subj. 2sg.
nosoe 44b15, sec. fut. 2sg. no soife 33a1, perf. 2pl. with infix. 3sg. rasoisit
62
The gloss is innascnam. DIL separates this as inna ascnam, but this is incorrect. In a
masc u-stem, the only forms identical to the nom. sg. can be acc. sg. or dat. sg., and that
does not work with the article. The Latin is acc., and that would fit the OI form perfectly
with an acc. sg. art, nasalising, and the acc. of ascnam.
346

103c15, 3pl. rosoisset 124c6, pass. sg. rusoad 101a6, part. soithe 94c15, VN
(m u) soud, acc. sg. soud 86c4, dat. sg. soud 99d7; conto (turns,
converts): Milan: pres. ind. 3sg. toi 65a1, 3pl. rel. contoat 46c1, perf. 3sg.
conruthi 51c22, contoroe 123b7, VN comthd, comthoud 20a16, acc./dat.
sg. comthod 53c14, 111a5; dointa (turns back, returns): Wrzburg: perf.
3sg. with rel. n donintarri 16b18, pres. subj. 2pl. manitintith 12d2, pass.:
pres. subj. 3sg. manitintither 12c46, Milan: pres. subj. 2sg. duintae 3a13,
nadtintae 3a15, past subj. 3sg. duintad (this form not in DIL) 89d6, perf.
3sg. (rel.) dorintai 3a7, duintarrae 54d3, VN (m u) nom. sg. intud 2a6 (2x),
3a11, acc. sg. tintuda 3a14, dat. sg. tindud 2a15 (2x), 37a10, gen. sg. tintuda
2b3, 3a8, tintda 45d2, acc. pl. tintudai 3a14, nom. pl. tinduda 2d11;
immsi (turns around, moves): Milan: pres. ind. 2sg. rel. immesoisiu
111a6 pass. sg. rel. immesoither 61a26, past subj. 3sg. connachnimpad
122a14, verb nec. impaithi 125a8, VN impd, impd 61a28, impud 120c8,
acc. sg. impd 93a18.
sred scatters, hurls

