Concubinage and Consent

In our imperfect world, rape happens frequently but nearly no one publicly defends the legitimacy of forcible or nonconsensual sex. So pervasive is deference to some notion of consent that even Daʿish supporters who uphold the permissibility of enslaving women captured in war can insist that their refusal or resistance makes sex unlawful. ¹ Apparently, one can simultaneously laud slave concubinage and anathematize rape. A surprising assertion about consent also appears in a recent monograph by a scholar of Islamic legal history who declares in passing that the Qurʾan forbids nonconsensual relationships between owners and their female slaves, claiming that “the master–slave relationship creates a status through which sexual relations may become licit, provided both parties consent.” She contends that “the sources” treat a master's nonconsensual sex with his female slave as “tantamount to the crime of zinā [illicit sex] and/or rape.” ² Though I believe in the strongest possible terms that meaningful consent is a prerequisite for ethical sexual relationships, I am at a loss to find this stance mirrored in the premodern Muslim legal tradition, which accepted and regulated slavery, including sex between male masters and their female slaves. ³

Western scholars have generally assumed that in Islamic jurisprudence, milk al-yamīn, typically rendered “ownership by the right hand,” automatically granted free male owners licit sexual access to enslaved females whom they owned. ⁴ My research on marriage and divorce in formative-period Sunni legal texts paid close attention to the jurists’ frequent analogies between marriage and slave ownership, as well as to doctrines governing marriages involving enslaved persons. I showed that jurists understood milk al-nikāḥ (marriage) and its attendant spousal claims through analogies with gendered and sexualized slavery. I never explored the possibility that the jurists considered an enslaved female's consent necessary for a licit sexual relationship outside of marriage. I do so now in this brief essay. Although I limit myself to formative-period sources, the main contours of shared legal doctrines on milk al-yamīn persist until the modern era.

Notably, Qurʾanic passages on slavery differ strikingly in terms of their terminology and main preoccupations from later jurisprudential texts. ⁵ That the text of the Qurʾan does not permit sexual access simply by virtue of milk al-yamīn is a defensible theological claim. ⁶ Whether jurists took this stance is a historical question. If—as I have assumed—they did not, then to accept the former claim means that the jurists misunderstood or departed from scripture by disregarding enslaved women's consent. The other possibility is that generations of scholars, including me, have misunderstood the legal tradition.

Did a man who wanted to have sex with his own female slave need to obtain her consent for that relationship to be licit according to early Muslim jurists? It is difficult to prove a negative, but the answer seems to be a clear no. Any argument must be largely from silence, as the sources simply do not discuss the issue. I recall no instance in any Maliki, Hanafi, Shafiʿi, or Hanbali text from the 8th to 10th centuries where anyone asserts that an owner must obtain his female slave's consent before having sex with her. Indeed, I am aware of no case where anyone asks whether her consent is necessary or even asserts that it is not required. The mere absence of discussion proves nothing, of course. Sometimes things escape mention because they are universally accepted. What jurists take for granted—particularly across madhhab boundaries—is often more telling than what they state explicitly. One could perhaps argue that slaves’ consent to sexual relationships with their masters was such an obvious requirement that no one thought it necessary to mention. Yet in sharp contrast to their silence about slaves’ consent to sex with their owners, scholars paid significant attention to consent to marriage. They agreed unanimously that an enslaved female's consent was never required for a marriage contracted by her owner. Al-Shafiʿi (d. 820) is typical: “He may marry off his female slave without her permission whether she is a virgin or non-virgin.” ⁷ It strains logic to suggest that an enslaved woman is subject to being married off without her consent or against her will to whomever her owner chooses but that he cannot have sex with her himself without her consent. It is even more of a stretch to accept that the need for consent within concubinage was so obviously a condition for its legitimacy that no one considered it necessary to say so, but that the absence of the need for a slave's consent to her marriage required explicit affirmation.

A slightly different example reinforces the legal distinction between marriage and concubinage. In discussing withdrawal (ʿazl) as a method of contraception the jurists distinguish between consent (possibly) required from wives and that (never) required from enslaved concubines. They disagreed about whether husbands needed their enslaved wives’ agreement to practice ʿazl or that of their wives’ masters. (A person cannot simultaneously own and be married to the same slave, though people can under certain circumstances marry other people's slaves.) All accepted—sometimes tacitly, sometimes explicitly—that a man could practice withdrawal with his own female slave without seeking her permission. ⁸

As this example demonstrates, the rights of wives and slaves differ. Milk al-nikāḥ and milk al-yamīn are incompatible; they cannot be combined. If a man who is married to someone else's slave comes to own her (e.g., via gift, purchase, or inheritance) she ceases to be his wife, but sex remains lawful by virtue of milk. ⁹ No jurists discuss her consent to the transfer of ownership; none suggest that her consent is preferred let alone required for the continuation of their relationship under its new regime. If, on the other hand, a man wishes to marry his own female slave, he must either sell her to someone else, who in turn must consent to the marriage, or he must free her, making her own consent to the marriage necessary. (The manumission must take place before the marriage, else the marriage would impermissibly mingle the two sorts of milk; if she agrees but after being freed refuses to marry him, she may owe compensation but is not re-enslaved.)

