¹ Robinson, D. M., “A Preliminary Report on the Excavations at Pisidian Antioch and at Sizma,” AJA. Ser. II, Vol. XXVIII (1924), 435 ff.CrossRefGoogle Scholar; “Roman Sculptures from Colonia Caesarea (Pisidian Antioch),” The Art Bulletin, IX, i (1926), 5 ff.Google Scholar; “Greek and Latin Inscriptions from Asia Minor,” Trans. Amer. Phil. Assoc. LVII (1926), 195 ff.Google Scholar; “Notes on Inscriptions from Antioch in Pisidia,” JRS. XV (1925), 253 ff.Google Scholar; cf. Ramsay, W. M., “Some Inscriptions of Colonia Caesarea Antiochea,” JRS. XIV (1924), 172 ffGoogle Scholar.

² This journey in Turkey was undertaken as Thomas Whitcombe Greene Scholar of the University of Oxford and with the help of a generous grant from the Oxford University Craven Committee.

³ CIL. III, 292, 6814Google Scholar ( = ILS. 8976a), 6815, 6816 ( = ILS. 8976); JRS. VI (1916), 130Google Scholar, fig. 13; XLVIII (1958), 75 f.

⁴ CIL. III, 296 f.Google Scholar, 6835 f., 6837 ( = ILS. 5081).

⁵ CIL. III, 6810Google Scholar ff. (ILS. 7198); JRS. II (1912), 101, no. 33Google Scholar.

⁶ On the patronatus see Sebastian, E., De patronis coloniarum atque municipiorum Romanorum (Halle, 1884)Google Scholar, and Hardy, E. G., Three Spanish Charters (Oxford, 1912), 42 f.Google Scholar, note 87.

⁷ CIL. III, 6817Google Scholar ( = ILS. 998). For the dates of these governors, see Magie, D. M., Roman Rule in Asia Minor (Princeton, 1950)CrossRefGoogle Scholar, II, 1596 f., and Sherk, R. K., Legates of Galatia (The Johns Hopkins University Studies in Historical and Political Science, Ser. LXIX, no. 2, Baltimore, 1951)Google Scholar.

⁸ JRS. XIV (1924), 179 ff.Google Scholar, no. 6; XV (1925), 255 ff.

⁹ JRS. III (1913), 302, fig. 74Google Scholar; for the years of his governorship, see MAMA. VIII, 211Google Scholar.

¹⁰ See above, note 5; for his career, PIR ² A 1095. He was a senator, but not from Antioch, as his tribe alone would show (Quirina, not Sergia); perhaps then governor of Galatia. This view, advanced by SirRamsay, W. M., Social Basis of Roman Power in Asia Minor (Aberdeen, 1941), p. 27Google Scholar, is rejected by R. K. Sherk, op. cit., 109, because of the absence of the title leg. Aug. pr. pr. from the dedication. The patronatus would properly be offered after retirement from the governorship (ILS. 6087, section CXXX), and all the posts mentioned in these dedications to Arrius are domestic; omission of the legateship need cause no surprise.

¹¹ Sterrett, J. S., Epigr. Journ., 128Google Scholar; cf. JRS. VI (1916), 97Google Scholar, fig. 9a, and CIL. III, 6834Google Scholar. By the Lex Coloniae Genetivae Iuliae (ILS. 6087), section XCVII, the patronatus was inherited from the founder of the colony by his children and descendants.

¹² Année épigraphique, 1910, 36Google Scholar.

¹³ JRS. III (1913), 260Google Scholar, no. 4. This inscription does not mention the consulship that Caristanius held as suffect in 90 (Année épigraphique, 1949, 23Google Scholar).

¹⁴ For Dottius, see above, note 2; for the Flavonii, , JRS. XLVIII (1958), 74 ffGoogle Scholar.

¹⁵ JRS. III (1913), 289Google Scholar, no. 17. For an unknown patronus coloniae of uncertain date, gymnasiarch in the colony, see JRS. XIV (1924), 198, no. 32Google Scholar.

¹⁶ ILS. 6087.

¹⁷ For the functions of curiae in local elections, see the Lex Malacitana, ILS. 6089, section LII ff., and for their relation to the functions of the comitia curiata, comitia centuriata, and comitia tributa, see Hall, U., Historia XIII (1964), 273 ffGoogle Scholar.