PIE *sterh3- (LIV2 599-600, IEW 1029-30)
CC *stre- (not in Schumacher)
?as-sri (sprinkles): Milan: VN assreud (GOI p. 508) 48c33,
63c19.
347
Bibliography
8 Bibliography
Bailey, C-J. N. 1996: Essays on Time-Based Linguistic Analysis. Oxford.
Beekes, R.S.P. 1988. Laryngeal developments: a survey, in
Bammesberger, A. (ed.), Die Laryngaltheorie und die Rekonstruktion
des Indogermanischen Laut- und Formensustems, pp. 59-105.
Heidelberg.
Bergin, O.J. 1916. Irish sirid seeks, riu 8, 196.
--- 1938. Varia I, riu 12, 215-235.
Best, R.I., and Bergin, Osborn (ed.). 1929. Lebor na hUidre: the Book of the
Dun Cow (First edition, third reprint (with corrigenda, by the School
of Celtic Studies, Dublin Institute for Advanced Studies, 1992).
Dublin. Accessed at CELT, http://www.ucc.ie/celt/online/G301900.
html, 12 Sept. 2011.
--- 1956. Book of Leinster, formerly Lebar na dachongbla. Dublin.
Accessed at CELT, http://www.ucc.ie/celt/ online/G800011B.html, 12
Sept. 2011.
Binchy, D.A. 1960. IE *qe in Irish, Celtica 5, 77-94.
Bisagni, Jacopo and Warntjes, Immo. 2007. Latin and Old Irish in the
Munich Computus: a reassessment and further evidence, riu 57, 133.
Bisagni, Jacopo. 2008. Amrae Coluimb Chille: a critical edition. Unpubl.
PhD thesis.
Boling, Bruce. 1972. Some problems of the phonology and morphology of
the Old Irish verb, riu 23, 73-101.
Breatnach, Liam. 1986. Varia VI, riu 37, 191-4.
Bronner, Dagmar. 2005. Code-switching in medieval Ireland: the case of
the Vita Tripartita Sancti Patricii, Journal of Celtic Linguistics 9, 112.
Carney, James. 1958. Two Old Irish poems, riu 18, 1-43.
Charles-Edwards, T.M. Mi a dynghaf dynghed and related problems, in:
Eska, Joseph F., Gruffydd, R. Geraint and Jacobs, Nicolas (eds.),
Hispano-Gallo-Brittonica: Essays in Honour of Professor D. Ellis
Evans on the Occasion of his sixty-fifth Birthday, 1-15. Cardiff.
349
Bibliography
Cowgill, W. 1975. The origins of the Insular Celtic conjunct and absolute
verbal endings in H. Rix (ed.), Flexion und Wortbildung: Akten der
V. Fachtagung der Indogermanischen Gesellschaft, Regensburg, 4070. Wiesbaden.
--- 1983. On the prehistory of Celtic passive and deponent inflection,
riu 34, 73-111.
Crystal, David. 1991. A Dictionary of Linguistics and Phonetics (3rd edition,
updated and enlarged). Oxford.
De Bernardo Stempel, Patrizia. 1999: dominale Wortbildung des lteren
Irischen: Stammbildung und Derivation. Tbingen.
Delamarre, Xavier. 2008. Dictionnaire de la langue gauloise: Une approche
linguistique du vieux-celtique continental. Paris.
DIL: Dictionary of the Irish language based mainly on Old and Middle Irish
materials. Compact edition. Dublin 1998.
Dillon, Myles. 1947. Celtic and the other Indo-European languages,
Transactions of the Philological Society, 15-24.
--- 1971. Irish drgaid spreads (coverings), makes a bed, Celtica 9,
205-209.
Duran, J. 1995: Branching trees, waves, and rising tides: some reflections
on the notions of speech community and diffusion of linguistic
traits, Journal of Celtic Linguistics 4, 77-88.
Evans, Ellis. 1967. Gaulish Personal dames. Oxford.
Fleuriot, Lon. 1976-7. Le vocabulaire de linscription Gauloise de
Chamalires, tudes Celtiques 15, 173-90.
Fowkes, Robert A. 1947. Celtic Etymological Notes, Language 23, no. 1,
120-2.
Geiriadur Prifsygol Cymru. Cardiff 1950.
Genee, Inge. 2005. Latin influence on Old Irish? A case study in medieval
language contact, Journal of Celtic Linguistics 9, 33-72.
GOI: Thurneysen, Rudolf. 1946 (reprinted 1998). A Grammar of Old Irish.
Revised and enlarged edition. With supplement. Dublin.
Greenberg, J. 1963. Some Universals of Grammar with Particular
Reference to the Order of Meaningful Elements, in Greenberg, J.
(ed.) Universals of Language, pp. 58-90. Cambridge.
Greene, David. 1962. Varia I, riu 19: 111-113.
350
Bibliography
Griffith, Aaron. A Dictionary of the Old-Irish Glosses in the Milan Codex
Ambrosianus C 301 inf. http://www.univie.ac.at/indogermanistik/
milan_glosses.htm.
Hamp, Eric, P. 1975. Varia II, riu 26, 168-174. (consecha etc)
--1976. Some Gaulish names in -ant-, riu 27, 1-20.
Hull, Vernam. 1947. Notes on some Early Irish verbal forms, Language 23
no. 24.
IEW: Pokorny, J. 1959. Indogermanisches etymologisches Wrterbuch.
