Nigdelis Greek, Roman, and Byzantine Studies 49 (2009) 515–533 © 2009 GRBS The Gens Varinia in Macedonia: On the Serrai Decree SEG LIV 617 Pantelis Nigdelis HE GENS VARINIA is one of the approximately 560 Italic gentes recorded to have members living in the province of Macedonia.1 Nineteen Latin and thirteen Greek in- scriptions spanning the period from the first century B.C. to the third A.D. indicate that members of this gens had settled only in cities in the province, specifically Perinthus, Philippi, Thasos, Thessaloniki, and Serrai. One of these was Varinius Rebilus, who is honoured in the decree under consideration here.2 The history of the decree and its reconstitution is a long one. Charles Edson was the first to record fragments A and B, as number 727 in his Notebooks (hereafter NB), on a visit to 1 For the gentes in Macedonia see Ath. Rizakis, “L’émigration romaine en Macédoine et la communauté marchande de Thessalonique: perspectives économiques et sociales,” in Chr. Müller and Cl. Hasenohr (eds.), Les Italiens dans le monde grec IIe s. av. J.-C. – I er s. ap. J.-C. (BCH Suppl. 41 [2002]) 109– 132. 2 For the evidence on the gens Varinia see A. Tataki, The Roman Presence in Macedonia. Evidence from Personal Names (Meletemata 46 [2006]) 432–434, who however is unaware of the debate over the identification of Varinius Rebilus (his no. 18) with [- - -]nium Rebilum of Thasos and does not mention the Varinii of Perinthus (prob. 1st c. B.C.). Other names that should be added to his list are [- - -] Varinius [- - -], attested by a fragment of a sarcophagus in the necropolis of Eleutheroupolis (near Philippi) and dated to A.D. 56: Ch. Koukouli-Chrysanthaki, “ΈΈνα αρχαίο πόλισμα στην Ελευθερούπολη Νομού Καβάλας,” Tekmeria 4 (1998/9) 49–52 [AE 1999, 1445]; and Varinius Pothus and P(ublius) Variniu(s), mentioned in funerary inscriptions from Thessa- loniki: P. Nigdelis, Επιγραφικά Θεσσαλονίκεια (Thessaloniki 2006) 277 (late 2nd/early 3rd c.) and 283 (2nd half 2nd c.). For the probable relationships among the members of the gens who settled in the different cities of the province, see Nigdelis 283–287 (with corrigendum for the name Rebilus). T 516 THE GENS VARINIA IN MACEDONIA Serrai on 24 July 1937;3 they were published a few years later by Greek and Bulgarian scholars (see the lemma of the in- scription below). The decree excited interest because it was thought that the honourand, given his unusual cognomen, [- - - ]ινιος Ῥέβιλος, was the same person as the [- - -]νιος Ῥέβιλος who appears in a long inscription from Thasos, possibly dating from before A.D. 69, preserving a decree honouring a prom- inent citizen who bequeathed certain fields to the city, and a fragmentary official addendum probably relating to the execu- tion of his testament.4 The identity of the Rebilus in the two inscriptions and the determination of his gens divided scholars for many years: some thought that the gentilicium in the Thasian inscription should be [Καν]ίνιος, associating him with the fab- ulously wealthy C. Caninius Rebilus (A.D. 56+) who was consul suffectus in 37;5 some thought [Οὐαρ]ίνιος, considering him to be a member of the gens Varinia. The main argument of this latter group was the mention in the Thasos testament of at least four persons, presumably heirs or legatees of the testator, whose names appear with this gentilicium in full.6 The gens name and the identity of the person remained open questions until 2004, when Christian Habicht, on the basis of Edson’s squeeze, read the name Varinius in the first line of fr. A 3 Edson saw the two fragments “in the basement of the Gymnasium” and records that he does not know where they came from. 4 C. Dunant and J. Pouilloux, Recherches sur l’histoire et les cultes de Thasos II (Paris 1958) 76–82 no. 185 [SEG XVIII 350], with the emendations of G. Daux, “Quelques noms, quelques texts,” in Thasiaka (BCH Suppl. 5 [1979]) 351–373, at 358, 360. 5 This is the view of Dunant/Pouilloux, Recherches 81–82, followed by H.- G. Pflaum, “Histoire et cultes de Thasos,” JSav 1959, 75–88, at 80; Daux, in Thasiaka 358, 360; and D. Samsaris, “La vallée du Bas-Strymon à l’époque impérial,” Dodone 18 (1989) 203–382, at 234. 6 J. and L. Robert, Bull.épigr. 1948, 106, and 1959, 328 (p.232), followed by F. Papazoglou, “Le territoire de la colonie de Philippes,” BCH 106 (1982) 89–106, at 102 n.62 (who thinks that the full name is M. Varinius Rebilus), and Fr. Camia, “Il ‘testamento’ di Rebilus e l’epistola di Vinuleius Pataecius ai Tasii,” ZPE 146 (2004) 265–271, at 267. For other suggested completions of the name see the apparatus criticus below on line 10 of the inscription. PANTELIS NIGDELIS 517 of the Serrai decree, putting an end to the debate.7 We are also indebted to Habicht for the final and till now most complete version of the inscription, with the addition of a previously published fragment of considerable importance for compre- hending its content. As Habicht rightly observes, this fragment, which Edson found and identified on his second visit to Serrai, on 27 July 1938,8 preserves part of the beginning of the decree, which earlier editors had not realised.9 Understanding of the inscription was further aided by several restorations proposed by Angelos Chaniotis, concerning the crucial fragment C, in SEG LIV. Despite these