Originalveröffentlichung in: Scripta Classica Israelica 21, 2002, S. 209-242. The Jews of Aphrodisias: New Evidence and Old Problems Angelos Chaniotis 1. The 'donor inscriptions' and their puzzles1 Aphrodisias, site of a famous sanctuary of Aphrodite, important center of urban life in Roman Asia Minor, and the capital of the province of Caria in Late Antiquity,2 has at- tracted considerable attention among students of Judaism since the publication by Joyce Reynolds and Robert Tannenbaum of two important inscriptions conceming the Jewish community (Fig. 1). The two texts give the names of 68 Jews, three proselytes, and 54 theosebeis ('god-fearers'), thus attesting the existence of a large and apparently prospering Jewish community at Aphrodisias.3 Unfortunately, this is the only uncontro- versial Statement one can make about these inscriptions, which I will call — for the sake of convenience — the 'donor inscriptions'. Almost all issues related to this monument, including the date of the two texts, their relation to one another, the interpretation of the introductory text written on one of the two inscribed faces,4 the interpretation of the word patella or patelläs5 and the nature of the 'memoriaF (mnemeiori) set up 'for the relief of the people from grief ,6 the question whether the 'godfearers' {theosebeis) con- stituted a separate group (persons that attended the synagogue without being fully 1 All dates are CE, if not otherwise stated. Abbreviations of epigraphic corpora are those of the Supplementum Epigraphicum Graecum (see Index XXXI-XLV). 2 Recent studies on the urban development and history of Aphrodisias (with further bibliogra- phy): RouecM 1989; Roueche 1993; Ratte forthcoming I and II. 3 Reynolds and Tannenbaum 1987. It is difficult to estimate the number of the donors, since in a few cases it cannot be determined whether a word is a personal name or designation of an occupation. The bibliography on these inscriptions is immense (cf. SEG XXXVI 970; XXXIX 1100, 1105 and 1841; XLI 918; XLIV 862; XLV 1503). I mention only a few im­ portant studies: van der Horst 1990; Trebilco 1991: 107-10, 152-5, 179, 182f; Murphy-O'Connor 1992; Williams 1992; Botermann 1993; Bonz 1994; van Minnen 1994: 255-7. 4 Face A in Reynolds' terminology (face II, here): 9eös ßonGös ITATEAAAAO[..] l oi ÜTTOT€T<ryue|voi TTfc 6€Kav(ias) I T&V (J>iXoua0ü)[v] II twv kc TTaiT€uXoY(oüiTioi'?)l eL? dnevöriaiav I Ty ttXtiGi eKTiaa[v] I ei; L8iwv uvijua. 5 A soup kitchen for the poor: Reynolds and Tannenbaum 1987: 26-8; cf. Botermann 1993: 192-4; Levine 1999: 1009. An association of cooks or 'Imbissinhaber': van Minnen 1994: 256f. Cook-shop customer: Mussies 1991: 293-5. See next note. 6 The object of the donation has been interpreted as a philanthropical institution (if the word TOTeXXa in 1. 1 means 'soup kitchen for the poor'; cf. note 5), a funerary institution or as­ sociation (McKnight 1991: 158 note 64; cf. Williams 1992: 306-10: a synagogal triclinium initiated by a burial society), or a synagogue (G.W. Bowersock apud Feldman 1993: 575 note 116). In light of the vocabulary used {apenthesia, mnema), I regard the second Sugges­ tion as the most plausible. Scripta Classica Israelica vol. XXI 2002 pp. 209-242 210 THE JEWS OF APHRODISIAS Figure 1. Face I of the 'donor inscriptions' converted),7 and the understanding of several terms and names, have excited a great deal of controversy. In the editio princeps, Joyce Reynolds discussed in a very clear manner all the problems concerning the date and the genesis of the two texts and presented all the possible options. If she finally favored a date in the Severan period (c. 200) for both texts, she did this not without warning other scholars about the problems involved and about other possibilities (fourth or fifth Century). Many scholars have chosen, however, Reynolds and Tannenbaum 1987: 48-66. On this question see also Kant 1987: 687-90; Cohen 1989: esp. 31-3; van der Horst 1990: 169-71; Molthagen 1991: 46f.; Trebilco 1991: 152-66; Murphy-O'Connor 1992; Rajak 1992: 20f; Bonz 1994: 291-9; Lieu 1995; Rutgers 1998: 219f.; Stanton 1998: 267-91; Wander 1998: 8-12, 65-128. Mitchell's recent studies (1998, 1999b: 115-21) have substantially endorsed the view that the sebomenoi ton theon of the literary tradition are identical with the theosebeis of the inscriptions (but cf. note 49 below). ANGELOS CHANIOTIS 211 to disregard her explicit warning,8 and the attribution of both texts to the early third Century has become almost canonical.9 The question of chronology may seem a rather technical matter, but it has very important historical