¹ The term ‘tribe’ is used here only as an English equivalent of the Arabic Banū or Awlād.

² See Trimingham, J. Spencer, A History of Islam in West Africa (Glasgow, 1962), 156–60.Google Scholar

³ el-Bekri, Abou Obied, Description de l'Afrique septentrionale, rev. ed. (Paris, 1965), 324. Translated by MacGuckin de Slane.Google Scholar

⁴ Désiré-Vuillemin, G. M., Histoire de la Mauritanie (Nouakchott, 1964), 90–1;Google ScholarNorris, H. T, ‘Znāga Islam During the Seventeenth and Eighteenth Centuries’, Bulletin of the School of Oriental and African Studies, xxxii (1969), 496–526, see p. 498.CrossRefGoogle Scholar

⁵ Hamet, Ismaël (Ed. and translator), Chroniques de la Mauritanie Sénégalaise (Paris, 1911), 45;Google ScholarMarty, P., Etudes sur l'Islam et les tribus maures, les Brakna (Paris, 1921), 4–5;Google Scholarde La Chapelle, F., ‘Esquisse d'une histoire du Sahara occidental’, Hesperis, xi (1930), 35–95 see pp. 85–6.Google Scholar

⁶ ech-Chinguite, Ahmed Lamine, El Wasit (Saint Louis, 1953);Google ScholarMiské, Ahmed Baba, ‘Al Wasit. Tableau de la Mauritanie à la fin du xix^e siècle’, Bulletin de l'IFAN, xxx (ser. B) (1968), 117–64;Google ScholarNorris, , ‘Znāga Islam’, 497–9.Google Scholar

⁷ El-Bekri, , Description de l'Afrique, 328–9.Google Scholar

⁸ Mahmoûd Kâti ben El-Hâdj El-Motaouakkel Kâti, , Tarikh el-Fattach, rev. ed. (Paris, 1964), 314–15.Google Scholar Translated by Houdas, O. and Delafosse, M.Monteil, C., Les Khassonké (Paris, 1915), 357–60.Google Scholar

⁹ This combination of religious specialization, political neutrality, and commerce is also found with the Diakhanké trade network in the far western Sudan. See Curtin, P. D., ‘Pre-Colonial Trade Networks and Traders: The Diakhanké’, in Meillassoux, C. (ed.), Trade and Markets in West Africa (London, forthcoming).Google Scholar

¹⁰ See, for example, Hunwick, J. O., ‘Religion and State in the Songhay Empire, 1464–1591’, in Lewis, I. M. (ed.), Islam in Tropical Africa (London, 1966), 296–315.Google Scholar A few insubstantial hints suggest that states under clerical rule had existed still earlier in the history of the Senegambia. At the beginning of the sixteenth century, for example, a Pulo named Eli Bana and identified as a cleric in some accounts ruled over the region of Kahone near the present-day city of Kaolak. Whatever the nature of his state, it was overthrown by the invading Guélowar, who then set up the kingdom of Saloum. (Brigaud, , Histoire traditionelle, 165–6;Google ScholarBoulègue, Jean, ‘Contribution à la chronologie du royaume du Saboum’, Bulletin de l'IFAN, xxviii (ser. B) (1966), 657–62, see pp. 660–1.)Google Scholar

¹¹ Norris, , ‘Znāga Islam’, 520–1.Google Scholar

¹² Norris, (‘Znāga Islam’, 511, n.Google Scholar) states only that later Mauritanian authors refer to him as Amīr al'Mu'minīn, but evidence that he also used the title himself comes from French sources. Sieur Chambonneau, who was in Saint Louis during a part of the jihad, reported that he used the title, ‘bourguli, … which in their language means the great master of prayer’. (Chambonneau, ‘L'histoire du Toubenan, ou changement du Royes, et réforme de religion des nègres du Sénégal coste d'Afrique depuis 1673 qui est son origine, jusqu'en 1677’, in Ritchie, C. A. (ed.), ‘Deux textes sur le Sénégal (1673–1677)’, Bulletin de l'IFAN, xxx (ser. B) (1968), 338–53, see pp. 339–40.)Google Scholar Chambonneau was evidently confused here, since the other title, Imāmūna, is the one meaning great master of prayer. Bourguli is clearly bur juulit the Wolof translation of Amīr al-Mu'minīn. The similar Fula title, laamido juulbe, was the title used by later Fulbe leaders, such as Sheikh Umar. The French spelling used by Chambonneau follows the seventeenth-century French custom of representing the hard ‘j’ in African names by ‘g’, rather than the ‘dj’ or ‘di’ that became common in the nineteenth century.

