Slavery and Islam: What is Slavery?

Is there slavery in Islam? When people pose this question they usually assume it’s the Islam part that needs clarification. Everyone already knows what slavery is. Actually, it’s quite the opposite. The Islam part is relatively straightforward. The real problem is trying to pin down what we mean by slavery. The more we scratch the surface of that word and try to define it, the more we find that our assumptions and even our words fail us. What we think we mean by slavery means little outside our own American experience, and the moment we try to fix what slavery is as a human phenomenon we find a hall of mirrors reflecting our own assumptions back at us. We all think we know what slavery is, but would we really know slavery if we saw it?

Imagine we could explore the phenomenon of slavery throughout history. Imagine that, as huge Doctor Who fans, we hitch a ride in the Tardis, which allows us to travel across space and time. Our first stop is an exotic, desert land where slavery is common. We visit a well-off home, where we find certain people performing domestic work while an older man sits drinking tea. Everyone has the same dark skin color. Suddenly the lounging tea-drinker shouts at a young man serving him and smacks him hard with a fly swatter. We are eager to know who all these people are. Fortunately, the TARDIS translates all languages directly to your brain. We ask one of the men serving tea his name, and he says his name is Saffron and that he is “one of the delicate folk” working in the household. He has worked in this house for five years, but he tells us that, in one year’s time, he’ll have saved enough money to move on and start his own teashop. We ask about the young man getting smacked. “Oh, that poor boy… he’ll be here till the old man dies.”

Where has the TARDIS taken us in our exploration of slavery? The first place we visited was the city of Mecca in the 1400s. The ‘soft and delicate (raqīq)’ man Saffron was a slave in the wealthy man’s household who had an agreement with his master to buy back his freedom on installments (mukataba). Raqīq was the standard term for slave, and epicurean names like Saffron were typical. The younger man being smacked for bad service, who was tied to the household seemingly forever, was the wealthy man’s own son.

The second place we visited was the capital of the Ottoman Empire in 1579. The minister was Sokollu Mehmet Pasha, the grand vizier and de facto ruler of the empire during the time of three sultans. At the time of our visit, he had already been one of the empire’s richest and most powerful men for almost two decades.  He was also a slave of the sultan. He was born in Bosnia, as were all his guards, who were also slaves of the sultan.^[1]

The land where we met the man working in a clock factory was England in 1860.  Although the worker was a free man, according to labor laws in England at the time a worker who failed to show up for work was guilty of stealing from his employer and was tried and sentenced as a criminal. Finally, the last place we visited was a land in which slavery had long been illegal: rural Arizona in 2004, where the local sheriff was overseeing a juvenile chain gang.

If we are searching for the phenomenon of slavery, what are we really looking for? Is it the label ‘slave’ that matters? Or is it the reality of the condition behind it? The soldiers and administrators of China’s Manchu Qing dynasty (1644-1912) were technically slaves (aha) of the dynasty and proudly referred to themselves as such. The title of slave was later applied to anyone of Manchu descent in Qing China. But the word had no link to the reality of any servile condition.^[2] Up through the 1800s, the upper administration of the Ottoman Empire was in the hands of people technically classified as kul (privileged sultanic slaves) who had more power and esteem than their free counterparts.^[3]

When we come across a word that translates as ‘slave’ in English, does that word necessarily mean what we mean by slavery? Our word ‘slave’ in English comes from the Medieval Latin word for Slavic peoples, Sclavus, since they were the population in the Balkans from which European slave traders drew their cargo up through the thirteenth century.^[4] A common English dictionary definition of a slave is ‘someone who is legally owned by another person and is forced to work for that person without pay.’ This notion of slavery as reducing human beings to things owned by other people has been a major theme in how the concept has been understood in the West. It was crucial to how abolitionists understood slavery in the eighteenth and nineteenth centuries when the movement to end slavery began. But the roots of this definition go further back to the roots of Western heritage. They lie in Roman law, which divided people into two categories: the free (a free person has the ‘natural right’ to ‘do as he pleases, unless prevented by the force of law’) and slaves, who exist as the property of others.

