We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 27 September 2013
Evidence for the Ptolemaic occupation of the Cycladic island of Keos in the 3rd century B.C. is both contentious and exiguous. A recent archaeological surface survey of both the territory and the polis-centre of Koressos (Ptolemaic Arsinoe, apparently the principal Egyptian foothold on the island) invites a review of relevant historical and epigraphical evidence, old and new. Contrary to some opinion, all the available evidence implies that the later Hellenistic period was not a time of material prosperity for Koressos. It is suggested that its incorporation into the Ptolemaic empire may have contributed to its demise as a city-state, and to its eventual transformation into a minor adjunct of the neighbouring polis of Ioulis.
Acknowledgements. An earlier version of this paper was presented at the First Joint Archaeological Congress (Baltimore, Maryland, January 6 1989), in a symposium organised by Davis, Jack L. and Reger, Gary, entitled ‘Ptolemaic Conquest and Influence outside Egypt: The Hellenic Response’ (see AJA 93 [1983] 247–8).Google Scholar We are grateful for helpful comments received from the discussants at the symposium, Roger S. Bagnall and Stella G. Miller, and subsequently from Sue Alcock, Lin Foxhall, Gary Reger, Anthony Snodgrass and Dorothy Thompson. The fieldwork on Keos described here was conducted in 1983–84, in collaboration with Eleni Mantzourani of the University of Athens, with permission of the Greek Archaeological Service and with the support of funding from: the National Geographic Society (Grant 2291–81); the British Academy; the Institute for Aegean Prehistory; the Faculty of Classics and the Museum of Classical Archaeology, University of Cambridge; the University of Illinois at Chicago; the Dr M. Aylwin Cotton Foundation; and various private donors. For a full account, see Cherry, J.F., Davis, J.L. and Mantzourani, E., Landscape Archaeology as Long-Term History: Northern Keos in the Cycladic Islands [Monumenta Archaeologica 16] (Los Angeles 1991)Google Scholar; and, for a brief summary, Davis, J.L., Cherry, J.F., and Mantzourani, E., ‘An Archaeological Survey of the Greek Island of Keos’, National Geographic Society Research Reports 21 (1984) 109–16.Google Scholar
Abbreviations in addition to those in standard use
Bagnall = Bagnall, R.S., The Administration of the Ptolemaic Possessions Outside Egypt [Columbia Studies in the Classical Tradition 4] (Leiden 1976).Google Scholar
Graindor = Graindor, P., ‘Kykladika,’ Musée Belge 25 (1921) 78–125.Google Scholar
Keos = Cherry, J.F., Davis, J.L. and Mantzourani, E., landscape Archaeology as Long-Term History: Northern Keos in the Cycladic Islands [Monumenta Archaeologica 16] (Los Angeles 1991)Google Scholar
Mendoni = Mendoni, L.G., Κέα Ι: Ὁι ἐπιγϱαφές καί οἱ ἄλλες γϱαπτές μαϱτυϱίες γιά τό νησί, πϱοσωπογϱαφία – τοπογϱαφία – παϱαδόσεις – ἱστοϱία (Athens 1988).Google Scholar
Robert = Robert, L., ‘Sur un décret des Korésiens au Musée de Smyrne’, Hellenica 11–12 (1960) 132–76.Google Scholar
1 Bagnall 141, Several other Aegean towns, too, were re-founded and re-named at this time as Arsinoe (after the Egyption queen Arsinoe II, wife and full sister of Ptolemy II Philadelphus), notably Methana in the Argolid and Rethymna on Crete: see Foxhall, L., Mee, C., Forbes, H. and Gill, D., ‘The Ptolemaic base of Methana’, [abstract] AJA 93 (1989) 247–8Google Scholar; Wartner, S., ‘The Ptolemies and Crete’, [abstract] AJA 93 (1989) 248.Google Scholar For a map of Ptolemaic possessions overseas, see Bowman, A.K., Egypt after the Pharaohs: 332 BC–AD 642, from Alexander to the Arab Conquest (Oxford 1986), fig. 2.Google Scholar