improvements, some points in the text remain problematic. And given this inscription’s unique importance in the documentation of the social history of Serrai, an exhaustive and substantiated discussion of its reconstruction and inter- pretation is warranted.10 To enable the reader to follow the discussion more easily, I give here the full text of the decree, based on Edson’s squeezes, photographs, and transcriptions, which I was able to study at the Institute for Advanced Study ( figs. 1–4). Fr. C: Edson, NB 727 C; G. V. Kaphtantzis, Ἱστορία τῆς πόλεως τῶν Σερρῶν καὶ τῆς περιφερείας της I (Athens 1967) 96–97, no. 22 (with photograph); G. Mihailov, “Inscriptions de la Thrace égéenne,” Philologia (Sofia) 6 (1980) 9, no. 13 [SEG XXX 614]; Sam- saris, Dodone 18 (1989) 256, no. 75; Habicht, ZPE 148 (2004) 286 7 “Das Ehrendekret von Serrai für Rebilus,” ZPE 148 (2004) 283–288, esp. 287. 8 NB 727 C. Edson recorded it as “lying inside of the ἀρχειοφυλάκιον of the now abandoned old Metropolis.” 9 ZPE 148 (2004) 286. Earlier editors, e.g. Kaphtantzis, Mihailov, and Samsaris (see the lemma below) thought that the fragment came from a funerary inscription. For the position of the fragment in the inscription see n.12 below. 10 A critical edition of the (ca. 75) inscriptions from Serrai and its environs —most of them funerary—is still a desideratum. For useful collections of these inscriptions see the works of Kaphtantzis and Samsaris cited in the lemma below, and cf. L. D. Loukopoulou, “Sur la structure ethnique et sociale de Serrai à l’époque impériale,” in Poikila (Meletemata 10 [1990]) 173–187, at 181–183. 518 THE GENS VARINIA IN MACEDONIA [SEG LIV 617 (A. Chaniotis); AE 2004, 1333 (M. Sève)]. Frs. A+B: Edson, NB 727 A+B; V. Besevliev and G. Mihailov, Belomorski Pregled 1 (1942) 326, no. 18 (photograph no. 11);11 J. and L. Robert, Bull.épigr. 1948, 106; Kaphtantzis 91, no. 17 (with photo- graph); Daux, Thasiaka 356–360; Mihailov 7, no. 10 [SEG XXIX 775]; Samsaris 233–235, no. 36; Habicht 284 [SEG LIV 617; AE 2004, 1333). Cf. Dunant and Pouilloux, Recherches II 80–82 (J. and L. Robert, Bull. épigr. 1959, 328); Pflaum, JSav 1959, 80; Papazoglou, BCH 106 (1982) 102 n.62 [SEG XXXII 842]; D. S. Samsaris, Ἱστορία τῶν Σερρῶν κατὰ τὴν Ἀρχαία καὶ Ρωμαϊκὴ Ἐποχή (Thessaloniki 1999) 291, no. 184; Camia, ZPE 146 (2004) 265–271; Tataki, The Roman Presence 434, no. 18. Nigdelis, Επιγραφικά Θεσσαλονίκεια 286 n.60. – – – – – – – – – – – – – C [ ca. 2-3 ] | | | | [ ca. 12 τὸ]12 ἄγαλμα αὐτο[ῦ ca. 8 δα]- πανᾶσθαι ἐν τῆ ̣[ι ca. 9 ] 4 ἡμέρᾳ εἰς τὸ ̣ν δ[ ca. 9 ] εἰ δέ τις εἰση[γήσεται ca. 2 ] – – – – – – – – – – – – – A [ 7 ] Ο ̣ὐαρ ̣ί ̣νι ̣ο ̣ν ̣ Ῥ ̣έ ̣[βιλον] [κα]λ ̣εῖσθαι τε καὶ τοὺς ἀπὸ ̣ [τῆ]ς οἰκίας αὐτοῦ ἀπελευθέ- 4 ρου ̣ς, στηλο[γ]ρ ̣α ̣φη ̣θ ̣ῆναί τε τὸ ψήφισμα τοῦτο γενομε- vv B [.] ͡ Ν ̣Ε ̣Ι ̣ [ ca. 3 ]'/ΩΝΤΩΝΤΕ βουλευ- [τῶν] | \ Α ̣| | Ο ̣ | Δ ' ' /[ ca. 3 ] ' ' ' ^ | Τ . θ ̣ῆνα[ι] 8 [τὴν] στήλην ἀπέναντι του ̣[ 3-4 ] 11 Who are in error concerning the number of lines of the inscription, adding a further line where the two fragments A and B meet. 12 According to Habicht’s calculations (284), the average number of miss- ing letters is 21. He also gives the dimensions of the three fragments based on Edson’s measurements. The necessary completion [γενεθλίωι] in line 3 shows that his calculation is correct. Also, comparing this fact with the larger number of letters in the lines in the lower part of the stele (e.g. AB.7 and 8, with ca. 24 letters each), we may conclude that the monument tapered slightly towards the top, as was customary with Macedonian funerary steles (see e.g. the decree SEG LI 786 from Amphipolis in honour of the gymnasiarch Philippos). PANTELIS NIGDELIS 519 [ ca. 3 ]ι ̣ου ἔχουσαν ἐπιγραφὴν [Οὐαρ]ι ̣νίωι Ῥεβίλωι εὐεργέτηι [ vac. τ]ῆς πόλεως [ vac. ] 12 [ vac. ]ˉ Γ D ^ [ vac. ] C: 1 [ ] | | | | lapis; Edson discerns traces of a round letter before the four vertical lines. 2 ΑΓΑΛΜΑΥ | 6 lapis, Edson; ἄγαλμα αὐτ ̣ο ̣[ῦ - - - δα] Kaphtantzis, Mihailov, Samsaris; ἄγαλμα αὐτο[ῦ ca. 8 δα] Habicht, Sève; ἄγαλμα αὐτο[ῦ καὶ προσδα] Chaniotis. 3–4 ΠΑΝΑΣΘΑΙΕΝΤΗ ̣ [ ] / ΗΜΕΡΑ lapis, Edson; νᾶσθαι ἐν τ[ῇ τῶν Ῥοζαλίων] / ἡμέρᾳ Kaphtantzis; νᾶσθαι ἐν τ[ῇ (τῶν) Ῥόδων?] / ἡμέρᾳ Mihailov; νᾶσθαι ἐν τῆ[ι (τῶν) Ῥόδων?] / ἡμέρᾳ Samsaris; νᾶσθαι ἐν τῆ[ι ca. 9 ] / ἡμέρᾳ Habicht, Sève; νᾶσθαι ἐν τῆ[ι γενεθλίωι] / ἡμέρᾳ Chaniotis. 4 ΕΙΣΤΟ ̣ΝΔ lapis, Edson; εἰς τ[ὸ]ν δ[εῖπνον πλησίον τοῦ τάφου μου] Kaphtantzis, Mihailov, Samsaris; εἰς τ[ὴ]ν δ[ ca. 9 ] Habicht; εἰς τ[ὴ]ν δ[ιανομὴν] Chaniotis. 5 ΕΙΔΕΤΙΣΕΙΣΗ lapis, Edson; εἰ δέ τις εἰς ἡ[ρώιον Kaphtantzis; ἡ[ρῷον Mihailov, Samsaris; εἰ δέ τις εἰση[γήσεται ca. 5 ] Habicht; εἰση[γήσεται] Chaniotis. A+B: 1 ¸ | Α ̣ | | N ̣ | ¸ | N ̣ | | lapis, Edson; Ρ ̣Α ̣Ι ̣Ν ̣Ο ̣ Besevliev/Mihailov, Mihailov, Samsaris; Ο ̣ὐ ̣α ̣ρ ̣ί ̣ν ̣ι ̣ο ̣ν ̣ Ῥ ̣έ ̣[βιλον] Habicht. 2 | | ΙΣΟΛΙ lapis, Edson; Ε ̣ΙΣΘΑΙ Besevliev/Mihailov; -εῖσθαι τε J. and L. Robert; [..]ε ̣ισθαι Kaphtantzis, Mihailov, Samsaris; [..]λ ̣εῖσθαι Habicht; [κα]λ ̣εῖσθαι Chaniotis. 2–3 ΚΛΙΤΟΥΣΑΠ6[ ]ΣΟΙΚΙΑΣ lapis, Edson; ΚΑΙΤΟΥΣ ̣ΑΠ ̣Ο / [τῆ]Σ ΟΙΚΙΑΣ Besevliev/Mihailov; καὶ τοὺς ἀπὸ / [τῆ]ς οἰκίας J. and L. Robert; καὶ τοὺς ἐντὸ[ς / τῆ]ς οἰκίας Kaphtantzis; καὶ τοὺς ἀπὸ / [τῆ]ς οἰκίας Mihailov, Samsaris, Habicht. 