The East African Coast, c. 780 to 1900 C.E.
Randall L. Pouwels
The Swahili of East Africa have one of the most venerable Islamic traditions in sub-Saharan Africa. Much as elsewhere on the continent, Islam made its first impressions through commercial exchanges between Africans and Muslims, though jihad never became the important instrument of conversion or enforcement in East Africa that it was in other places. Following the first conversions on the northern coast, the tenth through the fifteenth centuries was a time of steady and peaceful expansion southward in the wake of developing trade. By 1500, and as early as 1200 in a few locations, Islam had become the majority religion of coastal peoples. Arab, Persian, and even Indian immigrants played their parts in this process, yet it is false to see the emergence of Islam in East Africa as an extrinsic phenomenon. Rather, these were centuries when Islam was adopted, adapted, and internalized, as a coastal “African” religion. By 1500, Islam had become central to Swahili society and identity everywhere.1
The truth of the last statement is most apparent in the primacy given to religion in the Swahili struggle against Portuguese attacks in the middle period of coastal history, 1500 to 1800. Not only did coastal Muslims succeed in driving out their Christian nemeses, they renewed and added new foundations to their faith. Ties with the Middle East were strengthened, literacy assumed an unprecedented position, and Arabizing fashions began at this time. Most of these new developments again radiated southward from the northern coast, where the Pate region had become the focal point of Swahili culture and religion during these centuries.
Such tendencies were both augmented and changed in the nineteenth century when the Zanzibar sultanate gave new luster to everything “Arab.” In addition, all the sultans, especially Barghash b. Said, evinced a greater interest in enforcing more control over religion and law by appointing qadis, while maintaining closer ties with the heartlands of Islam. More people went on pilgrimage. More ʿulamaʾ sought training at the feet of scholars in Arabia, Mombasa, Lamu, and Zanzibar. The tariqas appear also to have played a greater part in many coastal people’s religious practices, which contributed to local religious tensions. Yet, by the end of the century the first encounters with European colonialism strengthened the sultanate and the position of Zanzibar as the new fount of coastal religion.
Foundations, c. 780 to 1500
The earliest and most detailed evidence so far discovered of an Islamic presence in East Africa has been at Shanga in the Lamu archipelago. Horton there excavated a mosque and Muslim burials dated between 780 and 850 C.E.2 From the time of its original construction to its abandonment early in the fifteenth century, this mosque went through twenty-five refurbishments, evolving from a tiny, “tent-like” structure, built from impermanent materials pegged to posts, to one made of a timber and daub, then to a porites coral-and-mud building, and finally to a substantial edifice made of coral rag and lime plaster. Continual reconstructions were done to correct the qibla line and to enlarge the mosque for a community whose Muslim population was growing.
From slightly later, archaeologists have found evidence of Muslim settlement of Pemba and Zanzibar. At what might have been Qanbalu (Ras Mkumbuu), Horton has excavated a large Friday mosque dated to about the tenth century. Dating from the tenth and eleventh centuries, mixed Muslim and non-Muslim burials have been unearthed at nearby Mtambwe Mkuu, while twelfth-century mosques bearing Kufic inscriptions have been discovered at Tumbatu Jongowe and Kizimkazi.3
Likewise, numismatic findings have provided clues to the earliest appearance and spread of Islam on the coast, and again the evidence is strikingly rich. Ninth-century silver coins have been exhumed at Shanga, bearing inscriptions of local Muslim rulers. Others of the later-tenth and eleventh centuries from Shanga and Mtambwe Mkuu strongly resemble Fatimid currencies, attesting to the later influences of the Red Sea region on the coast. Testifying to Kilwa’s later prominence, large numbers of Islamic coins minted there in the twelfth to fifteenth centuries have been unearthed at many locations.4
So far, this dating sequence suggests that Islam materialized on the northern coast late in the first millennium, then spread southward with the direction of later commercial development. The earliest mosque at Kilwa dates to the thirteenth century, but like Shanga it underwent continual expansion and renovation thereafter.5 Mosques and Islamic burials remained relatively few until the thirteenth century, but dramatic proof of the expansion and growth of Islam is provided at locales like Shanga and Kilwa, which saw continual augmentations and improvements to existing mosques, as well as new additions. By the fourteenth century, it is estimated, there were more than thirty communities that had at least one mosque, many with several. Furthermore, Islam had extended to the Comoro Islands and Madagascar sometime during this period, perhaps as early as the eleventh century.6
Here it is appropriate to inquire into the nature of early Islam on the coast and ask what changes it underwent. First, there is little doubt that it was a by-product of the growing incorporation of East Africa into the world of the western Indian Ocean, a process that had begun centuries before the Islamic era but that reached its greatest development between 1000 and 1500. This encompassed many peoples, some as far away as the headwaters of the Congo River and the fringes of the Kalahari Desert. The crucial vehicle of this process was trade. While many scholars in the past recognized this fact, they viewed Islamization in East Africa as solely the result of Arab commercial initiative, which included the founding of coastal sites like Shanga, Mombasa, Kilwa, and others. Contacts with other peoples of the Indian Ocean did play an important part in the mechanism of change; however, it is important now to recognize that to understand the conversion of coastal peoples and the nature of coastal Islam, the broader context of cultural and religious change among Africans has to be taken into consideration.
By the time the earliest Swahili settled locations like Shanga late in the first millennium C.E., their Eastern Bantu—speaking ancestors already had experienced centuries of steady expansion and migration from the Western Rift region of East Africa, where they originally dwelt. Moreover, change through cross-cultural borrowing was not new to them. Their ancestors had established contacts with neighboring eastern and central Sudanic and Southern Cushitic-speaking societies, and they had been exposed to a wide range of physical environments. Though originally a Neolithic people who lived from a subsistence tradition that included hunting and planting in well-watered areas, they had habituated to the drier conditions of East African savannas by learning livestock-breeding, cereal cultivation, and iron-making techniques.7 Significantly, most of these skills had been acquired from non-Bantu-speaking neighbors. Arriving on the coast c. 100–350 C.E., their descendants the Proto-Northeast-Coastal Bantu-speakers (PNEC) merely continued the already ancient habits of cultural adaptation to new conditions. Through the normal process of dialect and language formation, between 350 and 650 C.E., the PNEC-speakers split into three major language communities, including the Proto-Sabaki (PSA). Finally, between 650 and 1000 C.E., the PSA cluster split into five dialects, including the ancestors of the Swahili, the Comorians, the Mijikenda, the Elwana, and the Pokomo. Throughout this entire period, as PNEC gave rise to PSA, and as PSA gave rise to Proto-KiSwahili (PSW), the long-familiar pattern of adapting their cultures to new surroundings—this time to a coastal environment—continued. The dynamics of this process emerge clearly from the kinds of lexical innovations introduced into PNEC, then PSA, then finally into the Proto-Swahili between 100 and 1000 C.E., as shown in table 1.8
Especially noteworthy is that the number of innovations increased with time, indicating that the pace and extent of environmental adaptation accelerated.
It must also be emphasized that these Africans’ acculturation to the coast was not restricted to their physical surroundings, but included the human environment as well. Cultural vocabulary from other Bantu- and Cushitic-speaking communities continued to be borrowed, but for the first time in their experience their environmerit included exogenes. Therefore, some early innovations, indicated in the table by starred items, originated in Persian and Indian languages, suggesting that even before the Islamic era, coastal Bantu-speakers’ experience of their surroundings included significant foreign components. It is important here to note that, though these components were extrinsic, their inclusion was intrinsic to the larger processes of adaptation and change that coastal Africans had experienced for many centuries. The Islamization of the Swahili-speaking peoples was innate to their emergence as a distinct culture.