Ownership is as much about how we imagine relationships as exercising real control. As the famous social historian Orlando Patterson points out, who and what we say we own is really only a matter of our customs and manners.^[5] Modern Americans would gasp at the notion of ‘owning’ their children, but from the Roman through the medieval period in Europe parents could and did sell their children off as slaves to creditors in order to pay debts. Moreover, poor parents abandoning their children was a regular source for slave markets in Europe.^[6] Yet all these children started off as technically ‘free’ in the legal sense, not legally owned by anyone. In the US, wives and husbands have numerous claims on and powers over each other and their labor, as becomes clear during divorce.^[7] But we would never speak about marriage as a relationship of ownership. Conventions in early imperial China were different. There, husbands regularly listed their (free!) wives as property in their will, bequeathing them to some friend.^[8] Astoundingly, between 1760 and 1880—less than a century and a half ago—there were 218 cases of Englishmen holding auctions to sell off their wives, even advertising these auctions in the newspaper.^[9]

What would it mean to ‘own’ a person? Does it mean to have total control over them? We have full control over our young children, but, unlike a chair or a pen, we cannot seriously physically harm them without legal consequence. In fact, this distinction between ownership and control is not very helpful for defining slavery. As with our children today, it was impermissible for Muslims to kill or seriously injure their slaves, and those who did faced legal consequences under the Shariah. In some contexts, ownership might fail completely as a concept for understanding slavery. Slavery existed in imperial China, but it was not conceptualized through ownership. Slaves were not legally ‘owned’ at all for the very technical reason that Chinese law could not categorize people as ‘things.’^[10]

We usually think of slavery as something that exists in a dichotomy with freedom. But what does freedom mean? As the legal scholar Vaughan Lowe jibes, inverting Rousseau’s famous line about man’s natural state of freedom, “Man is born in chains, but everywhere he thinks himself free.”^[11] Almost no human being is free of dependence on others and on society as a whole. Almost everyone is forced to work in order to earn wages to buy food. The son in the household we visited in Mecca was technically free but he depended on his father for all his support and had to obey him or face his anger. If he fled his home to get away from his nasty father, he’d be ostracized by all those he knew and loved. The man’s slave, meanwhile, had evenings off to earn his own money and would soon be free of his master. Who was free in this situation?

At a theoretical level, how we understand freedom in the West is inherited from Classical Greece and Rome, where ‘free’ was the legal category of citizens of a democratic republic. A free person is autonomous, at liberty to do whatever he or she wants unless the law prohibits it. Everyone else is a slave. But even in Classical times, this legal definition of freedom was no more than a “rhetorical argument,” as one scholar puts it, since in reality few people in the Greek and Roman world were ‘free’ by this definition. Almost everyone was constrained by powerful social, economic, and even legal bonds.^[12] Ironically, even in theory this notion of freedom only applies in liberal democracies. In autocracies—perhaps a majority of societies in human history—almost no one is free by this definition.^[13]

Nor does freedom exist on a single plane. It is often relational, expanding or contracting depending on the relationship in question. In the ancient and medieval Mediterranean world (both Europe and Islamic civilization), a slave’s intense subordination was not absolute. He or she was subordinated to his or her master, not to society as a whole. So Roman and later Byzantine masters used slaves to run their shops and to be the public faces of their businesses, negotiating and arguing with countless ‘free’ customers and contractors on a daily basis.^[14] The slave was not the lowest rung on the ladder in the streets of Rome or Constantinople/Istanbul. If their master was a powerful or wealthy person, the slave enjoyed the status of that connection in public life. The status of the slave depended on the status of his or her master.

By now you should see that any question about slavery is very complicated. One of the biggest challenges that historians and anthropologists interested in slavery face is whether there is even some single institution of slavery that exists across time and space that they can even study.^[15] It’s tempting to assume that, though the details might differ, there is something called slavery out there, popping up throughout history, and that we’d know it if we saw it. But, of course, as our hypothetical trip in the TARDIS shows, what we would recognize as slavery is determined by our own cultural memory of what the English word \ˈslā-v(ə-)rē\ means to us.

When Americans think of slavery we think of Twelve Years A Slave and Roots. The images are seared marks in our mind: African men, women, and children being seized by ruthless slave traders, torn from their homes and each other, packed like chattel into the holds of stifling slave ships, sold like cattle at auction to white plantation owners, who worked, oppressed, and lashed them mercilessly for the rest of their lives. Slavery in our cultural memory is ‘the original sin’ of America: the reduction of a person, against their will, to the status of property, owned by another person who had absolute right over their labor and who deprived them of the natural right to freedom and family.

