Navigation – Plan du site
études et essais


A Soninke Muslim Diaspora in the Mande World
Andreas W. Massing
p. 887-922


Baghayogho, une diaspora soninké dans le monde mandé. — Cet article retrace le rôle et l’implication d’une seule famille — dont les membres se réclament les derniers des Wangara aujourd’hui — dans la réformation de l’islam suite au pèlerinage du roi Askia Mohammed et dans l’introduction de la loi islamique aux villes maliennes du xviie. Aussi semble-t-il que la « mission » auprès des chefferies des zones aurifères des régions forestières — dans des zones aussi diverses que le Gabu, le Konyan de la Guinée, le Worodougou, les royaumes mossi et dagomba — est basée sur un dessin conscient de la part de la dynastie songhay et de ses conseillers musulmans de convertir, à partir d’une famille de savants de Tombouctou, les animistes du Sud à un islam qui abandonne les traditions maraboutiques de la première vague de l’islamisation, et fondée sur les anciennes connections commerciales des Wangara avec les zones de production d’or et de kola. Le mythe du pèlerinage de savants de plusieurs dyamou — tous d’origine soninké — indique que chaque famille de savants a choisi une « aire de mission » et celle des Baghayogho s’est concentrée sur les axes fluviaux le long du Bani et de la Volta pour réformer l’islam chez des peuples aussi divers que les Bambara (païens), les Senoufo, les Mossi, et les Dagomba. Des recherches restent à faire afin d’établir les liens concrets entre les fondateurs des branches dans la diaspora avec la branche principale à Tombouctou.

Haut de page

Texte intégral

1The present study of the Baghayogho1 family covers a particular aspect of the Mande diaspora namely the history of a particular family involved as agent for the expansion of Islam in West Africa2. Previous work on the Wangara took me to Djenné, Timbuktu, the Wasulu, the Ivory Coast and Ghana, with some sources pointing even to Nigeria and Benin3. I became interested in the Baghayogho family and its diaspora in West Africa through their last representatives in Timbuktu. In writing the history of this particular family I would like to establish a model dyamou—history—and plan to follow up with others. Third, as the Baghayogho are closely tied to the propagation of Islam in West Africa, I aim to demonstrate their strategy. It appears that during his jihad against the kharedjite dynasty of Sonni Ali, Mohammed Askia and his Muslim advisors designed a strategy to convert the animists which was further elaborated during and after his pilgrimage to Mecca; the strategy was based on the century-old Wangara long-distance trade network used for the penetration and establishment of Muslim colonies in the south through the conversion of traditional ritual practitioners of the Komo, with their knowledge of divination, magic and sorcery, and their adoption into what I may call a “maraboutic” Islam. This led to a first wave of conviction or conversion of chiefs4. I have previously shown the early spread of Islam through the Wangara network to have occurred between the 13th and the 15th centuries, but want to document here its expansion during the 16th and 17th centuries through the Baghayogho, among others, giving rise to Soninke-based diasporas such as the Diakhankhe or Maninka-Mory. Wilks (1963: 87-98) and Levtzion (1968: 171) have previously documented the ramifications of the Baghayogho family among such Mole-Dagbane groups as the Mossi and Dagomba, and Marty (1922) did this for the Ivory Coast and Guinea, while Skinner & Giesing (Conteh et al. 1987; Giesing 2003) have shown the network in Guinea-Bissau. Hunwick (1990) can be credited with the first biography of a Baghayogho, namely Mohammed I (1524-1596), but he believed the family to be of Malinke stock even though the sources are clear on their Wangara-Soninke origin.


2The African gold trade was well established as the southern link of a chain whose northern link were the salt caravans, described by V. Fernandez (1506) (Baiao & Bensaude 1940) for the 15th century5. The Azelaï or Idelaï left Timbuktu for Teghaza and later Taodenni, and returned with camels laden with salt plates which were shipped to Djenné to be exchanged for gold brought from the mines in the south by Wangara merchants. Ibn Battuta mentions a group of black merchants, called Wanjarata, ten days south from Oualata6. The Tarikh el-Fettach (Houdas & Delafosse 1981), exploiting unknown earlier sources, links the Wangara to the royal Kayamagan dynasty of Ghana—of 22 kings before the Hejra and 22 kings thereafter. It quotes Demir el Yakoub who claims the Songhay, Wakore and Wangara to be descendants of a certain Taras ben Haroun from Yemen, who died and left his sons in the hands of their uncle. They emigrated, and the elder son, Ouakoré-ben-Taras, became ancestor of the Wakore, while the second, Songai-ben-Taras, became ancestor of the Songhay, and the youngest son called Ouangara-ben-Taras became ancestor of the Wangara, while one of his slave-girls called Meinga became the ancestor of the Maïga, the Songhay royal clan7. They scattered across the land, and Wakore became their king and took the title Kayamagan “whose heritage will last long” (ibid.: 40-41), or according to another etymology “king of gold”, kaihou meaning gold, and maga king. “Formerly the chief of Mali was one of their vassals like one of their high functionaries or ministers. They were among the most powerful kings. Their capital was Kombi” (ibid.: 74). “Qaiamagha was the first prince who ruled in this region and the capital was Gâna, a big city in the Baghena country […]. They were a white race but we ignore where they trace their origin. But as for their subjects, they were Wakore. When this first dynasty disappeared it was replaced by that of Melli whose princes were of the black race” (ibid.: 18). “There is no agreement on which tribe their princes (of Kayamaga) belonged to; was it that of the Wakore, like some say, or the tribe of the Wangara, like some others state; according to others they were Senhadja, which is closer to the truth. What is most certain is that they were not blacks” (Houdas 1981: 78).

3Curiously, the Tarikh further claims that all chiefs like the Mossi-koï, Djenné-koï, etc. were from the west, from the Wakore or the Wangara (Houdas & Delafosse 1981: 94)8. When in 1509 the Askia Daoud made an expedition into the Ouagadou, he captured the cattle and the Zendj9, who spoke Wankore and took them to Gao (Houdas 1981: 218).

4For 1468 the Tarikh es-Sudan reports: “We have learnt that in the Tekrur country a person has appeared under the name of Sonni Ali who kills the men and pillages the country. This person began to rule in 869 (1464/1465). He occupied the throne for 27 years” (Houdas 1981: Chap. 12). Ali Kolon, son of Sonni Mahmoud Da’ou10, came from the dynasty of the Za (Da’o?), i.e. of the Kayamagan of Tekrour and took the title Sonni Ali. With his brother he had been a hostage at the court of Mali where they had been raised until their adulthood11. The circumstances of the rise of Sonni are unclear but could be seen in the context of insecurity in the northern provinces of Mali, threatened by the Mossi and Touareg. Mali was no longer able to defend these provinces, and the rich merchant communities looked elsewhere for help. It is likely that Timbuktu called on Sonni Ali in order to get rid of the Touareg, as its sheikh Mohammed Naddi offered to surrender the city.

“The end of Touareg rule was marked by odious and limitless exactions and acts of violent tyranny. The Touareg caused desolation everywhere. Sultan Akil forbade the inhabitants of Timbuktu to pay their customary dues to the Timbuktu-Koï. […] Very irritated by these words, the governor employed all means to revenge himself with the sultan. In secret he sent a messenger, Sonni-’Ali, to commit him to come to Timbuktu, and promised him to deliver the town into his hands and make him the sovereign. […] Akil resolved to flee and left with the jurisconsultes of Sankore to gain Biro. The authority of the Touareg had never extended beyond the River. The Timbuktu-Koï promptly sent boats across to allow the cavalry to cross the river. When Sonni-Ali arrived coming from the direction of Haoussa, Ammar took his turn to run away to Biro, fearing that Sonni-Ali wanted to punish him for the resistance which he had made previously” (Houdas 1981: 40-41).

5Mohammed Naddi had died in 1468 and left as his successor his son Omar who was apparently opposed to Sonni Ali. Touareg Sonni Ali persecuted the ulema (religious leaders) of Sankoré probably for religious reasons12 and most of them fled to Biro, the commercial section of Oualata. Sonni Ali then named Mokhtar Mohammed Naddi, Omar’s brother, Timbuktu-Koï, thus confirming the succession of Mohammed Naddi, and sent him in pursuit of Omar and the ulema to battle them at Tadjit. Omar was later reinstated because Sonni Ali fell out with Mokhtar and imprisoned him around 1487, and Omar had been active in pursuing the Mossi and recovering most of the prisoners (July-August 1480). This occurred during the second Mossi invasion under the 12th Naba Niandefo (1441-1511) to Oualata which enslaved its inhabitants. In 1483, Sonni Ali battled the Mossi south of the Niger and in 1485 even launched an expedition to the frontiers but perished on his return from another campaign into the Gourma against the Fulani, in a late flash flood (6 November 1492).

6His son Sonni Abou Bekr Da’ou, also called chi Baro, was proclaimed successor at Diakha on 21 January 1493, but was challenged by Mohammed ben Abou Bekr et-Touri, the provincial governor of Gao the later Askia Mohammed. It seems that he declared a jihad because Sonni Da’ou was twice in vain asked to convert to Islam by the Askia’s envoy Salih Diaoura (Houdas 1981: 118). Da’ou lost the battle at Diakha where ten of the ministers or princes had joined him (April 1494), following which Mohammed was proclaimed king. Omar Komjago, Mohammed’s brother, took Diakha in 149413. Mohammed el-Askia made the orthodox Muslims return from Biro to Timbuktu and ordered the release of Mokhtar ben Mohammed Naddi who, however, had died in prison after seven years’ captivity, and thus reinstalled Omar as Timbuktu-Koï.

7Mohammed el-Askia’s reputed origin was in the Futa Toro from Silla, a Soninke settlement and clan west of Kaedi, but according to others he was born in Zinder in Hausaland, and had an affinity with the Wankore. One of his wives, Mariam Dabo, mother of the Askia Ismail, was Wankore (Houdas 1981: 152, 161). It was Ismail who liberated his father in 1538, ten years after being overthrown and blinded by his son Musa.

8El-Askia appointed a number of Wakore to important functions: thus Mour Salih Djaoura (Diawara, or Soumaré?, in my view the famous Salim Souaré) became his councillor and was sent on religious missions to Sonni Da’ou and the Mossi kings to convince them to accept Islam14. Salih had been born of Wankore origins in Tauta-allah near Tendirma, a Wangara town 10 km west of Tonka15, and after “jihad” against the “infidels” was given the task to organize the hajj (pilgrimage) for Mohammed el-Askia.

The Wangara and Baghayogho of Djenné

9After the conquest of Djenné16, El-Askia made his pilgrimage to Mecca in 902-903 A. H. (1496/1497-1497/1498)17, and it was on his return that he presumably extended his jihad to the animist Bozo and other Djennenké and appointed a Muslim judge.

“One of the Sultan’s dignitaries had a dream, three times consecutively, that this man from Toura would, wherever he and his descendants settled, save the town from trouble, and that his grave would protect it against all attacks. […] Sidi Mohammed. b. Omar b. Mohammed b. Aqit came to Djenné, he saw Foudiya and admired much his conduct. […] The sultan, when he heard of the dream, ordered the man to be brought to him and asked him to remain in Djenné and settle there” (Houdas 1981: 31)18.

10Apparently the faqih was considered a powerful marabout and the Askia Mohammed invested Foudiya with the office of cadi (judge) in 1498-1499. He at once “began to destroy the temple of the idol, which the pagans had adored, and razed the houses, which were found in his walls” (ibid.). This I believe to be the shrine where the Bozo had buried Ta Jenepco, a Bozo virgin sacrificed in order to prevent the new Djenné from attacks by evil spirits19. Thus Foudiya Mohammed Fodiki Sanou, al Wankori20 became the first cadi of Djenné21 having come from Bitou at the end of the 9th century A. H.22 and “established himself at Toura, halfway between Chini23 and Djenné, where he went every Friday to fulfill his religious obligations, without people knowing who he was” (ibid.). The location of Bitou is somewhat in doubt: Wilks believes it to be Bitugu or Bigo, the Diula merchant settlement preceding Bondoukou, and marking for the Tarikh es-Sudan the southern limit of Mali (Houdas 1981: 30-33), but the name could also refer to Biro-Oualata24.

