The Brick Palace of Machina

One of the disappointing aspects of the study of Sayfawa royal architecture is the absence of adequate photographs, plans, and sketches of the old palace complex at Machina, Nigeria. Long known to be the only fired-brick construction sponsored by the Sayfawa that is still in use (the others have been plundered for building material or ravaged by time and sackings), Machina's palace structure has undoubtedly been modernized or restored in the 20th century (and likely the 21st, too). But finding photographs of what the palace complex looked like has led us to Google and social media accounts by Nigerians from Yobe State or Machina itself. These images, however, do not provide a complete overview of the palace complex. Nor does Google Earth clearly display an outline of the palace complex. 

In spite of these aforementioned limitations, the photographs online do offer some clues as to how the brick structure might have looked in the past. In addition, it was also similar to other sites in Borno and possibly the larger "Central Sudan" region. For instance, the circular structure in which the Machinama (king of Machina) is seated while talking to guests may be how circular, brick structures were used in Birni Gazargamo, Gambaru, and other sites. These brick buildings with a courtyard for hosting visitors or relatives were part of a larger palatial complex of buildings with brick walls. Besides this feature, the old brick walls for a structure that appears to have been retained in front of newer concrete walls may have been the reception space for larger audiences. Alternatively, it may have once been part of the walls of the original house for the king, possibly lost when the ruling family favored concrete and other materials to modernize the palace.

In terms of historical sources and studies, Machina's palace was noted by colonial-era officials, travelers, and subsequent, post-colonial scholars. C.J. Lethem, whose 1919 report on the Marchena District report has been digitized, was told the palace was unique for having a main gate facing the east rather than the west. Supposedly, this was done so the palace's gate could face toward Birni Gazargamo. Moreover, Lethem believed the bricks at Machina's palace were smaller than those of Gazargamo and Gambaru. Yet despite its smaller size, the structure in the early 20th century had one hall, which was 20 feet high, plus 54 layers of brick in its walls. The structure was in need of repair, however, especially the roof and arches. Unfortunately, Lethem did not include a sketch or plan for the palace. To his credit, however, he recorded an early version of Machina history based on a chronicle lost when Zinder attacked the town in c. 1896 (or was it just a girgam?). This chronicle, rewritten based on memory, extended the history of the town 988 years. It similarly made the original rulers of the town Sao (So). 

But the chronicle sought to connect Machina with the origin of Sayfawa consolidation at Gazargamo in the 15th century. According to this tradition, the first mai to settle at Gazargamo sent troops to the west to capture a woman who could bear a son for him (no previous wife had sons). The woman the Bornoan forces seized was actually a Machina woman who was the wife of the Machina Sao hunter, Bolo. Unbeknownst to the ruler of Borno, this Machina woman was already pregnant with a son, born in Borno. She later bears the Sayfawa king another child, named Arigu like her previous son. Then, in a case of mistaken identity, her first son succeeded to the throne. After calamity, people in Borno realize her second son was the legitimate heir. With his half-brother on the throne, the son of the Machina chief returns to Machina. But missed by his dear half-brother in Gazargamo, he is visited by the Sayfawa mai. In order to make sure his brother lives in an appropriately royal setting, the Sayfawa mai "sent for a supply of the famous Birni bricks, or, according to another account, hade them made at Mijirik, near Gumsi "(Lethem 4). If taken at face value, this account would make Ali Gaji the step-father of Arigu, the first Machina leader to live in a brick palace. Since there are no sources to corroborate what is clearly a legend of origin, and we have no records of a Sayfawa mai named Arigu who ruled after Ali Gaji, this account is nonetheless interesting for not claiming direct descent from a Sayfawa mai. Even if its attempts to connect the Machina dynastic line to an earlier period in Borno history, the fact that they don't assert a Sayfawa origin may represent some historical truth. Indeed, according to Lethem, the chiefs of Machina were Walu, a group possibly tied to the Dagira peoples.

The next person to describe the palace of Machina was Migeod, whose Through Nigeria to Lake Chad was published in 1924. Migeod saw the palace, describing it as a brick house with a brick wall around it. He also thought the bricks were carried from Gazargamo a little over 100 years ago, when the Sayfawa capital was abandoned. Moreover, the Machina palace was in poor condition by the time Migeod saw it. Consequently, the Machinama was in the process of rebuilding it. Nonetheless, Migeod still observed large pillars and arches and a square-shaped reception hall with 4 arches (Migeod 278). Is it possible this is the most imposing of the brick structures seen today? It is quite clear that Migeod was wrong to suggest the brick palace was only built in the 1800s and with plundered bricks from Gazargamo. This is contradicted by Lethem's sources and may have been a reflection of the limited knowledge of fired-brick structures west of Lake Chad at the time. Interesting, Migeod's book included 2 photos of walls from the Gambaru palace structure, whose bricks do bear a resemblance to those of Machina. 

With the exception of John Lavers, whose research on Machina's palace is difficult to find, Wilhelm Seidensticker seems to the most recent scholar to investigate the site. In a short essay on brick technology in Borno ("Borno and the East: Notes and Hypotheses on the Technology of Burnt Bricks"), he referred to Machina's palace as a complex with several impressive brick buildings. Furthermore, Machina tradition attributes the bricks to Gazargamo. The people there even described the bricks differently, based on color. In another essay, entitled "Archaeometric investigations on burnt bricks from six archaeological sites in Nigeria and Niger," Seidensticker et al. wrote of preliminary chemical analysis of bricks from different parts of Borno. The Machina bricks, though said to have come from Gazargamo by oral tradition, were actually unlike the samples from every other site (Lergam, Gazargamo, Gambaru, Kukawa graves of al-Kanemi shehus, Garumele, Garukime). Although further analysis is necessary, the chemical analysis of the bricks from these sites may suggest that masons and bricklayers sponsored by the Sayfawa were sent to various locations. These artisans may have used local materials for their bricks. In fact, this theory is the most sensible since it is difficult to imagine large quantities of bricks being moved from across vast distances if local materials could be sourced and prepared to complete elite structures at Machina, Gambaru, Garumele, Lergam, Garukime, and other sites. Of course, actually dating the bricks and a more thorough chemical analysis are required to understand the chronology and when Machina's palace came into being. 

