A Commercial City in the Economic and Cultural Cross-Currents
MANY PEOPLE VISIT THE SITES OF SWAHILI CITIES. ON KENYA’S NORTH COAST Lamu is an essential stop for backpackers, fashion photographers, and the European jet set. Gedi, a couple of hundred miles south of Lamu, is billed as the Pompeii of East Africa and is a regular feature on the tourist route. Visitors come to Mombasa because it is the most important port city in East Africa and to Malindi because it has a thriving beach (and sex) tourism industry. Just to the south in Tanzania tourists flock to the island of Zanzibar, some to visit the beaches and others to see the historic Stone Town with its narrow streets and romantic decrepitude. On the mainland the Kunduchi ruins just outside the city of Dar es Salaam are a frequent stop on the way to the beach hotels north of the city. Kilwa, however, is visited by only the most committed Swahili ruins enthusiasts. To get there one must make a fairly arduous overland journey from Dar es Salaam and then go by boat to the island, where the ruins of the city lie. Though there is now a town on the mainland called Kilwa, the original site was on the island, abandoned since the sixteenth century. Kilwa’s isolation today means that one really has to want to visit the place to end up there. I have made it to all the sites listed above along with dozens of more obscure sites on Zanzibar, Pemba, and Mafia islands, but Kilwa has thus far eluded me.
I first became interested in Kilwa and the other cites of the East African coast as an undergraduate. I had spent most of my childhood living in West Africa, moving from Ghana to Nigeria to Cameroun. In each of these places we lived in cities, although in the course of our travels we saw many villages and, very occasionally, wildlife. Thus, my experience of Africa was essentially urban. When we were back in the United States on leave I realized that cities were not something that my American peers associated with Africa. Kids I met would ask if I saw lions. This is rather like an African meeting someone from, say, Philadelphia, and asking if he worries about the grizzly bear menace. Other people would ask if my family lived in a hut—we did not. Through these kinds of questions, I realized that cities were not the first thing that came to people’s minds when they thought of Africa; instead, they thought of villages and wilderness. But my Africa, the Africa of my memories, was one of cars, buses, crowds, noise, and buildings. It was all about the city.
Likewise, my scholarly interest in Africa has always been about cities. I took my first African history class in college at a time when my parents were living in Dar es Salaam, Tanzania. During a visit on summer break, while on our way to the beach, we stopped to visit a small group of ruins. I had seen ruins before in Africa—slave trade castles in Ghana, some pre–World War I German plantations in Cameroun, and the abandoned Safari Club, a colonial-era nightclub on a mountaintop outside of Yaoundé, Cameroun, that looked like a place where Humphrey Bogart’s Rick might have set up shop after giving up the lease on his café in Casablanca. But these ruins were different. They predated the colonial period by a good three hundred years and it was obvious that what we were looking at was not a village—it was a small city. It was a bit of the African past that resonated with my experience of the African present. Those ruins stuck with me, and when I went to graduate school, it was the Swahili coast and its cities that I studied. I wrote my dissertation and first book about Zanzibar, but I always remained interested in the once-bustling, now-abandoned and isolated Kilwa, which to me embodied the Swahili world at its peak.
Kilwa served as a major trade center that effectively monopolized the East African gold trade, reaching the height of its cultural and commercial success from the thirteenth to fifteenth centuries. When the Portuguese arrived in Kilwa shortly after 1500, they were awestruck by the built environment of the city and determined to take advantage of the rich resources and trade on the Swahili coast. Kilwa, a commercial city, was unprepared for military assault and quickly fell. This chapter will review Kilwa’s rise and fall as well as the legacy left behind through historical records and its modern-day ruins.
KILWA AND THE WORLD:
A City in the Cross-Currents of Indian Ocean Trade
Kilwa is on a small island about 90 miles (145 kilometers) south of Dar es Salaam, the capital of the modern nation of Tanzania. (See Map 9.1.) It is near the Rufiji Delta, where Tanzania’s biggest river enters the Indian Ocean. The channel that separates the island from the coast is shallow and not very wide. Like many Swahili cities, Kilwa is situated in a place where land and sea intermingle. The island is part of a complex estuary with many small inlets and islands. (See Map 9.2.) Kilwa was among the most important of the Swahili cities, and its location serves as a sort of metaphor of the nature of Swahili society. The Swahili occupy the narrow coastal strip from southern Somalia to northern Mozambique, with the majority of the Swahili people living in Kenya and Tanzania. On the most basic level, to be Swahili is to be Muslim, to speak the Swahili language, and to be urban (or at least to be in the political or economic shadow of a city). But there is more to it than that. Just as the estuarine site of the city of Kilwa blurs the line between land and sea, the Swahili world is rooted on the land but strongly influenced by the sea. What distinguished the Swahili from the people of the interior, whom they dismissively referred to as Washenzi, or savages, was their access to the trading world of the Indian Ocean. Islam, luxury items like cloth and ceramics, and various foods that were unique to the coast, all came to the Swahili world courtesy of the trade networks that linked towns like Kilwa into a system that stretched from the Mediterranean to the South China Sea.
Map 9.1. East African Coast, ca. 1450
Kilwa was one of many city states on the East African coast. During the period between 1200 and 1500, Kilwa in the south and Mombasa in the north were the largest and commercially dominant towns in the region.