Bern und Mnchen.
Irslinger, Britta Sofie. 2002. Abstrakta mit Dentalsuffixen im Altirischen.
Heidelberg.
Isaac, Graham. 1993. Issues in the reconstruction and analysis of Insular
Celtic syntax and morphology, riu 44, 1-32.
--- 1996. The Verb in the Book of Aneirin. Studies in Syntax, Morphology
and Etymology, Tbingen.
--- 2000. The most recent model of the development of absolute and
conjunct flexion, riu 51, 63-8.
--- 2004a. Placenames in Ptolemys Geography (cd-rom). Aberystwyth.
--- 2004b. Scotland in de Hoz, J., Lujn, E.R. and Sims-Williams, P.
(eds.) dew approaches to Celtic place-names in Ptolemys
Geography, pp.189-214. Madrid.
--- 2005. Insular Celtic vs. Gallo-Brittonic: an empirical or a
methodological question? in Gillies W. and Harding D.W. (eds.)
Celtic Connections: Papers from the Tenth International Congress of
Celtic Studies, Edinburgh, 1995, volume 2: Archaeology,
numismatics, historical linguistics, pp. 190-202.
--- 2007a. A new conjecture on the origin of absolute and conjunct
flexion, riu 57, 49-60.
--- 2007b. Celtic and Afro-Asiatic, in Tristram, Hildegard L.C. (ed.)
The Celtic Languages in Contact: Papers from the Workshop within
the Framework of the XIII International Congress of Celtic Studies,
Bonn, 26-27 July 2007, pp. 25-80. Potsdam.
--- 2007c. Studies in Celtic Sound Changes and their Chronology.
Innsbruck.
--- 2009. Die Urgeschichte der verbalen Morphosyntax im Keltischen:
eine Schnittstelle zwischen Grammatik, Semantik und Pragmatik, in
Ricken, E. & Widmer, P. (eds.) Pragmatische Kategorien. Form,
Funktion und Diachronie. Akten der Arbeitstagung der
indogermanischen Gesellschaft vom 24. bis 26. September 2007 in
Marburg. Wiesbaden.
Joseph, Lionel S. 1987. The origin of the Celtic denominatives in *-sag-,
in Watkins, C. (ed.), Studies in memory of Warren Cowgill (19291985), 113-159. Berlin.
351
Bibliography
Kavanagh, Samus and Wodtko, Dagmar (ed.). 2001. A Lexicon of the Old
Irish Glosses in the Wrzburg Manuscript of the Epistles of Saint
Paul. Vienna.
Klingenschmitt, G. 1978: Zum Ablaut des indogermanischen Kausativs,
Kuhns Zeitschrift 92, 1-13.
--- 1982: Das altarmenische Verbum, Wiesbaden.
Koch, John T. 1987. Prosody and the Old Celtic verbal complex, riu 38,
141-76.
--- 1992a. Gallo-Brittonic vs. Insular Celtic: the interrelationships of
the Celtic languages reconsidered, in Le Menn, G. and Le Moing, J.Y. (eds.) Bretagne et Pays Celtiques. Langues, Histoire, Civilisation.
Mlanges offerts la mmoire de Lon Fleuriot. Rennes / SaintBrieuc, pp. 471-95.
--- 1992b. Further to tongu do dia toinges mo thuath, tudes Celtiques
29, 249-61.
Kortlandt, Frederik. 1979. The Old Irish absolute and conjunct endings and
questions of relative chronology, riu 30, 35-53.
Lambert, Pierre-Yves. 2002. Recueil des inscriptions Gauloises. Vol. II.
Facs. 2. Textes Gallo-Latins sur Instrumentum. Paris.
--- 2003. The Old Welsh glosses on Weights and Measures, in Russell,
Paul (ed.) Yr Hen Iaith. Aberystwyth, pp. 103-34.
LEIA: Vendryes, J., Bachellery, E., Lambert, P.-Y. 1974-1996. Lexique
tymologique de lIrlandais Ancien. Lettres A B C D M d O P R S T
U. Dublin. (LEIA)
Lewis, Henry and Pedersen, Holger. 1937 (reprinted with corrections and a
supplement, 1974). A Concise Comparative Celtic Grammar. 3rd
Edition. Gttingen.
LIV: Rix, Helmut (et al.). 1998: Lexikon der indogermanischen Verben: Die
Wurzeln und ihre Primrstammbildungen. Wiesbaden.
LIV2: Rix, Helmut (et al.). 2001: Lexikon der indogermanischen Verben:
Die Wurzeln und ihre Primrstammbildungen. Wiesbaden.
Loth, J. 1892. Les mots latins dans les langes brittoniques. Paris.
Marm, Ursula. 1995. (unpubl. PhD thesis) Die Verbalkomposition im
Altirischen (unter besonderer Bercksichtigung des Prfixes frhgoid.
*uts-). Bonn.
Maurer, F. 1952. dordgermanen und Alemannen. Studien zur germanischen
und frhdeutschen Sprachgeschichte, Stammes- und Volkskunde, 3rd
ed. Mnchen & Bern.
352
Bibliography
McCone, K. 1979. Pretonic preverbs and the absolute verbal endings in Old
Irish, riu 30, 1-34.
--- 1985. The Wrzburg and Milan Glosses: Our Earliest Sources of
Middle Irish, riu 36, 85-106.
--- 1986. From Indo-European to Old Irish: conservation and innovation