3–4 ΑΠΕΛΕΥΘΕ/ΡΟΥ ̣Σ lapis, Edson; ΑΠΕΛΕΥΘΕ/Ρ[oυ]Σ Besevliev/Mihailov; ἀπελευθέρους J. and L. Robert; ἀπελευθέ/[ρ]ους Kaphtantzis; ἀπελευθέ/ρο[υ]ς Mihailov, Samsaris; ἀπελευ- θέ/ρους Habicht. 4 ΣΤΗΛΟ ̣[.] | ^ΦΗΟΗΝΑΙΤΕ lapis, Edson; ΣΤΗΛΟ [γρα]Φ| ̣| ̣C ̣|ΝΑΙΤΕ Besevliev/Mihailov; στηλογραφεῖναι τε Kaphtantzis; στηλο[γρ]αφηθῆναι Daux; στηλογραφηθῆναί τε Mihailov, Samsaris; στηλο [γρ]α ̣φη ̣θ ̣ῆναί τε Habicht. 5 ΤΟΨΗΦΙΣΜΑΤΟΥΤΟI lapis, Edson; ΤΟΥΗΩΙΣΜΛΙΟΥ.Ο.ΕΝΟ ̣ ΛΕ Besevliev/Mihailov; τὸ ψήφισμα τοῦ γορενομέω ̣ Kaphtantzis; τὸ ψήφισμα τοῦτο γενομε. Mihailov; τὸ ψήφισμα τοῦτο γενο Samsaris; τὸ ψήφισμα τοῦτο γενόμε Habicht. 6 [.] ͡ Ν ̣Ε ̣ | [ ca. 3 ] |/ ΩΝΤΩΝΤΕ lapis, Edson; …Ε… ̷ ΟΝΤΩΝΤΕΒΟΥΛΕΥ Besevliev/Mihailov; τῶν τε βουλευ J. and L. Robert; ς βουλευόντων τε βουλευ Kaphtantzis; - - -ε- -οντων (ή - - -ε—υον τῶν) τε βουλευ Mihailov; [βουλε]υ ̣όντων τε βουλευ Samsaris; [ν]ον ….υόντων τε βουλευ Habicht. 7 [ ] \ Λ ' ' ¸ ' Δ ' ' /[ ca. 3 ] ' ' ' L | Τ . ¯ ΗΝΑΙ[ ] lapis, Edson; ……… ͡ ΗΝΑΙ Besevliev/Mihailov; [σαμένων, ἐδεδόχθαι] τεθῆναι Kaphtantzis; [τῶν - - -]τ ̣ε ̣θῆναι Mihailov; [τῶν ἐδεδόχθαι] τ ̣ε ̣θῆναι Samsaris; [τῶν] κ ̣απ ̣..δ 8 θῆναι Habicht. 8–9 […] ΣΤΗΛΗΝΑΠΕΝΑΝΤΙ | ΟΙ[ 3-4 ] / [ ] | ΟΥ lapis, Edson; [καὶ] ΣΤΗΛΗΝ- ΑΠΕΝΑΝΤΙΤΟ[ῦ βου/λευτη]Ρ ̣ΙΟΥ Besevliev/ Mihailov; στήλην ἀπέναντι το[ῦ βουλευτη]ρ ̣ίου J. and L. Robert; στήλην ἀπέναντι το[ῦ ἡρω/ε]ίου Kaphtantzis; [τὴν] στήλην ἐπέναντι το[ῦ βου/λευτη]ρ ̣ίου Mihailov; [τὴν] στήλην ἀπέναντι το[ῦ βου/λευτη]ρ ̣ίου Samsaris; [τὴν] στήλην ἀπέναντι το[ῦ 520 THE GENS VARINIA IN MACEDONIA …] / […]ίου Habicht; [τὴν] στήλην ἀπέναντι το[ῦ Σεβα]/[στε]ίου Chaniotis. 10 |NIΩΙ lapis, Edson; [Ἀντω]ΝΙΩΙ Besevliev/Mihailov; [Οὐαρει]νίωι J. and L. Robert ; [Ἀν]ν ̣ίωι Kaphtantzis; [Γ. Καν]ι ̣νίωι Daux; [- - -]νίωι Mihailov; [Κανει?]νίωι Samsaris; [Οὐαρ]ι ̣νίωι Habicht. 11 [ ] ' ' Σ lapis, Edson; [τῆ]Σ Besevliev/Mihailov; [τ]ῆς Kaphtantzis, Mihailov; [τῆ]ς Daux, Samsaris; [τ]ῆς Habicht, Sève, Chaniotis. 12 – Γ D ^ lapis, Edson; [τῶν Σ]ε ̣ρ ̣α ̣[ίων] Kaphtantzis; [τῶν Σι]ρ ̣ρ ̣α ̣[ίων] Samsaris.13 The starting point for deciphering the decree is the word ἡμέρᾳ in C.4. On the basis of the number of missing letters (nine) the only possible completion of the preceding line is [γενεθλίωι]. As Chaniotis correctly proposes, we must accept that the decree concerned the anniversary of the birthday of Varinius Rebilus.14 The third line, where the infinitive [δα]πα- νᾶσθαι evidently depends on a verb of the type ἔδοξεν found in the decision of the boule and the demos, leaves no room for doubt that the subject is a sum of money that was to be distributed on that day. The custom of prominent citizens leaving a sum of money to their city to be distributed on the anniversary of their birthday is attested by a considerable number of inscriptions and causes no surprise: a typical example comes from Gortyn in Crete: the boule and the demos erect a statue in honour of the prominent citizen Titus Flavius Xenion (whose death is placed in the period A.D. 174–182). A later inscription on the second side of the base informs us that Xenion left money to the city to be distributed on eight different anniversaries, including his birthday and the birthdays of three other members of his family.15 13 For the views that have been expressed concerning the end of the in- scription see below. 14 Sève (AE ) thinks that the decree concerned the erection of a statue, a view that is only partially tenable, see below. 15 I.Cret. IV 300 B: στηλογ ̣ρ ̣α ̣[φία διανομῶν] τῶν καταλε[ιφθεισῶν ἐπὶ] κωδικίλλοις Φλ[α(ουίου) Ξενίωνος] ἡμερῶν η΄ … πρὸ ια΄ Καλανδ(ῶν) Δεκεμ- βρίων Φλα(ουίου) Ξενίωνος γεν[εθλίῳ], Εἰδοῖς Ὀκτωβρίαις Λαμπριοῦς καὶ Ξενοφίλου γεν[εθλίῳ], πρὸ ζ΄ Καλανδ(ῶν) Αὐγούστων Ζηνοφίλου γενεθλίῳ, πρὸ α΄ Καλανδ(ῶν) Αὐγούστων Κλ(αυδίας) Μαρκελλείνης γε[νεθλίῳ]. Cf. M. Guarducci, Epigrafia greca III (Rome 1974) 299–300, and for the prom- inent citizen J. Oliver, Hesperia 21 (1952) 397–399. Distributions of money to members of the boule of Miletos upon the anniversaries of the birthdays of private citizens and of the emperor, with a host of parallels, have been PANTELIS NIGDELIS 521 Details regarding the distribution to be made on the anniver- sary of Varinius’ birthday are given at C.4. The reconstitution of this line presents difficulties, however, since the letter deter- mining the gender of the article of the last word is imperfectly preserved. The completion εἰς τ[ὴ]ν δ[ιανομὴν X]16 is syn- tactically justified if we assume that this was how the sum to be distributed to each beneficiary was stipulated and that this sum was, as in similar instances, a single-digit number. It conflicts, however, with Edson’s reading, who discerned traces of the letter omicron on the stone, traces that are confirmed by the squeeze.17 This being the case, the most probable restoration— again taking into account the number of missing letters—is εἰς τὸ ̣ν δ[ῆμον  numerus (vel sim.)].18 Whether this sum was the whole of the bequest or the amount to be distributed to each person cannot be determined with any certainty; both gestures are documented in similar inscriptions.19 It is also possible that the terms of the bequest were recorded in the (preceding and missing) rationale to the decree. There appears to have been a connection between this distri- bution of money and the statue of the honorand mentioned at C.2 (ἄγαλμα αὐτο[ῦ]).20 One possible completion for this line is ___ collected and studied by N. Erhardt, “Ein milesische Festkalender aus severischer Zeit,” IstMitt 34 (1984) 371–404. 