Table 1
PNEC (100–350) |
PSA (350–650) |
PSW (650–1000) |
God |
Doum palm |
a free person |
rice |
Borassus palm |
slave (two terms) |
island |
*water pot |
*barbarian |
short rains |
coconut leaf |
*lady (bibi) |
cloth |
coconut juice |
*gentleman (bowana) |
*mush (ugali) |
shark |
seagoing ship |
*steel |
ship mast | |
*fishing net |
*fortified place (boma) | |
*rope |
*canoe | |
*fishing net | ||
*cuttlefish | ||
*oyster | ||
rock cod | ||
fish weir | ||
*incense |
That conversions to Islam occurred early in the emergence of a Swahili civilization comes as no surprise, a fact demonstrated by Horton’s discovery of a late-eighth-century mosque at Shanga. For starters, Ehret’s evidence suggests that the Swahilis’ Eastern Bantu ancestors (1000–500 B.C.E.) already had a concept of a Supreme God; however, the focus of their daily religious “observance” was on spirits that were considered to be nearer than this Creator God. The innovation of a new term for God (mulungu) in Northeast-Coastal Bantu further suggests that shifting to a new religion in which God played a central role would have demanded relatively little intellectual change—something closer to a reorientation of spiritual solemnization than to a conceptual shift.
Reorientation, of course, did not require wholesale abdication of ancient beliefs and rituals. As in other parts of the continent, for centuries coastal settlements harbored “mixed” populations of believers and unbelievers, while Muslims themselves continued honoring the ancestors and fearing malevolent forces, both Muslim and non-Muslim.9 Verification of the appearance, mixing, and enlargement of Muslim communities from archaeology has been cited already. Arab sources reveal how gradually mixed populations gave way to communities in which the majority had converted. In the tenth century, Masudi was the first to register the presence of Muslims among the coastal “Zanj” population, and it was only at Qanbalu that Muslims seem to have had a majority by that time. Even in Qanbalu, though, Muslims and “infidels” lived side by side, and where they had the majority, they remained surrounded by non-Muslim neighbors (in the significantly named ard al-zanj, land of the zanj—blacks—the East African equivalent of the Sudanic Bilad al-Sudan) against whom they sometimes waged jihad, as reported by Ibn Said in the thirteenth century and Ibn Battuta in the fourteenth.
Coastal Islam developed its own distinctive flavor right from the beginning, one that mixed local with universal practices. The Shanga mosque provides exquisite archaeological detail about how this happened.10 First, not only were the early mosques quite small, but, as Horton’s evidence strongly documents, their appearance was entirely within a context rooted in a non-Muslim, presumably Bantu-speaking, past, situated as it was in a Mijikenda-like central enclosure that appears to have been restricted to community, clan, and ritual (including non-Islamic) use. Even long after they had become the majority, Shangas Muslims continued building their mosques and tombs in their own style and from natural local materials like coral, coconut by-products, and a particularly sticky form of mud (kinyokae). Only after the eleventh century did foreign influences creep into some details of mosque architecture at Shanga, Kizimkazi, and Tumbatu. Also, Arabic and European notices attest that local Islam continued to include practices that had non-Muslim origins both before and after 1500. Ibn Battuta observed scarification among Muslims at Kilwa, for example, and Figueroa mentioned seeing uncircumcised Muslims at Kilwa. Santos was only the first of many who observed widespread coastal practice of sorcery, and in the seventeenth century Dapper reported that the sultan of Malindi relied on Muslim advisers (walimu) who employed necromancy.11 Even today, practices similar to these remain common among African Muslims everywhere (see chapter 21).
Were there foreign influences on early coastal Islam? Certainly there were, especially in the great commercial centers like Mogadishu (see chapter 11), Mombasa, and Kilwa, as well as at secondary locations like Shanga, Mozambique, the Comoro Islands, and Madagascar. Archaeologists have found many varieties of imported pottery, particularly at northern sites in the earlier phases (to 1100), then at southern sites later. Among these are pre-Islamic Sassanian and post-Islamic Sassanian-Islamic, lead-glazed, monochrome, and sgraffiato wares from the Persian Gulf; several kinds of Indian pottery and glass beads; and post-thirteenth-century black-on-yellow wares from Yemen. The testimonies of local traditions that tell of “Shirazi” and “Debuli” immigrants provide echoes of this early era of dominance of the coastal trade by merchants from Siraf and Sind.12 The inclusion of Indian and Persian loanwords in PSA and PSW lends support to the evidence. Arab influences came later, after the eleventh century, when Fatimid initiatives reinjected energy into the Red Sea trade eastward to India and the East Indies, and northward to Byzantium and the Mediterranean. On the coast, evidence of this development was found by Horton in the Fatimid-like coins minted at Shanga and Pemba, cited above. However, the most enduring Arab Islamic influences came from Yemen after the thirteenth century, the hallmark of which was black-on-yellow pottery. This also was an age conspicuous for the first of three major waves of migrations of shurafaʾ families from Yemen and the Hadhramaut that would occur in subsequent centuries. The most important of these migrants was the Mahdali clan, who eventually founded the Abul-Mawahib dynasty at Kilwa and reigned over that city’s rise as an influential center of trade and Islam in the fourteenth century.
A case already has been made for the essentially local flavor of early coastal Islam, but there were influences from foreign Muslims, particularly immigrants. There is evidence, much as in the western Sudan, that some minority forms of Islam were exported to East Africa in the early centuries. An eighteenth-century account tells of a putative migration of the Ibadi Julanda family from Uman to the coast in the eighth century following their defeat at the hands of the Umayyads. Horton believes he has found evidence that Ibadis were present at Tumbatu, and Wilkinson has discovered clues that Ibadis were present at Kilwa and engaged in proselytizing among the local population.13 Other clues suggest that Shiʿis also found their way to the coast. It appears unlikely they ever managed to become a majority anywhere, though testimonies of their presence lingered well after 1500.
Barros tells of the “Emozaidi,” a Fiver Shiʿi sect of Yemen who, he claims, emigrated to the coast and eventually settled inland among the “Baduys.” Dapper specifically cites Malindi as a Zaidi stronghold.14 Elsewhere, he alludes to other Shiʿi groups who settled the coast after being driven from their lands because they were of the “cult of Hali [sic].” Santos claims that many coastal Muslims of the sixteenth century followed the “Persian” sect, who were said to have been “at great variance” with the official sect of the “Turks.”15 Given the prominence of Oman, Siraf, Kish, Yemen, and Fatimid Egypt in the early coastal trade, the likelihood that heterodoxy existed in these regions at that time seems high; therefore, these accounts have some ring of truth about them. On the other hand, it seems likely that some Portuguese confused local traditions of “Shirazi” origins with sectarian differences, or they simply could have misunderstood the local Islamic practices to be manifestations of Ibadism or Shiʿism. Whatever their truth, the long-term effects of early Ibadism and Shiʿism on coastal Muslims are nebulous.
The decline of heterodoxy in the Islamic heartlands after 1100 and the continued importance of the Red Sea, Yemen, and the Hadhramaut in the commerce of the western Indian Ocean, however, helped assure the conversion of most coastal Muslims to Shafiʿi Islam. (Sind might have been an early orthodox influence on the coast, also, since by the ninth century Daybul appears to have become a source of “official” religious propaganda.)16 When Ibn Battuta visited the coast in 1331, he mentioned encountering only Shafiʿis, and in the sixteenth century Santos observed that many followed “the Xaphaya sect.”17 If Shiʿi Islam established a presence on the coast and hung on for several more centuries, it appears that Shafiʿi Sunnis had a lasting influence on the major commercial centers. The early connection Islam had with trade in East Africa clearly continued after 1200. At Shanga, for example, as Horton has suggested, the piety of the commercial classes lay behind the construction of two new neighborhood mosques in the fourteenth century.18 The emergence of a wealthy and influential commercial class contributed to the proliferation of neighborhood mosques all along the coast and to the popularity of the Shafiʿi madhhab. It was trade that attracted clans like the Mahdali to the major entrepôts, and because clans like the Mahdali demonstrated an ability to bring prosperity wherever they settled, the example they set of lives devoted to pious Shafiʿi observances would have greatly influenced many coastal Africans and explains why they were given dynastic titles like “The Father of Gifts (Abul-Mawahib).”
What was the nature of this influence? Aside from creating prosperity, the merchant-sultans of Mogadishu and Kilwa assumed responsibility for constructing and expanding community religious buildings like the Friday mosque at Kilwa. From Ibn Battuta’s account, both cities had sultans who governed by Sunni principles by including wazirrs, qadis, and muhtasibs in their court entourages. Abu al-Muzzafar Hasan, the Mahdali sultan visited by Battuta, was noted for his many charitable gifts, as well as his veneration of “holy men” who frequented his court from as far away as the Hijaz and Iraq. Several from Kilwa’s ruling family personally received training in the religious sciences. They waged jihad against the “infidels” of the mainland opposite the island community, and we are told by the author of the Kilwa Chronicle that one of the sultans was “martyred” in such a raid.19
The paucity of evidence makes it is difficult to comment extensively on the level of scholarship before the eighteenth century; however, the few indications available point to its relative superficiality. At least in the major commercial centers, ulama apparently could be found. Mogadishu achieved an early reputation as a center of learning, and Ibn Battuta reported that many “lawyers” resided there, where disputes were settled by qadis according to the Shariʿa. At Kilwa, he encountered numerous ʿulamaʾ and shurafaʾ who were revered and patronized by the sultans. Abuʾl-Mahasin mentioned a qadi from Lamu whom he described as “a man of great erudition in the law according to the Shafiʿi rite.”20
Our sources are equally inconclusive concerning the extent to which Arabic was understood and used prior to the eighteenth century. It was understood by a few, perhaps as a second language, at Kilwa, Malindi, and Mogadishu, while religious texts were known and prized among the ruling classes.21 On the other hand, spoken and written Arabic was hardly known even among the elite in remote locations. General illiteracy is what one would expect since KiSwahili was not yet a written language, and religious instruction seems to have been available only for those who could manage a voyage to Arabia.22 When some could go, their primary purposes were for trade or pilgrimage (hajj). It is impossible to speculate knowledgeably even about this, although Abul-Mahasin’s qadi comes to mind since he was met while on the hajj, and it should be remembered that the sultans of Kilwa took advantage of their pilgrimage for instruction. Nothing in any sources prior to the eighteenth century mention locally available forms of religious tutelage.
A very telling indicator of Arabic literacy is that there was no significant borrowing of Arabic loanwords into the Swahili language before at least the seventeenth century. According to Nurse and Hinnebusch, this was true not only for cultural taxonomy, where Arabic later made an impact—that is, for terms relating to law, literature, and religion—but was especially the case involving basic vocabulary.23 One is forced to draw the conclusion that, to whatever extent Arabic was known, until the seventeenth century it remained confined to first-generation immigrants and to the few who obtained religious tutelage abroad. It appears that Arabic and Swahili coexisted by performing parallel roles: for most, Swahili remained the language of daily discourse, while Arabic, as a “canonical” language, remained confined to religion and correspondence.
The Middle Period: 1500 to 1800
When without warning Vasco da Gama appeared off the East African coast in 1498, he came as the representative of a king and people who were interested in commerce and religious crusade, in that order. With a view to raking profits from the fabled gold trade, Sofala, Mozambique, and Kilwa were subjugated and fortified in 1505. The expected riches failed to materialize, so in 1511 Mozambique and Kilwa were abandoned, and almost the entire focus of Portuguese commercial attention shifted to India and the Far East. As for crusade, initially most of the coast was affected but little. However, coastal Muslims and Portuguese Christians immediately assumed positions of mutual distrust, and the Portuguese, following practices that long had existed between Muslims and Christians of the Mediterranean, perpetrated several acts of egregious brutality. In one notorious episode, they attacked and scuttled an entire shipload of Muslim pilgrims.24
Wherever soldiers and sailors went, missionaries accompanied them in the global Christian effort to outflank Islam. In the fleet of 1500, eight Franciscan friars made the crossing to India, and the first church was built at Cochim in 1503.25 Several religious orders established mother houses in Goa with a view to bringing the sign of the cross to Indian Ocean peoples, and Mozambique became a host to Jesuit and Dominican friars who soon began proselytizing among Muslims and non-Muslims of the southern coast and the Zambezi River valley. They largely ignored the northern coast until the late sixteenth century, except for sporadic attacks on Mombasa, Ozi, and Pate. However, in 1516 Sultan Selim (“the Grim”) captured Egypt for the Ottomans, and Suleyman the Magnificent’s Muslim counteroffensive in the Red Sea and the Indian Ocean began in the 1520s. Against this background, the situation on the coast played itself out and preordained a determined Christian riposte.
A problem for the Portuguese was their failure to capture Mogadishu, Zeila, and Aden. Combined with the extension of Ottoman sovereignty over Egypt, the Red Sea, and the Hijaz, this gave heart to the coastal people’s tendency to seek new champions and sources of reassurance from the religious epicenters of the Red Sea and Arabia during the dark, humiliating centuries of “infidel” rule. Many who opposed the “infidels” naturally considered themselves to be the subjects of the Ottoman sultans and said the khutba in their names. Throughout much of the sixteenth century, numerous bloody, closely fought battles on land and sea blazed between Muslims and Christians from Ethiopia to Goa. By midcentury, especially with the defeat of Ahmad Gran’s jihad in Ethiopia, it seemed that the Portuguese and their Ethiopian coreligionists had seized the high ground. However, between the 1540s and 1560s, fleets from the Red Sea region attacked Ormuz, Calicut, and Malabar.26 The east coast seems not to have been directly involved in this fighting; however, the threat to the Portuguese remained throughout the century, and twice, in 1588 and 1592, the Ottomans attempted forays against Pate and Mombasa; they were, however, defeated.
Pivotal developments resulted from these occurrences. First, the old links with southern Arabia and the Red Sea were sustained, extending northward from Madagascar, Angoche, and Grand Comoro Island as far as the Straits of Mecca (the Red Sea). Besides the Swahili (from the “coast of Malindi”), apparently “Arabs from Mecca” played an important part in this linkage, and for this reason, according to Portuguese accounts, Islam and coastal civilization were making new inroads everywhere in the sixteenth and seventeenth centuries.27 The slave trade from Madagascar and the southern coast seems to have grown, a crucial fact because enslavement for Africans often meant their transportation to Arabia and conversion to Islam.28 For this reason, in 1620, the Inquisition tried to stop Arab and Swahili merchants from trading in slaves.29
The importance of these commercial ties was, if anything, augmented because they assumed a new significance with the decline of Kilwa Kisiwani as a commercial and religious center. Pate stepped in as one of the key elements in the chain linking the coast, via Mogadishu, and the Red Sea. Situated just off the northern coast of present-day Kenya, Pate was a natural way station of the coastal trade, and its friendly ties with powerful mainland peoples added greatly to its middleman position in the supply of ivory and other animal by-products, most of which were supplied by Oromo and Aweera hunters. These amicable relations also permitted residents to farm extensively in mainland areas in relative safety, producing grains like rice and millet in exportable quantities. This made Pate not only a major funnel of the slave trade, but a substantial consumer of slaves as well. This dual status both as a producer and as a crucial link in the coasting trade, which put it in direct contact with the southern coast on the one hand, and the Red Sea and Southern Arabia on the other, made Pate inviting for immigrants.
Probably beginning after the defeat of the jihad of Ahmad Gran in 1542, and extending into the next century, Pate became the new home for various, high-profile Muslims who, besides having trade connections throughout the Indian Ocean as far as Java, strongly adhered to the Alawi tariqa of the Hadhramaut. Among these were a number of sharif clans and others noted for their learning and piety, the most noteworthy of whom were the al-Husayni, the Jamal al-Layl, the Abu Bakr bin Salim, and the al-Masila Ba Alawi clans. Many of these undoubtedly arrived directly from southern Arabian homelands, while others, according to Martin, arrived on the coast after having fought unsuccessfully in the jihad of Ahmad Gran.30 Not only, therefore, were they active in the slave trade, according to Portuguese accounts, but they introduced the militancy of Gran’s holy war to Islamic resistance to the Portuguese Christians in East Africa. In Pate, these clans settled in their own quarters, one of which was significantly named Inati, after the Hadhrami town of their origins. In 1569, Barreto already alluded to the presence of a Muslim “priest” at Pate whom he reported to have already achieved the status of being the “principal one” for the entire coast. The significance and activities of Pate’s shurafa, however, were articulated with the greatest precision by Father Thomas do Domingos in 1630: “Ships come to this island of Pate, on their way from Mecca to the Island of San Laurenco [Madagascar], with sharifs [Xarifos] who are their qadis [Cassizes], who go there to spread their faith and to obtain many Malagasy, native Gentiles, to take to Mecca and to make them Muslims [Mouros].”31
The Portuguese religious offensive began in 1560 during the height of the Muslim attacks against their Persian Gulf and Indian bases. Pope Paul created the See and Archbishopric of Goa, and Cochim and Malacca each became bishoprics, and two inquisitors accompanied the fleet of 1559–60.32 Augustinians shouldered the task of leading the offensive, establishing monasteries at Tana, Cochim, Chaul, Malacca, Macao, Isfahan, Muscat, Bassein, Goa, and Ormuz.33 On the East African coast, the response was a little delayed, but in 1598 they established the convent of Santo Antonio near the new stronghold (Fort Jesus) they were building in Mombasa.34 Soon, three new residencias followed in Faza, Pate, and Zanzibar, and the brethren were occupied in “converting many to the faith of our Lord.” Initially, successes were few, but in time the friars claimed to have had a mixed congregation at Mombasa of four thousand, and moderate success among the neighboring, non-Muslim population.35
The fact that a few Muslims (mostly women) numbered among the converts was not the only vexation the Augustinian and Dominican brothers caused.36 At Pate, the Augustinian residencia was forced on the city in 1598, very much over the pleadings and objections of its sultan, and one can easily imagine that they were no more welcome elsewhere. Even their allies, the sultans of Malindi, complained numerous times of interference by missionaries, probably not unlike the intolerant behavior exhibited by Father dos Santos at Mozambique.37 By the 1620s, it appears that inquisitors and “Pais do Christaos” from India were active on the coast to shore up laxity among Christians and to take action against the Muslim slave trade. Local fear of the inquisitors was high enough that guarantees against them were added to a treaty between the Portuguese and the citizens of Pate, Lamu, and Siyu in 1636.38
According to Dapper, coastal Muslims “everywhere” revolted and “made every effort to put themselves under the protection of the Turk” because the Portuguese would not allow them freedom of religion.39 Above all, it appears that the shurafa of Pate were at the center of this resistance, and Pate itself remained a major thorn in the side of the Portuguese throughout the seventeenth century. The church there was subject to occasional clandestine desecrations, and priests were surreptitiously poisoned; some of these victims were forced to return to Goa to expire.40 From Pate, resistance spread elsewhere. Ave-Maria presents an entire list of Augustinian padres who died, many of them murdered, at Zanzibar between 1649 and 1694.41 At both Pate and Zanzibar, the churches were attacked and eventually had to be abandoned. Apparently, only the ones at Mombasa and Faza remained beyond 1650.
It is important to note that the nature and extent of leadership by the Hadhrami immigrants were not expressed only in the form of resistance: indeed, after 1600 the beginning of a religious renaissance blossomed in the Lamu archipelago. Thanks especially to the shurafaʾ, a scholarly tradition in the advanced religious sciences was established locally for the first time, though this remained an avocational path closed to most ordinary Swahili until the nineteenth century. An even more momentous Hadhrami contribution was their origination of a written coastal literature. At first, religious didactic and artistic works were written in their native Arabic, but as they became indigenized, coastal scholars of Hadhrami background began writing in KiSwahili; consequently, the earliest written poetry from the coast dates from the middle of the seventeenth century.42 One by-product of this was the development of a handicraft industry that produced copies of religious texts of very high quality at Pate, Lamu, and especially Siyu in this period.43
Thanks chiefly to the Hadhramis, the Lamu archipelago thus became the religious and cultural heartland of the coast between 1550 and 1800, and the influence of the Lamu region and the Pate shurafaʾ became widespread. After having settled in Pate and its immediate neighbors for several generations, sharif families further migrated to other parts to the south, taking their spiritual charisma and scholarly traditions with them wherever they relocated. By the 1700s, new ʿAlawi ruling houses appeared all along the coast—at Ozi, Vumba Kuu, Tumbatu, Zanzibar, Kilwa, the Comoro Islands, and elsewhere—all claiming to have connections with Pate.44 Nurse and Hinnebusch have found further evidence of this: according to their monumental study of the history of the Swahili language, the dialect of the Pate-Lamu region had a large preponderance in the southern coastal dialects at that time—indicating a southward migration of a “large or prestigious group of [northern] immigrants” from the Pate region.45
The Nineteenth Century
The final expulsion of the Portuguese from the coast north of Mozambique in 1728 brought yet another new era of Arabization and religious developments to East Africa. In some respects, it continued changes that had been occurring over the previous two centuries, especially those involving the intellectual and religious importance of the northern coast and migrants from southern Arabia (another significant migration occurred in the 1800s). Above all, the Arabization of Swahili culture and religion was the hallmark of the nineteenth century. Of course, in some respects this had begun earlier, perhaps even as early as the fourteenth century, and certainly by the seventeenth century. Yet it seems to have been the arrival of the Omanis in the eighteenth century and the establishment of the Zanzibar sultanate in the 1820s that really made the difference.
