¹ McGinn, T., Prostitution, Sexuality and the Law in Ancient Rome (1998)Google Scholar is so recent it appeared after this article was initially completed, and provides a much more systematic (though circumscribed) study of the legal rules affecting Roman prostitution than hitherto. Stumpp, B., Prostitution in der römische Antike (1998)CrossRefGoogle Scholar appeared a little earlier, but follows Herter, H.'s articles, ‘Dime’, in RLAC III (1957), 1149–1213Google Scholar, and ‘Die Soziologie der antiken Prostitution im Lichte des heidnischen und christlichen Schrifttums’, JbAC 3 (1960), 70–111Google Scholar, in providing a useful collection of material organized in a set of rather confused analytical categories; Leontsini, S., Die Prostitution im frühen Byzanz (1989)Google Scholar, is a more coherent study but of a later period. A still earlier methodological tradition is represented by e.g. Balsdon, J., Roman Women: Their History and Habits (1962), 224–9Google Scholar, following Schneider, K., ‘Meretrix’, in RE XV.1 (1931), 1018–27Google Scholar, and the rather more liberal Kiefer, O., Sexual Life in Ancient Rome (1934), 55–63Google Scholar, who includes prostitutes in the section on ‘free love’ rather than ‘less reputable women’. Both, however, consider prostitutes who had liaisons with famous poets to be the most blessed of women and worry about the public health consequences of prostitution in ways that never occurred to the Romans, but rather reflect the concerns of nineteenth-century reformers such as A. J. B. Parent-Duchâtelet, whose work De la prostitution dans la ville Paris considerée sous la rapport de l'hygiéne publique, de la morale et de l'administration (1836) provided a model for similar works produced elsewhere in Europe and America, and has exerted considerable influence on historical studies of prostitution ever since.

² Among the more notable of these are: Walkowitz, J., Prostitution and Victorian Society: Women, Class and the State (1980)CrossRefGoogle Scholar; White, L., The Comforts of Home: Prostitution in Colonial Nairobi (1990)CrossRefGoogle Scholar; Hill, M. Wood, Their Sisters' Keepers: Prostitution in New York City, 1830–1870 (1993)Google Scholar; Mazo Karras, R., Common Women: Prostitution and Sexuality in Medieval England (1996)Google Scholar; and Hershatter, G., Dangerous Pleasures: Prostitution and Modernity in Twentieth-Century Shanghai (1997)Google Scholar.

³ Karras, op. cit. (n. 2), 10; Karras, R. M. and Boyd, D. L., ‘Ut cum muliere: a male transvestite prostitute in fourteenth-century London’, in Fradenburg, L. and Freccero, C. (eds), Premodern Sexualittes (1996) 104Google Scholar.

⁴ White, op. cit. (n. 2), 10–21.

⁵ White, op. cit. (n. 2), 6–10; Hill, op. cit. (n. 2), also uses prostitutes' letters to help produce a study of the sex-trade in mid-nineteenth-century New York in which prostitution also appears as a rational economic choice for women, with its own ‘positive appeal and rewards’ (5).

⁶ Karras, op. cit. (n. 2), 9; see also the similar remarks of Hershatter, op. cit. (n. 2), 3–33.

⁷ Dio, , Or. 7.133Google Scholar, and see also Cod. 8.50.7. Stories of enslavement and prostitution feature in Sen., , Contr. 1.2Google Scholar, the Hist. Apoll. Tyr. 33–7, Xen. Eph. 5.5–9, and Apul., , Met. 7.9–10Google Scholar. Women's fierce resistance to the logical consequence of such a fate—that she should lose her virginity to the wrong man in the wrong circumstances — is the centrepiece of such stories, so there are some limits to her victimhood.

⁸ Scheidel, W., ‘Quantifying the sources of slaves in the early Roman Empire’, JRS 87 (1997), 156–69Google Scholar, argues that slave reproduction was actually the greatest source of supply. Whether any such reproduction would have occurred within the brothel itself is unclear, certainly the implication in most sources is that prostitutes successfully strove not to reproduce, which recent studies on the efficacy of ancient methods of contraception and abortion might help explain, see esp. Riddle, J., Contraception and Abortion from the Ancient World to the Renaissance (1992)Google Scholar.