16 Proposed by Chaniotis, who follows Habicht’s reading. 17 Traces of a curve may be seen in the upper left and upper right portion of the letter. Traces of a round letter can also be seen in the photograph published by Kaphtantzis. 18 Denarii (assuming that the sum was expressed in denarii) could be indi- cated in a variety of ways. There is nothing to preclude the possibility that the word was indicated on the stone by an abbreviation, e.g. δην. For the expression δαπανᾶσθαι (or equivalent) εἰς + accusative of person see L. and J. Robert, Claros I (Paris 1989) 15 [SEG XXXIX 1243], IV.20–23, ἐκδεχό- μενος δὲ καὶ Ῥωμαίους (meaning the officers who visited the city) καὶ τελῶν τὴν ἰς τούτους δαπάνην ἐκ τῶν ἰδίων. 19 See Erhardt, IstMitt 34 (1984) 393 ff. 20 For the use of ἄγαλμα for statues of prominent citizens in Macedonia, see e.g. SEG XΧXV 744 with the restorations of M. B. Hatzopoulos and L. D. Loukopoulou, Recherches sur les marches orientales des Téménides (Meletemata 11 [1992]) 77–80, no. K2, lines 43–46: ἐψηφίσθαι αὐτοῦ καὶ τοῦ πατρὸς αὐτοῦ Ἀπολλωνίου καὶ τῆς μητρὸς αὐτοῦ Στραττοῦς ἑκάστου ἄγαλμα 522 THE GENS VARINIA IN MACEDONIA [καὶ προσδα]/πανᾶσθαι,21 but it is difficult to determine pre- cisely what the connection between the two might have been. Greek inscriptions of the imperial period do of course attest cities giving prominent citizens the right to erect statues, on which occasion they distributed sums of money;22 but since what we have here is a gift marking a birthday, this version (i.e. a direct distribution) does not hold. There are many possible ways of completing this line: the passage could for example refer to the place where the distribution was to be made upon the decreed day, viz. the person’s statue, as was the custom in other regions, e.g. Thyateira in Lydia, where the city honoured a distinguished citizen inter alia because he left  ͵ςφʹ πρὸς τὸ δίδοσθαι ἀπὸ τῶν τόκων αὐ ̣τῶν ἑκάστῳ βουλευτῇ καὶ τετειμημένῳ (sc. magistrate) ἐν τῇ γενεθλίῳ τοῦ ὑοῦ αὐτοῦ Αἰλιανοῦ κατ’ ἔτος · μη(νὸς) Ξανδικοῦ· ιηʹ  α΄ ἐπὶ τοῦ ἀνδριάντος αὐτοῦ.23 It could also refer more simply to the erection of a statue in honour of Rebilus, presumably in a prominent position in the city, as a precondition for the distri- bution of sums of money to the demos on the anniversary of his birthday. In this case the condition would be expressed with one of the customary formulas, e.g. ἐφ’ ᾧ τε. Consequently C.1–4 should be completed as follows: ___ λίθινον, σταθῆναι δὲ τὰ ἀγά ̣λ ̣ματα καὶ τὸ ψήφισμα τοῦτο ἔνθ’ ἂν αὐτὸς ὁ ἀγωνοθέτης ἐπισημοτάτῳ τῆς ἀγορᾶς αἱρῆται τόπωι. The possibility of this referring to the statue of an emperor can be excluded: such a statue would have been designated as ἄγαλμα Σεβαστοῦ, and it is very doubtful that in that case protective clauses ensuring the bequest would have been neces- sary. Note τοὺς ἀπὸ τῆς οἰκίας αὐτοῦ ἀπελευθέρους in AB.2–3. 21 Proposed by Chaniotis (SEG ), without explanation. 22 E.g. the honorary inscription of the third century from Tenos IG XII.5 951.12–14, ἐτείμησεν καὶ δευτέρῳ ἀνδριάντι, δόντα καὶ πάλιν πᾶσιν τοῖς κατοικοῦσιν διανομήν. 23 TAM V.2 926.8–13. Compare two examples from Philadelphia, V.3 1457 ἀναθέντα τῇ ἱερωτάτῃ βουλῇ  βφ΄ καὶ [τῷ] συνεδρίῳ τῶν πρεσβυ- τέρων  αφ΄ πρὸς τὸ ἀπὸ τῶν τόκων κατ’ ἐνιαυτὸ[ν] ἐπὶ τοῦ ἀνδριάντος τοῖς ἐλθοῦσιν ἐξ αὐτῶν δίδοσθαι διανομὴν τῇ γενεθλίῳ αὐτο[ῦ ἡ]μέρᾳ, ἥτις ἐστὶν μ[η](νὸς) Π[ε]ρειτίου η΄, and 1475 [ἀνα]θεῖσαν τῇ κρα(τίστῃ) [β]ο ̣υλ[ῇ] χ ̣[ω]ρίον πρὸς τὸ νέμεσθαι̣ τ ̣ὴν ἀπὸ αὐτοῦ πρόσοδον τοὺς βουλευτ ̣ὰς ἐν μη(νὶ) Ἀρτεμισίῳ γι΄ τ<ῇ> γενεθλίῳ ἡμέρᾳ τοῦ [ἀ]δελφοῦ πρὸ τῶν ἀ ̣νδρι- άντων. PANTELIS NIGDELIS 523 C [ 2-3 ] | | | | [ ca. 12 τὸ] ἄγαλμα αὐτο ̣[ῦ, ἐφ’ ᾧ τε / ἐφ οὗ ἀεὶ δα]- πανᾶσθαι ἐν τῆ[ι γενεθλίωι] 4 ἡμέρᾳ εἰς τὸ ̣ν δ[ῆμον X(?)]. As regards the nature of the bequest, the first editors of the Thasos decree assumed, on the basis of the reference to freed- men in A.2, that the bequest in the Serrai text was testamen- tary in nature, like that on Thasos.24 The Serrai decree and the interpretation proposed above lend significant additional weight to this assumption, since in most known cases, dona- tions in the form of distributions of money on the anniversary of a birthday are testamentary bequests that take effect after the death of the person honoured.25 In any case, it is clear that what we have here is a bequest which its author has taken care to safeguard with a series of provisions that have been incor- porated into the text of the decree. This may be concluded from the fragmentarily preserved C.5, whose wording εἰ δέ τις εἰση[γήσεται is the same as that of the first line of the Thasos decree, which prohibits any change in the use of the bequest to which it refers, rendering null and void any proposal to sell or mortgage all or part of the donated farmland.26 It is very likely that in both cases there were other safeguard clauses as well, 24 This argument is also accepted by Camio, ZPE 146 (2004) 266 n.10, although it is mistakenly attributed to the Roberts. 