11The political circumstances of the Foudiya’s investiture were those of the first conquest of Djenné by Sonni Ali in 1468 and its re-conquest by Mohammed Askia in 1494, because Djenné had remained loyal to the Sonnis and to its traditional animist beliefs. This appointment marks the introduction of Islamic law by the Askia, who no longer trusted the local chiefs to administer justice, because of their allegiance to the Sonni dynasty and its religious heresy—the legitimate successor still being a refugee in the neighbourhood25. Even in 1656 the Djenné-koïs still adhered to kharedjite doctrine26.

12Thereafter four of the thirteen cadi of Djenné in the Songhhay period (1493-1593) were of Wankore origin27. Mohammed Sanou appears to be the grandfather or father of Mahmoud Baghayogho, the third cadi, and Benba (Bamba?) Kenati was “the last of the Sudanese dynasty” (Houdas 1981: 243-245)28, because he led a revolt against the Askia in 1591.

13Not much transpires of the next twenty years of Askia Mohammed until his overthrow in 1518. His son Musa usurped power in 1528 but was assassinated in 1531, and Mohammed Bonkano, nephew of Mohammed Askia followed him but was himself overthrown in 1537 and succeeded by Ismail. He ended the imprisonment of his father but died unfortunately early in 1539. His brother Ish’aq was enthroned as Askia (1539-1549), while his brother Daoud, the Kourmina Farin, campaigned against Mali (1542-1544), and he appointed Abu Bekr, the father of Mahmoud Baghayogho, as new cadi. Would he not follow his father’s example by appointing one from the same family?

14Mahmoud ben Abou Bekr al Baghyo’o al Ouankori, a remarkable legal consultant and theologian, was named cadi of Djenné between 1539 and 1542. Mahmoud may have been born around 1508 and his father Abou Bekr around 1490, which would coincide with the time of Mohammed Foudiya’s arrival in the region. Like his father, Askia Ish’aq was sure about his choice.

15A conference was held at Djenné on Isa’aq’s return from Taba (in 1539 according to the Tarikh es-Sudan, and in 1542 according to the Tarikh el-Fettach). The battle of Taba took place as the new Askia Isa’aq followed Mohammed Bonkano, Moh. Askia’s brother—who had been overthrown in 1536 by Askia Ismael and had fled into the Malian province of Bendougou. At that conference in the Djenné mosque Mahmoud ben Abou Bekr Baghayogho sat next to the Askia. When the latter asked who was the greatest oppressor […] Mahmoud said “indeed we don’t know anybody here who is a bigger oppressor than you, since you are the father of all the oppressors”.

“The Askia Isa’aq returned to Gao and decided to invest Mahmoud Baghayogho with the office of cadi and sent one of his officers with the investiture. By the time of his arrival the chief of Djenné and his subjects had assembled the inhabitants and the legal scholars, and called Mahmoud who ignored the cause of the meeting. They took him and held him by force and pulled two boubous which the Askia had sent over his shoulders, and wrapped a turban around his head, while he cried and complained, and read him the Askia’s letter. Then they brought him a horse, as the letter ordered, to take him to his house” (Houdas 1981: 168).

“His wife, mother of his son and jurist Sidi Ahmed Baghayogho, came towards him and asked: ’How could you accept the function of cadi […] It would have been better if you had chosen death and told him ’you may kill me but I won’t accept’. Mahmoud formulated curses against the Askia and died in the same month saying: ’Isa’aq has deprived my eyes of sleep and forced me to keep them open at all times. God may disturb his existence and send him reasons for worries.’ When the messengers returned to Gao, the cadi was already dead” (ibid.: 169)29.

16Mahmoud’s sons, Ahmed and Mohammed, apparently from different mothers, were born around 1523 and raised in Djenné30. Like their father they were offered the cadi’s office by the new Askia Daoud (1549-1583) but refused31. When they travelled to Gao, they were received by Ahmed b. Moh. Said, descendant of the cadi Mahmoud b. Omar b. Moh. Aquit and teacher of Mohammed (the father in law of Ahmed Baba).

“The Askia was happy at their arrival and his face shone with pleasure. Daoud solicited the erudite law consultant Mohammed-Baghyo’o el Ouankoré, to accept the function (of khatib); but the latter refused and addressed his master and brother, the saint of God, the law consultant Ahmed ben Mohammed-Said who agreed to go with him to Kagho (Gao) to beg the prince not to force Mohammed to fill this post. Little time after their return, the cheikh-of-islam, the jurisconsulte Ahmed, who had interceded on behalf of Mohammed Baghyo’o, died; may the Lord have mercy with them” (Houdas 1981: 177).

17In the 1550s Daoud campaigned in Kanem, Bornou, Katsina and Bousa, but in the 1560s the North of the Songhay Empire with the Teghaza salt mine was lost twice to the Moroccans (1561 and 1566) (Houdas 1981: 165, 171). The annual salt caravan was interrupted shortly but another mine was found at Tauten (Taodenni), some 700 km away.

The Baghayogho in Timbuktu

18Sometime in the 1580s Mohammed made the pilgrimage to Mecca with his maternal uncle. It is not known whether he ever returned to Djenné, but we find him installed in Timbuktu with his brother Ahmed in 1583, when Askia al Hajj succeeded his father Daoud, killed at Tondibi. It seems that at this time, Mohammed became Imam of the Sidi Yahya mosque, which office the Baghayogho have held to the present day32. The Sidi Yahya mosque is probably the oldest Timbuktu mosque, built around 145033, and has special significance: when the Touareg under their leader Akil took control of Timbuktu in 143334, they gave the chieftaincy to Mohammed Naddi, a Senhaja from Chinguetti35 who built the mosque, which was named after him, but then entrusted the imamate to his companion Sidi Yahya, who was venerated like a saint ever afterwards36. “Mohammed Naddi had this well-known mosque built to have his friend and companion, the eminent saint, and perfect pole, Sidi Yahya-ed-Tadelsi, fill the post of Imam. The two friends died at the same time, towards the end of Touareg rule” (Houdas 1981: 39). His surname ed-Tadelsi indicates Sidi Yahya’s origins from the Tadla in Central Morocco37. Nothing is known about the mosque and its Imams in the interval from 1468 to 1583, only that it was restored in 1569 by cadi el-Aqib. The reason for the association of the Baghayogho with this mosque seems to lie in their Wankore origins, and their alliance with the Aqit family from Chinguetti.

19Mohammed Baghayogho settled a dispute between Oumar, son of the cadi el-Aqib, who had died in 1582, and the new Askia. Aqib had given asylum to Daoud’s eldest son Bonkano, when El Haj usurped the succession, and kidnapped him, so that Aqib refused to accept the judgeship from him. “During that period, Mohammed Baghayogho tried to maintain the harmony among the inhabitants and prevented the development of tendencies the people had to ruin or rob each other and squander the property of the orphans”38.

20In 1587, Mohammed Bano deposed his brother Askia El-Hadj and killed two of his brothers in exile, but was deposed himself one year later, and died on the eve of a decisive battle against his brothers. The winner, the balama Sadiq, gained the universal support of merchants and ulema and returned from Gao. “As soon as he entered Timbuktu, the balama went to the house of the eminent jurist Mohammed Baghayogho and introduced himself […] and passed the night there […]” (Houdas & Delafosse 1981: 251).

21Prince Mahmoud was chosen as successor of Bano and mounted the throne under the name Ish’aq II. He undertook a campaign into the land of the Mossi who were raiding the Songhay hinterland, but was assassinated in 1590 in Bilanga in the Gourma39.

22On 30 May 1591 the Moroccan army arrived in the hinterland of Timbuktu after taking control of the trans-Sahara routes. One year later, on 5 October 1592, it defeated the Askia Mohammed at Tondibi near Gao. The pasha Mahmoud pursued the remains of the Songhay army into the Gourma— where the Dendi Farin Nouhou withdrew and established an independent Songhay chiefdom40. In 1594 the pasha returned to Timbuktu to punish it for its alleged support of Askia Nouhou41. He sent the caid Mami to Timbuktu to suppress a revolt, which may have been led by the loyal Aqit judges.

“Pacha Mahmoud was in pursuit of Nouhou when the news came from caid et-Tourki about the rebellion in Timbuktu which had killed 76 of his musketeers. The caid Mami who was to punish the town arrived at night […] which the population spent in fear without closing an eye, expecting a disaster. The caid entered the fort this same night when he was received by the caid el-Mustafa […] Yet, after the first prayer of the afternoon, the legal consultant and cadi Omar had the honourables of the town summoned, who answered and all went to him, including the jurist Mohammed Baghayogho. When he arrived and saw them all with their arms, Mohammed Baghayogho called on their chiefs and told them […] come all down and meet us” (Houdas 1981: 302-303).

23Mohammed negotiated a peace with el-Mami and prevented the looting of Timbuktu and organized next market day. But the pasha Mahmoud arrived himself from Gao and summoned the ulema to the Sankoré mosque under the pretext of renewing their oath of loyalty towards the sultan (23 October 1593).

“As they had reached Zim-Konda, one of them, a Ouankoré called Andafo, took the sabre of one of the soldiers […]. At the time when they had been made prisoners, two men from the tribe of Wangara […] refused to have the rope placed around their neck, and one of them slapped one of the Moroccan soldiers; the latter drew his sabre and hit him; but the brother of the first Wangara took the sabre off his hands and hit him in return. This was the occasion for a massacre.
Our master Mohammed ould-Kourtam told me that the Moroccans killed 14 inhabitants: two men of the Wangara tribe, two from the Soudan, one of the caste of blacksmiths called Abdallah Niabaly, and from the class of San42, among them the sage, the saint, the god-instructed, the jurisconsulte Ahmed Moya, Mohammed el Amin, son of the caid Mohammed ben Mahmoud ben Omar ben Mohammed Aqit, others again and finally Mohammed El-Mokhtar” (Houdas 1981: 259)43.

24Most of the ulema were deported to Morocco; only Moh. Baghayogho and his family were spared because of their association with a Moroccan saint, and the mosque was saved from destruction by the army. In the following months Timbuktu “became like a corpse without soul […] there remained nobody in the town who observed the laws except Mohammed Baghayogho who consoled the relatives of the victims in captivity” (and held the funeral services for the slain victims).

25The pasha wanted M. Baghayogho to sign a false declaration that the ulema had remained loyal to the Askia and plotted against him, but Baghayogho resisted these pressures (Houdas 1981: 308-309)44. He died Friday 19 Shawwal 1002 (October 1594), his brother Ahmed having preceded him in 1585.

The Descendance

26Mohammed’s sons were known as Abdullah and Abderahman but nothing else is known about them. However, Ahmed’s sons, Ibrahim (Abou Ishaq) and Ahmed (Mohammed) left a record (Houdas 1981: 70-71)45. The former appears to have died in 1638, when his obituary appears:

“On 8 Rabi of 1048 (19 August 1638) died the most eminent, wise and believing cheikh, legal advisor and scientist Abou Ishaq Ibrahim, son of Ahmed Baghyo’o el Wankori” (Houdas 1981: 449).

27The latter, also called Mohammed ben Ahmed Baghyo’o el Ouankori, was renowned and held the oratory in the mosque of Mohammed Naddi (Sidi Yahya) (6 September 1650) for the caid Ali ben Rahmoun en-Monehebbi who had died at Goundam. He himself died in 1655.

“This Wednesday evening of the month of safar (9 September 1655) died the lord of our epoch, our master, and sheikh of Islam, a man useful to mankind, the jurisconsultant Mohammed, son of the jurisconsultant Ahmed, son of the jurisconsultant and cadi Mahmoud Baghyo’o el Ouankori. The last prayers were said by the son of his brother, the jurisconsultant Mohammed ben el Mostafa es Sahara, musalla of the notables and saints” (Houdas 1981: 457, 487).

28El Mostafa es Sahili, or also es-Sahara appears to be a third son born in 1565/1566 and died in 161646. Below we will see that a Mustafa is credited to be the first Imam of Mossi in 1681.