Ultimately, we can only hope for complete photographs of the site, dating of the bricks, and further study and comparison with Borno and Kanem fired-brick sites. The traditions of Machina suggest that the palace was built by a Sayfawa mai for a half-brother who ruled at Machina. This tradition may not stand up to scrutiny, but is interesting nonetheless for providing an example of Sayfawa elite agriculture "extended" to Machina. This brings to mind the Sayfawa sponsorship of a brick mosque for the Kalumbardo community, although possible Sayfawa descent for the second movement's founder has been suggested by Bobboyi. Further, the Machina palace structure, despite modernization, has elements comparable to other fired-brick structures in the Central Sudan. Further study of the largest brick building and the use of arches or pillars may help us understand how Gazargamo, Gambaru or Njimi structures looked. Since the bricks used at Gazargamo and Gambaru were larger, we should probably expect Machina's palace to have been much smaller than what the Sayfawa rulers used for their own elite residences. 

Ouara, ville perdue (Tchad)

Jean-Paul Lebeuf and Johannes Hermann Immo Kirsh's Ouara, ville perdue (Tchad) is a study of the elite palatial complex at Wadai's former capital, Wara. Unfortunately, the site does not seem to have received further investigation and one wonders about the current conditions of the palace, mosque, and other structures. As elite architecture made of bricks in the Central Sudan, one cannot help but wonder to what extent the earlier fired-brick building traditions of Kanem and Borno influenced Wadai. Lebeuf and Kirsch did not fully explore this question, but they raised a number of intriguing theories about parallels with Kotoko architecture and African royal architectural traditions. 

Using the few written sources, oral sources, and past and contemporary plans, photographs, and sketches, the authors did endeavor a basic overview of the site. That said, given the paucity of written sources from the period when Wara was occupied by the ruling dynasty of Wadai, one must interpret the sources carefully. Nonetheless, a general overview is possible. The larger site includes an area attributed to the period of the Tunjur, but the site is most associated with Abd al-Karim and his descendants. A royal cemetery, a maison de marabout brick structure, the palace structure, elite residences for upper-echelon dignitaries like the djerma and the counsel hall structure are still impressive ruins. A partially ruined pyramidal structure on Mount Treya and a central tower that may have been an observatory aligned with the Pleiades indicate some intriguing possible motives for the site's design. Unfortunately, precise dating of many of these structures has not been attempted, although a brick from the mosque structure dated to 1796 was found. It would seem that successive kings of Wadai built or sponsored the construction of some brick buildings over the centuries the site served as a royal capital. Furthermore, it was likely not a large city, with its growth circumscribed by the mountainous terrain. 

Despite its small size, the city's astronomical alignments for some structures and the larger cosmological and ritual significance of the space's design are worth exploring. The building remembered traditionally as the maison de marabout included a tower or column structure of brick. The authors speculate that this may also have been a symbol of the kingdom's axis. Indeed, the division of Wadai into 4 quadrants based on the cardinal directions with the center, or axis, at this structure was no coincidence. Furthermore, the Salle de Conseil with 4 sides, square-shape may have been another reference to the 4 provinces of the state: North, South, East, and West. While the authors may need further proof for the larger cosmological points about square and round buildings and towers as references to the land and the sky, possible similarities with the Kotoko states is not impossible. Clearly, part of the capital was designed for ritual reasons. For instance, the platform on Mount Trey was used for accession rituals of the Wadai rulers. The royal cemetery was also ritually significant, even though none of the tombs feature inscriptions. 

Hopefully with the dating of fired-brick structures in Kanem, similar research can be undertaken for Wara. Even if the brick mosque was built in the 1790s, some of the other structures may date from the 17th or early 18th centuries. This late date also raises the question about influences from Kanem, Bagirmi and Wadai. The influence of Kanem-Borno in Wadai can be detected in areas like political titles, such as the djerma. The word birni was also used in Wadai to refer to the palace, perhaps another instance of political vocabulary (and royal architecture?) being influenced by Kanem-Borno. It is not inconceivable that rulers of Wadai also borrowed from the Sayfawa dynasty (and its successors in Kanem, the Bulala) brick structures for elite residences and mosques. A close analysis of brick construction techniques and architectural styles must be launched for Kanem, Wadai and Borno to verify this hypothesis. Massenya, the capital of Bagirmi, should likewise be thoroughly investigated. 

Hausa Influences in Oyo

Whilst perusing Samuel Johnson's History of the Yorubas, one notices multiple examples of Hausa influence in the Oyo state. Even when referring to the mythical or semi-legendary rulers like Sango, examples of this Hausa or northern presence can be detected in Yoruba historical tradition. For example, one of the Yoruba legends of origin traces the people's descent from Lamurudu of Mecca. This figure was said to have been the father of Oduduwa as well as two kings of Gogobiri and Kukawa in the Hausa region (3). Kukawa may mean Borno. Even the (legendary) Sango was said to have had a Hausa slave, too (149).

Some titles used in the Oyo government were also words of Hausa origin. Take the Magaji, for instance. This title was said by Johnson to go to the heir of a great estate, usually the eldest son of that family. The alafin of Oyo's court also included an official called the Magaji Iyajin. He functioned as an official "elder brother" who defended the interests of the king (69). Well, magaji in Hausa means heir or successor, according to An English-Hausa Dictionary. Another word of Hausa origin used by Oyo's government was the seriki, derived from the Hausa word for king. According to Johnson, the seriki was an military leader who was comparable to the Balogun but more for the young warriors. A successful seriki could even succeed the Balogun (133).

Some of the specific trumpets and instruments associated with kingship also came from the Hausa or via Nupe peoples. The kakaki and ivory trumpets were only used for the king (121). Johnson also referred to a cymbol (aro) or the Bembe drum as instruments introduced by the Hausa. The use of these instruments may have been through the Nupe as an intermediary, but it demonstrates how the northern peoples exerted a significant influence on Oyo.

As noted by various scholars, such as Adamu in The Hausa Factor in West African History, Hausa captives and artisans were also a significant presence. Hausa captives were the main labor force for taking care of the cattle owned by the chiefs. In addition, Hausa and Fulani peoples were often the barbers and ropemakers in Oyo (123). Hausa traders were also a familiar presence long before the 19th century. One of them, named Alaja-eta by Johnson, had his goods seized in Oyo until Aole, the alafin, ordered them returned. The failure to return the Hausa merchant's Koran contributed to the political tensions of Aole's reign after the zenith of Abiodun (190). Later, the Hausa slaves in Oyo joined Afonja as soldiers for Ilorin, but that development is better known or documented.

