173 Arheološki vestnik 70, 2019, 173–183 Some notes on two Greek inscriptions from Montenegro Komentar k dvema grškima napisoma iz Črne gore Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ Izvleček V članku sta ponovno predstavljena grška napisa, odkrita v obalnih mestih Perast in Ulcinj v Črni gori. Že v primar- ni objavi sta bila napisa ovrednotena kot pomemben historični vir, a zaradi manjkajočih in nenatančnih najdiščnih in kronoloških podatkov je bila vrednost obeh zgodovinskih dokumentov omejena. Avtorja v prispevku podajata revidirano branje in novo interpretacijo obeh napisov ter drugačno datacijo spomenikov. Napis iz Perasta je posvetilo cestnih stra- žnikov in njihovega poveljnika, verjetno izvira iz grško govorečega mesta na južnem Jadranu. Oltar iz Ulcinja, napisan v dorskem dialektu, je združenje kamnosekov postavilo Artemidi ubijalki jelenov. Od kod spomenik izvira, ni znano. Paleografski, lingvistični in historični argumenti kažejo na kasnejši čas izdelave, kot se je sprva domnevalo, tj. na čas okoli 1. st. pr. n. št. Ključne besede: Črna gora, Perast, Ulcinj, kamniti spomeniki, grška epigrafika, 1. st. pr. n. št., peripolarh, peripoloi, grška poklicna združenja Abstract This paper discusses two Greek inscriptions found in the modern-day Montenegro, in the coastal towns of Perast and Ulcinj. Already at the time of their initial publications the inscriptions were recognized as valuable historical docu- ments. In spite of this, their usage as historical sources was limited, because their proper geographical and chronologi- cal context was lacking. In this paper we offer updated reading and interpretation of the texts, as well as an alternative dating to those that have been proposed previously. The dedication (?) from Perast was erected by peripoloi (patrolmen) and must have originated from a Greek speaking town along the southern Adriatic. The altar from Ulcinj, dedicated to Artemis Elaphabolos (the Deer Slayer), was made by an association of stone-cutters, who must have come from a Doric speaking settlement. The combined weight of paleographical, linguistic and historical evidence points to the later date for the inscriptions than it was previously assumed, most likely ca. 1 st century BC. Keywords: Montenegro, Perast, Ulcinj, stone monuments, Greek epigraphy, 1 st century BC, peripolarch, peripoloi, Greek professional associations 174 Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ 1. THE PERIPOLARCH INSCRIPTION Description of the monument (Fig. 1) The first document is the well-known “Peripo- larch inscription” from Zmajević’s collection. The text is inscribed on a limestone slab (dimensions: 28.5 × 30.5 × 15.5 cm). The inscription is poorly preserved; the end is unreadable, and the entire right portion of the text is lost. It is difficult to assess the exact width of the fragmented lines, but, if the supplement in l. 7 is accurate, there are 29 letters in the same line, 9 of which are lost. It seems that approximately half of the text is lost in ll. 1–6 and one-third in ll. 7–10. The stone block Fig. 1: “Peripolarch inscription” from Perast, kept in the lapidarum on the island of Gospa od Škrpjela (Our Lady of the Rocks), at the entrance to the Bay of Risan. Sl. 1: Napis z omembo peripolarha/peripolarhov iz mesta Perast, hranjen v lapidariju na otoku Gospa od Škrpjela na vhodu v Risanski zaliv. was probably brought to Perast, Montenegro from elsewhere, most likely as a part of the collection of antiquities of the 17 th -century archbishop and antiquarian Andrija Zmajević (1624–1694). It was lost, reused as building material, and discovered by accident after the Second W orld W ar. Following its discovery, it was transported to the lapidarum on the island of Our Lady of the Rocks (Gospa od Škrpjela), at the entrance to the Bay of Risan, where it remains today. 1 Previous publications: Rendić-Miočević 1987 (editio princeps, republished in Rendić-Miočević 1989); Cabanes 1991, 220, no. 6; Masson 1991, 353–359 (summarised in SEG 41.546); SEG 38.572 1 Rendić-Miočević 1989, 181–182. 175 Some notes on two Greek inscriptions from Montenegro (based on Cabanes’ reading); Martinović 2011, 278–279 (with errors, both in reading and inter- pretation); Łajtar, Martinović 2012, 87–91, no. 2. Date: 1 st century BC (?) Inscription 1 περιπόλαρχο[ι/ς - - - - - - - - - - - - - -] Ἐπιδοον Θείρων[ος - - - - - - - - Ἀν]- τάλλου, Μᾶρκος [- - - - - - - - - - - -] περίπολοι Γωρος [- - - - - - - - - - - ] 5 ΟΥ Τρίτος Τρίτου [- - - - - - - - - - -] ΟΛΑΠΟΣ Βάτωνος, ΑΝΕ[- - - - - -] Πλάτωρ Σαλλῆνος, Μεδεστ[ις Μεδεστι-?] νος, Μεδεστις Πλατουρίου, ΠΑ[- - - - -Δα?] ζου, Τρίτος Πιθείου, ΙΣΤ[- - - - - - - - - -] 10 ΝΟΣ. Οὗτοι οἱ περίπολοι τοῖς [- - - - - - -] ΑΠΑΙΣΤΥ ἐποίησαν ΙΟΥ[- - - - - - - - - - -] ΤΟΣ v ΑΧ vac. ΣΟΙΣ v Ο[- - - - - - - - - - - ] vac. ΣΑΛΛΑ vac. ΑΘΛ[ - - - - - - - - - - - - - - -] - - - - - - - - - - - - - - - - - - - - - - - - - - 1 probably περιπόλαρχο[ι- - - unless the lost right portion is so long to allow for a full name of a single peripolarch, perhaps followed by another title in plural, in which case Ἐπιδοον, Ἄν]ταλλος and Μᾶρκος are not peripolarchs. However, hav- ing several peripolarchs remains the simplest and therefore the preferred explanation. 