in the verbal system, in Evans, D. Ellis, Griffith, John G. and Jope,
E.M. (eds) Proceedings of the seventh International Congress of
Celtic Studies, 222-66.
--- 1991a. Oir. -ic reaches, ithid eats, rigid stretches, directs, rules
and the PIE Narten presents in Celtic, riu 42, 1-11.
--- 1991b. The Indo-European Origins of the Old Irish dasal Presents,
Subjunctives and Futures. Innsbruck.
--- 1994. An tSean-Ghaeilge agus a Ramhstair, in McCone, Kim,
McManus, Damian, Hinle, Cathal, Williams, Nicholas and
Breatnach, Liam (eds) Stair na Gaeilge: in ms do Pdraig
Fiannachta, pp. 61-219. Maynooth.
--- 1996. Towards a relative chronology of ancient and medieval Celtic
sound change. Maynooth.
--- 1997. The Early Irish Verb. Second Edition Revised with Index.
Maynooth.
--- 1998. Double nasal presents in Celtic, and Old Irish licid
leaves, in Jasanoff, J., Melchert, C. and Oliver, L. (eds.) Mr
Curad: Studies in Honor of Calvert Watkins, pp. 465-476. Innsbruck.
--- 1999. OIr. erbaid entrusts, orb heir and orbae inheritance, in
Anreiter, P. & Jerem, E. (eds.) Studia Celtica et Indogermanica:
Festschrift fr Wolfgang Meid zum 70. Geburtstag, pp. 239-242.
Budapest.
--- 2005a: A First Old Irish Grammar and Reader, Including an
Introduction to Midlde Irish. Maynooth.
--- 2005b: L'irlandese antico e la sua preistoria (translated from SnG by
Elisa Roma). Alessandria.
--- 2006: The origins and development of the Insular Celtic verbal
complex. Maynooth.
McKee, Helen. 2000. The Cambridge Juvencus Manuscript, Glossed in
Latin, Old Welsh and Old Irish. Text and Commentary. Aberystwyth.
McManus, Damian. 1983. A Chronology of the Latin Loan-words in Early
Irish, riu 34, 21-71.
Meid, Wolfgang. 1963. Die indogermanischen Grundlagen der altirischen
absoluten und konjunkten Verbalflexion. Wiesbaden.
Meier-Brgger, Michael. 2003. Indo-European Linguistics. Berlin.
Meillet, A. 1908. Sur laoriste sigmatique, in Mlanges de linguistique
offerts m. Ferdinand de Saussure, 79-106. Paris.
Meyer, Kuno. 1907. The Expulsion of the Dssi, riu 3, 135-142.
--- 1997. Sanas Cormaic. Felinfach.
353
Bibliography
Nagy, G. 1970: Greek Dialects and the Transformation of an IndoEuropean Process, Harvard.
OBrien, M.A. 1956. Etymologies and Notes, Celtica 3, 168-184.
Crualaoich, Conchubhar. 1997. Syncope Patterns in Denominative
Verbs, riu 48, 239-264.
Cuv, Brian. 1958. Varia, igse 9, part I, 4-9.
ORahilly, T.F. 1942. Notes, mainly etymological, riu 13, 144-219.
Orel, Vladimir E. 1998. Albanian Etymological Dictionary, Leiden, Boston,
Brill.
Pedersen, Holger. 1909. Vergleichende Grammatik der keltischen Sprachen.
1. Band. Einleitung und Lautlehre. Gttingen.
--- 1913. Vergleichende Grammatik der keltischen Sprachen. 2. Band.
Bedeutungslehre (Wortlehre). Gttingen.
Pokorny, J. 1920: Zur Stammbildung der irischen Kausativa, Zeitschrift
fr vergleichende Sprachforschung 49, 74-82
--- 1936. Mittelir. catd einwickeln, ZCP 20, 96.
--- 1959. Indogermanisches etymologisches Wrterbuch. Bern und
Mnchen.
Rix, Helmut. 1977. Das keltische Verbalsystem auf dem Hintergrund des
indo-iranisch-griechischen Rekonstruktionsmodells, in Schmidt, Karl
Horst & Kdderitzsch, Rolf (eds), Indogermanisch und Keltisch:
Kolloquium der Indogermanischen Gesellschaft am 16. und 17.
Februar 1976 in Bonn, 132-58. Wiesbaden.
--- (et al.). 1998. Lexikon der indogermanischen Verben: Die Wurzeln
und ihre Primrstammbildungen. Wiesbaden. (LIV)
--- (et al.). 2001. Lexikon der indogermanischen Verben: Die Wurzeln
und ihre Primrstammbildungen. Wiesbaden. (LIV2)
Russell, Paul. 1988. The Celtic Preverb *uss and related matters, riu 39,
95-126.
--- forthcoming. uocridem: a new British word from Vindolanda, Studia
Celtica 45.
Sarauw, Chr. 1900. Irske Studier. Copenhagen.
Schmidt, Karl Horst. 1981. The Gaulish Inscription of Chamalires, The
Bulletin of the Board of Celtic Studies, vol. 29, part 2.
--- 1986. Zur Rekonstruktion des Keltischen: festlandkeltisch und
inselkeltisches Verbum, ZCP 41, 159-79.
354
Bibliography
Schrijver, Peter. 1991: The Reflexes of the Proto-Indo-European Laryngeals
in Latin. Amsterdam, Atlanta.
--- 1993. On the Development of Vowels Before Tautosyllabic Nasals in
Primitive Irish, riu 44, 33-52.