Overall, the Hausa influences in Oyo seem to be of a much deeper antiquity than we initially thought. While much of it was likely indirect through Nupe and Bariba peoples, the actual Hausa presence in Oyo was clear. And it was not restricted to slaves and Hausa or Fulani-oriented crafts and trades. It even included titles and political offices that used words of Hausa origin. Some were said to marry with the elite of the Oyo kingdom. Thus, Abiodun, remembered in tradition as one of the greatest kings of the Oyo Empire, married his daughter to a Muslim who had secretly harbored a son of Gaha for several years (186)! According to Johnson, Abiodun was previously a trader in potash before ascending the throne and defeating Gaha (187). Is it possible the potash trade increased Abiodun's dealings with Muslim traders from Hausa land and the north? 

Mahamat ibnou Dirmi and Wadai

Whilst perusing Abd el-Karim ibnou Djamé, propagateur de l'Islam et fondateur du royaume du Ouaddaï by Marie-José Tubiana, Issa Hassan Khayar, and Paule Deville, we noticed yet again the Borno/Kalumbardo connection for Wadai's founder. According to various oral traditions, Abd el-Karim studied theology or religion with a Mahamat ibnou Dirmi. This name appears in Muhammad Bello's account of the first Kalumbardo's destruction at the hands of Umar b. Idris of Borno. It also appears in Borno oral traditions and in Wadai. We have written about this elsewhere in our endeavor to analyze Umar b. Idris's decision. Yet the traditions cited by Tubiana et al. strengthen the Borno identification.  For example, the account of Adam Mahamat el-Amin of Wadai referred to Mahamat ibnou Dirmi in the following manner: "Il fut tué injustement par ses enemies"(27). Wadai tradition is adamant that Mahamat ibnou Dirmi was buried in Dar Bornou (20). 

Interestingly, however, Wadai tradition remembers a shaykh named Dede who was buried at the Bahr El-Ghazal (27). This is possible since Bagirmi (where Bidderi is located) in the late 1500s and early 1600s was not disconnected from this area and Islamic scholars sometimes taught in different locations as they moved. More importantly, if the teacher of Abd el-Karim was indeed the same leader of the first Kalumbardo, is it not unreasonable to postulate a totally political motive for the Borno mai's decision to execute him and exile Waldede? Consequently, does this mean that Wadai's foundations possibly preceded the reign of Umar b. Idris of Borno (r. 1619-1639, in Lange's chronology)? This could mean that Barth's chronology for the kings of Wadai is more accurate than that of Nachtigal's, although further confirmation is necessary. 

Tantalizing clues, however, may be found in the sparse data about one of the predecessors of Umar b. Idris. According to Palmer's Bornu Sahara and Sudan, Muhammad b. Idris (1596-1612) died in Kanem. According to Palmer, "He was slain in Holy War, and buried at Dagana Daniski" (244). This could have been related to the rise of Abd el-Karim's rise to power in Wadai. His political ascent led to some of the Tunjur leaving Kanem, although pressure on the Bulala sultans of Kanem may have also been a factor since it is unclear if a Tunjur exodus from Wadai had already begun in c.1610 or 1611. Either way, it does seem rather likely that Mahamat ibnou Dirmi was the very same leader at the first Kalumbardo. If one of his students, Abd el-Karim, later overthrew the Tunjur dynasty, it is no wonder that Umar b. Idris viewed Waldede and al-Jirmi as a political threat. 

Revisiting the Siècles Obscurs

Part of Gustav Nachtigal's map of Kanem and the Lake Chad Region (Gallica)

Although the paucity of written sources makes it exceedingly difficult, the so-called siècles obscurs of the 17th and 18th centuries are worth revisiting. Far from being obscure, 17th–18th century Kanem was politically, economically, and socially dynamic. Several changes took place politically and ethnically as Kanem and the Bahr el-Ghazal witnessed population movements, the replacement of the Bulala sultans by an alifa under the auspices of the Sayfawa, and the growing threat of Wadai in the east. Kanem’s integration in a Borno-centered regional economy continued through the lively trade in livestock and horses while trona production and trans-Saharan trade continued. Kanem’s political conflict between the alifate based at Mao and the Tunjur leadership at Mondo likewise contributed to a period of political conflict. Nonetheless, Kanem’s social, ethnic, economic, and political transformations in this era provide another vantage point for viewing the gradual decline of Borno and the political revolutions that shocked the Central Sudan in the 19th century. Thus, the picture that emerges from an analysis of the 17th and 18th centuries is one of political and economic dynamism. This was a provincial dynamism that gradually declined with metropolitan Borno’s loss of the Bilma salt mines, insecure trans-Saharan routes, and a reorientation of the Central Sudan’s metropole with the rise of the Sokoto Caliphate.

First, the political changes. We have already attempted to create a more coherent chronology for the alifa rulers at Mao, using oral traditions and attempting to corroborate it with other sources. But, what was the state of the Dalatoa alifas? According to Landeroin, the institution began with Dala Afouno. Whether or not he was from a village in Kano, his descent from a Magumi mallam and appointment by the Sayfawa seems to have begun in the 17th century. Indeed, Zeltner has echoed Lavers in preferring a date by c. 1642 for the establishment of the caliphate at Mao.[1] As for the court at Mao, it consisted of court officials called kokenawa. The court dignitaries used titles like those of Borno: kaigamma, jerma, dugma, yerima, etc. In addition, the post of dima was a once-powerful position with authority over the Kanembu tribes and requiring ratification by Borno.[2] Conte has also stated that the court at Mao was modeled on that of Borno, with “fiefs” for some officials while others did not receive any land. The highest dignitary title, the zegbada, was quasi-hereditary and vested in a descent group of the alifas. Other court titles included Mala, N’goa, Kajala, gerema, dalado, maina, magiri, Nogona, and khalifa. The jerma was a minister of justice and the interior. As for the Mala, it was for the head of the Dalatoa Kingiru.[3] Moreover, a Kogona clan, established in the 17th century, was originally a military and administrative caste under the authority of the alifas.[4] 

 