¹⁸ ILS. 6770,6770a and b. Dean, L. R., AJA. Ser. II, Vol. XXII (1918), 196Google Scholar, regards Lilybaeum as a colonia founded by Augustus, but cf. Mommsen, T., CIL. X, p. 742Google Scholar, Ziegler, K., RE. XIII, 544Google Scholar, and Vittinghoff, F., Römische Kolonisation und Bürgerrechtspolitik unter Caesar und Augustus (Wiesbaden, 1952), 121Google Scholar; Lilybaeum seems to have been raised to colonial status by Pertinax (cf. CIL. X, 7205Google Scholar); the tribes are attested in the reign of Commodus.

¹⁹ West, A. B., Corinth, VIII, ii (Harvard, 1931), p. 91Google Scholar: almost all the tribes are named after friends and relatives of Caesar and Augustus. See also L. R. Dean, AJA., loc. cit.

²⁰ ILS. 9415.

²¹ See below, p. 56 f.

²² Curiae at Malaca (above, note 17) and at Acinipo (CIL. II, 1346Google Scholar); in the municipium at Tarentum (ILS. 6086); in African towns (ILS. III, ii, p. 681Google Scholar); in the municipium of Lanuvium (ILS. 6199); and at Turris Libisonis (ILS. 6766); centuriae at Panhormus (ILS. 5055).

²³ Juristische Schriften, I (Berlin, 1905), 213Google Scholar.

²⁴ Municipalities of the Roman Empire (Cambridge, 1913), 439Google Scholar; cf. Hardy, E. G., Three Spanish Charters (Oxford, 1912), 46Google Scholar, note 101, 100, note 8.

²⁵ Mommsen, loc. cit., attributed this to the late date of the constitutions of the African cities: the distinction between colonia and municipium had become blurred.

²⁶ The exception is Urso, but freedmen were admitted to that colony (ILS. 6087, section CV); many of them will have been of Greek or oriental origin. At Lilybaeum the Greek element seems to have been much weaker than the Punic (Mommsen, , CIL. X, p. 742)Google Scholar.

²⁷ For phylae in Greek cities, see Jones, A. H. M., The Greek City (Oxford, 1940), 158 f.Google Scholar; for their political functions, ibid., 163 ff.

²⁸ So L. R. Dean, AJA., loc. cit. The tribus would of course function in the same way as the thirty-five tribes, not like Greek phylae.

²⁹ J. S. Reid, op. cit., 438, presumably follows Liebenam, W., Städteverwaltung im römischen Kaiserreiche (Leipzig, 1900), 214Google Scholar, when he states that the Augustan colony of Berytus was also divided into curiae. If this were correct, it would support our view of the institution: Berytus was originally a Phoenician town. The evidence for Liebenam's assertion was an inscription, published by Renier, L. in Bibliothèque de l'école des hautes études XXXV (1878), 299Google Scholar, which mentions a commune tricensimae. Mommsen, republishing the inscription in CIL. III, 6671Google Scholar, rejects Renier's interpretation and shows that the commune consisted in “socii portoria exigentes Beryti”. This view is accepted by de Laet, S. J., Portorium (Brugge, 1949), 340Google Scholar.

³⁰ Pol. II, 11, iii: τὰ μὲν συσσίτια τῶν ἑταιριῶν τοῖς Φιδιτίοις.

³¹ Gsell, S., Histoire ancien de l'Afrique du Nord, II (Paris, 1920), 233Google Scholar, regards this as an “hypothèse peu sûre”. See Clermont-Ganneau, , Comptes rendus de l'académie des inscriptions, Ser. IV, Vol. XXVI (1898), 348 ffCrossRefGoogle Scholar.

³² Die Phönizier, II, i (Berlin, 1849), 479 ffGoogle Scholar.

³³ Hal., Dion, Ant. Rom., IIGoogle Scholar, 23. Both Aristotle and Dionysius compare the institution they are describing with that of the Spartans.

³⁴ E.g., Toutain, J., Les cités romaines de la Tunisie (Paris, 1895), 278 ff.Google Scholar; contra, Schmidt, J., Rheinisches Museum, XLV (1890), 599 ff.Google Scholar, and CIL. VIII, Suppl. 14683, followed by Kübler, , RE. IV, 1819 f.Google Scholar, and by Gsell, op. cit., 232. Mommsen, , Römische Geschkhte, I, 503Google Scholar, thought they might be “oligarchisch geleitete Zünfte”. Warmington, B. H., Carthage (London, 1960), 123Google Scholar, gives a verdict of non liquet.