2 The singular nominative is περιπολάρχης in the Greek litera- ture but περιπόλαρχος (pl. περιπόλαρχοι) in the Hellenistic inscriptions. 3 2–3 Ἀν]|τάλλου was supplemented by Rendić-Miočević and accepted by Masson and Łajtar. 5 ΟΛΑΠΟΣ is assumed by Rendić-Miočević to be a complete personal name 2 Cabanes 1991, 202 points out that it is unlikely that a small city would employ several peripolarchs: “C'est seulement dans l'inscription 6 que D. Rendic-Miocevic suppose que se rencontrent plusieurs péripolarques, com- me à Athènes (cf. IG II 2 , 204, 1. 20-21; 2973) II3 1, 292; autant ce partage de la charge apparaît nécessaire dans une grande cité comme Athènes, autant il est douteux que des communautés plus restreintes aient besoin de plusieurs péripolarques” . However, since the origin of the inscription is unknown, we are in no position to speculate on the size of the city where the peripoloi were employed. The number of guardsmen involved in the safekeeping of the territory of a city is more likely to be in a direct correlation to the size of the territory and the length of the border, rather than to the population of the city. 3 E. g. IG II 2 1193, l. 2; 2968, II l. 5; SEG 35.153; 38.175 II l. 4; 38. 521, l. 5; 41.148 l. 8 etc. (thus Ὀλαπος Βάτωνος), and this was generally accepted by other scholars (and duly recorded in LGPN IIIA as Ὄλαπος, p. 339, a singular instance). 8–9 - - -Δα]ζου is a supplement suggested by Masson, accepted by other scholars. 13 Read as vv ΣΑΛΛΑ v ΑΘΛ [vel ΑΘΑ] by Rendić-Miočević and Cabanes; vv ΣΑΛΛΑ v ΛΘΛ by Masson; . .ΛΛ . . . ΘΑ[. . . .]Ν[- - - -] by Łajtar. Translation Peripolarch(s) (Patrol commander(s))… Epidoon, son of Theiron… son of Antalos, Markos… peripoloi (patrolmen) Goros... Tritos, son of Tritos… son of Batonos… Plator, son of Salenos, Medestis, son of Medestis, Medestis, son of Platourios... Tritos, son of Pitheios… These peripoloi (patrolmen)… made... Commentary The document is significant for two main reasons. The first one is that it supplies a highly interesting list of personal names, Illyrian (Τριτος, Βάτων, Πλατωρ, Πλατουριος, Δαζος, probably Σαλλην, as well as Ἐπιδοον, qualified by Mason as “très bizarre”), Greek (Θείρων, Ἄνταλλος, Πιθειóς), or otherwise unknown (Γωρος, Ὀλαπος [?], Μεδεστις). 4 There is even a Latin name (Μᾶρκος), the importance of which will be discussed later. The second reason is the mention of peripolarchs (περιπόλαρχοι) and peripoloi (περίπολοι), otherwise rare in inscriptions. It is difficult to judge the type of document, but the most likely solution is a dedication to a deity (Illyrian or Greek?) by this group of guardsmen. Apart from the fragmentary nature of the surviving text, the main problem in any attempt at its interpretation is the lack of an established context: we know neither the place nor the time of its creation. 4 For a discussion of these names, see: Krahe 1929, 16–20, 40, 92–94, 118; Mayer 1957, 80–82, 114, 273–275, 344; Alföldy 1969, 163–164, 267, 314; Rendić-Miočević 1989, 184–185; Masson 1991, 357–359; Łajtar, Martinović 2012, 89–91. Previously unattested names are probably Illyrian too. Generally, on Illyrian names: Katičić 1962; Alföldy 1964; Katičić 1976, 178–184; Wilkes 1996, 74–87. Mason suggested that letters in l. 13 (read as ΣΑΛΛΑ) represent a variant of another indigenous name Σαλλας, which is well attested: Krahe 1929, 99; Mayer 1957, 290–291; Mas- son 1991, 359. 176 Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ Origin Several suggestions have been brought forward concerning the origin of the document. Rendić- Miočević opts for Dyrrachium or its immediate vicinity, but also considers areas further north (Scodra, Lissus) as equally likely, and does not even rule out the far south of the Illyrian lands (the region of Apollonia). 5 In line with this is the opinion of Cabanes who favours an origin “from southern Illyria”. 6 In contrast, Masson opts for a location far to the north, somewhere in the cen- tral regions of Dalmatia. 7 Łajtar proposes that the document was inscribed in the Bay of Risan itself, 8 although the first editor rejected this possibility in strong words: “We need to reconcile ourselves with the fact that the monument was brought to Perast from far away and that its contents can in no way be connected with Perast itself nor, it seems, with this historical Illyrian bay.” 9 Since direct information is lacking, the only indication of the monument’s provenance is the Illyrian personal names of the peripoloi and their commander, and this, on its own, obviously leaves too much room for speculation. However, the equally important question of the dating of the inscription did not receive the same attention. Paleography W e will now address the supposed paleo-graphical similarities between the Ashmolean tablet and the Perast stone slab. Upon closer examination, these seem to be only partial while the differences are equally significant. The two texts are, after all, inscribed on different materials, a bronze tablet op- posed to a limestone slab. On the whole, the Perast inscription seems somewhat crude in comparison. In it, Ο is in most cases smaller than the other let- ters, but not nearly so distinctively small as in the Ashmolean tablet. Δ in the Perast inscription, unlike in the Ashmolean tablet, has discreet apices formed by the elongated right hasta of the letter, elongated 5 Rendić-Miočević 1989, 186–187. 