implications. The significance of the 'donor inscriptions' as a source for the Jewish Community at Aphrodisias — but also for the suspected influence of Mishnaic rabbis on the Jews of the Diaspora, for Jewish euergetism, for the social Status of adherents to Judaism, for the Service of Jews and sympathisers as councillors, for the popularity of Biblical names, for the attraction of the synagogue, the 'visibility' of proselytes and the tolerance of proselytism, for the mean- ing of the term theosebes (and its possible evolution), etc. — depends entirely on the historical context(s) in which we place them. The correct dating of the texts, therefore, has enormous significance for students of Judaism. In addition to this, if the 'donor in scriptions' have not hitherto been discussed in the context of the religious interaction amongst Christians, Greeks, and Jews in Late Antiquity,10 this is entirely due to the early datings. The aims of this paper are to clarify the relation between the two texts, to establish a more accurate chronology, and to present hitherto unpublished evidence for the Jewish Community at Aphrodisias in Late Antiquity, pointing out its historical implications. 2. The relation between the two texts of the 'donor inscriptions' The monument on which the two texts are inscribed is a 2.80 m high marble block (probably a free-standing stele), tapering a little towards the top.11 Two opposite faces are 46 cm wide (faces I and III, or B and D in the ed. pr.), the other two are slightly nar rower — 45 cm (faces II and IV, or A and C in the ed. pr.). The immediate temptation would be to designate the wider faces as the front and back faces and the narrower as the lateral faces. This impression is strengthened by the fact that only three of the faces are carefully smoothed, whereas one of the wider faces (face III) is neither smoothed nor inscribed; one cannot avoid the conclusion that the smoothed and inscribed face (face I, opposite to face III) is the front side and the rough and uninscribed face III is the back side, originally intended to remain invisible — e.g., to be placed against a wall. Further observations make this interpretation more attractive. The inscribed face I has a drafted margin or rebate down both sides, whereas the other inscribed face (face II, to the left of face I at right angles) lacks this treatment and has a fillet with rough-dressed treatment instead. The appearance of the inscriptions on the two faces strengthens the 8 Reynolds and Tannenbaum 1987: 22: 'the position is not, unfortunately, susceptible of proof. It may be wrong; but it seems to us likely'. 9 The early chronology is followed, e.g., by Strubbe 1989: 194f. (with regard to the Integra tion of the Jews and their Service as members of the Council); van der Horst 1990; van Minnen 1994: 255; Trebilco 1991: 152-5; Murphy-O'Connor 1992; Rajak 1992: 20; Williams 1992; Bonz 1994 (only for one of the two texts); Braun 1998; Levine 1999: 1009; Williams 1999: 93 (with regard to the popularity of Hebrew names; cf. below note 70). More sceptical Goodman 1988: 261f.; Molthagen 1991: 47 note 27; now also Williams 2000:318. 10 The 'donor inscriptions' are not mentioned in two important studies on Aphrodisias in Late Antiquity: Roueche 1989 and Trombley 1993/94. 11 Cf. the description by Reynolds in Reynolds and Tannenbaum 1987: 3 and 19. 212 THE JE WS OF APHRODISI AS impression that face I has the more clean and neat appearance that one expects for the front side of a monument: the inscription on face I begins at the very top of the block (Fig. 1), while the text on face II begins 15 cm lower (Fig. 2); the text on face I is in- scribed with carefully engraved letters, with standardised letter-heights within guide-lines; on the contrary, the text on face II lacks this care, there are no guidelines, the letter-heights vary, in a few cases the text goes beyond the right-hand margin, and the first line is oblique.12 Figure 2. Face II of the 'donor inscriptions' Reading this description one can come to only one conclusion: face I was written first, possibly while the stone was still lying on the ground; this made it possible for the 12 Van Minnen 1994: 256, has suggested that the first line was written later, but a close exami- nation of the stone confirms Reynolds' view that the first line was written together with the rest of the text on this face. ANGELOS CHANIOTIS 213 mason to draw the guide-lines and to Start inscribing the text at the very top of the block, as he could bend over it and did not have to climb a ladder. Sometime later, after the block had been set up, a