The Diwan

Mai Ibrahim and courtiers as depicted in Denham, Clapperton, and Oudney's Narrative of Travels and Discoveries in Northern and Central Africa, in the Years 1822, 1823, and 1824.

Although his study of the document contains many speculations built on flimsy foundations, Dierk Lange's critical translation and interpretation of the Dīwān des sultans du (Kānem- )Bornū: chronologie et histoire d'un royaume africain (de la fin du Xe siècle jusqu'à 1808) is required reading for all interested in the history of Kanem-Borno. Using both copies of the Diwan given to Heinrich Barth, Lange translated the text and used a variety of internal and external sources to corroborate the historical accuracy of it. Indeed, specific external sources with approximate or nearly precise dates can support the chronology of Kanem-Borno by looking at the relatively accurate length of reigns of various mais. Undoubtedly, Lange's chronology is probably the most accurate we possess for the history of Kanem-Borno. 

By beginning with the end of the Sayfawa dynasty in 1846, and working backwards in time, one can arrive at a generally accurate timeline for the heads of state back to the medieval era. While less precise for the time of Hummay or the previous so-called Duguwa dynasty, the rough chronology gives us an idea of who was in power in early Kanem during moments of trans-Saharan interaction with polities in Tunisia, the Fezzan, Egypt, and beyond. Lange also suggests some probable theories on the nature of the original lost "Chronicle of Borno" that was likely the basis for the Diwan and how it pushes back the origins of Kanem to the 6th or 7th century. Thus, the fictive reign lengths of hundreds of years attributed to a few early Kanem kings (Duku, Katur) really represent multiple generations of rulers (11 or 12) before Ayuma (r.987-1007), perhaps pushing the origins of Kanem back to the 500s (Lange 141).

Unfortunately, Lange confidently asserts some questionable theories about the origins of Kanem and the question of dynastic continuity throughout the history of the civilization. Despite the Diwan, Ahmad b. Furtu, Ibn Sa'id (drawing on Ibn Fatima), and Kanuri oral traditions, Lange insists on seeing the Sayfawa as newcomers not lineally descended from the Duguwa rulers of early Kanem. To Lange, the Sayfawa, beginning with Hummay, were actually Berber in origin. Drawing on al-Yaqubi, al-Idrisi, Muhammad Bello, the romance of Sayf (with its allegedly anti-black or at least racialized character) Leo Africanus, the Diwan alluding to red-skinned early mais, and some linguistic assumptions, Lange comfortably claims, "Hummay était en toute vraisemblance un Tarki (sing. du nom arab Tawarik) issu d'un milieu islamisé et même, en partie, arabisé" (111). According to Lange, Hummay's very name could be connected to the Tuareg name Hammai, while his mother and grandfather could have names of Arabic origin modified by Berber speakers (98). 

While one could could not deny the possibility of ancient Saharan contacts between Berbers and the ancestors the Kanembu and Kanuri, the Diwan clearly states Hummay's mother was a Kay (Koyam) and his father was the previous king of Kanem, Abd al-Djalil. Consequently, we here at the blog differ in our interpretation of a passage in the Halle manuscript of the Diwan (97). To us, the passage is emphasizing a distinction between the descendants of Hummay and Duku because of the former's deeper attachment to Islam. While earlier mais appear to have been Muslims (Abd al-Djalil, Hawa), perhaps it was only with Hummay and his descendants that Islam was more deeply established as an official religion of Kanem (or its ruling elite). For example, we know al-Bakri, writing in the mid-11th century, saw Kanem as a land of "pagan" Sudan. However, if the Diwan is reliable for 11th century kings, Hawa (r. 1067-1071) was already a Muslim. Hawa, if male, was the son of Arku, whose mother was a Tomaghera from Kawar, a region with a deep history of Islamic traders (including Berber merchants). Is it not likely that the ascension of Hummay may have represented a strengthening of a Muslim faction among the ruling dynasty, perhaps one more deeply attached to Islam than Hawa and Abd al-Djalil? Furthermore, the possibly Arabic names of Hummay's mother and maternal grandfather (Takrama and Ghumar) could simply reflect the transmission of Islam through Berbers (and Arabs). Based on the sources we currently have, there's simply no evidence for Hummay being Tuareg or some other sort of Berber. However, Lange could be correct that Hummay came from a background already Islamized, as indicated by the Diwan mentioning 2 previous Muslim rulers of Kanem.