6 Cabanes 1991, 202. 7 Masson 1991, 359. 8 Łajtar, Martinović 2012, 91. 9 Rendić-Miočević 1989, 182 („Isto tako se moramo pomiriti i s činjenicom da je spomenik u Perastu odne- kud donesen i da se njegov sadržaj ne može povezati ni sa samim Perastom, niti, kako se čini, s ovim povijesnim ilirskim zalivom.“). forms being more common in the later Hellenistic and Roman times. In both inscriptions, we see Α with the broken cross-bar, which appears from the 3 rd century BC onward, but this characteristic is much more pronounced in the Perast inscription. Hastae of the letter Μ in the Ashmolean tablet are curved; in the Perast inscription, they are straight. Σ in the Perast inscription has straight upper and lower hastae, unlike their angled equivalents in the Ashmolean tablet. This transition between the two variants of the letters Μ and Σ took place during the 3 rd and 2 nd centuries BC. In the Ashmolean tablet, the right hasta of the letter Π is slightly shorter than the left one and slightly curved. In the Perast inscription both hastae are straight and of equal length. Again, it is a feature that appears in the later Hellenistic period and becomes widespread from the 1 st century BC onward. 10 To date any particular Greek inscription solely on the basis of its paleog- raphy is notoriously unreliable, and much more so when the proper context is lacking. That said, sometimes this is the only means of dating available. Judging by the letter-forms, the Ashmolean tablet most likely belongs to the earlier Hellenistic age, while the Perast inscription was probably inscribed in the late Hellenistic or the early Roman period. The 1 st century BC seems to be the safest conclusion. Date Without some chronological frame, this document (as well as any other) loses most of its value. The first editor, Rendić-Miočević, dated the text in the middle of the 2 nd century BC. His initial claim was that this inscription shares all the paleo-graphical features with the rather more famous bronze tablet in the Ashmolean museum, which mentions a single peripolarch, and which was approximately dated in the 3 rd century BC by Louis Robert. 11 10 Cf. McLean 2002, 42–45; Woodhead 1992, 64–65. 11 Robert 1955, 284: “La gravure semble être le seul argument pour la datation, en l’absence de toute donnée prosopographique. Je ne crois pas que le document remonte au IV e siècle, malgré la sobriété des caractères ; la petite dimension des omicron et des oméga suspend au sommet des lignes, les formes courbes de la barre transversale des alpha et de la haste droite des pi, le caractère général de l’ écriture me paraissent propres à faire dater ce texte de la haute époque hellénistique, disons du III e siècle. ” The tablet was originally dated to the 4 th century BC (cf. Erwerbungen des Ashmolean Museum of Art and Archaeology of Oxford 1912, Jahrbuch des Kaiserlich Deutschen Archäologischen 177 Some notes on two Greek inscriptions from Montenegro The alleged paleo-graphical similarity was strongly emphasised. “Our inscription from Perast shows all of the same characteristics”, Rendić-Miočević wrote. 12 However, somewhat contradictory to his previous statement, he settled for a significantly later date, sometime after the Third Macedonian war (171–168 BC). The reason given is that the document obviously belongs to the period in which Rome had already consolidated its presence on the eastern shores of the Adriatic (i.e., there is a Hellenized Latin name among the peripoloi). It was certainly the right line of reasoning, but we think that it was not taken quite far enough. The dating to the mid-second century BC was accepted by other scholars without question. 13 As far as the contents of the inscription are concerned, there are only two points that have any significance for the dating: the presence of the titles of peripolarchs and peripoloi (sometimes used to set terminus ante quem for the document) and the appearance of a Hellenized Latin name. The first point does not contradict the proposed dating; the second actually supports it. Peripolarchs and peripoloi are not frequently mentioned either by the ancient authors 14 or in epigraphical docu- ments. On the basis of these rare references, one may gather that peripoloi were a type of guards- or patrolmen, charged with watching and protecting the χώρα of a city. They could be citizen-soldiers Instituts 28, 1913, p. 471). It was republished by Cabanes 1976, 563, no. 41. 12 Rendić-Miočević 1989, 187 („Sve te karakteristike pokazuje i naš natpis iz Perasta.“). 