mason (certainly a different one) inscribed the second text on face II; his work was impeded by the height of the block and by the fact that it was Standing. This explains why he had to Start lower, why his lines are not horizontal, and why the script gives the impression that less care was given to it. The mason of face II was not less experienced, worse paid, or simply lazy; he was working under unfavorable conditions. Why then did such an experienced epigrapher as Joyce Reynolds favor the assump- tion that the text on face II was inscribed first, but without excluding the alternative pre- sented above?13 The reason is simple and at first sight persuasive: the text on face II has a heading, that on face I Starts with a list of names. For the same reason Reynolds con- cluded, again very cautiously and without excluding other possibilities (e.g., the exis- tence of a crowning capital or even of another stele carrying an explanatory text), that the names of Jews and theosebeis inscribed on face I continue the list of donors which Starts on face II (Reynolds' face A); consequently the two texts belong to the same his- torical context. Reynolds' argument is based, however, on the assumption that the block has 'quite extensive damage'14 at the top and that only one line is missing. This is, how ever, not the case: the top of the block has indeed been broken off and, therefore, a sepa rate heading could have been written on the lost part of face I (possibly on a moulding) introducing the names of the Jews and the theosebeis and indicating the nature of their donation. This heading could still be read at a height of c. 3 m, if it had been written with slightly larger letters than the rest of the text (only 2 cm). We may, therefore, conclude that the text on face I was written first. It had a separate heading, and the commemoration of the 55 Jews and 52 theosebeis, whose names are preserved on this face, is separate (and possibly of a different nature) from the donation mentioned in the text of face II. We should now turn to the chronological relation between the two texts and the date of the text on face I. 3. The date of the 'donor inscriptions' In most references to the 'donor inscriptions' in studies concerning the history of Juda- ism the chronology cautiously proposed by Joyce Reynolds (c. 200) is accepted without comment or indeed any notice of the many doubts Reynolds herseif raised regarding her 13 Reynolds and Tannenbaum 1987: 19. Reynolds' interpretation has been generally accepted, with a few exceptions: Bonz 1994: 285-91, has reached the same conclusion as mine, but with a different argument, observing that the formulaic expression theos boethos on face III is not attested earlier than the fifth Century; for the text on face I she follows the traditional date in the early third Century. Doubts on whether the two texts belong together have also been expressed by van Minnen 1994: 255 and Ameling 1996: 31 note 4. 14 Reynolds and Tannenbaum 1987: 3. A second argument (the presence of the extremely rare name Antipeos in both texts) is not conclusive: Antipeos in face I does not necessarily be long to the same historical context nor is he necessarily related to Antipeos on face II; cf. Bonz 1994: 287. 214 THE JEWS OF APHRODISIAS early date (see notes 8-9 above). In the last few years the early chronology has been doubted by several scholars,15 and it is now time to review the dating criteria closely. The main argument for an early date is the fact that after the Constitutio Antoniniana of 212 the free population of the empire received Roman citizenship, and the recipients added to their name the Latin name Marcus Aurelius. Since none of the 123 persons listed in the 'dpnor inscriptions' has this (or any other Roman) nomen, Reynolds plau- sibly concludei that the texts were written either before 212 or long after that date, after Roman Citizen; nomenclature had been abandoned for a single-name system, i.e., in the fourth or fifth Century. She preferred the earlier date, because some Aurelii (and deriva tives of the name Aurelius) continue to appear in inscriptions of Aphrodisias in the fourth Century.16 With regard to another common, but not always conclusive, dating criterion, namely the letter forms, Reynolds rightly observed that most individual fea- tures of the palaeography can be reconciled with a date any time between c. 200 and c. 450. Some features, such as the good alignment on face I, some letter forms, and the use of stops seem earlier, but other features, such as the Variation of letter sizes, the poor alignment on face II, and the many