Urban East African Slavery

Introduction

Histories of urbanization and histories of slavery are closely intertwined.¹ In nineteenth-century East Africa, many towns were born out of or thrived from wealth derived from slavery. Moreover, slave labor was primarily responsible for the building of these towns. Enslaved persons quarried the stone, carried materials long distances, and constructed the buildings, including some of the more delicate architectural detail. Slaves made the towns, and significant portions of town populations consisted of slaves. Nevertheless, identifying and understanding urban slavery in East Africa remains a challenge for a contemporary scholarship, as do identify what towns of East Africa in the nineteenth century could be described as “urban” and establishing criteria to identify “urban” spaces.

The stone towns of the Swahili Coast—with access to trans-oceanic trade—hold the most visible archeological and architectural traces of East African urbanism. The connection between stone and urbanism is not arbitrary; there was a clear correlation between the ability to build in stone and coastal power, influence, and wealth, which was in turn largely derived from the ability to trade and own slaves. Yet, even the great stone towns like Lamu and Zanzibar town had significant and growing areas filled with buildings made of natural and degradable materials such as thatch, timber, and earth‬. These were the dwellings of slaves, ex-slaves, and their descendants who tended to serve the wealthier individuals living in stone houses.

The main stone towns include those of the Zanzibar and Lamu archipelagos, as well as Mombasa. There were also the smaller towns of Malindi, Pangani, Kilwa Kivingje, and Bagamoyo. All these stone towns, whose wealth was derived largely from slavery, were governed by Omani, Zanzibari, and British rule in the nineteenth century. There were also many towns that emerged from the nineteenth-century trade routes, such as Tabora, Ujiji, Kirando, Kasongo, Riba-Riba, and Mtaka. In Ujiji in the 1880s, for example, the majority of the population were enslaved persons and did not only work on plantations but also in the homes of wealthy merchants as domestic staff or as part of harems.² Although many of the people captured in the slave trade were incorporated into these relatively urban hinterland economies and societies between the Great Lakes and the coast, little is known about what it was like to be enslaved in these towns.

The social history of East Africa’s towns provides perhaps the most important insight as it suggests the existence of an urban consciousness closely but complexly linked to slavery. In Pangani, Jonathan Glassman tells us that there existed an urban “plebian” consciousness potent enough to revolt against and overthrow the Omani hegemony in 1888. Whether these individuals, described at the time as “barbarians” (washenzi) and “young men” were indeed slaves who perhaps even created a community out of their slave identity is less clear. Yet, it is certain that newcomers (many of them slaves) flocked to urban towns, in search not only of wealth but also of the cosmopolitan cultures of town life. Most struggled to share in the wealth of the dominant patrician elite, often being reduced to menial labor or sinking into unpayable debt. Swahili proverbs often imply that slaves could not become part of respectable (urban) society. For example: “free men hold together, assist each other in word and in everything; slaves do not and cannot, because they are dependent on their master and cannot join others.”³

However, newcomers were able to integrate into existing community institutions from which they were by default excluded. When turned away, they established their own institutions, such as prayer groups and Koranic schools. Dance societies, too, were a way for new arrivals to network, find communities, and even challenge the political hegemonies.⁴ Laura Fair and Patricia Romero have observed in Zanzibar and Lamu respectively distinctive fashions, music, and social events suggesting an urban consciousness among slaves and ex-slaves in the post-abolition period.⁵ Whether this urban consciousness was specifically linked to an identity of enslaved persons is difficult to track, largely because these individuals tended to use the towns to remove themselves from slavery, rather than to solidify and prolong their slave identity. Mostly, it seems that people with slave status sought to integrate and camouflage themselves into the hegemonic urban Muslim culture, and so they did not form easily-definable groups of acculturated (ex-)slaves. If there did exist an urban slave consciousness, it would have been a complex one in which people with slave status bound themselves together, while also endeavoring to bring themselves closer to the cultures of the people who had enslaved them.

Even as we come closer to determining criteria for “urban,” distinguishing between slavery in urban areas and slavery in rural areas remains very challenging as the two were always closely related and networked.⁶ For example, the Lamu Archipelago was dependent upon the rural mainland for agricultural crops. These spaces were key meeting points between Swahili traders and non-Swahili groups in which goods were collected from the slightly more distant mainland. Moreover, urban and rural populations were far from static and there was a lot of travel for work purposes for both free and enslaved persons. Enslaved persons who worked in urban settings might live in peri-urban or rural areas. Equally, enslaved persons who worked on their master’s plantations were likely to have masters residing in town. Sometimes they lived with their masters in town and traveled to the plantations for work. Movements between town and country could also be seasonal. For example, many slave owners in the town of Zanzibar would hire out their slaves to help with the intensive harvesting periods of cloves. “Urban slavery” is, therefore, a complex phenomenon to define, and this complexity shows in itself how interrelated East African towns, villages, and hamlets were.

This chapter analyzes these interconnections through three sections. The first relates to the ways in which people came to be enslaved and were brought (eventually) to work in urban settings. The second section is about what it was like to live and work as an urban slave and how experiences depended upon gender, social status, skills, and the socio-economic status of one’s master. The third and final section explores how urban slaves became emancipated largely through their own efforts, but also with the support of their masters, and anti-slavery legislation.