13 Cabanes 1991, 220; Masson 1991, 357; Łajtar, Martinović 2012, 87. Martinović 2011, 279, understood the words of Rendić-Miočević in a very literal fashion and dated the slab immediately after the fall of the Ardiaean king Gentius (in the same year, 168/167 BC): “ As to the time of the creation of that inscription, it certainly belongs to the first half, or better still to the middle of the 2 nd century BC, immediately after the final collapse of the Illyrian state under the king Gentius, in the year 167 BC.” (our italics) („Što se tiče vremena nastanka tog natpisa, on svakako pripada prvoj polovini, ili bolje sredini II vijeka stare ere, neposredno posle definitivnog sloma ilirske države pod kraljem Gentijem, 167. godine stare ere“). There is nothing about this inscription or its contents that would suggest such a precise date. 14 All literary references to peripolarchs and peripoloi come from classical Athens: Thuc 4.67 (the Athenian peripoloi are a part of the force sent into Megaris); 8.92 (a mercenary peripolos assassinates Phrynicus); Aesh. 2.167 (peripoloi are made up from the members of Athenian youth); FHG II 112 (the same information, by Aristotle). Cf. n. 15–17. or mercenaries, or a mixture of both. 15 The institu- tion seems to be Athenian in origin, 16 but it was emulated by other Greek cities. 17 Some scholars used the very mention of peripoloi as an indica- tion that the Perast document could not have been inscribed later than the middle of the 2 nd century BC. The implication is that the peripoloi are a kind of military unit and thus a feature of independ- ent or semi-independent cities. Allegedly, with the establishment of the Roman authority on the east coast of the Adriatic and Ionian Seas, these should have ceased to exist. This is an unneces- sary assumption: the self-governing Greek cities possessed their own armed forces until late into the Imperial age. While peripolarchs and peripoloi are not recorded after the Hellenistic period, there were many analogous institutions in the autono- mous cities of the Eastern Roman Provinces. 18 Finally, there is a Latin name, Μᾶρκος. As already noted by Masson and Łajtar, it almost certainly does not form a part of the usual Roman trinominal system. Instead, it is a Roman name assimilated into the standard Greek nominal system, and it was almost certainly followed by a Greek-style patronymic. Such practice was hardly uncommon during the Roman Imperial era when Roman and Roman-derived names were widely used by the populace of the Greek-speaking provinces, but there are precious few examples of Hellenized Latin names in the 2 nd century BC (merely one or two examples of the name Μᾶρκος) and only from the late 1 st century BC do these become more widespread. If the Perast inscription is actually from the middle of the 2nd century BC, it would be one of the earliest instances of a Hellenized Latin name (and easily the earliest example of the name Μᾶρκος). According to LGPN, there are only two instances of this Hellenized name dated with any certainty before the 1 st century BC: SEG 38.478 ll. 9, 11 (Ἀφροδίσιος Μάρκου, Bouthrotos, after 15 On peripolarchs and peripoloi, see: Griffith 1935, 86–88; Robert 1955, 284–285; Cabanes 1991, 210–215; Chaniotis 2008, 105–106, 132–137. 16 There is a reference to peripoloi in the 7 th century BC Sicyon (FGH 105 F2), but it comes from a much later source and seems anachronistic, cf. Chaniotis 2008, 132, n. 110. 17 For the extensive list, see Chaniotis 2008, 132–133 and 132, n. 115. 18 Most of the evidence on the armed forces of the Greek cities under Roman rule comes from either Egypt or Asia Minor, see; Magie 1950, 647–648 (cf. 1514–1516, n. 46–47); Jones 1966, 212–213 (cf. 348–349, n. 2–4); Robert 1970, 97–108; Mitchell 1995, 195–197; Fuhrmann 2012, 66–82. 178 Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ 163 BC) and I. Délos 1924, l. 7 (Λεύκιος Μάρκου, 123/2 BC). 19 However, there are no reasons to as- sume this. The safest assumption is that the Perast inscription comes from the time when Hellenized Latin names became more common, i.e., from the 1 st century BC or later. The former date would agree well with the previous conclusion we made, based on the letter-forms. 2. INSCRIPTION MENTIONING THE ASSOCIATION OF STONE-CUTTERS Description of the monument (Fig. 2) The second text is found on the surface of a small limestone altar found in the town of Ulcinj, Montenegro (Roman Olcinium). Dimensions: 31 × 21 × 9.5 cm. The text is shallowly inscribed in three lines, inside four separate square spaces. The letters are of uneven size; those in l. 3 were cramped in the remaining space in the second, third and fourth squares. Previous publications: Mijović, Kovačević 1975, 32 (editio princeps); Martinović 2011, 284–285; Harland 2016. Date: 1 st century BC (?) Inscription 1 τὸ κοινὸν τõν λατό- μον Ἀρτέμιτι Ἐλα- φαβόλοι Translation The association of stone-cutters to Artemis Elaphabolos (the Deer Slayer). Commentary Unlike the preceding document, this short text is preserved in its entirety. In every instance, omicron is inscribed instead of omega. There 19 We deliberately exclude the rather special case of the 4 th century BC tyrant of Katane (of Italian, probably Oscan origin), Mamercus (Μάμερκος), gen. of whose name was given in IG IV 2 , 1, 95, l. 72 as Μάρκου. was a tendency of shortening the long vowels in the Greek pronunciation of the later Hellenistic and Imperial periods: thus, sometimes the long o vowel (written in Classical Greek as omega) was articulated as a short o (written as omicron) in the same position. 