Additional details on the administrative structure of the Dalatoa-Magumi ruling group can be found in an ethnographic work by Robert Bouillé. His Les coutumes familiales au Kanem was published in the 1930s, after the momentous changes of the 19th century and the French colonial conquest. Nonetheless, it is likely reliable for the structure of the government before the 1800s, too. Thus, the alifas divided Kanem into sous-commandements led by chiefs with titles like Fougbou, Melinia, Kadela, etc.[5] While it is possible that the Fougbou at Mondo was very autonomous in the past, it is noteworthy that the leader of the Tunjurs there used a Kanembu title. Moreover, oral testimony reported to Landeroin clearly establishes how difficult it was for the alifas to ensure payment of tribute or taxes. Nonetheless, the state appears to have been modeled on the Sayfawa court of Borno (which, of course, first developed in Kanem several centuries earlier). As noted by Nachtigal, the alifas appointed dignitaries who oversaw several valleys from Mao to Gala, including Billema, Kimbageri, Niggara, Billangara, and more.[6] The family of the zegbada, the highest post, was said to be part of the Samarous tribe and had intermarried with the family of Mustafa (of the alifas) for around 200 years by the 1930s.[7] Intriguingly, Nachtigal believed the djegebada, or zigibada, was a slave official who functioned as a royal messenger in Borno.[8] In the case of Kanem’s alifas, this title was not associated with servile status and actually possessed a close relationship with the ruling family. Further confirmation is necessary, but we suspect the family that monopolized the post of zegbada may have included the father of the 19th alifa, Ahmadou Kalli (though he reigned more than once during the 19th century). In Landeroin’s genealogy, Ahmadou Kalli’s father is named jagata Moustapha.[9] Was this jagata Moustapha of the same family as the zegbada family?

In addition to taxes on agricultural produce and livestock, notables and village chiefs had to present gifts to the alifa for Ramadan, harvest times or other occasions. Failure to do so meant their goods were seized.[10] The principle of taxation paid to the alifa may have been linked to the idea of him as the sole “owner of the land” of the Kanembu. This justified their tax on outside groups, the kiski. Land was distributed in lougan, a term for a cultivated plot. Beneficiaries of the alifa were allowed access to land and paid a moud at harvest, 6 zakats per lougan.[11] In addition, the state used officials called mara to function as a village chief. Others, called Déguedji, were sent by chiefs to collect taxes.[12] Furthermore, Kanembu farmers sometimes paid several other taxes, such as the dougouchi, mararom, and sadaka.[13] In exchange for the burden of various taxes and forced “gifts” to the alifa, the state also maintained a Bit el mal while Borno still held Kanem. This was a caisse publique designed to feed the poor, provide gifts to mallams, aid foreigners, and to pay masons for the construction and maintenance of mosques.[14] Returning to the origins of the state based at Mao, the first alifa, Dala Afouno, had to submit most groups by force to receive tribute. This included the Tubu, Daganas, Koukas west of Fitri, and Shuwa Arabs. All of his successors similarly had to launch campaigns from Mao to force recalcitrant groups to pay the required tribute, save Hadji, the seventh alifa.[15] Consequently, the issue of taxation was a persistent problem for the rulers of Kanem during the 17th and 18th centuries. Indeed, by the end of the 18th century, Méle Koura b. Beker, the 14th alifa, was deposed by the mai of Borno for pillaging and attacking villages of the Yabouribous, a Magumi group that refused to pay tribute since they were of Magumi descent.[16] This led to the deposed alifa seeking the aid of Sabun of Wadai to seize his old position.

In the midst of the state’s struggle to ensure tribute payments, Kanem’s population structure was also changing. Zeltner, in his history of Kanem, wrote of the Awlad Himayd and Rashid in the Bahr el-Ghazal region in the 17th or 18th centuries. The Daganat united with the Awlad Alwan while the Bani Wa’il under Far’on and Idris were the most prosperous. The Kreda also pushed south. By the mid-18th century, the Asali had left the Bahr el-Ghazal, likely due to Tubu pressure.[17] Unsurprisingly, the Daza presence, migrating from Borku to Kanem, was also felt during the 18th century.[18] Growing pressure from nomadic groups like the Kreda also pushed the Bani Wa’il to disperse in c. 1780-1800.[19] Of course, the aforementioned Tunjur were also established in Kanem during the 17th century. The 18th century even witnessed the migration of various Kanembu populations to the west of Lake Chad. The Tsougourtis, for instance, were said to have left Kanem during the 18th century. New villages of Borati, Dogochi, Kaoua, Kounguia and others, of which Kaoua was the most important, were settled. Kaoua was said to have become a village during the reign of Ali b. al-Hajj Dunama of Borno.[20] The 19th century witnessed the arrival of the Awlad Sulayman, another nomadic group, while Borno and Wadai engaged in conflict with each other over suzerainty of Kanem. Undoubtedly, the perennial conflict between the Tunjur of Mondo and the Dalatoa of Mao continued.

Despite the population shifts with nomadic or semi-nomadic Tubu or Arab populations and the departure of Kanembu farmers, Kanem was believed to be a prosperous region by some outsiders. One of the rare textual sources that allude to the region in the 18th century referred to Kanem as a “city” 5 days from the town of Domboo. Kanem, or perhaps Mao, was seen as the capital of a fertile province where cattle and horses for the ruler of Borno were raised.[21] It is also quite likely that trona was collected in the Foli region, an area where a Magumi presence had been established for centuries and was filled with many villages.[22] Thus, despite the chronic conflict between Mondo and Mao or the pressure from Tubu and Arab nomads, Kanem continued to participate in trans-Saharan trade, promote a pastoralist economy, and likely exported trona to Borno and beyond. This trade in livestock, horses, salts, and other products ensured Kanem occupied a role in the larger Borno-centered Central Sudan economy. When peaceful relations were maintained, the Buduma likely participated in this exchange through providing canoes and barks for the movement of people and goods across Lake Chad. Even as ecological and population shifts made Kanem less capable of supporting larger cities or towns like Borno could, the region was likely more important than is often remembered for the larger Bornoan economy. On the other hand, the gradual decline of Borno’s economic power and the loss of control of the Kawar salt trade in the mid-18th century likely had a negative impact on Kanem. In fact, growing insecurity on the trans-Saharan routes may have decreased trans-Saharan trade that once reached Kanem. 