29The Tedzkiret en Nisian covers the early 18th century. The family members mentioned in the Tedzkiret (Houdas 1966: 98-237) are the following:

— Nana Khidj, daughter of El Fa47 Ahmed, son of Imam Boubo b. Mohammed Bagayogo b. Ahmed b. Mahmoud El Ouankori, deceased 20 July 1748;
— El Fa-es-Siddiq b. Imam Moh. Bagayogo, son of Imam Koured, deceased 6 October 1741;
— l’Imam Baba ben el Fa Moh. Baghyo’o;
— El-Fa Ahmed ben Abderrahman b. Ahmed b. Imam Moh. Koured, deceased 3 July 1744;
— Aicha, daughter of Baghyo’o, b. Imam Ishaq Ibrahim b. Ahmed b. Mahmoud Baghyo’o, deceased 28 December 1742;
— el-San El-Fa Ahmed b. Abdallah b. Ahmed Tagh b. Ibrahim b. Ahmed b. Mahmoud; Baghyo’o el Ouankori, after the incident at Minia Kaina Yendi, 22 February 1743;
— Mohammed b. Mohammed Baghyo’o, son of Imam Moh. Koured;
— El Abbas Ed Cherif son of the Imam Baghyo’o, son of the law consultant and Imam Moh. Koured, deceased 19 March 1729;
— Imam Ibrahim Koured (Kouradi);
— el-Ouankori Abou el Abbas Ahmed b. Mahmoud Baghyo’o b. Ibrahim b. Ahmed Baghyo’o deceased in 1738.

30Unfortunately we know little about the Imam Mohammed Koured (Kouradi), who had two sons: Mohammed and Ahmed, who cannot be placed in the genealogy, unless he was a son of Abou Ishaq. Thus we can represent the first seven generations of the Baghayogho clan in a diagram:

Fig. I: Genealogy of the Baghayogho of Djenné and Timbuktu

Fig. I: Genealogy of the Baghayogho of Djenné and Timbuktu

Relationship with Other Families

31Between the Baghayogho and the author of the Tarikh el-Fettach, Mahmoud Mutawwakil el Kati, existed family relations: El Kati was married to Kadi, daughter of Mokhtar Timeta al-Wankori, whose son Mohammed had married one of the women of the Kâti family. “On dohor Sunday 17 (30 January 1644) died my wife Kadi-bent-el Mokhtar Timeta el-Ouankori” (Houdas & Delafosse 1981: 452). “Saturday 5 of the month safar (2 April p1645) died in the town of Bina, my brother and friend, my relative by alliance in a double sense, Mohammed, son of cheikh Mokhtasar Timeta el-Ouankori” (ibid.: 454). “The 4th (1st April 1645) died my brother Mohammed, son of cheikh El Mokhtar Temt el-Ouankori” (ibid.: 419)48.

32Between the Aqit and the Bagayogo family existed close teacher-student relations: Mohammed Baghayogho studied under Ahmed b. Hadj Ahmed b. Oumar Aqit and taught Ahmed’s son Ahmed Baba and looked after his relatives during his exile; Ahmed Baba taught Ahmed’s son Mustafa es Sahara (Saad 1983: 246, n. 186).

33From the late 18th to the early 20th century the family needs to be better documented. Alfa Ahmed Bagnio b. al-Hasan b. Ahmad Bagnio49 b. Al-Imam Mustafa was Imam of the Sidi Yahya mosque at the end of the 19th century; Mustafa’s brother had died in 1842/1843, (Marty 1920, t. 2: 65; Saad 1983: 222). Mohammed ben al-Fa Wangara, called Banyon, born toward 1858 and alive at the end of the First World War, is mentioned by Marty as Wakore of Kong origins, belonging to the Tidiania sect of the zaouia of Fez (Marty 1920, t. II: 20-21). He seems to be the grandfather of Hasseye Mahmoud B. of Timbuktu (see next section).

In Search of Origins

34When looking for the last Wangara in Djenné, Seriba Touré of the Sankoré quarter tried to convince me that the Wakore were Arabs, but Ba Hasseye Maiga, the chief of the Songhay quarter of Djenné insisted they were merchants of Soninke origin and still represented by certain families in Konofia quarter—still today called Wangarakunda. And that some lived in the Bajinde quarter of Timbuktu, also called Wangarakunda, next to Djamaikunda, quarter of the blacksmiths, behind the Sidi Yahya mosque. Therefore, I visited Timbuktu in early 1997 and looked for the guardian of the Sidi Yahya mosque and was introduced to Hasseye Mahmoud Baghayogho. He claimed to be the last descendant of the Wangara and the Baghayogho in Timbuktu and that his ancestors were Soninke from Ghana installed in Kong for centuries. He added that his family had been the owners of the caravanserail of the Azalai, the annual salt caravan to Taoudenni50, until the Touareg revolt of 1991-1996. He further mentioned a branch at Séguéla which he visited occasionally51.

35In Bamako I met other Baghayogho families—mainly noumou also called Doumbia52who claimed their origin from Kélénya, 70 km from Bamako. Following up these claims I visited the chiefs and elders of Faragouaran in the Wasulu in 1997—a village entirely peopled by Baghayogho, noumou and horon. They told me53 their origins were with pilgrims to Mecca who visited the kalif at Bagdad, and therefore took the name Bagadaji, or Baga-yoro (place of Bagdad). They also claim to have given Imams to the Diakhanké, a claim which could not be verified in Dia, where the dyamou (clan) is not represented, but for Diakha near Kayes54.

36In the Malian Wasulu all Baghayogho say to have come from Kélénya and spread from there to Danou, Bolou, Kelenya and Koro in Côte-d’Ivoire. They claim descent from five brothers of Kélénya: Wombo (Wahab), Zanifing, Danou Moussa, Nfamoussa and Malab. While Wombo remained at Kelenya, Zanifing went to Niakoni and his descendants founded Bolou, Moussa installed himself at Danou, and Nfamoussa migrated south in search of richer lands and founded Koro. Here the family furnished the chefs de canton during the colonial period. Blondiaux (1897: 371) confirms:

“Le village de Koro a été fondé par M’fa Moussa Bakaioko, grand-père du chef de village actuel. C’est à Mfasirafa Bakaïoko qu’a été légué le pouvoir à cause de sa parenté plus directe avec le prédécesseur, mais en réalité le vrai chef est Famodiän Bakaïoko, son cousin, qui jouit à Koro et dans ses environs d’une grande autorité. Les relations ont été bonnes entre Samory et Koro qui a bénéficié de la destruction de Borotou [Boron?] et des villages de Gouyé par l’almamy.”

37The foundation of Koro is shrouded in myth, but it seems that Fa Moussa, the founder of Koro, was the Nfamoussa from Kélénya. At the end of the 19th century, we find the Baghayogho not only in Koro, but also in Séguéla, Kani and Sarhala. At Séguéla, karamoko Yousouf Baghayogho was recognized by the French as one of the most influential Muslims of his time, who claimed first his origin from Timbuktu, and second that the Diomandé (Kamara) had called the Baghayoghos to Séguéla in order to convert the animist Bambara.

“Autrefois à Séguéla, Yousouf Bakayoko était un chef religieux très influent. Cette influence a disparu à sa mort, et personne à Séguéla n’a relevé cette puissance. L’ancêtre de Yousouf Bakayoko, Mfafé, Bolé Bakayoko vient de Tombouctou sur la demande du chef des Diomandé qui venaient de s’installer à Léguéla [sic!] vers le commencement du xviie siècle.
Frère d’un marabout célèbre, il fut, paraît-il, un thaumaturge de quelque mérite.
Lorsqu’après un séjour à Séguéla il voulait retourner dans sa patrie, son frère qui était resté à Tombouctou, lui donna l’ordre de placer un bâton en travers de ses épaules et de marcher jusqu’à ce que les deux extrémités de ce bâton fussent arrêtées par des arbres. Il se mit en route et arriva à Dabala, près du confluent du Boa et de la Férédougouba. Les piroguiers refusèrent de le passer malgré l’offre de quatre esclaves.
Mfafé Bolé frappa l’eau de son bâton, l’eau s’écarta et forma deux murailles entre lesquelles il passa mouillant le bord de son manteau […]. Il appuya son bâton contre le tronc d’un baobab qui est resté penché et qu’on montre encore aujourd’hui […]. Il remit ensuite son bâton sur ses épaules et s’arrêta sur l’emplacement actuel du village de Koro où son bâton s’accrocha entre deux arbres […]. Il fut enterré là et son tombeau est un lieu de pèlerinage” (Gouvernement général pour l’Afrique française 1906: 283).

38We conclude thus that the older branch of the Baghayogho are at Koro, but Yousouf of Séguéla pretended to be the religious chief of his relatives at Koro and those of the Koyaradougou. His ancestors have indeed converted the Fofana Mande who settled these lands and founded Mankono (Gouvernement général pour l’Afrique française 1906)55.

39Mfafé Kolé, Mfafé Bolé or N’fa Moussa seem to me one and the same person. At Séguéla, the Barhayorho, Binanté and Timité, are the prominent Muslim families while the animists are Diomandé (Kamara) and Soumahoro (Kanté).

“La personnalité la plus remarquable de Séguéla est Vanzoumana Barhayorho, […] né vers 1870. Son père Amadou était assez connu dans toute la Côte-d’Ivoire […] il est décédé en 1900. L’almamy de Séguéla est Baba Sarhoyoro, né vers 1865 et occupant cette place depuis la mort de son frère Oussoumana […]. Le Gouaran n’est pas très islamisé en soi: mais il renferme une grande quantité de dioula de passage […]. Mankono comprend une majorité de musulmans où dominent les Cissé, les Bamba, les Sérifou, mais surtout les Barhayorho et les Fofana. Les Fofana sont la famille la plus nombreuse et la plus turbulente du Ouorodougou qui a été islamisée par les Barhayorho, il y a un siècle; ils se sont détachés avec le temps de cette obédience et notamment de leur maitre Yousoufou Barhayorho, de Séguéla, qui les a dirigés” (Marty 1920: 156-159)56.

40The centres of Diomandé colonization are found in the Konyan in Guinea, whose capital Musadougou is named after the marabout Musa Baghayogho; in the Mahou around Touba, the Barala around Boko, and the Worodougou with Worofla as capital. The Konyan was occupied in two waves according to Moundekeno:

“Le Konya: dont l’occupation s’est effectuée en deux vagues: a)  La première constituée de Condé, Konaté et Kourouma qui a très tôt pris souche entre le Niger et le Dion-Sankarani. b)  La seconde vague, où dominent les Camara ou Dyomandé, dont le chef migrateur était Fonikaman parti de Tabon dans le premier quart du xvie siècle à la tête d’un groupe Mandinka fuyant l’anarchie grandissante qui règnait alors dans le Moyen Niger” (Moundekeno 1979: 20)57.

41The Koné are recognized as colonizers of the Sankaran (called Dion in Bambara). The Kamara called themselves Dion-Mande, i.e. Mande from the Sankaran, when they left the area to colonize the mountains further east, where they found such animist groups as Gbéressé (“Kpelle” in Liberia), Dan, Bambara, Senoufo, Nafana and Ligbi. The conquest of these animist areas could not take place without magic practices, namely by marabouts, the Maninka Mori.

“Les Maninka Mory ne sont pas à proprement parler des Maninka. Ce sont des Sarakollés du Moyen Niger très anciennement islamisés, commerçants depuis une date assez reculée. Les Sarakollés avaient comme pays d’origine le Wagadou ou pays de Ghana ancien. De Ghana, ils s’installèrent d’abord dans le Dyafounou et pour des raisons multiples, ils amorcèrent de nouvelles migrations qui les conduirent dans l’entre deux fleuves, le Batè. La date du xviie siècle peut être retenue comme date de leur infiltration” (Moundekeno 1979: 20-22).

42In the Konyan, Konsaba, brother of Feren Kaman called upon his marabout Soumahoro who, according to the legend, sent a fish with a message to Timbuktu which was captured by the karamokos Mori Kane and Musa Baghayogho. They travelled together to help Konsaba but separated at Odienné. According to le Campion, twelve saints settled the region after their joint pilgrimage to Mecca.

