Readers of the English poet Lord Byron will no doubt remember his treatment of Ottoman-period slavery in oriental tales such as The Giaour and The Corsair, as well as his hero’s being sold into slavery in the poem Don Juan.
While Byron was a great lover of modern Greece, his travels through the Eastern Mediterranean in the early decades of the 19th century took him across the then Ottoman world. Fascinated by his experiences in Turkey, the Balkans, and the borders of the Levant, Byron worked them up into poems featuring many classically orientalist themes, including piracy, slave markets, and women’s confinement in harems.
In his best work, Don Juan and Childe Harold’s Pilgrimage, Byron adds satire and his enormous facility with words to create two of the most important English poems of the first half of the 19th century. While his “Turkish tales” do not figure in a current exhibition on Mediterranean slavery, Esclaves en Méditerranée, XVIIe-XVIIIe siècles, at the Institut du Monde arabe in Paris, they could certainly be seen as relevant to its theme.
The slavery that was a feature of most Mediterranean societies in the early modern period, roughly speaking from the 16th to the end of the 18th centuries, was a consequence of the division of the region into European states in the north and Ottoman-controlled ones in the south that sometimes came to blows over territory or influence.
While Ottoman attempts at expansion in eastern Europe ended with the siege of Vienna in 1683, and European colonial control of the countries of the southern Mediterranean did not begin in earnest until the 19th century, a kind of rivalry reigned throughout the 17th and 18th centuries, with this giving rise to state-sanctioned privateering on both sides of the Mediterranean.
Ports such as Genoa and Livorno in Italy and Algiers in Algeria became well known for their role in processing the booty from pirate expeditions and coastal raids, with human captives being brought back for sale in slave markets in the Italian ports or Ottoman cities such as Tunis or Algiers.
These expeditions were not part of organised campaigns. They were nothing like on the scale of the activities of the European companies that built trading stations on the West African coast and shipped enslaved Africans across the Atlantic to sugar and cotton plantations in North and South America in the same period.
Instead, the Mediterranean trade, though tolerated or even encouraged by the states concerned, was opportunistic and small in scale, probably numbering a couple of million people during the centuries of its operation as opposed to the more than ten million Africans sold into slavery during the transatlantic trade.
Even so, the exhibition gives some possibly surprising figures, saying that the slave population of Livorno and Valetta (Malta) in the 17th-18th centuries stood at around 20 per cent of the whole, though there is no information on how these figures have been arrived at or whether these were household slaves, prisoners of war, galley slaves, temporary captives, or members of other categories.
Mediterranean slavery co-existed with free labour, mostly as an economically unimportant addition and not the foundation of the wider system. As a result, it was less organised and, unlike Atlantic slavery, was not part of a racialised and hereditary hierarchy. There was a religious component – Christians would often enslave Muslims, and Muslims would enslave Christians, but neither would generally enslave members of their own communities. Slavery could strike almost anyone if they were in the wrong place at the wrong time.
In his famous novel Robinson Crusoe, for example, published in 1719, UK author Daniel Defoe describes Crusoe selling a young companion into slavery to a Portuguese sea captain encountered off the African coast. In Don Quixote by Cervantes, one of the most celebrated of all early European novels, Don Quixote evokes the experience of Christian captives among the Moors, echoing Cervantes’s own experience of being held as a slave in Algiers between 1575 and 1580.
What these experiences had in common were elements of bad luck but also the recognition that Mediterranean slavery was not a permanent condition. In this respect it was a bit like Roman slavery, also not the basis of the wider economy, in that if one could be sold into slavery, one could also be bought out of it. The payment of a ransom was an established way of regaining freedom and a passage home.
According to M’hamed Oualdi of the European University Institute in Florence, one of the curators of the exhibition, the European or North African sovereigns of the time profited by the activities of the corsairs, or pirate-slavers, even if they did not explicitly organise them. “They could benefit from between five and 20 per cent of the value of the booty (the quint [fifth] or khums in Arabic),” he says, “and then there was the value of the ransom that could be collected before they were returned home.”
The exhibition includes letters sent by Muslim captives in Naples pleading with the authorities in Tripoli to arrange their return home, presumably helped by the payment of a ransom. Another letter, dated 1735, contains the pleas of 48 Moroccan captives in the French port city of Marseilles to the Moroccan sultan Moulay Abdallah asking for their exchange for Christian captives in Morocco.
Slavery in the Mediterranean, never the basis of the wider economy unlike the plantation slavery in the southern United States, eventually disappeared. There was no lengthy campaign against it, as there was against the transatlantic trade. Once the European states became serious about ridding the Mediterranean of piracy, as they did in the wake of Napoleon Bonaparte’s military campaigns, the slave-trading associated with entrepots like Genoa, Livorno, Malta, and Algiers also began to disappear.
Napoleon’s troops cleared out the religious orders that had previously controlled Mediterranean islands like Malta, the United States and the European powers took action against the corsairs, and pressure was brought to bear on the Ottoman authorities to suppress slave-raiding.
In captivity: UK writer Justin Marozzi’s book Captives and Companions is also about slavery, though this time mostly in the Middle East and taking in a much greater historical range.
