Dan Dana — Dávid Petruţ

A military graffito on a pottery plate from the auxiliary fort at Buciumi (Roman Dacia)

Plates 3–4

The Roman auxiliary fort from Buciumi (Sălaj county, Romania)[1], situated in Dacia Porolissensis, was extensively excavated between 1963 and 1976[2]. During the archaeological investigations carried out in 1969, a pottery vessel fragment with an ownership inscription cut after firing (post cocturam) was discovered in the southern half of barracks no. 2 (B2) in the praetentura of the fort (pl. 3 fig. 1). No further contextual data is available regarding the find, which has until recently eluded the archaeological community, remaining hitherto unpublished. The fragment was part of a medium-sized plate, closely associated from a formal point of view with the type Curle 15 terra sigillata vessels[3]. Based on the characteristics of the fabric and slip, the vessel most likely belongs to the category of local tableware which draws on various types of the aforementioned category, known also as “imitations of terra sigillata”[4]. The graffito (pl. 4 fig. 2–3) was placed just below the carination of the vessel, running approximately parallel with it. There is a certain degree of disparity regarding the size of the letters, typically their height varying between 4 and 6 mm, except for the opening letter which is somewhat larger (8 mm). A total of 16 letters can be identified and, except the last one, all are completely preserved, albeit the graffito continued beyond the perimeter of the currently preserved sherd. In the second part of the inscription the alignment of the letters is defective, possibly suggesting that the author attempted to correct the trajectory of the text. The comprehensibility of this part of the graffito is also affected by surface damage mostly in the form of scratches which often interfere with the letters.

Regarding the palaeography, the graffito consists of relatively roughly cut capital letters, however occasionally the influence of the cursive script is noticeable (I, L, S, Z)[5]. The opening letter, Z, displays a very rare and peculiar cursive form, produced by rotating the upper half of the letter; A is in both preserved instances “open”, i.e. lacking the cross-bar; the first I is rendered through a winding vertical line; L is com­posed of two oblique strokes in a very similar fashion to the cursive variants; at this point the course of the text changes slightly, the graffito gradually getting closer to the carination, the letters which follow displaying a more casual style; S is typical for the cursive script, although fairly angular, composed of three strokes; G is very similar to an angular C, composed of three strokes ending in a serif, which was crucial in identi­fying the letter; the third I ends in two serifs as well, while the last visible letter, although poorly preserved can be identified as an A. The remaining letters (N, M, X) correspond essentially to the capital script. In addition to the letters described above, the symbol “7” (usually depicted as an inversed C) indicates the designation of a centuria. Based on the identifications outlined above, the following reading can be put forward:

“Zanax, soldier from the century of Genialis”. As expected, the inscription is an indication of ownership[6]. The name of the owner is expressed in nominative, although in most cases the genitive form is used, e.g. the graffito from Corbridge (Coria?), in Roman Britain, bearing the name of a Thracian eques of the coh. I Fida Vardullorum milliaria civium Romanorum equitata: t(urma) Vindalici, Eptacenṭ[i] (RIB II.7 2501.617).

Ownership inscriptions on pottery vessels are relatively well represented on all categories of Roman provincial sites, however the highest concentration of such graffiti can be found in military installations (forts and fortresses) as well as the adjacent civilian settlements [7]. The number of graffiti reported in Roman Dacia so far is not particularly high, even so a couple of notable instances can be mentioned. Two examples can be quoted from the auxiliary fort at Cristeşti (Mureş county), in Dacia Superior, garrison of ala I Bosporanorum: (1) IDR III.4 163: CRIISCIINTIS, that is Crescentis, on a pottery flagon; (2) IDR III.4 165: possibly SITAL (?) (presumably a Thracian name), on a pottery vessel. A further graffito, discovered in the Roman auxiliary fort at Arcobara/Ilişua (Bistriţa-Năsăud county), in Dacia Porolissensis, and recently published, was inscribed entirely in cursive letters on a pottery beaker: Prisci dupl(icarii) [8].

The category of ownership inscriptions cut after firing is mostly present on table­ware, especially bowls and plates, while their occurrence on beakers and cups is somewhat lower. Their prevalence in the military environment can be linked primarily with the communal lifestyle of the contubernia, as well as with a higher rate of literacy among soldiers[9]. The ownership marks were most likely intended to aid the possessor in identifying and recovering his vessel after the conclusion of the communal meals, and to avoid theft. In most cases the names of the owners were inscribed in an inverted position, as is the case with the vessel from Buciumi. From a technical point of view, this allowed greater access to the surface of the vessel’s wall, also increasing its stability during the process. The high number of graffiti inscribed within the area of the footstand also suggests that the vessels were usually stored in an overturned position, thus facilitating their identification [10].

The name which follows the centurial sign is easily restorable in the form of GENIẠ[LIS]. This graffito provides the fifth occurrence of the abovementioned name (OPEL II 164–165) in Roman Dacia[11]. Much more interesting, however, is the name of the owner, Zanax, of peregrine status. The name is followed by the indication of his military rank (ordinary soldier, miles), and finally by the reference to his subunit within the cohort. The popular form milis was preferred instead of miles, since the unaccented e is prone to substitution with the unaccented i[12]. This extremely rare name was recorded previously only once, on a Latin epitaph from Byzantium, dating to the first half of the 3^rd century AD (possibly in 214, on the occasion of the Parthian expedition of Caracalla), the monument being erected by an aquilifer of the legio II Adiutrix (from Pannonia Inferior, garrisoned at Aquincum), named Aur(elius) Zanax, for his deceased colleague [13]. Adam Łajtar, the last editor of the inscription, inter­preted the name as being part of the well-known series of Greek names compounded with the termination -αξ, consisting mostly of nicknames. These forms betray a familiar or expressive language (Βύβαξ, Κλέαξ, Λάλαξ, Πλάταξ, Σάλαξ, Στράβαξ and so on)[14], but the author also highlighted the difficulties of finding an etymological explanation regarding the name Ζάναξ/Zanax[15].

There is possibly another occurrence of the name Zanax, hitherto unnoticed (pl. 4 fig. 4). A catalogue of names from Rome, presumably of soldiers (CIL VI 32923, from the 3^rd century AD), contains the transcription AVRELIVS ZANA (col. B, l. 10). In fact, no other occurrence of a masculine name Zana is known up to now; the right edge of the plaque is lost, and the last letter of each nomen and cognomen is relegated to the end of the column, cf. the next name, col. B, l. 11, OTACILIVS FELI[X]; therefore, it is quite certain that the complete cognomen was in fact ZANA[X]. Together with the new example from Roman Dacia, the total number of known occurrences of this enigmatic name amounts to three, all in Latin.

Similar names starting with zan- can be traced to different linguistic areas without providing however much assistance in explaining the variant at hand: a) the peculiar name Ζάνατος, attested at Beroia in Macedonia (I.Beroia 134, line 23, from 118 BC) (LGPN IV 141) and at Beit Nattif (Pella/Betholetepha) in Judaea (IEJ 61 [2011] 210); b) the Illyrian name Zanatis, with three occurrences: twice in Dalmatia (CIL III 14620 and ILIug I 106); and a sailor from Ravenna, Murcius Zanatis (CIL XI 111); c) the Semitic name Zanis, attested at Arrabona , in Pannonia Superior (CIL III 4371 = RIU I 141), decurio of the ala I Augusta Ituraeorum sagittariorum; d) another Semitic name, Ζανα (feminine name), attested in Palaestina (SEG XXXVII 1489).

A further possibility is the Iranian origin of the name, given the Persian dialectal differentiation /x/, cf. zanax “chin”, from OP *zanaka-[16], but this is quite improbable. We may invoke the curious variant Pharnax, occurring instead of the well-attested Pharnaces/Φαρνάκης. Several instances can be reported: the freedman A. Barbius P(h)arnax at Aquileia (CIL I 3411 = I.Aquileia I 3); a legionary [Ul]p(ius) Pharnax at Viminacium, in Moesia Superior (CIL III 8110 = IMS II 51). This interesting variant occurs twice in Roman Dacia: C. Iulius Farnax, Augustalis of the municipium Aurelium Apulum in 180 AD (CIL III 986 = IDR III.5 20); a legionary Aur. Farnax at Potaissa (CIL III 7688).

It is difficult to pass judgement regarding the origin of the name at this moment. At any rate, whether the name is Greek (with an expressive suffix) [17] or Iranian, the soldier in question came most probably from a Hellenophone region, even from the Near East, being recruited and enlisted to a cohort garrisoned at Buciumi in Dacia Porolissensis, sometimes during the second century AD.

This military graffito reveals thus the names of a new soldier and centurion in the fort at Buciumi, Zanax (a miles of peregrine status) and Genialis[18]. However, since the auxiliary unit is unnamed, it is impossible to decide between the coh. II Augusta Nervia Pacensis Brittonum milliaria, garrisoned at Buciumi since Hadrian’s reign[19], and other auxiliary units present probably temporarily at Buciumi, such as coh. I Flavia Ulpia Hispanorum milliaria civium Romanorum equitata[20] and coh. I Britto­num milliaria Ulpia torquata equitata pia fidelis civium Romanorum[21], some of them mentioned on bronze labels (tesserae militares)[22].