25 See e.g. the example from Gortyn (n.15 above), with the technical terms καταλείπω and κωδίκιλλοι, indicative of a testament. For similar examples from Macedonia see the Derriopos decree IG X.2.2 300 (A.D. 95), which provides for monetary gifts to be distributed to the boule on the anniversary of the birthday of Vettius Bolanus, which he ἀφῆκεν αὐτῇ κατὰ διαθήκην (11–12), or the votive inscription 336 (A.D. 126/7) from the same city, which tells us that the heirs of one Anthestia Fusca δηνάρια πεντα- κιχείλια ἠρίθμησαν τῇ βουλῇ ἐκ διαθήκης (2). For examples from other areas see Erhardt, IstMitt 34 (1984) 396. 26 Dunant/Pouilloux, Recherches no. 185.11–14: ἐὰν δέ τ[ις εἰση]γήσηται περί τ[ι]νος τούτων ἢ γράψῃ ἢ ἐπιψ[ηφίσῃ ἢ ἐ]νγράψῃ εἰς τὸ τῆς πόλεως γραμματοφυλα[κῖον τ]ὰ μὲν γραφέντα καὶ τὰ ψηφισθέντα ἄκυρα εἶναι. Cf. Habicht, ZPE 148 (2004) 286, who concluded from this Schutzklausel in both texts that “die beiden Urkunden ein- und derselbe Rebilus im Zentrum steht und dass die beiden Urkunden gleichzeitig sind.” 524 THE GENS VARINIA IN MACEDONIA perhaps fines to be paid to some authority (in the case of Thasos, the priests responsible for the cult of the emperor).27 The question that arises is whether the Serrai bequest was protected by other safeguards as well. It is my belief that one such safeguard is concealed in the part of the inscription that has not yet been read, lines 6–7, where the two fragments A and B meet. Further study of Edson’s squeeze and transcription leads to the following new reading and restoration:28 τὸ ψήφισμα τοῦτο, γενομέ- vv B [ν]ων ἐν[όρ]κων τῶν τε βουλευ- [τῶν] καὶ τοῦ δή [μου], καὶ τεθῆνα[ι] 8 [τὴν] στήλην On this reading the council members and the demos were required to take an oath before the decree was engraved and published. The phrase γενομένων ἐνόρκων is surprising, par- ticularly in the position it occupies, for this is its first occurrence in a Macedonian decree,29 although it is known from treaties between cities, which of course is not the case here.30 What the nature of this oath was, and how, to whom, and before whom it was to be sworn, are questions that cannot be answered with any assurance. Given, however, that the text concerns a be- 27 Lines 14–18: [τὸν δὲ] εἰπόντα ἢ {δὲ} γράψαντα ἢ ἐπιψηφίσαντα ἢ ἀναγ[ράψα]ντα τὴν γνώμην εἰς τὸ πόλεως γραμματοφ[υλ]ακῖον ὀφείλειν τοῖς τῶν Σεβαστῶν ναοῖς στατῆ[ρ]ας ἀτι[μ]ήτους [δ]ισμυρίους. 28 The main difference between these and earlier readings has to do with the letters in the words ἐνόρκων (specifically, in the squeeze the lower right- hand stroke of the letter omega—which is rendered in open form with two horizontal strokes—appears clearly, while of the kappa what can be dis- cerned is part of the downstroke and the upper part of the wedge shape) and δήμου (where the delta is unmistakeable). ἐνόρκων is the only possibility that makes sense. 29 For the decrees of Macedonian cities and their wording, see P. J. Rhodes and D. M. Lewis, The Decrees of the Greek States (Oxford 1997) 187– 194. 30 E.g. the letter of Antigonus Monophthalmus to Scepsis of 311 B.C. (OGIS 5; Welles, Royal Corres. 1; Staatsvertr. III 428) 58–61: εἰς δὲ τὸν λοιπὸν χρόνον ἐνόρκων γενομένων τῶν τε Ἑλλήνων πάντων καὶ τῶν ἐν τοῖς [π]ράγ- μασιν ὄντων μᾶλλον ἂν καὶ ἀσφαλέστε[ρ]ον διαμενεῖν τοῖς Ἕλλησιν τὴν ἐλευθερίαν. PANTELIS NIGDELIS 525 quest, it is fair to assume that the oath would have to do with complying with its provisions, which means that the question should rather be whether there are any indications that bene- factors required that beneficiary cities swear to respect the terms of the bequest. And indeed we do have such a case, in a first-century testa- ment from Κibyra, where the provisions for safeguarding a legacy include the swearing of an oath: Quintus Veranius Philagros, a prominent citizen and former high priest of the imperial cult, left the sum of 400,000 drachmas to the city to provide for the maintenance in perpetuity of the gymnasium (αἰώνιος γυμνασιαρχία). In order to safeguard this bequest the donor stipulated that every year on the date of the “vows” (κατευχῶν) the ephebes and the demos should take separate oaths before the magistrates and the scribe of the demos, swear- ing to maintain the gymnasium and its endowment δι’ αἰῶνος. According to the most likely interpretation, the day of the “vows” was the official first day of the year, when sacrifices for the health of the emperor were made in Rome and the provinces (vota annua).31 In this context it is obvious that the oath would have been made to the Tyche/Genius of the Augustus, especially since it is specified later in the text of the bequest that compensation for any infringement must be made in honour of the emperor or the senate, which suggests that the bequest was connected with the imperial cult.32 31 I.Kibyra 43.5–11: ὀμνύτωσαν δὲ καθ’ ἕκαστον ἐνιαυτὸν οἱ ἔφηβοι ἐν τῶι γυμνασίωι τὸν πάτριον ὅρκον συνφυλάξειν τὴν γυμνασιαρχίαν καὶ πάντας τοὺς πόρους αὐτῆς. ὀμνύτω δὲ καὶ ὁ δῆμος ἐν τῆι τῶν κατευχῶν ἡμέραι διὰ τῶν ἀρχόντων καὶ τοῦ γραμματέ[ω]ς τοῦ δήμου, ὡς ὑπὲρ σωτηριωδεστάτου πράγματος, τηρήσειν τὴν γυμνασιαρχίαν ταύτην καὶ τὰ χρήματα αὐτῆς. Cf. P. Herrmann, “Kaiserliche Garantie für private Stiftungen,” in W. Eck, H. Galsterer, H. Wolf (eds.), Studien zur antiken Sozialgeschichte (Cologne/Vienna 1980) 339–356, at 348–350, who however does not exclude the possibility that such a day of the “vows” might exist in the local calendar. The oath stipulated here, like that of Gythion (see below), is construed by S. R. F. Price, Rituals and Power (Cambridge 1984) 119, as an oath to the Tyche of the emperor. For the oath to the Tyche/Genius of the emperor, see Fr. Bömer, “Der Eid beim Genius des Kaisers,” Athenaeum 44 (1966) 77–133. 