Entry into Slavery

The peak of the slave population in Zanzibar and Pemba was reached in the 1840s and 1850s, soon after the increase in slave imports in the mid-1840s at the time of the Hamerton Treaty (among other anti-slave trade measures). In the early nineteenth century, the total volume of the East African slave trade was between 6000 and 13,000, much of which was absorbed into the Indian Ocean trade. In the 1850s, the trade between 14,000 and 15,000, and in the 1860s it increased further to 20,000.⁷ When Seyyid Barghash, the second Sultan of Zanzibar, prohibited the importation of slaves to the island in June 1873, there was a shift in demand for slaves for plantations to women and children for domestic use in Zanzibar and Oman, so urban slavery actually retained its relevance in relation to plantation slavery.⁸

The people who were traded, whether they were destined to work in urban homes or on plantations, originated mainly from the rural East African mainland. According to estimates from 1860, the majority of slaves (about 15,000 out of 19,000) came from Lake Malawi. Most of the remaining 4000 slaves came from the Mrima coast and a minority of slaves in urban elite homes came from Uganda, Ethiopia, and Sudan.⁹ As the demand for slaves multiplied and as hinterland populations depleted in the nineteenth century, slave traders were increasingly opportunistic nearer the more urban coast.¹⁰

There were many ways in which people came into the slave trade, including war, abduction, sale, debt, or even sale of oneself. There was extensive depopulation on the East African mainland, in both towns and rural areas. This suggests that the demand for slavery in urban centers on the coast robbed hinterland areas of the chance to develop larger urban centers, unless they were along the caravan route. Some enslaved persons did not enter the coastal slave trade immediately but were rather pawned and passed several hands in the hinterland, until at some point they were sold into the coastal trade. The story of a boy of about eight years of age from the Makua region, which straddles modern-day Tanzania and Mozambique, gives some insight into the turmoil of undergoing pawnship and living as a slave in multiple homesteads:

I kept on thinking and thinking, and fancying, “I shall never get to a quiet, settled place, where there is no more going away and being sold over and over again.” I kept on brooding over this, and I could not get my food down; yet some of those people pitied me, but I refused to eat. I used to say I had had enough, because I was very, very sad indeed; and, besides, I had no one to play with.¹¹

Some accounts of entering the slave trade are particularly detailed, such as the story of Swema recounted by herself in Swahili and translated by Père Anton Horner in 1866. Swema was born around 1855 in Yaoland and remembers eating well and that her family had been able to barter rice for beads, cotton, and salt with passing traders. Swema’s life started to fall apart when her father, a hunter, died after being attacked by a lion. Not only did she lose her father, but locusts came and stripped the fields bare. There was famine and disease that killed all her siblings. Swema and her mother moved to a different area where they incurred debts that she was unable to pay back. As a result, Swema was seized and taken into the slave trade. Her mother struggled and pleaded to join and stay in the caravan Swema had been taken up in but, being considered well past her prime and very weak, she was unwanted by the slave traders, who beat her incessantly and eventually abandoned her on the route to the coast, revealing much about the brutality of the economics of the slave trade. Listless from her unimaginable grief, Swema finally arrived in Kilwa, where she and the other slaves were kept in a dark room for some days before departing again for Zanzibar. The arduous journey to Zanzibar by boat took six days, leaving Swema even weaker. At the slave market, orders were barked at her in a language she did not understand. She was so weak and despondent that the slave traders cut their losses, bundled her up in a straw mat, and buried her alive in a shallow grave. It was a young Creole man from Reunion who heard her muffled cries and took her to the Spiritans’ mission in Zanzibar, where she became Christian and later embarked upon a career as a nurse.¹² Narratives such as this belong to enslaved children who were declared legally free with the help of Christian missionaries on arrival in Zanzibar. Although they may not have been destined to work as urban slaves, their journeys to the coast are likely to be similar to those of many others who did end up as urban slaves.

It was logical that capturing people from distances far off from the coast reduced the risk of flight once on the coast, but not all urban slaves originated from so far into the hinterland. Some of the enslaved persons originated from areas much closer to the coast. This was true of Mombasa, for instance, as a flight to the nearby mainland was not always easy as patrons were well connected and frequently helped each other trace runaway slaves (this was less true of Zanzibar). Moreover, there are stories of people from the nearby hinterland being tempted onto ships with promises of better livelihoods further north, only to be taken out to sea to be disorientated, then taken to Mombasa.¹³ Especially for enslaved young people who had never traveled beyond their homesteads, this kind of trickery may well have worked.

Sometimes, the centers in which slavery was most institutionalized could be vulnerable to slave raiding. In the 1850s, there was a spate of traders from Oman and the Persian Gulf abducting slaves from both town and country in Pemba, Mombasa, and Zanzibar.The Consul of Zanzibar reported on 28 March 1860 that:

Zanzibar resembles a city with a hostile army encamped in its neighborhood. Every person who is able to do so, sends his children and young slaves into the interior of the island for security, people are afraid to stir out of their houses after dark, reports are daily made of children and slaves kidnapped and in the suburbs of the town. They even enter the houses and take the children away by force.¹⁴

Ironically, these abductions seem to have been a consequence of the enhanced patrolling of the slave trade, which had made acquiring slaves even more desirable.¹⁵

So, once enslaved, slaves could be vulnerable to re-enslavement and abduction, even in urban areas like Zanzibar. Some arrived as slaves in urban centers having already worked as slaves in rural areas. However, most of the enslaved persons who had recently been taken from their homes on the mainland were destined for plantation slavery on the coast or the Middle East. There was little appetite among slave owners living in the towns of East Africa for slaves recently arrived via the caravan route. Urban slaves who lived and worked in the homes of their masters tended to be wazalia (s. mzalia), meaning “born here.”¹⁶ “Here” could either mean the actual house of their owner, or it could carry a more general meaning of “here in the town.” Thus, most enslaved persons who entered urban slavery were second or third generation slaves. Therefore, their entry into slavery was usually from birth. These wazalia spoke Swahili and were more assimilated than slaves who had recently arrived. This notion of being more assimilated is central to how urban slavery has been conceived by scholars and will be explored in the following section that deals with the spectrum of exploitation, coercion, and violence that people living and working as urban slaves experienced.