¹³ Khaldun, Ibn, The Muqaddimah. An Introduction to History, 3 vols., i (New York, 1958), 465–72. Translated by Franz Rosenthal.Google Scholar

¹⁴ The principal translations of the Arabic accounts are Hamet, , Maurìtanie sénégalaise, 164–242;Google ScholarBasset, René, Mission au Sénégal (Paris, 1909), 435–541;Google ScholarSidia, Cheikh, ‘Tableau historique de Cheikh Sidia’, Bulletin de la comité d'études historique, et scientifique de l'AOF, ii (1921), 76–95;Google Scholar and a detailed report in Norris, ‘Znāga Islam’. French sources are especially Chambonneau, , ‘L'histoire du Toubenan’ with briefer notices in Cultru, P., Premier voyage du Sieur de la Courbe fait à la coste d'Afrique en 1685 (Paris, 1913), 132–3;Google ScholarBarbot, J., A Description of the Coasts of North and South-Guinea (n.p., 1732), 62.Google Scholar The last two report the millennial aspects, not mentioned in other sources.

¹⁵ Norris, , ‘Znāga Islam’, 514–15;Google ScholarChambonneau, , ‘L'histoire du Toubenan’, 339–40.Google Scholar The Satigi in question is named (as Bubakar Samba Lamu or Bokar Tabakali) only in the oral traditions collected by Captain Steff and preserved in Fords Gaden, IFAN, Dakar.

¹⁶ The religious revolution in Cayor is a fortunate instance where Arabic sources and Wolof traditions concur as to the main lines of the events. Arabic sources identify Ndiaye Sal by name, but one discrepancy is that Wolof tradition has Ma Fali assassinated on orders of Ndiaye Sal, while the French reports say he was driven into exile in the neighbouring kingdom of Baol. See Chambonneau, , ‘L'histoire du Toubenan’, 341;Google ScholarNorris, , ‘Znāga Islam’, 514;Google ScholarBrigaud, F., Histoire traditionelle du Sénégal (Saint Louis, 1962), 100.Google Scholar

¹⁷ Chambonneau, , ‘L'histoire du Toubenan’, 340–1:Google ScholarCultru, , Voyage du Sieur de la Courbe, 132–3.Google Scholar The traditional history of Walo does not conform to French and Arabic sources. Both Yerim Kodé and Fara Koumba are common names in the royal line, but neither name occurs at this period of time in the traditional king-lists. This is not altogether surprising, since Wolof rulers had alternate names, and the French may well have known them by one of these. Tradition states, however, that the pretender who ruled for Nasīr al-Dīn came from the Teedyek matrilineage. With some uncertainty about the named individuals and their roles, tradition nevertheless supports the view that opposition was stronger in Walo than any of the other three sedentary kingdoms. See Norris, , ‘Znāga Islam’, 515 n.;Google ScholarBrigaud, , Histoire traditionelle, 69–71;Google ScholarMonteil, V., Esquisses Sénégalaises (Dakar, 1966), 19 and 38.Google Scholar

¹⁸ Norris, , ‘Znāga Islam’, 516–17;Google ScholarChambonneau, , ‘L'histoire du Toubenan’, 343–5.Google Scholar

¹⁹ The kingdom is phonetically bundu, but it often appears on modern maps as Bondu or Bondou.

²⁰ For this account I have depended chiefly on Ndiaye, Saki, ‘The Story of Malik Sy’, edited and translated by Skinner, A. N. and Curtin, P. D., to appear in Cahiers d'études africaines;Google ScholarDieng, Doudou, ‘Le cercle de Bakel’ (unpublished MSS., about 1911, CRDS, Saint Louis);Google ScholarRoux, Emile, Notice historique sur le Boundou (Saint Louis, 1893);Google ScholarRançon, A., ‘Le Bondou’, Bulletin de la société de géographie de Bordeaux, xvii (n.s.) (1894), 433–63, 465–84, 497–548, 561–91, 593–647, see pp. 473–;6.Google Scholar

²¹ Terminology for the Fulbe, Peul, Fulani, Fula, or Tukulor is thoroughly confused. The usage here approximates the usage in Bundu, where the French terms Peul and Toucouleur merely distinguish occupation among those who speak Fula and are considered to be a single people. Thus, Fulbe (sing. Pulo), their own name for themselves, is used for all Fula speakers. Futankobe (sing. Futanke) are the Fulbe from Futa Toro, just as the Bundunkobe (sing. Bundunke) are Fulbe from Bundu. Phonetic spellings of African words other than Arabic follow the orthography recommended by the Bamako conference of 1965 under Unesco sponsorship.

²² Rançon, , ‘Bondou’, 469–72;Google ScholarRoux, , Le Boundou, 1–3;Google ScholarTardif, J., ‘Kédougou: aspects de l'histoire de la situation socio-économique actuelle’, Bulletin et Mémoires de la société d'anthropologie de Paris, viii (11th ser.), (1965), 167–230, see pp. 169–;71;Google Scholar Maimouna Mamadou Sy, Curtin collection of historical traditions of Bundu, tape T1(1), available at IFAN, Dakar or African Studies Association, Center for African Oral Data, Indiana University, Bloomington, Indiana.