⁹ Justin, , Apol. 1.27Google Scholar. Exposing children thus raises the alarming prospect of committing, not only fornication, but also incest, when visiting a brothel, and so is definitely not to be practised by Christians. Lactantius, (Inst. 6.20.22Google Scholar) repeats the accusation, but with the qualification that exposed children end up either in slavery or the brothel (‘vel servitutem vel ad lupanar’) though this opposition should probably not be pushed too hard.

¹⁰ Cod. 1.4.12 and 11.41.6 ( = CTh 15.8.2). The measure deals with masters as well as fathers.

¹¹ BGU IV 1024.8–18 and Luc, DMeretr. 6. On the more imaginative aspects of the Alexandrian court case, see Keenan, J. G., ‘Roman criminal law in a Berlin papyrus codex (BGU IV 1024–1027)’, AP 35 (1989), 15–23Google Scholar.

¹² Firm., , Math. 6.11.6Google Scholar; Apul., Apol. 75–6.

¹³ Firm., , Math. 3.6.22Google Scholar; see also e.g. 7.25.9.

¹⁴ Dig. 23.2.43.5; Lact., , Inst. 5.8.7Google Scholar; see also the judge's remarks at BGU IV 1024.20–30.

¹⁵ Høigård, C. and Finstad, L., Backstreets: Prostitution, Money and Love, trans. Hansen, K., Sipe, N. and Wilson, B. (1992), 40Google Scholar; and see also e.g. McKeganey, N. and Barnard, M., Sex Work on the Streets: Prostitutes and their Clients (1996), 26Google Scholar. These are, of course, sociological studies of prostitution itself rather than of popular views of it and its causes.

¹⁶ Emp., Sext., P 3.201Google Scholar.

¹⁷ Hershatter, op. cit. (n. 2), 181–209, discusses circumstances and contexts for the selling or pawning of daughters into prostitution in early twentieth century China, which may give some clues in this respect.

¹⁸ Kiefer, op. cit. (n. 1), 59, quotes, for example, a comment by a contemporary German sexologist that, ‘One is born a wife just as one is born a prostitute; and no woman who is meant for free-love becomes a wife by being married’.

¹⁹ With one partial exception, slaves fill all literary brothels where the status of their denizens is clear; the sort of exception is provided by the empress Messalina at Juv. 6.116–32, though the status of the other, more regular, women in the brothel is not clear, and there are also several ambiguous cases.

²⁰ Sen., , Contr. 1.2Google Scholar.

²¹ The nastiest, most avaricious, literary leno is undoubtedly that in the Hist. Apoll. Tyr. 33–7; Xenophon of Ephesus' Tarentine pornoboskos, on the other hand, shows some redeeming qualities at 5.7–8. The slave collar of a meretrix from Bulla Regia in Africa (ILS 9455) also emphasizes the coercive element of prostitution.

²² The tituli and cellae are the most regular feature of Latin literary lupanara, see e.g. Petr., Sat. 7–8, Juv. 6.116–32, and Mart. 11.45. Cellae are also found in the famous Pompeian lupanare at VII.12.18–20.

²³ PSI IX 1055a and Artemidorus 1.78, cf. 4.9. For Greek brothels of an earlier age see Davidson, J., Courtesans and Fishcakes: The Consuming Passions of Classical Athens (1997), esp. 83–91Google Scholar.

²⁴ The initial praise is quoted by Hor., , Sat. 1.2.31–2Google Scholar; the sequel is provided by a scholion in the pseudo-Acronic collection.

²⁵ [Quint.], Maj. Decl. 14–15.

²⁶ Lib., Decl. 25; the general theme seems to be a standard one, see Rhet. Gr. 8.409 (Walz).

²⁷ Ath. 13.5666–718.

²⁸ Dio, , Or. 7.133–52Google Scholar.

²⁹ Muson., Fr. 12 (Lutz) and see e.g. Jones, C. P., The Roman World of Dio Chrysostom (1978), 12–18CrossRefGoogle Scholar for his relations with Dio.

³⁰ See e.g. Juv. 6.121–32 and 11.172–3; Hor., , Sat. 1.2.30Google Scholar; Sen., , Vit. Beat. 4.7Google Scholar.