Extraction of Labor During Slavery

Enslaved persons who lived and worked in urban areas in East Africa experienced varied levels of exploitation, coercion, violence, and social stigma. What it was like to be an urban slave depended on a number of factors, including the slave’s gender, the slave’s social status, whether the slave lived in their master’s home, as well as the social and economic status of the masters. The conditions of a slave depended greatly upon the financial security—or lack thereof—of their owners. Some of the worst abuses towards slaves occurred when slave owners had turned out slaves who had become sick, as they could not afford to look after them. This was increasingly the case towards the end of the nineteenth century when the economic impact of the collapse in clove prices truly made its mark. To take an example from 1895, in which a slave mistress beat Jamili, her slave of about eleven years, then tied her by the arm to a tree. Following this incident, Jamili’s arm swelled so much that the rope could not be removed. Jamili’s arm was then cut off and she was sent out to live on the street.¹⁷ This shows that not all slave owners in the towns were wealthy and that this lack of wealth sometimes aggravated the conditions of enslavement.

The slaves who lived with their urban elite masters played mostly non-commercial functions. In 1842, Hamerton wrote that, “a man’s wealth and respectability in the dominions of the Imam of Muskat is always estimated by the number of African slaves he is said to possess.”¹⁸ Likewise, Ephraim A. Emmerton, a Salem merchant who regularly visited Zanzibar, declared in 1849 that, “[s]laves are owned here because it is fashionable to have them, not because it is profitable.”¹⁹ The social value of having slaves visible in the household, rather than hidden in the clove farms that gentile society did not visit, is reflected in prices for slaves as plantation slaves were sold for only £5–8, while domestic servants were sold for £12–25.²⁰ Not all elite slave owners thought this way. Some considered the accumulation of slaves as a demonstration of wealth was a needless extravagance. They would therefore have their slaves trained in various trades, potentially hired out for a profit, and thus reduced their household expenditure.²¹ No doubt slave owners were increasingly likely to think this way amidst the economic downturn of the late nineteenth century.

In Swahili coastal towns, slaves who lived in the households of their masters were usually wazalia and this, along with their proximity to the urban centers, granted slaves some social status. It is noteworthy that “mshenzi” (a “raw slave” or, more generally it could mean “barbarian” or “foreigner,” pl. washenzi) and “mjinga” (“idiot slave,” pl. wajinga) carried the suggestion of rural origins and were obviously derogatory. So, through these terms, we see that the social status of slaves varied depending on whether they were seen as coming from the country or the city. These views were not just held by elites, but even by slaves. While studying the impact of cholera in Zanzibar, Dr. James Christie also observed that “The town negroes look down upon their country cousins with a good deal of contempt, and consider themselves a superior class.”²² Representing the other end of the social scale are words like “uungwana” (“civilization”) and “uarabu” or “ustaarabu” (“Arabness”) that are associated with Islam and the architectural and clothing fashions of the urban coast. The wazalia urban slaves occupied a position somewhere between the extremes of enslaved ushenzi (“barbarianism”) and freeborn uungwana (meaning something akin to refined urban Muslim civilization). They seem to have been treated like fictive kin, were trusted to work independently, and usually had some knowledge of Islamic teaching. Wazalia, by demonstrating their loyalty through years of service, could climb to positions of considerable authority as politicians, plantation supervisors, caravan leaders, traders, dhow captains, or skilled artisans. For instance, Emily Ruete (then Seyyida Salme of Zanzibar) was taught calligraphy by one of the Sultan’s slaves (most likely a mzalia) “who was notoriously proficient in the art, to the dignity of writing master.”²³ This is all to say that the social status of urban slaves was extremely varied and linked closely to their proximity to the town and assimilation to urban culture.

Urban slaves were also sometimes allowed to hire themselves out, this was less often the case in very rural parts. Hire slaves were known as vibarua (s. kibarua) and had to share a portion—usually half—of their wages with their master. The owner’s financial share of the kibarua’s labor often changed over time. For example, in the case of Rashid bin Hassani, who worked as a kibarua in the late nineteenth century, he gave over all his wages to his mistress, Bibi Zem-Zem, but once he was married, he kept his entire wage. As he described it, he was “merely under Bibi Zem-Zem’s protection.”²⁴ Male vibarua work was very varied and could include building work, carpentry, boatbuilding, smithing, sailing, and load-carrying. Meanwhile, although female vibarua did sometimes work as porters, they tended to be more commonly forced into sex work, or performed household work such as cooking and water fetching. Vibarua tended to be wazalia who had gained the trust of their owners. Many of them lived with their owners, but some lived in their own homes. In the latter cases, these slaves could accumulate enough wealth to rebuild their homes with lime and stone, or to have some domestic slaves of their own. For landless or newly arrived people—whether slaves or free—acquiring slaves to hire out was a common strategy that could help garner both wealth and status. There was a significant market for this kind of work, especially when European and Indian traders could not legally own slaves (from 1860 as per the Indian Penal Code), while also not being able to easily acquire non-slave labor.

