1. Frederiksen, M. W. in a discussion recorded in Dialoghi di Archeologia iv–v (1970–1971) 326Google Scholar.

2. CIL X 1614, 3707, 6126, 6339Google Scholar; Lugli, G., Forma Italiae 11. i. 82Google Scholar.

3. Cicero, ad Att. ii 3. 2, 4. 7Google Scholar, ad fam. vii 14. 1, ad Att. xiv 9. 1.

4. CIL V 346; X 3393.

5. The Avianii have recently been disentangled by D'Arms, J. H., ‘CIL X 1792: A Municipal Notable of the Augustan Age’, HSCP lxxvi (1972) 207Google Scholar.

6. Hatzfeld, J., ‘Les Italiens résidant à Demos’, BCH xxxvi (1912) 5CrossRefGoogle Scholar; id., Les Trafiquants Italiens dans l'Orient Hellénique (1919) 107, 228.

7. A. J. N. Wilson, Emigration from Italy in the Republican Age of Rome (1966) 96–7; S. Treggiari, Roman Freedmen During the Late Republic (1969) 138; T. P. Wiseman, New Men in the Roman Senate (1971) 199. W. W. Tarn and G. T. Griffith, Hellenistic Civilization ³ (1952) 317 are too sweeping.

8. Vitruvius vii praef. 15, 17.

9. IG III 1 561 (IG II–III² 4099). Not Δέκμος κοσσόυτιος Δέκμου, as Wilson, loc. cit., from Hatzfeld, Trafiquants 228.

10. IG III 541 (IG II–III² 3426).

11. Thompson, Homer A., ‘The Odeion in the Athenian Agora’, Hesperia xix (1950) 88Google Scholar (Mr. Ward-Perkins tells me that he entirely agrees with Professor Thompson on the Campanian influences here). This suggests that Cossutius was chosen by Antiochus for his engineering skill with roofs.

12. ILLRP 712 (105 B.C.) C. Cossutius C. 1. Gent(ius); 711, Cos?]sutius C. 1. Eud (———).

13. D'Arms, J. H., ‘Puteoli in the Second Century of the Roman Empire’, JRS lxiv (1974) 104Google Scholar: a Cossutius Rufus decurion in 187 A.D., a N. Coss[ut]ius (or perhaps Coss[in]ius – less probably Coss[on]ius or Coss[id]ius)IIvir in 59; CIL X 1784 2–3.

14. CIL IV 5396, ?5403, ?5410 (both referring to the man's discentes); P. Castrèn, Ordo Populusque Pompeianus (1975) 159.

15. Vetter, E., Handbuch der Italischen Dialekte I (1953) no. 222Google Scholar – a meddix at Volscian Velitrae. CIL X 169 (Potentia in Lucania); 1575 (Misenum); 5985 (Signia); 1945 (Puteoli, a paenularius); 3744 (Atella); 5873 (Ferentinum, the only M.); 6149 (Caieta). Add Gianetti, A., ‘Epigrafi inediti di Casinum e dintorni’, RAL xxvi (1971) 790Google Scholar = AE (1971) no. 102, a C. Cossuntius (sic) Pamphilus. By the time of Antiochus Epiphanes the Campanians had probably recovered at least the civitas sine suffragio lost in the Hannibalic War; they perhaps got it back in 189, while a few had never lost it, and some had perhaps been allowed to join the citizen colonies founded in Campania after the war (P. A. Brunt, Italian Manpower (1971) 63, with Livy xxxviii 28. 4, 36. 5).

16. ‘No trace of Roman influence shows in his work’, A. W. Lawrence, Greek Architecture (1957) 212; assuming, what is not certain, that extant columns and capitals date all or in part to his time – see Wycherly, R. E., ‘The Olympieion at Athens’ GRBS v (1964) 161Google Scholar. Lawrence, like others, thinks Cossutius could be of Greek origin, but freedman background seems unlikely.