--- 1994. The etymology of the Celtic adverbs for against and with
and related matters, riu 45, 151-89.
--- 1995: Studies in British Celtic historical phonology. Amsterdam,
Atlanta.
--- 1997. Studies in the History of Celtic Pronouns and Particles.
Maynooth.
--- 2003. The etymology of Welsh chwith and the semantics and
morphology of PIE *k(w)sweibh-, in Russell, Paul, Yr Hen Iaith, pp. 123. Aberystwyth.
--- forthcoming. Old British, in E. Ternes (ed.) Brythonic Celtic.
Bremen.
Schulze-Thulin, Britta. 2001. Studien zu den urindogermanischen o-stufigen
Kausativa/Iterativa und dasalprsentien im Kymrischen. Innsbruck.
Schumacher, Stefan. 1995. Old Irish *tucaid, tocad and Middle Welsh
tynghaf tynghet re-examined, riu 46, 49-57.
--- 1999. Randbemerkungen zu absolut und konjunkt: Mittelkymrisch
hanfot, in P. Anreiter and E. Jerem (eds.), Studia Celtica et
Indogermanica: Festschrift fr Wolfgang Meid zum 70. Geburtstag,
453-64. Budapest.
--- 2000: The Historical Morphology of the Welsh Verbal doun,
Maynooth.
--- 2004: Die Keltischen Primrverben. Ein vergleichendes,
etymologisches und morphologisches Lexikon. Innsbruck.
Seebold, E. 1970. Vergleichendes und etymologisches Wrterbuch der
germanischen starken Verben. The Hague / Paris.
Sihler, Andrew J. 1995. dew Comparative Grammar of Greek and Latin.
Oxford.
Sims-Williams, Patrick (2007). Common Celtic, Gallo-Brittonic, and
Insular Celtic, in Gauloise et celtique continental, P-Y Lambert, G-J
Pinault (eds.) Geneva. pp. 309-54.
Sjoestedt, Marie-Louise. 1925. Les Verbes Causatifs et Itratifs dans les
langues Celtiques, Mlanges Linguistiques Offerts M. J. Vendryes,
323-340. Paris.
Stokes, Whitley. 1876-8: On the Celtic Comparisons in Bopps
Comparative Grammar, RC 3, 31-9.
--- 1905a. The Evernew Tongue, riu 2, 96-162.
--- 1905b: Flire engusso Cli D: The Martyrology of Oengus the
Culdee. London.
355
Bibliography
Stokes, Whitley and Strachan, John (eds.). 1901 (repr. 1987): Thesaurus
Palaeohibernicus: a collection of Old-Irish glosses, scholia, prose and
verse, volumes 1 and 2. Dublin.
Strachan, John. 1893a. Keltische Etymologien, Indogermanische
Forschungen 2.
--- 1893b. Contributions to the history of the deponent verb in Irish,
Transactions of the Philological Society 22, issue 3, 444-566.
--- 1949. Old-Irish paradigms and selections from the Old-Irish Glosses,
with notes and vocabulary. Fourth edition, revised by Osborn Bergin.
Dublin.
Stber, Karin. 1998: The Historical Morphology of N-stems in Celtic.
Maynooth.
Thes.: Stokes, Whitley and Strachan, John (eds.). 1901 (repr. 1987):
Thesaurus Palaeohibernicus: a collection of Old-Irish glosses,
scholia, prose and verse, volumes 1 and 2. Dublin.
Thurneysen, Rudolf. 1913 Review of Pedersens Vergleichende Grammatik
der keltischen Sprachen 2. Band. Anzeiger der indogermanischen
Forschungen 33, 23-37.
--- 1918. Miszellen, ZCP 12, 408-414.
--- 1921a. Grammatisches und etymologisches, ZCP 13, 101-108.
--- 1921b. Allerlei Keltisches, ZCP 13, 297-304.
--- 1921c. Tochmarc Ailbe Das Werben um Ailbe, ZCP 13, 251-282.
--- 1636. Irisches. Altir. ro-geinn hat Platz Zeitschrift fr
vergleichende Sprachforschung 63, 114-7.
--- 1946. A Grammar of Old Irish. Revised and enlarged edition. With
supplement. Dublin. (reprinted 1998)
Uhlich, Jrgen. 2002. Verbal Governing Compounds (Synthetics) in Early
Irish and Other Celtic Languages, Transactions of the Philological
Society 100:3, 403-33.
Vendryes, J. 1938. Lexpression du rire en celtique, tudes Celtiques 3,
38-45.
--- 1955-6. Neddamon delgu linda tudes Celtiques 7, 9-17.
--- 1974. Lexique tymologique de lIrlandais Ancien. Lettres R S.
Dublin. (LEIA)
--- 1978. (with Bachellery, E. and Lambert, P.-Y.) Lexique tymologique
de lIrlandais Ancien. Lettres T U. Dublin. (LEIA)
--- 1981a. Lexique tymologique de lIrlandais Ancien. Lettre A. Dublin.
(LEIA)
--- 1981b. Lexique tymologique de lIrlandais Ancien. Lettre B. Dublin.
(LEIA)
--- 1983. Lexique tymologique de lIrlandais Ancien. Lettres M d O P.
Dublin. (LEIA)
--- 1987. (with Bachellery, E. and Lambert, P.-Y.) Lexique tymologique
de lIrlandais Ancien. Lettre C. Dublin. (LEIA)
356
Bibliography
---
1996. (with Lambert, P.-Y.) Lexique tymologique de lIrlandais