abbreviation marks, seem later. She preferred the earlier date observing that 'with a date in the late fourth or fifth centuries it is difficult to reconcile the letter forms and still more the layout of face B [= face I]'.17 Reynolds never concealed the fact that the arguments for an early date are not con clusive. The problems of the first argument have been demonstrated by Helga Botterman, who has pointed out that the absence of Aurelii cannot be used as a dating criterion, since the use of Roman Citizen nomenclature is not consistent; many persons used the name Marcus Aurelius after 212, others did not.18 Therefore, the absence of Aurelii in the two texts cannot serve as an indication of date, let alone as proof of an early date. Similarly, it is true that individual forms of letters (and to a great extent the overall appearance of the text on face I) resemble those in Aphrodisian inscriptions of the Severan period. This does not exclude, however, a date in the fourth Century for the text on face I or the fifth Century for the text on face II.19 In my table of letter forms (Table 1), one finds forms of the letters alpha, sigma (both angular and lunate), and upsilon (with a horizontal bar) as well as an abbreviation mark which reappear in an inscription that Reynolds has plausibly dated to the fifth Century.20 If the palaeography can be reconciled with any date between c. 200 and 500, a decisive argument for a late 15 Williams 1992: 297 note 4, 301 (mid-third Century); Botermann 1993: 187-92 (fourth Cen tury); Bonz 1994 (ca. 200 for the text on face I, fifth Century for the text on face III); Mitchell 1998: 64; 1999a: 73 note 72; 1999b: 117 note 108 (fourth Century). Cf. G.W. Bowersock apud Feldman 1993: 577 note 138; Ameling 1996: 31 note 4. 16 Reynolds and Tannenbaum 1987: 20. 17 Ibid. 18 Botermann 1993: 187-9; cf. Williams 1992: 297 note 4; Bonz 1994: 286. For the Constitutio Antoniniana and the diffusion of the name Aurelius in the Greek East see Buraselis 1989: esp. 120-48. 19 Cf. Bonz 1994: 286f., who observes that the similarity of the letter forms on face II with those of the synagogue inscriptions of Sardis Supports a date in the fourth Century or later. 20 Reynolds and Tannenbaum 1987: 137f. no. 10 (here, Appendix II no. 25). For other fourth- and fifth-century inscriptions with similar letter forms as in faces I and II (esp. E, H, 2, T, and Q) see, e.g., Roueche 1989: nos. 14, 16, 17, 19, 21, 22, 32, 42 and 73. ANGELOS CHANIOTIS 215 Letter Face I FflCG II Fflce I Face II X X A A Ä A A AAA AAA B J? t /U rr r r / n TMTM rT7T A AA AA c r rr r rp E 6£ <r zc c ac c C LL r r L T TT T ZI y Ul u II Y H n n n ¥ y 0 A Kl* U v V v I i 1 1 A A A K K K K T T i AA A LA-/ A A MM M M N N abhr. sign Table 1 date is the use and the form of the abbreviation signs, which in Reynolds' words 'at first sight seem very Byzantine';21 Reynolds is also right that they 'are all attested in use by the third Century, although more freely in papyri than in inscriptions'; but in the early third Century the sign s is never used in the papyri to abbreviate words or names, and in the inscriptions it is attested in this function only from the fourth Century onwards.22 Therefore, there is no compelling reason for dating the two inscriptions to the Severan period. But are there reasons for preferring a later date? This question can be easily answered for the text on face II (face A in Reynolds' edition). For this text there are other criteria which lead to a date around the fifth Cen tury. The decisive argument has been provided by Marianne Palmer Bonz,23 who has observed that the formulaic expression theos boethos is — to the best of our knowledge — not attested earlier than the fourth Century and becomes common only after c. 350. Some other difficulties with an earlier date had already been pointed out by Reynolds 21 Reynolds and Tannenbaum 1987: 21. 22 I owe the information on the papyri to my colleague Professor D. Hagedorn; see, e.g., the indices of P.Lond. vol. I-IV. All the epigraphic examples for the abbreviation sign s from Aphrodisias are from the fourth and fifth centuries: Roueche 1989: nos. 65, 68, 72, 116, 212; cf., e.g., SEG XLVII 908 (Macedonia, fifth Century). 23 Bonz 1994: 289f. 216 THE JEWS OF APHRODISIAS herseif: the mention of a psalm-singer (1. 15) would be problematic24 and the word Palatinos (1. 11) — whether as a designation of an official, a Status, or a personal name — makes better sense in the context of Late Antiquity.25 The presence of three proselytes (11. 