Cultural and religious life during the “obscure centuries” can be glimpsed, too. After all, it was in Kanem where the father of the Shehu, Muhammad al-Amin al-Kanemi, was born. Said to be a reputable mallam from Fahi, a village near Mao, Muhammad Ninka later married an Arab woman from Fazzan.[23] Kanem likewise produced Goni Musa Burmama, born in Mao in 1754/5. Although he came to Gazargamo for further study, his background points to the existence of Islamic scholars and prestigious religious figures in Kanem.[24] The region’s Kanembu peasantry also inherited certain customs from the Dalatoa or Bornoans. One of these customs, a form of mutual aid called debdou, illustrates how the peasantry engaged in mutual aid by sharing resources. The custom may very well have had deeper Kanembu origins, but the Dalatoa and their allies were believed by Bouillé to have brought it to Kanem.[25] The growing insecurity of the 19th century unsurprisingly shaped settlement patterns, too. The evidence for that in the 17th and 18th centuries is ambiguous, but the expansion of the Tubu and Arab presence may have fueled more conflicts as central authority weakened with the gradual decline of Borno. 

In conclusion, the political, economic, demographic, and social picture of Kanem during the 1600s and 1700s is one of great dynamism. While some continuity in administrative structure is evident through the deeper common history of Kanem and Borno’s origins under direct Sayfawa administration, Kanem in particular witnessed the fall of the Bulala sultans, the brief rise of the Tunjur, and the subsequent ascent of the Dalatoa at Mao. Further, Kanem’s economy likely became even more centered on pastoralism as ecological shifts and the immigration of various Arab and Tubu groups transformed the region’s demography. Nonetheless, ongoing trade, links to Borno and trans-Saharan networks, as well as the trade in livestock, horses, and salt likely made its economy dynamic. To what extent, if any, the alifas controlled artisanal production by the Haddad, a caste group, is unclear but another possible factor for Kanem’s political economy. In short, there was much more going on than what Zeltner initially labeled the siècles obscurs.

________________

[1] Zeltner, Pages du histoire du Kanem, pays tchadien, 193.

[2] Nachtigal, Sahara and Sudan III, 9-11, 21.

[3] Conte, Marriage Patterns, Political Change and the Perpetration of Social Inequality (in South Kanem, Chad), 120-122.

[4] Ibid, 129. 

[5] Robert Bouillé, Les coutoumes familiales au Kanem, 279. 

[6] Nachtigal, Sahara and Sudan III, 74-75.

[7] Bouillé, 220, 240.

[8] Nachtigal, Sahara and Sudan II, 252-253.

[9] Landeroin, “Notice historique” in Documents scientifiques de la Mission Tilho, 1906-1909, 392.

[10] Bouillé, 168.

[11] Ibid, 189.

[12] Ibid, 192.

[13] Ibid, 193.

[14] Ibid, 196.

[15] Landeroin, 380. 

[16] Ibid, 381.

[17] Zeltner, 193.

[18] Conte, 60.

[19] Zeltner, 194. 

[20] Landeroin, 391.

[21] Association for Promoting the Discovery of the Interior Parts of Africa, Proceedings for the Association for Promoting the Discovery of the Interior Parts of Africa I, 129.

[22] Landeroin, 389.

[23]Brenner, The Shehus of Kukawa, 39. 

[24] Bobboyi, The Ulama of Borno: a study of the relations between scholars and state under the Sayfawa, 1470-1808, 25.

[25] Bouillé, 146.

Garumele (Wudi) as a Sayfawa Capital

 

Plans of the Garumele site by Binet, Bivar & Shinnie, Haour and the area displayed with geophysics (Source)

One important site in modern Niger, Garumele, is said to have once been a Sayfawa capital. Heinrich Barth, drawing on oral traditions, believed it had been a royal site. Nachtigal and other others similarly believed this. Landeroin, for the Tilho Mission's Notice historique, compiled traditions that referred to the site as a capital of Ali b. Dunama before he established Gazargamo. Palmer, Urvoy, Bivar & Shinnie, Lange, Gronenborn and others have all, in one form or another, argued for Garumele (or Wudi, the name of the successor settlement) as a Sayfawa capital before Gazargamo. Despite its alleged importance in the history of the Sayfawa dynasty, the area was not seriously surveyed or excavated until recently. Fortunately, Haour, Magnavita, Robertshaw, and others have been able to examine middens, find faunal remains and ceramics, engaged in chemical analysis of glass bead finds and get radiocarbon dates from charcoal samples. The best evidence clearly shows use of the site for inhabitants in the 17th and 18th centuries, but was it a Sayfawa capital in the 15th or even late 14th century?

Satellite image of the site (Source)

The historical sources, primarily oral, strongly suggest a deeper antiquity for the site. Landeroin, for instance, was told that the Sayfawa went to Woudi (Wudi) after leaving Kanem (Landeroin 354). However, the site was said to have been unhealthy and caused frequent fevers. In connection to this tradition, a Magumi woman was said to have complained to Ali b. Dunama about Tuareg raiders. So, the mai left Mankinta in charge at Wudi while he attacked the Tuareg from Agram, or Fachi (355). While it is likely that oral tradition may have compressed a number of different reigns of Sayfawa monarchs, Tuareg oral traditions of Air do remember a period of Bornoan attacks and raids. Furthermore, the Kano Chronicle includes a reference to a Bornoan attack on Asben (Air). This suggests that the traditions associating the Mankinta as the leader of Wudi appointed by Ali b. Dunama may be accurate. Furthermore, Bedde traditions reported by H.R. Palmer similarly refer to the Mankinta (makinta) of Wudi in relation to the foundation of Gazargamo. In Palmer's Sudanese Memoirs, leaders named Makinta Baro, Makinta Yatku and Kaloma Arge were remembered as brothers who came to Borno from Wudi. The site of Wudi may very well have been under the administration of a makinta of Bedde origin who served the Sayfawa. This official may have been responsible for provisioning the palace kitchen of the mai. It is possible that in the 15th century, the fired-brick structure was used by the maiwa but a Bedde official was appointed to administer the territory. 