Yet as we have seen, ownership, freedom, and exploitation come in shades of gray. They exist on spectrums. Historians and sociologists have attempted to delineate categories on this spectrum, in part to determine if we can really talk about slavery as something separate from other forms of forced labor or involuntary servitude. The main categories on this ‘continuum of dependency’ other than slavery are:^[16]

Serfdom: In Europe, this tradition goes back to ancient Greece. Laborers, usually peasant farmers, were free in the sense that they owned their own clothes, tools, livestock as well as the fruits of their labor. But they were bound to the land on which they lived or to their landlord wherever he might go.^[17] Serfdom in Europe developed as the status of free peasants and settled Barbarian prisoners of war in the late Roman Empire collapsed into a single class of “quasi-servitude” not too different from slavery.^[18] Serfdom disappeared in most of Western Europe in the wake of the Black Death in the 1300s, though it continued in the institution of villeinage in England until around 1600 and continued into the 1800s in mining areas of Scotland and German-speaking lands. Serfdom is most associated with Russia, where it came to replace slavery in agriculture and domestic spheres in the late 1600s and early 1700s.^[19]

Master/Servant Relationship: When serfdom disappeared from Western Europe, it was replaced by the relationship between the laborer and the landowner/employer. Unlike our modern notion of a worker’s contract, however, failing to live up to this contract was a criminal offense. Only in the British colonies in North America did a notion of free labor eventually appear in the 1700s, and this did not make its way back to Britain until 1875.^[20]

Debt servitude: This has been one of the most widespread forms of coerced labor. When a person is unable to repay a debt, he or she becomes the slave of the creditor. This was extremely common in Southeast Asia, where our Western model of slavery was extremely rare.^[21]

Bonded labor/indentured servitude: This is similar to debt servitude and has been very common in history. A person willingly enters into an agreement to exchange their labor and a loss of some freedoms for a fixed period of time in return for some service or up-front payment. This differs from debt servitude because the person willingly surrenders their labor and a degree of freedom.

These categories are not fixed or hermetically sealed. They bleed into each other, making it very hard to come up with a clear line distinguishing slavery from other forms of coerced labor. Scottish mining serfs often wore collars with the names of their masters on them, for example, something we’d probably associate more with slavery.^[22] Indentured servants from Britain, who made up two-thirds of the immigrants to British North America before 1776, could be sold, worked to exhaustion and beaten for misbehavior. They could not marry and, in Virginia at least, could be mutilated if they tried to escape. In Maryland the punishment was death.^[23]

Slavery in colonial America was worse, but only in that it was permanent. On the other hand, as early as the 1400s in the Ottoman Empire people captured in war were sometimes settled to work lands owned by the sultan. Although technically slaves, their condition was closer to serfdom. These slaves formed families that lasted generations and passed down the land they worked to their children. Only if a head of household died without any children would his estate revert back to the imperial treasury. Later on, as Ottoman cities industrialized, factory owners preferred using slave labor because slaves would not leave for seasonal work elsewhere. By agreeing to mukātaba contracts with these slaves—in which the slaves bought their own freedom by installments—these factory owners were able to maximize the slaves’ productivity.^[24] They were, in effect, more like wage laborers working for a set term in a master/servant relationship than slaves.

We might think of slavery as distinguished from other types of coerced labor by the question of choice. Indentured servants chose to enter into those contracts. Slaves would never choose to become slaves, right? But realities are much more complicated. Outside of slavery in the Americas, ‘voluntary slavery’ was not uncommon at all.^[25] In Ming China many impoverished tenants sold themselves into slavery when they could not pay rent.^[26] In 1724, the Russian czar abolished slavery and converted all of Russia’s slaves into serfs because serfs were offering themselves as slaves to avoid paying taxes; serfs paid taxes, slaves did not.^[27] Earlier, in the fifteenth-century duchy of Muscovy, what scholars term ‘limited service contract slavery’ became common. In such a contract, a person asks someone wealthy for a loan for a year, at which point the person will pay them back and will also work for them in the meantime instead of paying interest. If the borrower cannot pay the creditor back in a year, they become their slave. Most often, they became a lifetime slave. This type of slavery replaced all other forms of slavery in Russia. And yet there was also indentured servitude at the same time, differing from slavery only in that an indentured servant could not be physically harmed by their master.^[28]