2 Bagnall 143: Merker, I.L., ‘The harbor of Ioulis’, AJA 72 (1968) 383–4.CrossRefGoogle Scholar
3 Dittenberger first argued that the Αρσινοει̑ς mentioned in SIG 3 562 were to be located in the Aegean islands, and his view has been widely accepted. Recently, however, Mendoni (820. n.55) has suggested that they were Cretans, arguing that the ᾿Αρσινοει̑ς. if citizens of Koressos, should have been listed next to the Ioulietans and Karthaians. Her objection carries little force given the fact that in the same document Amorgians are also listed in three separate places, rather than side by side as might be expected for poleis on the same island: Σάμιοι οἱ ἐμ Μινοίαι (l. 80); Αἰγιαλει̑ς (l. 83); and Αρϰεσιεει̑ς (l. 81. It is also worth noting that of the 16 identifiable names preserved in this list, all are in the Aegean islands and none is Cretan. Robert, L., ‘Les Asklepieis de l'archipel’, REG 46 (1933) 423–42 [at p. 423–5],CrossRefGoogle Scholar has identified the ᾿Ασϰληπιει̑ς. the single name left unidentified by Dittenberger, with the Θερμαι̑οι of Ikaria. ᾿Αρσινοει̑ς are recorded among cities of Crete in a Magnesian inscription of ca. 200 B.C.: Kern, O., Inschriften von Magnesia am Maeander (Berlin 1900), no. 21Google Scholar: cf. Le Rider, G., ‘Les Arsinoéens de Crète’, in Kraay, C.M. and Jenkins, G.K., eds., Essays in Greek Coinage Presented to Stanley Robinson (Oxford 1968) 229–40.Google Scholar It seems more likely that names were recorded in the order in which states accepted requests of the Magnesian theoroi.
4 While it is certainly true that the finest harbour on the island is found there, it is not the only one that might have been useful to the Ptolemies. Georgiou, H. and Faraklas, N. (‘Ancient habitation patterns of Keos: Locations and nature of sites on the Northwest part of the island’, Ariadne 3 [1985] 207–66. at p. 220)Google Scholar have recently described the bay of Poieessa (at modern Poisses) as being of some local importance, but it is not suitable for large boats and is not protected from the prevailing northwest wind. It is worth noting, however, that there are well sheltered moorings nearby at Koundouros (certainly part of the territory of ancient Poieessa) and that the small bay of Kambi served until recently as a shipment centre for export of the velanidi acorn, an important component of the island's economy.
5 Robert 132–76. On Patroklos' expedition in the Aegean and arguments for the date of the Chremonidean War, see Walbank, F.W., ‘Macedonia and Greece’, in CAH 2 VII.1. (Cambridge 1984) 236–40Google Scholar: Will, E., Histoire politique du monde héllénistique (323–30 av. J.C.) Vol. I (2nd ed., Nancy 1979) 219–31.Google Scholar For Callimachos, see Epigram v (Pfeiffer, R., Callimachus, Volumen II: Hymni et Epigrammata [Oxford 1949] 81–82Google Scholar:
ἔστ᾿ ἔπεσον παρἀ θῖνας ῾Ιουλίδας ὁφρα γένωμαι σοἰ τὀ περίσχεπτον παίγνιον. ᾿Αρσινὁη
Poieessa is not, in fact, much farther from Ioulis than is Karthaia, and might in some sense be thought to be ‘on the shores of Ioulis’. In recent centuries, land there has been regularly cultivated by farmers who have been permanently resident at Ioulis (the site of the modern Chora of the island). The association implied by this verse may reflect not geographical distance, so much as a political nexus promoted between the two poleis at a time when the island was under Egyptian control.
6 Lasserre, F., Strabon. Géographie, Tome VII [Livre X] (Paris 1971) 14–16.Google Scholar Robert (160. n.1) has discussed this and similar cases in the work of Ptolemy.