17. IG III 2 2873 (IG II–III² 10154).

18. Mommsen, , Ephem. Epigr. i (1872) 286Google Scholar says the spelling is found till Augustus, but the only evidence he brings is the undated signature of M. Cossutius Cerdo, for which see below. Th. Eckinger, Die Orthographie Lateinischer Wörter in Griechischen Inschriften (1892) 8–11 says the last bestimmt datiert is from 45 B.C. but gives no references. I note IGRRP iv 401, referring to L. Antonius Μαάρκου ύιός in or soon after 49 B.C.; Miss J. M. Reynolds points out to me a dedication to Artemidorus of Cnidos (who tried to warn Caesar on the Ides) by a Μαάρκος Ἀιφικιος Μαάρκου Ἀπολλώνιος (Collitz, , Sammlung d. gr. Dialekt-Inschriften IV i (1889) 3527Google Scholar). B. Meinersmann, Die Lateimschen Wörter u. Namen in d. gr. Papyri (1927) 84 confirms that ‘die Schreibung ist in den Papyri riicht belegt’. In Latin the double a, an Italic feature found in various words and names, dies out about the same time: ILLRP 594 is 49–46 B.C., and M. W. Frederiksen tells me he would put 787 and 1271c about the same period; cf. M. H. Crawford, Roman Republican Coinage (1974) nos. 494 and 514, of 42 and 41 B.C.

19. Malalas viii p. 205 Bonn.

20. IGLS 825; Antioch-on-the-Orontes ii, Greek and Latin Inscriptions, 160–1, no. 90. I owe the reference to M. H. Crawford; it has not been much noted.

21. And Campbell, W. A., AJA xlii (1938) 205CrossRefGoogle Scholar thinks the name scratched before the cement was dry.

22. G. Downey, A History of Antioch in Syria (1961) 102.

23. Polybius xxvi 1.

24. Downey argues that Antiochus may have restored Seleucus I's temple of Apollo at Daphne. Livy (Polyb.) xlii 20 says that he gave money for fortifications at Megalopolis, and began a marble theatre at Tegea. The bouleuterion at Miletus was built on his behalf by two brothers from the town, his favourites and ministers Timarchus and Heracleides; there has been speculation that it was copied from or was model for that at Epiphaneia (Knackfuss, H., Milet II (1908) 99Google Scholar). The brothers were well known and influential at Rome, Diod. Sic. xxxi 27a.

25. Hatzfeld, opp. citt,; Inscr. de Délos. 1738, 1739, 1767. Competaliastae of the name would of course be freedmen.

26. Wilson, op. cit. 118–9. Inscr. de Délos 2473, a dedication to Apollo Marmarios.

27. Bürchner, IosPW IX 1930Google Scholar.

28. A critical edition of Ciriaco's remains is expected from E. W. Bodnar; meanwhile see his Cyriacus of Ancona and Athens, Coll., Latomus xliii (1960Google Scholar). Our list is IG XII 5. 11.

29. For Romans referred to by Greeks with praenomen alone, see Fraser, P. M., Ptolemaic Alexandria II (1972) 1074–5, 1116Google Scholar.

30. L. Vidman, Isis und Sarapis bei den Griechen und Römern (1970) 76 ff.; note the famous description in Apuleius, Met. xi 16Google Scholar.

31. IG XII Suppl. 557 line 25: ‘Litt. saec. la’, and again not a Julius or a Claudius to be seen.

32. Strabo C446–8; Wilson, op. cit. 98. In fact the epigraphic and numismatic evidence suggests that Eretria was more important than Chalcis: J. Day, ‘The Value of Dio Chrysostom's Euboean Discourse for the Economic Historian’, Studies in Roman Economic and Social History in Honour of Allen Chester Johnson, ed. P. R. Coleman-Norton (1951) 225.

33. Carystan marble was employed by Caesar's prefect Mamurra for his house, Pliny, NH xxxvi 48Google Scholar; R. Gnoli, Marmora Romana (1971) 157–9 connects a Byzantine reference to Chalcidic marble with the quarries c. 3 km. from Eretria but c. 20 from Chalcis, the stone from which he does not find used in Rome before the Flavians.

34. Hatzfeld, Trafiquants 107, from J. Keil, Jahreshefte des öst. Arch. Inst. in Wien, Beiblatt, 59 no. 25 (a P. Avianius C.f. is no. 21), who observes that the inscription (in Greek) does not suggest a rich family. He dates it to the 1st century A.D.On the quarries at Teos, Dio Chrys. lxxix 2; D. Magie, Roman Rule in Asia Minor (1950) 808 n. 52; and now Ballance, M. H., ‘The Origin of africano’, PBSR xxxiv (1966) 79Google Scholar, and Gnoli, op. cit. 147–151, showing this was the place of origin of ‘africano’ (variegated grey, white and red) used in the earlier first century B.C. on the Basilica Aemilia (and identified by Gnoli with ‘Lucullan’ marble); as well as of a less valuable grey marble.

