Thinking about the Magumi

A topic we have been thinking about whilst revisiting sources on the Bulala Sultanate of Fitri is the clan name Magumi. We had not considered it in much depth since encountering a gloss of its meaning as "free-born" in Koelle's dictionary. Although the name or word magumi very well could have also signified free-born status or people, it's also clearly associated with the Sayfawa Dynasty as the royal clan of Kanem and Borno. The form of the word could also be affiliated with an archaic version of the word mai. The -mi suffix at the end connotes descent from, as in "son of." Thus, Magumi should literally mean son of or descendants of Magu (or Mague? Maga?). This can be seen in the divisions of the Magumi named after various maiwa the branches claimed descent from (see Nachtigal's Sahara and Sudan for a list of them). So, if not originally a clan name, it emerged as one through the numerous descendants of the rulers of Kanem and Borno. Perhaps Magu, Maga, or Mague was the original form of the word mai? Or did it become the word for king due to the growing power of the royal clan?

Another form of Magumi survived among the Bulala royal clan as Maga. According to Henri Carbou, the Bulala in Fitri referred to themselves as Maga, Magga, or Maggue, clearly related to the Magu in the Kanuri Magumi. It also manifested among the Bulala memories of a leader named Ali Gatel Magabirni. Carbou also proposed an etymology for Maga based on the Tubu language, with Mâga as containing the word maï (sultan) and the  suffix ga to denote relation. Carbou may be more likely correct about the meaning of Toumagera (masters of the Tou, or Tibetsi). 

Yet one also notices speakers of other Nilo-Saharan languages seemingly referring to themselves with words that almost sound like Maga or Magu. For instance, in medieval Nubia, the people of Nobadia were Migi (ⲙⲓⲅⲓ). In Old Nubian, a word for people is Mugi or Magi, reminiscent of the Magu(mi) of Kanem and Borno. One can also point to the Maiga of the Songhay for the nobility. There may be no connection at all, but one wonders if the word could be related to something deeper in Nilo-Saharan languages.

The Bulala Sultan of Fitri

Whilst recently perusing a book on Chad, we saw the Bulala sultan of Fitri in Les treize préfectures de la République du Chad by Pierre Gentil (1962). Although we could not find his name in the pamphlet, we wonder if it is Oumar since the book was published in the 1960s. Or perhaps Hassan Absekin who became sultan in 1967 according to Hagenbucher. 

Kandakma and 'an.d.k.ma?

The title of Dunama, the man who killed Kaday b. Dunama (Augustin Holl, The Diwan Revisited: Literacy, State Formation and the Rise of Kanuri Domination (AD 1200-1600)). 

Something we have been reconsidering whilst re-reading parts of the Diwan is the dignitary who murdered Kaday b. Dunama (in c.1277, according to Lange's reckoning). As he indicated in his footnotes, the Arabic text suggests that the man who killed the mai, named Dunama, held a title or perhaps was a dignitary in the Sayfawa court. Lange, however, did not venture speculating on what the title may have been, yet his French translation rendered it as the above.

H.R. Palmer, in his Bornu Sahara and Sudan, endeavored to identify the title. According to Palmer, who did not explain his reasoning, it was a "Dikuma" who was supposedly the chief of the N'Galma Diku (N'Galma Duku). Using evidence from the gargams, Palmer also believed that Kaday (or, Kadai) was killed at Zui Zui in Borku. 

Muhammad Nur Alkali was apparently influenced by Palmer's thinking here for he raised the possibility of Zuwi Zuwi being an oasis known as Zau Kura along the Kawar-Tripoli trade route. Thus, he rendered the title in the Diwan as "Ndikuma." This is more reminiscent of the Dirkuma of the Tura and difficult to reconcile with Palmer's claim the title was held by the chief of the N'Galma Duku. 

As for Heinrich Barth, he wrote the title as ʿAndákama. Barth's reading is close to Lange's. Here is how the Arabic text looks for this part of the Diwan on page 38 in Lange:

Lange wisely avoided a guess at the possible meaning or origin of the title. We, however, wonder if this obscure title or official disappeared in the Sayfawa court but survived among the Bulala sultans in Kanem (and later, Fitri). Our evidence is slim, but Hagenbucher's study of the Bilala of Fitri found a title called kandakma. Described as the premier adjoint of the galadima, the title was occupied by the N'guidjim. Part of the Birni dignitaries of the sultanate, they were on the "left" of the court. Among the Birni or "people of the left" were the N'Guidjim, Kassawa, Gaswa, Madawa, Fadawa, Kiliti, and Argumba. Since many of these titles appear to be of Kanembu/Kanuri origin, and the Bulala sultanate in Kanem likely borrowed many titles from the Sayfawa court, we wonder if the kandakma is another borrowing. Even if we are correct, we cannot automatically assume the office meant the same thing given the vast chronological distance between 13th century Kanem and 20th century Fitri.