In the fourteenth century when the Indian Ocean trade was booming, Kilwa was anything but a backwater. Its sultans, who sometimes claimed origins in the Persian city of Shiraz and at others in the Hadhramaut region of Yemen, minted their own coins, built a mosque that is believed to have been the largest enclosed space in sub-Saharan Africa, and hosted visitors and merchants from around the Indian Ocean. When the Swahili coast was at its cultural and economic height from the thirteenth through the fifteenth centuries, Kilwa’s only real regional rival was Mogadishu, located in present-day Somalia. When the Portuguese arrived in Kilwa shortly after 1500, they were awestruck by the built environment of the city. An early Portuguese account of the Southeast Asian city of Malacca (near modern Singapore) mentioned the presence of merchants from Kilwa, a sign of the city’s far-reaching engagement with Indian Ocean trade, though it appears that the Swahili hosted merchants in their cities more often than they ventured as far afield as Malacca.
Map 9.2. Kilwa Kisiwani, ca. 1400–1600
This map shows a detailed view of the site of Kilwa Kisiwani (Kilwa on the Island). The city’s island location offered access to the sea and the coast, while making the city difficult to invade from the mainland.
Kilwa’s history can open a number of windows for us. As the archetypical city of the Swahili coast when it was at its maximum, it offers us a view of the rise of the Swahili in the context of the Indian Ocean world. From 1000 to 1500 the Indian Ocean was the economic center of gravity for most of Eurasia and Africa, and the rise of the Swahili sites occurred in the economic and cultural cross-currents of the Indian Ocean. As a place that was urban—even urbane—Kilwa’s story also challenges some of our deeply held ideas about Africa’s past. Kilwa has also been at the center of one of the great debates among African historians: the argument over the role of non-Africans in the development of states and other institutions in the continent and the opposing view that sees Africans as the primary forces in their own history. Finally, Kilwa’s decline is as historically interesting as is its rise. Its transformation from major commercial center and thriving port to isolated ruins brings into view major shifts in the economic and political geography of the Indian Ocean that took place shortly after 1500.
The World of the Swahili
From very early times the east coast of Africa has been in contact with the other lands of the Indian Ocean rim. The evidence for the earliest of these contacts is indirect. A cluster of food crops of Asian origin, including a type of banana as well as other less familiar crops, appeared in West and Central Africa in the first millennium and presumably passed through East Africa to get there, although no evidence of that passage has been found. The first documentary evidence that mentions the world of the Swahili comes to us in the form of a merchant’s guide to the Indian Ocean. Dating from the early part of the first millennium, the Periplus of the Erythrean Sea was written in Greek, probably by a Greco-Roman merchant from Egypt. It describes the routes that merchants should follow to get to the various trading ports of the Indian Ocean and what to expect when they arrive. The portions that deal with East Africa indicate that trade to East Africa was in the hands of Arab merchants, who sailed down the coast to a place called Rhapta, where they bought ivory, tortoise shell, beeswax, and slaves in exchange for cloth, grain, wine, and iron. Archaeologists remain uncertain where Rhapta was or even whether it was a permanent settlement or only occupied seasonally during the trading season. Most readings of the distances indicated in the Periplus suggest that Rhapta was somewhere in either modern Tanzania or Kenya. A prominent Tanzanian archaeologist, Felix Chami, considers the most likely location to be near the mouth of the Rufigi River in Tanzania—right in the heart of the region where the Swahili civilization would eventually take shape and very close to the eventual location of Kilwa.
The Periplus does not say much about the people with whom these Arab merchants traded, only that they built sewn boats and fished with a type of fish weir still in use in East Africa today. The name Rhapta may derive from the Arabic word for bound or sewn, and as such it may refer to the boats these people made. The consensus now is that by 100 to 300 CE, shortly after the time of the Periplus, Bantu-speaking farmers inhabited the coast. Prior to their arrival the coast was occupied by foragers and herders who did not disappear with the arrival of farming; herders in particular seem to have been in fairly close contact with the emerging farming communities on the coast.
By the middle of the first millennium a society that we might describe as proto-Swahili, in that they lived on the coast (the term Swahili means “people of the coast” in Arabic) but had not yet accepted Islam or begun to build cities, emerged on the East African coast. Small communities that not only farmed but also exploited the area’s resources dotted the coast, but there is little evidence of maritime trade. Between 800 and 900 CE, however, pottery from the Middle East, beads, and other items become increasingly common in Swahili communities, suggesting that they were engaged in long-distance trade with other regions of the Indian Ocean.
The increasing tempo of Indian Ocean trade and its growing importance in coastal communities had a transformative effect on the towns of the Swahili coast. Some of them grew in scale, becoming cities rather than towns. They became much more hierarchical and were dominated by a merchant elite, conventionally called “patricians,” who used imported goods and foods as markers of their social status. Between 800 and 1000 their inhabitants accepted Islam, making them members of a vast religious, intellectual, and cultural community that not only reached across the ocean to Asia and North Africa but also distinguished them from their non-Muslim neighbors on the coast and in the interior. In time they began to build public buildings, like mosques, in stone. Later the homes of the wealthy came to also be built in stone, though common people continued to live in mud-and-coral rag houses. They also began to use a language that is clearly ancestral to the modern language of Swahili. By 1000 the coast had become Swahili, and Kilwa was growing into one of the new cities that characterized the Swahili coast.
Kilwa’s Origin Myths
Because most sub-Saharan African societies were nonliterate, historians of Africa have few internally generated written documents to work with, especially for the earlier periods of the continent’s history. The Swahili coast is an exception. The Swahili embrace of Islam brought with it the religious obligation to read the Quran, which led to the spread of literacy in Arabic, at least among religious scholars and leaders, and eventually to the use of Arabic script to write both Swahili and Arabic. Although there is not the wealth of written documentation that one would find in European or Chinese archives, the city of Kilwa has two written histories. The first and oldest is known as the “Arabic History of Kilwa,” and the second and probably more recent is the “Swahili History,” both named as such because the former was written in Arabic and the latter in Swahili. The Arabic history came from the papers of an Islamic scholar, and though the only surviving manuscript dates from 1867, the original text is thought to date from the early 1500s. A German linguist collected the Swahili history in 1889. It was mostly likely an oral history that he had someone recite to him and then transcribed. Although the outline of both texts is similar, there are some differences in the names given to sultans and the details of events. It is also worth noting that any histories or chronicles composed during the period we are examining (900–1500) have been lost; a tropical environment does not favor the preservation of paper or parchment. There are, however, many brief dated inscriptions on tombs and mosques that help historians to flesh out the chronology of the various towns. The Swahili were also among a very small group of sub-Saharan Africans who produced coins. Because Islam discourages the representation of humans or animals, these coins have the names of the rulers who struck them and the dates of their production inscribed on them. The rulers of Kilwa were particularly prolific coin makers, so the names and reign dates of Kilwa’s kings are well established.
Both contain a story about the origins of Kilwa, but because the Arabic history is more detailed, we will use its version. The story begins in the Persian city of Shiraz in the ninth century. Interestingly Shiraz is quite far inland, so it’s hard to imagine that it ever had direct trade contacts with East Africa; here it seems to be representing Persia in a sort of generic sense. According to the Arabic History the sultan of Shiraz had a dream that foretold the collapse of his kingdom. In response, Sultan Husein and his six sons took to the sea, each in his own ship. Each ship stopped at different places on the East African coast, and the ship that came to Kilwa was the vessel of Ali ibn Husein (Ali son of Husein).
When Ali arrived at the island of Kilwa, he met a man named Muriri wa Bari. Muriri was a Muslim, and Kilwa already had a mosque. However, it was by a non-Muslim (an “infidel” in the text) from the mainland ruled Kilwa. The king was away on the mainland hunting but was expected back soon. When the king returned, Ali asked him if he could buy the island and settle on it. The king agreed and asked that, as payment, the island be encircled in cloth. This he collected and took with him to the mainland. In the Swahili version Ali actually covers the area between the island and mainland with cloth, allowing the king to walk to the mainland on a cloth causeway. That Ali had access to so much cloth earned him the epithet Nguo Nyingi, which is Swahili for “many clothes.” Apparently, however, the king was not satisfied even with this vast quantity of cloth and decided to retake the island by force. Muriri wa Bari suspected the king might do something like this and warned Ali of the danger. In the Arabic history Ali responds by having his followers dig out the channel between the island and the mainland, making it too deep for the king and his army to cross. The Swahili history depicts Ali using a much more interesting and symbolic defense: instead of digging out the channel, Ali reads from the Quran and offers sacrifices to make the channel impassible to the king, in effect using the power of religion rather than shovels to protect the island from threats from the mainland. Also noteworthy in the Swahili history is that when he first arrives, Ali marries one of the infidel king’s daughters.
These are stories that tell us more about how the Swahili saw their place in the world than they do about the origins of Swahili towns. However, there is much that is historically interesting about these stories, and archaeology can corroborate some of the material in them. First, a major aspect of Swahili identity concerns the idea of foreign origins. Elite families all over the coast claim either Arab or Persian origins. In fact, many of the people whom scholars (and books like the one you are reading) designate as Swahili refer to themselves as “Shirazi.” Although there has certainly been migration to East Africa from the Arab and Persian worlds, most of the coastal population is of local, African origin. What defines them as a people, however, is their cultural and commercial connections to places on the other side of the Indian Ocean. Thus, the cloth that Ali brought with him and used to purchase his new home is imported cloth, a mainstay of Indian Ocean trade and an important marker of social status for Swahili elites. But it is not simply cloth that distinguishes Ali from the king. Islam—especially in the Swahili history—another defining quality of Swahili identity, completes the separation of Ali’s island realm from the non-Muslim mainland. Thus, the importance of things foreign—people, objects, and religion—to the Swahili is established in these two histories.
At the same time, it is worth noting that when Ali, Kilwa’s “founder,” shows up, there is already a community of some sort on the island, including a Muslim population. Muriri wa Bari’s name suggests that he was not an Arab or Persian but rather a Swahili Muslim. That the Kilwa chronicle has Ali marry the king’s daughter suggests that the social distance between the Muslim people of the island and the non-Muslim people of the coast was not that great. Even Ali’s epithet, Nguo Nyingi, is telling in that though it refers to his access to foreign trade goods, it is also in the local language, Swahili, not in Persian or Arabic. Ali may have been from Persia, but the histories are quick to embed him in local culture. The society that developed in Kilwa and on the rest of the Swahili coast reflected the mixing that is apparent in the two histories. The Swahili towns were neither African (in the usual essentialized sense of the term) nor were they Arab or Persian colonies; rather, they are probably best described as part of the Indian Ocean world.
Kilwa’s Transformation
According to the Arabic history the founding of Kilwa occurred in the middle of the third century of the Islamic calendar, which would be roughly 875 CE. This date matches up with the archaeological record in an interesting way. Kilwa and the region surrounding it were certainly settled prior to the late ninth century. Kilwa itself was one of a number of small villages in the region, many of which appear to have been repeatedly occupied and abandoned, perhaps because their inhabitants practiced shifting cultivation. But between 750 and 1200 Kilwa, like many other sites on the Swahili coast, experienced a transformation. It grew in size relative to its neighbors, an elite class emerged, and some of its buildings began to be made of stone rather than mud. In short, it became a city.
How this process occurred is not entirely clear, but both written and archaeological records suggest that Kilwa’s connection to the commercial and cultural world of the Indian Ocean triggered the change. The written record speaks of the importance of imported cloth probably from Gujarat in India and of Islam as defining qualities of the Swahili cities. The archaeological record indicates that during this time imported pottery became available in places like Kilwa that were emerging as cities but not in the surrounding communities. Combined, these suggest the centrality of Indian Ocean trade to the development of Kilwa and the other Swahili towns.
Around 900 CE the Indian Ocean as a whole was experiencing a maritime revolution. In the western Indian Ocean this meant the development of new types of sails and the replacement of fragile steering oars with much more effective and sturdy rudders. The type of vessel that emerged as the characteristic ship in the region is called a dhow, and these vessels served to integrate the world of the Swahili into the broader Indian Ocean region. Dhows carried trade goods and passengers within the western Indian Ocean. Dhow captains used their knowledge of the Indian Ocean wind system to make long-distance voyages of a type that would be almost unthinkable for their contemporaries in the Atlantic and Pacific Oceans: both the Atlantic and Pacific are much bigger than the Indian Ocean, and because their “tops”—the northern land or island bridges that allowed early sailors to first cross these oceans—are at much higher and colder latitudes, they were more challenging to sailors wishing to stay near land while making their voyages. Chinese trade goods made it across the Indian Ocean long before they reached the Americas.
By contrast the Indian Ocean’s winds and weather were much better suited to early navigators. Because the Indian Ocean has monsoon winds that blow one way for part of the year and another way for the rest of year, it is possible to make long-distance voyages in the Indian Ocean and to have a tailwind going both ways, provided you have the patience to wait a couple of months for the winds to change directions. For ships coming from the Persian Gulf or India to East Africa, Kilwa is the southernmost point one can sail to with reasonable confidence of making it home on the next monsoon. If one goes further south, there is the possibility that the return voyage will take two monsoon seasons rather than one. So the area around Kilwa represented a sort of pivot point in Indian Ocean trade patterns. In the same way that straits in the entrance to the Red Sea or the Persian Gulf helped make Aden and Hurmuz important trading cities during the same period, the wind patterns that drove Indian Ocean trade helped make Kilwa an important trade center. They also enforced long layovers for ships’ crews who were waiting for the winds to turn. The result was a pattern in which sailors and merchants often maintained long-term social relationships in the ports they visited, and in many cases had wives and families in them too. At first Kilwa, like most Swahili cities, acted mostly as an intermediary between the people of the interior, who provided ivory, beeswax, slaves, and other goods, and the visiting merchants, who sought these goods. However, when gold from Zimbabwe began to reach the southern Indian Ocean port of Sofala, Kilwa emerged as the port where that gold was concentrated for shipment to the Middle East and India.
Thus, there were two phases in the growth and expansion of Kilwa. The first, which was associated with the emergence of an elite class through trade, took place between 800 and 1200 and occurred at numerous sites along the coast. During this time Kilwa grew from a village to a trading town as a result of its role as an intermediary with the interior. The second phase of growth occurred after 1200, when Kilwa became one of the dominant trading ports of the East African coast, a process that happened to only a few of the cities that had gone through the first phase of transformation. Its growth into the biggest and richest city on the Swahili coast likely stems from Kilwa’s role as an intermediary in the gold trade. Although the gold trade from Sofala, a city on the coast of modern Mozambique that was linked to the gold-producing areas of south central Africa, was originally dominated by the city of Mombasa, Kilwa wrested control of the trade from its rival sometime early in the thirteenth century. The result was Kilwa’s growth into one of the dominant cities of the East African coast.
The First Phase
Stephanie Wynne-Jones, an archaeologist who has worked in and around Kilwa, has taken a particular interest in Kilwa’s first transformation (though she does not make so sharp a distinction between the two periods as I do). Her work has shown that Kilwa was pretty much like any other small coastal settlement until 800, when new types of pottery, imported ceramics from the Middle East and China, appear in Kilwa’s archaeological record but not in the hinterlands. Chinese ceramics in particular came to East Africa though an interlocking system of trade circuits across the Indian Ocean. They would have changed hands many times before they arrived in places like Kilwa and, as a result, would be quite valuable after all the middle men took their cut. They seem to have functioned as markers of elite status, a sort of visual representation of their owners’ connection to the commercial world of the Indian Ocean. Thus, the trade connections that made the emerging elite wealthy also gave them access to the symbols of wealth and status. Bowls and plates were displayed in the public parts of merchants’ homes, the mihrabs (the niche that indicates which wall of a mosque faces toward Mecca) of Swahili mosques often had imported bowls set into the mortar around them, and the tombs of Muslim holy men had bowls set into their surfaces as well. This seems to speak eloquently of the value that the Swahili placed on these items, though it can be jarring to the modern observer to see ceramics with Chinese characters on them set into the tombs of Muslim holy men.
