Introduction
Slavery in the Ottoman Empire is an incredibly complex subject that both heavily overlaps and diverges from our notions of slavery in the Americas. While the rhetoric of this historical inquiry often argues that slavery in the Ottoman Empire was more “mild” than the Americas, this line of reasoning does little justice to the countless people who were brutally enslaved, sexual abused, and forced to work. More recent inquiry, such as that by Toledano, have challenged this idea that slavery in the Middle East during this time period was less problematic than the extensive Atlantic slave trade.
Enslaving: Warring, Raiding, and Trading
Starting in the early 14th and 15th centuries with the rapid expansion of the Ottomans into the Balkans and Greece, thousands of Orthodox Christian peoples were enslaved. The nature of Ottoman warring and raiding in the Balkans and Greece enriched the treasury through trading slaves with Christian states, including: Genoa, Venice, the Knights of Malta, Cyprus, and more.

We have archival evidence of this early Ottoman slave trade. Fleet details the prices of slaves sold in these markets spotted throughout the Balkans, Greece, and Anatolia. She posits that by 1438, the slave market had become saturated from a massive, continuous influx of slaves brought to the market by the Ottomans (European and Islamic Trade in the Early Ottoman State : The Merchants of Genoa and Turkey. (Fleet) See page 46).
Of course, not all slaves were sold to foreign Christians. Many slaves found their way forcefully led to the interior of the Ottoman Empire. For example, according to the archival research of Canbakal and Filiztekin, between 1460 and 1480, 18.4% of free people owned slaves in Bursa (Slavery and Decline of Slave-Ownership in Ottoman Bursa 1460–1880 (Canbakal and Filiztekin) See page 63, table 2A).
This would have accounted for around 180 slaves split between about 90 slave owners. Almost all of these slaves would have been urban or domestic workers.

As for ownership, from the same source, we see that slaves were predominantly owned by the richest of society. However, there was a strong middle-class presence in slave-ownership until the 1700s.

The way in which the Ottoman Empire obtained slaves would change with the stagnation of Ottoman expansion into Europe during the 17th century. Here is where we begin to see a larger emphasis on trading for slaves. Much of this would have been conducted through East Africa, the Caucuses, and around the Black Sea.

One such source is looked at by Wilkins in his analysis of slaves in Aleppo between 1640 and 1700. Wilkins shows that the slave population composed of Georgians and Russians, who both formed about 1/3 of the total population. The remaining third was formed from a variety of East African peoples, Hungarians, Poles, and more (Slavery and Household Formation in Ottoman Aleppo, 1640-1700 (Wilkins) See page 354, table 1).
Raids into the fractured and contentious middle-ground between the Ottoman and Safavid Empires would have resulted in a number of Georgian slaves. Likewise, Russians, Hungarians, and Poles were frequently kidnapped through widespread raiding situated around the Black Sea by the Crimean Tatars. These slaves would then be traded into the Ottoman Empire. East African slaves would have been worked through the already established slave trade in the region to Ottoman Cairo and beyond.

Meanwhile, back in Bursa, the African slave population had only begun to see a significant increase by the 1600s. Before this, slaves in Bursa were predominately Europeans centered around the Black Sea. However, numerical accounts are limited until the early 1700s on African slaves in this region. And unfortunately, even though we know that African slaves would have been in Bursa from 1500 and increased by the 1700s, we have few records that denote ethnic origins of slaves in this region.

Regarding the relative cost of slaves, Caucasian – especially Circassian – white women were often the most highly sought after. The price for white slaves, in general, was two to three times higher than their African counterparts. White, female slaves would have been 10%-20% more expensive than white men. Other white peoples and Ethiopians were also sought after. The cheapest slaves would have been other African peoples.

From this discussion, we can see that slaves came from a wide variety of ethnic backgrounds and were enslaved in many different regions of Europe, Asia, and Africa. Ottoman sources for slavery changed over the history of the empire. Originally, slaves were primarily gathered through warring and raiding. However, by the 1600s, this had primarily changed to trading and, to a lesser extent, raiding.
Concubines, Soldiers, and Laborers: The Many Roles of Ottoman Slaves
Slaves had a myriad of roles within Ottoman society. This is actually where it gets rather complicated and diverges significantly from the Atlantic slave trade.

