The jihâd of cUthmân dan Fodiye in Hausaland and the rise of the Sokoto federation of emirates has been studied from the point of view of the background of jihâd movements throughout 18th and 19th century West Africa,1 the historical succession of events in the Sokoto jihâd itself,2 the Islamic ideology underlying these jihâds,3 and the social grievances particularly of the Fulani herdsmen against the Hausa rulers.4  Likewise the story of Oyo‘s succumbing to a jihâd waged from Ilorin has been told and the reasons debated.5

To my knowledge, however, the link between Hausaland and Oyo has not yet been explored from the standpoint of the economic dynamics of West African politics.  The links were there in the 18th century before the Sokoto jihâd.  The Sokoto state, I argue, arose in response to the Oyo state in a contest over control of the Atlantic trade.  Oyo lost, but other factors prevented Sokoto from pursuing its victory to fruition.

I have no new historical facts to present, but utilize the monumental works of D. Murray Last6 and Robin Law7, along with many less known references. It is only a question of gathering and utilizing the facts in a new synthesis.  It will be necessary to summarize Oyo and Sokoto history as already established, along with the interpretations of it that have been advanced. In the course of doing so, and in the last section, I will present my own interpretation.

The political economy of Oyo until the jihâd8

      The expansion of Oyo

Whatever may be said of Ife as the “cradle of the Yoruba” or the greater early importance of other Yoruba states, old Oyo was the paramount Yoruba power from the 17th to the early 19th century.  The capital, called Katunga in the time of Clapperton and otherwise known as Oyo Ile, was dangerously exposed to Borgu and Nupe enemies.  In spite of its destruction in the 16th century it was reoccupied because of its commercial advantages which outweighed the risk its site presented.9  It lay within the savanna zone where horses could be kept for easy communication with Nupe, Borgu and other Sudanic kingdoms to the north, with the forest kingdoms to the southeast and the coastal kingdoms of Ardrah and Whydah which were accessible through a corridor of savanna reaching to the sea.

Horses obtained from Borno gave the Alafins the military capability to secure their borders against raids from Borgu and Nupe and to consolidate and expand their territory towards the south.  The Oyo heartland supplied foodstuffs, notably grains and cattle in the northern parts and yams and palm oil in the southern parts, iron tools and weapons, pottery, and cotton cloth.10  Such tribute enabled the Alafins to maintain a staff and an army; the populace, in return, enjoyed security and the prosperity generated by a wider market.

In the first half of the 17th century the Alafins tried to expand into Ijesa, Ikiti and other Yoruba kingdoms to the southeast.  They were unsuccessful, because the forest terrain was unsuitable for cavalry, and in the rains the tsetse fly was especially dangerous for the horses.  The motive for attempting to subjugate the forest states must have been to exploit their agricultural products which were different from those of the savanna.  Ijesa kola nuts were particularly valuable for resale to the north in competition with the Gonja supply, although Yoruba production of the preferred nitida variety of kola was not very important until colonial times.11

In the second half of the 17th century Oyo began to expand to the southwest, all the way to the coast.  The savanna conditions permitted the employment of cavalry, but the tsetse fly and the problem of providing fodder for the horses prevented Oyo from permanently conquering or occupying widespread territory in the south.12  Yet some colonization of the coastal routes did take place, first in the Anago area, west of the river Yewa between Badagry and Porto Novo, at the beginning of the 18th century, then in the mid-18th century in the north of Egbado country, and finally in southern Egbado, including the towns of Ijanna and Ilaro.13

Oyo’s first dealings with Atlantic trade were through the independent kingdoms of Allada, Porto Novo and Hueda (= Whydah).  Oyo raided Allada several times in the late 17th century,14 presumably to secure better conditions for its own trading activity.   Trade continued to be hampered, however, because Allada and Whydah were at war with each other from 1712 to 1722,15 and in 1717-18 they declared royal monopolies over the slave trade in their respective ports.  This action provoked Dahomey to annex Allada in 1724 and Whydah in 1727.16  Dahomian control of these port kingdoms was a threat to  Oyo’s interests, and Oyo responded in two ways.  First, it campaigned against Dahomey from 1727 to 1730 until Dahomey agreed to become a tributary state, keeping the whole of Whydah and most of Allada except Ajase (a Yoruba name for Porto Novo). 17  Secondly, because of the recurring instability of Whydah and Allada, it developed easterly trade routes leading to Badagry and Lagos through gbado country.