Another interesting observation made about the changing nature of ceramics in Kilwa and other sites is that bowls become more common, meaning they may have been associated with new dietary habits, possibly public feasting. Food and how one eats it is always a major marker of status, and as a new elite class emerged in Kilwa, they appear to have changed the way people ate. It is also possible that what they ate changed as well. Excavations at Swahili sites on the island of Pemba indicate that right around 1000, elites in Pemba began to consume a new staple food—rice. Commoners, however, continued to eat sorghum and millet, grains that are African domesticates consumed all over the continent. Sativa rice (the variety cultivated and eaten in East Africa) is not native to the African continent and presumably came to East Africa from Asia, most likely from the Indian subcontinent. Few aspects of human culture are more class and culture bound than what we eat, how we eat it, and with whom we eat. Thus, the appearance of new foods, along with new dishes from which to serve them, appears to mark the emergence of a new elite class even before this new class began to build new types of houses for themselves. It’s also worth noting that this new elite culture was not a package imported wholesale from across the ocean; rather, Kilwa’s emerging elites combined imported goods and foods from several regions of the western Indian Ocean as markers of their status. There is nothing in the record that would suggest that this new elite culture was transplanted to the coast from elsewhere.
The Second Phase
Kilwa’s role in the gold trade allowed for further expansion and elaboration of the elite culture that had emerged. Not until the thirteenth century did Kilwa come to have a significant number of stone buildings. Even after its rise to prominence the city continued to have more mud houses and fewer stone houses than those of the northern Swahili coast, though no one has a compelling explanation for this difference.
When Kilwa was at its peak, prominent merchants built their houses in stone. Sometimes these were two-story houses, but more often they were single story. Swahili patrician houses were elaborate affairs meant to house the merchant and his family as well as to contain areas where trade goods could be stored and processed. They also had apartments in them that appear to have been used to house visiting merchants and their goods.
These houses were as comfortable and well designed as any in the world. Houses at Gedi, which is much better preserved than Kilwa, had toilets in them, something few houses in the world could claim in 1300. They also had systems for channeling water from nearby wells to cisterns in the houses. Set into the bottoms of these cisterns were Chinese ceramic bowls, often with the Chinese character for “long life” inscribed on them.
In addition to these domestic structures there were a number of prominent public buildings in Kilwa. The most famous of these were the Husuni Kubwa and the Great Mosque. Every Swahili city had a congregational mosque where the city’s Muslims could all pray together on Fridays, but none could claim anything on the scale of Kilwa’s Great Mosque. Constructed in the twelfth century, Kilwa’s Great Mosque went through a major expansion in the reign of Sultan Hasan bin Sulaiman in the 1330s. It had a barrel-vaulted ceiling and a dome—probably the first true dome constructed in East Africa—which remained the coast’s largest until the nineteenth century. It also had an enclosure used exclusively by the sultan called a maqsura, allowing him to pray in kingly privacy. When the mosque was completed, it was the largest enclosed space south of the Sahara. Unfortunately, shortly after this expansion parts of the roof collapsed and were not fully repaired until the fifteenth century.
Hasan bin Sulaiman also constructed the Husuni Kubwa. Although the Great Mosque is a larger and grander version of a type of building found in any Swahili city, the Husuni Kubwa has no real parallels in other cities. It is a royal palace, built on a staggering scale, with multiple levels, courtyards, a water basin large enough that some archaeologists call it the swimming pool, vast storage areas, and a complex of apartments that all surround a large courtyard. It appears that Hasan bin Sulaiman was building a royal palace to be not only a visible symbol of his wealth and power but also as a place that might house many visiting merchants and their trade goods. The word husn from which Husuni derives is an Arabic word that refers to storage; Kubwa is Swahili for big. Thus, the Husuni Kubwa was the big storehouse. The courtyard surrounded by apartments seems to have been intended as a place where visiting merchants could live and trade as guests of the sultan. Its scale suggests that Hasan may have intended to dominate some part of the city’s trade. If that was his intention, however, he failed; the building was never finished. The usual interpretation of this is that the city’s patrician class was not pleased with this effort to infringe on their rights, so they put a stop to it. It is noteworthy, however, that Hasan’s failed bid for greatness took the form of a vast warehouse and self-contained market. Hasan was as much a merchant as a king, aspiring more to the status of a Venetian Doge than a Louis XIV.
