A Roman Fleet Soldier's Gravestone from Scarbantia (TRHR 38

ELECTRUM * Vol. 31 (2024): 143–151 https://doi.org/10.4467/20800909EL.24.011.19161 www.ejournals.eu/electrum A Roman Fleet Soldier’s Gravestone from Scarbantia (TRHR 38) Péter Kovács  https://orcid.org/0000-0002-8727-1588 Department for Classical Philology Pázmány Péter Catholic University Abstract In his paper the author deals with a funerary verse epitaph from Scarbantia (Sopron) published by him earlier (TRHR 38). Based on his new restoration, the gravestone (stela) was erected to a veteran by his wife who served earlier as fleet soldier. He was most probably called Britto. Several vulgar Latin phenomena can be observed in the inscription that was erected in the second half of the third century or at the very beginning of fourth c. The type of the metrical formula resiste viator et lege can be found in several Pannonian verse inscriptions. Based on the pieces of information given in the problematical parts of lines 2 and 3 that concern the wife, at least one line is missing at the beginning of the text and she erected the grave monument and the funerary epitaph to herself as well. Keywords: Latin epigraphy, Pannonia, Roman army, Carmina Latina, vulgar Latin. In the supplement volume (TRHR) of the series Die römischen Inschriften Ungarns I published a funerary epitaph in 2005 that was found in the late Roman city wall of Scarbantia in Pannonia superior (today Sopron) in 1990 (Színház, Street 23) used as spolia (TRHR 38).1 Based on lines 3–6, it is sure that the funerary epitaph was erected by the wife to his husband who was a veteran. It was clear from the very beginning that the inscription contains metrical parts, however, it was omitted in the corpus of the Pannonian verse inscriptions.2 On the other hand, the exact reading and interpretation of the gravestone have remained unclear. Basically, I correctly deciphered the letters twenty years ago, but with minor changes the entire text can be reconstructed. In this paper I intend with this epigraphic text and the possible reconstructions. 1 2 Gömöri 1993. CLEPann. Péter Kovács 144 Lower part of Roman stela with the epigraphic field. Limestone. Heigh: 72, width: 48, depth: 18 cm. Letter height: 3.5 cm. The upper part and the left frame of the gravestone are cut off, below, it is broken. The surface of the epigraphic field and the letters are totally worn off, damaged by a longitudinal fissure. Besides this fissure empty spaces were left in the lines 1, 2 and 4 (only the oblique stroke of the letter V in line 3follows exactly the fissure) that means it existed already before the carving. The letters are carelessly carved, they can be observed between the auxiliary lines of the ordinatio. No interpunction remained, but there are several ligatures. The entire epigraphic field is lined. In lines 2–4 several superfluous ligatures can be found: in the verbs lege ET, gessit TE, in the name Brittone: ONTE. Sopron, Soproni Museum, Lapidarium (without InvNr.). My former reading of the epitaph was as follows: [---]uca d(efuncta) est a(nnorum) IIII [---] Passer e[t] / [cons]iste viator et lege qui T[---]IT / [---] aessi et de caro cun^iuge (!) Crit/on^t^e qui vixit a^n(nos) LXX et fecit / stipen^dia XXXII. Lit.: THRH 38 (AE 2005.1216); Fehér 2007 (suggested the completions quit (!) (instead of quis), cuuiugi (!) and gaessit (!)), 112 j. 2, 277 j. 86, 331 j. 12, 511 no. 26. Some new observations Three observations must be considered in order to restore the entire text: 1. The expansion of line 1 is surely erroneous and there is no abbreviation in the line with only single exception. 2. Only one single abbreviation can be observed in the text (in the most frequent formula vixit ann(os)) and another one must be supposed in line 1. 