Prior to the (effective) relocation of the sultanate of the al-Busaidi from Oman to East Africa, the observation of strict religious practices and the administration of the shariʿa appears to have been fundamentally an informal and haphazard affair in the majority of coastal centers. In most instances, local ʿulamaʾ and other persons variously educated in law and theology (for example, wanazuoni) served in merely advisory capacities; local rulers reserved for themselves the actual rights of adjudication.46 This situation changed gradually under the first three Busaidi sultans.47 While the sultans retained the right of final appeal, in a growing number of instances through the nineteenth-century heyday of their rule, adjudication was delegated to court-appointed qadis. The first two sultans, Sayyids Said (1832–56) and Majid (1856–70), appointed two official court qadis, one Ibadi, for their Omani subjects, and one Shafi i, for the Swahili majority. There also appear to have been other ʿulamaʾ present at their courts who served in other advisory capacities. Official qadis like Shaykh Muhammad b. Juma al-Barwani and Shaykh Muhyi ad-Din al-Qahtani al-Waʾil sometimes were required to undertake diplomatic commissions.
Elsewhere, as in Mombasa, local elders were permitted to nominate their qadi, who then was officially confirmed in office by the sultan. The third sultan, Barghash (1870–88), extended and strengthened the religious leadership of the sultanate in coastal life far beyond that of his predecessors. Disposed temperamentally more toward religious idealism than to pragmatism, in the early years of his rule Sayyid Barghash displayed a much greater tendency to follow the exhortations of his ʿulamaʾ and to take an independent line from the “advice” of the British consul. By 1875, however, circumstances had forced him to rely on British support. That, and a desire to augment and streamline his governing apparatus, led him to appoint many more officials throughout his East African dominions, including additional religious functionaries not only in Zanzibar but in all the major coastal towns. Besides local walis, Barghash and his successors directly appointed qadis for Lamu, Mombasa, Bagamoyo, Lindi, and Kilwa. Perhaps most important, Sayyid Barghash provided annual passage to Mecca for indigent pilgrims, a fact that won for him the accolade of the “Harun al-Rashid” of East Africa.48
The practice started by Sultan Barghash became established under his successors. Naturally, the colonial administrations added their own wrinkles. While German rule lasted, they directly appointed administrators (among them, the qadis) and limited their authority. The British preferred indirect rule: they tried to preserve local customs and judicial institutions as much as possible—wherever they were consistent with “civilized” behavior. In Zanzibar and the coast, for the most part this meant support for the existing system of qadis courts.49 However, there were changes. For one, the authority of the qadis courts was gradually curtailed to extend only to matters of domestic and family law; the more serious criminal matters were left to secular administrators, including mudirs, who were appointed either directly by the colonial authorities with the approval of the sultan (in Kenya and Tanganyika), or by the sultan with the approval of colonial authorities (in Zanzibar). There were also higher tribunals to handle the most serious matters and to act as courts of appeal: these were “mixed” courts, consisting half of Muslim judges and half of British jurists, and they administered versions of the Indian civil and penal regulations.
The more important aspect of the colonial system is that it further institutionalized the formal nature of the system started by Barghash. While some ʿulamaʾ refused positions and refused to cooperate with this application of kafir regulation, many others continued to go along with it actively and accept positions whenever they were offered. Thus, for example, the British formalized the positions of chief qadi for Zanzibar and Kenya, and an entire procession of Mazrui shaykhs was none too proud to accept the pay of their colonial masters for serving them in this capacity.
But it was not only the sultanate that transformed religious life at Zanzibar. The new merchant and landowning class, wealthy from the plantation and trading economy of the coast, contributed to the religious revival of the nineteenth century. As documented by Berg and Walter (1968) and Sheriff (1992), the era saw a dramatic increase in the number of mosques, mosque officials, Quran schools, and religious endowments. It was the new “bourgeoisie” who were responsible for this “unofficial” expansion of Islam, driven as they were by piety and a desire to leave named memorials and endowments for their descendants.50 Though many of the new mosques were built by Ibadis principally for their use, Zanzibar and other locations seem to have enjoyed a pervasive atmosphere of sectarian liberality, and often Shafiʿis were allowed by their Arab and Indian builders to enjoy the use of these new religious monuments. Moreover, powerful clans like the Barwani and the Harthi could afford to employ their own religious advisers, thus increasing the number of opportunities for employment among the learned class.
The older, established centers had more than their quota of wanazuoni and ʿulamaʾ, but Zanzibar nevertheless attracted some of the ablest scholars of Mombasa, the Lamu region, the Comoro Islands, and the Benadir towns. Mombasa was home to the distinguished Mazrui ʿulamaʾ, mostly from the line of Shaykh ʿAli b. ʿAbdallah Mazruʿi. The Comoro Islands contributed many scholars, the most notable of whom were scions of Hadhrami clans who had migrated there from Pate or directly from Arabia. The most noteworthy of these was Sayyid Ahmad b. Abu Bakr bin Sumayt, who served as the qadi al-qudat for all the sultans from the time of Barghash until his death in 1925, as well as Sh. Hasan b. Yusuf Mngazija and Sh. ʿAbdallah b. Wazir Msujini. From Lamu came Sh. ʿAbdallah Bakathir, Sayyid Ahmad’s contemporary; Barawa and other Benadir towns contributed several distinguished ʿulamaʾ, among them Sh. Muhyi ad-Din, Sayyid Said’s first Shafii court qadi, and ʿAbdul-ʿAziz b. ʿAbdul-Ghani al-Amawi. The greatest source of information on the Shafii ʿulamaʾ of East Africa over the past 175 years, Sh. ʿAbdallah Salih Farsy, provides biographic information on more than 170 individuals, most of whom were connected with Zanzibar at some time in their lives.51 Prosperity and the chance of finding employment made Zanzibar an attractive destination for religious scholars from all over the region.
Given these conditions, throughout the century the standard of religious education and scholarship continued to wax higher. Quran schools offered basic instruction to anyone who desired it, and most ʿulamaʾ offered advanced training in some specialized science. The quality of religious education benefitted especially from men like Sh. ʿAli b. ʿAbdallah Mazruʿi, Sayyid Ahmad bin Sumayt, and Sh. ʿAbdallah Bakathir who traveled to southern Arabia and the Hijaz to receive training that they then passed on to students. Unfortunately for a wider dissemination of advanced teaching, in most cases the shaykhs of “Arab” families such as the Mazrui, bin Sumayt, Maawi, Husayni, and Mandhiry offered higher-level lessons only to select scions of “Arab” clans. They provided such classes in the privacy of their homes and only at special hours. Little or no such instruction was given to women; once women had been trained to memorize the Quran, no further efforts were made to include them in the private classes—though one catches a glimpse in Farsy’s work of an occasional exceptional daughter of a shaykh who managed to learn a lot through eavesdropping.