That Hasan bin Sulaiman attempted to build so grand a palace was quite out of the ordinary in Swahili towns. In the ruins of many Swahili archaeologists have difficulty determining which house belonged to the ruler, and often they end up designating a house that is slightly larger than the others as a “palace.” Some Swahili towns treated their rulers with elaborate deference: some of these rulers were carried by slaves in a bed when they left their homes and proceeded by men playing siwas, side-blown horns made from elephant tusks. The last Swahili ruler of Zanzibar, the Mwinyi Mkuu, was approached by his subjects on their hands and knees. Some Swahili rulers were also seen as sources of magical power. When the Omani Arabs who took over Zanzibar in the nineteenth century imprisoned the Mwinyi Mkuu, it did not rain until he was released, a fact attributed to the Mwinyi Mkuu’s power. However, the Mwinyi Mkuu’s palace, though structurally different from other houses in that it had an audience chamber, is hardly vast. It’s only slightly larger than the typical merchant’s house. Even in places where kings were treated with great deference, they were usually selected by their peers rather than inheriting the office on strictly hereditary terms. Other Swahili towns, like those on the north coast of Kenya, have no tradition of kingship at all; they were oligarchies ruled by their patrician families. Similar patterns of oligarchic rule are found in other trade-oriented societies. For instance, merchant families dominated many Italian city-states as well as the Netherlands at its seventeenth-century peak. Intriguingly, there are also a few references to Swahili queens. When the Portuguese arrived in Zanzibar, the island was purportedly ruled by a queen, but it is not clear how common female rulers were, nor is much known about the general status of women on the coast during this period, though at least in historical times Swahili women had valued privacy but were not in any sense isolated or marginal members of society.
Kilwa, by contrast, had kings, but such evidence as there is suggests that they were not treated much differently from other patricians. The Arab traveler Ibn Battuta visited Kilwa in 1331 and reported having a conversation with the sultan (Hasan bin Sulaiman) in the street. A beggar, who asked the sultan for his clothes, interrupted their conversation. These the sultan handed over, earning him the respect of the crowd of onlookers as well as Ibn Battuta. This suggests that Kilwa’s rulers, even an exceptional character like Hasan bin Sulaiman, did not cultivate much of an air of mystery about themselves and instead seem to have been intent on using their wealth to curry favor in their subjects. If anything, Kilwa’s rulers—and most Swahili rulers for that matter—seem to have been first among equals in a ruling class rather than hereditary royalty.
Although the various cities of the coast were rivals and fought occasional wars with each other, they don’t seem to have been a terribly warlike people. Some accounts mention a holy war being waged against the peoples of the interior, and the cities and island, as was the case with Kilwa, were usually walled, presumably for defensive reasons. In general little is known about the relationship between the Swahili towns and the people of the interior. Some scholars see the Swahili as having very limited contact with the interior, mediated through nearby allies, whereas others think there may have been more extensive contact between the coast and interior. Major conflicts with the peoples of the interior occurred when drought and famine triggered movements of people that threatened coastal cities. The best known example of this occurred in the sixteenth century when the arrival of ten thousand marauding Zimba from the interior interrupted the struggle between Ottoman Turks and the Portuguese over the Swahili town of Mombasa. The Zimba, who had menaced many other cities on their way north to Mombasa, were most likely refugees from drought who had taken to raiding in an effort to obtain food. Believing that the Zimba were cannibals, the Turks fled the city in the only direction they could—toward the ships of the Portuguese—leaving the inhabitants of the city to face the Zimba on their own.
The town chronicles say little about the military exploits of towns and their rulers, though vicious intrigue and conspiracy seem to have been the norm. These were merchant princes, not war leaders. There was never a Swahili empire, and any effort from one town to dominate another was usually short lived. The coastal cities saw occasional military threats from the interior, but these cities usually depended on their walls, location, and allies in the hinterland to protect them. Occasionally threats appeared from the sea. In one case, during the thirteenth century, Sakalava raiders from Madagascar attacked the city and carried off a number of people. Later they made a second attack and were repelled.
GLOBAL ENCOUNTERS AND CONNECTIONS:
Kilwa’s Place within the Indian Ocean Trading Network
In 1498 raiders who could not be so easily repelled slipped by Kilwa without stopping. During the sixteenth century, as part of a broader expansion that included the discovery of the Americas, Iberian sailors entered the Indian Ocean by sailing around the southern tip of Africa. They brought with them maritime technology in the form of ships built to carry cannons and ideas about the use of naval power to control sea trade that, though common in the Atlantic and Mediterranean, were novel in the Indian Ocean. The first Portuguese navigator to reach the Swahili coast, Vasco da Gama, passed close enough to see Kilwa on his way north toward India. He had entered the Indian Ocean by sailing around the southern tip of Africa and was on his way to complete the first roundtrip voyage from Europe to India. The next Portuguese expedition into the Indian Ocean, led by Pedro Cabral in 1500, did stop briefly in Kilwa, but after a nervous and inconclusive meeting between Cabral and the sultan in their respective boats, Cabral went on his way. Da Gama returned in 1502 and forced the sultan to become a tributary of the king of the Portugal. In 1505 Francisco d’Almeida sacked the city, installed a new king, and built a fortress, leaving a small garrison behind to protect his puppet ruler. Perhaps most devastating to Kilwa, the Portuguese seized Sofala and took over that city’s gold trade. Portuguese power on the Swahili coast never reached much farther inland than the range of their ships’ guns, but Kilwa and other Swahili cities had no prior experience of naval bombardment, and the Portuguese, despite their small number, were formidable and determined soldiers.