“Vers le milieu du xviiie siècle, douze marabouts partirent ensemble de la boucle du Niger et se rendirent à la Mecque. Leur pélerinage terminé, ces pieux personnages se fixèrent en des régions différentes:
Hadji Salimou Souaré et Mohammadou Barhayorho demeurèrent à Tombouctou;
Hadji Mohammadou Fofana et Maliki Somono demeurèrent à Kankan;
Hadji Bamba Fadiga demeura à Dienné;
Hadji Bakari Traoré demeura à Ségou;
Hadji Mohamadou Dialé à Vamala;
Hadji Yssifou Kamaraté à Bouandougou (cercle de Ouorodougou);
Hadji Morifing Bamba à Toulé (cercle de Ouorodougou);
Hadji Diaya à Samatiguila, près d’Odienné;
Hadji Férémourou Kanaté à Féréfougoula près d’Odienné;
Enfin Hadji Moussa Barhayorho (Bagayogo), des douze compagnons le plus saint, le plus illustre, vint dans la région de Touba et fonda Koro.
À vrai dire, l’histoire rapporte seulement qu’un chef fétichiste nommé Gonsalia Diomandé, habitant le pays de Beyla, tant pour obtenir la victoire dans les luttes des clans, voire de famille, que pour repousser vers les forêts les Yafouba, populations autochtones, fit appel à la protection des prières de Moussa Barhayorho (Bagayogo) et surtout à l’appui de ses hommes venus en exode, soit volontaire, soit forcé, des régions (de) la boucle du Niger. Moussa prêta son concours, les Yafouba furent razziés et la victoire acquise aux Diomandé. Moussa, pour sa part du butin, eut beaucoup (131) de captifs et l’existence florissante de Koro fut dès lors pour longtemps assurée” (Marty 1922: 132).

43Traditions collected in Koro and published by Derive and Nguessan (1976: 25-26) report a similar story:

“Le marabout qui a fondé ce village et qui s’appelait El Hadj Moussa Bakayoko (Baayo in korokakan) est venu de Tombouctou […]. Alors que celui-ci était à Tombouctou, un des chefs Diomandé du Gbeka (région de Borotou) demanda à un marabout de Sifuula, du nom de Mya Samoo Kiaaté [Kouyaté], de venir faire des prières afin que son village devienne puissant. Ce dernier répondit qu’il ne pouvait pas mais qu’il connaissait un homme, avec qui il avait fait le pélerinage à la Mecque qui pouvait sûrement accomplir cette tâche […] il écrivit une lettre qu’il placa dans la bouche d’un poisson […] qui descendit le Djoliba jusqu’aux rives de Tombouctou. Des pêcheurs […] avertirent El Hadj Moussa Bakayoko. Ce dernier s’en alla informer son frère aîné, un autre grand marabout de Tombouctou El Hadj Alpha Mahamoud. Celui-ci lui dit de partir. C’est ainsi que El Hadj Moussa Bakayoko se dirigea d’abord dans le Gbé où il rencontra Fuengana [Feren Kama] et Konsaba Diomandé qui étaient les chefs de la région à l’époque.”

44Derive (1990: 55-56) also collected the traditions of the region of Odienné (Wo-Djenné according to local etymology) in the Noolu:

“Les fondateurs d’Odienné seraient incontestablement les Kamagate (ou Kamate) et les Komara. Les traditions font état d’un voyage entrepris par un groupe de douze pèlerins originaires de la boucle du Niger, partis à pied à la Mecque. Leur chef s’appelait Sani Mosouaren ou encore Mosuaren Samanssi [Salimu Souaré?]. Cet événement se serait produit vers la fin du seizième siècle. Ces hommes auraient suivi des routes diverses, mais deux d’entre eux El Hadj Youssouf Kamagate and Sani Mosuaren, après être restés sept ans à la Mecque, établirent leur résidence en un emplacement situé sur la rive gauche du Baoulé, au Nord Ouest de Logouanasso (Logonasso sur les cartes au 1/200.000), près d’Odienné. Avant de suivre l’histoire de ces deux fondateurs de la ville d’Odienné, disons quelques mots de deux autres de leurs compagnons: El Hadj Morifere Kante and El Hadj Va Moussa or Moussa Bakayoko […] le deuxième compagnon s’aventure plus loin et s’établit à Koro. […] Ces deux villages Ferefougoula and Koro sont devenus des lieux de pèlerinage où les croyants viennent invoquer les noms de ces deux saints.”

45Thus the subjection of the animists in the Mahou and Konyan mountains is credited to Mousa Baghayogho—apparently the same who founded Koro. Unfortunately, the name Mousa is conspicuously absent from the genealogy of the Timbuktu family.

46Thus we have a constellation of Mande elements from the West—the Kone and Kamara—and from the North—the Maninka Mori of Soninke or Wangara origins—to subdue and convert the animists of Guinea and north-western Ivory Coast. While the myth indicates a Kamara request for help rather than a conscious “missionary” effort by the Muslims, I argue for a conscious plan for a jihad against the southern pagans—to whom in particular the Wangara gold traders had privileged access. As other Soninke clans like Kamaghate, Timité, Fofana, Sanogo, and Bamba were involved—all of whom are not only represented in the north-western but also the north-eastern Ivory Coast (in Kong, Bouna or Boundoukou), and Ghana together with Sylla, Diaby, Doukouré and Konaté—the Soninke predominance in the diaspora does not appear to be a coincidence. They were recognized as “old in Islam” and benefited from the “Wangara connection”, namely ancient and recent contacts through the trading networks. I argue thus that the conversion was first planned after Mohammed Askia’s hajj to Mecca as an instrument to introduce Islam and thus better control the auriferous and kola-producing regions of the south, and renewed in later centuries.

The Baghayogho Diaspora in Guinea-Bissao

47In the Mandinka kingdoms of Gabu and Biafada along the Geba river, the Baghayogho have established two lines in Bijini and Janna, whose founders were Sulaiman Ba and his son Daouda Baghayogho or “Baayo” (Tarikh Bijini, in Conteh, Gomes & Skinner 1987; Giesing 2003). Daouda’s three sons and Fode Baayo established the four Baayo qabila in Bijini.

48The Tarikh Bijini reports that Suleiman Ba was the son of the Kunti Sheykh Ahmed Bekkai in Timbuktu, and emigrated with slaves and cattle during the time of the jihad of El-Hadj Omar (c.1850), and arrived at Kanjaara (Kanjaga) and Bijini during the reign of Sanka Nanki I of Biafada. Interestingly, the Baghayogho all claim to derive from Mohammed Baghayogho I of Timbuktu and his two sons, who are the joint ancestors of the Baayo in Senegal (Diakha), in the Gambia and of the Gabu. Bijini is said to have been founded by the Baayo, the Daabo, Saama and Kasama families whom we will meet later. As related by Fr. Barreira and Almeida, the Mandinka kingdoms in the Gambia and Sierra Leone were established in the 1550s and the kings baptized around 1655 (Massing 1980).

The Baghayogho Diaspora in Mossi

49Levtzion (1968: 91) has collected evidence on the introduction of Islam to Mamprusi, Mossi and Dagbon:

“When Moro-Naba Dulugu learnt that the Mamprusi chief had an Imam among his ministers, he wished to have an Imam at his court also. Therefore he invited a Muslim of Zitenga and named him the first Imam. They adopted Islam from him. This Muslim Mustafa Bagayogo was member of a family of Timbuktu origin, who installed himself first at Mane and at Zitenga, both in Mossi. The Islam had been introduced to Mamprusi during the reign of Na Atabia at the beginning of the 18th century, more than half a century before Na Dulugu One says that Dulugu was sent with his brothers by their father Moro Naba Sagha to study the Koran with the Muslims. Two of his brothers, Ngadi and Sigiri, became Muslims.”

50But Delobsom, Levtzion’s source, wrote “the emperor had a friend, a marabout in the region of Zitenga who was called Mustapha, and Doulgou made him come to Ouagha” (Delobsom 1934: 205). Besides, he dates the first arrival of the Muslims to the 14th century:

“The Yarnaba—chief of the Yarcés, whose ancestor Koanda (a synonym of Koné) came to the Mossi under the reign of Koumdoumié, 8th Mogho Naba—is in charge of sacrificing the bulls and rams given to him by the Emperor (at the occasion of Id-el-Kebir). Their installation in Mossi dates to the reign of Koumdoumié (5th successor of Oubri). This race furnished to the Mossi the marabouts and dioula, but a certain number among them remained what they were from the origin that is fetishers” (ibid.).

51According to Tindrebeogo, Koundoumié reigned from 1335 to 1358. Marc (1909: 111), however, puts the arrival of the Yalsé only into the 18th century:

“Les Yarsé (au singulier Yarga) sont des Mandés venus s’installer au Mossi vers le milieu du xviiie siècle et actuellement assez transformés par l’influence du milieu et par des mariages fréquents avec des femmes mossi. […] On trouve des Yarsé dans tous les villages où se tiennent de grands marchés; on en trouve également assez régulièrement d’étape en étape dans les localités qui jalonnent les principales routes des caravanes. Certains villages sont presque exclusivement habités par eux et ont pour chef de village le chef des Yarsé: Caya (qui s’appelle également Salmatenga), La, Pouitenga, Rakay sont dans ce cas.”

52There is evidence of religious persecutions and refugees in the wake of Ngolo Diarra’s victory over the Ton Massa between 1746 and 1760:

“[A]rrivée des Mandé-dioulas après la mort de Sagone en 1754, les Mandé-dioulas établis dans le pays de Ségou qui avaient pris parti contre Ngolo vainqueur de Sagone, s’enfuirent à la crainte de représailles et vinrent s’installer au Mossi. Ngolo essaya de les poursuivre et vint à deux reprises attaquer le Mossi, […] au cours de la seconde, en 1787, Ngolo mourut” (ibid.: 142-143).

53The apparently conflicting evidence is consistent with history because from the 14th century onwards the Mossi invaded Mali and brought Muslim captives to Mossi. Later Naba Doulgou—from 1785-1802 according to Tiendrebeogo (1990: 30)—, as the 26th naba and the first Muslim and contemporary to Ngolo’s wars, had the religious duty to welcome Muslims persecuted in another land, and probably did. But he was himself at war with the Muslim chief of Mane, who had a conflict with his own son Zitenga Naba Kougri.

“Le Naba dulgou se mit donc en campagne contre le chef de Mané qui était en conflit avec le Zitenga Naba Kougri, fils cadet de Doulgou. La région de Zitenga était d’ailleurs la plus riche de l’Empire. Naba Doulougou fut blessé mortellement par une flèche empoisonnée au cours de cette campagne et mourut à Loumbila Bagrin où il avait été transporté” (ibid.).

54Thus Naba Doulgou, a Muslim, died at the hand of other Muslims. In 1999 I contacted the Mane Yalse who confirmed their origin from Mali and showed me a certain number of Arabic tarikh, which we had no time to translate. In 2001, I interviewed the Zitenga Naba, Salif Kongo, and his Imam Yalga Salif Baghe, who showed me the copy of an Arabic manuscript with the following text:

“This is Musa the Believer who was the first in Islam (called Imam Delwende). Abd’ el Mumin Baghya came from Bagdad in Iraq, at the time of Mogho Naba Souré, in 168158. In that year Abd’ el Mumin came and invited everyone to adopt Islam. His children scattered. His first son was Abubakar who called himself Imam Youré and installed himself in the land of Zitenga, which is under the region of Ziniaré.

55The manuscript gives the following genealogy of the Baghya Imams:

“Musa le Croyant
Abd el Mumin Baghya
I.  Abubakar, son fils, Imam Youré;
II.  Mustafa, un des fils d’Abubakar;
III.  Mahmoud, un des fils d’Abubakar;
IV.  Idriss, frère de Mahmoud, de la même mère;
V.  Mokhtar, enfant direct d’Abd el Mumin;
VI.  Ahmadi, de même mère que Mokhtar;
VII.  Ousman, de même mère que Ahmed;
VIII.  Mahamadi, fils de Ousman;
IX.  Ismaila, fils de Ousman;
X.  Mahama, fils de Tahir (Tasiré), mort en 1967.”

56As we can see, the above genealogy has no dates and appears highly telescoped since three or four generations could hardly hold the imamate for 300 years but states certain facts: Mane was the first village given to Muslims by the Moro Naba and Abd el Mumin was sent from Mane to Zitenga. At Naba Doulgou’s time Muslims had been in Mossi for two generations and the Zitenga Naba sent Abd el Mumin’s son Mustafa to the Naba’s’s court in Wogodogo. Even though Naba Doulgou, had been trained as a Muslim59, only Naba Koutou, the 30th Naba, was the first and only Muslim Naba.

57The Zitenga Naba, Salif Kongo, himself a Muslim, and his Imam both claim that the entire Baghya family descends from the Baghayogho of Timbuktu, and had Musa as ancestor. Unfortunately, we have not yet been able to identify either Musa or Abd el Mumin in the Timbuktu genealogy. Descendants of the Zitenga Imams still administer the Sankariaré mosque—rather than the grand mosque of Ouagadougou, whose Imams are Hausa—and hold the original of the Zitenga manuscript and confirm it was written by one of them.

58Thus, while Muslim yalsé came to Mossi as soon as the 14th century, the Moros only accepted an Imam at their court in the late 18th century, and with one exception never converted to Islam. Further research on the Mande-Muslim traditions at Wogodogo, Mane, Zitenga and Sanmatenga is needed to document Mossi-Muslim interaction.

Baghayogho in Mamprusi and Dagbon

59Islam came to Dagomba via Mossi according to Levtzion (1968: 91). “At the time of Na Zangina, the Sabali yarna (yar-na, a contraction of yarse-na the chief of the Muslims) was regarded as the senior Muslim in Dagomba […] the yarna of Sabali belongs to the Boghyo or Baghayogho patronymic group, whose fame as ulema goes back to the sixteenth-century scholar MuHammad Baghyogho of Timbuktu. The Imams of the Moro-Naba of Wagadugu belong to the same group60 […]. Zangina was the first king to be circumcised, the first to become Muslim, and it was he who built a mosque in Sabari. The name of the Imam was Yamusa. Muhammad al Kashnawi [Katsina-wi] introduced Zanjina, king of Daghbun to Islam […] Yamusa Saghyardu al Wangari was made Imam of the mosque” (Wilks 1963: 88 n. 6 & 7)61. From our note on Séguéla the reader will remember that Saghyardu corresponds to Baghayogho, which opens up the interesting possibility that Yamusa might refer to Nfamusa from Koro. Na Zangina ruled from 1648 to 1677, according to Tamakloe, and from 1680 to 1713 according to Wilks, and might be a contemporary of Na Atabia, who died in 1741/1742. Tait62 wrote on the inauguration of the first mosque of Dagbon:

“Yamusa, Yari-Na of Sabari, went and opened the door of the mosque and told Na Zangina to enter inside the mosque. Na Zangina entered. Yamusa, Yari-Na of Sabari took a Quran and put it in the hands of Na Zangina. Na Zangina took the Quran and held it. He worshipped toward Hausaland, and worshipped towards the land of Ashanti and worshipped towards Moshiland.”

60According to Malam Adam ben Uthman Baghayogho of Sabari, who was interviewed by Wilks, all the yarnas belonged to the lineage of Baghayogho, and Yamusa was the son of a Souleyman who arrived from Salmatenga (i.e. Kaya) with a Buba Moli in an embassy by Na Zangina to the Mogha Naba (Wilks 1963: 93 n. 44, 94 n. 48, 94 n. 50-51). Kaya harbours an important Yarsé colony, some 50 km from Mane.

61If we accept 1681 as the date of arrival of Abd el Mumin at Zitenga, and 1713 as the date of Na Zangina’s, then the arrival of Islam at Dagomba during Na Zangina’s late and Na Atabia’s early reign is possible. If we assume that Musa le Croyant was identical with Yamusa and continued from Kaya with his father Souleyman to Dagbon and Mamprusi to introduce Islam there between 1681 and 1713 (year of Na Zangina’s death), we conclude that Islam was introduced to Dagbon and Mamprusi chiefdoms around 1700, few years after its introduction into Mossi, and half a century after its introduction into the Konyan and Mahou. Evidently further research on the Baghayogho is necessary in Dagbon and Mamprusi.

Islam in Sansanne-Mango and Gonja

62Some 50 km east of Mamprusi lies the Moba capital Bunkpurugu, and further 50 km to the south-east lies Sansanne-Mango, capital of the small Anno (Chokossi) enclave. Binger (1892) reported that around 1730 the Na-yiri, the Mamprusi Paramount Chief, wanted to subdue his sub-chief at Nalerigu and called on the Bono (Abron) of Bondoukou, the Anno (Agni) from Groumania on the Comoé river, and the Gonja to give him military assistance against Nalerigu. In recompense the Mande of Groumania were given the land to settle which they called Sansanne Mango in remembrance of their hometown Mango-Toura.

63Historical manuscripts of the migration have been written by the Imams, and copied and transcribed by the Germans Seefried (1913), Asmis (1912) and Zech (1904) and preserved in the National Archives of Togo63. These manuscripts have been copied by Roouveroy (1976) with comments on recent history.

64The account “The trek from Mango Tura to Sansanné Mango” tells of the departure of Bema Bonsafo from Mangu Tura on Friday 12, of the Badara people one week later, of Soma Breima and his people on the 22nd, and the remainder of the warriors on the 29th. The groups conquer Ngorumi, kill the chief, and stay there for 13 months; thereafter they conquer Sauwa, where they stay for another 13 months, after having killed its chief. The king of Jabo, hearing of them, sends his son and his Mishirasi Imam with 73 slaves to ask them to come to his assistance against Kandia. They kill the Kandia chief and settle there for 13 months. Meanwhile, the king of Mampurugu sends a message to the king of Jabo, through Kurabaso Imam of Kusawgu, who sends his son Idrissa to the king of Jabo. The latter sends his son Breima and his Shinsina-Imam to Soma and Breima at Kandia. He informs them of the message from Mampurugu and gives them his son Weio as guide to Gambaga. They go there and first conquer Kantindi (near Dapaong) and then Kuniogu (Sansane).

65The account itself was written on behalf of Imam Amadu Bandawiu, who made the trip with his elder brother Gazama (Gejuma in Seefried’s transcription), who became the first Imam of Sansanne64. The Roouveroys (1976) note that both belong to the Kambaya lineage which they associate with the Kamaghaté of Groumania65. Marty (1920: 58) reports that Groumania signifying “ancient okra field” is a Mande-dioula and Muslim colony of Kong.

“[L]a première famille qui s’installa dans le ’champ de gombos’ fut les Karamaté, ils sont restés les chefs de village; les guerres de Samory y amenèrent des fugitifs; puis vinrent des gens de Bouna et des Haoussa... les trois quarts des habitants y sont dioula et perpétuellement en route. […] Groumania est une ville mandé-dioula fille de Kong, partant nettement musulmane...” (Marty 1920: 58).

“Les Mango ou Anno sont des métis de Ngan, de Mandé, et d’Agni […] les Anno ou Anonofoué sont venus ici et se sont constitués par des invasions successives de Ngan, de Mandé et d’Agni Ashanti […] ce qui finit par amener son unité, sous l’autorité du chef mandé-dioula, Dyan Ouattara, lors des attaques de Fofié, roi des Abron vers 1830.
Ces Mande-Dioula, maîtres politiques du pays, sont tous de diamou Ouattara. Ils venaient, soit de Sansanné Mango (Togo), soit d’Anno (Dagomba anglais) et c’est ainsi que le pays a pris d’eux le nom de Mango ou d’Anno. Ils étaient tous fétichistes et le sont restés à l’exemple de leurs cousins mandé du nord ne les ayant pas convertis à l’islamisme, sauf exception” (ibid.: 49-50).

66According to Levtzion (1968: 82),

“the Kamaghte Imams were associated with the Jermabu chiefs, the descendants of Beima Bonsafo [the first leader of the trek]; the second leader Soma Ibrahima founded the second branch of the ruling family, the Sangbana, who have their own Muslims, from the Dao clan who came directly from Kong.”

67Yet, the Kamaghate are not only associated with Mango-Groumania, but also with Begho, from where they came to Gonja to constitute the Sakpare Imams66 of the Paramount Chiefs. They also had been the Imams of the Bouna Friday mosque67. The relations between Bouna and the Imams of Gonja, especially Buipe, are close, and the authors of Kitab Gonja appear to have come from Bouna. But following Marty (1920: 166) they were also associated with the Baghayogo:

“Cette famille des Kamarhaté, comme celle des Barhayorho, est originaire de Tombouctou. Les deux familles y entretenaient, disaient-elles, des relations suivies, qui ont cessé lorsqu’elles sont venues dans la région. On cite un Kamarhaté, mort et enterré à Tombouctou, qui aurait fait le pèlerinage de la Mecque. […] Les Kamarhaté fondèrent le village de Sendou à proximité de Bouandougou, et en furent chassés par les fétichistes. Ils vinrent alors demander asile aux Kandé, ancêtres du chef de canton actuel, et s’installèrent à Bouandougou. Cet établissement remonterait à treize générations, soit au xvie siècle.”

68Of further interest in the Sansane account is the relation between the king of Jabo (Gonja) and Bema and Soma: he sent them his Mishirase—derived from misiri-si, i.e. “mosque people”—this most likely refers to the Buipe Imams, who were the first Imams of Gonja and of the same clan as Bema’s Imam, whereas the Kusawgu chief had a Sinsina Imam, who was not a Sakparé.

69Roouveroy (1976) mentions that the Koranti (of the family of Bema and Soma) of Mango68 were descended of the Gazama lineage, that of the first Imam; they seem to me related to the family of Imoru Konandi of Bouna, who wrote the Kitab Ghunja. This seems to imply that Kamaghate, Gazama and Korandi are one and the same family69.

70The Kitab Ghunja records for 1732/1733: “Now in this time the people of Mango were called Zuma” (Soma’s people). This dates the arrival of the Mango people to 1732. For 1736/1737 the killing of the chiefs of Kandia and Kong is mentioned—both former divisional chiefdoms, a few miles from each other on the road to Wa—who were killed in battle together with a malam Fadiga, grandfather of the Kitab’s authors, whom we believe to be Imam Konandi from Bouna. Thus, the conquest of Kandia by the Anno could be dated to 1736, and their arrival in Mamprusi toward 1740. Thus they could well have founded Sansanne Mango by 1750.

71In the small states of Wa and Bouna, created by Dagombas before the Gonja incursion, when the western border of Dagomba still ran along the Volta River, we should expect similar influence by Wangara Muslims:

“But the yarna of Nasa introduced the first Imam Ya’umaru Taraore to Wa, whose origin was from Dia ’the Islamic metropolis of the Macina’. […] In Wa, the yarna, who is also known as the Shaykh al-Wangara is still regarded as the head of the community, and the community is predominantly Wangara: Kong was the centre of learning, and Imam Sa’id from Wa went to Kong in the 19th century. Wa is part of the Dyula zone, but the Wala Muslims have adopted the local language and been integrated in the state. In Bouna, the Muslims speak Dioula—while the chiefs claim Dagomba origin but have assimilated to the Kulango; in Wa there is a structural dichotomy between the chiefly lineages of the later invasion and the commoner clans who had come first and expelled or absorbed the Lobi-speaking peoples” (Levtzion 1968: 135-136).

72According to the manuscript “The History of Wa”70, the first settlers were Mande immigrants:

“The first among them came from Mande […]. They divided into three: some settled at the village of Palwogho, others at Vise and Nasa. Those who settled at Palwogho were of three clans: Senu, Dao, and Juna. Those who settled at Vise belonged to one clan, Kunate. Those who settled at Nasa also belonged to one clan, Taraore, These were the first Muslims in the country of Wa […]. Among those who came to Wa, Kunata, Juna and Dao are in the Kabanya ward, while Sambaleyiri is the ward of the Senu.”

73While it is difficult to identify all the Muslim dyamou in Wa, I note some contradictions in the traditions: whereas the introduction by the yar-na and Traoré points to the Yalsé of Burkina—prior to the establishment of the Dagomba hegemony in Wa and Bouna—the other dyamou show little relation with Bouna and Kong which might be expected from the traditions.

74While the Baghayogho are historically well documented for Timbuktu from the 16th to the 18th century, and all members of the Baghayogho dyamou feel a certain unity, the links from Timbuktu to the different diasporas are difficult to establish. Thus Musa, Mustafa or Suleiman, do not appear in the genealogy established for Timbuktu, because our knowledge of 17th- and 18th-century Timbuktu Baghayogho is sparse, and we know even less about the emigrants—and appear as mythical ancestors. So far the three separate Baghayogho branches: the western branch in Gabu, the central branch in the Wasulu which gave rise to those in the Konyan and Worodougou, and the eastern yarse branch in Mossi and Dagbani cannot be connected to common ancestors. Yet the split into branches seems concerted and well coordinated with other Muslim families, and we suggest that it represents the result of a mission plan for the animist peoples of the south undertaken by Timbuktu religious leaders and organized after Mohammed Askia’s hajj for whom the actual name of the missionary was less important than his relationship to a renowned Muslim of almost holy reputation. Further research in Timbuktu itself, but also in such diaspora centres as Diakha, Kelenya, Koro, Mane, Kaya, Ouagadougou, Nalerigu and Sabari—and why not in Baghdad—would not only add more details to the family history but perhaps shed light on the missing genealogical links between the diaspora and its founders.