Ancien. Lettre D. Dublin. (LEIA)
Walde, A. & Hofman, J.B. 1982. Lateinisches Etymologisches Wrterbuch.

Erster Band. Heiderberg.
Wartburg, Walther von. et. al. 1922-2002. Franzsisches Etymologisches
Wrterbuch: Eine Darstellung des galloromanischen Sprachschatzes.
25 vols. Basel.
Watkins, Calvert. 1962a. Varia II, riu 19, 114-118.
--- 1962b. Indo-European Origins of the Celtic Verb: 1. The Sigmatic
Aorist. Dublin.
--- 1963a. Indo-European metrics and Archaic Irish verse, Celtica 6,
194-249.
--- 1963b. Preliminaries to a historical and comparative analysis of the
syntax of the Old Irish verb, Celtica 6, 1-49.
--- 1964-7. Old Irish cssaid, tudes celtiques 11, 131-4.
--- 1966. The origin of the f-future, riu 20, 67-81.
--- 1969. Indogermanische Grammatik 31. Heidelberg.
--- 1991. A Celtic-Latin-Hittite etymology, in: Abusch, Tzvi,
Huehnergard, J., Steinkeller, Piotr (eds.) Lingering Over Words:
Studies in Ancient dear Eastern Literature in honor of William L.
Moran (Harvard Semitic Studies 37). Atlanta. pp. 451-453.
Wodtko, Dagmar. 1995. Sekundaradjektive in den altirischen Glossen.
Innsbruck.
Wodtko, Dagmar, Irslinger, Britta, and Schneider, Carolin. 2008. domina
im indogermanischen Lexikon. Heidelberg.
Zimmer, H. 1890. Keltische Studien 7. Die schicksale des
indogermanischen s-aorists im irischen und die entstehung des
keltischen
s-prteritums,
Zeitschrift
fr
vergleichende
Sprachforschung 30, 112-224.
Zupitza, E. 1901. Die Vertretung der u-Diphthonge im irischen und
Verwandtes, ZCP 3, 275-82.
357