Unlike bonded laborers or serfs, we might think of slaves as people with little or no legal right to protection. This has often been true. In Ming China, slaves were often referred to as “not human.” Not only could they not own property, marry or have legitimate children, but killing one of them also posed no legal problem.^[29] Among the Toraja people of Sulawesi (today in Indonesia), someone who had been convicted of a capital crime could have one of his slaves executed instead of himself.^[30] A judge in South Carolina in 1847 declared that a slave “can invoke neither magna carta nor common law”; for the slave the law was whatever the master said.^[31]

Yet not only were legal realities often quite complicated, so were the social realities behind the laws. In Roman law, slaves were conceptualized as people with no rights. Since they were, in theory, prisoners of war who had been spared execution, they were legally dead anyway.^[32] And during the period of the Roman Republic (6th-1st centuries BCE), there was no legal restriction on a master’s treatment of his slaves. But such laws are not very helpful in distinguishing free from slave, however, since Roman heads of household at that time also enjoyed the theoretical ‘power of life and death’ over every man, woman and child in the family.^[33] As the number of slaves in the expanding Roman Empire increased, however, laws were put in place to protect them. Under the Emperor Hadrian (d. 138 CE) excessive punishment was forbidden, as was killing a slave without a legal ruling. The Emperors Antoninus Pius (d. 161 CE) and later Constantine (d. 337 CE) made it clear that if a master killed his slave in cold blood or by excessive punishment he was guilty of homicide. And in the legal code of Emperor Justinian (d. 565 CE) it was clear that the master’s rights to do violence to his slave were limited to reasonable discipline.^[34]

In early America, all thirteen colonies had laws regulating race and slavery, which were occasionally updated. Although ten states in the South had slave codes making it a crime to mistreat slaves, mistreatment was understood in relation to the severity of the disobedience or infringement that the master was punishing. Amputating limbs, castration, and execution were all allowed as punishments when the alleged crime was severe. And it was almost impossible for slaves to challenge any treatment in court, since they could not even testify. Nonetheless in North Carolina and Virginia a handful of white slave-owners were executed or imprisoned for murdering or cruelly treating their slaves.^[35]

As a leading scholar of slavery, David Davis, observed, “The more we learn about slavery, the more difficulty we have defining it.”^[36] A trans-historical definition of slavery has indeed proven very hard to find. As a leading scholar on Ottoman slavery has remarked, it is difficult to treat slavery as one definable phenomenon just in the Ottoman Empire, let alone globally (though he stresses that the varieties of slavery in the Ottoman realm were different in degrees not different in kind).^[37]Nur Sobers-Khan has observed about slavery in Ottoman Istanbul, that it was so diverse that it doesn’t make sense to talk about slavery as a unified phenomenon even in one city, let alone in the whole Mediterranean region.^[38] Scholars don’t even agree on where to start. Many historians, proceeding from a Marxist paradigm, have sought to explain slavery as a purely economic phenomenon. Others, especially scholars of slavery in the Islamic world, have stressed that slavery is often much more of a social phenomenon. Definitions of slavery have tended to revolve around three notions: the slave as a family-less outsider, the slave as property, and the slave as the object of violence.^[39] But for a definition to fit all the things that people today commonly associate with slavery, that definition has to be so vague that it’s almost useless. So slavery is “the forced labor of one group by another,” according to some social scientists.^[40] Others have suggested that the slave is always an outcast.^[41] According to Davis, to apply across human history, slavery can only be defined as extreme social “debasement”; whatever the hierarchy, slaves are always at the bottom.^[42]

Some scholars have proposed more specific definitions for slavery as an economic, legal, and social condition. One argues that slavery is a mode of exploitation that is uniquely characterized by its means of reproducing itself, namely through political violence or captivity in war.^[43]

The most influential, specific definition comes from Orlando Patterson, who defines slavery as always exhibiting three features. First, slavery involves perpetual domination ultimately enforced by violence. Second, slavery involves a state of natal alienation, “the loss of ties of birth in both ascending and descending generations” that preclude making claims of birth or passing them on to one’s children and that cuts the slave off from family and community except as allowed by the masters. They inherit no protection or privilege and can pass none on to their children. Finally, slaves are denied any honor. Slavery is thus defined as the “permanent, violent domination of natally alienated and generally dishonored persons.”^[44]