Kilwa might have bounced back from this, but the Portuguese also seized Sofala, the source of the gold that had made Kilwa so rich. The Portuguese made a serious effort to control the trading system of the Indian Ocean, a policy that no Indian Ocean power had ever tried to implement on so grand a scale. By 1550 they had captured the most strategically placed trading cities of the Indian Ocean, save for Aden, and required that any ship sailing on the Indian Ocean purchase a cartaz, a sort of passport, or face attack from Portuguese ships. It was, in effect, a region-wide extortion scheme.
Most scholars think that the overall effects of the Portuguese effort to reshape the economy of the Indian Ocean were fairly small. There were never enough ships to enforce the rules about the cartaz, and merchants simply added the cost of dealing with the Portuguese to the expense side of the ledger and carried on with businesses. Although this may have been true of the broader region, the Portuguese did have a more profound effect on the Swahili coast than they did elsewhere. They were able to control the southern reaches of the Zambezi River and actually had territorial possessions in Mozambique. With their relatively poor hinterlands, Swahili cities were more dependent on sea trade than on the trading port of Malabar, so the diminution of their sea trade had larger effects for a place like Kilwa than it did for Cambay or Calicut.
In the seventeenth century Omani Arabs drove out the Portuguese. The Omanis had been building a trading state based in the port city of Muscat, and their influence reached from East Africa to India. Swahili elites were pleased at first but soon realized that the Omanis had come to stay. Under Omani control, Swahili cities were able to retain various levels of autonomy, but most had Omani governors imposed on them, and in the nineteenth century there was significant migration from Oman to the Swahili coast. Swahili elites became increasingly impoverished, and despite cultivating a strong sense of cultural superiority to the new arrivals from Oman, the elites were steadily marginalized.
The Omanis occupied the same fort in Kilwa that the Portuguese had used to dominate the town. By the eighteenth century the city, for reasons unknown, had been abandoned and a new town, Kilwa Kivinje, had emerged on the coast. Kilwa Kivinje, however, never rivaled its namesake for size, wealth, or power, though it had a brief boom as a slaving port late in the eighteenth century, supplying slaves to the Omani capital in Zanzibar and to the French sugar islands of Reunion and Mauritius.
Kilwa’s vulnerability to the Portuguese and the Omanis tells us something about both the benefits and risks of trade. Kilwa depended on its access to Indian Ocean trade networks for income as well as maintenance of its unique social order. Kilwa’s merchants traded African gold for the cloth and ceramics that not only distinguished the merchants from their social inferiors within the Swahili world but also distinguished the Swahili as a group from their non-Swahili neighbors. Although there are passing references to Kilwa merchants turning up in other parts of the Indian Ocean, they seem to have specialized in one small branch of the Indian Ocean trade—their control of the gold trade from Sofala. When the Portuguese, global leaders in naval warfare at the time, severed that link, Kilwa was dealt a fatal blow.
Unlike other Indian Ocean ports such as Aden or Malacca, Kilwa, with its dependence on the gold trade, was a one-trick pony. Aden, which was, if anything, less connected to its hinterlands than Kilwa, survived the Portuguese incursion easily. Aden sat at the entrance to the Red Sea and was a transit point for trade among the Mediterranean, the Persian Gulf, India, and East Africa. Severing a link in that system would be a problem, but not a fatal one. Other Swahili cities like Mombasa proved much more resilient than Kilwa. Mombasa had connections to the interior as well as to the islands of Pemba and Zanzibar, and it likewise fared much better under the Portuguese.
Seeing the Swahili as highly receptive to cultural influences from people with whom they came into contact is tempting. After all, one thousand years ago they became Muslim due to the influences of their engagement with the largely Muslim trading world of the Indian Ocean, adopted the Arabic script, and showed a real enthusiasm for the material goods of the Indian Ocean; however, their experience of the next five hundred years tempers that fact. Between 1500 and the present the Swahili have been exposed to Portuguese, German, and British rulers and missionaries espousing many different types of Christianity. Yet people on the coast, the Swahili, have remained resolutely Muslim. In the nineteenth century the British Empire’s expansion to include both East Africa and India encouraged large numbers of Indians to migrate to East Africa. There are now Hindu temples in almost every coastal city, but the Swahili have shown no interest in Hinduism. Although the Swahili language has absorbed some vocabulary from Gujarati and Urdu, it is East African Indians who have adopted Swahili as their mother tongue, not the reverse.
Unlike Kilwa, many other Swahili cities survived into the present day. Mombasa is Kenya’s second largest city and East Africa’s busiest port, but the Swahili are a minority, overwhelmed in a population swollen by migrants from the interior. Lamu lives on as a sort of Swahili version of Colonial Williamsburg, Virginia, but its elites are poor and much of the prime property belongs to Europeans. In the nearby town of Siyu Swahili patricians have not had the wherewithal to maintain their stone houses and have instead moved into mud houses; however, the status of owning a stone house remains, even when it has not been occupied for a century. The crumbling structures still stand in Siyu, and everyone knows which houses belong to which families. Swahili wealth and political power have vanished, but Swahili culture survives.
Kilwa’s status as a busy trading port with commercial connections that stretched across the western Indian Ocean meant it had many visitors. Few, however, put pen to paper to describe the city. An exception was Ibn Battuta (1304–1377), who was one of the world’s great travelers.