Has Rabiria any connection with the great negotiator C. Rabirius Postumus, defended by Cicero? There are too many Volumnii for it to be possible to identify her husband further.

35. IGRRP IV 1092; R. Herzog, Koische Forschungen und Funde (1899) 117. After 88 B.C. many Delian merchants may have migrated to Cos. Grey marble thence, Gnoli op. cit. 152, apparently post-Augustan (and Cossutia may just have been married to a man who lived there).

36. See Smith, A. H., ‘Gavin Hamilton's Letters to Charles Towneley’, JRS xxi (1931) 306Google Scholar. Hamilton thought the two inscriptions recorded a father and son.

37. Cerdon or Cerdo, nickname of the art-dealer Damasippus (Horace, Sat. ii 3–25Google Scholar, and Pseudo-Aero ad loc.) and of a slave of Trimalchio (Petronius Sat. 60. 5) is as these passages show derived from κέρδος, gain, not the rare κερδώ, fox. Not uncommon epigraphically for slaves and freedmen.

38. BM Cat. nos. 1666, 1667; G. M. A. Richter, Three Critical Periods in Greek Sculpture (1951) fig. 85. Mr. J. B. Ward-Perkins very kindly confirms that the Catalogue is right to distinguish the marbles: no. 1666 appears to be fine-grained, with traces of ‘bedding’, almost certainly second quality Pentelic, no. 1667 is medium-grained, and he feels it could well come from the upper (Roman) quarries on Paros, where the stone is not of exceptional quality. Renfrew, C. and Peacey, J. Springer, ‘Aegean Marble: a Petrological Survey’, BSA lxiii (1968) 45Google Scholar seem too pessimistic about identification; for the capacities of scientific analysis, Craig, H. and Craig, V., ‘Greek Marbles: Determination of Provenance By Isotopic Analysis’, Science vol. 176 no. 4033 (April 1972) 401CrossRefGoogle ScholarPubMed.

39. Rubensohn, O., ‘Parische Kunstler’, JDAI 1 (1935) 56Google Scholar believes Cerdo made one in Paros, one in Rome, and dates them on the letter forms to the first half of the first century B.C., as still in PW XVIII 4 1868 s.v. Paros.

40. Coarelli, F., ‘Classe Dirigente Romana e Arti Figurativi’, Dial, di Archeol. iv–v (1970–1971) 241Google Scholar; he holds the turn to classicism came about 100B.C, not after Sulla.

41. By Mr. Ward-Perkins in a letter to me.

42. CIL XII 423, with D. M. formula, perhaps post-Flavian in Gaul, according to J. J. Hatt, La Tombe Gallo-romaine (1951) 18–19. For a sculptor who worked in Rome and Gaul, Pliny, NH xxxiv 45–6Google Scholar.

43. IG XIV 1250.

44. Kaibel, , ‘Zu den griechischen Kunstlerinschriften’, Hermes xxii (1887) 156Google Scholar; PW xv 838, Menelaus 18 (Lippold); Enc. di Arte Antica II 871. Hatzfeld, Trqfiquants, loc. cit., denies the identification.

45. See van Essen, C. C., ‘De Bildhower Pasiteles’, Meded. v. h. Niederlandsch Hist. Inst. te Rome, 2 vii (1937) 29Google Scholar, and lit. quoted there. Menelaus has been made responsible, on grounds of style, for various works, including even the Ara Pacis!

46. Helbig, W., Führer⁴ (ed. Speier, H.) III (1967) 2352Google Scholar and lit. quoted there: at least Neronian, possibly Flavian. The group, in theTerme Museum, is described as of ‘Griechischer, vielieicht Parischer Marmor’.

47. CIL VI 16528, from Rome, provides us with a M. Cossutius M. 1. Stephanus, but the inscription is probably too humble to record a distinguished sculptor, pleasant as it would be to annex him for our family.

48. IG XII 5 1049, with dating; PW Suppl. Ill 262, Cossutius 3a (Lippold); Enc. di Arte Antica, Cossutius. M. Squarciapino, La Scuola di Afrodisia (1943) 13 no. 8, cf. 19ff. Her Κοσσούθιος appears to be a slip, but I have not seen the stone. She supposes that the sculptor worked in Rome, or possibly Aphrodisias, and the work was sent thence to Paros – rather unlikely, though, as the statue itself is lost, not to be disproved by examination of the marble.