But Patterson’s definition fails to apply to many instances of what we would otherwise think of as slavery. Sometimes it was the slaves who dominated free people, as in the case of the Turkish slave soldiers of the Abbasid caliphs in the ninth and early tenth centuries. Even before the Ottomans began their system of imperial slaves, Egypt and Syria were ruled by the Mamluk (literally, ‘slave’) state (c. 1260-1517). Although they were freed after they finished their military training, the Mamluk dynasty of Turkic or Caucasian warlords reproduced itself generation after generation by importing new slave soldiers into a ruling military elite that defined itself by its military slave experience.^[45] Far from being dominated by anyone, they were their own masters and dominated the whole of the state and society. Patterson argues that slave elites in Islamic civilization were still effectively powerless because their fate still hung on the whim of their masters. But the frequency with which Abbasid Turkic slaves, Egyptian Mamluks, and Ottoman Janissaries summarily executed their masters when it suited them strongly suggests otherwise.

Nor have those who identify as slaves always been natally alienated. Byzantine imperial slaves could own property and bequeath it to their children.^[46] The Ottoman agricultural slaves settled on imperial lands passed their estates on to their children for generations. Unlike Roman slavery, where the status of a child’s mother determined its status, the main position in the Shariah was that a slave woman who gave birth to her master’s child became free when her master died, as did her child. Until then he could not sell her. Far from being natally alienated from her child, its status as the child of a freeman ensured the mother’s own freedom. Elite imperial slaves like Sokollu Mehmet Pasha were technically natally alienated in the sense that, according to the letter of the Shariah in Ottoman lands, their wealth reverted back to the treasury (bayt al-mal) upon death. But in reality, when an elite imperial slave like Sokollu died, what transpired was a form of negotiation between state officials and the heirs. Since many of these slaves had amassed—and stashed—immense wealth, it was more efficient for the state to negotiate for a portion of it in return for allowing the heirs to receive the remainder without legal problems.^[47] Here the slave’s natal alienation functioned more as an irregular estate tax than a total deprivation of their right to pass on their property to their heirs. There were also other easy means of circumventing the natal alienation of wealth. Like many wealthy citizens of the Ottoman Empire, imperial slaves could place their wealth in endowments (Ar. waqf, pl. awqaaf) and make their descendants the beneficiaries.^[48]

Furthermore, the children of Ottoman imperial slaves could not pass their wealth on to their children (it reverted back to the imperial treasury upon their death), but their children retained the privileges of their fathers’ proximity to power as well as the status of their mothers. Sokollu Mehmet’s wife was the daughter of the sultan, so his sons attained high office. What is more striking is that, in many cases, Ottoman imperial slaves maintained their relationships to their original families in the Christian areas of the Balkans, using their newfound power to elevate their relatives.^[49] Sokollu Mehmet appointed his brother as Orthodox Patriarch in the Balkans, and his cousin later followed him to the office of grand vizier.^[50] Later, in the late eighteenth century, the Georgian slave elite in charge of administering the Ottoman province of Egypt maintained close relations to their families back in the Caucasus and even received visits from them.^[51]

Sometimes exploiting family connections was one of the major purposes of enslavement. Though technically slaves, Christian Europeans captured by the Ottoman naval forces of Algiers in the eighteenth century were often more like hostages. They could send and receive mail from their families and, if their masters were lucky, their families paid ransoms to free them. In the meantime, they could own property, make money (those assigned to elite jobs like cofeegi, coffee pourer, might live better than in their home country), and mix freely.^[52]

Like the word terrorism, slavery is also a deeply, deeply political issue, not in the sense of politics as what we see on the nightly news, but rather in the sense that it is inherently tied to questions of power. Just as the practice of slavery is an extreme exercise of power by some human beings over others, wielding the language of slavery is a claim to moral authority over others. It is no surprise that advocates of ending brutal or unacceptably exploitative labor practices such as sweatshops, child sex trafficking, forced marriage, and organ trading refer to such phenomena as ‘modern day slavery.’ The reason for invoking the word ‘slavery’ instead of other definitions such as bonded labor or child labor is clear: slavery provokes an emotional reaction that spurs people into action and support for a cause. From students to rock stars, who wouldn’t support ending slavery?