7 Dunant, C. and Thomopoulos, J., ‘Inscriptions de Céos’, BCH 78 (1954) 316–48 [at p. 334–5].Google Scholar
8 No facsimile of the inscription has been published. The stone is ‘cassé de tous côtes, sauf à droite (anathyrose).’ Lines 4–6 are relevant to its assignment to a particular polis. From the published photograph (see Dunant and Thomopoulos [n.7 above] 335, fig. 13), it appears that there is room for at least an additional letter at the beginning of line 5. K, the first preserved letter, falls directly beneath the I in [ε]ἰ̑οιν of line 4; in line 4, 15 letters falls to the left of this position in the inscription as it has been restored by the editors, but in line 5 only 14. There would thus seem to be no reason why the ethnic of another Keian polis could not be restored here. Each would be only 9 letters long (as opposed to the 8 of ΚΟΡΗΣΙΩΝ e.g., ΠΟΙΗΣΣΙΩΝ, ΚΑΡΘΑΙΕΩΝ, or, for that matter. ΑΡΣΙΝΟΕΩΝ. Following Mendoni, L.G. (‘More inscriptions from Keos’, BSA 84 [1989] 290 and n.1).Google Scholar the characteristic phrase in the heading of this inscription seems to exclude a Karthaian origin, but permits Ioulis or Poiessa as alternatives to Koressos. Mendoni (Ibid. 293) also publishes a Ioulietan inscription found in two parts, one in Ioulis and one at Poiessa, and comments that this ‘should not be considered strange, [since] the travelling of stone is a well-known phenomenon’; her remark applies in equal measure to the findspot of SEG xiv 541.
9 See n.12 below for a discussion of the controversy over the dating of this inscription.
10 Graindor 125.
11 Bagnall 141–5.
12 Bagnall (65. n.119: 142. n.91) adduced supporting evidence for a date in the 190s B.C. in the form of a prosopographical link between a Cypriot dedication and the Delphic list. Arguments for a date ca. 245 B.C. have been set out by Daux, G.: ‘Trois remarques de chronologie Delphique (IIIe et IIe siècles avant J.-C.).’ BCH 104 (1980) 115–25CrossRefGoogle Scholar: and ‘La grande liste Delphique des théorodoques’, AJP 101 (1980) 318–23. The question of chronology is of obvious importance for determining the duration of Ptolemaic control over Koressos, since it is unlikely that the Koressians would have assumed their former identity so long as an Egyptian garrison remained. The prosopographical link no longer seems to clinch the dating of Column I to the 190s B.C., because the name in question. 'Αριοτος Τιμ[οδ]ἡμου Χι̑ος, appears to have been inscribed in rasura and thus offers no more than a terminus ante quem for the original document (Daux, G.BCH 104 (1980) 120–3Google Scholar).
13 Daux's restoration of the phrase [ἐν] ᾿´Α[ρσιν]ό[ηι] in Column I. line 37 is to be rejected: Robert (161–73) has plausibly suggested [ἐ]ν [Καρ]θ[αίαι] in its place, not the name of Koressos (Bagnall 142. n.90) which is clearly present in line 38.
14 See n.3 above.
15 Robert 168.
16 Cook, B.F., ‘A Dated Hadra Vase in the Brooklyn Museum’, The Brooklyn Museum Annual 10 (1968–1969) 115–38 [p. 137. no. 18]Google Scholar; SEG xxiv 1176.
17 See Braunert, H., ‘Auswärtige Gäste am Ptolemäerhofe’, JDI 65–66 (1950–1951) 231–63 [p. 235. no. 15]Google Scholar: Rönne, T. and Frazer, P.M., ‘A Hadra vase in the Ashmolean Museum’, JEA 39 (1953) 84–94.Google Scholar
18 Huss, W., Untersuchungen zur Aussenpolitik Ptolemaios' IV [Münchener Beiträge zur Papyrusforschung und Antiken Rechtsgeschihte 69] (München 1976) 227Google Scholar; Cook (n.16 above) 128–30; see also Bingen, J., ‘Vases d'Hadra et prosopographie ptolémaïque’, Chronique d'Egypte 43 (1968) 389–92 [389, no. 1CrossRefGoogle Scholar; 391, no. 4].
19 On the mints of Rithymna and Methana, see Le Rider (n.3 above), especially p. 231; for the 4th century B.C. bronze coinage of Koressos, see G. Reger and M. Risser. ‘Coinage and federation in Hellenistic Keos’, in Keos (Ch. 15): for the evidence for a later 3rd century B.C. Keian federation, see also J.F. Cherry. J.L. Davis and E. Mantzourani, ‘Introduction to the historical and epigraphical evidence’, in Keos (Ch.10).
