Ch. Edwige's "Notes sur la subdivision de Melfi" is a report from the 1920s based largely on oral traditions. While certainly aware of earlier works on Bagirmi's history, such as Nachtigal, Edwige's list of Bagirmi kings before the 19th century is much shorter. Perhaps due to the nature of his informants in Melfi, which included Sokoro, Kenga, and Dekakire, much of Bagirmi's pre-1800s past was not retained in local traditions. Exceptions, of course, could be found. For instance, the chiefly lineage of the n'gares of the Sokoros claimed to know the names of about 13 different chiefs. Besides this family lineage which inherited a position, the heterogeneous population of Muslims and non-Muslims provides fewer instances of this type of well-established genealogy.
Despite the obvious errors and incomplete nature of Edwige's list of Bagirmi kings, it is nonetheless a possible opportunity to explore how traditions from this area of the kingdom perceived their distant past. In some ways, it follows the general narrative with occasional omissions or what may be legendary traits added to the tradition. Take Dalou-Birni, for example. Clearly identifiable with Nachtigal's Dalobirni, who reigned in the 17th century, he is also meant to be Birni Besse, one of the founding figures of Bagirmi's royal dynasty. Edwige's informants added extra details. First, according to Edwige, Dalu was of Arab origin from the Kuraysh (though it is telling that his brothers included one named Kenga, perhaps an admission of the shared roots of the Barma and Kenga). This Middle Eastern origin story is hardly surprising. But Edwige also shared a more narratively rich background for Dalou-Birni, whose accentuated color sparked the feeling from his father that his mother had committed adultery. Perhaps this is an attempt to explain the darker skin of Dalou-Birni despite his Arab father. After his father threw the infant Dalou-Birni into a fire and he miraculously survived, his father gifted his mother 100 cows. This is an attempt to link the etymology of the toponym Bagirmi with the words "Bagara mia!" that his mother shouted after the gift.
Later on, upon arriving in Bagirmi, Dalou-Birni attracted followers through his skills as a hunter. However, a local ruler of the Bidili (Bideri?), which included locals already in the region such as the Birkete, was forced to pay a tribute to the Bulala sultan of Fitri, Saboun (is this tradition remembering Wadai's Saboun, who invaded Bagirmi in the early 1800s?). Since the tradition also turns the Bulala rulers into tributaries of Wadai, one must suspect we are dealing with anachronisms. After all, if one uses the conventional dating for Dalou-Birni and early Bagirmi, the 16th century, there was no Wadai. Indeed, the Bulala sultans would have been based in Kanem in the early 1500s. Unless one wishes to follow Carbou's theory, which would turn the Fitri-based kingdom which imposed tribute on Bagirmi into the Kuka kingdom, what exactly was going on in the area that became the center of Bagirmi's royal dynasty remains unclear. A connection to the Bulala sultans, however, may make more sense given their reference to Bagirmi in the context of a Bulala sultan in the 1500s by Ahmad b. Furtu.
Regardless, Dalou-Birni's success at hunting and sharing of the meat continued to attract followers. Meanwhile, he also began preparations for ending the payment of tribute to Bagirmi. Dalou-Birni prepared the revolt by secretly preparing shields and spears to use against Saboun. When 30 horsemen were sent by the Bulala to collect the tribute (in strips of cloth and clothes), Dalou-Birni and 100 of his followers attacked and killed most of them. Unsurprisingly, Saboun later sent 300 horsemen to plunder the herds of the Arabs and Fulani in Bagirmi. Once again, Dalou-Birni defeated these forces and became king upon his return to Massenya. As king, he imposed tribute in horses to strengthen the army and gather the tribute in strips of cloth. It seems unlikely that he would not have engaged in any slave raiding, but supposedly he avoided it.
Unfortunately, the list of subsequent kings by Edwige omitted so many names that it would only go back as far as c. 1660 instead of the early 1500s. This meant that for Edwige, Dalou Birni must have lived in the 1600s whilst his successors only included Allah Hadji I, Laouan, Allah Hadji II, Gaourang I Kabir, Bourkou-Manda-Kabi, Abd-el-Kader, Mohamed Abssakin, Bourkou-Manda and Gaourang II. A few of these names are instantly recognizable, although others are less so (Allah Hadji). His informants may have telescoped many kings from the period before the 19th century. One sees this through Allah Hadji I's reign, who is presumably based on Abd Allah (r. 1568-1608 by Nachtigal's reckoning). However, this Allah-Hadji I of Edwige appears to be a composite figure combining a number of different Bagirmi kings and their exploits. Thus, Allah-Hadji I was said to have begun to attack the Massa, but also attacked the Saroua, Boua, Logone, and imposed vassal status on Logone, Kousseri and Goulfei. Through a successful siege of the town of Mahamat, the Kotoko king, these aforementioned Kotoko towns became vassals of Bagirmi.
The next named king, Laouan, easily recognizable as Loen or Loel, would have reigned in c. 1741-1751. Like Loel, this Laouan was killed and replaced by his younger brother. Although not named correctly, this brother was Muhammad al-Amin, whose reign Nachtigal placed from c. 1751-1785. Here is where traditions become interesting for Hadji. This mbang was living in exile in a place called Sounar. Presumably, Sounar is what Edwige's informants thought to be the city of Sennar to the distant east. However, tradition here has turned Muhammad al-Amin's life in the Eastern Sudan into one of becoming a local dyer in a village and marrying the chief's daughter. Details also diverge on the tale of his return. Since his "brother" Laouan was so oppressive, one of his musicians sought Allah-Hadji (Muhammad al-Amin), working together with him to return to Bagirmi and seize the throne. Needless to say, this version is quite distinct from Nachtigal's retelling, although it follows the same basic sequence. Indeed, Allah-Hadji uses Moito as a base and then defeats Laouan at Sado before becoming ruler of Bagirmi.
Additional problems can be found. Unlike Nachtigal, whose informants stressed the cruelty of Muhammad al-Amin and his numerous military expeditions, Allah-Hadji II was liked by his subjects and nothing else stood out from his reign. A similar problem emerges with the reign of Gaourang-Kabir, the son of Muhammad al-Amin. Absolutely no mention is made of the Wadai invasion launched by Saboun or any of the military ventures of these rulers.
Clearly, Edwige's recounting of Bagirmi history before the 19th century is very flawed. But the specific area he collected information in may have played a role. Despite these aforementioned issues, his article nonetheless raises new questions about how the Bagirmi mbangs were remembered in different parts of the kingdom. Nachtigal, who was able to collect information from political elites tied to the court of the mbang, could access a more controlled royal genealogy in somewhat informal settings. Edwige, on the other hand, may have heard a less official version which, in spite of its omissions, still coherently overlapped with parts of the "official" version Nachtigal heard from elite informants.