³¹ See e.g. Mart. 1.34.5–6 and 2.53.7; Hor., , Sat. 1.2.83–5Google Scholar and 101–5;and Ath. 13.5698-f.

³² Wallace-Hadrill, A., ‘Public honour and private shame: the urban texture of Pompeii’, in Cornell, T. J. and Lomas, K. (eds), Urban Society in Roman Italy (1995) 39–62Google Scholar.

³³ CIL IV.2173–296 and CIL IV.3929–43; see on the identification of brothels etc. in Pompeii, Wallace-Hadrill, op. cit. (n. 32), 53 with notes.

³⁴ Hor., , Sat. 1.2.30Google Scholar; [Quint.], Decl. Maj. 14.8, and see also 15.7 and 9–10.

³⁵ Dio, , Or. 7.133Google Scholar and 77–78.4. There is no hint of anything other than an economic motive being involved in prostitutes' presence at these festivals, as also for their presence around temples. Prostitutes did also participate in the religious life of the community like any other group but none of this bears any resemblance to notions of ‘sacred’ or ‘temple’ prostitution that have come to form such a potent part of the historical imagination (on which see generally Beard, M. and Henderson, J., ‘With this body I thee worship: sacred prostitution in antiquity’, Gender and History 9 (1997), 480–503CrossRefGoogle Scholar).

³⁶ Such cellae—single rooms with masonry beds that open directly onto the street — have been found in Pompeii, see Wallace-Hadrill, op. cit. (n. 32), 53 with notes, and their operation seems to be referred to in Ov., , Am. 3.14.9–10Google Scholar, Mart. 1.34.5, and Juv. 3.134–6. The prostitutes at the circus are at Juv. 3.65.

³⁷ See e.g. Prop. 2.23.13–16; Juv. 6.O16; and Mart. 3.93.15 for references to these outdoor operations.

³⁸ All are from Elephantine: O. Wilck. 83 and 1157 are both dated to A.D. III; and SB VI 9545 n. 33 and IV 7399 to A.D. 142 and 144 respectively. Any travelling worker in Egypt needed local leave to pursue their trade in this way.

³⁹ Gell. 4.14.

⁴⁰ [Quint.], Maj. Decl. 14.7.

⁴¹ Ov., , Am. 1.10.21–4Google Scholar.

⁴² On the Latin vocabulary see Adams, J. N., ‘Words for “prostitute” in Latin’, RhM 126 (1983), 321–58Google Scholar; in which, partly through an emphasis on the Republican period, a somewhat different overall view of the terminology is taken than that expressed here.

⁴³ Tert., , Apol. 6.3Google Scholar and Amm. 28.4.9 both use prostibulum (or the feminine prostibula) to designate meretrices; the word is used for a brothel at Vulg. Ezech. 16:34 and Hist. Apoll. Tyr. 33 (earlier recension).

⁴⁴ Paul the Deacon summarizes Festus: ‘scorta appellantur meretrices, quia ut pelliculae subiguntur. omnia namque ex pellibus facta scortea appellantur’ (‘scorta are what meretrices are called, because they are worked like hides. For all things made from hides are called scortea.’) (443.6–7 Lindsay; the Festan text at 442.13–17 is longer but badly fragmented). Later writers offer two explanations for the figurative formation of lupa: the commentator Servius (In Virg. Aen. 3.647) considers the connection between woman and beast to be a shared obscenity and odour, while Isidore of Seville (Etym. 10.163) refers to the rapacity of the she-wolf as she captures her unfortunate prey.

⁴⁵ Davidson, op. cit. (n. 23), 77. See also Hershatter's description of the professional hierarchy in imperial China, which raises the possibility that these hierarchies are always, to a degree, things of the past, that nostalgia is part of the ‘shared imaginary’ through which they are constituted (op. cit. (n. 2), 34–65).

⁴⁶ For Cytheris see e.g. Cic., , Phil. 2.58Google Scholar; Plut., , Ant. 9.4Google Scholar; and Serv., Comm. In Verg. Buc. 10. If Suetonius' work, entitled by John Lydus in the only remaining reference to it Peri Episêmôn Pornôn/On Illustrious Prostitutes (Mag. 3.64), had survived it might (or might not, since what Lydus takes from it is entirely mythical) alter this picture.