³⁵ This point was made by Gsell, op. cit., 231 f.

³⁶ Tertullian, , Apol. XXXIX, 15, cf. XXXV, 2Google Scholar.

³⁷ At Thuburbo Maius there seem to have been eleven curiae (Merlin, A., Inscriptions latines de la Tunisie (Paris, 1944), no. 728Google Scholar).

³⁸ Sterrett, J. S., Wolfe Expedition, 352Google Scholar.

³⁹ JHS. XXIV (1904), 113Google Scholar, no. 150; republished, Social Basis of Roman Power, 185, no. 171.

⁴⁰ MAMA. VIII, p. 2Google Scholar.

⁴¹ MAMA. VIII, no. 3; originally published in Social Basis of Roman Power, 167 ff., no. 164.

⁴² For the naming of tribes, see A. H. M. Jones, op. cit, 158, 172, and 334, note 3. The names of phylae are usually in the masculine plural or in the form of feminine adjectives: see Szanto, E., Die griechischen Phylen (Wien, 1901, 72 ff.)Google Scholar; sometimes they govern the name of a deity in the genitive case: see Robert, L., Hellenica VII (1949), 194 ffGoogle Scholar. Thiasus, however, is a collective noun, and the title may at a pinch be counted a member of the first category; cf. τὸ Μέγα Συνέργιον, one of the phylae of Side (IGR. III, 810Google Scholar). At Antioch Thiasus Lib. dedicated an inscription to a governor of Galatia, CIL. III, 291Google Scholar (=6818), where Mommsen interpreted Lib. as Liberi (genitive singular) rather than as libertus; cf. the Thiasi Lib. Pat. from the territory of the Roman colony of Philippi (CIL. III, 703 f.Google Scholar), where the cult was indigenous (Collart, P., Philippes, Ville de Macédoine (Paris, 1937), 413 ff.Google Scholar).

⁴³ At Philadelphia and elsewhere in Lydia guilds were given the title phyle (A. H. M. Jones, op. cit., 162; E. Szanto, op. cit., 65 f.), and collegia were dignified with the title ἱερός (see, e.g., IGR. III, p. 656Google Scholar). There is nothing but the form of the name to distinguish τὸ Μέγα Συνέργιον at Side from its fellows οἱ βωμεīται and so on, which must have been called after striking landmarks in the city (see Robert, L., Rev. Ét. Gr., LXIV (1951), 193Google Scholar).

⁴⁴ JHS. XXIV (1904), 115Google Scholar, no. 158; CIL. III, 14400a.

⁴⁵ Social Basis of Roman Power, 184, no. 170. His reason was that “(tribus) meant Roman tribe in the Augustan coloniae.”

⁴⁶ Cf., for example, CIL. III, 6974Google Scholar, of which the last three lines read “Aquila leg. suo/pro pr. fecit/CXXII”; in 14401 a–c praetore is written in full. In CIL. the provenance of the fragment is given as Zolaera, i.e., the tepe on which Lystra stood, and it is unlikely that any inscription would have been moved to that deserted site; but in JHS. no provenance is given, and the stone is wrongly identified with CIL. III, 14400d, which is ascribed in the Corpus to Hatun Saray. In Social Basis of Roman Power the fragment is said to have been found in the village. If this is correct, there is no reason why it should not have been brought from a distance to be used in the village for building purposes.

⁴⁷ Calder, W. M., JRS. II (1912), 102Google Scholar, following Ramsay, W. M., Historical Commentary on the Galatians (ed. 2, London, 1900), 204Google Scholar. For reversion to phylae at a later date, see Ramsay, W. M., Social Basis of Roman Power, 185Google Scholar. See also A. H. M. Jones, op. cit., 172, and 338, note 30.

⁴⁸ For references, see above, notes 3 ff. The Vicus Herculis (JHS. L (1930), 272Google Scholar) is spurious. For vici at Sinope see Année épigraphique, 1916, 120Google Scholar.