Some female slaves in both town and country were classed as “suria.” “Suria” were female slaves who had sexual relations with their owners, who they were not legally married to. Their status was both variable and precarious. Although the children of suria were legally free according to Islamic law, their status could be ambivalent. In Lamu it was believed that suria contaminated her offspring. In Zanzibar, suria and their children were more likely to suffer from status struggles if they belonged to non-elite families. The account of Emilie Ruete suggests that concubines in Sultan’s palace were highly regarded, and it is noteworthy that her own mother was a Circassian concubine, and Ruete had nonetheless inherited a sizeable fortune upon the death of her father Seyyid Said. Many of Said’s sons, such as Seyyid Barghash, who became Sultans, were children of concubines.

Vijakazi (female slaves working within the household, s. kijakazi²⁵) were mostly young women who were also wazalia and tended to live with their masters in urban areas. These slaves were greatly valued, fetching higher prices than unskilled male slaves of the same age group and also higher than older women.²⁶ Vijakazi would carry out domestic work such as cooking, cleaning, washing laundry, fetching water, and sometimes fieldwork. In less affluent homes, the mistress of the house would perform these tasks alongside them.²⁷ Sometimes vijakazi were given gifts of money or earned money from selling foodstuffs.²⁸ Vijakazi (and, for that matter, suria) who refused the sexual demands of their owners were sent out to the fields to work. There was a proverb that went: kijakazi kina meno chauma/sikitaki tenna, kipileke shamba/kikalime, meaning, “the slave girl has teeth, she bites/I do not want her any longer, send her to the fields/to do agricultural work.”²⁹ This proverb is telling about both the potential for the sexual exploitation of vijazaki, but also the likely preference for work in the town as opposed to the plantations. Conversely, though, it may have been that work in more rural parts offered some an escape from sexual abuse.

For many elites, who tended to live in towns, slaves acted as an extension of the self and their owners shaped their dress and comportment in order to represent their own perceived identity. Elite women, in particular, had their vijakazi do things that were considered improper for them to do themselves, such as going outside, spreading gossip, or attending dances. The female slaves of elite families carried umbrellas to hide their mistresses from the sun and male onlookers. For more complete coverage, groups of slave women sometimes carried coverings known as ramba or shiraa around their mistresses. Other tasks might include massaging their mistress to sleep, while another gently fans her. At the weddings of freeborn women, they would get their skin rubbed with ground sandalwood, and have their hands and feet painted by slave women. The kinds of female slaves who performed these tasks were often referred to as wapambe. The term comes from the verb “-pamba” (to adorn) but it is unclear whether it was the wapambe themselves who were adorned or did the work of adorning their mistresses, or both. The meaning of the term has changed dramatically since, to mean “bodyguard,” though the sense of a person serving their master or employer closely still remains.³⁰

Household slaves, who were mostly women, often held very intimate relationships with their town-dwelling owners. The case of slave women who looked after elite slave owners’ children illustrates this point. In Ruete’s autobiography, she explains that slaves who looked after royal children were “highly esteemed and honoured as long as they lived” and “as a rule they are freed as reward for their fidelity and devotion.” They were also seen as “second mothers” to the children. Ruete believed that the slave nurses of Zanzibar were far superior to the nurses available for employment in Germany, the country she fled to in order to marry her German husband. She mused that this was probably because German nurses tended to have their own children, and working away from home was a huge sacrifice for them. Meanwhile, for the enslaved childcarer in Zanzibar:

She has been in [her master’s] service for years, may even have been born in her house: thus, her own interests and those of her master’s are closely knitted together. And further, there may be added a circumstance of much weight—a black nurse is not required to part with her child, but frequently, if not always, she retains it. The child of the nurse receives the same nourishment as its little foster brother or sister, shares its pap.³¹

Likewise, Bi Kaje, a woman who grew up in a slave-owning family in Mombasa and whose life has been recorded in oral history, remembered many of the slaves in her household affectionately. In many ways, they seemed to be treated much like part of the family, especially when they had become Muslim and acculturated to freeborn Swahili norms and practices.

Despite the widespread notion of slaves working in the master’s urban household as being “family,” status distance between them was always maintained. Slaves had to show deference to their masters at all times. For instance, they could not wear shoes in the presence of their masters and were not allowed to wear any kind of headdress or veil.³² In addition, they were rarely given names of the Prophet and His descendants. Although slaves were able to marry, they required consent from their masters. In addition, male slaves usually had to pay about five Maria Theresa Thalers, or a kilemba (turban). When showing their masters obedience, they would say “shikamoo” (a contracted version of “nashika miguu yako,” or “I clasp your feet”). Composed in the 1850s, the poem, “The Advice of Mwana Kupona upon the Wifely Duty,” instructed that a proper freeborn wife should ensure social distance was maintained between themselves and their slaves. It is noteworthy that this was an extremely popular poem that was often recited in the 1930s. It is difficult to determine how slaves perceived and personally experienced their social status, though there is evidence to suggest that slaves themselves used particularly dehumaniszng language to describe enslaved persons. For instance, an Anglican missionary in Zanzibar observed that “among the poorer class and slave population of Zanzibar,” a slave might be called “chitu” (a “vulgar pronunciation” of “kitu” meaning “thing”; newly arrived slaves in Zanzibar tended to pronounce “ki” as “ch.”³³ This implies a sense of feeling like inalienable and poorly treated property.