Haut de page


Asmis, R.
1912 “Die Stammesrechte von Sansanne Mangu”, Zeitschrift für vergleichende Rechtswissenschaft, 27 (1 & 2): 73-89.

Baiao, A. & Bensaude, J.
1940 O manuscrito Valentim Fernandes (Lisboa: Editorial Atica).

Barth, H.
1849-1957 Travels and Discoveries in North and Central Africa (London: Royal Geographical Society).

Bernus, E.
1960 “Kong et sa Région”, Études Éburnéennes, 8: 240-321 (Abidjan: Ministère de l’Éducation nationale, Direction de la recherche scientifique).

Binger, L. G.
1892 Du Niger au Golfe de Guinée par le pays de Kong et Mossi (1887-1889) (Paris: Hachette).

Blondiaux, F.
1897 “Du Soudan à la Côte-d’Ivoire”, L’Afrique française, 10: 339-345; 370-372.

Boutillier, J.-L.
1993 Bouna, royaume de la savane ivoirienne (Paris: Karthala).

Bowdich, T. E.
1819 Mission from Cape Coast Castle to Ashantee (London: Murray, repr. F. Cass).

Clozel, F. J. & Villamur, R.
1902 Les coutumes indigènes de la Côte-d’Ivoire (Paris: A. Challamel).

Conteh, I., Gomes, M., Sidibe, B. K., Skinner, D. E. & Wood, N.
1987 A History of the Migration of the Baayo Family from Timbuktu to Bijini in Guinée-Bissau, Tarikh Bijini mimeo.

Delobsom, D.
1933 L’Empire du Mogho Naba (Paris: Loviton & Cie.).
1934 “Notes sur les Yarcé”, Revue anthropologique, XLIV (10): 326-337.

Derive, M.-J.
1990 Étude dialectologique de l’aire manding de Côte-d’Ivoire (Paris: CNRS et ACCT, Peeters).

Derive, M.-J. & Nguessan, A.
1976 Chroniques des grandes familles d’Odienné (Abidjan: Institut de Linguistique appliquée, LVII).

Dieterlen, G.
1955 “Mythe et organisation sociale en Afrique occidentale”, Journal de la Société des Africanistes, XV, partie I: 39-76; partie II: 125-131.

Fernandez, V.
1940 Descripçao da Costa de Africa. Ed. by Balao & Bensaude, Manuscript in the Royal Archives of Munich.

Gallais, J.
1967 Le Delta intérieur du Niger (Dakar: IFAN).
1973 Pasteurs et paysans du Gourma: la condition sahélienne (Paris: CNRS-Bordeaux: CEGT).

Geysbeek, T.
1996 “A traditional History of the Konyan (15th-16th Century): Vase Camara’s Epic of Musadu”, Printed Paper Dept. of History, Michigan State University, 48-85.
2004 History from the Musadu Epic: The Formation of Manding Power on the Southern Frontier of the Mali Empire, Ph. D. Dissertation, Ann Arbor, Michigan State University.

Giesing, C.
2003 Stammbaum (lasili) der Baayo von Bijini und Janna, personal communication.

Goody, J.
1964 “The Mande and the Akan Hinterland”, in J. Vansina, R. Mauny & L. V. Thomas (eds.), The Historian in Tropical Africa (London: Oxford University Press): 195-218.

Gouvernement général pour l’Afrique française
1906 La Côte-d’Ivoire (Paris: Larose).

Green, K. L.
1984 “The Foundation of Kong: A Study in Dioula and Sonongui Ethnic Identity”, Ph.D. Dissertation (Evanston: Northwestern University).

Hamdoun, S. & King, N.
1975 Ibn Battuta in Black Africa (London: Rex Collings).

Houdas, O. (French translation by)
1966 [1901] Tedzkiret en-Nisian fi akhbar moulouk es-Soudan (Paris: Librairie d’Amérique et d’Orient).
1981 [1900] Tarikh es-Sudan, par Abderrahmane ben Abdallah ben ’Imran ben ’Amir Es-Sa’di (Paris: Maisonneuve).

Houdas, O. & Delafosse, M. (French translation by)
1981 [1913] Tarikh el-Fettach,  par Mahmoud Kâti ben El-Hadj El-Motaouakkel Kâti (Paris: Maisonneuve).

Hunwick, J.
1990 “A Contribution to the Study of Islamic Teaching Traditions in West Africa: The Career of Muhammad Baghayogho 930/1523-24 to 1002/1594”, Islam et Sociétés au Sud du Sahara, 4: 149-163.
1999 Timbuktu and the Songhay Empire (Leiden-Boston: Brill).

Ibn Khaldoun
1968-1969 Histoire des Berbères et des dynasties musulmanes de l’Afrique septentrionale, 4 vol. (Paris: Geuthner).

Innes, G.
1976 Kaabu et Fuladu: Historical Narratives of the Gambian Mandinka (London: University of London School of Oriental and African Studies).

Kamian, B.
1959 “Une ville de la République du Soudan: San”, Les Cahiers d’Outre-Mer, Bordeaux, 12 (45), janvier-mars: 216-231.

Levtzion, N.
1968 Muslims and Chiefs in West Africa (Oxford: Clarendon Press).

Mangin, E.
1921 Essai sur les us et coutumes du peuple mossi au Soudan occidental (Paris: Challamel).

Marc, L.
1909 Le Pays Mossi (Paris: Larose).

Marty, P.
1915-1916 L’Islam en Mauritanie et au Sénégal (Paris: Leroux).
1920 Études sur l’Islam et les tribus au Soudan, vol. II, Tombouctou (Paris: Leroux).
1922 Études sur l’Islam en Côte-d’Ivoire (Paris: Leroux).

Massing, A. W.
1980 “The Mane, the Decline of Mali, and Mandinka Expansion Towards the South Windward Coast”, Cahiers d’Études africaines XXV (1) 97: 21-55.
2000 “The Wangara, an Old Soninke Diaspora in West Africa?”, Cahiers d’Études africaines XL (2) 158: 281-308.

Monod, T., Teixeira da Mota, A. & Mauny, R.
1951 Description de la Côte occidentale d’Afrique par Valentim Fernandes (1506-1510) (Bissao: Centro de Estudios de Guiné Portuguesa).

Monteil, C.
1971 [1932] Une cité soudanaise: Djénné (Paris: Édition Anthropos; Londres: Institut international africain).
1924 Les Bambara du Segou et du Kaarta (Paris: Larose).

Monteil, P. L.
1894 De Saint-Louis à Tripoli par le lac Tchad: voyage au travers du Soudan et du Sahara accompli pendant les années 1890-91-92 (Paris: Alcan).

Monteil, V.
1968 “Al-Bakrî (Cordoue 1068), Routier de l’Afrique blanche et noire du Nord-Ouest (Kitâb al-Masâlik wa-l-Mamâlik)” Bulletin de l’IFAN, série B, XXX (1): 39-116.

Moundekeno, S.
1979 “Essai sur la mise en place des populations de Haute-Guinée” (Kankan: Centre culturel): 1-24.

Nagel, L.
1978-1995 Traditions from Barala, Côte-d’Ivoire (unpublished manuscript in Geysbeek (2003, Annexe X)).
1986 “Focus on Barala”, The Malinke People of Ivory Coast, (Odienné: WEC Mission, unpubl. paper): 11-13.

Ndiaye, B.
1973 Les castes du Mali (Bamako: Éditions populaires).

Niane, D. T.
1982 Sundiata ou l’Epopée mandingue (Bamako: Éditions populaires).

Person, Y.
1964 “En quête d’une chronologie ivoirienne”, in J. Vansina, R. Mauny & L. V. Thomas (eds.). The Historian in Tropical Africa (London: Oxford University Press): 322-337.
1968 Samori, une révolution dioula (Paris: Presses universitaires de France).

Randau, R.
1934 “Les Yarcé”, Revue Anthropologique XLIV (10): 324-325.

Roouveroy van Nieuwaal, E. & E.
1976 Ti Anufo (Leiden: KITLV).

Saad, E. N.
1983 Social History of Timbuktu: The Role of Muslim Scholars and Notables (1400-1900), (Cambridge: Cambridge University Press).

Seefried, A.
1913 “Beiträge zur Geschichte des Mangu Volkes in Togo”, Zeitschrift für Ethnologie, 421-435; also in Togo National Archives, Reichs-Kolonial-Amt, vol. 4 (3), Togo: 307-370.

1975 “Table ronde sur les origines de Kong”, Annales de l’Université de Côte-d’Ivoire (Abidjan).

Tauxier, L.
1912 Le Noir du Soudan (Paris: Larose).
1917 Le Noir du Yatenga (Paris: Larose).
1921 Le Noir de Bondoukou (Paris: F. Leroux).
1924 Nouvelles notes sur le Mossi et sur le Gourounsi (Paris: Larose).
1942 Histoire des Bambara (Paris: F. Leroux).

Tiendrebeogo, Y.
1990 “Histoire traditionnelle des Mossi de Ouagadougou”, réd. par R. Pageard, Journal des Africanistes, XXXIII (1): 7-46.

Wilks, I.
1963 “A Note on the Early Spread of Islam in Dagomba”, Transactions of the Historical Society of Ghana VIII: 87-98.
1989 Wa and the Wala: Islam and Polity in Northwestern Ghana (Oxford: Clarendon).

Wilks, I., Levtzion, N. & Haight, B.
1986 Chronicles from Gonja (Cambridge-New York: Cambridge University Press).

Zech, Graf v.
1904 “Land und Leute an der Nordwest Grenze von Togo”, Mitteilungen des deutschen Schutzgebietes 17: 107-135.