⁴⁹ ILS. 6663 f. Bormann, E., CIL. XI, i, p. 77Google Scholar, suggested that the seven vici of Ariminum represent half the fourteen regiones of Rome, but Augustus' renumbering of the city districts did not take place until about 7 B.C. (RE. IA, 482), later than the foundation of Ariminum. Bormann's view that there were the same number of vici at Antioch as at Ariminum is supported by the fact that two of their vici bear the same name: Cermalus at Antioch (CIL. III, 296Google Scholar, = 6835) and at Ariminum (ILS. 6663), Velabrus at Antioch (CIL. III, 6810Google Scholar, = 289) and at Ariminum (ILS. 6661). Antioch, like Rome, Ariminum, Nemausus, and, later, Constantinople, was founded on seven hills (RE. XVI, 2295Google Scholar); hence both the seven vici and the fourteen regiones of Rome (so Mommsen, , Staatsrecht, III, i, 114Google Scholar, note 4).

⁵⁰ ILS. 1018 mentions the tenth vicus of the colony. Ten or twelve was a favourite number for Greek phylae (A. H. M. Jones, op. cit., 158).

⁵¹ Sterrett, W. S., Epigraphical Journey, p. 143Google Scholar, followed by Calder, W. M., JRS. II (1912), 102Google Scholar, suggested that the twelve mahalleler into which Yalvaç is divided correspond to twelve vici in the colony and to twelve phylae which the vici superseded. (As the result of subdivision there are now eighteen mahalleler at Yalvaç.)

⁵² Lanuvium has both vici (ILS. 5683) and curiae (ILS. 6199).

⁵³ Properly a section of a regio as at Rome (ILS. 6073) and Ostia (CIL. XIV, 4298Google Scholar). We are concerned only with the second of the three meanings ascribed to the word by van Buren, A., RE. VIIIA, 2090Google Scholar. Duncan-Jones, R. P., Historia XIII (1964), 204 f.Google Scholar, regards the vicani of Ariminum as “a series of exlusive groups, not merely an organisation of the plebs;” but his evidence (juxtaposition of vicani and collegia in CIL. XI, 377Google Scholar) is not incontrovertible.

⁵⁴ Mommsen, , Ephem. Epigr. VII (1892), p. 436Google Scholar. Antioch also had plateae: JRS. VI (1916), 106Google Scholar, no. 6 (Augusta); XIV (1924), 180, no. 6 (Tiberia).

⁵⁵ A. H. M. Jones, op. cit., 158 and 172.

⁵⁶ ILS. 1018, 2718.

⁵⁷ E.g., at Cyrene, Herodotus IV, 161. For other examples see the index in E. Szanto, Die griechischen Phylen. Note the Regio Romana at Colonia Felix Augusta Nola (ILS. 6349), where the original inhabitants lived “loco inferiore et iure honorum carentes” (Mommsen, , CIL. X, p. 142)Google Scholar. So Ramsay, W. M., Social Basis of Roman Power, 168Google Scholar, considered that at Lystra “ἱερὸς θίασος was the phyle, probably, in which the entire native population was enrolled.”

⁵⁸ ILS. 6770a and b. There were at least seven tribes at Corinth (Corinth, VIII, ii, p. 159Google Scholar).

⁵⁹ For the practice at Rome, see Cicero, , de Off. III, 20, 80Google Scholar; Seneca, , de Ira, III, 18Google Scholar, i: “M. Mario cui vicatim populus statuas posuerat;” Pliny, , Nat. Hist. XXXIII, 132Google Scholar: “tam iucunda plebei lege ut Mario Gratidiano vicatim tota statuas dicaverit;” XXXIV, 27: “statuerunt et Romae in omnibus vicis tribus Mario Gratidiano.” For the connexion of the Roman vici with the cult of the Lares Compitales, see the references collected in RE. VIII A, 2093Google Scholar.

⁶⁰ At Antioch the formula decreto decurionum is used in the Novius series, postulante populo in the Arrius Calpurnius Frontinus Honoratus series, a combination of both in the Dottius series. Mommsen, , Ephem. Epigr. VII (1892), p. 437Google Scholar, remarks that the vici do not act independently. At Corinth, on the other hand, the tribes seem to have used their own initiative in honouring patrons (see the note on Corinth, VIII, ii, 75, no. 89Google Scholar).