3. The length of the epigraphic field and the missing part can be established based on the metrical formula in line 2. It is clear that in line 2 the addressing formula of the traveller resiste/siste/consiste viator et lege with imperative verbs can be read that relatively frequently occurs in prosaic funerary inscriptions containing metrical elements.3 In Pannonia the formula was used with the verbs consiste (Tit. Aq. 1965 from Tác) or resta (AE 2010.1264 from Carnuntum; CIL III 15195 from Aequinoctium) and siste (AE 2008.1094 from Carnuntum), and it can be observed especially in the area of Carnuntum near Scarbantia.4 Earlier I completed the end of line 1 as the cognomen Passer et, but in my opinion (based on better new photos), we must read re/[s]iste viator et lege.5 Based on this observation, only the very beginning of the lines is missing (one, max. two letters). The word with the ending -asse could have been most probably in or ex classe, ex asse or fortasse, but in our case, Meyer 2022, 194–216. Resiste viator et lege: CIL III 6155 = 7571; VIII 9642; AE 1973.231. Cp. Fehér 1998, 96; CLEPann, p. 151, Fehér 2007, 34–35. 5 Among the numerous verse inscriptions of this type, the same formula is attested only in reverse order: viator resiste et lege … (CIL II 6155 = 7571 = CLEMoes 27), but cp. resta viator et lege (CLE 1878, 1902; CLEPann 9). 3 4 A Roman Fleet Soldier’s Gravestone from Scarbantia (TRHR 38) 145 the only real possibility is classe. Another fact is that line 1 instead of the highly uncertain abbreviations should be read as follows: huc (instead of hoc) adest mil(es) i[n cl]asse. The word ‘soldier’ is most probably abbreviated, and the fissure divides the letters of preposition in. That means the husband who was surely a veteran (with stipendia XXXII) was a fleet soldier most probably before the late Roman military reform. He was buried here (hic adest, but instead of the adverb hic, huc was used (V pro I).6 The verb adest with the same meaning can be observed several times in funerary verse epitaphs (mainly in Christian ones: hic … adest: e. g. ILCV 1050; AE 2003. 255; CIL V, p. 622,13).7 The formula miles in classe is attested in Eburo’s Pannonian stela from Aquincum (Tit. Aq. 609). The length of his stipendia (32 years) also agrees with the usually long service of the fleet soldiers.8 Unfortunately, the question of the place of his service (in the imperial fleet in Ravenna or Misenum as many Pannonians did9 or in a provincial (Danubian) fleet as the classis Flavia Pannonica)10 remains unsolved as it is not mentioned in the remained part of the text. It seems more probable that he served in the near in Vindobona or Carnuntum where fleet soldiers also garrisoned during the Principate (AE 2010.1261)11 and in the Late Roman period as well (milites liburnarii of the legio XIIII Gemina (Carnuntum and Arrabona) and the classis Histrica: Not. Dig. Occ. XXXIV, 26–28).12 The veteran most probably returned to his hometown following his discharge. The name Previously I read the veteran’s name as Crit/on^t^e based on the ligature NTE, but the letter T is superfluous, therefore similarly to the ligatures ET it is mistakenly carved. The cognomen Crito(n) is of Greek origin, its Roman use attested (but extremely rare) in the Imperial period13, but it would perfectly suit to the veteran’s service in the fleet where several Greek-speaking Orientals served. On the other hand, now I see the first letter of the name rather as a damaged B (the superior curve remained) and the first letter in the next line is missing, therefore I suggest the reading Brit/tone. The ethnicon Britto could have also been used as cognomen.14 Unfortunately, neither of them (Criton or Britto) is attested in Pannonia. V pro I: CIL III, p. 2572, 2676; Mihăescu 1978, 176–177, no. 120. The elliptic formula in hoc (sc. tumulo) cannot be excluded either: cp. in huc (!) tumulo: e.g. AE 1921.47; ILCV 2455; IScM VI,2, 545; Galdi 2004, 322, 369. 