The texts these shaykhs taught were for the most part, as observed by Becker, the standard Shafiʿi legal and theological works that formed part of the “Indian Ocean corpus.” Most of these were produced in Cairo, the Hadhramaut, Yemen, and Java, and for the most part they were based on Nawawi, Ibn Hajar, Ghazali, and Ba Fadhl.52 But these scholars also authored religious tracts of their own: Sh. Muhyi ad-Din composed a theological work (on tawhid, proofs of God’s singularity), a commentary on Nawawi’s Minhaj al-Talibin, and a short work on Arabic morphology. Perhaps the largest corpus of locally written work was that of Sh. ʿAbdul-ʿAziz b. ʿAbdul-Ghani, whose writings dealt with a variety of subjects, among them tawhid, Arabic grammar, and rhetoric. Interestingly, he seems to have avoided anything on law (fiqh).53
Not only did the nineteenth century see growing depth to local standards of literacy and scholarship, but in some respects the new standards became more popular and widespread. This was due to the more significant part the tariqas had begun to play. For several centuries, beginning with the era of the Hadhrami immigrations after the sixteenth century, the Alawiyya already had made contributions toward this end; however, that sodality being unusually devoted to venerating the shurafaʾ, it tightly restricted religious education only to descendants of the Prophet and a few other families. In the Benadir towns and the Comoro Islands, on the other hand, the Qadiriyya and the Shadhiliyya orders had grown popular over the previous centuries, so when ʿulamaʾ from those places turned up in growing numbers in East African locations, these religious orders arrived with them. The end of the nineteenth century and the early decades of the twentieth thus witnessed the appearance of a new type of missionary and teacher in East Africa: this new breed devoted themselves to the spiritual and educational needs of disadvantaged Muslims and non-Muslims. Most, but not all, were members of the two brotherhoods named above, the most noteworthy being Shaykh ʿAbdul-ʿAziz b. ʿAbdul-Ghani and Shaykh Uways b. Muhammad al-Qadiri from Somalia, Shaykh Yahya b. ʿAbdallah (Sh. Rumiyya) of Bagamoyo, Sayyid ʿAbdur-Rahman b. Ahmad al-Saqqaf of Siyu, and Habib Salih b. ʿAlawi Jamal al-Layl of Lamu.54 Tariqa shaykhs like Shaykh Rumiyya and Shaykh Uways, through their efforts to spread the teachings of their religious orders, aided in the conversion of neighboring non-Muslim peoples to Islam. But more important, all these men trained a new class of Swahili shaykhs for whom any education in the religious sciences would have previously been unobtainable.
The coming of the popular tariqas had two other results: it helped to popularize Islam and it contributed to the ethnic tension and conflict that was characteristic of the colonial era; it still echoes today in East Africa. Most of the adherents of the tariqas managed to get some degree of advanced education; thus, today one hears of many Swahili and non-Swahili wanazuoni wakatikati, wanazuoni wadogo, and walimu—“middle-level” and “little” ʿulamaʾ of former times whose religious education, though sometimes considerable, did not quite rival the scope of that of the “great” shaykhs. Many of them mixed elements of literacy with related sciences such as amulet-making, numerology, and falak (Islamic divination). This “Africanization” of popular Islam even extended to changing perceptions of the prophet Muhammad, who was seen by some as black. Such opinions and practices, along with public, and frequently noisy, performances of their dhikrs, aroused the opprobrium of the ʿulamaʾ. Because most ʿulamaʾ who opposed the popular orders were from the traditionally trained, immigrant clans, rivalries between “orthodoxy” and “heterodoxy” became an expression of a kind of class conflict. Later, in the postcolonial era, these rivalries took on an ethnic appearance that pitted “Africans” against “Arabs”; in similar fashion, the race for sinecures in the new administrations produced competition between scions of “Arab” families (for example, the Mazrui, the bin Sumayt, and the Bakathir) and the prized Swahili pupils of men like Sh. Ahmad al-Saqqaf and Sh. Ramiyya.
In other ways, as Zanzibar became the new focal point of the cultural and religious life of the coast in the nineteenth century, coastal Muslims of all sects came to look to the Arabs, the sultans, the sultans’ courts, and the new Zanzibar ʿulamaʾ and their schools as the new cynosures of Islamic probity and prestige. The old term for a civilized or cultured person, uungwana (literally, a freeborn, city person), thus gave way to ustaarabu, “to be Arab-like,” and it was between 1650 and 1900 that most of the Arabic loanwords presently in the language were absorbed. In the 1800s, the Zanzibar dialect, Kiunguja, replaced that of the Pate-Lamu archipelago as the prevalent dialect of KiSwahili.55 These changes in the Swahili language itself reflected the escalating cultural and religious prestige of the Arabs and Zanzibar.
This growing trend toward greater literacy and attention to the written law had other consequences. Landownership, family inheritance practices, and the general status of women underwent subtle changes, as Caplan, Strobel, and others demonstrate.56 Prior to the nineteenth century, Swahili clansmen and clanswomen in most locations shared equal rights of land use—rights that derived solely from principles of kinship. And, for the most part, it appears that locally born and reared elders, sultans, and even qadis conformed to local property-rights customs. Beginning with the Mazrui arrival in Pemba in the 1700s, but much more so in the 1800s under Busaidi rule, land that previously had been used to grow food crops for local consumption was converted to cash cropping. As subsistence practices rapidly yielded to a cash-based and mercantile economy, in locations like Pemba, Zanzibar, and the Malindi and Lamu regions especially, rights to land use rapidly yielded to rights of land ownership. Customary inheritance rights and practices concerning such land gradually were replaced literally by the letter of the shariʿa.
The imposition of stricter standards and the appointment of walis, mudirs, and qadis to apply those standards by the sultans, and later by their colonial masters, extended their reach into other areas of family law all over the coast. For all women, but especially for those of the upper class, this meant both a decline in personal status vis-à-vis male family members and in their share of inherited property. Indeed, ironically in certain respects women of slave or lower-class status, and women living in the shrinking areas into which the written law was not yet applied (the so-called underdeveloped areas), fared better than their wealthier, upper-class sisters: they were left with rights that the more prosperous lost.
The long march of Islam in subequatorial Africa had its beginnings in, and for many centuries remained largely confined to, the coast of East Africa. The latest evidence indicates that it permanently established itself in the region earlier than it did in any other part of sub-Saharan Africa. But it was extremely slow in making headway beyond its coastal foothold.
The twelve-hundred-year-long history of Islam on the coast produced a civilization that was as rich as it was deep in its tradition—a tradition that therefore proved to be exceedingly resilient in times of crisis. Conversion to Islam in the early centuries proceeded slowly and unevenly, but by the time the first Europeans arrived to challenge Islamic civilization in East Africa, the faith already had achieved majority status in the coastal towns, which by then existed by the dozens.
The Islamic tradition, rooted in both its African and Islamic past, proved to be highly adaptable, and this characteristic assured its survival against determined efforts by European missionaries to subvert it. Under Arab and British rule in the nineteenth and twentieth centuries, moreover, when coastal Islamic civilization moved away from its robust African underpinnings toward one displaying significantly greater foreign influences, it was not undermined. The new influences somewhat altered the character of East African Islam, making it more literate and outward-looking than before. But like many African religions, coastal Islam not only managed to survive, absorbing and adapting powerful outside trends, it thrived.57
1. Mazrui and Shariff 1994, 25.
2. Horton 1996, chapters 9–10.
3. Horton 1991; Flury 1922.
4. Horton et al. 1986; Horton 1996; Freeman-Grenville 1971.
5. Chittick 1974, 34.
6. Wright 1992, 1986; Couto 1778–88, 311; Barros 1945, 9.
7. See Ehret 1998, 171 passim.
8. Most of this information and the data are taken from Nurse and Hinnebusch 1993, 288–97.
9. See Pouwels 1987, 63–97.
10. Horton 1996, chapters 9–10.
11. Devic 1975, 86–87; Figueroa 1974, 592; Santos 1891, 1:427–31; Dapper 1686, 401.
12. See Khan 1969; Kirkman 1983; Whitehouse and Williamson 1973; Whitehouse 1974; Whitehouse 1977; Whitcomb 1975.
13. Badger 1871, 18; Wilkinson 1975; Wilkinson 1981; Horton 1992.
14. Barros, in Freeman-Grenville 1962, 83; Dapper 1686, 401.
15. Dapper 1686, 396; Santos 1891, 1:381.
16. See Abdul-Ghafur 1966.
17. Ibn Battuta, in Freeman-Grenville 1962, 31; Santos 1891, 1:381.
18. Horton 1996, 62.
19. Ibn Said, quoted in Guillain 1856, 268–69; “Ibn Battuta and the ‘Kilwa Chronicle,’” in Freeman-Grenville 1962, 28–32, 39–42.