Haut de page


1 In general—except in quotations—I followed the full spelling in the Tarikh es-Sudan, which writes Baghayogho, with double ghain, but occasionally drops the second inter-vocalic gh to yield Baghayo’o.
2 “Mande diaspora”: I have shown the Wangara to have Soninke roots, but through the assimilation of the latter into the Mande mainstream—where they are designated as Marka, or Dafin—they can be considered today as part of the wider Mande culture.
3 See Massing (2000), for a discussion of the sources referring to Gâna and Wangara.
4 These maraboutic practices are still widespread. In 1969, I witnessed myself how such a Molly man tried to convince a chief in a Gola village, at the Sierra Leone—Liberia border, of the powers of Islam by healing his wife with a combination of charms and verses taken from a Koran, written on a tabloid with ink and washed off with water.
5 The gold mines of the south are scattered along a line corresponding to the present border of Mali with Guinea the Ivory Coast—from Kenieba, across the Manding mountains to the Tinkisso valley with its placer mines, and Yanfolila, Sikasso and along the Comoe and Black Volta rivers into the North West Akan areas at Banda and Bono. At the Siama mine, 100 km south of Sikasso, the prospective works revealed placer mines estimated at 800 years of age.
6 Ibn Batuta first writes Zaghari, then Zagha, but appears to refer to the same town. Hunwick (1999) suggests Diakha as a transcription, which would point to Dia on the western branch of the Niger, one of the oldest towns of Mali. “[t]he people of Zagha are old in Islam and very religious and seekers of knowledge” (Ibn Batuta, Voyage 1352-1354).
7 For me, this implies that the Sonrhay royal clan were descendants of Wangara.
8 From the Ouagadou, the territory of ancient Ghana. Articles in Tapama, nos. 2 et 3, 1998, claim that Oua in Songhai means “milk”, and koré “very white”, meaning “white as milk”; the term is said to be derived from the Awkar, the area formerly inhabited by the Soninké. Source: elders of Diré, Goundam and Niafounké and Tindirma. What is important here that Da’o, Dabo are dyamou associated with Wankore, and Tendirma is a frequently mentioned Wangara town.
9 Zendj are captives in the sole ownership of the royal clan.
10 Also called Za Yasiboi (Houdas 1981: 9); on the basis of his family name, Da’ou or Dabo, which like Baro is associated with the Ouangara.
11 Like the Romans, the Malian kings requested their vassals to attach their children as hostages to the sovereign’s court, usually the daughters as secondary wives, the sons as warriors. “When the children reached the age to enter service, the sultan of Melli took them with him. At that time these princes were vassals and it was custom that the sons of the kings were forced into the service of their suzerain. Some of these young people returned to their country after having served for a certain time; others on the contrary, continued to stay with their suzerain till their death” (Houdas 1981: 10).
12 The religious establishment were Malikites, while Sonni Ali was Kharedjite, who appointed Imams of local origin, while the others insisted on Imams related to the prophet’s family.
13 This apparently was not the Diakha in Mali but a homonymous settlement near Ayorou, south-east of Gao (Houdas & Delafosse 1981: 106, n. 1).
14 According to Delobsom (1933: 207) the ancestors of the Yarsé-nabas, the Koanda— or Konaté—had already come to the Mossi under the 8th Naba, Koudoumié, 1337-1358, at Mansa Musa’s time. Was this the first attempt to convert the animists?
15 Marty (1920, t. III: 65) says of Tendirma: “capitale du kourmina, construite sur la ville juive de Daro (Da’o) en 1496-1497.” Therefore, were the Wangara white because they were of Jewish origin, Yemen, and later mixed with blacks? There are still today families of Jewish origins in Niafounke, Diré and Tendirma, as in 1996 an announcement in the Malinet was seeking Jewish teachers for that region.
16 According to local history, Djenné, which means Dia-ni, i.e. small Dia, has been founded my migrants from Dia (C. Monteil 1971).
17 The Tarikh es-Sudan gives his departure as Safar 902 (Sept.-Oct. 1496) and his return in 903 (1497/1498); the Tarikh el-Fettach gives the departure as Sept. 903. The hajj was made in company of several governors and seven experts, viz. alfa Saleh Diawara, môri Mohammed Haogaro, alfa Mohammed Toulé [Touré], Gao Zakaria Sanafa, môri Mohammed Tenkou [Tenenkou?], cadi Mohammed Niédobogho and Mahmoud el Kati, the uncle of the author of the Tarikh es-Sudan.
18 Cadi is the judge who settles legal disputes among Muslims according to Islamic law; the customary judge of Djenné who settled, in the two centuries since the “conversion” of the chief of Djenné Koboro, the legal disputes among the inhabitants had been called khatib.
19 The old Djenné, today Djené Djeno, was abandoned, and the new town suffered from frequent collapses of its walls believed to be caused by spirits (C. Monteil 1971)—due to the annual filling up of the arm of the Bani, along which Djenné is built, the water rising in the earthen walls makes them collapse. The shrine is still a sanctuary today located at the beginning of the hospital area.
20 The Arabic text spells “Fodi al faqih Muhammad Saquwa al Wankari”. Houdas (1981: 30-33), in his translation mentions that manuscripts A and B omit “Fodi”, and Sanou replaces Saqu in manuscript B. On page 34 appears “Mohammed Foudiya Sanou”; while the Arabic gives the first two words as honorific titles— fode faqih (excellent saint).
21 Djenné and the djenné-were had remained loyal to Sonni Ali since his conquest of 1468, and Mohammed Askia preferred a foreigner as cadi not only for religious but also for political reasons.
22 The Arabic text is in fact ambiguous: “rahal min baladih fi ard bit” does not necessarily mean “in the land Bit” but could also be read as “fi ard beit”, i.e. “in his homeland”. A Bitu is mentioned at the beginning of the chapter in the Tarikh es-Sudan as the southern border of Mali, which extends “from Bitu to Fangaso, from Kanyaga to Singhilu”. Elsewhere Bitu is designated as the land of gold; the name probably referred to the auriferous southern regions, in particular Lobi and Ashanti. But it could also refer to Biro (Walata), which was also in strife at the end of the century. For Fodiki, the Arabic text writes Fudi al-faqih Muhammad Saquwa al-Wankari. Houdas (1981: 30) mentions in his translation that this name lacks in the manuscripts A and B, and that Sanou replaces Sagou in the manuscript C; on page 34 of the Tarikh es-Sudan he is called Mohammed Foudiya Sânou. The difference between the Arabic and the translation is considerable, the Arabic establishing the first two words as titles, placed before the name fode el faqih, i.e. the excellent saint. But Foudi could also refer to a proper name or a dyamou, namely Fodé or Fodi, evidently derived from the Arabic, and Fadiga or Fodiga. Person (1968: 168-169) mentions Fondiya: “Appelés par les Kamaghaté, les Bambara, liés à Kong et à Ségou, repoussèrent les Sénoufo dans le Nöölu où la frontière allait se stabiliser jusqu’au xxe siècle. Ils imposèrent leur autorité aux Malinké animistes (Kuruma, Fani, Fondya, Bamba et Koné) de façon à contrôler tout l’Ouest du futur cercle d’Odienné”. The dyamou mentioned as those of “Malinke animists” are also those of noumou, i.e. blacksmiths. Particularly the Bambara and Senoufo blacksmiths were the guardians of such initiation traditions of the Komo or Poro as circumcision and scarification, and resisted Islam in the rural areas. Thus Foudya might have been a converted blacksmith. The name Sanou seems to refer to Bobo-Dioulas or Marka in the region of Bobo. One of theirs, Fama Mohammed Sanou at Boussoura is mentioned by P.L. Monteil (1894: 93). The Sanou had sold Tieba as slave in his youth, which cost them his revenge and the war with them in 1893.
23 Probably Tene, in Bobo country, 60 km from Djenné.
24 Whatever the actual referent, the fact that Bigho in Ghana and Biro near Walata are homonyms does not seem coincidental to me; as so often the founders of the new settlement Bigho appear to have named it after the terminus of the trans-Saharan trade thus pointing toward their own origins.
25 Apparently, chi Baro was not dead.
26 “On the 16 of Djoumada 1 (12 March 1656) news was received from the caid […] who was at Djenné […] that the nefarious, rebellious, kharedjite Djinni-Koï-Bokar had armed troops and et began to march on Kanba’a with the plan to take Sorya and conquer his town, in order to intercept the route of those who were travelling to Djenné” (Houdas & Delafosse 1981: 488).
27 Besides Mohammed Sanou, there were Abbas Kibi, Mahmoud Bagayogo and Mohammed Benba Kanâti; the following four, Fouko, Kanadji, Tanata’o, Sonqomo, do not bear the eponym el-Wankori; two others, Omar and Ahmed Torfo, father and son, appear to be djennenké, and Bokar Terouari, the previous to last, seems to have come “from the royal family of the Kaladougou”, that is the Traore from Kala (Houdas 1981: 35), the island between the Niger and the Bani.
28 Should one take “dynasty” literally and assume that all came from the same family? They were all black and Wankore origins are specifically mentioned for four of them. I assume Fode to be the father or grand father of the 3rd cadi Mahmoud.

C. Monteil (1971: 31-32) quoting a version of the Tarikh es-Sudan belonging to the emir of Douentza, gives precious information on the origins of the Djenné population: “Par Guirga et plus exactement Guiriga on désigne en soninké les individus plus ou moins métissés de ’rouges’ (guiré); par ’rouges’, les noirs désignent tous ceux qui ne sont pas nègres: arabes, berbères, amaures, peuls et les métis des uns des autres. Les Nono étaient de ces Sanhadja dits Guirga […]. Les Soninké dits aussi Wakoré, Sarakolé, marka sont également des Guirga. Le mot Azer ou Adjer n’est qu’une forme berbère désignant les Guirga de Dia qui étaient aussi des Nono. La tradition rapporte, en effet, que les gens de Nono sont venus du nord, du sahel par bassikounou. Arrivés à Dioura, une scission se produisit: sous le commandement d’une femme nomée Fatma Diara certains de ces Nonos vinrent à Killa et Kossala; plus tard ceux de Killa gagnèrent Koubay, où les Niafogou actuels prétendent être les descendants de Fatma Diara. Les autres dissidents de Dioura, ayant pour chef le frère de Fatma, vinrent à Nono, puis à Dia-Kolo et enfin fondèrent Dia, métroplole célèbre d’où sont parties des colonies fameuses telles que: Sansanding qui reçut de Dia les familles Khoma, Sissé, Berété, Diabi, Komé; Diakha-sur-Bafing (Haut-Sénégal), cité considérable qui était formée par la réunion de dix villages dont sont sortis tous les Diakhanké de l’ouest (Haut-Sénégal, Gambie, Guinée); d’autres dans le Djimbala, le Bara, le Fitouka avec, dans cette dernière région, un nouveau village de Nono ou Nounou, près de Niafouké. Vers le Sud, une de ces migrations de Dia fonda Djénné […]. Ainsi à Djaboro, village de Bozo dont l’emplacement est peu éloigné du village actuel de Pérou, une fraction de Nono, commandée par un certain Namankéni, s’imposa et fonda, là, la première Djénné. Il y a eu, en effet, deux Djénné comme le remarque en ces termes, le Tarikh es Soudan [Houdas 1981: 23]: ’À l’origine, la ville (de Djénné) avait été bâtie à Djaboro. Plus tard, on la déplaça pour la transporter au lieu où elle se trouve actuellement. L’ancienne ville était située près de la ville moderne dans la direction du Sud’.”