13, 7, and 22) would be surprising only a few years after the reinforce- ment of the anti-conversion laws under Septimius Severus.26 By calling to mind that the toleration of Christianity from 311 onwards improved the conditions of the Jews, H. Botterman has supported a date in the fourth Century.27 An even later date, however, remains quite possible (cf. below). The text on face I is certainly earlier and more difficult to date, since there are almost no internal dating criteria other than the names and occupations of the men listed here. The fact that at least 29 of them are not further identified by their father's name, but by their occupation, seems a late feature,28 but would not exclude a priori a date in the early third Century.29 The most important, and hitherto not fully exploited, dating criterion is the onomastic material. The majority of the persons in this inscription have names so typical for Late Antiquity that one would immediately be tempted to date the text on face I to the fourth Century or later.30 Reynolds recognized this problem, and in her onomastic survey she observed that many names are not attested earlier than the third 24 Reynolds and Tannenbaum 1987:46. 25 Reynolds and Tannenbaum 1987: 42f., who observe, however, that there are already first-century attestations of the Latin form palatinus to designate persons in the emperor's Service. This does not change the fact that the word is attested in Greek inscriptions only after the fourth Century. For its late use, see, e.g., Frey 1952: no. 1006; SEG XXIX 636; XXXVIII 817; XLII 639; XLIV 1599. Cf. Cotton and Geiger 1989, no. 724 verso and com- mentary ad loa The personal name Palatinos is already attested in the second Century, but it becomes common only in Late Antiquity. For isolated second- and third-century attestations see SB VI 9017 (second Century); IGSK 17, 3817 (second or third Century); IG II2 2239 1. 211 (late third Century); P.Oxy. I 43 VI,8 (third Century); for late attestations (fourth-sixth Century) see Preisigke 1922: 260 (3 cases); Foraboschi 1971: 70 (3 cases). 26 Cf. Reynolds and Tannenbaum 1987: 43-5, who doubt, however, whether Roman laws were automatically in force at Aphrodisias, a free city. This is not the place to discuss this issue, but the idea that the Aphrodisian Jews challenged the Imperial legislation in the early third Century seems to me improbable. For the anti-conversion laws of Septimius Severus and the question of their historicity see Smallwood 1976: 500-2 and Braun 1998: 154f. 27 Botermann 1993: 190-2. For the improvement of the position of the Jews after 311 cf. Noethlichs 1996: 101. For the strength of the Jewish Community in the fourth Century at Antioch see Hahn 1996. 28 Mitchell 1999a: 73 note 72. For this practice in Aphrodisias in Late Antiquity cf. Roueche' 1989: nos. 91,113, 169, 189-91, 195, 206, and 214. 29 For examples earlier than c. 212 see, e.g., SEG XL VI 737 (Beroia), 2170.2 and 33 (Terenouthis). 30 This has already been observed by Mitchell 1999a: 73 note 72, who points out that some names (Amachios, Eusebios, Heortasios, Eugenios, Praoilios, Acholios, Eutychios, Gorgo- nios, Paregorios, Gregorios, Polychronios, Politianos, Leontios, Prokopios) are not attested until the fourth Century; some of these names are in fact attested earlier (see Appendix I), but Mitchell is right in his Observation that it seems incredible that so many names typical of the fourth and fifth Century should occur in an early third-century document. ANGELOS CHANIOTIS 217 Name I-II II/III III IIK IV V-VII undated Acholios 1 2 1 1 Adolios 1 Amantios 1 1 Amazonios 1 1 6 14 Anikios 1 1 Anysios 7 Arkadios 2 1 2 7 9 1 Eugenios 4 6 10 19 32 27 4 Eupeithios 1 I Eusebios 20 20 54 77 29 5 Eutropios 2 4 11 14 9 Gorgonios 1 5 11 14 18 1 Gregorios 1 1 4 29 5 32 2 Heortasios 1 2 4 1 Manikios 1 Oxycholios 1 1 Paregorios 3 5 5 10 4 2 Patrikios 1 2 7 22 Polychronios 5 5 8 24 19 12 7 Prokopios 1 2 3 2 7 Romanos 5 1 2 8 7 16 3 Strategios 2 2 2 3 22 44 29 67 182 229 196 26 Table 2. The onomastic habit in face I of the 'donor inscriptions' This table shows the chronological distribution of some of the names attested in face I of the 'donor inscriptions' from the first to the seventh Century. For the Corpora surveyed for this table see Appendix I. None of these names is attested earlier than the first Century. The date of many inscriptions is not certain. In order to avoid a manipulation of the evidence in favor of a late date, in cases of doubtful chronology I have adopted the earlier alternative; those Christian inscriptions (with a cross or another Christian Symbol) that can not be dated securely, are regarded as belonging to the fourth Century, but may be later. I-II = c. first and second Century IIK = c. 220-300 II/III = late second or early third Century Iv = c- fourtn Century III = c. third Century V-VII = c. fifth-seventh Century Century and are rare in our evidence before Late Antiquity.31 In many cases the only early (i.e., second or third Century) attestation she could find for a name was that of a related form,32 the Latin form,33 or an attestation in Rome.34 However, the onomastic 31 Reynolds and Tannenbaum 1987: 97 (Amachios), 98 (Anysios, Gorgonios), 99 (Heortasios), 103 (Oxycholios), 106 (Arkadios), 109 (Patrikios and Prokopios), and 110 (Strategios). 