20. For Ibn Battuta and Abul-Mahasin, see Freeman-Grenville 1962, 28–30, 33. For evidence on Mogadishu, see Ibn Sai d, in Ferrand 1903, 322–23.
21. Trindade claimed that Arabic was spoken “widely” at Kilwa, and Barbosa reported that the sultan of Kilwa possessed “books of the Alcoran.” At Malindi, Lopes reported that the king corresponded in Arabic. See Trindade 1866, 2:8; Stanley 1866, 11; Lopes 1812, 106, 168.
22. Gois reported that most people at Kilwa spoke only “gibberish” (algaravia); that is, the vernacular that probably was not initially understood by the Portuguese. Gois 1978, 148.
23. Nurse and Hinnebusch 1993, 286, 328. For a more recent discussion of the possible early impact of Arabic on Swahili, see Hinnebusch 1996.
24. See Rezende [n.d.].
25. Ibid., fl. 1.
26. Ibid., 23–32; Barros, 22.
27. Many so-called Moorish towns, built in the style of the coast, were mentioned by Barros (9). On the “Arabs of Mecca,” see Dapper 1686, 398–99; Trindade 1866, 43. On the trade with the Red Sea, see Santos 1891, 1:288.
28. At the Madagascar town of Mafialage, according to Faria e Sousa, “the Moors used to buy boys whom they sent to Arabia to serve their lust.” Faria e Sousa 1971, 266. For more on the slave trade, see Correa 1975, 665.
29. Fernao de Albuquerque to the king, 14 Feb., 1620, in Os Livros das Moncoes, Torre do Tombo (National Archives, Lisboa), 22B, fls. 438–40; Os Livros das Moncoes, 17, fl. 10; Strandes, 1961, 243.
30. Martin 1974; Pouwels 1987.
31. Barreto de Rezende, in Theal 1898, 216. For the quotation, see Espirito Sancto 1630, 13.
32. Rezende [n.d.], 31–32.
33. Alonso 1988, 29.
34. This and most of the following information on the Augustinians is based on Alonso 1988 and Hartman 1967.
35. Felix of Jesus, in Hartman 1967, 59ff. Also, Ave-Maria 1948.
36. Strandes 1961, 175.
37. Q.v. Santos 1891, 2:242–46.
38. On the Pais dos Christaos, see the regimento of the king to the viceroy, Feb. 27, 1568; also, Fernao de Albuquerque to the king, 14 Feb., 1620, both in Livros, (note 29), vol. 22B, 438–40; and vol. 40, 260, 267, 272, 272v. Also, Strandes 1961, 210, 243.
39. Dapper 1686, 401.
40. “Index Parochiorum,” in Alonso 1988, 245; and Hartman 1967, 60. The so-called Pate chronicles provide numerous evidences of the shurafa leadership during this period.
41. Ave-Maria 1948, 473, 500, 508, 521, 537, 571, 664.
42. E.g., Knappert 1979.
43. Allen 1981.
44. See Pouwels 1987, 138ff. Given the Hadhrami migrations of the thirteenth century, however, some coastal ʿAlawi dynasties might have pre-dated the seventeenth century. The Pate Chronicles also allude to supposed conquests of mainland areas, including the jumbeates of the Mrima coast. Quite possibly, given that there is no evidence of any such actual conquest by the ruling Nabahani of Pate, these would more likely refer to the influence Pate would have wielded over these southern ports.
45. Nurse and Hinnebusch 1993, 30, 328–29.
46. Much of this material is discussed in greater detail in Pouwels 1987, 63–96.
47. Ibid., 97–144.
48. Farsy 1989, 107–8.
49. See Anderson 1970, xi, 61–67; Pouwels 1987, 163–90.
50. Berg and Walter 1968; Sheriff 1992.
51. Farsy 1989.
52. Becker 1968; Pouwels 1997.
53. Farsy 1989, 43–44. Many of these works can be found in the library of the East African Centre for Research in Oral Traditions and African National Languages (EACROTANAL) in Zanzibar.
54. All these are discussed at length in Pouwels 1987, 158–62 and 196–201. See also Martin 1976, 164–69; Nimtz 1973; Zein 1972; Lienhardt 1959.
55. Nurse and Hinnebusch 1993, 328–31.
56. Caplan 1982, 29–44; Strobel 1979, 43ff.
57. The author wishes to thank the Research Council of the University of Central Arkansas, the American Philosophical Society, and the National Endowment for the Humanities for financial support toward the research and writing of this chapter.
Bibliography
ʿAbdu’l-Ghafur, M. 1966. “Fourteen Kufic Inscriptions of Banbhore, The Site of Daybul,” Pakistan Archaeology 10: 65–90.
Allen, James deV. 1981. “Swahili Book Production.” In Kenya Past and Present, 13:17–22.
Alonso, Fr. Carlos. 1988. Los Agustinos en la Costa Suahili. Valladolid: Estudio Agustiniano.
Anderson, J. N. D. 1970. Islamic Law in Africa. London: Cass.
Ave-Maria, M. de. 1948. “Manuel eremetico de congregacao de India Orientale dos eremitas de N. P. S. Agostinho.” In Documentaca para a historia das missoes do padroado Portugues do Oriente, ed. A. de Silva-Rego, 12:160–73. Lisbon: Agencia Geral do Ultramar.
Badger, G. P. 1871. History of the Imams and Sayyids of Oman from A.D. 661 to 1856. London: Hakluyt Society.
Barros, J. dos. 1945. Asia: Dos feitos que os Portugueses fizeram no decobrimento e conquista dos mares e terras do Oriente, Dec. 2. Lisbon: Agencia Geral das Colonias.
Becker, C. H. 1968. “Materials for the Understanding of Islam in German East Africa,” Tanzanian Notes and Records 68, trans. and ed. B. G. Martin: 46–51.
Berg, F. J., and B. J. Walter. 1968. “Mosques, Population, and Urban Development in Mombasa,” Hadith (ed. B. A. Ogot; Nairobi) 1:47–100.
Caplan, A. P. 1982. “Gender, Ideology, and Modes of Production on the Coast of East Africa.” In Paideuma 28: From Zinj to Zanzibar, in Honour of James Kirkman, ed. J. de Vere Allen and T. H. Wilson, 29–44. Wiesbaden: Steiner.
Chittick, H. N. 1974. Kilwa: An Islamic Trading City on the East African Coast. London: British Institute in Eastern Africa.
Correa, G. 1975. Lendas da India, vol. 1. Oporto: Lello e Irmao.
Couto, D. de. 1778–88. Decadas da Asia. Lisbon: Regia Officina Typografica.