Though such practices are indeed reprehensible, with ‘modern day’ slavery we run across some familiar problems. If we took the definitions of slavery used by activists fighting ‘modern slavery’ (the main one is it’s slavery ‘if you can’t walk away’) and applied them to just Western history we’d find that almost no one was free by their standards.^[53] As some scholars have observed, the most prominent advocates for ending modern day slavery have not applied the label to the forced labor of criminals in the American penal system.^[54] This is no doubt a very political choice since fewer rock stars and students would be as willing to accuse the US government of engaging in ongoing slavery. So even when invoked for noble causes today, ‘slavery’ is still a deeply political word, both in the emotional reaction it triggers and in the self-censorship that people use in when and where they apply it.

The political nature of slavery is particularly pronounced in the history of Islam and the West. During the eighteenth and even nineteenth centuries the fear of being captured by Muslim pirates in the Atlantic and western Mediterranean loomed large in the Western European (particularly British) imagination. And indeed thousands of British and Americans were taken as slaves in such a way. We still see the cultural imprint of this fear in movies like Never Say Never Again (1983), where James Bond rescues Kim Basinger from a remarkably out of place Arab slave auction, and Taken (2008), where Liam Neeson finally rescues his daughter from first (Muslim) Albanian traffickers and finally from a lascivious Arab sheik. But, like the selective use of the term ‘modern day slavery,’ this conversation is selective in its claim to Western moral authority. During the same era that Europeans and Americans were decrying capture and enslavement by Muslim pirates, the enslavement by Europeans of Muslims from the Ottoman Empire was booming.^[55] And our Western cultural memories are even more selective.  Western theater-goers likely felt no outrage in The Spy Who Loved Me (1977) when Bond visits the harem of his Arab sheik friend and is offered one of the women (when in the Orient, says the sheik, “one should delve deeply into its treasures”). From the British tabloids to then private citizen Donald Trump, in 2015 many parroted the claim that Muslims in northern England were luring young white girls in as sex slaves. Some Muslims were doing this, but few media reports stated that the majority of offenders were actually white men.^[56]

The word slavery has been political even when it has been invoked for the best of causes. And the political forces that have shaped how slavery is understood have often hobbled the best efforts of those fighting against the extreme exploitation of fellow human beings. Abolitionists in the nineteenth century chose to define slavery as treating human beings as property in part because, if they defined slavery as harsh deprivation or exploitation, their pro-slavery opponents would just point to the factory of conditions of industrial England and America and note that ‘free’ workers were being treated just as badly.^[57]

Having emphasized that slavery consisted of humans being treated as property, abolitionists were left with no objection to continued exploitation of the same people they had just freed once it became technically illegal to own people. British abolitionists succeeded in ending slavery in the Indian Ocean in the 1830s. But then they found that laborers were still being transported to East Africa from India in the same horrid conditions as slaves and with the same high mortality rate. They were just called ‘coolies’ rather than slaves.^[58] Today, decades after the legal right to own other human beings was abolished globally, activists referred to as new abolitionists, seeking to mobilize public concern over exploitative labor, have redefined slavery as ‘not being able to walk away.’^[59]

Ultimately, the word ‘slavery’ can mean so many things that it’s not very useful for accurate communication. It often ends up referring to things we don’t mean when we think of slavery, or it fails to match things we do associate with slavery. As such, the word slavery has limited use as a category or conceptual tool. It’s much more useful to talk about the extreme exploitation of human beings’ labor and the extreme deprivation of their rights. In any society, whether it has ‘slavery’ or not, we are likely to find such conditions. Instead of fixating on a word or ill-defined category, it is much more useful to focus on regulating conditions and protecting people’s rights in order to prevent extreme debasement. And, as our next essay will show, this is precisely what the Shariah aimed to do.

[1] The Ottoman tradition of elite slavery may have been inherited from the late Roman and Byzantine Empires, where imperial slaves (often eunuchs) could rise to high positions in the military and the administration; Youval Rotman, Byzantine Slavery and the Mediterranean World, trans. Jane Marie Todd (Cambridge, MA: Harvard University Press, 2009), 104; Cam Grey, “Slavery in the Late Roman World,” in The Cambridge World History of Slavery: Volume I The Ancient Mediterranean World, ed. Keith Bradley and Paul Cartledge (Cambridge: Cambridge University Press, 2011), 499.

[2] Pamela Kyle Crossley, “Slavery in Early Modern China,” in The Cambridge World History of Slavery: Volume 3 AD 1420-1804, ed. David Eltis and Stanley Engerman (Cambridge: Cambridge University Press, 2011), 200.

