Slaves

In the description of Roman trade given so far, the main thing which differentiates it from the trade of, say, sixteenth-century Europe is that the Roman

64  Tchernia (1986b: 27-8).

65  Petronius, Sat. 76.

66  Strabo iv.197 and v.218 (with Harris 1989b: 124).

67  [[Camodeca (1999: no. 51) (superseding previous editions).]]

68  For Juvenal’s evidence about Rome itself see Courtney (1980: 112). On Roman vegetables: Andre (1961: 13-51).

69  See Petronius, Sat. 40, Martial viii.33.11, etc. They are the only common foodstuff at Pompeii which was definitely imported: Meyer (1988: 200, 215).

70  Tchernia (1986b: 293). At Pompeii too the garum amphorae far outnumber the olive-oil ones: Manacorda (1977: 121, 131).

71  See Ponsich and Tarradell (1965), Curtis (1978), Ponsich (1988).

Empire included three cities which were each twice the size (or nearly so) of the largest cities which existed in Christian Europe around 1600. But there were other important differences too, one of which was the relative lack of a commercial bourgeoisie in the Roman Empire, a fact which was closely linked to the very large role in Roman commerce of slaves and freedmen. Another difference is that the Romans carried on an extensive slave trade in order to maintain this element in their economic and social structure.

The common notion that the slave population of the Empire now consisted mostly of the children of slaves cannot be sustained.72 In any case such slaves might be bought and sold, like those who became slaves by other means such as through capture in war, through the enslavement of foundlings, or through importation across the frontiers.

Augustus had probably expected that his 2 per cent tax on slave sales would produce annual revenue on the order of five million sesterces, which implies that as many as a quarter of a million sales were believed to take place each year.73 Subsequently the number of prisoners-of-war who came on to the market in an average year declined (though it could sometimes be very high: the total numbers enslaved in Trajan’s Dacian wars and at the end of Bar-Kochva’s revolt were in six figures). In all probability, however, neither demand nor supply declined significantly at least until late in the Antonine period.

While Rome and Italy undoubtedly constituted the largest regional market, slaves were in demand all over the Empire. The main areas of ‘surplus’ within the Empire were Thrace, Asia Minor, with the exception of the provincia Asia, and Syria. Slaves were imported from time to time across practically all the frontiers, and regularly through Palmyra into Syria, as we know from the local fiscal law of 137.74 The route through Zarai into Numidia, which we happen to know of from the Zarai Tariff of 202,75 was merely one of the other channels of importation. Slaves were bought and sold in every Roman city, as fairly extensive testimony shows. However the most active centre of the trade, apart from Rome itself, was Ephesus,76 which drew most of its slaves from the interior of Asia Minor and probably still exported them in large numbers to Italy.

Imperial intervention was slight, as far as we can tell, even though the imperial household must itself have been a large source of demand. The sales tax had risen to 4 per cent by ad 43,77 and probably stayed at that level. By the time of the Zarai Tariff, however, the rate of duty charged on the

72  For this and the other views expressed in this section see Harris (1980a; 1999) [[Chapters 3 and 4 respectively]].

73  The tax: Dio lv.31.

74  OGIS 629 = CIS ii.3.1.3913. On the import of slaves from Colchis see Braund and Tsetskhladze (1989).

75  CIL VIII.4508.

76  Harris (1980a: 127); see also Achilles Tatius v.17.

77  CIL VI.915 = ILS 203; cf. Tacitus, Ann. xiii.31.

Importation of slaves was lower, at least in Numidia, than for any other major commodity, which can only have come about because the government wished to improve supply.