20 The dative Ἀρτέμιτι is a late Doric/Northwest form. 21 This form (Ἄρτεμις, Ἀρτέμιτος, Ἀρτέμιτι, etc.) is fairly frequent, both in the Peloponnese and Western Greece. 22 Origin No firm claims were ever made about the origin of the monument. 23 In fact, there is nothing that can be said with certainty about it, save that it probably comes from the wider area of the Adriatic/ Ionian Seas. It is unlikely that it was inscribed in the vicinity of its find-spot. Doric/Northwestern forms can hardly aid in determining the possible origin: there was a multitude of Doric speaking settlements on the shores of the Ionian and Adriatic Seas. The cult of Artemis Elaphabolos is also fairly common throughout the Greek world. 24 Date As with the previous example, the dating of this inscription is something of an issue. 25 To obtain 20 Petrounias 2007, 602, 604–605. Cf. McLean 2002, 350 (and n. 30). 21 The earlier editors did not recognize the dative Ἀρτέμιτι as Doric/Northwestern and assumed it was an engraver’s error. Mijović, Kovačević 1975, 32 read the text (with the aid of S. Kisas) as ΤΟ ΚΟΙΝΟΝ ΤΟΝ ΛΑΤΟΜΟΝ ΑΡΤΕΜΙ(ΔΙ) ΤΙ ΕΛΑΦΑΒΟΛΟΙ. Martinović 2011, 285 treated the absence of omega as another error that needs to be corrected, but he made several errors of his own in the process. His reading: Το κοινών τῶν λατομῶν Ἀρτέμιδι τη Ἐλαφαβόλοι (note the garbled and missing accents). 22 Cf. n. 26. 23 Mijović, Kovačević 1975, 32 assumed that the monu- ment was made locally, although by a foreign group of stonemasons, that probably came from Delphi. They gave no arguments in support of this claim. 24 Wernicke 1895, 1384; Burkert 2011, 103, 345; Budin 2016, 1–2, 19–20, 48–67, 143–144; Janda 2016, 114–115. 25 Mijović, Kovačević 1975, 32 dated the inscription between 5 th and 3 rd centuries BC, but this was based on an (unwarranted) assumption that the monument is local and that it is roughly contemporary with the construction of the walls of Olcinium (which itself was dated on the basis of the building technique). Martinović 2011, 284 made an assertion that “the inscription comes from a very early, 179 Some notes on two Greek inscriptions from Montenegro anything like a precise date based on letter-forms is nearly impossible in this case. It is a short and roughly inscribed text. At first glance, it appears to be fairly archaic, but its crudeness is deceiving. Omicron is smaller than other letters, although barely smaller in some cases. Ε is almost lunate. archaic age”, but then decided, without further explana- tion, to date it to the late 5 th or the early 4 th century BC. Both claims are unsustainable and can be rejected outright. Arnaoutoglou 2016, 7, 16 dated the inscription to 3 rd or 2 nd century BC but without stating the reasons (letter-forms?). Harland 2016 designated the document as “undated” . Α is inscribed with a single horizontal cross-bar. Hastae of the letter Μ are straight and outwardly projected. Except for the curved epsilon, these features seem to point to the early or middle Hellenistic period, but such conclusion would be anything but certain, especially given that many indicative letters are missing (there are no Π, Ξ or Σ-s). The only safe conclusion that can be gained from paleo-graphical features, as well as from the general shape of the monument, is that it postdates the Classical age. It is up to the language and the contents to furnish more. Fig. 2: Ulcinj (Olcinium), the altar of Artemis Elaphabolos, kept in the Museum of Ulcinj (a); detail of the inscription (b). Sl. 2: Ulcinj (Olcinium), oltar Artemide Elaphabolos, hrani ga muzej v Ulcinju (a); detajl napisa (b). 180 Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ The dialectical features of the text provide some, albeit very rough, indications of its chronological frame. Ἄρταμις (gen. Ἀρτάμιτος etc.) is the original Doric/Northwest form, recorded in all dialects belonging to this group, except the Cretan. As we approach the end of the Hellenistic age, it becomes increasingly common to see texts in Doric or Northwest Greek dialects infused with Attic koine elements. In the late Hellenistic and Imperial Doric inscriptions, those that are already influenced by the Attic koine, the name of the goddess changes to Ἄρτεμις (gen. Ἀρτέμιτος, dat. Ἀρτέμιτι, etc.). 26 This inscription is clearly one of those. 27 The fact that the language of the inscription is not completely assimilated to koine is less helpful for dating than it might seem: Doric inscriptions are encountered in the later Hellenistic as well as the early Imperial periods. 28 Taking all this into account, it is likely that the text was inscribed in the 1 st century BC, though the preceding century is not excluded. Judging solely on the basis of the letter-forms or the dialect employed, there is little about this text that would prevent us from assigning it to the 2 nd century BC. However, the mention of a professional association of stone-cutters makes such a dating less than likely. Professional as- sociations appear only gradually and hesitantly in the Hellenistic age, and only in certain places; contemporary private and cult associations are 26 Wernicke 1895, 1336; Bechtel 1923, 60, 118, 185, 247, 339–340, 427, 483, 537, 579, 638, 727–728, 880; Buck 1955, 24, 154. 