20 Bagnall 145: Buraselis, K., Das hellenistische Makedonien und die Ägäis (Munich 1982) 87, n.201Google Scholar; Graindor, P., ‘Inscriptions des Cyclades’, Musée Belge 11 (1907) 97–113 [at p. 104–6]Google Scholar: Graindor 121–2; Dow, S. and Edson, C.F. Jr., ‘Chryses: a study of the evidence in regard to the mother of Philip V’, HSCP 48 (1937) 127–80 [at p. 134 n.1].Google Scholar The discussion of this inscription by Hammond, N.G.L. and Walbank, F.W. (A History of Macedonia, Vol. III, 336–167 B.C. [Oxford 1988] 294)Google Scholar classes it with others that explicitly mention a king Antigonos, but cautions that it does not constitute independent evidence for domination of the Aegean by Antigonos II.
21 Graindor (n.20. above) emphasized that the transcription is ‘très exacte’, but it is obviously not absolutely precise at the broken edges of the inscriptions: e.g. τ for π in B, line 1: λ for ν in B, line 3: and τ for Ζ in A, line 7, immediately above the two letters that remain from the name of the King on A.
22 Geagan, D.J., ‘Inscriptions from Nemea’, Hesperia 37 (1968) 381–4CrossRefGoogle Scholar: cf. Robert, L., Bulletin épigraphique (1969) no. 236Google Scholar.
23 Caskey, J.L., ‘Koroni and Keos’, in Studies in Attic Epigraphy. History, and Topography Presented to Eugene Vanderpool [Hesperia Supplement 19] (Princeton 1982) 14–16 [pl.2, c].Google Scholar
24 Briscoe, J., A Commentary on Livy Books XXXI–XXXIII (Oxford 1973) 99.Google Scholar
25 Dunant and Thomopoulos (n.7 above) 344.
26 Buraselis (n.20 above) 183 and n.4: cf. Huss (n.18 above) 226, where IG XII 5 1069 could ‘… auf makedonischen Einlluβ hinweisen.’
27 Robert 155–6.
28 Ibid.
29 For general discussion of the coinage of the Ptolemaic possessions overseas, see Bagnall 176–212. On the Delian coinage, see Bruneau, P. et al. L'ilot de la maison des comédiens (Délos XXVII) (Paris 1970) 404–5Google Scholar [coins F 502–9], 412 [table].
30 Bagnall 204, n. 112.
31 Bagnall 205. For the coinage from Tenos, see Etienne, R. and Braun, J.-P., Ténos I: Le sanctuaire de Poseidon et d'Amphitrite [BEFAR 263] (Paris 1986)Google Scholar; the authors also suggest (p. 210–11), on the basis of ceramic evidence from the mid- to late-3rd century B.C. that as a result of the Chremonidean War and Antigonid control over Athens ‘… Ténos vit en circuit fermé pendant un bon tiers de siècle’. In addition to Ptolemaic coinage (no. 3 in our list), Mendoni (819–20, n.52) has reported the discovery of six lead sling bullets (μολυβδίδες) in recent excavations at the polis centre of Koressos; but these do not necessarily constitute evidence for a Ptolemaic presence, and are widely distributed throughout the Greek mainland and in the islands: Varoucha-Christodoulopoulou, E., ‘Συμβολὴ εἰς τὸν Χϱεμωνίδειον πόλεμον 266/5–263/2 π.Χ.’ AE (1953–1954) 321–49 [at p. 332–4].Google Scholar
32 For Manthos' map, see Valetas, G. and Varthalites, D., Ἀφιέϱωμα στὴν πνευματικὴ Κέα (Athens 1986) 34.Google Scholar Archaic temple: Welter, G., ‘Von griechischen Inseln: Keos I’, AM (1954) 48–93 [at p. 64–69]Google Scholar; Lauter, H., ‘Bermerkungen zum archaischen Tempel von Koressia’, AA (1979) 6–16Google Scholar; Schuller, M., ‘Dorische Architektur der Kykladen’, JDI 100 (1985) 361–83.Google Scholar Fortification walls: Welter, ibid.; Maier, F.G., ‘Stadtmauern auf Keos’, AM 73 (1958) 6–16.Google Scholar Excavations in the town: Zapheiropoulou, Ph., ‘Κέα (Κοϱησσία)’, ADelt 27 (1972) 606–11.Google Scholar