⁶¹ Corinth, VIII, ii, 15, no. 16Google Scholar. Agrippa may have been patron of the entire colony or, as other Corinthian tituli suggest (e.g. 50, no. 68), only of the tribe. Here at any rate the tribe seems to have been acting on instruction from the decurions.

⁶² This is a disputed point and one which cannot be argued here. See Levick, B. and Jameson, S., “C. Crepereius Gallus and his Gens,” JRS. LIV (1964)Google Scholar. For the composition of Antony's legions, one of which took part in the colonization of Antioch, see Cuntz, O., Jahrb. Öster. Arch. Inst. XXV (1929), 70 ff.Google Scholar, and, contra, Tarn, W. W., Class. Quart. XXVI (1932), 75 ffCrossRefGoogle Scholar.

⁶³ See Vittinghoff, F., Römische Kolonisation und Bürgerrechtspolitik, 24 ff.Google Scholar, and Hampl, F.. Rhein. Mus. XCV (1952), 52 ff.Google Scholar, for general principles. This is another topic too large to be dealt with here.

⁶⁴ CIL. III, 14399bGoogle Scholar (Iconium).

⁶⁵ Salutaris, ILS. 6073; Tuscus, ILS. 7575; Cermalus, , Varro, , Ling. Lat. V, 54Google Scholar Müller; the Velabrum, ILS. 7485Google Scholar. Mommsen, , Staatsrecht, III, i, 114Google Scholar, note 4, associated the Vicus Patricius with the Esquiline Hill.

⁶⁶ I hope to deal with this question at greater length elsewhere.

⁶⁷ For help with this inscription I am greatly indebted to Mr. John Matthews of The Queen's College, Oxford.

⁶⁸ Sterrett, W. S., Epigraphical Journey, 95Google Scholar; JRS. II (1912), 86 ff.Google Scholar, nos. 5 and 6.

⁶⁹ For the dates given in this paragraph, see Seeck, O., Regesten der Kaiser und Päpste (Stuttgart, 1919)Google Scholar.

⁷⁰ ILS. 777 and 5592.

⁷¹ CIL. III, 13619ff.

⁷² Symmachus, , Relationes, XIII, 2Google Scholar; ILS. 766; CIL. IX, 5946Google Scholar; Ephemeris Epigraphica IV (1881), 799Google Scholar.

⁷³ RE. VII, 1833Google Scholar.

⁷⁴ RE. VII A, 2115Google Scholar.

⁷⁵ Ammianus Marcellinus XXVII, 9, vi f.; Zosimus IV, 20. As Mr. Matthews says, the fulsome eulogy given to Valens (though we do not know what was said about Valentinian I) suggests that the inscription celebrates one of his achievements; the stone may have been crowned by a statue of the emperor. There is a fourth century male head in the Museum at Yalvaç, which looks as if it has been carved separately for insertion into its torso. Did a statue of the original second century honorand have a new head fitted in the fourth century?

⁷⁶ CIL. XI, 6665Google Scholar, seems to be relevant.

⁷⁷ CIL. XI, 6632bGoogle Scholar (Valentinian I and Valens).

⁷⁸ CIL. X, 516Google Scholar (Constantine).

⁷⁹ CIL. XI, 6669Google Scholar (Julian).

⁸⁰ CIL. XI, 6625Google Scholar (Constantius II).

⁸¹ Cf. ILS. 765 (Valentinian I) and 5555 (Valentinian I, Valens, and Gratian).

⁸² For the history of the epithet Victor, see Weinstock, S., “Victor and Invictus,” Harvard Theological Review, L (1957), 211 ffCrossRefGoogle Scholar.

⁸³ Ramsay, W. M., Classical Review, XXXIII (1919), 1 ff.CrossRefGoogle Scholar, nos. I and II, dated to 340–380 on stylistic grounds.

⁸⁴ Mr. Matthews tells me that one of the sixteen known praefecti of Egypt, 353–374, was a native of the province, and that of nineteen known vicarii and comites of the fourth century Spanish provinces two, perhaps three, are of Spanish origin. The practice was forbidden by Theodosius in 380 (Cod.lust. IX, 29, 3Google Scholar; Seeck, op. cit., 95; Jones, A. H. M., The Later Roman Empire (Oxford, 1964), I, 389)Google Scholar; presumably it was common enough to be worth prohibiting.