32 E.g., in the case of Anysios she refers to Anytos/Anyte (ibid. 112 note 30), in the case of Heortasios to Heorte (ibid. 99), in the case of Eupeithios to Eupeithes (107), although the ending -ios is a characteristically late feature. 33 Reynolds and Tannenbaum 1987: 97 (Amantios), 105 with note 113 (Amazonius), and 109 with note 157 (Patricius). 218 THE JEWS OF APHRODISIAS habits in late second- or third-century Rome should not be regarded as representative of the Greek East, despite the wide circulation of some of the names attested in Rome; furthermore, it should not be surprising that Latin forms of names are attested earlier than their Greek Version. Still, such parallels would not be so problematic if they did not concem fully a quarter of the names attested on face I. In addition, some of the third-century attestations of a name are dated to the later part of the Century and not to the period around 200. Because of the great importance of the onomastic habit for the dating of the inscription on face I, I have undertaken a detailed treatment of the most characteristic names in Appendix I; here, I present only a summary of the results. Table 2 reveals beyond any doubt that the overall onomastic habit of this text is that of Late Antiquity (fourth Century). For 12 names we only have isolated attestations before 212, primarily in Rome; these names become common in the East only after the mid-third Century; five names (Acholios, Adolios, Anikios, Oxycholios, and Patrikios) appear in our record at least one generation after the Constitutio Antoniniana, while another four names (Amantios, Anysios, Eupeithios, and Manikios) are not attested until at least one Century after the early date. A few characteristic cases should suffice. In the onomastic lexica and corpora I have surveyed, Amazonios is attested only once before the third Century, becoming common only long after the Severan period (21 attestations); in the case of Eusebios the ratio of attestations before and after c. 200 are 20 to 180 (of which 106 are of the fourth Century or later), in the case of Eutropios 4 to 38, in the case of Gorgonios 6 to 48, in the case of Gregorios 2 to 67, in the case of Polychronios 10 to 63, in the case of Romanos 6 to 33; in all these cases the bulk of the evidence is frorn the fourth Century or later. The great number of late attestations becomes even more important if we take into consideration the fact that the number of inscriptions generally decreases after the third Century. We either have to assume that face I of the 'donor inscriptions' was in- scribed around 200 to commemorate men whose names deviated radically from the contemporary onomastic habits, or that the inscription dates to some time after c. 250. In light of all the other evidence, the latter conclusion is compelling. It seems impossible to me to come to a more accurate date for the two texts within the period we call Late Antiquity. The religious tolerance in the period between Galerius' decree (311) and the more aggressive measures for the establishment of Christianity as State religion under Theodosius I seems a plausible historical context for the commemoration of at least 55 Jews and 52 theosebeis on face I. Such a date can without difficulty be reconciled with the text's palaeographical features and with the mention of bouleutai?5 Its differences from the text on face II (palaeography, larger number of biblical names) support the assumption that the second text was inscribed much later, certainly after c. 350 (because of the acclamation th'eos boethos) and probably sometime in the fifth Century.36 34 E.g., for Acholios, Gorgonios (ibid. 98 with notes 35 and 39), Eusebios (100 with note 53), Oxycholios (103 with note 87), Paregorios (103 with note 89), Amazonios (105 with note 113), Arkadios (106 with note 121), Gregorios (107 with note 129), Eutropios (107 with note 136), Prokopios (109 with note 161). 35 For the bouleutai cf. Mitchell 1999a: 73 note 72. 36 Cf. Bonz 1994: 289f.