⁴⁷ For discussion of the poetic evocations of these mistresses and their relationship to Roman social reality see, e.g. Griffin, J., ‘Augustan poetry and the life of luxury’, JRS 66 (1976), 87–105Google Scholar; Kennedy, D., The Arts of Love (1993), esp. 83–103Google Scholar; and Wyke, M., ‘Mistress and metaphor in Augustan elegy’, Helios 16,1 (1989), 25–47Google Scholar.

⁴⁸ Hor., , Ep. 1.18.34–5Google Scholar; Sen., , Brev. Vit. 16.4Google Scholar, the corrupting combination of scorta and vinum is standard, see also e.g. Livy 23.45.2 and Nep., , Dion 4.4Google Scholar.

⁴⁹ Juv. 3.66; Ov., , Am. 3.14Google Scholar; Mart. 1.34. The figure of the whore who services her customers in cemeteries, or stands naked in a stinking brothel, is also used as a most extremely unlikely guardian of basic standards of propriety, standards scandalously let slip by her social superiors, at Juv. 6.O14–O16 and 11.171–3. Apuleius also uses the shocking juxtaposition ‘uxor lupa’ to describe the wife of Rufinus at Apol. 75.

⁵⁰ The evidence is collected by Stumpp, op. cit. (n. 1), 216–17 and McGinn, T., Prostitution and Julio-Claudian Legislation: The Formation of Social Policy in Early Imperial Rome (unpub. PhD diss., University of Michigan, 1986), 22–3Google Scholar.

⁵¹ CIS II iii.3913 and see Matthews, J., ‘The tax law of Palmyra: evidence for economic history in a city of the Roman East’, JRS 74 (1984), 157–80Google Scholar and further discussion below. It is worth mentioning that the earlier Palmyrene version of the law translated hetairai as ‘slave girls’ (P 122) implying that prostitutes were slaves in Roman Palmyra.

⁵² Mart. 2.53 and 1.103; Juv. 3.132–6. The ludicrous price put on Tharsia in the lupanar of the Hist. Apoll. Tyr. (33) — half a pound of gold initially, one aureus/100 HS thereafter — surely simply befits her status as a beautiful princess, rather than anything else. As Duncan-Jones, R. says about the novel in general: ‘Figures that have dramatic meaning but no external significance predominate among the prices’ (The Economy of the Roman Empire (2nd edn, 1982), 252Google Scholar).

⁵³ This is the kind of distinction Davidson builds his hierarchy on, op. cit. (n. 23), 109–36.

⁵⁴ Ov., , Am. 1.10Google Scholar.

⁵⁵ See e.g. Høigård and Finstad, op. cit. (n. 15), 130–1, and also the remarks of White, op. cit. (n. 2), 12–15. Of course, street-walkers may decide to pursue their career in a more relaxed fashion.

⁵⁶ Høigård and Finstad, op. cit. (n. 15), 129.

⁵⁷ White, op. cit. (n. 2), 13–16 and 103–25.

⁵⁸ Priap. 40 presents itself as a poem from the prostitute Telethusa marking the purchase of her freedom, but this is the only hint at such a practice.

⁵⁹ Juvenal (10.236–9) lambasts a man who left all his money to a whore who had been displayed for many years within the walls of a fornix, but the extremities of the case suggest that even more modest versions would have been rare, though the general risk of fortunes being lost in brothels is reflected in the various stories about the over-involvement of young men with meretrices which usually turn on their disinheritance, e.g. Quint., , Inst. 7.4.20Google Scholar and 11.1.82–3, and Calp. Flacc. 30.

⁶⁰ McGinn, op. cit. (n. 50), 24; Macr., , Sat. 1.10.16Google Scholar; this is, of course, not the only story told about Larentia and her lands, but that it could be told is significant none the less.

⁶¹ On infamia see generally the remarks of Levick, B., ‘The senatus consultum from Larinum’, JRS 73 (1983), 109Google Scholar; and see also, on the particular issues discussed here, McGinn, op. cit. (n. 1), 26–69, Gardner, J., Being a Roman Citizen (1993), 110–54Google Scholar, and Edwards, C., ‘Unspeakable professions: public performance and prostitution in ancient Rome’, in Hallett, J. and Skinner, M. (eds), Roman Sexualities (1997), 66–95Google Scholar.