In Bursa, most slaves would be in the urban centers performing a variety of domestic and commercial jobs. Canbakal and Filiztekin note several occupations that the majority of slaves would be involved in: general shops; textile shops; tool production, including agricultural and textile tools; land; livestock; commercial textile stock; commercial agricultural stock. By far, most slaves would be involved in the textile industry (Slavery and Decline of Slave-Ownership in Ottoman Bursa 1460–1880 (Canbakal and Filiztekin) See page 67, table 5).
Akin to Bursa, in Aleppo, most slaves would be involved in urban occupations. Primarily, slaves found themselves under military-administrative officials, who both participated in managing the Ottoman army and participated in the civil bureaucracy. Under these officials, some slaves would have worked in civil offices, while others would have augmented the army. A sizeable number of slaves would have also found themselves under wealthy merchants as commercial agents for a household. The third largest group would have been domestic servants under Muslim women (Slavery and Household Formation in Ottoman Aleppo, 1640-1700 (Wilkins) See page 357, table 2).
Besides private ownership, there was also a variety of state slaves, who were bonded to the Ottoman government. These would have been the kul, military administrative slaves; and the janissaries, slave soldiers. In both cases, these would have been young, Christian boys, often taken from the Balkans, who were then raised in Ottoman society. They would have learned Turkish, converted to Islam, and worked for the Ottoman government as they grew older. Here, many kul and janissaries actually became rather economically successful.

In the popular imagination of Ottoman slavery somehow being more “mild” than Atlantic slavery, the harem slaves are often pointed to as evidence of this “mildness.” However, this neglects to mention the traumatizing enslavement of the woman, the societal slave status of the woman, and the sexual abuse or sexual slavery that the woman must endure.

Nonetheless, concubines of the master could sometimes achieve some societal status. If they were to bear a son for the master, the woman would have the umm al-walad (“mother of the child”) status. Upon the realization of the pregnancy, the woman could no longer be sold and the child would become a free Muslim upon birth. In some cases, the concubine was manumitted by the master upon learning of the pregnancy and the now-freed slave and former master would wed. This allowed for some societal advancement for female slaves.

While most slaves would have been urban, domestic workers, this is not to imply that agricultural slaves did not exist. We know that by 1620 in Bursa, a majority of slaves would be in rural areas. Domestic work could still be done here, but there is little doubt that slaves would have participated in agriculture. Additionally, we have evidence of slaves growing cotton in Ottoman Egypt to meet European demands during the American Civil War ( As If Silent and Absent : Bonds of Enslavement in the Islamic Middle East (Toledano) See page 13).
From this discussion, we can see that slaves had a multitude of different roles within Ottoman society and that these roles often had an imbalance of power and agency when compared to one another. Many historians currently draw a line between kul/harem slaves and the rest of the slave strata during the Ottoman Empire, because of the severe imbalance in supposed quality of life and rights.
Ottoman Slaves: Treatment in Society
Toledano proposes the follow criteria for roughly determining the treatment of slaves within the Ottoman Empire:
“the tasks the enslaved performed — whether domestic, agricultural, menial, or kul/harem • the stratum of the slaver — whether a member of an urban elite, a rural notable, a smallhold cultivator, an artisan, or a merchant • location — whether in the core or a peripheral area • habitat — whether urban, village, or nomad • gender — whether male, female, or eunuch • ethnicity — whether African or not” (As If Silent and Absent : Bonds of Enslavement in the Islamic Middle East (Toledano) See page 14)
This does a pretty good job of showing the complex slave-master relationship and the varying levels of treatment based on a slew of different factors. From this criteria, he concludes:
Enslaved domestic workers in urban elite households were better treated than enslaved workers in other settings and predicaments. • The lower the stratum of the slaver and the farther from the core and the less densely populated the habitat, the greater the chances the enslaved had to receive bad treatment. • The lives of enslaved Africans and enslaved women were more often than not harder than the lives of enslaved whites and enslaved men (As If Silent and Absent : Bonds of Enslavement in the Islamic Middle East (Toledano) See page 15)
Periodically, slaves were manumitted by their masters. This worked in one of three ways. The first was an immediate freeing of the slave whenever the slave owner felt was appropriate. The second was the freeing of a slave upon the master’s death. The third was the freeing of a slave upon the completion of a set amount of work or number of years worked.

In Aleppo, for example, 259 manumissions are recorded in court documents between 1640 and 1670. 96% of these would have been the first type, the immediate release of the slave. We should be careful when looking at these numbers though. These numbers only look at manumissions that occurred in the courts. The freeing of a slave was generally not done through the courts. But, what we should take away from this is that 96% of slaves were immediately release. Thus, marking the other two forms as rather rare in Aleppo.