27 Dat. Ἀρτέμιτι begins to appear in Doric/Northwest- ern dialect inscriptions in the late 3 nd century BC. Most of the examples come from the epigraphic record of the Roman era. 28 As is well known, Attic koine became the dominant Greek dialect during the early Hellenistic age, but the traditional dialects persisted for a long time. While it is indisputable that koine eventually triumphed (the Byzantine and Modern Greek are its descendants), absorbing the local dialects, this was a protracted struggle. In the most regions, koine replaced the local dialects on public monuments by the 1 st century BC, but in private inscriptions this transition is often postponed, especially in the Doric-speaking areas, while dialectal features in the koine inscriptions themselves endure even longer (to the 3 rd century AD and beyond); there were even artificial attempts at revival of the local dialects during the Imperial Roman era; cf. Buck 1955, 173–180; Bubenik 1989, 73–90; Adrados 2005, 180–184; Bubenik 2007; Horrocks 2010, 84–88. much more prominent. The number of profes- sional associations rises sharply only in the 1 st century AD; the majority of the source material concerning them (mostly inscriptions and papyri) comes from the first three centuries AD. This can be (and sometimes is) explained simply as a lack of epigraphical expression, but there are strong reasons to think that associations of workers and craftsmen, insomuch as they existed at all, were underdeveloped before the Roman age. They owe their flourishing to the favourable conditions of the Early Empire. 29 The associations of stone- cutters or quarry-workers (κοινὰ των λατόμων) are referred to very rarely and the few attestations that exist belong to the Imperial age. 30 If the altar from Olcinium were actually from the 2 nd century BC, the association of λατόμοι mentioned there would precede the earliest known example by a full two centuries. T o conclude, the very rough indications given by the letter-forms indicate an early to mid-Hellenistic date, but these are far from being conclusive. The appearance of the association of stone-cutters is, however, a strong indication of a later date, prob- ably late Hellenistic or even early Roman. The possibility that the monument and the letters were not created at the same time should be considered, the text perhaps being inscribed at a later date. 29 Arnaoutoglou 2016, 10–11. 30 Terms such as λατόμος are of course used from a very early time to designate craftsmen as such (cf. table of examples given by Ruffing 2008, 623–626); however, examples of professional associations of quarry-workers are very rare: Θρακικά 6, 1935, 302, no. 1 (mid-1 st century AD); Roesch 1982, 182–183, no. 29 (3 rd century AD, after 213); perhaps also Pralong 1980, 259–262 (4 th century AD or later), although context is uncertain. Other examples may be hidden under different terminology, because the word λιθοξοός (stone-mason, stone polisher or sculptor) was occasionally used as synonymous with λατόμος (see Robert 1960, 30–37, especially 32, n. 2 and 3; cf. Zimmer- mann 2002, 138, n. 953). There are documented examples of the associations of λιθοξόοι (cf. IGBul. II 674), though usually there is a clear distinction between the two types of craftsmen (see Pralong 1980, ll. 5–6). A general lack of information on the associations of quarry-workers should be no surprise considering the facts that such workers were among the most impoverished of the free population, and that the quarries were often worked by slaves, convicts or other types of compulsory-laborers. 181 Some notes on two Greek inscriptions from Montenegro ADRADOS, F. R. 2005, A History of the Greek Language: From Its Origins to the Present. – Leiden-Boston. ALFÖLDY , G. 1964, Die Namengebung der Urbevölkerung in der römischen Provinz Dalmatia. – Beiträge zur Namenforschung 12, 55–104. ALFÖLDY, G. 1969, Die Personennamen in der römischen Provinz Dalmatia. – Heidelberg. ARNAOUTOGLOU, I. 2016, Were there Craftsmen Asso- ciations in the Hellenistic W orld? – In: Statuts personnels et main-d’ œuvre en Méditerranée hellénistique, Colloque 17–18 mars 2016, Clermont-Ferrand, 1–12 (in print). BECHTEL, F. 1923, Die griechischen Dialekte II: die west- griechischen Dialekte. – Berlin. BUBENIK, V . 2007, The Decline of Ancient Dialects. – In: A. F. Christidis, M. Arapopoulou, M. Chriti (eds.), A History of Ancient Greek from Beginnings to Late An- tiquity, 482–485, Cambridge. BUBENIK, V. 1989, Hellenistic and Roman Greece as a Sociolinguistic Area. – Amsterdam, Philadelphia. BUCK, C. D. 1955, The Greek Dialects: Grammar, Selected Inscriptions, Glossary. – Chicago, London. BUDIN, S. L. 2016, Artemis. – London, New York. BURKERT, W . 2011 2 , Griechische Religion der archaischen und klassischen Epoche. – Stuttgart. CABANES, P. 1976, L’Épire de la mort de Pyrrhos à la conquête romaine (272–167 av. J.C.). – Paris. CABANES, P . 