Porterage increasingly offered distinct opportunities for emancipation and upward mobility that many male slaves felt were worth the risk in the face of the great dangers of joining caravans. Slave masters rarely exercised their legal right to stop their slaves from joining caravans, so many were able to do so. Just as entering slavery involved a migration from the hinterland to the coast, escaping slavery often involved joining trade caravans that would take slaves from the coast into the hinterland, and with good fortune, they were likely to return with increased wealth to spend in the towns. Porters did not only gain wealth through transporting goods; many of them traded goods they had produced themselves or offered services wherever they ventured. Fundi (pl. mafundi) was the term designated to trained artisans or craftspersons, who were often wazalia. Mafundi rarely stayed put in the towns. In fact, it was through travel and porterage that they acquired wealth, status, and networks.³⁴ However, life on the road was hard and limited the more leisurely time people preferred to spend in towns. Proverbs such as “msafiri masikini ajapokuwa sultani,” meaning “a traveler is poor, even if he be a sultan,” evoke this sense that the pursuit of wealth through travel could undermine one’s enjoyment of life.³⁵ Another drawback was that Indian merchants, who were not allowed to own slaves themselves, often had slaves bound to them through debt. Because entrepreneur slaves required capital to conduct independent trade while working as porters, for instance, they sought loans from Indian financiers. Thus, some slaves in pursuit of wealth accumulation could find themselves entrapped by more than one master, though legally slave owners were obliged to foot the bill of their slaves’ debt.

A unique feature of urban slaves in East Africa was their tendency to be assimilated into hegemonic Arab culture. This came with many benefits as higher social status tended to open doors to better and more secure livelihoods. However, the proximity to one’s master, coupled with the high level of surveillance characteristic of the Swahili towns, may have felt oppressive. As for the nature of the urban slave’s labor, it was more variable than the kind of work that existed for slaves in the countryside. Urban slaves could live quite independently by hiring themselves out or working in intimate contact with their masters whom they maintained contact with long after abolition. Indeed, the nature of the terms upon which the urban slave was bound to their master had an enormous bearing upon their options for exiting slavery, as shall be discussed in the next section.

Exits from Slavery

There were varied ways in which urban slaves could cease being slaves, and some of these predated anti-slavery legislation. One method was to physically escape. Naturally, watoro (fugitive slaves) would aim for destinations where they could hide from their owners and avoid re-enslavement.³⁶ Some fled to Christian mission stations, often, where they would usually have to conceal (often without great difficulty) the fact that they had been enslaved. These mission stations would usually be situated on the outskirts of urban centers (so as to be removed from the temptations to sin those cities presented, as the European missionaries’ logic had it), but some became very large. Typically, it was rural slaves who fled to urban areas where they were more likely to be absorbed into the growing urban population with little chance of retribution from their former owners. However, in some areas, it was more likely for urban slaves to flee to comparatively rural areas. On the Kenyan coast especially, both urban and rural slaves fled to watoro settlements, which tended to be heavily barricaded and afforded them some protection from their masters. Though these settlements were in rural locations, some of them were extremely large and powerful. The best example is that of the Sultanate of Witu, which was founded in 1867 by Fumo Lotti, a chief who had been outlawed by the Sultan of Zanzibar and welcomed over 10,000 watoro. However, slavery continued in this area, and some watoro were sent back to their former owners, so it was not necessarily a route to emancipation. Still, many slaves tried to flee their masters, even after the enactment of anti-slavery legislation, which was often partial in its coverage. As for the urban slaves who fled, it has been argued that they were not fleeing the urban environment but rather the masters and terms of their enslavement; many of these urban slaves would look to other urban settings to benefit from a fuller role in urban institutions of commerce and community.³⁷

According to Islamic law and custom, slaves could and should be manumitted upon the owner’s death. Manumission was much more commonplace in the town than in the countryside, where slaves were working on financially critical plantations. For example, Sultan Said manumitted thousands of concubines, soldiers, domestics, and messengers, but his plantation slaves remained enslaved. Seyyid Barghash, too, ensured that all his “town slaves”,³⁸ who numbered more than 3000, as well as the slaves in his army and the concubines in his harem, were freed upon his death. As such, Dr. Christie estimated that nearly half the town’s population was made up of manumitted slaves.³⁹ Ironically, anti-slavery legislation made slave owners less willing to manumit their slaves as slaves were in increasingly high demand and owners were particularly reluctant to part with them.⁴⁰ Still, even after Seyyid Hamoud, the Sultan of Zanzibar between 1896 and 1902, passed the 1897 abolition decree, which contained a promise of monetary compensation, some slave owners did opt to manumit their slaves under Islamic law, perhaps following the lead of the sultan himself who manumitted hundreds of slaves after passing this decree.⁴¹

In Zanzibar, most manumitted slaves lived on land given to them by their former owners. In the 1830s, years before moving his capital from Muscat to Zanzibar, Seyyid Said (Sultan of Oman, 1807–1856) took possession of such a piece of land, called Kiungani (meaning “in the suburbs”), and used it as a way to maintain ties between himself, the Busaidi dynasty, and his manumitted slaves.⁴² Successive sultans who reigned over Zanzibar followed suit, sometimes charging the former slaves rent. Indeed, the practice of charging rent became increasingly commonplace through the emancipation process, which was in turn met with objections on the part of ex-slaves and their descendants. The united refusal to pay rent in Ng’ambo, Zanzibar, culminated in the 1928 rent strike.