      Motives for expansion to the Atlantic

If we look for the motive of Oyo‘s intervention in or occupation of these coastal areas we find different opinions.  For P. Morton-Williams, the reason is simple; it was to “participate in the coastal trade in slaves with the Europeans”.18  The same opinion was expressed by R. Smith: “The Oyo had become by the late seventeenth century exporters on a large scale of slaves for the Atlantic trade.”19  In spite of the evidence which these authors offer, I.A. Akinjogbin rejects this view altogether.  He says, “While Oyo would have derived certain economic advantages from its conquests, it does not appear to have benefited much from the slave trade with the Europeans in the 17th century... For Oyo at that date the coast was really a backwater.”20  Akinjogbin does, however, admit that Oyo was heavily engaged in slave trading in the late 18th century.21   He says: “The Oyo derived their wealth largely from the export of slaves, a trade which, since the beginning of the reign of Abidun, seems to have been the main prop of Oyo economy.”22

The more recent work of Robin law argues more closely the case that Oyo’s expansion to the Atlantic was motivated by the slave trade.  The witness of Dapper in the 1640s, as well as documents of 1723, 1727 and 1753, refer to the kingdom of “Olukumi” (or one of its variants), meaning a Yoruba kingdom, likely Oyo, as a supplier of the considerable volume of slaves sold through Allada.23  This evidence confirms the theory that Oyo had little local economic advantage to gain by annexing or exerting influence over neighbouring states towards the coast, since the products of these states were not significantly different from those of the Oyo heartland (apart from salt from evaporated sea water), all these states being in a savanna region.  The only advantage of any consequence from this reach to the sea was the booming Atlantic trade.

This argument is bolstered by further documentation and activity which has no plausible explanation apart from  Oyo’s interest in the slave trade.   Oyo’s intervention against Dahomey in 1726 to 1730 made good sense as a way of reopening Oyo’s access to the sea which Dahomey’s annexation of Allada and Whydah had cut off. The resumption of  Oyo’s supply of slaves to Whydah is indicated by a 1754 document blaming a slump in trade to political troubles within Oyo. 24   Other documents refer to the slave trade in the 1770s.25  In the 1780s the Dahomians closed their ports again to Oyo traders, forcing them to move to the newer ports which had arisen in the west: Epe (not the Nigerian Epe, but present-day Cotonou) in the 1720s, Badagry in 1736, Porto Novo in the 1750s, and Lagos in the 1760s. Between 1760 and 1770 Portuguese ships stopping at Principe include 13 from Epe, 12 from Porto Novo, 14 from Badagry and 2 from Lagos.26  At first this trade passed through the kingdom of Wem.  Then, as we have seen, to have more direct control of the routes to the new ports Oyo colonized southern gbado in the 1770s and 1780s.27  Dahomian attacks on Epe in 1782, Badagry in 1784, and Wem in 1786 only served to make Whydah languish and to drive the bulk of the slave trade to Porto Novo and Lagos. Lagos became the most important port in the 1790s, with supply routes coming by lagoon from Porto Novo and eventually Oyo, or from Ikosi and Ikorodu and eventually the Ijbu kingdom.  The Ijbu slaves were obtained either by condemning local criminals or by purchase from Oyo.

The slaves sold by Oyo were from four different sources: 1) native criminals, who were not so numerous, 2) war captives, who were the majority,28 3) slaves given in tribute by vassal states, such as Dahomey, who were not so many, and 4) slaves bought from Nupe, Borgu and even the Hausa states and Borno, especially in the late 18th century. Adams (1786-1800) refers to the large numbers of Hausa slaves for sale at Porto Novo.29

Oyo paid for slaves and horses from the north with European goods obtained from the coast, particularly cowry shells, but also iron bars, cloth, earthenware, beads, rum and guns.30  Guns, however, were not of much use in Oyo which relied mainly on cavalry and did not know how to use firearms effectively.31  On the coast slaves were almost the only commodity the European traders were interested in, although Oyo may also have sold a little cloth, ivory and palm oil.32

The main beneficiaries of  Oyo’s economic system were the Alafin, the trading magnates and their hangers-on. The majority farmer people benefited from the empire mainly because its peace and security permitted local markets to prosper.  They were not heavily taxed, since the Alafin got food supplies for his retainers and the army from farms run by slaves.  Ordinary people also could obtain imported iron for tools and cloth in exchange for kola and palm oil.  Some found their fortune by joining the army or becoming retainers of big traders.