Dapper, O. 1686. Description de l’Afrique. Amsterdam: Wolfgang, Wassberge, Boom, & van Someren.
Devic, L.-M. 1975. Les pays du Zendj, ou la cote Orientale d’Afrique. Amsterdam: Oriental.
Ehret, Christopher. 1998. An African Classical Age: Eastern and Southern Africa in World History (1000 B.C. to 400 A.D.). Charlottesville: University Press of Virginia.
Espirito Sancto, Fr. Thomas do Domingos. 1630. Breve relacao das Christandades que os religiosos de nosso Padre Sancto Agostinho teem a sua conta nas partes do Oriente e do fruyto que nellas se faz. Lisbon: Imprensa dos Padres de Sancto Domingos.
Faria e Sousa. 1971. The Portuguese Asia, vol. 3. Farnborough, U.K.: Gregg International.
Farsy, Shaykh Abdallah S. 1989. The Shafi’i ʿUlamaʾ of East Africa (1820 to 1970): A Hagiographic Account. Ed. And trans. Randall L. Pouwels. Madison: University of Wisconsin African Studies Center.
Ferrand, G. 1903. Relations du voyages et textes géographiques Arabes, Persans, et Turcs relatifs a l’Extrème Oriente de VIIIe au XVIIIe siècles. Paris: Leroux.
Figueroa, M. F. de. 1974. “Chapters Relating to East Africa in the Account of Martin Fernandez de Figueroa.” In Documentos sobre os Portugueses em Mozambique e na Africa Central, vol. 3. Lisbon: Centro de Estudos Historicos Ultramaronos.
Flury, S. 1922. “The Kufic Inscriptions of Kizimkazi Mosque, Zanzibar, 500 A.H. (A.D. 1107),” Journal of the Royal Asiatic Society 21:38–46.
Freeman-Grenville, G. S. P. 1962. East African Coast, Select Documents. London: Oxford University Press.
———. 1971. “Coin Finds and the Significance for East African Chronology,” Numismatic Chronicle 11:283–301.
Gois, D, de. 1978. Cronica de Dom Manuel. Vol. 1. Lisbon: Edicao Amigos do Livro.
Guillain, Charles. 1856. Documents sur l’histoire, le géographie, et le commerce de l’Afrique Orientale, vol. 1. Paris: Bertrand.
Hartman, Fr. A. 1967. “The Augustinians in Golden Goa, According to a Manuscript by Felix of Jesus, O.S.A.” In Analecta Augustiniana 30:5–167.
Hinnebusch, T. J. 1996. “What Kind of Language is Swahili?” AAP no. 47: Swahili Forum 3 (Sept.), ed. R.-M. Beck et al.
Horton, Mark. 1991. “Primitive Islam and Architecture in East Africa,” Muquarmas 8:103–18.
———. 1992. “Archaeology and the Early History of Zanzibar.” Paper presented at the International Conference on the History and Culture of Zanzibar (Dec.).
———. 1996. Shanga, The Archaeology of a Muslim Trading Community on the Coast of East Africa. London: British Institutute in Eastern Africa.
Horton, Mark, et al. 1986 “The Mtambwe Hoard,” Azania 21:115–23.
Khan, F. A. 1969. Banbhore: A Preliminary Report on the Recent Archaeological Excavations at Banbhore. Lahore: Dept. of Archaeology and Museums.
Kirkman, J. 1983. “The Early History of Oman in East Africa,” The Journal of Oman Studies 6:41–58.
Knappert, J. 1979. Four Centuries of Swahili Verse. London: Heinemann.
Lienhardt, Peter. 1959. “The Mosque College of Lamu and Its Social Background,” Tanganyika Notes and Records 53:228–42.
Lopes, Thome. 1812. “Navegacao as Indias Orientes.” In Colleccao de noticias para a historia e geografia das nacoes ultramarinas que vivem nos dominios Portugueses. Lisbon: Academia Real das Sciencias.
Martin, B. G. 1974. “Arab Migrations to East Africa in Medieval Times,” International Journal of African Historical Studies 7:367–90.
———. 1976. Muslim Brotherhoods in Nineteenth Century Africa. Cambridge: Cambridge University Press.
Mazrui, A. M., and I. N. Shariff. 1994. The Swahili: Idiom of an African People. Trenton, N.J.: Africa World History.
Nimtz, A. H. 1973. The Role of the Muslim Sufi Order in Political Change. Ph.D. diss., Indiana University.
Nurse, Derek, and Thomas Hinnebusch. 1993. Swahili and Sabaki, a Linguistic History. Los Angeles: University of California Press.
Pouwels, Randall L. 1987. Horn and Crescent: Cultural Change and Traditional Islam on the East African Coast (800–1900). Cambridge: Cambridge University Press.
———. 1997. “Islamic Law and Theology.” In Encyclopedia of Sub-Saharan Africa. New York: Scribner’s.
Rezende, P. Barreto de. [n.d.] “Breve Trotado e Epilogo Todos os Visoreys que tem travido no Estado da India.” Bibliothèque Nationale de Paris, Manuscripts Portugais no. 1.
Santos, dos J. 1891. Etiopia Orientale. Lisbon: Biblioteca de Classicos Portugueses.
Sheriff, A. H. M. 1992. “Mosques, Merchants, and Landowners in Zanzibar Stone Town,” Azania 27:1–20.
Stanley, H. E. J. 1866. Duarte Barbosa: A Description of the Coasts of East Africa and Malabar in the Beginning of the Sixteenth Century. London: Hakluyt.
Strandes, Justus. 1961. The Portuguese Period in East Africa. Trans. J. Wallwork. Nairobi: East African Literature Bureau.
Strobel, Margaret. 1979. Muslim Women in Mombasa (1890–1975). New Haven: Yale University Press.
Theal, George M. 1898. Records of South-eastern Africa: Collected in Various Libraries and Archive Departments in Europe. Vol. 3. London: Government of the Cape Colony.
Trindade, Paulo de. 1866. Conquista espiritual do Oriente. Lisbon: Centro dos Estudos Historicos Ultramarinos.
Whitcomb, D. 1975. “The Archaeology of Oman: A Preliminary Discussion of the Islamic Periods,” Journal of Oman Studies 1:123–57.
Whitehouse, D. 1974. “Excavations at Siraf,” Iran 12:12–30.
———. 1977. “Maritime Trade in the Arabian Sea: The Ninth and Tenth Centuries A.D.” In South Asian Archaeology (1977), ed. Maurizio Taddei, 865–88. Naples: Instituto Universitario Orentale.
Whitehouse, D., and A. Williamson. 1973. “Sasanian Maritime Trade,” Iran 11:29–49.
Wilkinson, J. C. 1975. “The Julanda of Oman,” Journal of Oman Studies 1:97–108.
———. 1981. “Oman and East Africa: New Light on Early Kilwan History from Omani Sources,” International Journal of African Historical Studies 1:272–305.
Wright, H. T. 1986. “Early Communities on the Island of Maore and the Coast of Madagascar.” In Madagascar, Society and History, ed. D. P. Kottak et al., 53–88. Durham: Carolina Academic Press.
———. 1992. “Early Islam, Oceanic Trade, and Town Development on Nzwani: The Comorian Archipelago in the Eleventh to Fifteenth Centuries A.D.,” Azania 27:81–128.
el-Zein, A.-H. 1972. The Sacred Meadows, a Structural Analysis of Religious Symbolism in an East African Town. Ph.D. diss., University of Chicago.