8 Starr 1960, 88–96. 9 Acrudoae 2012, 127–160. 10 Rummel 2008, 28–86. 11 Weber 2013, 377–381. 12 For the problem of the bases of the classis, see Hobel 2018, 65–69. 13 Lőrincz 1999, 86; Solin 2003, 266, 2449. 14 Kajanto 1965, 201; Lőrincz 2002, 129. 6 7 Péter Kovács 146 The interpretation of lines 2–3 The most problematical part of the inscription is the second half of line 2 following the imperative lege that can hardly be read. The following letters are remained: … QVIT or +[-2-]+++C or G IT + are all vertical hastae. Between the words lege and qui there would be enough room for two letters, in this case the most plausible completion would be [si] qui(s) and it would concern to the traveller (the formula si quis es occurs together with resiste, viator, lege-type texts) as well (e. g. AE 2008.853; CIL VÍII 3109), sometimes instead of the vocatives viator or hospes. (e. g. CIL II 952 the epitaph of another Britto from Baetica) The problem is that I could not observe any trace of letters between lege and qui and it seems probable the room was made because of the transversal fissure of the stone monument. The structure of the sentence is clear. Two subordinate clauses with the question word qui(s) depend on the main predicate lege.15 The use of the relative pronoun qui instead of the question word quis can be several times observed in vulgar Latin inscriptions, even in the Danubian provinces.16 The next word following qui begins with a T and three letters are missing. On the other hand, the relative pronoun was mistakenly used, instead of the feminine quae the masculine qui was used that is also typical late vulgar phenomenon.17 The last letters of the line are surely I and T therefore one must suppose they belong to the end of a verb in present perfect third person singular. The C or G before them decides the question as it can only be completed as fecit and the damaged letters before the C do not exclude this verb. Between the the qui(s) and the predicate the subject in the accusative must have been mentioned that begins with the letter. The subject was the grave/ epitaph itself that was erected by the wife, therefore most probably the word tumulus or titulus was used. The problem is that in both cases, there is no enough room for the entire word, therefore the abbreviation tumul(us) or titul(us) must be supposed. The formulae titulum (CIL IX 5762 = CLE 551; CIL VI 39444 = 39445 = CLE 1961) or tumulum fecit (CLE 1431 = ICUR 19255) are used in verse inscriptions as well. The use of abbreviations can frequently be observed in carmina, even the same ones. Both possibilities are attested even in Pannonian verse funerary epitaphs: titul(um): CIL III 3241 = CLEPann 50 = CLE 1208 (Sirmium); RIU I 80 = CLEPann 25 = CLE 578 (Savaria). tumul(um): CIL III 3241 = CLEPann 50 = CLE 1208 (Sirmium). The obviously metrical sentence should mean that the traveller must read who (obviously the wife) erected the epitaph. In line 3 one can clearly read AESSIT (at the very beginning of the line the traces of an unidentifiable letter can be seen above) that can only be completed (as suggested by B. Fehér18) as the third person singular verb gessit that can only concern to the wife who erected the gravestone. The use of the dipthong AE instead of E can be observed several 15 16 17 18 Fehér 2007, 277. Mihăescu 1978, 232, n. 212; 255, n. 255; Galdi 2004, 338–340. Mihăescu 1978, 231–232, n. 212; Galdi 2004, 332–333. Fehér 2007, 511, no. 26. A Roman Fleet Soldier’s Gravestone from Scarbantia (TRHR 38) 147 times in vulgar Latin inscriptions even in Pannonia.19 