1991, Recherches épigraphiques en Albanie: péripolarques et peripoloi en Grèce du Nord-Ouest et in Illyrie – la période hellénistique. – Comptes rendus des séances de l’ Académie des Inscriptions et Belles-Lettres 135, 197–221. CHANIOTIS, A. 2008, Policing the Hellenistic Countryside: Realities and Ideologies. – In: H. van Wees, C. Brélaz, P . Ducrey (eds.), Sécurité collective et ordre public dans les sociétés anciennes, 103–153, Genève. FGH = Jacoby, F . 1923–1958, Die Fragmente der griechischen Historiker I–III, Berlin. FHG = Müller, K. W . L. 1871–1873, Fragmenta historicorum Graecorum I–V, Paris. FRASER, P . M., E. MATTHEWS (eds.) 1997, A Lexicon of Greek Personal Names IIIA: The Peloponnese, Western Greece, Sicily and Magna Graecia. – Oxford. FUHRMANN, C. J. 2012, Policing the Roman Empire: Soldiers, Administration and the Public Order. – Oxford. GRIFFITH, G. T. 1935, The Mercenaries of the Hellenistic World. – Cambridge. HARLAND, Ph. 2016, Dedication to Artemis Elaphabolos. – In: Ph. Harland, Associations in the Graeco-Roman World. An Expanding Collection of Inscriptions, Papyri and Other Sources in Translation (on-line edition, based on editio princeps [http://philipharland.com/greco- -roman-associations/dedication-to-artemis-elaphabo- los-by-an-association-of-stonecutters-undated/] last access 27. 12. 2018). HORROCKS, G. 2010, Greek: A History of the Language and Its Speakers. – Oxford. I. Délos = Durrbach, F . et al. (eds). 1926–1950, Inscriptions de Délos, Paris. IG II 2 = Kirchner, J. 1913–1940 2 , Inscriptiones Graecae II: Inscriptiones Atticae Euclidis anno posteriores, Berlin. IG IV 2 = Von Gaertringen, F . H. 1929, Inscriptiones Graecae IV: Inscriptiones Argolidis, Berlin. IGBul. II = Mihailov, G. 1958, Inscriptiones graecae in Bulgaria repertae II: Inscriptiones inter Danubium et Haemum repertae, Sofia. JANDA, M. 2016, Artemis mit der goldenen Spindel. – Innsbruck. JONES, A. H. M. 1966 2 , The Greek City from Alexander to Justinian. – Oxford. KATIČIĆ, R. 1962, Die illyrischen Personennamen in ihrem südöstlichen Verbreitungsgebiet. – Živa antika 12–1, 95–120. KATIČIĆ, R. 1976, Ancient Languages of the Balkans. – Hague, Paris. KRAHE, H. 1929, Lexikon altillyrischer Personennamen. – Heidelberg. ŁAJTAR, A., J. J. MARTINOVIĆ 2012, Greek Inscriptions in the Bay of Kotor. – Palamedes 7, 81–107. LGPN = Fraser, P. M. et al. (eds.). 1987–2014, A Lexicon of Greek Personal Names I–Vb, Oxford. MAGIE, D. 1950, Roman Rule in Asia Minor I–II. – Princeton. MARTINOVIĆ, J. J. 2011, Antički natpisi u Crnoj Gori. Corpus inscriptionum Latinarum et Graecarum Mon- tenegri. – Kotor. MASSON, O. 1990, À propos d’inscriptions grecques de Dalmatie. – Bulletin de Correspondance Hellénique 114, 499–512. MASSON, O. 1991, Notes épigraphiques : Thessalie et Dalmatie. – Bulletin de Correspondance Hellénique 115, 353–359. MAYER, A. 1957, Die Sprache der alten Illyrier I: Einleitung. Wörterbuch der Illyrischen Sprachreste. – Wien. MCLEAN, B. H. 2002, An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods from Alexander the Great Down to the Reign of Constantine (323 B.C – A.D. 337). – Ann Arbor. MIJOVIĆ, P., M. KOVAČEVIĆ 1975, Gradovi i utvrđenja u Crnoj Gori. – Beograd, Ulcinj. MITCHELL, S. 1995, Anatolia. Land, Men, and Gods in Asia Minor I: the Celts and the Impact of Roman Rule. – Oxford. PETROUNIAS, E. B. 2007, Development in Pronunciation during the Hellenistic Period. – In: A. F. Christidis, M. Arapopoulou, M. Chriti (eds.), A History of Ancient Greek from Beginnings to Late Antiquity, 599–609, Cambridge. PRALONG, A. 1980, T rouvailles dans une carrière phrygienne inconnue : une inscription rupestre et un sarcophage “in situ” . – Revue Archéologique, 1980/2, 251–262. RENDIĆ-MIOČEVIĆ, D. 1987, Helenistički natpis iz bivše Zmajevićeve zbirke u Perastu. – Arheološki radovi i rasprave 10, 155–165. RENDIĆ-MIOČEVIĆ, D. 1989, Iliri i antički svijet. – Split. RIDIGER, Š. 2005, Uvod u grčke dijalekte. – Novi Sad. ROBERT, L. 1955, Péripolarques. – Hellenica 10, 283–292. ROBERT, L. 1960, Épitaphes et acclamations byzantines à Corinthe. – Hellenica 11–12, 21–52. ROBERT, L. 1970, Études Anatoliennes. Recherches sur des inscriptions grecques de l’Asie Mineure. – Amsterdam. ROESCH, P. 1982, Études béotiennes. – Paris. RUFFING, K. 2008, Die berufliche Spezialisierung in Handel und Handwerk: Untersuchungen zu ihrer Entwicklung 182 Nemanja VUJČIĆ, Olga PELCER-VUJAČIĆ und zu ihren Bedingungen in der römischen Kaiserzeit im östlichen Mittelmeerraum auf der Grundlage griechischer Inschriften und Papyri II. – Rahden. SEG = Supplementum Epigraphicum Graecum, Brill, Leiden. WERNICKE, K. 1895, Artemis. – In: G. Wissowa (ed.), Paulys Realencyclopädie der classischen Altertumswis- senschaft I–3, 1336–1440, Stuttgart. WILKES, J. 1996, The Illyrians. – Oxford, Cambridge MA. WOODHEAD, A. G. 1992, The Study of Greek Inscriptions. – Norman, London. ZIMMERMANN, C. 2002, Handwerkervereine im Griechi- schen Osten des Imperium Romanum. – Mainz. Prvi od obeh napisov, na katerem se omenja peripolarh (sl. 1), je bil po drugi svetovni vojni odkrit v kraju Perast, kamor ga je najverjetne- je že v 17. st. prinesel škof in antikvar Andrija