33 Isolated towers: Welter (in n.32 above) 87–90; J.F. Cherry, J.L. Davis and E. Mantzourani, ‘The towers of Keos’, in Keos, Ch. 13. Keian miltos: Bent, J.T., Aegean Islands: The Cyclades, or Life Among the Insular Greeks (London 1885Google Scholar; reprint Chicago 1966) 464; J.F. Cherry, J.L. Davis and E. Mantzourani, ‘Miltos and metallurgical extraction’, in Keos, Ch. 14. Shrine of Dionysos at Ayia Irini: Caskey (in n.23 above); Caskey, M.E., ‘Ayia Irini, Kea: the terracotta statues and the cult in the temple’, in Hägg, R. and Marinatos, N. (eds.), Sanctuaries and Cults in the Aegean Bronze Age [Skrifter Utgivna av Svenska Institutet i Athen 28] (Stockholm 1981) 127–35.Google Scholar
34 Keos, passim.
35 T.M. Whitelaw and J.L. Davis, ‘The polis center of Koressos’, in Keos (Ch.12, with Appendix 12.1, ‘The finds from Koressos’, by R.F. Sutton Jr.). This research is in keeping with the recent extension of intensive survey techniques from purely rural contexts to urban centres: see, e.g., Bintliff, J. and Snodgrass, A., ‘Mediterranean survey and the city’, Antiquity 62 (1988) 57–71CrossRefGoogle Scholar; S.E. Alcock, ‘The polis of Phlius and urban survey’, Hesperia (forthcoming).
36 Welter (in n.32) 54.
37 Graindor 125.
38 Ruschenbusch, E. (‘IG XII 5 609: Eine Bürgerliste von Iulis und Koresia auf Keos’, ZPE 48 [1982] 175–88)Google Scholar has recently calculated, on the basis of epigraphical evidence, a total population of 3,840–4,090 for the whole island of Keos in the 4th century B.C. His calculations rely heavily on a list (IG XII 5 609), found near Ioulis, in which male names are recorded under several headings; many are described as men of Koressos, but most are clearly Ioulietans. Originally 482 or more individuals were included (at least 154 of Koressos and at least 320–40 of Ioulis) and these are argued to represent the total number of citizens living in the two poleis. It is far from certain, however, that the list records a complete census of citizens of the two city-states: its heading is not preserved and thus offers no clue why the names were inscribed. A military interpretation of the document, for example, might suggest male populations as high as 260 for Koressos and 570 for Ioulis. Ruschenbusch's interpretation of an early 3rd century B.C. decree of Koressos, (IG XII 5647)Google Scholar, dealing with the distribution to the citizens of Koressos of sacrificial meat at a festival is also not without difficulties. It is not possible to be certain that any more than 180–200 individuals were given meat, as he has calculated, or to determine what percentage of those so provided lived within the polis. However, these and various other arguments (for which see Cherry et al. in Keos [Ch.10]) do suggest populations in the range of 4,000–6,700 for the island as a whole, and on the order of 1,000–1,300 for Koressos. It is in any case implausible that sufficient food could have been produced on the land that lay within the territory of Koressos to feed very many more than about 1,000 individuals.