[3] Christoph K. Neumann, “Whom did Ahmet Cevdet represent?,” in Late Ottoman Society, ed. Elisabeth Özdalga, 117-134.  London: Routledge, 2005), 117.

[4] David Brion Davis, Challenging the Boundaries of Slavery (Cambridge, MA: Harvard University Press, 2003), 17-18.

[5] Orlando Patterson, Slavery and Social Death (Cambridge, MA: Harvard University Press, 1982), 22.

[6] Grey, “Slavery in the Late Roman World,” 496; Rotman, Byzantine Slavery, 174-76.

[7] Patterson, Slavery and Social Death, 22.

[8] Crossley, “Slavery in Early Modern China,” 191.

[9] Julia O’Connell Davidson, Modern Slavery: The Margins of Freedom (New York: Palgrave Macmillan, 2015), 162.

[10] Crossley, “Slavery in Early Modern China,” 187.

[11] Vaughan Lowe, International Law: A Very Short Introduction (Oxford: Oxford University Press, 2015), 1.

[12] Here quoting Youval Rotman, Byzantine Slavery, 19.

[13] Rotman, Byzantine Slavery, 17-18.

[14] Rotman, Byzantine Slavery, 97-98.

[15] Joseph C. Miller, The Problem of Slavery as History (New Haven: Yale University Press, 2012), 12.

[16] David Eltis and Stanley Engerman, “Dependence, Servility, and Coerced Labor in Time and Space,” in The Cambridge World History of Slavery Volume 3, 3.

[17] Richard Hellie, “Russian Slavery and Serfdom, 1450-1804,” in The Cambridge World History of Slavery Vol. 3, 276-77.

[18] Cam Grey, “Slavery in the Late Roman World,” 484-6.

[19] Hellie, “Russian Slavery,” 284, 292-93.

[20] Eltis and Engerman, “Dependence, Servility, and Coerced Labor,” 7; Davidson, Modern Slavery, 68.  In England this issue was governed by the Statute of Artificers, which the American colonies only adopted in a limited way.

[21] Kerry Ward, “Slavery in Southeast Asia, 1420-1804,” in The Cambridge World History of Slavery Volume 3, 165-66.

[22] Eltis and Engerman, “Dependence, Servility, and Coerced Labor,” 6.

[23] Kenneth Morgan, Slavery and Servitude in Colonial North America (New York: New York University Press, 2000), 8-9, 20; David Galenson, “The Rise and Fall of Indentured Servitude in the Americas: An Economic Analysis,” Journal of Economic History 44, no. 1 (1984): 4.

[24] It was in the Ottoman state’s interest to keep this agricultural system stable; Y. Hakan Erdem, Slavery in the Ottoman Empire and its Demise, 1800-1909 (New York: St. Martin’s Press, 1996), 12-13, 15.

[25] Stanley Engerman, “Slavery at Different Times and Places,” American Historical Review 105, n. 2 (2000): 481

[26] Crossley, “Slavery in Early Modern China,” 189.

[27] Hellie, “Russian Slavery,” 284, 293.

[28] Hellie, “Russian Slavery,” 279-80.  The author notes the similarity between this Russian contract and the ancient Persian custom of antichrisis (as named by Greek authors).

[29] Crossley, “Slavery in Early Modern China,” 191.

[30] Ward, “Slavery in Southeast Asia,” 171.

[31] Lawrence M. Friedman, A History of American Law, 2nd ed. (New York: Simon & Schuster, 1985), 225.

[32] W.W. Buckland, The Roman Law of Slavery (New York: AMS, 1969, reprint of 1908 Cambridge U. Press edition), 2-3.

[33] Yan Thomas, “Vitae Necisque Potestas: Le Père, La Cité, La Mort,” Publications de l’École Française de Rome (1984): 499–548.

[34] Buckland, The Roman Law of Slavery, 36-8.

[35] Kenneth Morgan, Slavery and Servitude in Colonial North America, 35, 77; Ira Berlin, Many Thousands Gone: The First Two Centuries of Slavery in North America (Cambridge, MA: Belknap Press, 1998), 116; Paul Finkelman, “Slavery: United States Law,” in Oxford International Encyclopedia of Legal History, 5:258-262; Friedman, A History of American Law, 225-6.

[36] David Brion Davis, Slavery and Human Progress (Oxford: Oxford University Press, 1984), 8.