Before the verb, we can suppose the use the conjunction et between the two predicates. What this line can mean? The meaning of the verb gero with the preposition de is unclear because it is surely a vulgar form. In this case, the original meaning of the verb gerere cannot be used as to ‘carry’, ‘bear’, ‘manage’.20 Its unclassical use with the meaning ‘to live’ frequently occurs in the epigraphic material (vitam or annos gerere), but never used with the preposition de. The original meaning of the preposition as ‘about’, ‘according to’, ‘away from’ or ‘with regard to’ cannot be used either, but the vulgar use (instead of the genitive) has no sense in this line either. On the other hand, the preposition can be used as instrumental ablative21, therefore the sociative ablative sense in this epitaph cannot be ruled out. In this case the line would mean that ‘she lived with his beloved husband’. Another possibility is that here the verb [di]gessit would have been used with the meaning ‘to demonstrate’, ‘enumerate’ or ‘relate/recount’.22 In this case the line can mean that ‘she related about (the deeds of) his beloved husband’. Both interpretations are unique ones. Fig. 1. The gravestone from Scarbantia (photo: Ortolf Harl) CIL III, p. 2570, 2676; Mihăescu 1978, 184–185, n. 31. ThLL VI,2, col. 1928–1950, esp. 1941. 21 ThLL V,1, col. 42–80, esp. 62–64 and 80; Väänänen 1981, 114, no. 50; Mihăescu 1978, 246–247, no. 236–237. 22 ThLL V,1, col. 1115–1120, esp. 1119–1120. 19 20 Fig. 2. Detail of the damaged part of line 2 (photo: Dénes Józsa) Fig. 3. Detail of the beginning of line 3 (photo: Dénes Józsa) A Roman Fleet Soldier’s Gravestone from Scarbantia (TRHR 38) 149 Fig. 4. The drawing published in the TRHR How many line(s) are missing? Theoretically, it cannot be excluded that all lines of the inscriptions would have remained, but in this case, the epitaph would not have mentioned the wife’s name. In my opinion, this would be highly improbable, therefore at least one line is missing that should have contained more information about the wife. This one must have been metrical too. It is very unusual even in verse epitaphs that the reader (traveller) had to read (lege) more information about the person who erected the inscription (in this case the wife) than the dead person (his husband). This fact probably means that the wife erected the gravestone and the epitaph to his husband and herself as well, when she was alive and the first line(s) commemorated her deeds in the third-person singular. Date The tombstone had to be erected before second half of the fourth century when the town walls of Scarbantia were built and several earlier Roman stone monuments were built into the walls as spolia.23 The stela cannot be dated much earlier because the palaeography of the letters (narrow, relatively high, tilting to the right and irregular letters) (cp. e.g. the very typical cursive-like G and L letters of these decades with longer oblique strokes)24, the unexperienced hand of the stonecutter and the lack of abbreviations refer to a date 23 24 Póczy 1967, 137–153, but see Soproni 1978, 146 Anm. 60. See Fehér – Kovács 2018. Péter Kovács 150 following 260 AD as in this period a total decline of the epigraphic habit in Pannonia can be observed.25 Very similar letters can be observed in an altar from Scarbantia erected by a beneficiarius during the first tetrarchy (RIU I 178). The unusual formulae cited above and the also infrequent fecit stipendia (mentioned in five other epitaphs: CIL III 7553; AE 1976.669; 1977.791; EE 8, 801; CIL VI 3609 (the earliest one from the year 221)) are also refer to the Late Roman period. From Scarbantia two other verse inscriptions are known (RIU I 185, 239 (also mentions the word viator)), but all of them can be