Any conclusion about Oyo’s involvement in the Atlantic slave trade rests not merely on direct evidence but also on the principle explained earlier that economic interest is at the base of political expansion.  Simple ethnic togetherness or the glory and prestige of the nation or even the spread of a religion are not enough.  The export of slaves to the coast is the most plausible explanation of  Oyo’s military and political action in the coastal region, even when we have no record of the deliberations and planning behind this action.33

As a powerful middleman between the north and the Atlantic, Oyo had to maintain an efficient trading complex which we could name “Oyo Incorporated”. The centralization that this required led, from the beginning of the 18th century, to a greater concentration of power in the hands of the Alafin and a curtailing of the power of the metropolitan nobles or Oyo Mesi. 34  The inbuilt human weakness of the whole system appeared in a power struggle between these two parties.  The issue was one of free business interest versus government restrictions, and not of military expansionism versus trading interest, as Akinjogbin maintains,35 because expansion of the empire could serve its economic interests, both by widening the trading zone and by capturing slaves which could be sold.  The question was how would the profits of the trading operations be shared.  Would most go to the Alafin, whose central authority ensured the unity of the empire and the possibility of long-distance trace, or would the provincial governors take a larger share, and thereby weaken the imperial trading network by introducing their own tolls and restrictions on the traders?

      The fall of Oyo

The struggle between the Alafins and the Oyo Mesi came to a head in 1754 when the Basôrun Gaha (Gaa) grabbed effective power and took over administration of the empire.  The Alafin Abidun dislodged him only in 1774 by enlisting the support of the provincial towns.  All was not yet well, however, and Nupe and Borgu took advantage of the disturbed situation to launch several successful attacks on Oyo in the 1780s.  When the Alafin Abidun died in 1789, according to the chronology accepted by Akinjogbin and Law,36 his children were driven out of Oyo and the succession went to the weak Awole, whom the Oyo Mesi thought they could manipulate.  Instead, Awole proceeded to quarrel not only with the Oyo Mesi but also with the provincial governors, notably with their leader, Afonja of Ilorin.

Awole‘s plot to get rid of Afonja by ordering him to attack the impregnable town of Iwere backfired when Afonja marched on Oyo instead.  Awole was therefore compelled to commit ritual suicide; this was in 1796, according to the afore-mentioned chronology. Another Alafin named Adebo was installed, but lasted only 120 days; another named Maku lasted three months. The military then ruled for five years, that is, to 1802, when the Alafin Majotu took power.  During all this time Afonja grew stronger in Ilorin and the surrounding area.  Oyo and Ilorin held each other at bay until 1817, when Ilorin launched a successful Muslim rebellion against Oyo.  Internal dissension and external aggression were to bring Oyo down. 37

Afonja was not a Muslim himself, but found his opportunity to break the stalemate with Oyo by exploiting the sentiments of the considerable Muslim community living within Oyo territory.  These included: 1) Hausa, Fulani and Arab clerics and traders from the north who had settlements all along the trade routes to the coast, 2) Fulani herdsmen, who at least from ethnic solidarity would support a jihâd led by Fulani clerics, 3) an important group of Yoruba Muslim traders who worked hand in hand with the northerners, and 4) a large slave population of Muslims from the north.

Afonja invited a Fulani cleric named Sâlih, also called Alimi, to rally these Muslim groups against Oyo.  Sâlih proclaimed a jihâd, which greatly disrupted normal life all the way to the Oyo capital.  The Alafin responded by executing many of the Muslims in Oyo Ile, but this was only a momentary reprisal; Muslim traders and clerics were soon freely moving again within Oyo Ile and its dependencies.  Fulani bands, however, continued to raid the countryside, creating many abandoned towns and making life insecure and intolerable in the Oyo heartland.

Meanwhile Afonja ran into difficulty with his Muslim allies. Sâlih apparently had died when his son cAbdassalâm declared his allegiance to Sokoto and was in return recognized as “Emir of Yoruba” in 1823 or 1824.  Either before or after this event, Afonja became alarmed and tried to get the Fulani and Hausa to leave Ilorin and settle elsewhere.  They then rose up and killed him together with his Yoruba Muslim assistant lagberu.  The Fulani Muslims were thenceforward in charge of Ilorin.