So, manumission did not equal complete emancipation, though ties to one’s former owners were not necessarily undesirable for ex-slaves. In some cases, personal relationships between former masters and former slaves were very long-lasting. This was probably especially true for unmarried women or aging individuals. As such, this is the story of Baadadhiki recounted by her mistress, Bi Mariamu in Mombasa:

Baadadhiki, her work was to cook in the kitchen. She didn’t marry. She was married where she came from [Yao territory]; when she came here she didn’t marry again. She was granted her freedom by Mwana Hadija. She freed her in writing: “She is not my slave. I have used her many days. It’s over. Now she is a huru [freed slave].” Still Baadadhiki accompanied her to places. For example, if there were somewhere to spend the day, a wedding, Mwana Hadija said, “Let’s both go.” They both spent the day. If there were food and she wanted to take some, she would do so. [When Mwana Hadija died,] she stayed with a slave from the same household (my father’s concubine who had been married off to another man and lived in her own quarters).⁴³

Here, we might conclude that maintaining master–slave relationships was desirable for personal and economic reasons. Manumission was, for many, preferable to claiming legal freedom through the colonial courts as the ex-slaves freed by the abolition decree were thought of as slaves of the government, who had, the logic went, paid for them with the compensation money.⁴⁴

In many ways, claiming emancipation through anti-slavery legislation did shift the control from slave owners to the government. For example, article five of the 1897 abolition act dictated that ex-slaves were liable to be charged as vagrants if they did not have an employer, landlord, or patron to vouch for them. Consequently, individuals in the town who the British officials identified as “vagrants” were rounded up and taken to work on plantations. This vagrancy article was the first of many legal measures to attempt to control the labor of able-bodied workers.⁴⁵ Perhaps the greatest limitation was the lack of provision for concubines, which was due to British fears that they would turn to sex work if freed. Both the 1897 and 1907 abolition decrees stated that the law would only allow concubines to claim freedom on the grounds of cruelty. The 1907 ordinance was amended in 1909 to include concubines, but they required the consent of their owners to both claim freedom and “continue to be entitled to all the rights and privileges which they have previously enjoyed under the Mohammedan Law” (e.g. inheritance). Also, concubines who left their masters without their consent would lose custody of their children.

Despite the skepticism towards abolition and its significant limitations, anti-slavery legislation was an important way in which slaves—both urban and rural—were able to emancipate themselves. In the late nineteenth and early twentieth century, several decrees were issued that allowed slaves to claim legal freedom in East Africa. These included those of 1890, 1897, 1907, and 1909. Little is known about how these decrees affected urban and rural areas differently but it seems that urban slaves were more likely to claim legal emancipation. Following the 1897 decree, 64 percent of the slaves on Zanzibar island who claimed emancipation lived in the city. It is thought that many of these urban slaves were vibarua who saw an opportunity to rid themselves of the duty to pay part of their earnings to their owners. For other vibarua—especially those who were older—they preferred to continue paying their owners so as to maintain their relationship with their owner in case they needed to turn to them in hard times.

Yet, equally, there is also reason to believe that urban slaves—especially those who had converted to Islam and integrated into coastal culture—were less likely to claim freedom through the anti-slavery legislation. In Lamu, urban slaves who had converted to Islam were taught that these secular laws had no bearing on their situation: according to Muslim law, slaves could only be freed by their masters. In addition, slaves of elites (many of whom resided in urban centers) were less likely to claim freedom because they had more to lose from cutting off ties with their patrons. This was especially true for the slaves of elite owners who were status symbols. These individuals often lacked the training and transferrable skills that could help them get work elsewhere, and many had little choice but to remain dependent on their owners. Thus, although the majority of slaves who claimed legal freedom resided in urban areas, it did not follow that most urban slaves claimed legal freedom.

Legal abolition sparked the migrations of ex-slaves from plantations to the towns, not necessarily for paid work. In Kenya, young Mijikenda males would travel to Mombasa to work unpaid and gain experience as watumishi (servants) while also taking Koranic studies. When they completed this training, they would often search for well-paid jobs as domestic servants in colonial households. In Zanzibar, many ex-slaves gravitated to the town, which was a refuge from dependent relations. Many of the ex-slaves educated by missions did so, too, finding and developing new networks beyond the small Christian community, much to the despair of missionaries.⁴⁶ Wage labor was relatively plentiful in import–export houses, as well as in domestic service and government. Urban wages were also much higher than those that could be found in rural parts. When wages rose due to unusually heavy crops, urban workers found temporary work on the plantations.

Although many household slaves in urban areas were women, it was also true that females were less likely to escape agricultural slavery than males. The main ways out of agricultural slavery were to join the caravan trade or become a craftsperson. However, women were excluded from these opportunities, largely because their male counterparts prohibited them from doing so. For women slaves, concubinage was the most likely route out of agricultural slavery but this was rife with risk and, even if they were to reach the highest possible status in a household as a concubine, they would probably never be able to manage their own household.⁴⁷

Post-abolition, many ex-slaves maintained relationships with their former owners. Interviews with Muslim women in Mombasa in the 1970s by Margaret Strobel revealed as much. One woman who belonged to a slave-owning family spoke of the closeness between her and one of their former slaves:

She was like my mother, so in the morning she would come and sweep for me. She lived in her own rented house. In the morning she came to sweep for me, she drew water for me, she cooked food for me. That’s it, children have been born there, and she has raised them and made them like her own. You can’t come and tell her, “do this.” […] Once the British had come, you didn’t dare to call a person a slave. Mama, that’s all. You send her on errands, she is useful to you if she wants to be herself, out of her own kindness.⁴⁸

An anonymous informant from Mombasa conveys a different way of looking at these kinds of situations as evidence of the persistence of the slave-master dynamic:

Nowadays people are not sold [as slaves], but something persists. They know that they cannot come to us inappropriately and say, “I want my child to marry yours.” We attend weddings and dances together, but they know their place. There is one thing, however; if a child of ours wants to set up household with them, it is the child’s own business. [...] Even nowadays, people are not [considered to be] of one kind [i.e., equal]. Even if we see that his thoughts are good, he will not receive a wife from us. We go to weddings together, but the matter of intermarriage creates problems. Older people still do not approve.⁴⁹

As Strobel’s informants suggest, even when slaves managed to attain legal emancipation and secure livelihoods, the hardest thing to shift was slave status, which even today can compromise a slave descendant’s claim to citizenship. In Kenya, the descendants of slaves are still referred to as “wageni” (foreigners). A good example of contemporary discrimination based on slave genealogies is the case of the self-named “Freretownians” of Mombasa, who are mobilizing politically to be recognized as an ethnic group. The importance of being part of an ethnic group officially recognized by the Kenyan government is colossal as without this (or without marrying a person who is part of a recognized ethnic group), one cannot get an identity card, which is essential for gaining access to state social welfare, schooling, and healthcare. The Freretownians are descended from the ex-slaves and watoro, living in Frere Town or other surrounding mission settlements, who the CMS gave land in the late nineteenth and early twentieth centuries. At this time, Frere Town would not have been considered “urban.” Moreover, the inhabitants of the mission station were, at least in theory, legally freed slaves rescued through anti-slavery patrols and free persons from the surrounding areas, though many were likely to have been slaves fleeing their masters in Mombasa. Despite its rural beginnings, Frere town eventually became an important part of Mombasa city, by which point the land was classified as church property and much of it was sold to private enterprises, though it had been maintained for decades by the slave descendants who had valid titles for the land. The Freretownians are not an isolated case. Similar challenges are faced by people on the coast who have struggled with stigma as they live on land (much of which had been rural and is now urban, and thus more valuable) owned by former slave owners. So, while slavery as a labor practice is rare in East Africa, the social, economic, and political impact of slavery system is still great. And, although the ascendants of these individuals may not have been urban slaves, the difficulties of their predicament, namely their landlessness, are enhanced by recent urbanization.

Conclusion

There were several urban centers of East Africa, and their prosperity, size, and influence were derived largely from slavery. Yet there was significant overlap between urban and rural slavery, so much so that “urban slavery” proves to be a problematic analytical category. Many slaves who had resided with their masters in towns moved to peri-urban settlements when they were manumitted, often benefitting from access to the land courtesy of their former masters. Many rural slaves flocked to the towns once they had their freedom papers, but were then subject to coerced labor under local businesspeople or the colonial government, who might label them “vagrants.” Urban slaves often shared intimate relationships with their masters, which suggests that towns fostered a sense of loyalty between the enslaved and the slavers. However, towns were also a refuge for watoro who would hide themselves in the growing maze-like mud and thatch townships that developed on the edges of the main Swahili towns. Equally, on the plantations there were some slaves whose owners entrusted them with much responsibility, to help oversee and manage the slave labor force, so there was some variation in the roles of slaves in rural areas.

Still, slavery in urban spaces was distinctive compared to the forms of slavery that existed in rural parts. Firstly, most urban slaves had been enslaved for a long time or were second- or third-generation slaves. This was in contrast to the enslaved who were exported into the Indian Ocean trade, or those who populated plantations. Secondly, slavery in urban areas was more varied according to the more varied social and economic standing of slave owners and forms of labor they tried to extract from their slaves. Thirdly, in terms of exiting slavery, slaves of elites—who tended to live in urban environments—had a good chance of being manumitted. Hire slaves who had already carved out ways of making independent livelihoods were also likely to claim emancipation once slavery became illegal. The distinction between town and country would have felt very stark for people at the time, even if in practice lives and livelihoods constantly transgressed it. The enslaved who rebelled or resisted were threatened with exile to the countryside. Meanwhile, many rural slaves sought the opportunities available in the town. Lacking ties in the city could make slaves more vulnerable. The town and the communities within it that some urban slaves belonged to, were assets. Yet, like all communities have the potential to be, they could be oppressive and force slaves deeper into relationships in which they were irrevocably confined. These dependent relationships characterized Swahili towns and, in some cases, also contributed to their downfall as they became increasingly impoverished and comparatively insignificant in the shadow of leading economic centers such as Nairobi and Dar es Salaam in the twentieth century.

Notes

• 1. This paper was written as part of the project, African Abolitionism: The Rise and Transformations of Anti-Slavery in Africa (AFRAB), which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. [885418]).

• 2. Paul E. Lovejoy, Transformations in Slavery: A History of Slavery in Africa (Cambridge University Press, 2011), 225–26.

• 3. Johann Ludwig Krapf, A Dictionary of the Swahili Language (London: Trübner and Co., 1882), 269.

• 4. Glassman, Feasts and Riot: Revelry, Rebellion, and Popular Consciousness on the Swahili Coast, 1856–1888 (Portsmouth, NH: James Currey, 1995).

• 5. Laura Fair, Pastimes and Politics: Culture, Community, and Identity in Post-Abolition Urban Zanzibar, 1890–1945 (Athens, OH: Ohio University Press, 2001); Patricia W. Romero, Lamu: History, Society, and Family in an East African Port City (Princeton: Markus Wiener, 1997).

• 6. G. H. O. Abungu and H. W. Muturo, “Coast-Interior Settlements and Social Relations in the Kenya Hinterland,” in The Archaeology of Africa: Food, Metals and Towns, eds. Bassey Andah, et al. (London: Routledge, 2014), 694–704.