49. Op. cit. 55; he dates the base to the first half of the first cent.-B.C.

50. Treggiari, S., op. cit. Appendix 2, with ILLRP ii p. 500Google Scholar; Gordon, A. E., Epigraphica I: On the First Appearance of the Cognomen in Latin Inscriptions of Freedmen, U. of Calif. Pubs, in Class. Archaeol. i 4 (1935Google Scholar) shows that the omission of the cognomen is commoner in Latin inscriptions than in Greek ones; ours is of course Greek. A freedman's town of origin is perhaps never given (ILLRP 799 is the exception proving the rule).

51. Cicero pro Balbo 30 (cf. Hatzfeld, Trafiquants 74–5); Atticus refused it for this reason, Nepos, Atticus 3. 1Google Scholar (but Gesner brackets the explanation). However, would a Roman advertize his citizenship of an unimportant place? And Aphrodisias was too far inland to export marble to Rome.

52. First suggested to me by M. H. Crawford.

53. The sculptor Aphrodiseus registered in PW I 2727 turns out, alas, to be a Koblanos of Aphrodisias, PW XI 932 s.v. Koblanos (Lippold), cf. Squarciapino op. cit. 16. The peculiar Άϕορδίοιιυς seems to be a dialect form of Aphrodisios from darkest Pamphylia, Heubeck, A., Beiträge zur Namenforschung vii (1956) 8Google Scholar.

54. Montuoro, P. Zancani, ‘Repliche Romane di una Statua Fidiaca’, Bull. Com. lxi (1933) 43Google Scholar; cf. Pliny, NH xxxvi 38Google Scholar.

55. F. Bechtel, Die Historischen Personennamen des Griechischen bis zur Kaiserzeit (1917) 536–46 gives a long list (‘elle doit d'ailleurs être critiquée’, says Robert, L., Hellenica I (1940) 123Google Scholar, reiterating however that the phenomenon is a common one). We could, of course, in theory have an enfranchised peregrine called Aphrodeisieus rather than a freedman.

56. Varro, LL viii 21Google Scholar.

57. Erim, K. T., ‘The School of Aphrodisias’, Archaeology xx (1967) 18Google Scholar, and for works under Hellenistic influence there, ‘De Aphrodisiade’, AJA lxxi (1967) 238Google Scholar. I am grateful to Professor Erim for confirming his view with reference to Cossutius Aphrodeisieus. After isolated instances from the archaic period, the first work on the site is perhaps early Augustan, and Aphrodisian sculptors are only attested in Rome a good deal later. Note however Cornelius Aphrodeisieus (whose name provokes the same problems as that of Cossutius) at Corinth, probably in the early first century A.D., Squarciapino op. cit. 12.

58. IG XII 5 422.

59. So Robert, C., PW IV 1674Google Scholar, Cossutius 3. The spelling Μάρκος is often found in inscriptions contemporary with those showing Μάρκος – or even in the same ones.

60. It was borne by Aemilius Avianianus, that other patron of sculptors at least after his adoption by an Aemilius. His freedmen are C., which suggests that this was his original praenomen too.

61. Marshall, F. H., Collection of Ancient Greek Inscriptions in the British Museum iv 2 (1916) 217Google Scholar; cf. Enc. di Arte Antica s.v. Maarkos (Guerrini, L.); PW XIV 1648Google Scholar (Lippold).

62. G. M. A. Richter, op. cit. 45 n. 2; dated by Adriani, , Not. Scavi xvi (1938) 172Google Scholar, figs. 7–10, and by Meritt (ace. Richter) to the 2nd. century B.C. Meritt suggests Μαάρκου for Adriani's Μάλλιου, and this looks right to me. The men could presumably be his freedmen, using only his praenomen, as would be possible at this early period, to identify him; and perhaps described as Athenians not because they were citizens but as a recommendation of their artistic qualities. But it is perhaps simpler to think that they were sons of a Greek with a Roman given name. (Adriani notes ‘marmo bianco a cristalli lucidi piuttosto grossi’.)

63. Paribeni, R., Not. Scavi xi (1933) 431Google Scholar no. 33, with photograph. Picard, Ch., REL xiv (1936) 161Google Scholar and C. C. Essen, op. cit., accept that Zeuxis is probably an artist.

64. G. M. A. Richter, Catalogue of Greek Sculptures in the Metropolitan Museum (1954) no. 190, and op. cit. 45, 49, and fig. 98 (with epigraphic notes by B. Meritt). Richter describes the statue as of Pentelic marble, and Dr. D. von Bothmer kindly tells me that this is probably but not certainly right.