32 Lines 11–19: ἐὰν δέ ποτε καθ’ ὃν δήποτε οὖν τρόπον καταλυθῇ ἡ γυμ- νασιαρχία, ὑπεύθυνος ἔστω ὁ δῆμος τῶι αὐτοκράτορι καὶ τῆι συγκλήτωι εἰς 526 THE GENS VARINIA IN MACEDONIA A similar reading appears to apply to a bequest made in the Macedonian city of Kalindoia: in A.D. 87 Flavia Mysta and her children rebuilt the damaged part of the city’s Augusteum, where the family’s statues had been placed, and demanded that their fellow citizens swear an oath to the Genius Augusti to take care of the temple, and presumably of the statues as well.33 This is the earliest known instance of an oath in the name of the Tyche of the emperor in Macedonia.34 The same oath was used in the case of a bequest made at Gythion in 41/2.35 Here the donor was a wealthy woman Faenia Aromation, descendant of a freedman, member of the famous gens Faenia, and a trader in aromatic oils and perfumes. Having left the sum of 8000 denarii for the operation of the gymnasium, she entrusted her freedmen and slaves to the city and the councillors of Gythion, for them to look after and keep ___ τὸ ἀποκαταστῆσαι τὴν γυμνασιαρχίαν παρ’ ἑαυτοῦ καὶ τοὺς πόρους αὐτῆς ταῖς τειμαῖ[ς τῶ]ν Σεβαστῶν καὶ τ[ῆ]ς συνκλ[ή]του, ὥστε μένειν τὴν γυμνασι- αρχίαν ἐν τῇ πόλει καθὼς ἔταξεν Κόιντος Οὐηράνιος Φίλαγρος, with the comments of the editor (T. Corsten). 33 K. Sismanidis, ΑΕΜΘ 18 (2004) [2007] 217–218 [SEG LIV 606; AE 2004, 1329; Bull.épigr. 2006, 253; EBGR 2007, 251): ἐνευχόμεθα τὴν τῶν Σεβαστῶν τύχην τοῖς πολείταις ἡμῶν … πρόνοιαν ποιεῖσθαι τοῦ ναοῦ, ἐν ᾧ τὸ γένος ἡμῶν ἀνάκειται, ὃν κατεσκευάσαμεν ἐκ θεμελίων ἐκ τῶν ἰδίων, ὅπως μένῃ ἀκατάφθορος. Sismanidis thinks that the oath is associated with the inauguration of the temple. I believe that the text of the votive inscription is based on a prior bequest and that one (likely sole) safeguard attached to it was the oath by which the citizens of Kalindoia swore to maintain the temple and the statues of the family that were in it. 34 For other inscribed examples of this oath in Macedonia, which concern however the protection of tombs, see Nigdelis, Επιγραφικά Θεσσαλονίκεια 391 ff. 35 IG V.1 1208, with important corrections and the dating of the text by Ad. Wilhelm, Griechische Inschriften rechtlichen Inhalts [Athens 1951] 90–100 [SEG XIII 258]. Of the abundant literature on this inscription see most recently K. Harter-Uibopuu, “The Trust Fund of Phaenia Aromation (IG V.1 1208) and Imperial Gytheion,” Studia Humaniora Tartuensia 5.A.4 (2004) 1–17 (especially for the legal problems of the inscription and some different supplements from those of Wilhelm), and A. D. Rizakis, “Les affranchi(e)s sous l’Empire: Richesse, évergétisme et promotion sociale,” in V. Ana- stasiadis and P. Doukelis (eds.), Esclavage antique et discriminations socio-culturelles (Bern 2005) 233–241. PANTELIS NIGDELIS 527 from harm, presumably at the hands of third parties, both for the remainder of her life and after her death, at which time the slaves were to be freed. The security demanded was an oath to be sworn by all the citizens and the councillors in the name of all the gods and the Tyche of the Augusti.36 Given these examples, the Rebilus bequest can certainly be interpreted in the same way, that is, as a wealthy citizen requiring his com- patriots to swear by the Tyche of the emperors that they would abide by the terms of his bequest. This interpretation of ἐνόρκων γενομένων is also perfectly in keeping with the other safeguards Rebilus inserts in the Thasos decree, in the sense that they reveal a man who is strongly attached to the imperial cult: the sanctions for breach of the terms of this bequest are not restricted to payment of a mon- etary fine, but include prosecution for infamy and sacrilege towards the divine Augusti, while the bequest itself was placed under the protection of the imperial cult.37 Whether the part of the Serrai decree that has been lost contained similar pro- visions we do not know, but it certainly cannot be excluded. The mere fact of Rebilus’ particular attachment to the imperial cult is enough to warrant such an assumption. His relationship with the imperial cult appears to be confirmed by yet another of the clauses in the Serrai decree (AB.7–9), stipulating that a 36 Lines 48–56: παρακατατίθεμαι δὲ τῇ [πόλει καὶ το]ῖς συνέδροις καὶ τοὺς θρεπτούς μου καὶ ἀπ[ελευθέρους] πάντας τε καὶ πάσας. ἐνεύχομαί τε ὑμεῖν θε[οὺς πάντας] καὶ τὴν τῶν Σεβαστῶν τύχην, καὶ ζώσης ἐμοῦ [καὶ ἐὰν <ἀν>θρώπιν]όν τι πάθω, καὶ κατὰ ἄνδρα καὶ κοινῇ, τὴν ἀρίστην [τῆς βουλήσεω]ς μου καὶ ὧν ἐγὼ τειμῶ καὶ τετείμηκα θρε[πτῶν καὶ ἀπελε]υθέρων διὰ παντὸς ὑμᾶς ποιήσασθαι πρόνο[ιαν, ὅπως ἀεὶ ἀν]επείλη<π>τοι διὰ τὴν ἁπάντων ὑμῶν εἰς ἐμὲ ε[ὔνοιαν καὶ ἀνενόχλ]ητοι φυλαχθῶσιν. For the legal interpretation of the passage see Harter-Uibopuu, Studia Humaniora Tar- tuensia 5.A.4 (2004) 13–14. For the social origin of the benefactress, see Rizakis, in Esclavage 239–240. 