Garumele's outer wall of egg-shaped bricks in "Old Kanuri Capitals"

As for the earlier history of the site, it is currently impossible to determine if it was identical with the Kagha capital of Umar b. Idris (reigned c. 1382-1385) after the Sayfawa left Kanem permanently. Lange has suggested Kagha was likely the Kaka of al-Qalqashandi and the Jaja of Ibn Sa'id, but the bulk of the archaeological evidence points to a later occupation of Garumele. Another historian, Muhammad Nur Alkali, has also described Wudi briefly. Wudi was said to have been attacked by the Bulala in c. 1471, during the reign of Ali b. Dunama (Alkali 88). According to this historian, Wudi, also called Abadam, was a site that grew through its fishing industry (121). This fishing industry may have also been supplemented by a pastoral herding economy in the Sahelian landscape. Linseele's study of animal remains from Garumele noted the greater number of sheep than goats and has proposed that a pastoral nomadic element was a source of livestock for the site. This may have been why the site was said to be under the control of a makinta, an official who may have gathered fish and herds of sheep or cattle for feeding the Sayfawa royal court.

A view of the eastern tubali wall (Source)

What does the archaeological evidence suggest? Haour, using one charcoal sample that requires further confirmation by other types of evidence, has date ranges from 1280-1330 or 1350-1390 for Garumele (Haour 361). This would suggest that Garumele was either the Jaja/Kagha/Kaka of medieval Arabic sources and the Diwan or settled in the late 13th or 14th centuries. Despite this early date, the Garumele pottery was of a more recent type (366). It was also closest in type to the Yobe and Lake Chad sites north of the firki soils (368). Haour has even proposed identifying Garumele with Gatiga, a settlement mentioned by Ahmad b. Furtu (372). The Magnavitas, in "Garumele Revisited: Retracing Vanished Fired-Brick Elite Constructions and New Data on Settlement Foundation," found evidence of Garumele's occupation in more recent times. For example, radiocarbon dating of wood charcoal suggest construction of the town wall between 1459-1644 (166). The finding of a pipe fragment found in a layer was also consistent with a late 1500s or early 1600s foundation of Garumele as an enclosed urban settlement (172). Of course, this does not mean there were no earlier, non-brick structures and use of the site. Further evidence of more recent occupation of Garumele can be seen in the chemical composition analysis of glass beads by Robertshaw et al. Their conclusions suggest a 17th and 18th century occupation based on the European beads (Robertshaw 602). Overall, the bulk of the evidence favors a later date for the site's wall and brick structures. An earlier occupation in the 1400s, or perhaps even the 1300s, is still possible if one recalls the perilous state of the Sayfawa dynasty during an era of internal conflicts and wars with the Bulala rulers of Kanem or the Sao in Borno. Thus, the earlier Sayfawa use of Garumele as a capital or royal settlement may not have included fired-brick structures or walls. Later, however, after the consolidation of Sayfawa power in Borno, subsequent maiwa may have sponsored the brick structures at Garumele or Wudi.

Evidence of the widespread looting of Garumele for bricks (Source)

Despite the chronological uncertainties and the lack of dated bricks from the elite structure, the site of Garumele presents a number of unique features. The use of egg-shaped bricks for the outer wall is more similar to Hausa architectural practices. This raises the question of possible influences from Hausaland in this site, since it differs so markedly from the outer walls of other Borno towns or cities. Furthermore, the Magnavitas noted that the mounds or middens of Garumele included ruins of clay buildings or home structures. In addition, smaller, fired-brick structures outside the elite compound were present (Magnavita 165). This suggests the town included smaller homes of more perishable material as well as circular brick structures like that of Gambaru. The area was, of course, much smaller than Gazargamo, but Garumele's distinctive outer wall and fired brick structures indicate the site was definitely occupied by an elite group and may have had influences from the west. Acquiring dates for the elite brick structure may finally establish when the Sayfawa sponsored elite architecture at the site. Meanwhile, historians should revisit Kanuri and Bedde tradition and the written sources. If Garumele was not the Jaja, Kaka or Kagha of medieval sources, it was likely located nearby and this region near the northern shore of Lake Chad deserves close scrutiny for any understanding of early Sayfawa expansion by the 13th century. 

Bibliography

Alkali, Mohammad Nur. 2013. Kanem-Borno under the Sayfawa: A Study of the Origin, Growth, and Collapse of a Dynasty (891-1846) . Nigeria: University of Maiduguri.

Bivar, A. D. H., and P. L. Shinnie. “Old Kanuri Capitals.” The Journal of African History 3, no. 1 (1962): 1–10. http://www.jstor.org/stable/179796.

Haour, Anne, and Boube Gado. “Garumele, Ville Médiévale Du Kanem-Borno? Une Contribution Archéologique.” The Journal of African History 50, no. 3 (2009): 355–75. http://www.jstor.org/stable/25622052.

Lange, Dierk. Le Dīwān Des Sultans Du (Kānem-)Bornū: Chronologie Et Histoire D'un Royaume Africain (de La Fin Du Xe Siècle Jusqu'à 1808). Wiesbaden: F. Steiner, 1977.

Linseele, Veerle, and Anne Haour. “Animal Remains from Medieval Garumele, Niger.” Journal of African Archaeology 8, no. 2 (2010): 167–84. http://www.jstor.org/stable/43135516.

Magnavita, Carlos, and Sonja Magnavita. “Garumele Revisited: Retracing Vanished Fired-Brick Elite Constructions and New Data on Settlement Foundation.” The African Archaeological Review 34, no. 2 (2017): 155–75. http://www.jstor.org/stable/44988621.

Palmer, H.R. Sudanese Memoirs: Being Mainly Translations of a Number of Arabic Manuscripts Relating to the Central and Western Sudan. 1st ed., new impression. London: Cass, 1967.

Robertshaw, Peter, Marilee Wood, Anne Haour, Karlis Karklins, and Hector Neff. 2014. “Chemical Analysis, Chronology, and Context of a European Glass Bead Assemblage from Garumele, Niger.” Journal of Archaeological Science 41 (January): 591–604. https://doi.org/10.1016/j.jas.2013.08.023.

Tilho, Jean (editor). Documents Scientifiques De La Mission Tilho. Paris: Imprimerie Nationale, 1910.

Kanem's 13th Century Eastward Expansion

A map of Kanem at its apogee based on a problematic map by Waziri.