“A Visit to Zeila, Mogadishu, Mombasa, and Kilwa Kisiwani,” by Ibn Battuta
Originally from Morocco, Ibn Battuta spent much of his adult life on the road. He completed the haj, or Muslim pilgrimage to Mecca, as a young man and served as a qadi, or Islamic judge, first in Delhi and then in the Maldives Islands. He went to China, crossed the Sahara to visit Mali in West Africa, and made a voyage down the coast of East Africa to the Swahili coast, where he stopped in Kilwa.
Certain themes run through Ibn Battuta’s work. He tends to be interested in the way Islam is practiced in the places he visits, approving of practices that he deems orthodox while noting his disapproval of those he thinks contrary to proper Islamic practice. His other recurring interest is the generosity or lack thereof that his hosts show to him. He expected to be given lavish gifts and became a bit tetchy if the gifts received were not quite up to snuff. His memoirs make him sound like a fairly unpleasant person, but they are an invaluable resource for understanding the world of the Indian Ocean in the centuries before the Portuguese arrived. The following excerpt is from his description of East Africa, which he visited in 1331.
•Does Ibn Battuta seem surprised by anything he finds in Kilwa? Is he in an exotic foreign land, or is this familiar territory for him? What does this tell you about the world of Islam in fourteenth century?
•Ibn Battuta lists the names of a cluster of Islamic scholars who are in residence in Kilwa. Where do these people come from? What does this tell you about Kilwa’s oceanic connections? Why do you suppose they came all the way to Kilwa?
Then I set off by sea from the town of Mogadishu for the land of the Swahili and the town of Kilwa, which is in the land of the Zanj. . . .
We spent a night on the island and then set sail for Kilwa, the principal town of the coast, the greater part of whose inhabitants are Zanj of very black complexion. Their faces are scarred, like the Limiin at Janada. A merchant told me Sofala is half a month march from Kilwa, and that between Sofala and Yufi in the country of the Limiin is a month’s march. Powdered gold is brought from Yufi to Sofala.
Kilwa is one of the most beautiful and well-constructed towns in the world. The whole of it is elegantly built. The roofs are built with mangrove poles. There is much rain. The people are engaged in holy war, for their country lies beside that of the pagan Zanj. The chief qualities are devotion and piety: they follow the Shafi’i rite.
When I arrived, the Sultan was Abu al-Muzaffar Hasan surnamed Abu al-Mawahib [father of gifts] on account of his numerous charitable gifts. He frequently makes raids into the Zanj country, attacks them and carries off the booty, of which he reserves a fifth, using it in the manner prescribed by the Quran. That reserved for the kinfolk of the Prophet is kept separate in the treasury, and, when the Sharifs come to visit him, he gives it them. They come from Iraq, the Hijaz, and other countries. I found several Sharifs from the Hijaz at his court, among them Muhammad ibn Jammaz, Mansur ibn Labida ibn Abi Nami and Muhamma ibn Shumaila ibn Abi Nami. At Mogadishu I saw Tabl ibn Kubaish ibn Jammaz, who also wished to visit him. This Sultan is very humble: he sits and eats with beggars, and venerates holy men and descendants of the Prophet.
I found myself near him one Friday as he was coming away from prayer and returning to his house. A faqir from Yemen stopped him and said: “O Abu al Mawahib.” He replied: “Here I am, O beggar! What do you want?” “Give me the clothes you are wearing!” And he said: “Certainly you can have them.” “At once?” he asked. “Yes, immediately!”
He returned to the mosque and entered the preacher’s house, took off his clothes and put on others. Then he said to the faqir: “Come in, and take these.” The beggar entered and took them, wrapped them in a cloth and put them on his head. Then he went away. Those who stood by thanked the Sultan most warmly for the humility and generosity he had displayed.
His son and successor designate took back the clothes from the faqir and gave him ten slaves in exchange. When the Sultan learnt how much his subjects praised his son’s action, he ordered that the beggar should be given ten more slaves and two loads of ivory. In this country the majority of gifts are ivory: gold is very seldom given. . . .
Source: Ibn Battuta, “A Visit to Zeila, Mogadishu, Mombasa, and Kilwa Kisiwani,” in The East African Coast: Select Documents from the First to the Earlier Nineteenth Century, edited by G. S. P Freeman-Grenville, 31–32 (Oxford: Clarendon Press: 1962).
Chittick, Neville. Kilwa: An Islamic Trading City on the East African Coast Nairobi, Kenya: British Institute in Eastern Africa, 1974.
Freeman-Grenville, G. S. P. The East African Coast: Select Documents from the First to the Earlier Nineteenth Century. Oxford: Clarendon Press, 1962.
Gilbert, Erik, and Jonathan Reynolds. Africa in World History, 2nd ed. Upper Saddle River, NJ: Prentice Hall, 2008, ch. 7.
Horton, Mark, and John Middleton. The Swahili. Oxford: Blackwell, 2000.
Kusimba, Chapurukha. The Rise and Fall of the Swahili States. Walnut Creek, CA: Altamira Press, 1999.
Spear, Thomas. “Early Swahili History Reconsidered.” International Journal of African Historical Studies 33, no. 2 (2000): 257–290.
Ruins of Kilwa Kisiwani and Songo Mnara, UNESCO, http://whc.unesco.org/en/list/144.
World Heritage Site: Ruins of Kilwa Kisiwani and Songo Mnara, www.worldheritagesite.org/sites/kilwakisiwanisongomnara.html.