⁶² Dig. 3.1.1.5 — women; 3.1.1.8 — infames, the others (mostly family members) on behalf of which they can act are specified at 3.1.1.11.

⁶³ Dig. 3.2.1.

⁶⁴ Calp. Flacc. 5; the same controversia is referred to at Sen., Contr. 10.1.13–15, where a selection of sententiae abusing the leno is preserved, and, more briefly, at Rhetores Latini Minores 83.1. For similar attacks on the ethical and legal persona of the leno, see also [Quint.], Min. Decl. 385, where a leno is attempting to bring an action datnnum iniuria datum against a man who administered a love-potion to one of his slave meretrices.

⁶⁵ Dig. 22.5.3.5.

⁶⁶ [Quint.], Decl. Maj. 14.5.

⁶⁷ Other declamations featuring prostituted slaves include Calp. Flacc. 37; [Quint.], Decl. Min. 356, and, of course, Sen., Contr. 1.2.

⁶⁸ Dig. 3.2.4.2–3.

⁶⁹ Dig. 5.3.27.1, for instance, ensures that rents from brothels can be claimed as part of an inheritance, for such establishments are run on the properties of many honesti viri; and see also CIL 111.13750 which is also discussed below.

⁷⁰ For discussion of this provision, and the problems that surround it, see e.g. McGinn, op. cit. (n. 1), 70–104, Treggiari, S., Roman Marriage (1991), 62–3Google Scholar, and Gardner, op. cit. (n. 61), 123–6.

⁷¹ Dig. 23.2.43.pr– 3 and 7–9; cf. 23.2.41. pr, where Marcellus implicitly interprets the legislative ‘palam’ rather differently, as signifying an openness of publicity, rather than of accessibility, as for Ulpian; other juristic usage also tends to favour Marcellus in this respect. However, since what Marcellus is arguing is that the ‘palam’ is an unnecessary qualification, he agrees with Ulpian that it is accessibility, not publicity, that is at issue in incurring infamia and the penalties of the Julian law. These points are also discussed at McGinn, op. cit. (n. 1), 123–39.

⁷² Quint., , Inst. 8.4.2Google Scholar; cf. Cic., , Cael. 16.38Google Scholar.

⁷³ Sen., , Ben. 6.32.1Google Scholar.

⁷⁴ On the poetic figure of the lena see Myers, K. Sara, ‘The poet and the procuress: the lena in Latin love elegy’, JRS 86 (1996), 1–21Google Scholar; and see CIL IX.2029 for Vibia.

⁷⁵ Dig. 37.12.3.pr covers the removal of turpes personae as heirs; and the Gnomon of the Idiologos actually sets a minimum inheritance below which the Lex Julia does not apply, which may reflect practice elsewhere even if not the letter of the law, see Treggiari, op. cit. (n. 70), 78–9. There are sufficient legal (and other) references, however, to indicate that meretrices did marry, or form other kinds of legally recognized unions, see e.g. Dig. 23.2.24 and 47; 25.7.3pr and 48.5.14.2.

⁷⁶ See Buckland, W. W., The Roman Law of Slavery (1908), 70–1Google Scholar and 603–4 and McGinn, T., ‘Ne serva prostituatur: restrictive covenants in the sale of slaves’, ZRG 107 (1990), 315–53Google Scholar and now also McGinn, op. cit. (n. 1), 288–319.

⁷⁷ SHA, Hadr. 18.8; Dig. 1.12.1.18. The phrasing of the Severan instruction suggests that the Praetor's duties were connected with the enforcement of ‘ne prostituatur’ covenants.

⁷⁸ Dig. 3.2.24 and 38.1.38pr.

⁷⁹ McGinn, op. cit. (n. 76), offers a similar explanation, but one which focuses more on the preservation of the household's honour in all this.

⁸⁰ Tac., , Ann. 2.85Google Scholar.

⁸¹ Suet., , Tib. 35.2Google Scholar; Dig. 48.5.11.2; and see also Levick, op. cit. (n. 61).

⁸² Sen., , Vit. Beat. 6.7.3Google Scholar; the case of the meretrix Manilla and the aedile Hostilia has also been used as evidence of this register, but is less clear in this respect than is often suggested.