[37] Ehud Toledano, Slavery and Abolition in the Ottoman Middle East (Seattle: University of Washington Press, 1998), 164-65; Toledano, As if Silent and Absent: Bonds of Enslavement in the Islamic Middle East (New Haven: Yale University Press, 2007), 21.

[38] See also Nur Sobers-Khan, Slaves without Shackles: Forced Labour and Manumission in the Galata Court Registers, 1560-1572 (Berlin: Klaus Schwarz Verlag, 2014).

[39] Martin Klein, “Introduction,” in Breaking the Chains: Slavery, Bondage, and Emancipation in Modern Africa and Asia, ed. Martin Klein (Madison: University of Wisconsin Press, 1993), 4-5.

[40] Rodney Coates, “Slavery” in Blackwell Encyclopedia of Sociology, ed. George Ritzer (Oxford: Blackwell, 2007).

[41] A. Testart, “The Extent and Significance of Debt Slavery,” Revue Française de Sociologie 43 (2002): 176.

[42] Davis, Slavery and Human Progress, 17-19; Brenda Stevenson, What is Slavery? (Malden, MA: Polity, 2015), 8.

[43] Claude Meillassoux, The Anthropology of Slavery (London: Athlone, 1991).

[44] Orlando Patterson, Slavery and Social Death (Cambridge, MA: Harvard University Press, 1982), 7-8, 13.

[45] Nasser Rabbat, “The Changing Concept of the Mamlūk in the Mamluk Sultanate in Egypt and Syria,” in Slave Elites in the Middle East and Africa, ed. Miura Toru and John Edward Philips (London: Kegal Paul, 2000), 89, 97.

[46] Rotman, Byzantine Slavery, 104.

[47] See Ali Yaycıoğlu, “Wealth, Power and Death: Capital Accumulation and Imperial Seizures in the Ottoman Empire (1453-1839)” available at http://www.econ.yale.edu/~egcenter/Yaycioglu%20-%20Wealth%20Death%20and%20Power%20-%20November%202012.pdf.

[48] Leslie Pierce, Morality Tales: Law and Gender in the Ottoman Court of Aintab (Berkeley: University of California Press, 2003), 315; Toledano, As if Silent and Absent, 25; Ebru Boyar and Kate Fleet, A Social History of Ottoman Istanbul (Cambridge: Cambridge University Press, 2010), 147-48.

[49] Dror Ze’evi, “My Slave, My Son, My Lord: Slavery, Family and the State in the Islamic Middle East,” in Slave Elites in the Middle East and Africa, 75.  See also Metin Kunt’s short article, “Ethnic-Regional (Cins) Solidarity in the Seventeenth-Century Ottoman Establishment,” International Journal of Middle East Studies 5, no. 3 (1974): 233-39.

[50] Veinstein, G., “Soḳollu Meḥmed Pas̲h̲a”, in: Encyclopaedia of Islam, Second Edition, Edited by: P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, W.P. Heinrichs. Consulted online on 21 November 2016 <http://dx.doi.org.proxy.library.georgetown.edu/10.1163/1573-3912_islam_SIM_7090> First published online: 2012

[51] Daniel Crecelius and Gotcha Djaparidze, “Relations of the Georgian Mamluks of Egypt with Their Homeland in the Last Decades of the Eighteenth Century,” Journal of the Social and Economic History of the Orient 45, no. 3 (2002): 326.

[52] Christine E. Sears, “‘In Algiers, the City of Bondage’: Urban Slavery in Comparative Context,” in New Directions in Slavery Studies, ed. Jeff Forret and Christine E. Sears (Baton Rouge: Louisiana State University Press, 2015), 203, 207, 211.

[53] Julia O’Connell Davidson, Modern Slavery: The Margins of Freedom (New York: Palgrave Macmillan, 2015), 3, 6, 22-23, 37-39, 69, 169.

[54] Davidson, Modern Slavery, 100

[55] William Clarence-Smith and David Eltis, “White Servitude,” 139, 144.

[56] thestar.co.uk/news/majority-of-rotherham-child-exploitation-suspects-are-white-claims-new-report-1-739263

[57] Davidson, Modern Slavery, 33.

[58] Davidson, Modern Slavery, 32.

[59] Kevin Bales, Understanding Global Slavery (Berkeley: University of California Press, 2005), 52-54.