65. Possibility not noticed by Gross, W. H.PW Xa 351Google Scholar Zeuxis 2, or Moreno, P., Enc. di Arte Antica vii 1268Google Scholar Zeuxis.

66. S.v. Marmorarius.

67. CIL XI 1415.

68. In spite of Strabo C223, τά λιθουργεῖα in Pisa itself (also wood, now much used on building in Rome and ‘the villas’, he says).

69. Marmorarius can refer to sculptors of a decorative kind, see the Enciclopedia s.v. Contrasted with statuarii, Seneca Ep. lxxxviii 18, cf. Schroff, , PW XIV 2 1897Google Scholar s.v. It includes dealers in and sellers of marble, but plerumque de scalptoribus, acc. the Thesaurus s.v. See Robert, L., Hellenica xi–xii (1960) 28Google Scholar for its use by late Greek sources.

70. Pliny, NH xxxvi 14 and 48Google Scholar (Mamurra again); though ILLRP 325 suggests some local use from the mid second century.

71. CIL VI 597, dated to the (suffect) consulship of P. Calvisius Ruso and L. Caesennius Paetus, for whom see Degrassi, I Fasti Consolari dell'Impero Romano (1952).

72. CIL II 1066, VI 229, 461, XIV 4314 make this quite clear.

73. Epaphroditus also set up a tombstone to C. Cossutio Maec. Calidio Celeri, perhaps his patron, CIL VI 16518; he does not say that he made it himself, as inscriptions occasionally do. The tribus Maecia is not a very common one, covering Lanuvium, Hatria, Brundisium, Naples, Paestum, and a small town in north Italy; possibly we have here a pointer to the Campanian or south Italian origin of the Cossutii?

74. CIL VI 16534 a–b.

75. Guarducci, M., Epigrafia Greca I (1967) 455Google Scholar, with photograph of the side with the tools. The collection of tools is probably too complete to have the merely symbolic function common at any rate later on funerary inscriptions (G. Susini, The Roman Stonecutter (1973) 25).

76. Cn. is the commonest praenomen among the funerary inscriptions of Cossutii in Rome, all or mostly imperial in date. M. occurs, and one D. (CIL VI 16523).

77. Suetonius, DJ 1. 1Google Scholar.

78. Wiseman, op. cit. 227 lists him as a novus homo (no. 141).

79. Op cit. no. 359. with p. 603 n. 2.

80. Id. ib. no. 399, showing Amphitrite and Neptune, suitable types for a maritime trader (for the family see Levick, B. and Jameson, S., ‘C. Crepereius Gallus and his Gens’, JRS liv (1964) 98Google Scholar). Is it fanciful to note, in connection with L. Cossutius, that the effect of the Gorgon's head was to create stone figures (λιθίνων...άνδριαντων, Menander Dyscolos 159) and that Pegasus provided excellent transmarine transport (though not it seems a ship's name in antiquity)?

81. E.g. Cicero, ad. Q. f. III 1.4Google Scholar.

82. Enc. di Arte Antica s.v. Marmo.

83. Crawford, op. cit. no. 480, cf. S. Weinstock, Divus Julius (1971) 99, 200.

84. op. cit. 150, 227.

85. CIL X 8028; H. G. Pflaum, Les Carrières Procuratoriennes Equestres (1960–1) 1045.

86. CIL IX 4924 (the ancient Trebula Mutuesca).

87. CIL IX 4859, 4064.

88. In Verr. ii 3. 55, 185.

89. Ad fam. xvi 27. 2, with Tyrrell and Purser ad loc.

90. Spain, S. Gaul, Noricum etc; one or two in Etruria, including a prominent Q. Cossutius P. f. at Tarquinia, , CIL XI 3374Google Scholar and Not. Scavi ii (1948) 258Google Scholar.

91. PW IV 1673 (Groag); PIR ² 1543. D. Magie, op. cit. 1419 n. 68 for his position in Cilicia.

92. Cossi[ti]anus [Fi]rmus, a mid-third century procurator (CIL III 13240, Pflaum op. cit. 1067) should perhaps be Cossi[ni]anus? Cossinii and Cosinii are commoner in the imperial period than Cossutii, especially at official level, and there seems to be no example of the spelling Cossitius.

I am grateful to Mr. M. H. Crawford and Miss J. M. Reynolds for reading this paper, to the other scholars who so kindly answered my queries, and particularly to Mr. J. B. Ward-Perkins.