He takes a synoptic view of the forms of slavery that have existed in different societies where Islam was the majority religion. This makes his sampling unusually wide, covering everything from agricultural slavery in 9th-century Iraq to Mameluke slave soldiers in Egypt, slavery in Ottoman harems, and the slave girls of the mediaeval Abbasid caliphs.
Marozzi takes in forms of slavery that played very different roles in the societies concerned. Could 9th-century Iraq be described as a slave society, given the importance played in its economy by agricultural slaves? These were responsible for the most important slave revolt – the Zanj rebellion (869–883 CE) – ever recorded in the mediaeval Islamic world.
But this was a very different kind of slavery than that of Mameluke Egypt, where boys would be purchased at a young age in Central Asian markets and then trained as members of a ruling military caste. These slaves played an economically negligible role but a politically major one, since they made up the caste that ruled the country.
Slave girls at the courts of the Abbasid caliphs or slavery in Ottoman harems are forms of slavery that were of only marginal importance to the societies in which they existed. They are clearly very different from the chattel slavery that existed in the Americas, slavery in 9th-century Iraq, probably a bit like Roman slavery in which free and unfree labour existed side by side, slavery in Mameluke Egypt, where paradoxically it was the rulers who started out as slaves, or slavery in the early modern Mediterranean, mostly a form of piracy that faded away when the political conditions facilitating it disappeared.
Marozzi throws a lot of distinct phenomena together and calls them all slavery without too much regard for the particularities of the societies or systems concerned. Slavery is a rather imprecise concept, covering a wide range of cases. But his book does have the virtue of reminding readers of the existence of slavery in other societies besides those in North or South America during the transatlantic slave trade.
This is probably the most familiar to his readers aside from memories of Greek and Roman slavery, dramatised in Hollywood films like Spartacus about a slave revolt in the 1st century BCE, or from reading classical authors at school. Needless to say, this kind of slavery was quite unlike the system of plantation slavery on racial lines that existed in the United States and across the Americas until the late 19th century.
One learns from Marozzi’s book about the Zanj slave revolt in 9th-century Iraq, one of the most intriguing episodes in the rocky history of this never very stable regime. Large-scale agricultural slavery was not an established part of Arab or Muslim society, and probably the development of large projects run with slave labour was related to a shortage of free labour in the country at the time.
The Zanj, mostly Africans imported from East Africa as part of the developing coastal trade, were set to work draining the marshes of what is now southern Iraq and reclaiming the land for agriculture. It was backbreaking work, and in 869 they revolted against their Abbasid overlords, eventually controlling much of the south of the country, sacking the port city of Basra and massacring its inhabitants, and even threatening the caliphal capital of Baghdad.
“You can sense the entire Arab and Persian world shudder in the rage-filled prose… [of Arab chroniclers] Masudi and Tabari when they recounted the Zanj revolt,” Marozzi says. “It is impossible to know the final death toll but… estimates range from the 500,000 suggested by Masudi and the widely quoted figure of 1,500,000 from Al-Suli, a tenth-century Turkic scholar, to the less plausible 2.5 million claimed by the 13th-century historian Ibn Al-Taqtaqi.”
It was also the Abbasids that greatly expanded the use of imported slave soldiers to provide military manpower, regretting it like the later Ayyubid Dynasty in Egypt when those same slave soldiers took power themselves.
In his chapter on the Mameluke slave soldiers – the Arabic word means “owned” – Marozzi says that the idea of slaves becoming the ruling caste of a country “is one of the most remarkable episodes of slavery anywhere in the world.” There were many advantages to the system, at least at first, since the Mamelukes were professional soldiers who would sell their services to the highest bidder and had no conflicting loyalties.
However, perhaps rather like the Roman legions that flattered the ambitions of their commanders and eventually destroyed the system they were supposed to uphold, the Mamelukes eventually turned on their civilian masters. Marozzi writes that the success of these “slaves on horseback” represents “one of the most gravity-defying instances of social mobility ever witnessed. In this genre-bending model of slavery… their official slave status neither prevented them from owning their own slaves or from amassing fabulous riches and political power.”
The wealth and power of the Mameluke regime that ruled Egypt and Syria between the mid-13th century and its defeat by the Ottomans in 1517 can still be appreciated today in the monuments it has left behind in Cairo and other cities. Marozzi, an admirer, thinks of it as a kind of ultimate meritocracy, even if one that only operated for a restricted caste.
“Whereas any stupid, idle, and dissolute young man could and often did inherit the caliph’s throne from his father, this was effectively inconceivable under the Mameluke sultanate,” he writes. “From 1250 any man who aspired to be sultan of Egypt had to plot, manoeuvre, or fight his way to the top, demonstrating outstanding leadership [in a] system that was expressively designed to raise to power the most exceptional individuals” from an elite caste of slaves.
Esclaves en Méditerranée, XVIIe-XVIIIe siècles, Institut du Monde arabe, Paris, until 19 July; Captives and Companions: A History of Slavery and the Slave Trade in the Islamic World, Allen Lane: London, 2025, pp523
* A version of this article appears in print in the 30 July, 2026 edition of Al-Ahram Weekly
Short link: