which they cooperated and competed with one another; the risks and logistical challenges they faced; the rewards they received; and the role of states in constraining, directing, and taxing their activities. It shows that the Black Sea slave trade was not conducted by professional slave traders—there were no specialists who made their living chiefly by trading or shipping slaves. Instead, the slave trade was conducted by opportunists, buying and selling slaves alongside other commodities and transporting them in mixed-cargo ships.
The final chapter, “Crusade, Embargo, and the Trade in Mamluk Slaves,” situates the Black Sea slave trade within the religious and diplomatic contexts of the late medieval crusade movement and the broader struggle between Christian and Muslim powers for control of the Mediterranean. Christian proponents of the crusades asserted that Christian merchants, especially the Genoese, were strengthening the enemy by supplying the Mamluks with military slaves from the Black Sea. They advocated that slaves be included in the papal embargo policy against the Mamluks. Although Genoa did play an important role in facilitating the Mamluk slave trade, it was the state rather than individual merchants that negotiated the terms of engagement with the Mamluks, and it was the state that struggled to reconcile its slaving and crusading activities. Thus the Mediterranean trade in Black Sea slaves should be understood not only as an economic activity carried out by merchants but also as an area of state regulation with significant diplomatic and religious consequences.
Chapter 1
Slavery in the Late Medieval Mediterranean
When medieval Christians imagined an ideal society, order was among its guiding principles. The ideal society envisioned by Thomas Aquinas, for example, was set in the Garden of Eden, where he believed that humanity had existed in a sinless natural state.1 Because he also believed that humanity was naturally sociable, the inhabitants of Eden must have lived together in a society. Within that society, there would be naturally occurring differences of gender, age, and physical stature, as well as the differences of goodness and knowledge that would emerge as individuals exercised their free will. Those who cultivated goodness and knowledge would have dominion over the others, but without coercion. Dominion was necessary in Eden, according to Aquinas, for two reasons. First, society needed a ruler to look after the common good. Second, the wisest and best members of society were obligated to use their virtues for the benefit of all. Thus differences among people, both those naturally occurring and those developed through free will, led to social hierarchy. Accepting one’s place in the hierarchy was necessary for peace and order; refusing it constituted pride, the greatest sin.2
The ideal society imagined by medieval Muslims was different. Its setting was paradise, the Garden of the afterlife, where believers experienced never-ending joy surrounded by every kind of beauty and abundance. One aspect of this joy was reuniting with family members across generations. In early descriptions of the Garden, all the branches of an extended family were depicted living together in an infinitely spacious tent or pavilion, thus combining the pleasures of sociability and privacy. They would enjoy banquets together without the work of preparing and serving food. Instead, a retinue of beautiful male ghilmān and female ḥūr would serve the food, fill the cups, and provide the entertainment.3 The ghilmān and ḥūr were not human and had never experienced worldly life. Their sole function was to serve. Like splashing fountains and fragrant trees, they can be understood as living elements of the Garden’s adornment. Over time, depictions of the Garden evolved such that female ḥūr came to be portrayed as singers and sexual partners for the enjoyment of male believers, while female believers and male ghilmān faded into the background.4
Real societies in the late medieval Mediterranean were, of course, not anyone’s ideal. Real rulers used force to exercise dominion over their subjects, and real households depended on the labor of real people for their meals. The lowest position in the human hierarchy was in reality filled by slaves. It can be difficult today to understand how masters could recognize the humanity of their slaves while still denying them freedom, but it was precisely their humanity that made slaves valuable. Simply by existing at the bottom of the hierarchy, slaves enabled free people to feel superior. Their presence in a household enabled its head to display his skills as a ruler. Slaves’ reason and capacity for decisionmaking made them capable administrators and soldiers; their capacity for love and companionship made them suitable nurses for children and the elderly as well as sexual partners for their masters.
Christians and Muslims shared three fundamental assumptions about slavery in addition to its place at the bottom of the human hierarchy: that it was legal, that it was based on religious difference, and that it was a universal threat. These assumptions formed the core of a common culture of slavery in the late medieval Mediterranean. All free adults knew that they could legally purchase a slave. They expected the slaves available for sale to come from religious backgrounds different than their own. At the same time, they were aware that they themselves could become enslaved if captured by pirates or raiders in the wrong place at the wrong time.
The line between slavery and captivity in the late medieval Mediterranean was hazy. Captivity ended with a return home after the payment of a ransom or some other mechanism of exchange, and it was not necessarily associated with religious difference.5 Slavery was ideologically based on religious difference, and it was usually permanent. It is only clear in retrospect which individuals would be ransomed from captivity and which would remain in slavery. Ransoms, however, were easier to arrange for captives held in geographical or social proximity to their homes than for captives taken far away. In the western Mediterranean, ransoms and exchanges between Iberia and North Africa were mediated by friars, merchants, confraternities, and monarchs.6 In the eastern Mediterranean and the Black Sea, merchants and diplomats usually arranged ransoms.7 But for slaves transported over long distances, from the Aegean to Catalonia or from the Black Sea to Italy, no ransom could be expected.
This distinction between captivity and slavery shaped expectations regarding conversion. Captives taken across religious boundaries might or might not be pressured to convert, depending on how their masters believed conversion would affect their ransom.8 Slaves, for whom a ransom was not expected, were converted to the religion of their masters, and such conversions did not confer freedom. Christians who purchased Muslims, Jews, and pagans expected them to convert to Christianity while remaining enslaved. Muslims who purchased Christians, Jews, and pagans as slaves expected them to convert to Islam while remaining enslaved. Jewish law allowed partial conversion to Judaism for slaves, with the possibility of full conversion after manumission.9
One consequence of assuming slave conversion without manumission was a rhetoric of competition for slave souls. Conversion was one of the justifications given for slave ownership, because it increased the number of souls belonging to the “right” religion. In the late Middle Ages, some Christians also argued that it was better for Christians to purchase Christian slaves than to allow them to be purchased and converted by Muslims. In other words, they used the rhetoric of competition for slave souls to justify the enslavement of Christians by Christians. As a result, thousands of Christians from the eastern Mediterranean and the Black Sea were kept as slaves in the central and western Mediterranean. Some challenged their status in court, arguing that legal slavery had to be based on religious difference. A few won their freedom. Muslim scholars in the late medieval Mediterranean never presented an intellectual argument for the enslavement of fellow Muslims; whether they enslaved fellow Muslims through negligence is a different question. Given such complications, this chapter explores how slave status was defined in the late medieval Mediterranean and how the boundary between free and slave status was enforced.
Terminology
The English language has only two words, slave and serf, to indicate a person of unfree status. The vocabulary of Latin and Arabic is richer. In classical Arabic, all of the following terms may be translated as slave: ‘abd, raqīq, ghulām, fatan, khādim and khādima, mamlūk and mamlūka, waṣīf and waṣīfa, jāriya, and ama.10 In Mamluk-era texts, a male slave was usually called