Secondary Verbs in Old Irish

Uploaded by

Copyright:

Available Formats

Secondary Verbs in Old Irish

Uploaded by

Document Information

Original Description:

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

Secondary Verbs in Old Irish

Uploaded by

Copyright:

Available Formats

Provided by the author(s) and NUI Galway in accordance with publisher policies.

Please cite the published

Secondary Verbs in Old Irish: A comparative-historical study

PhD thesis, supervised by:

Aim, material, scope, and method ................................................. 17

Structure of the thesis .................................................................... 22

The terms Common Celtic, Brittonic, Goidelic;

Insular Celtic vs. Gallo-Brittonic ..................................................... 23

The Old Irish Verbal System................................................................ 27

Word order .................................................................................... 27

Primary versus secondary verbs .................................................... 27

Primary versus secondary endings ................................................ 28

Weak, strong and hiatus verbs ....................................................... 28

Strong verbs ........................................................................... 28

Hiatus verbs ............................................................................ 30

Weak verbs ............................................................................. 30

Simple and compound verbs ......................................................... 31

Absolute and conjunct ................................................................... 31

Historical background ............................................................ 32

Stem, tense and mood .................................................................... 34

Present indicative ................................................................... 35

Origin of the future formations ....................................... 38

Origin of the subjunctive ................................................ 37

Origins of the preterite .................................................... 41

Preterite passive ..................................................................... 43

Number and person ....................................................................... 44

Basic statistics ............................................................................... 45

Latin loans ..................................................................................... 48

Morphology of AI and AII verbs................................................... 52

Semantic motivation for formation as AI or AII verbs .......... 55

Semantic motivation for formation as deponent verbs ... 63

Morphology of and motivation for formation as -igidir verbs ...... 65

Derivations from nouns and adjectives in -ch ........................ 67

The appearence of unexplained -n- between base and suffix 68

Calques on Latin .................................................................... 69

Simple vs. compound verbs........................................................... 74

Arguments for age ......................................................................... 75

Primary verbs taking on weak flexion ........................................... 83

Semantics of derivative verbs ........................................................ 86

Semantic categories ................................................................ 87

Semantic relationship between subject or object of a verb

and its base ....................................................................................... 90

Subject causes [base] ...................................................... 96

Corpus of secondary verbs ................................................................. 107

Structure and notational conventions .......................................... 107

AI verbs ....................................................................................... 109

AII verbs ...................................................................................... 157

Appendix 1: Weak primary verbs ...................................................... 259

Appendix 2: Primary verbs ................................................................ 275

BII ................................................................................................ 314

BIII .............................................................................................. 324

BIV .............................................................................................. 327

Hiatus verbs ................................................................................. 334

methodological basis of the thesis. The second chapter is an introduction

Amrae Coluimb Chille

This verb is not attested in Wb. or Ml.

This verb is not attested in OI, or, this is a reconstructed root.

The Old Irish Verbal System

The Old Irish Verbal System