dated to the early Imperial period.26 Based on this argumentation, the gravestone was erected in the second half of the third century or at the very beginning of the fourth century and the unfortunately incomplete funerary text can be interpreted as follows: [------?] huc (!) adest mil(es) i[n cl]asse re[s]iste viator e^t lege qui (!) t[umu]l(um) or t[itu]l(um) fecit [e^}t gaessit (!) de caro cu^niug[e] (!) Brit[t]on^e qui vixit a^n(nos) LXX et fecit stipen^dia XXXII. Abbreviations CLE – F. Bücheler, E. Lommatzsch (eds.), Carmina Latina epigraphica, vols. 3, Leipzig 1895–1926. CLEPann – P. Cugusi, M. T. Sblendorio Cugusi, Studi sui carmi epigrafici. Carmina Latina Epigraphica Pannonica, Bologna 2007. EE – Ephemeris Epigraphica, Berlin 1872–1913. ICUR – I. B. De Rossi, Inscriptiones Christianae Urbis Romae septimo saeculo antiquiores, 3 vols., Roma 1857–1915. ILCU – E. Diehl, Inscriptiones Latinae Christianae Veteres, vol. 4, Berlin–Dublin–Zürich 1925–1967. RIU – L. Barkóczi, A. Mócsy et al., Die römischen Inschriften Ungarns, 6 vols., Budapest–Amsterdam–Bonn 1972–2001. ThLL – Thesaurus Linguae Latinae. Tit. Aq. – P. Kovács et al., Tituli Aquincenses, 3 vols., Budapest 2009–2011. TRHR – P. Kovács, Tituli Romani in Hungari reperti. Supplementum, Budapest–Bonn 2005. 25 26 Kovács 2004, 185–195. CLEPann, 53–55, no. 13–14; Fehér 1998, 65–66; Fehér 2007, 23–24. A Roman Fleet Soldier’s Gravestone from Scarbantia (TRHR 38) 151 Bibliography Acrudoae, I. (2012), Militaries from Pannonia in the Imperial Fleet at Misenum and Ravenna (FirstThird Centuries AD): Prosopographical Aspects, Studia Antiqua et Archaeologica 18: 127–160. Fehér, B. (1998), Poems and Versification in Pannonia, AAntHung 38: 65–102. Fehér, B. (2007), Pannonia latin nyelvtörténete, Budapest. Fehér, B., Kovács, P. (2018), Palaeographia Aquincensis. Aquincum görög és latinfeliratainak paleográfiája, Budapest. Galdi, G. (2004), Grammatica delle iscrizioni latine dell’impero (provincie orientale). Morfosintassi nominale, Roma. Gömöri, J. (1993), Sopron, Színház u. 23, Nr. 57/2, Archäologische Forschungen in Ungarn im Jahr 1990, Archaeológiai Értesítő 120: 113. Hobel, T. (2018), Classis Flavia Pannonica. Zur Geschichte einer Provinzflotte in der römischen Kaiserzeit, Magisterarbeit, Wien. Kajanto, I. (1965), The Latin Cognomina, Helsinki. Kovács, P. (2004), Late Roman Epigraphy in Pannonia (260–582), in: Gy. Németh, I. Piso (eds.), Epigraphica II. Mensa rotunda epigraphiae Dacicae Pannonicaeque, Debrecen: 185–195. Lőrincz, B. (1999), Onomasticon provinciarum Europae Latinaeum II: Cabalicus-Ixus, Wien. Lőrincz, B. (20022), Onomasticon provinciarum Europae Latinaeum I: Aba-Bysanus, Budapest. Meyer, A. (2022), Travelers and Texts: Reading, Writing, and Communication on the Roads of the Roman West, in: M. Young Mayers, E. Zimmermann Damer (eds.), Travel, Geography, and Empire in Latin Poetry, London–New York: 194–216. Mihăescu, H. (1978), La langue latine dans le sud-est de l’Europe, Bucureşti–Paris. Póczy, K. (1967), Scarbantia városfalának korhatározása – La datation de l’enceinte de Scarbantia, Archaeológiai Értesítő 94: 137–153. Rummel, C. (2008), The Fleets on the Northern Frontier of the Roman Empire from the 1st to 3rd Century, Nottingham (PhD Diss.). Solin, H. (2003), Die griechischen Personnamen in Rom: ein Namenbuch, Berlin–New York. Soproni, S. (1978), Der spätrömische Limes zwischen Esztergom und Szentendre. Das Verteidigungssystem der Provinz Valeria im 4. Jahrhundert, Budapest. Starr, Ch. G. (19602), The Roman Imperial Navy: 31 B.C.–A.D. 324, Ithaca. Väänänen, V. (19813), Introduction au latin vulgaire, Paris. Weber, E. (2013), Ein magister navaliorum in Carnuntum, in: W. Eck, B. Fehér, P. Kovács (eds.), Studia epigraphica in memoriam Géza Alföldy, Bonn: 377–381.