One topic in the annals of medieval Kanem that remains poorly understood is the eastward expansion of the 13th century. The eastward expansion likely began during the reign of Dunama Dibalemi (1210-1248, according to the chronology of Lange). Kanem's influence spread as far east as the land of the Tajuwa and Zaghawa. However, the nature of this control or influence is not clear. Nor can one easily determine its longevity or the reasons for why the Sayfawa maiwa believed extending this far was a worthwhile endeavor. Fortunately, some clues to the nature of this eastward expansion have been possibly revealed through chemical analysis of beads at the site of Njimi in Kanem. Data from stone ruins in Darfur dated to the era of Daju rule as well as evidence from Christian Nubian provide additional clues.

First, let us begin with the early written sources. The most detailed account from the 13th century, by Ibn Sa'id, draws from the earlier writings of Ibn Fatima. Ibn Fatima traveled to Kanem during the reign of Dunama Dibalemi. Writing later in the 13th century, Ibn Sa'id included the "Zaghawa" and "Tajuwa" as subjects of Kanem's Muslim ruler. Ibn Sa'id's brief description of this vast region is unfortunately limited, one can make some sense of it. The Zaghawa, who lived east of Manan (in Kanem) were said to be mostly subjects of Kanem. Tajuwa, said to be the capital of the Zaghawa, adopted Islam and accepted Kanem's suzerainty. Yet, later on, Ibn Sa'id wrote that the Tajuwiyyin were pagans who were refractory to the sultan of Kanem and kept to the deserts and mountainous terrain (Hopkins & Levtzion 189). This confusing account suggests that the land between Kanem and Nubia included Zaghawa and Tajuwiyyin peoples. The latter had a capital where Islam was accepted and the ruler recognized the authority of Kanem's sultan. It is very likely that Ibn Sa'id was also using the 12th century work of al-Idrisi here, particularly in his account of a town called Tajuwa. For al-Idrisi, however, the Tajuwin were pagans who kept camels and were raided by their neighbors (124). One of their towns was Samna, which was raided by the ruler of Bilaq (in Nubia). The news of this attack reached travelers to Kawar, who later related it to al-Idrisi (114). If Ibn Sa'id's muddled narrative is reliable, the town of Tajuwa (the capital of the Daju polity?) included Muslims and recognized Kanem's rule, yet other groups living in the region of Darfur (or its surroundings) were largely autonomous of Kanem or sought to maintain their autonomy.

Besides Ibn Sa'id, other sources merely reemphasize the Tajuwa as a type of "Zaghawa" (Ibn Khaldun) or, in the case of al-Maqrizi, add a few details. The latter wrote of a place called Tukama at the beginning of Taju land. Word of the Taju's penchant for working in stone and their wars against the enigmatic Wathku also reached al-Maqrizi in Egypt (354). Putting together all the aforementioned external Arabic sources, one can surmise that medieval Islamic geographers from the north saw the Taju as a branch of the Zaghawa. In addition, they lived between Kanem and Nubia and included Muslims among their pagan population. In the 13th century, some recognized the authority of Kanem while those living near deserts or the mountainous, rugged terrain, were able to at least resist Kanem's oversight. When one considers the vast distance between Darfur and Kanem and the rough terrain, outright military conquest was unlikely. Thus, it is perhaps more logical to see Kanem's interests in this region as one of securing an eastern trade route. Besides this trade route, occasional raiding outside of a small tributary zone may have been a common feature. Consequently, the "town of Tajuwa" with its Muslims may have been linked to the trade route while the rest of the area was very loosely controlled by Tajuwa.

In the absence of internal written sources from the 13th century, one must look to oral traditions and archaeology for further detail. Gustav Nachtigal, who traveled to Darfur and Wadai in the 19th century, collected oral traditions everywhere he visited. In the Sula kingdom ruled by a Daju dynasty, Nachtigal saw a list given by a prince which enumerated 21 rulers (Nachtigal 81). Another list of kings that Nachtigal saw included 13 Daju kings, 13 Tunjur kings and 22 Keira (Fur) sultans. Yet another list gave 5 Daju kings and 25 Tunjur and Keira rulers. The first Daju king was called Gitar (272). This early Daju kingdom in the Darfur region was based in Jebel Marra (273). On the other hand, Nachtigal was also told that the first Daju ruler was named Kosber, who lived in Debba. And in the first list of 21 Daju sultans Nachtigal received from a Sula prince, 6 were said to have been pagans based at Jebel Marra. These Daju kings were said to have controlled Darfur until the Tunjur, Ahmed el-Maqur, seized power (274). Based on Nachtigal and subsequent authors, O'Fahey's study of the Darfur Sultanate endeavored to use all known traditions on the Daju and Tunjur kings of Darfur to reconstruct a more coherent chronology. O'Fahey was able to locate written sources that allude to a Tunjur dynasty in Darfur by the 16th century (O'Fahey 31). Furthermore, the Daju title for a chief, shartay, survived in Darfur among the Keira sultans (37). This roughly suggests that the Daju kings in Darfur rose to power sometime before the 1500s, probably in the 1100s or 1200s.

In terms of archaeological evidence from Darfur, Andrew James McGregor has produced the detailed work to date. According to McGregor, the Daju rose to power in the 12th or 13th century (McGregor 34). The Daju were the Taju pagans of al-Idrisi, with only 2 towns. Moreover, al-Tunisi apparently believed the Daju were one of the 5 aboriginal Wadai groups 40). Yet as a polity, the Daju kingdom was remembered in Tunjur tradition for primarily subsisting on raiding its neighbors (47). Intriguingly, McGregor found a manuscript source that assigned the arrival of the Daju to Sila to 1692 (39). This date, however, must be too late since Tunjur oral traditions in Kanem contradict it. According to Gros, who studied Tunjur oral tradition in Kanem, the father of the Tunjur ruler, Daoud el Mireim, Omar, had subdued the Daju of Sila (Gros 274). If so, and Daoud el Mireim was the last Tunjur ruler of Wadai, then the Tunjur dynasty there would have conquered the Daju of Sila (Sula) by the early 1600s if not late 16th century. In other words, a Daju presence in Sila (Sula) likely predates the 17th century. Anania's late 16th century description of a polity in Darfur called Uri, presumably the Tunjur dynasty's capital or commercial town, included Dagio (Daju) among its subjects (Anania 343). 