Zmajević (1624–1694). Kljub fragmentarnosti je spomenik pomemben zaradi napisa, na katerem sta izpričana v epigrafskih virih ne prav pogosta vojaška naziva: peripolarh (περιπολάρχης oz. v tem primeru περιπόλαρχος, poveljnik cestne straže) in peripolos (περίπολος, cestni stražnik). Vsekakor ni nepomemben seznam osebnih imen, ki sledi besedilu. Za postavitev spomenika so poskrbeli cestni stražniki (peripoloi) in njihov poveljnik (ali poveljniki). Zaradi tega se domneva, da gre morda za votivni spomenik (posvetilo nekemu božan- stvu?). Seznam imen kaže na raznoliko etnično in kulturno ozadje dedikantov: prevladujejo imena ilirskega izvora, manj številna so grška, nastopa pa tudi eno rimsko in več neznanih imen. Ono- mastična analiza za določitev izvora spomenika ni bila dovolj. Raziskovalci, ki so spomenik proučevali v preteklosti, so predlagali različne hipoteze, vse enako možne in enako težko dokazljive. Najverjet- neje se zdi, da spomenik izvira s širšega ilirskega območja, iz enega od krajev ob vzhodni obali Jadranskega ali Jonskega morja, peripoloi pa so bili v službi grškega (ali grško govorečega) mesta s tega območja. Spomenik iz Perasta je bil v prvi objavi na podlagi primerjave z napisom na bronasti tablici iz muzeja Ashmolean, na kateri so prav tako nave- deni peripolarhi, pa tudi zaradi omembe rimskega imena Μᾶρκος datiran v 2. st. pr. n. št., čeprav domnevne paleografske podobnosti z bronasto tablico kažejo bolj na 3. stoletje. V endar pa omemba heleniziranega rimskega imena zahteva kasnejšo Komentar k dvema grškima napisoma iz Črne gore Povzetek datacijo, ko je bila rimska navzočnost na vzhod- nih obalah Jadranskega morja močnejša, torej po tretji makedonski vojni in padcu kralja Gentija. Na drugi strani prisotnost oboroženih stražarjev, domnevnih najemnikov v službi grškega mesta, nakazuje, da napis ni izhajal iz časa direktne rim- ske nadvlade. Lokalne oborožene sile bi lahko bile izraz neodvisnosti ali delne neodvisnosti grških mestnih državic. To je bil tudi razlog za datiranje spomenika v sredino 2. st. pr. n. št. Natančna analiza spomenika je ovrgla dosedanjo datacijo. Podobnosti med tablico iz muzeja Ashmolean in spomenikom iz Perasta so zgolj delne, več je med njima opaznih razlik, med drugim material, iz katerega sta spomenika izdelana. Datacija napisa iz muzeja Ashmolean je sporna: prvotno je bila tablica postavljena v 4. st. pr. n. št., Louis Robert pa je njen nastanek s tehtnimi argumenti umestil v helenistično dobo, v 3. st. pr. n. št. Ker so he- lenizirana rimska imena v 2. st. pr. n. št. izjemno redka, bi bil napis iz Perasta z datacijo v sredino tega stoletja eden najzgodnejših primerov napisa z omembo heleniziranega rimskega imena (in skoraj zagotovo najzgodnejši primer za ime Μᾶρκος). Verjetneje je čas nastanka spomenika iz Perasta kasnejši, ko so tovrstna imena pogosteje izpričana. Če združimo paleografske in historične argumente, kažejo na pozno helenistično ali zgodnje rimsko obdobje, zelo verjetno na 1. st. pr. n. št. Omemba peripolarhov in peripoloi na napisu ni datacijsko občutljiva, saj so bile lokalne oborožene sile v službi grških mest tako v helenizmu kot v rimskem času (op. 1–19). Drugi spomenik (sl. 2), ki je prav tako zanimiv in hkrati problematičen, je bil odkrit v mestu Ulcinj 183 Komentar k dvema grškima napisoma iz Črne gore (ant. Olcinium). Oltar je v celoti ohranjen, črke so vklesane plitvo in neenakomerno. Posvečen je Artemidi Elaphabolos (strelki jelenov), postavilo pa ga je združenje kamnosekov. Podatkov o izvoru in dataciji spomenika nimamo. Podobno kot smo domnevali pri prejšnjem spomeniku, verjetno tudi ta ni bil klesan v bližini najdišča. Edini argument, ki bi morda zožil območje provenience, je narečje napisa, namreč oblike besed, značilne za dorsko/ severozahodno jezikovno skupino. A je bilo na obalah Jonskega in Jadranskega morja več naselbin z dorsko govorečim prebivalstvom. Tudi Arte- mido Elaphabolos so častili po celotnem grškem prostoru. Narečne prvine (dodani elementi atiške koine) postavljajo spomenik v pozno helenistično dobo, vsekakor ni bil izdelan pred 2. st. pr. n. št., verjetnejša je datacija v 1. st. pr. n. št. Splošne pa- leografske značilnosti se s tem zaključkom ujemajo, čeprav je besedilo precej poškodovano, manjka pa tudi nekaj izpovednih črk. Omemba poklicnega združenja kamnosekov govori proti dataciji v 3. ali 2. st. pr. n. št. Poklicna združenja se namreč pojavijo šele proti koncu helenistične dobe, nji- hova številčnost in aktivnosti pa se okrepijo šele v 1. st. n. št. Združenja kamnosekov in delavcev v kamnolomih se pojavijo razmeroma pozno, v primerjavi z drugimi združenji so tudi redkeje izpričana. Če bi bil spomenik iz mesta Ulcinj res iz 2. st. pr. n. št., bi šlo za najzgodnejši primer takšnega združenja, kar dve stoletji pred pojavom sicer znanih tovrstnih združenj. A dejansko ni raz- loga za takšno domnevo, saj vsi našteti argumenti kažejo na 1. st. pr. n. št., čeprav ni izključeno niti 1. st. n. št. (op. 20–30). Prevod: Anja Ragolič Nemanja Vujčić Faculty of Philosophy University of Belgrade nemanja.vujcic@f.bg.ac.rs Olga Pelcer-Vujačić Historical Institute University of Montenegro olgapv@ucg.ac.me