²³ Several different traditional genealogies for Malik Sy's ancestors in the male line have been preserved. The longest tend to agree in tracing the line back to a Habiballāhi (Skinner, and Curtin, [eds.], ‘Malik Sy’, Alfa Ibrahīm Sow, Chroniques et récits du Foūta Djalon [Paris, 1968], 55),Google Scholar and most agree on Hamet as Malik's grandfather and Dauda as his father. One tradition separately collected in Futa Jallon and in Bundu, however, inserts the name of Eli Bans (among others) into the patrilineage between Habīballāhi and Dauda Hamet. (Narration of Farba Sek, in Sow, , Chroniques et récits, 55Google Scholar from the Vieillard collection; Narration of Demba Sali Dieng of Koussan in Bundu, Curtin Collection, tape T7[21].) The reference is almost certainly to the cleric, Eli Bana, who ruled part of Saloum in the early sixteenth century and the number of intervening generations would make this reference chronologically possible. While the evidence is too slight to make possible a firm conclusion, it may be that this tradition points to a real or fictitious connexion between Malik Sy and a much earlier cleric who held political power.

²⁴ Brue, André, 9 12. 1716, Archives Nationales, Paris, Colonies, C⁶ 5.Google Scholar

²⁵ Stone, T. G., ‘The Journey of Cornelius Hodges in Senegambia’, Eng. Hist. Rev. xxxix (1924), 89–95.CrossRefGoogle Scholar

²⁶ Hull, Thomas, ‘Voyage to Bundo’ (unpublished MSS. from the library of the Duke of Buccleuch, 1735), 18;Google ScholarCurtin, P. D., Africa Remembered (Madison, 1967), 28–9.Google Scholar

²⁷ I have used only three of the several existing Bundu king-lists accompanied by reign-lengths because these three are shown to be most accurate when checked against other evidence. They are the traditions recorded by Rançon, in about 1890 ‘Bondou’, 480–3;Google Scholar those published by Kane, Abdoulaye, ‘Histoire et origine des familles du Fouta-Toro’, Annuaire de la comité d'études historique et scientifique de l'AOF, i (1916), 325–43, see p. 341;Google Scholar and the narration of Demba Simbalou Sock of Sénoudébou, considered to be the most authoritative traditionalist in Bundu in 1966 (Curtin Collection, tape P2). Rançon's own calculation of the date of Malik Sy's death is 1699.

²⁸ Adam, G., Légendes historiques du pays de Nioro (Paris, 1904), 55;Google Scholar Demba Simbalou Sock (Curtin Collection, tape P2); Kane, , ‘Families du Fouta Toro’, 341.Google Scholar

²⁹ Rançon, , ‘Bondou’, 473–7.Google Scholar

³⁰ See Rodney, W., ‘Jihad and Social Revolution in Futa Djalon in the Eighteenth Century’, J. Hist. Soc. Nigeria, iv (1968), 269–84, see pp. 274–;7.Google Scholar

³¹ Rançon, , ‘Bondou’, 472.Google Scholar

³² Saki Ndiaye, ‘Malik Sy’, Toumané Kante (Curtin Collection T1[1]). The presence of the Sisibe as students in Futa Jallon is also reported in the traditions of the Futa Jallon (Bayol, Jean, ‘Voyage en Sénégambie’, Revue maritime et coloniale, xciv, 441–73; xcv, 72–104, 256–81, 438–66; xcvi, 155–81, 492–559 [1887–1888], xcv, 448.)Google Scholar In the tradition reported by Bayol, however, Boubou Malik is apparently confused with his son, Maka Boubou, who was too young to have studied with Tierno Samba and had his closest associations with Futa Toro.

³³ Traditions from the Vieillard collection, printed in Sow, , Chroniques et récits, 45 and 63.Google Scholar Still another kind of link is possible. The ruling dynasties of Futa Jallon are traditionally traced back to Alfa Kilala, who had two sons. One of these, Nouhou, became the father of Karamoko Alfa. The name of the other is sometimes given as Malik Sy, or else simply as Malik, and he became the father of Ibrahima Son, founder of the Sorīya. This was not, of course, the Malik Sy of Bundu, since the yetoode or family name of the Fnta Jallon dynasties was Bari, not Sy. It is possible, however, that Malik Sy was a given name, honouring a famous man. See Bayol, Jean, ‘La France au Fouta-Djalon’, Revue des deux-mondes, liv (1882), 902–32, see p. 916;Google ScholarMarty, P., L'Islam en Guinée (Paris, 1921), 8–9.Google Scholar

³⁴ Trimingham, , History of Islam in West Africa, 155–9, 167 and 171.Google Scholar

³⁵ Trimingham, , History of Islam in West Africa, 151–54.Google Scholar

³⁶ Gibb, H. A. R., Mohammedanism (London, 1949), 165–70.Google Scholar

³⁷ Rodney, Walter, A History of the Upper Guinea Coast 1545–1800 (Oxford, 1970), 236–9.Google Scholar

³⁸ Curtin, P. D., The Atlantic Slave Trade: A Census (Madison, 1969), 221, 224.Google Scholar

³⁹ See, for example, Labat, J. B., Nouvelle relation de l'Afrique occidentale, 4 vols. (Paris, 1728), iii: 262, 336;Google ScholarCultru, , Voyage du Sieur de la Courbe, p. lvi;Google ScholarSaint-Robert, , Despatch of 28 03. 1721, Archives nationales, Paris, Colonies C⁶ 6.Google Scholar