37 Dunant/Pouilloux, Recherches no. 185.18–20: ἄτιμον εἶναι καὶ αὐτὸν καὶ γέν[ος˙ ἐνέχ]εσθαι δὲ αὐτοὺς καὶ τῆι εἰς τοὺς Σεβαστοὺς [ἀσεβείαι]; cf. L. Robert, RevPhil III.10 (1936) 136–137. To the parallels for ἀσέβεια εἰς τοὺς Σεβαστοὺς add Iscr.Cos EV 279.3–4 ἐπιτί]μιον ἔστω τᾶς ἀσε[βείας τᾶς ποτὶ τὸν] Σεβαστὸν (A.D. 42), and Aphrodisias and Rome 62.b.7 ἀσεβείᾳ τῇ πρὸς τοὺς Αὐτοκράτορας καὶ ἱεροσυλίᾳ (A.D. 180–190). For asebeia towards Sebastoi see also K. Latte, Heiliges Recht (Tübingen 1920) 95. 528 THE GENS VARINIA IN MACEDONIA copy be placed beside a building, presumably one of the most important in the city, as in the case of the Thasos decree.38 Only the ending of the name of the Serrai building is pre- served. Earlier scholars proposed [βου/λευτη]ρ ̣ίου, but this was rejected, quite properly, as too long for the available space.39 Since calculations based on careful measurement of this space show that either three or four characters are required to com- plete line 8 and possibly three for line 9,40 there are only two plausible ways of completing the word: either [Σεβα/στε]ίου41 or [Και/σαρε]ίου. The latter seems less likely, as it requires four letters at the beginning of line 9, but cannot be totally ex- cluded.42 In either case, this would become the first inscrip- tional reference to a temple of the imperial cult in the city of Serrai.43 The date of Rebilus’ presence and activity in eastern Macedonia may have a bearing on the final problem con- nected with the Serrai decree, which relates to its length. Some scholars tend towards Daux’s view that the text ends at line 11 38 Dunant/Pouilloux, Recherches no. 185.20–22, ἀνα]γραφῆναι δὲ τόδε τὸ ψήφισμα εἰς στήλας τ[ρε]ῖς [καὶ ἀνα]θεῖναι εἰς τοὺς τῶν Σεβαστῶν ναοὺς μί[αν καὶ δύο εἰς τοὺς τόπους οὓς] ἂν δόξηι τοῖς [κ]ληρονόμοις αὐτ[οῦ]. 39 This completion was proposed by Besevliev and Mihailov, who be- lieved that they could distinguish traces of the letter rho at the beginning of line 9. Kaphtantzis (91) rejects both this restoration and others like ΓΥΜΝΑΣΙΟΥ (i.e. [γυ/μνασ]ίου) or ΠΡΥΤΑΝΕΙΟΥ (i.e. [πρυ/τανε]ίου) on the grounds of length. In his criticism of Kaphtantzis’ work, Mihailov (7) insists that there are traces of a rho on the stone, but Edson’s squeeze shows no trace of any letter before the iota ΙΟΥ. 40 So also Edson’s transcription. 41 Suggested by Chaniotis (SEG ). Chr. Habicht suggests to me [βου/λε]ίου (unattested in Macedonian inscriptions), but the word is short for the space. 42 Καισαρεῖον appears occasionally in inscriptions from Macedonia, as e.g. in the fragment of an honorific text from Charakoma in Mygdonia: P. Nigdelis, “Μακεδονικά Επιγραφικά II,” Tekmeria 6 (2001) 136 [SEG LI 808] line 13, and p.140. 43 The existence of such a temple in Serrai is inferred from the inscription honouring T. Claudius Flavianus Lysimachus who served as ἀρχιερεὺς καὶ ἀγωνοθέτης τῶν Σεβαστῶν: Samsaris, Dodone 18 (1989) 235–236, no. 38; cf. F. Papazoglou, Les Villes de Macédoine à l’époque romaine (BCH Suppl. 16 [1988] 379 nn.15 and 16, and Loukopoulou, in Poikila 184. PANTELIS NIGDELIS 529 and line 12 is the beginning of a new text or possibly an ad- dendum to the first (presumably along the lines of the Thasos decree).44 Others have read the traces of the surviving letters in a way that suggests to them the ethnic [Σι]ρ ̣ρ ̣α ̣[ῖος] / [Σ]ε ̣ρ ̣α ̣[ῖος].45 None of these solutions appears to be consonant with the actual evidence. Apart from one high horizontal stroke, the three letters that I can discern on the squeeze must be read as ΕΒΑ, giving part of the word Σεβαστός.46 Since the letters in line 12 are taller and wider than those in the other lines of the inscription, as Edson had already observed,47 I would propose: [ἔτους] Σ ̣ε ̣β ̣α ̣[στοῦ X (καὶ Y)].48 This means that the decree was dated by the era of Actium. While the gain from this new reading and restoration is not as great as it would have been had the numeral been preserved, if we accept it, as I believe we should, we gain a terminus post quem of 27 B.C. for the decree. This in turn has two consequences 44 Daux, Thasiaka 358: “la ligne 11 est aussi la dernière du texte (au- dessous, on a le commencement d’un autre texte ou d’une annexe, dont les restes misérables).” He is followed by Mihailov, Philologia 6 (1980) 9 (“sur la photo chez K[aphtantzis] on distingue ˉ Γ D ^ ̂ mais on ne peut être sûr, car les lettres sont passées par l’éditeur au crayon. Peut-être il s’agit d’un autre paragraphe: [τ]ῆς πόλεως suffit,” see SEG XXIX 775), and Habicht, ZPE 148 (2004) 284. The fact that the interval between lines 11 and 12 is 0.015 m. while those between lines 1–11 is around 0.009 could support this hypothesis, but see below. 45 [Σι]ρρα[ίων] is proposed by Samsaris, Dodone 18 (1989) 235, amending the completion [Σ]ερα[ίων] given by Kaphtantzis, Ἱστορία 91, because the spelling Σεραίων is not attested until much later. 46 Of the three legible letters the following marks can be discerned: (a) the upper part of one letter, formed of one vertical and one horizontal stroke, which could be gamma or epsilon; (b) the upper rounded part of what is more likely to be a beta than a rho (because of the distance between it and the following triangular letter, which shows that the bottom part of it cannot have been a vertical stroke); (c) the upper part of a triangular letter. 