39 For a general discussion of this issue, with critique, see Alcock, S.E., ‘Archaeology and imperialism: Roman expansion and the Greek city’, Journal of Mediterranean Archaeology 2[1] (1989) 87–135 [at p. 105–9]CrossRefGoogle Scholar; idem, ‘Roman imperialism in the Greek landscape’, Journal of Roman Archaeology 2 (1989) 5–34. Piracy has often been adduced to explain similar patterns of widespread retreat from rural residence. Yet some farmsteads, at least, appear to have remained in use in areas closer to the coast and more exposed than areas in which no such remains have been found. There has probably never been a time before the 20th century when brigands did not present some threat to islanders. Even in the 3rd century B.C., when the seas were controlled by the Ptolemies, piracy was still a problem, and yet this was a time when there was considerable rural settlement: see. e.g., Graindor 122; Bagnall 128; Petrakos, B.Ch., ‘Νέαι πηγαὶ πεϱὶ τοῦ Χϱεμωνιδείου πολέμου’, ADelt 22 (1967) 38–52 [at p.51].Google Scholar
40 Welles, C.B., Royal Correspondence in the Hellenistic Period (New Haven 1934) nos. 10–13.Google Scholar
41 Davies, J.K., ‘Cultural, social, and economic features of the Hellenistic world’, CAH 2 VII.1257–320 [at p. 269].Google Scholar
42 L. Robert (n.3 above) 440–2; also Rougemont, G., ‘Amorgos colonie de Samos?’, in Rougemont, G. (ed.), Les Cyclades: Matériaux pour une étude de géographic historique (Paris 1983) 131–4Google Scholar, on the matter of Samians on Amorgos.
43 Bagnall 229.
44 R. Osborne, ‘Landuse and settlement in Classical and Hellenistic Keos: the epigraphic evidence’, in Keos (Ch.16).
45 Bagnall 227.
46 von Gaertringen, F. Hiller and Wilski, P., Stadtgeschichte von Thera [Thera 3] (Berlin 1904) 106Google Scholar; Robert, L., Noms indigènes dans l'Asie-Mineure Gréco-Romaine I [Bibliothéque archéologique et historique de l'Institut Français d'Archéologie d'Istanbul xii] (Paris 1963) 411–18.Google Scholar
47 Bagnall 220.
48 The use of Cycladic military levies may have been a serious strain on human resources. For example, at the end of the 4th century B.C., Keians are catalogued among other islanders in an inscription from Nemea, (SEG xxv 357Google Scholar; Geagan, D.J., ‘Inscriptions from Nemea’, Hesperia 37 (1968) 381–5CrossRefGoogle Scholar; cf. p. 16 above). The precise number of πεζοί sent from Keos is not restorable, but the preserved endings (masculine plurals of the 2nd declension) of the cardinal numerals in line 5 and line 6 associated with entries between Keos and Kythnos, and between Kythnos and Mykonos, suggest that Cycladic islands were contributing levies of considerable size (i.e., 200 or more men); furthermore in line 3, the figure 700 is preserved. Even a force of 200 men could have represented as much as one-fifth of the total manpower of Keos. However, there is little direct evidence for the recruiting of mercenaries or colonists from Keos. Heichelheim, F., Die auswärtige Bevölkerung im Ptolemäerreich [Klio Beiheft 18, Neue Folge Heft 5] (1963) 84Google Scholar lists only two Keians in Egypt before the 2nd century B.C., the brother physicians Erasistratos and Kleophantos (and the presence of the former in Alexandria is not entirely certain; see Lloyd, G.E.R., ‘A Note on Erasistratos of Ceos’, JHS 95 [1975] 172–5).CrossRefGoogle Scholar Although, in total, a rather high percentage of Greeks in Egypt hailed from the islands (141 or 715 listed by Heichelheim), rather few came specifically from the Cyclades (15 of the 141). Probably there should be added a single Keian mercenary captain (see p. 15 above); and, as Dorothy Thompson has suggested to us, more Keian names may now have been collected from Egyptian papyri by the Prosopographica Ptolcmaica in Louvain. Even at nearby Athens there are few islanders attested among mercenaries of the 3rd century B.C., and only one Keian, (IG II 2 1958, line 15).Google Scholar
49 The full publication of the recent British survey of Methana should allow these questions to be addressed more fully; for a preliminary statement, see L. Foxhall et al. (in n.1 above). Also relevant, although not based on new fieldwork, are Reger, G., ‘The economic impact of the Ptolemies in the Cyclades’, [abstract] AJA 93 (1989) 248Google Scholar; S. Wartner (in n.1 above).
50 For a wider consideration of this issue with reference to Greece as a whole, see Alcock, S.E., Greek Society and the Transition to Roman Rule (Ph.D. dissertation, University of Cambridge, 1989)Google Scholar; idem (in n.39 above).
