AChAEMENID ANATOLIA AND ThE SLAVE SUPPLY 335
46 The kykloi were otherwise normally used for the sale of utensils: Poll. 10.18.
47 Poll. 7.11: καὶ κύκλοι δὲ ἐν τῇ νέᾳ κωμῳδίᾳ καλοῦνται ἐν οἷς πιπράσκεται τὰ ἀνδράποδα
‘and in New Comedy the place in which slaves are sold are called kykloi’’; Harp. s.v. kyk-
loi: ἐκαλοῦντο οἱ τόποι ἐν οἷς ἐπωλοῦντο τινες. ὠνομάσθησαν δὲ ἀπὸ τοῦ κύκλῳ περιεστάναι
τοὺς πωλουμένους ‘what they called the places in which certain persons were put up for sale,
so called because those being sold stood in the circle.’
48 οἴμοι κακοδαίμων τῆς τόθ‘ ἡμέρας, ὅτε εἶπέν μ’ ὁ κῆρυξ οὑτοσ<ὶ τί> ἀλφάνει; Fo r
the assumption that being an auctioneer was a shabby job, Theophr. Char. 6.5; cf.
Dem. 44.4.
49 οὐ θᾶττον αὐτὴν δεῦρό μοι τῶν τοξοτῶν ἄγων ἀποκηρύξει τις ὅ τι ἀλφάνοι; ‘make haste
and bring her here to me from the archers, and have her auctioned with “what’s the
bid?” ’
50 ἐγὼ μὲν ἤδη μοι δοκῶ, νὴ τοὺς θεούς, ἐν τοῖς κύκλοις ἐμαυτὸν ἐκδεδυκότα ὁρᾶν κύκλῳ
τρέχοντα καὶ πωλούμενον. One might interpret τρέχοντα as meaning ‘forced to exercise,’ cf.
Milton 2004 : 69 in relation to the Salé market: ‘now, they were stripped and put through
their paces. They were forced to jump and skip to test their agility.’ Cf. Xen. Hell. 3.4.19;
Ages. 1.28.
51 Descat 2012 : 204.
52 See Pritchett 1953 ; Pritchett and Pippin 1956.
53 Law on warranty: Hyp. 3.15 (cf. Pl. Leg. 11.916a; SEG 47:1026; IC IV 72 VII lines 10–15).
Records of slave sales: Hsch. s.v. en leukomasi. Slave-sale contract produced in court as evi-
dence: Lycurg. Leoc. 24; tax: Xen. Vect. 4.24.
54 Xen. Mem. 2.3.3 implies that some people do not have slaves to help them in their work.
For fifty slaves as an upper normative limit of holdings for the rich, see Pl. Resp. 9.578d-e; cf.
Men. Kolax 38. This dovetails with the range of figures preserved in the orators: Dem. 27.9
(two groups of slaves: 32–3 knife makers and 20 couch makers); Dem. 37.4 (30 mine slaves);
Lys. 12.19 (120 slaves co-owned by Lysias and Polemarchus).
55 Scheidel 2005 : 11.
56 Ducat 1990 : 19–29; Luraghi 2002 : 228–33. This is not the place to deal at length with this
question; for one thing, the issue of status is not essential for the analysis of this section, but
see my comments in Lewis 2013 : 393–5. I shall set out my position in more detail in Lewis,
forthcoming, chapter 6.
57 By this I do not mean to imply a teleological view of the evolution of slavery in Greece,
viewing helotage as an earlier ‘stage.’
58 To be precise: computation of the percentage of slaves exported to North America in rela-
tion to the Americas as a whole, using David Eltis’s online Trans-Atlantic Slave Trade Database,
yields a figure of 3.83064%.
59 Fogel and Engerman 1974 : 23 write ‘native-born blacks made up the majority of the slave
population in the U.S. colonies as early as 1680. By the end of the American Revolution,
the African-born component of the black population had shrunk to 20%. It hovered at this
share from 1780 to 1810 and then rapidly headed toward zero.’
60 Tadman 2000. For the demands of the labor sector to influence the gender ratios of a slave
population, cf. Origo 1955 who shows that Medieval Italian demand for domestic slaves
created a female-heavy population.
61 Harper 2011 : 76–7. For the work of helots and the structure of helotage ‘on the ground,’ see
Hodkinson 2008.
62 Hodkinson 2008 : 309–18.
63 Ephorus FGrHist 70 F 117 with Luraghi 2002 : 228–9.
64 On Cretan slavery in comparative perspective, see Luraghi 2009. Close analysis of Gortyn
in Lewis 2013 ; overview of Gortynian slave status in Lewis forthcoming, chapter 7.
65 Luraghi 2002 : 230–1; Hodkinson 2008 : 306.
66 Though Rihll 2010 : 216 notes that Lauffer’s estimate of 35,000 ‘is at the higher end of the
spectrum of possibilities.’