As for the alleged descent from Sayf, and the allegedly anti-black character of the romance of Sayf making it unlikely for "black Africans" to claim descent from him, we are not convinced (102). Obviously, a plethora of fictitious Yemeni or Arab genealogies exist across Muslim Africa. Moreover, the rulers of Kanem were known to have adopted the Sayfian nisba by the 13th century, centuries before the romance was written down in Egypt. We simply do not know if the ruling elite of Kanem were aware of the "racial" implications the legendary Sayf was associated with, and other factors could be relevant to the clearly fictitious political myth of origins. If, on the other hand, Lange's interpretation of Hummay as a Berber whose successors claimed descent from Sayf because of the "racial" character of the legendary Sayf defeating non-Muslim "black Africans" were true, one would have to wonder why they also claimed descent from the earlier, pagan dynasty. After all, Ibn Sa'id reported that the "pagan" ancestors of Dunama Dibalemi (r.1210-1248) were based in Manan, the ancient capital of Kanem mentioned by al-Muhallabi in the late 10th century (96). This would suggest Dunama Dibalemi was a descendant of the same kings who ruled from Manan and, later on, shifted the capital to Njimi. 

The main complication to the narrative of dynastic continuity from early Kanem to 1846 is al-Umari, who wrote of the Sayfawa seizing power after al-Hadi al-Uthmani established Islam in Kanem (99). In fact, this Hadi al-Uthmani who supposedly first established Islam, is not mentioned in the Diwan and supposedly claimed descent from the 3rd Rashidun caliph. Borno traditions and mahrams remember another person who introduced Islam (Muhammad b. Mani), and a close reading of al-Umari in Corpus of Early Arabic Sources for West African History never refers to Hadi al-Uthmani explicitly as a king. Instead, al-Umari wrote, "The first man to establish Islam there was Hadi al-Uthmani who claimed descent from Uthman b. 'Affan. It passed after him to the Yazanis, the descendants of Dhu Yazan" (Corpus 261). One wonders if the man al-Umari names as the first to establish Islam was an Islamic scholar affiliated with Hawa or Abd al-Djalil. The reference to the Yazanis or descendants of Sayf could be a reference to the Sayfawa introducing and cementing Islam throughout their domains as the primary patrons of the religion. After all, one of the sources of legitimacy of the dynasty was their support for Islam, Muslim scholars, and the faith. Hence, an alternative reading of al-Umari suggests the Sayfawa (or, really, the Duguwa but more explicitly Islamized) became the dominant patrons of the religion. 

Another sign of likely dynastic continuity from the Duguwa to the Sayfawa can be found in pre-Islamic forms of political legitimacy that continued throughout the history of Kanem-Borno. If the Sayfawa who took power in the late 11th century were Berbers from a family Islamized for quite some time, how does one explain their preservation of older, non-Muslim customs and traditions? Intermarriage with previous ruling elite? We find it more likely that they were indeed the same dynasty and they never completely dropped non-Muslim traditions of kingship. Thus, the belief in the power of the ruler to influence health and agriculture, the veneration and isolation of the mai, and widespread acceptance of the myth that the king doesn't require food, for instance, continued long after the medieval era. For example, Kanuri informants of Descourtilz in Saint-Domingue reported that the king of Borno never leaves his palace and any subject who dares to stare at him is punished by death (Descourtilz 146). Ibn Battuta in the 14th century, like al-Umari, described Borno's ruler as never appearing before the people and never addressing them unless from behind a curtain (although al-Umari mentioned 2 festivals in which the king did appear). 

In addition, W.G. Browne, who traveled to Darfur in the 1790s, mentioned a rainy season ritual in which the king of Borno, like the sultan of Darfur, was expected to dig holes into the ground with his hands, suggesting kingship in pre-Islamic Kanem was believed to have an effect on the harvest (probably explaining why al-Muhallabi thought the subjects of Kanem's "pagan" kings believed their mai provided food and health). This suggests a degree of continuity in kingship rituals, practices, and style from the time of al-Muhallabi (late 900s) to the late 18th century. In short, the mai was, usually, distant from his subjects, associated with extreme power, and believed to have some influence on agriculture (supernatural power associated with kingship in deep antiquity?) and these characteristics remained relevant long after mai Hummay ascended to the throne. If Hummay really was of external origin and/or Berber, he and his descendants must have quickly assimilated into the local ruling class and what became the Magumi (royal clan, magumi meaning free-born, according to Koelle). 

While more could be said on the questionable conclusions and guesswork of Lange with regards to Leo Africanus and Muhammad Bello on the origins of the Sayfawa dynasty, or even the speculative association of the Duguwa dynasty with blacksmiths, our hope is not to focus solely on the negative. The Diwan received a much better translation here than in Palmer's Bornu Sahara and Sudan. Fortunately, Lange is more careful and measured in his conclusions as well as providing a more accurate chronology than that of Urvoy or Palmer. For a superior translation with a plethora of useful comments and references, this French study of the Diwan is important for helping us to establish the basic timeline of an African civilization and how it developed over time. Though we are skeptical of the idea of two separate dynasties, this is essential reading for the history of Kanem-Borno.