With this successful challenge to Oyo so close to its capital, the vassal kingdoms and provincial towns gradually broke away. Consequently the long-distance trade routes from the Atlantic to the north, on which the economy of the empire depended, were gradually cut off.   Oyo’s export of slaves dropped dramatically, and its inability to import horses crippled its cavalry force.  International trade began to bypass Oyo, moving instead through Borgu and Dahomey to Whydah, or through Nupe, Ilorin and Ijbu to Lagos.  Oyo continued to suffer from internal dissension and from Fulani attacks until 1833, when Ilorin forces conquered the capital and forced the inhabitants to profess Islam.  Ilorin then proceeded to occupy the whole heartland of the Oyo kingdom.  The now tributary Oyo leadership staged one more effort to strike back at Ilorin with the help of Borgu.  Gwandu, headquarters of the Western part of the Sokoto caliphate, sent Nupe troops for the final assault on Oyo in 1836.38  The Borgu leaders were killed and the Alafin Oluewu was captured and executed.  Oyo Ile was then abandoned and the new Alafin Atiba established himself at Ago Oja, which was renamed the new Oyo. Nominally his rule extended over nearby towns such as Ibadan, which was founded by Oyo refugees, but effectively the Oyo empire had come to an end.

Ilorin attempted to extend its conquests over the whole of Yorubaland, but was checked by Ibadan at Osogbo in 1838.  Later in the 19th century its ambition came closer to realization, but was checked by the British.  Muslim Ilorin owed allegiance to Sokoto.  We must now review Sokoto history and the motives for the jihâd of cUthmân dan Fodiye.

In particular, we must ask why the jihâd continued unrelentingly towards the sea, while immediate northern neighbours, such as Argungu and Sabon Birni, were left undisturbed.

The Sokoto jihâd: a historical résumé 39

cUthmân dan Fodiye was born at Maratta in 1754 of a Fulani clan specializing in Islamic learning whose origins go back to Futa Toro.  His childhood, at Degel in Gobir territory north of present day Sokoto, included an elementary Islamic education under his father. In 1775 cUthmân went to Agadez to study under Shaykh Jibrîl, a Tuareg who had ambitions of carrying out a jihâd among his own people.   Jibrîl went on pilgrimage to Mecca and wanted cUthmân to accompany him, but cUthmân could not get his father’s permission.  Instead, he returned home and formed a jamâca, an embryonic group of followers and supporters, and went on preaching tours around Gobir, Zamfara and Kebbi.  In 1788 the Sarkin Gobir, Bawa Jan Gwarzo, invited him to the cId al-cadhâ prayer assembly where cUthmân’s followers outnumbered those of the Sarki.  Overawed, the Sarki granted him his five demands: 1) that he could preach without restriction, 2) that no one should be prevented from following him, 3) that Muslims wearing the turban should be treated with respect, 4) that Muslim prisoners would be freed, and 5) that taxes should be reduced.

Bawa Jan Gwarzo’s successor, Nafata (1801-3), opposed cUthmân, forbidding anyone but cUthmân himself to preach, forbidding conversion of those whose fathers were not Muslims, and forbidding turbans on men and veils on women.  Nafata’s son Yunfa (1803-8) began his reign friendly to cUthmân, but then tried unsuccessfully to assassinate him at Alkalawa.  CUthmân found a chance to retaliate when Yunfa sent out a raiding party to capture a wealthy Gobir man, cAbdassalâm, who had fled to Gunbara (near Jega) in the reign of his father.  Yunfa’s men failed to capture cAbdassalâm but they raided his village and carried off may prisoners.  cUthmân waylaid the raiding party on their return and freed the prisoners.  Yunfa thereupon ordered cUthmân to leave Gobir territory.

On 21 February 1804, with the severance of ties implied in a religious hijra, cUthmân moved westward to Gudu.  His followers, against the wishes of Yunfa, joined him and in May made formal submission (bayca) to him as their imâm.  Afraid of cUthmân’s build up of strength, Yunfa sent his army against him.  Although outnumbered and short of provisions, cUthmân’s men had the advantage of being in a wooded area next to Lake Kwotto.  Because of this position and their better morale they defeated the Gobir army.  cUthmân then moved camp to Magabshi (near Yabo) in Kebbi territory.  Many more Fulani joined him after his victory, but their raids for food alienated many of the Hausa peasants.