47 Edson, NB 727: “The letters of l. 12 seem probably to have been larger and more widely spread than the rest of the text.” This observation is con- firmed by the squeeze. 48 For the expression ἔτους Σεβαστοῦ see IG X.2.1 130, 448 (Thessa- loniki); EKM I 103, 136 (Beroea); I.Leukopetra 74. The position of the num- eral(s) is not always the same. 530 THE GENS VARINIA IN MACEDONIA that shed new light on a pair of problems relating to Varinius’ activity in eastern Macedonia. The first has to do with the date of his presence there. Francesco Camia has recently argued that the consul named in the date formula in the addendum to the Thasos decree as [Σ]ολπίκιος Γάλβα[ς] was not necessarily the emperor Galba (Servius Sulpicius Galba), who was consul for the second time in 69 (first consulship 33), as the editors of the inscription thought, but could have been his father C. Sulpicius Galba, consul in 5 B.C.; and further reminds us that the emperor’s elder brother, who had the same name, also served as consul, in A.D. 22.49 While the revised text does not confirm any of these three versions, it does at least show that the earlier ones are just as possible as the later and certainly places Rebilus in the final years of the first century B.C. or the early part of the next century. The second issue is the provenance of the decree, which Edson professed not to know. Dunant and Pouilloux main- tained that it came from Philippi.50 This now appears to be impossible: even if we accept that we have a decretum in Greek for the colony at this very early date, it would normally be dated in the usual Roman fashion (consuls, month, day) and not using the Actian era, which as far as we know was not used in any official document at Philippi.51 This conclusion does not 49 Dunant/Pouilloux, Recherches no. 185.37–38, [- - -] εἰδοῖς Μαρτίαις [- - - Σ]ολπικίωι Γάλβᾳ υἱ[ῶι] (in accordance with Daux, Thasiaka 359), and p.79. Cf. Camio, ZPE 146 (2004) 268–269. Camio places Varinius in the last years of the first century B.C. and the beginning of the first A.D., on the basis of the name Sulpicius Galba. 50 Recherches 80 (“selon toute vraisemblance, le décret trouvé pres de Serrès emane de cette dernière cité”), on the assumption that Serrai was part of the territorium of Philippi. For an overview of older views on this complicated subject see Camia, ZPE 146 (2004) 266–267. 51 For the Philippi inscriptions, all of them in Latin, which were trans- ferred to Serrai, see Loukopoulou, in Poikila 173–187. Neither the catalogue of Philippi inscriptions published by P. Pilhofer, Philippi II (Tübingen 2000), nor SEG and AE since 2000 contain a single honorific inscription from Philippi that is dated by the Actian era (there are unfortunately no extant decrees from the founding of the colony of Philippi). See e.g. the grave inscription edited by Koukouli-Chrysanthaki, Tekmeria 4 (1998/9) 49 (with emendations and observations in AE 1999, 1445), where a Varinius [libe]r(tus) PANTELIS NIGDELIS 531 of course preclude the possibility that Rebilus was indeed one of the first Varinii to settle in the colony, in the first century B.C.,52 and that like others of his fellow Roman citizens he carried out his affairs primarily through his freedmen (see AB.2–4 of the decree) in the fertile valley of the Serrai district.53 The Serrai decree in honour of Rebilus is certainly an inter- esting text, which together with the Thasos decree shows how some Roman colonists in Macedonia developed considerable economic activity in various cities in the province, although this is attested for only a handful of similar cases.54 With the new readings proposed above, however, the document acquires a more general interest: it enters the category of inscriptions at- testing that the safeguarding of bequests was sometimes secured at the benefactor’s insistence by an oath to be sworn by the beneficiary city in the name of the Tyche/Genius of the em- peror. These texts show, finally, that the imperial ideology in the form of the imperial cult had become entrenched in the cities of the Greek-speaking world on the personal as well as the civic level. A major role in this process would have been played by distinguished and influential Roman citizens who were devoted to the emperor and invoked him in their private documents, whether they were of Greek descent like Veranius Filagros and Flavia Mysta or merchants of Italic origin like Faenia Aromation and Varinius Rebilus.55 April, 2009 Aristotle Univ. of Thessaloniki / Institute for Advanced Study pnigdeli@hist.auth.gr ___ Aug(usti) VIvir Augustalis dated his sarcophagus using the official Roman system, i.e. the name of the month and the consuls for the year (A.D. 56). 52 For the first known Varinius attested in an inscription in the Philippi district see n.51 above. 53 For the composition of Serrai society in the imperial period see Loukopoulou, in Poikila 187, who concluded that there were no Roman merchants established in Serrai and that the city’s non-Greek citizens were either veterans or freedmen from known colonial families from Philippi. 54 See Rizakis, in Les Italiens 109–132. 55 I wish to thank Chr. Habicht, G. Thür, C. Bonnet, and G. Bowersock for their comments and useful suggestions. 532 THE GENS VARINIA IN MACEDONIA Fig. 1: fr.C, photograph Fig. 2: frs. AB, photograph PANTELIS NIGDELIS 533 Fig. 3: fr.C, squeeze Fig. 4: frs. AB, squeeze