29 Djenné probably recognized the legitimacy of the old Askia against the overthrow by his sons and brothers, and therefore called them “oppressors”.
30 According to the Tarikh es-Sudan (Houdas 1981: 43-47), in 903 A. H. (1523-1524), cf. biographical note by Ahmed Baba in Hunwick (1990).
31 “At the moment when the Askia Daoud had decided to give the function of cadi to one of the two brothers, and despite his insistence and the pressuring demands on the part of the inhabitants—both refused and persisted, until finally both took refuge inside the mosque and stayed there several months, visited daily by the Askia’s messengers until the latter decided to forgive them saying: ’the condition is that both will come to find me in Gao and let me enjoy the divine favours granted to them’” (Houdas & Delafosse 1981: 208).
32 Personal communication by Hasseye M. Baghayogho, guardian of the Sidi Yahya mosque, 10 January 1997, who invited me to his house to show me the manuscript of his History of Timbuktu and the manuscripts of his forefathers, among others a copy of the Tarikh el-Fettach by Abderrahman Mutawwakel el-Kati.
33 The great mosque, Zinguiber, and the Sankoré mosque were built around 1509 under cheik Aqib.
34 It had been founded by the Touareg at the end of the 5th century A. H., around 1122, to replace Biro (Walata) as commercial centre (Houdas 1981: 30).
35 The Tarikh es-Sudan in Chap. 8 says that they were Senhadja and closely related to the Lemtouna, Messoufa and Beni Djoddala, nomada—and ancestors of the Almoravids—between the Mauritanian coast, Atar and the Aoukar—and claims their origin from the Beni Hillal who moved from Yemen in the 10th century and, according to Ibn Khaldoun (1968-1969), founded Marrakech between 1062 and 1069. Chinguetti is 75 km east of Atar.
36 Houdas & Delafosse (1981: 171) lists Sidi Yahya’s as second in the list of eminent graves and says: “The jurisconsultes of Timbuktu whom I have encountered vistited him constantly, among others Sidi Ahmed Baba and the two eminent jurisconsultes Ibrahim and Ahmed, both sons of Ahmed Baghayogho”.
37 The Tadla is a plain between the Middle Atlas and Marrakech, with Beni Mellal as centre. His genealogy of 16 generations is given in Houdas & Delafosse (1981: 101).
38 According to Houdas & Delafosse (1981: 227-29), the Askia had refused the investiture of Timbuktu to el-Aqib’s son Omar, and only appointed him upon urgent remonstrance by Mohammed Baghayogho, while the inhabitants believed that he (Baghayogho) the same was interested himself, until the Askia sent a message saying “we would not have given Omar the functions of cadi during our lifetime without the intervention of Mohammed Baghayogho”.
39 Bilanga is a Wangara (Yalse) foundation and settlement, 45 km east of Kaya (Barth 1849-1857, vol. V: 583).
40 It seems that the Dendi chiefdoms of the Bariba of Northern Benin have their origins in these Gourma remnants of Songhai.
41 Nouhou had been freed from prison, where the Askia Isa’aq II had put him, but continued the Songhai resistance against the Moroccan army for 20 years in the Gourma-Dendi (south-west Niger, east Burkina and north Benin).
42 San means chief, lord, according to Houdas & Delafosse (1981: 366, n. 3).
43 In January 1997 Hassey Mahmoud Baghayogho, the present Imam of the Sidi Yahya mosque, gave a description of an incident which occurred at the corner of the street where the Sidi Yahya mosque is located: the two Wangara men were of his family, and were pushed along the road by the Moroccan musketeers at the barrels of their guns; when one of them wanted to stop at the street corner to urinate, the Moroccan soldiers knocked him down, whereupon his brother beat the soldier. The Moroccans believed this to be a sign of revolt and fired on the entire group, while the other group which was driven down a parallel street arrived safely at the mosque and was arrested and deported (personal communication).
44 Saad (1983: 56) is not right in his statement that the Moroccans destroyed the Baghayogho family.
45 On the death of Abou Ishaq: “On 19 August 1638 died the cheikh, legal consultant, wise, learned, perfect and practicing Abou Ishaq Ibrahim, son of the legal consultant Ahmed Baghyo’o El Ouankori” (Houdas 1981: 449).
46 Alternately called Mostafa es-Sahara or Mostafa es-Sahili, his filiation is contradictory: while in another place he is mentioned as son of Ahmed, and thus nephew of the great Mohammed B. “On Wednesday between midday and 3 o’clock of the 5th of the 1st of Rabi 1025 (19 March-18 April 1616) died the law consultant El Mostafa-ben-Ahmed-ben Mahmoud-ben-Abou-Bekr-Baghyo’o; he was buried the same day (God have mercy on him). He was a man of a soft character, tranquil, taciturn and bearing with patience the intrigues of the people. He had studied under the direction of his famous paternal uncle, the jurisconsultant Mohammed Baghayogho, who had made him study the Risala, the Mokhtasar and other works. But he had not completed the study of the Mokhtasar with his uncle. He also received lessons from the jurisconsultants Otsman el-Filali, Mohammed b Mohammed Korai and Abderrahman b. Ahmed el Modjtahid (called ’le juriste’). With the latter he studied the Modawwana and Mouatta.... At the beginning of his studies and during his uncle’s lifetime, he had learnt from Ahmed Baba, son of the jurist Ahmed, the elements of the Arabic language, the Mokhtasar and other subjects. The son of his paternal aunt, Mahmoud, had taught him the Alfiya and other sciences. When Ahmed Baba returned from Marrakech, El-Mostafa sometimes attended his conferences. He was named Imam of the Mohmmed Naddi mosque in the month of Cha’aban of the year 1008 (15/4-14/5 1600) and kept these functions till his death. He assisted the preacher of the mosque from the year 1016 (28 avril 1607) on. He was born in the year 973 (29/7/1565-19/7/1566)” (Houdas 1981: 365-6).
47 El Fa, from el-faqih, “the learned man” was later contracted to Alfa.
48 The Arabic T-m-t, vocalised as Timeta, is intriguing; there are only two dyamou in Mali with that sequence of consonants: Tomota (Bozo from Dia) and Timité, another Wangara family. My research in Mali identified Timité as being present in the Bobo settlements of Ouan, Fangasso, Tene and Sofara, all less than 50 km from Djenné. The Timité are said to reside by local informants from Ouan in the village of Poromo, near Fangasso. Of course, one should also explore the possibility of a relation between the Tomota of Dia and Timité in the Djenné region, given the closeness of the two settlements. Incidentally, Kamaghaté are also among the main families of Ouan. (I conclude that the region around Djenné is one of the main Marka settlement areas, where the Soninké who drifted down from the Mema, Goumbou and Nara to Banamba, and from there east, after crossing the Niger and Bani... Up to the present day a detailed study of the Marka-Dafin of Mali or Burkina Faso is lacking. The Timité are also among the Muslim families of Séguéla, Sarhala and Bondoukou, where they are the almamys come from Begho.
49 Bagnio or Banyon are other variant spellings of Baghyo’o or Baghayogho.
50 They provided camel, fodder, water and food for the camel drivers and guarding soldiers, and financed the salt and riverboat to Djenné.
51 A major branch of the Baghayogho is indeed located in Séguéla and Koro (Marty 1920).
52 Their synonym here is Kourouma.
53 Interview at Faragouaran canton N’tentou, 17 August 1997, with Nkoro Baghayogho, Baba B., Numuçe B., Abdoulaye B., Cemoko, Mamé Baghayogho. According to the manuscript Les Baghayogho, origine et histoire de leur installation dans le Wasulu, shown to me, they are represented at Kélénya, Faragouaran, Danou, Bolou. They say to have come from Timbuktu to install themselves at Kelenya. They descend from Arabs, are among the first who made the pilgrimage (from Kelenya). The name comes from Bagadaji, people of Bagdad. The elder pulls a sheet of paper from his boubou, entitled: “Bureau exécutif de l’Association des Bagayogos, alliés, cousins et amis.” The paper mentions five brothers at Kélénya: 1)  Wombo (Wahab), 2)  Zanifing, 3)  Danou Moussa, 4)  Nfamoussa and 5)  Malab Wombo established himself at Kelenya; Zanifing at Niakoni and his descendants form Bolou; Danou-Moussa installs himself at Danou; Nfamoussa leaves Kelenya and establishes Koro (Odienné) (founded at a sport where his walking stick got hooked on a tree called koro-koro, and thereafter his descendants left to Mankono, Séguéla, Touba. Malab, left to Farangouaran. The Baghayogho of Bamako come from Kélénya.
54 Marty (1920, t. IV: 14) signals several “Barayoro” of diakanké origin: Silatigui B. at Kayes Medine.
55 Practically the same citation appears in Clozel & Villamur (1902: 44-45) and Marty (1922).
56 The Diomandé (Mande from Dion) are the Kamara, Soumahoro, another name for Kanté, and Sarhayorho, a synonym for “Barhayorho”.
57 Fonikaman is Feren Kama(ra), see Massing (1980).
58 According to Delobsom (1933) and Tiendrebeogo (1990), the 23th Moro Naba was Naba Zombré, 1681-1744.
59 Tiendrebeogo (1990) gives the following list of 18th-century Nabas: 23.  Zombre 1681-1744; 24.  Kom I. 1744-1762; 25.  Saga I 1762-1783; 26.  Doulougou 1783-1802.
60 I have not yet had the time to go to Sabari to check out the Baghayogho claim.
61 Saghyardu points to Sarhayorho, a synonym of Barhayorho according to Marty (1922: 149, 157).
62 Tait manuscripts, collection B, in the Institute of African Studies, Legon University, quoted in quoted in Wilks (1963: 90) and Levtzion (1968: 88-93).
63 See Seefried (1913), Asmis (1912) and Zech (1904). Seefried speaks of the royale line in the Djeremabo quarter, the line of Soma in the Sanbana quarter, and of the Jokoë-line in Fomboro (written by Zech Jukui or Jakuba). The source: is a copy of a tarikh in the hands of the Imam Amadu, and a second manuscript signed by Imam Gasama, the imam of Bema Bonsafo.
64 Banda-wi, an Arabic form of “from Banda”, Banda was also called Fougoula in Dioula, designating the Muslim head cover. Old Begho was peopled by pagan Ligbi and Weila of Banda who, after the destruction of Begho, returned to Banda. The Kamaghate were the second dyamou after the Ligbi Bamba, and are still today the Imams in Namasa, the village founded by the Muslims after the destruction of Begho. In my history of Gonja (in preparation), I show that the Imams of Gonja are intimately linked to Kamaghate and the departure from Begho. I propose here the hypothesis that the beginnings of Gonja coincide with the Mango migration and the destruction of Begho. Ashanti attacked Banda in 1741 (Bowdich 1819: 170) which could be the cause of the fall of Begho—which was a colony of Kong, and that Kong dispatched the Dioula from Groumania to help the colony.
65 “Le lignage patrilinéaire Kambaya dont l’ascendance remonte à deux frères, Gazama et Amadu Bandawiu, fils de Yusuf, qui était resté dans la Côte-d’Ivoire, fait partie d’un grand clan patrilinéaire d’origine dyula qui est assez souvent dénommé Kamaghate, Kamatay, Jabakte, Kambakatiu. A N’zara le lignage se subdivise en trois segments” (Roouveroy 1976: 76).
66 “All Muslims claim descent from Fati Morukpe, and are called Sakpare. The Nsoa’wura is the chief of the Muslims, he is the chief’s spokesman and the head of the heralds, Dogte. There is a Nso’wura in each division, some dogte’s and Muslims at the same time. There were eight divisions: Buipe, Kpembe, Bole, Tuluwe, Kusawgu, Wasipe, Kong, and Kandia. The right of the latter two was revoked, and Wasipe’s right was restored in 1931. Each of the division chiefs was given a Sakpara Imam. Buipe is in charge of Jakpa’s tomb (misiri = referring perhaps to the first Mosque) and the handling of affairs after the death of a Paramount till the handing over to the new. He belongs to the Jabaghte group” (Levtzion 1968: 61; Goody 1964: 196). Kamagh’te is the patronym of most of the Sakpare, and also of the Imams of Namasa, who continue the line of the Imams of Begho, who founded the Kamaghte of Gonja. At Bole, the Sakpare Imams state that they came from Buipe after settling in Chama. Another group, the Dabo, say they are the descendants of Dabo who lived in a small village near Buipe. (Dabo is also a Mande patronym and represented among the Sansanne Mango.) The Sakpare Muslims of the Tuluwe chief reside at Chama, which is the residence of the Kabir-wur, priest of the Lansah shrine. (The Kagbir-wura of Kito, south of Kpembe, is actually a shrine priest). They say they were from Kafaba, a pre-Gonja Muslim settlement. The Kafaba chief was a peacemaker, Bengbangpo or councillor of Kpembewura who moved to Kito. Sakpare Muslims are attached only to the major divisions, but there are some at Kulepe. Only Kusawgu has no Sakpare Muslims as the first Muslim was a Hausa trader, Sanauta, the great-grandfather of the present Imam. Before the Daboya chief moved from Wasipe to Daboya, he had Sakpare Imams, but later dismissed them.
67 Personal information by the Bouna Imam Cissé. Boutillier (1993: 282) confuses them with the Kamara: “Les Kamara (on prononce aussi Kambara) semblent être le premier groupement étranger musulman à s’installer à Bouna, sous le règne de Kankru; ils seraient arrivés de Larabanga […] mais leur itinéraire antérieur à leur arrivée à Larabanga est très controversé. Ils furent […] les derniers musulmans de Begho xviie.” Ils furent les premiers imams de Bouna (ibid.: 283). “One finds Imams of Larabanga origin only in Gonja chiefdoms with a strong Dagomba cultural” (Levtzion 1968: 76). “Les relations entre les Kamara et le clan royal de Bouna sont particulièrement étroites: […] le clan royal donne des filles de leur propre clan en mariage à des hommes kamara […] ce qui a pour résultat, compte tenu de la nature des rapports oncle-neveu en milieu koulango, qu’un grand nombre de Kamara se sentent autant koulango que kamara et que la totalité d’entre eux parlent les deux langues. Le quartier kamara […] représente actuellement à lui seul plus du tiers de toute la population d’origine étrangère” (Boutillier 1993: 283). Marty (1922: 236) mentions Kambara-soura as 4th quarter of Bouna. Following Benquey’s list the 4th quarter is called “Bani-Koumbala”, i.e. Gbane from Kombala. Kombala is a Nafana village near Kong (Massing 2000), According to Boutillier (1993: 291) “Les origines de Bégho se trouvent dans neuf tribus (qabila) ayant differents nasab (noms de salutation). La première Bamba. La deuxième: Kamaghatay […]. La neuvième: Kawtay”. In the following table (ibid.: 292), the nasab Kawté is set on the line of the qabila Kamara. Pronounced like Kanuté, Kawte is the family name of the Imams at Larabanga, who also call themselves Kambara ou Marra. Thus, the Imams of Larabanga are either from Kombala or Kamaghate.
68 Seefried’s (1913) account mentions that Bema left after having buried his brother Koranti; further, that three other Koranti left with Soma, namely Koranti and Koranti Bosoro. Roouveroy (1976: 75) writes: “du lignage de Gazama—i.e. du premier imam—sont descendus l’imam Namo Koranti […]”.
69 I will investigate this hypothesis in a separate paper on the Kamaghaté and Imams of Gonja.
70 Manuscript in the Arabic Collection IAS/AR 152 of the Institute of African Studies of University of Legon, Accra, Ghana.
Haut de page

Table des illustrations

Titre Fig. I: Genealogy of the Baghayogho of Djenné and Timbuktu
Fichier image/jpeg, 58k
Haut de page

Pour citer cet article

Référence électronique

Andreas W. Massing, « Baghayogho », Cahiers d’études africaines [En ligne], 176 | 2004, mis en ligne le 17 avril 2008, consulté le 20 janvier 2017. URL :

Haut de page


Andreas W. Massing

Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ), Eschborn, Germany.

Du même auteur

Haut de page

Droits d'auteur

© Cahiers d’Études africaines

Haut de page