The jurists’ works, which both liken milk al-nikāḥ and milk al-yamīn and distinguish between them, provide no basis for the claim that nonconsensual sex within the latter is “tantamount to the crime of zinā and/or rape.” Jurists define zinā as vaginal intercourse between a man and a woman who is neither his wife nor his slave. Though seldom discussed, forced sex with one's wife might (or, depending on the circumstances, might not) be an ethical infraction, and conceivably even a legal one like assault if physical violence is involved. ¹⁰ One might speculate that the same is true of forced sex with an enslaved woman. This scenario is never, however, illicit in the jurists’ conceptual world. Nonconsensual sex—what contemporary Westerners would term rape—might be either a coercive subset of zinā, with blame lifted from the coerced participant, or a type of usurpation (ightisāb), a property crime that by definition cannot be committed by a husband or owner, who possesses an entitlement to, or ownership over, his wife's or slave's sexual capacity. Thus, as Hina Azam writes, “sexual coercion within clearly licit contexts (such as marriage or slavery) . . . fell outside the scope of zinā.” ¹¹

Azam's observation about classical Hanafi texts applies more broadly to the premodern fiqh universe: “concerns about consent in sex acts were secondary to concerns about the moral-legal status of those sex acts.” ¹² In other words, and in sharp contrast to our contemporary situation, consent was not a key “moral-legal concern.” A man's intercourse with a female slave might constitute zinā only if she belongs to someone else. Even if he marries off his own slave and no longer has lawful access to her, his having sex with her is a lesser transgression than zinā. The jurists’ occasional affirmations that a married female slave whose owner nonetheless has sex with her is not to be punished is the closest any of these texts comes to considering the relevance of an enslaved woman's consent. Notably, the issue emerges only because she is married to another man, a marriage for which jurists uniformly agree that her consent would have been unnecessary.

In sum, the books of marriage, divorce, and related topics in formative period Sunni fiqh compilations express no explicit concern whatsoever with the consent of an enslaved female to a sexual relationship with her owner. Further research might extend beyond the legal contexts of marriage, divorce, support, and manumission to other portions of fiqh texts, such as books of sales. Other types of texts, too, may preserve countervailing voices, in tension with the ideal fiqh model in which enslaved women's consent is simply disregarded. A handful of intriguing accounts in biographical and hagiographical texts portray (exceptional) female slaves who preserved their chastity. For instance, the first enslaved woman to be mother to a Shiʿi imam—the previous imams having been born to free women—preserved her virginity despite passing through multiple owners. Clearly intended to demonstrate miraculous workings in the lives of the imams, the story assumes that without divine providence, a desirable enslaved female would have no choice about her sexual partners. ¹³ Although not directly about consent to sex, one account in Ibn al-Saʿi's (d. 1275) Nisaʾ al-khulafaʾ (Consorts of the Caliphs), which mingles stories of Abbasid-era wives and enslaved concubines, discusses an enslaved woman sought for purchase by an elite man for an enormous sum. Her female owner asks her if she wishes to be sold—which would presumably entail becoming the purchaser's concubine. She says no; her owner frees her on the spot. ¹⁴ At least in literary retellings, then, some slaves had a certain amount of say in the conduct of their sexual lives. (Read in a certain light, a few passages in jurisprudential texts suggest submerged evidence of enslaved women's ruses to avoid sex with their owners. ¹⁵ If they do reflect such stratagems, they further show that a slave's simple “no” would not suffice.) In literary anecdotes, however, the point is not to record lived experience, but to affirm an imam's miraculous status, a concubine's cleverness, or an owner's largesse.

These stories also call into question the usefulness of the legal category of slave as an historical and social descriptor. Any enslaved female had a fixed set of enforceable rights (e.g., food and shelter) and duties (i.e., work as her owner chose, including sex with him). In practice, the sultan's favorite and a palace drudge might have had little in common beyond their formal status. Writing about a much later and better-sourced period, Ottoman historian Ehud Toledano presents a model of unequal but reciprocal relationships between slaves and owners. ¹⁶ One can think about the possibilities for resistance and agency in what were, after all, human—even if deeply unequal and unjust—relationships without succumbing to the temptation to read contemporary ethical norms into early Muslim texts which possess very different sensibilities.