So, based on oral traditions and written sources, we know there was a Daju polity in the Darfur region by the 13th century. It was said to have been based on Jebel Marra and likely raided neighboring groups in Darfur and perhaps further afield. Although the early rulers were said to be pagans, Ibn Sa'id believed that Tajuwa, the capital or principal town, was a Muslim city. Since the kingdom of Makuria allegedly raided one of the Daju towns in the 12th century, it is possible the polity had also disrupted a trade route or attacked parts of Kordofan closer to the Christian kingdom. Overall, sadly, it is difficult to reconcile the sources on this enigmatic polity. Yet it's rather loose administrative structure and reliance on raiding may be clues for a weak polity. If this was the case, why was Kanem interested in conquering this region?

Recent archaeological evidence and chemical analysis of artifacts found at the probable site of Njimi may provide the answer. The article "LA-ICP-MS analysis of glass beads from Tié (12th–14th centuries), Kanem, Chad: Evidence of trans-Sudanic exchanges" noted that the chemical composition of beads at Njimi were more similar to medieval beads found in eastern African sites than those of West Africa. While not the best smoking gun, this pecularity for medieval Kanem suggests that most of its beads were ultimately derived from the same sources of Eurasian beads that arrived in places along the Swahili Coast, southeastern Africa or the Red Sea coasts of Africa. This pattern, so different from sites in West Africa in the 10th-15th centuries, implies that Kanem was ultimately part of the Indian Ocean network that brought high alumina soda glass beads to eastern African sites. Since these types of beads were not common west of Kanem, despite trade routes through the Fazzan connecting western Africa to the Mediterranean, one must wonder if Kanem accessed the beads through a separate route to Egypt or Nubia. While further confirmation of the analysis of glass bead chemical composition in Makuria and Alwa is necessary, it is possible this eastward route from Kanem connected to the Red Sea/Indian Ocean networks through Nubia (or Egypt). Possible indications of a route to Nubia may be found in the analysis of beads at Soba, the capital of Alwa. In Soba, 20 high alumina beads of Indian origin were part of a study of 23 glass beads.

Besides the data from chemical analysis of glass beads, studies of cattle remains suggest Makuria imported cattle from the west. According to "The cattle factor. Faunal evidence for the study of social and economic processes in the medieval Middle Nile Valley. Sudan" by Marta Osypińska and Mariusz Drzewiecki, Makuria imported long-horned cattle from the west, probably the Central Sahel or Kordofan. Their analysis of faunal remains suggests that Alwa practiced a more conservative livestock farming based on ancient Kush while Makuria may have imported as much as 1/4 of its cattle from the west. While there is no evidence for Kanem playing any role in supplying cattle from the Central Sahel to Nubia (Makuria), it is possible that both Kanem and Makuria had an interest in secure trade routes through Kordofan and Darfur. This facilitated access to Egypt (and Nubia) for Kanem while the Christian Nubian kingdoms could be guaranteed of some security for trade routes through Darfur to Kawar and North Africa. Evidence from Nubian graffito writings or inscriptions found in Kordofan, such as one attesting to King Siti and likely from the 1330s, likewise demonstrate Makuria's interests in Kordofan and perhaps Darfur to the west. 

In summation, Kanem's eastward expansion in the 13th century was likely related to economic interests that likely included access to Indian Ocean networks via Nubia (and/or Egypt). While the northern, trans-Saharan route through Kawar and the Fazzan to the Maghreb and Egypt was still very important, the unique chemical composition of beads found at medieval Njimi suggest the use of an alternative route for at least beads. Since the Daju polity of the 13th century was possibly decentralized and divided, Kanem may have been able to expand its influence and promote an Islamic community of traders at the town of Tajuwa, likely somewhere near the Daju polity's base in Jebel Marra. Direct control or tribute-raising on much of the region would have been difficult due to the Zaghawa and Daju "pagans" who used the deserts and mountains to resist Kanem. Nonetheless, for the region to retain its significance to Kanem's rulers, Kanem may have engaged with the Daju polity for more security of the trade routes that led to Nubia (or Egypt). The possible evidence of a Nubian use of this trade route, or at least use by Muslim traders from Makuria and Alwa who also traded with the lands west of the Nile, can be seen in the cattle imports from the Sahel in Makuria. Furthermore, Makuria's rulers raiding a Daju town in the 12th century may be due to their raiders disrupting trade passing through the region. Lastly, the chemical composition of several beads at Soba, the capital of Alwa, resemble those of Njimi in Kanem. This may be due to interest in Makuria and Alwa in maintaining the trade routes to the west. Although any evidence for direct engagement between Kanem and the Christian Nubian states is lacking, Muslim traders active in all three kingdoms may have been the link.

Bibliography

Courtecuisse, L (editor). Les Arabes du Tchad. Paris: Centre de hautes études administratives sur l’Afrique et l’Asie modernes, 1971.

Hopkins, J. F. P., and Nehemia Levtzion (editors). Corpus of Early Arabic Sources for West African History. Princeton, NJ: Markus Wiener Publishers, 2000.

Lange, Dierk, and Silvio Berthoud.  "L'intérieur de l'Afrique occidentale d'après Giovanni Lorenzo Anania (XVIe siècle)". Cahiers d'histoire mondiale, 14, no. 2 (1972): 299-351.

Magnavita, S., MacDonald, B. L., Magnavita, C., & Oga, A. (2024). LA-ICP-MS analysis of glass beads from Tié (12th–14th centuries), Kanem, Chad: Evidence of trans-Sudanic exchanges. Archaeometry, 66(1), 100–118. https://doi.org/10.1111/arcm.12898

McGregor, Andrew James. The Stone Monuments and Antiquities of the Jebel Marra Region, Darfur, Sudan (dissertation). University of Toronto, 2000.

Nachtigal, Gustav, Allan G. B. Fisher, and Humphrey J (trans.). Fisher. Sahara and Sudan, Vol. 4. Berkeley and; Los Angeles: University of California Press, 1971.

O’Fahey, R. S. 2008. The Darfur Sultanate : A History. New York: Columbia University Press.

Osypińska, Marta, and Mariusz Drzewiecki. 2024. “The Cattle Factor. Faunal Evidence for the Study of Social and Economic Processes in the Medieval Middle Nile Valley. Sudan.” Journal of Archaeological Science Reports 55 (April): 104513. https://doi.org/10.1016/j.jasrep.2024.104513.

Then-Obłuska, J., & Dussubieux, L. (2023). Overseas imports on the Blue Nile: Chemical compositional analysis of glass beads from Soba, Nubia. Archaeometry, 65(5), 1018–1031. https://doi.org/10.1111/arcm.12863