• 7. Abdul Sheriff, Slaves, Spices, & Ivory in Zanzibar: Integration of an East African Commercial Empire Into the World Economy, 1770–1873 (London: James Currey, 1987), 60.

• 8. Edward A. Alpers, “Representations of Children in the East African Slave Trade,” Slavery & Abolition 30, no. 1 (2009): 27–40.

• 9. Lovejoy, Transformations in Slavery, 230.

• 10. Hideaki Suzuki, Slave Trade Profiteers in the Western Indian Ocean: Suppression and Resistance in the Nineteenth Century (New York: Palgrave Macmillan, 2017).

• 11. A. C. Madan, ed., Kiungani, or, Story and History from Central Africa (London: G. Bell and Sons, 1887), 41.

• 12. Edward A. Alpers, “The Story of Swema: Female Vulnerability in Nineteenth-Century East Africa,” in Claire Robertson and Martin A. Klein, eds., Women and Slavery in Africa (Madison: University of Wisconsin Press, 1983), 185–219.

• 13. Justin Willis, Mombasa, the Swahili, the Making of the Mijikenda (Oxford: Clarendon Press, 1993).

• 14. Cited by Suzuki, Slave Trade Profiteers, 102.

• 15. Suzuki, Slave Trade Profiteers, 103.

• 16. William Ernest Taylor, African Aphorisms: Or, Saws from Swahili-Land (London: Society for Promoting Christian Knowledge, 1891), 78; Carol Eastman, “Service, ‘Slavery’ (Utumwa) and Swahili Social Reality,” Afrikanistische Arbeitspapiere 37 (1994): 91.

• 17. J. P. Farler, “Second Letter from the Rev J. P. Farler,” Central Africa, September 1895.

• 18. Hamerton to the Secretary to the Bombay Government, Zanzibar, 2 January 1842, Zanzibar National Archives, AA12/29/43.

• 19. Norman Robert Bennett and George E. Brooks, New England Merchants in Africa: A History Through Documents, 1802 to 1865 (Boston: Boston University Press, 1965), 427.

• 20. Frederick Cooper, Plantation Slavery on the East Coast of Africa (New Haven: Yale University Press, 1977), 195–196.

• 21. Emilie Ruete, Arabian Princess (Leiden: Brill, 1993), 81.

• 22. James Christie, Cholera Epidemics in East Africa, from 1821 till 1872 (London, n.p., 1876), 312.

• 23. Ruete, Arabian Princess, 45.

• 24. Margery Perham, ed., Ten Africans (London: Faber, 1963), 99.

• 25. Older female slaves were called wajakazi (s. mjakazi).

• 26. Jan-Georg Deutsch, “Prices for Female Slaves and Changes in Their Life Cycle: Evidence from German East Africa,” in Gwyn Campbell, Suzanne Miers, and Joseph C. Miller, Women and Slavery, vol. 1 (Athens, OH: Ohio University Press, 2007).

• 27. Katrin Bromber, “Mjakazi, Mpambe, Mjoli, Suria: Female Slaves in Swahili Sources,” in Women and Slavery, 111–28.

• 28. Deutsch, “Prices for Female Slaves and Changes in Their Life Cycle”.

• 29. C. Velten, Desturi za Wasuaḥeli, na khabari za desturi za sheriʻa za Wasuaḥeli (Göttingen: Vandenhoeck & Ruprecht, 1903), 394.

• 30. Bromber, “Mjakazi”.

• 31. Ruete, Arabian Princess, 65–66.

• 32. Fair, “Dressing Up”.

• 33. Madan, Swahili-English Dictionary, 34, 122.

• 34. Glassman, Feasts, 75, 78.

• 35. Taylor, African Aphorisms, 70 no. 310.

• 36. Edward A. Alpers and Matthew S. Hopper, “Speaking for Themselves? Understanding African Freed Slave Testimonies from the Western Indian Ocean, 1850s-1930s,” The Journal of Indian Ocean World Studies 1, no. 1 (2017): 72.

• 37. Glassman, Feasts, 108.

• 38. The English translation of the will used the phrase, “town slaves.” No. 56 Euan Smith to Lord Salisbury, Zanzibar, 7 April 1888, FO 84/1906, UKNA.

• 39. Christie, Cholera Epidemics in East Africa, 303–5.

• 40. Elisabeth McMahon, “Trafficking and Reenslavement: The Social Vulnerability of Women and Children in Nineteenth-Century East Africa,” in Trafficking in Slavery’s Wake: Law and the Experience of Women and Children in Africa, eds. N. Lawrance Benjamin and L. Roberts Richard (Athens, OH: Ohio University Press, 2012), 39.

• 41. Frederick Cooper, From Slaves to Squatters: Plantation Labor and Agriculture in Zanzibar and Coastal Kenya, 1890–1925 (New Haven: Yale University Press, 1980), 74.

• 42. Fair, Pastimes, 118.

• 43. Strobel, “Slavery and Reproductive Labor,” in Women and Slavery in Africa, 126.

• 44. Cooper, Slaves to Squatters, 76.

• 45. Ibid., 116–17.

• 46. Michelle Liebst, Labor and Christianity in the Mission: African Workers in Tanganyika and Zanzibar, 1864–1926 (Oxford: James Currey, 2021).

• 47. Glassman, Feasts, 90–91.

• 48. Strobel, “Slavery and Reproductive Labor,” 126.

• 49. Strobel, Muslim Women in Mombasa, 1890–1975 (New Haven: Yale University Press, 1979), 18.