⁸³ The introduction of the tax is described in Suet., Calig. 40 and Dio ap. Xiph. 59.28.3; its collection in the second century A.D. by both publicani and troops is witnessed by at least two receipts on ostraka from Egypt (O. Wilck. 83, O. Cair. GPW 60, and see also O. Edfou 171), the bilingual inscription of the tax law from Palmyra, (CIS II.iii.3913Google Scholar), and an inscription from Chersonesus on the Black Sea (CIL 10.13750); I. Portes. 67 also records a harbour passage tax for prostitutes at Coptos on the Red Sea, but the structure of the tariff is too poorly understood to make much of it.

⁸⁴ McGinn, T., ‘The taxation of Roman prostitutes’, Helios 16 (1989), 87Google Scholar; and see the further comments of Bagnall, R., ‘A trick a day keeps the tax man away? The prostitute tax in Egypt’, BASP 28 (1991), 5–12Google Scholar and now also McGinn, op. cit. (n. 1), 248–87.

⁸⁵ McGinn, op. cit. (n. 84), 80–6 and 98–9.

⁸⁶ This is expressed most clearly by Evagrius at HE 3.39 when he discusses the eventual abolition of the tax by Anastasius in A.D. 498. And it is within the context of a Christian Empire with a new public policy towards prostitution that legitimation becomes an issue, as Evagrius points out.

⁸⁷ Its economic effects will have depended, in part, on whether the amount of the tax — set at the amount received for a single concubitus — accrued daily, as McGinn argues (op. cit. (n. 84), 90) or monthly, which is the more traditional view. The question is not really resolvable, but Bagnall (op. cit. (n. 84), 9–12 ) opens the way to a possible reconciliation between the evidence that suggests prostitutes were taxed much more heavily than other earners — that is the shocked report of Suetonius, the survival of the tax for so long under the Christian emperors, and Evagrius' account of its eventual demise — and the evidence of its collection in Egypt and Palmyra which puts it more in line with normal, monthly, accruals. For, as he points out, there is no reason to assume uniformity in implementation across the Empire; moreover, it seems to me suggestive that these latter pieces of evidence come from contexts where collection was by tax-farmers, not the military. McGinn also moves in a more flexible direction at op. cit. (n, 1), 264–8 and 274–86.

⁸⁸ CIL III.13750 = B. Latyschev, Inscriptiones Antiquae Orae Septentrionalis Ponti Euxini Graecae et Latinae I² No. 404; and see discussion by McGinn, op. cit. (n. 84), 88–90 and op. cit. (n. 1), 261–4.

⁸⁹ On the legal repression that led to the development of a pimp system in late nineteenth-century England, for instance, see Walkowitz, op. cit. (n. 2), 210–21.

⁹⁰ CIL III. 13750, 36 and NTh 18.

⁹¹ See Bagnall, op. cit. (n. 84) for details. I would further suggest that O. Edfou 171, which not only contains a broken text, but is also made out to a known poll-tax payer from a well-attested Jewish family, could well be made out to a profiteer from prostitution of the kind met elsewhere, rather than an actual pimp or simply the victim of a general apportionment of a shortfall in the telos hetairikon which are the two possibilities discussed by Bagnall.

⁹² P.Lond. inv 1562 verso, 19–20, published and discussed by Rea, J. in ZPE 46 (1982), 191–209Google Scholar.

⁹³ It is worth remembering, when reading Evagrius' account of arguments by those opposed to the abolition of this tax that it would leave the state unable to pay the army or serve God in the manner to which he was accustomed, the massive size of the global sex industry today and the fact that UN reports in the 1980s estimated that more money was being made in the trafficking of women, mostly for prostitution, than smuggling drugs or arms (see, e.g. S. Altink, Stolen Lives: Trading Women into Sex and Slavery (1995), 2).