Another obstacle is explaining the absence of the initial k in the Diwan. One wonders if what occurred here is a case in which the prefix was not included by the copyist of the Diwan through an error or differences between spoken forms of Kanuri in medieval versus more recent times? Assuming the Bulala spoke Kanembu during the centuries their sultanate controlled Kanem,  and the adoption of a Naba or Kuka-related language began after their exodus to the Fitri region, one might expect kandakma to be closer to an early Kanuri/Kanembu pronunciation of the word. Alternatively, there may be no connection whatsoever between the Bulala kandakma and the 'an.d.k.ma in the Diwan. But if the two words are indeed linked, which seems plausible, then the dignitary who killed Kaday b. Dunama in c. 1277 may have been someone who served the galadima and was close to the court. This, of course, assumes the deep antiquity of the title galadima and the post possibly being connected to the N'guidjim of Kanem (according to Carbou, the chief of the N'guidjim in Kanem was called galadima, although the uncertain but close ties to the Bulala may mean one cannot assume continuity from the earlier Sayfawa period in Kanem).

Another one of our incredibly speculative thoughts on this title is a possible link to a word in Kanuri, kə́ndə́gái, which the English–Kanuri dictionary (by Norbert Cyffer) gives for a reception room in a palace. Hutchison's study of Kanuri word formation analyzes kə̀ndə́gái ("royal reception room") and kə̀ndə́gà ("living") as derived from the verb dəgá, "live." If the Diwan's ʿan.d.k.ma and the Fitri kandakma reflect a form like kəndəg(a)-ma, the consonants k-n-d-g-m would account for the whole skeleton. The formation has a close parallel. Cigama (cikama), the Kanuri "gatekeeper" or chamberlain, from ciga, "town gate," plus -ma, so an office named after a part of the palace is exactly the kind of title that could explain the kandakma. We must stress that kəndəgama is not attested, that Lange has claimed control of audiences to the Cikama and correspondence and visitors to the Digma. If the link holds, the kandakma would be a palace-household officer attached to the reception room, with the kind of access that could explain how he reached Kaday.

Ultimately, our theory remains a tentative one. Clearly, Lange himself did not propose this in his study of the Diwan, and to our knowledge, the title has not been clearly elucidated or identified. The -ma suffix and the structure of the sentence suggest a title, and the overlap with the kandakma of the Bulala sultanate in Fitri is rather striking.  As for what the function of the 'an.d.k.ma may have been, it was perhaps a title for a first deputy of the galadima who may have been a "first minister" and thus close associate of the mai who advised the ruler. This is purely speculative, but one wonders if the word galadima is etymologically linked to an Old Kanembu word for advise (galā in Bondarev's dictionary of Old Kanembu). The kandakma Dunama was thus someone who was likely closely entwined in court affairs and possibly related to Kaday. Since Kaday's successor was his brother Bir, it is not inconceivable that 'an.d.k.ma Dunama was acting on behalf of Bir's faction. 

A Bulala Attack on Bilma

One of the interesting historical references in Maurice Abadie's Afrique centrale; la colonie du Niger is a possible Bulala raid that culminated in a massacre of people near Bilma, in Kawar. The alleged attack occurred in the 17th century, likely at a time when the Bulala sultans were already in the region of Fitri instead of Kanem. Thus, the attack probably took place when the Bulala were at a further distance from Kanem. Unfortunately, we could not locate any sources to corroborate this tradition. However, Nachtigal, in Sahara and Sudan, wrote a history of Bagirmi's kings based on informants who were descendants of the principal actors involved. One of the intriguing moments in his account of the reign of the Bagirmi king Burkomanda, whose regnal dates were probably 1635-1665, included far-ranging raids that reached Borku and Kawar. This same mbang of Bagirmi had a sister, Zara, married to the Bulala sultan. It is possible that the Bulala raid which massacred people in Bilma assembled for prayer was actually part of one of the Bagirmi king's expeditions.