⁹⁴ McGinn, op. cit. (n. 1), 140–215, argues (inter alia) that the Augustan adultery law deploys the categories of meretrix and leno against adulterous matronae and their complaisant husbands, both as deterrent and punishment. The relevant legal notions (especially that of lenocinium) were certainly part of the juridical matrix within which this legislation was formulated and elaborated, but McGinn's particular points are harder to make. For the main vehicle through which he claims the assimilation of adulterer to prostitute was achieved — the toga both were allegedly meant to wear — is problematic in many respects, and McGinn has to strengthen the slight contemporary evidence by recourse to later Christian writers who are, needless to say, participants in a rather different moral discourse. In a sense this just goes to emphasize that, in contrast with what was to follow, remarkably few women of the pagan Roman world were called ‘whores’, at least in polite society.

⁹⁵ For a recent discussion of these issues within the abolitionist tradition see e.g. Jeffreys, S., The Idea of Prostitution (1997)Google Scholar.

⁹⁶ This comparison is based on the definition of slavery formulated by Orlando Patterson in his Slavery and Social Death (1982), in particular his highlighting of the dynamics of honour and degradation within slavery. Of course, it is not the prostitutepunter relationship that is the slave-master one in actual slave prostitution.

⁹⁷ For a recent discussion within this tradition see e.g. Chapkis, W., Live Sex Acts: Women Performing Erotic Labour (1997)Google Scholar.

⁹⁸ See e.g. Walkowitz, op. cit. (n. 2), 13–41, and Hill, op. cit. (n. 2), 63–106, for the past and Chapkis, op. cit. (n. 96), 47–8 for the present.

⁹⁹ White, states categorically (op. cit. (n. 2), 6) that, ‘There were no pimps in Nairobi or anywhere else in Africa, outside of Johannesburg in the 1890s’. She also offers a range of evidence to support her wider claim that, ‘men and male control enter prostitution only after the state does’ (4). This claim ties in with the conclusions of other recent studies, but slavery provides an exception.

¹⁰⁰ Patterson, op. cit. (n. 96), 26.

¹⁰¹ Fogel, R. and Engerman, S., Time on the Cross: The Economics of American Negro Slavery (1974), 135Google Scholar. They offer only one statistic about the absence of slaves from prostitution — that in 1860 only 4.3 per cent of Nashville prostitutes were black (and none were slaves) while 20 per cent of the overall population was black — but that Southern prostitutes were overwhelmingly white, and even more overwhelmingly free is, however, not disputed.

¹⁰² See e.g. Bynum, V., Unruly Women: The Politics of Social and Sexual Control in the Old South (1992), 79–80Google Scholar and Clinton, C., The Plantation Mistress: Woman's World in the Old South (1982), 221Google Scholar.

¹⁰³ Muson., Fr. 12 (Lutz).

¹⁰⁴ I take support for this suggestion from the places, mostly busy ports, where prostitution, including of some slaves, did take significant root in the Americas, see e.g. Lockley, T., ‘Crossing the divide: interracial sex in antebellum Savannah’, Slavery and Abolition 18 (1997), 159–73CrossRefGoogle Scholar. This is also what seems to have happened to more slaves in Brazil, see e.g. Karasch, M., Slave Life in Rio de Janeiro, 1808–1850 (1987), 207Google Scholar, and the Caribbean, see e.g. Higman, B., Slave Populations of the British Caribbean, 1807–1834 (1984), 231–2Google Scholar and Beckles, H., ‘Black female slaves and white households in Barbados’, in Gaspar, D. and Hine, D. (eds), More than Chattel: Black Women and Slavery in the Americas (1996), 121–2.Google Scholar

¹⁰⁵ The complaint of Mary Boykin Chesnut, the abolitionist wife and daughter of major slave-owners, that, ‘we live surrounded by prostitutes’, is one of her key objections to slavery; that is the wrongs it perpetrates against white women married to men who make full use of these black ‘prostitutes’ (C. Van Woodward (ed.), Mary Chesnut's Civil War (1981), 29). Proslavery ideologue William Harper, on the other hand, praised this state of affairs, claiming that the ‘prostitution’ of all slave women was a benefit both to the women themselves and to the white men who were ‘less depraved’ by intercourse with them than they would be by intercourse with ‘females of their own caste’ (Faust, D. G. (ed.), The Ideology of Slavery: Proslavery Thought in the Antebellum South, 1830–1860 (1981), 104–7Google Scholar).

¹⁰⁶ On the general security of Roman slavery see Garnsey, P., Ideas of Slavery from Aristotle to Augustine (1996)Google Scholar.