In October 1804 and January 1805 the Fulani made unsuccessful attacks on the Gobir capital Alkalawa.   Then cUthmân moved his camp to Sabon Gari in Zamfara territory while his younger brother cAbdallâh attacked and captured Birnin Kebbi, the Kebbi capital, causing its Sarki, Muhammad Hodi, to flee and eventually set up headquarters at Argungu.  cUthmân’s son Muammad Bello began raiding Gobir and Zamfara villages for food, thereby making it dangerous for the Fulani to remain at Sabon Gari.  In July 1805 they moved to Gwandu and fortified the town as a permanent base.  In November 1805 a combined army of the Gobir, Tuareg, Zamfara and Kebbi marched on Gwandu. The Fulani went out to meet them and were severely defeated at Alwasa.  The enemy pursued the Fulani to Gwandu and might easily have taken the town and ended the jihâd, but a delay on their part permitted the Fulani to rally their forces and on the rougher terrain near the town to rout the enemy completely.  The Fulani were thereafter in the superior position.

In 1807 Muhammad Bello took Kiawa, the capital of Zamfara, ending the independence of that state.  On 3 October 1808 he took Alkalawa, but the Gobir Sarki, driven from his capital, maintained control of the northern part of Gobir, now in Niger Republic, and throughout the 19th century Gobir harassed Sokoto with frequent raids on its territory. The town of Sokoto was founded in 1809 when Muammad Bello chose the site as his headquarters.  In 1812 cUthmân divided the administration of the empire between his son Muammad Bello for the eastern part, with Sokoto as his capital, and his brother cAbdallâh for the west, with his capital first at Bodinga and later Gwandu.  In 1815 cUthmân moved to Sokoto, where he died on 20 April 1817.

When cUthmân began his jihâd he distributed flags to Fulani leaders throughout Hausaland and beyond.  In 1805 they overthrew the Hausa rulers of Daura, Kano and Katsina. At the end of 1808 Zaria fell and its Hausa rulers fled to Abuja (now called Suleja).  The Fulani also came to power in areas which never had Muslim rulers. Adama led the jihâd in the territory named Adamawa after him, most of which is now in Cameroon.  Hamman Ruwa took over Muri; Yakubu, a former student of cUthmân, established himself in Bauchi, and Buba Yero in Gombe. Other Fulani leaders established themselves in Keffi, Nasarawa, Lafia, Jema’a and Wase.  The adventurer Umaru Nagwamatse founded Kontagora, in whose surrounding territory he and his successors conducted intensive slave raiding.

Nupe, like the Hausa states, had Muslim rulers before the jihâd of cUthmân dan Fodiye.  The first was the Etsu Jibrilu (c1746-59), but shortly after his death a dynastic dispute led to a division of the kingdom.  In 1805 a Fulani cleric named Dendo conspired with Majiya, Etsu of Raba and the west, to defeat Jimada, the Etsu of Gbara and eastern Nupeland. Once Majiya secured power over a reunited Nupeland he began to fear Dendo’s influence and expelled him from the kingdom.  Dendo then joined Idrisu, son of the defeated Jimada, and together they deposed Majiya.  Dendo then installed Idrisu as a puppet king in Adama Lelu (near Eggan), while he himself held the real power in Raba. Idrisu revolted in 1830 and was killed, leaving the Fulani in complete control of Nupe.40 Besides Nupe, the Fulani set up emirates in Agai, Lapai, Pategi and Lafiagi.41

Lastly, as we have seen, a Muslim-Fulani coup was staged in Ilorin, and the heartland of the Oyo empire was incorporated into the Sokoto caliphate.

All these emirates were largely independent, so that we cannot speak of a Sokoto empire, but they recognized the primacy of Sokoto, and Sokoto or Gwandu had a hand in many of their internal affairs.42

Theoretical interpretations

Various models have been proposed to explain the genesis and disintegration of pre-colonial African states in general, or the causes of the Sokoto jihâd in particular.  Some of these models are mutually opposed, while others can be complementary, as can be seen by examining them one by one:

      The religious model

For the rise of any reformed Islamic state the religious model gets predominant attention.  It was the only model presented by the Sokoto jihâd leaders themselves and is the principal explanation offered by some modern writers.43  The religious model, which sees Islamic religious principles as the basic dynamic for the formation of an Islamic state, need not detain us here, since it is described in detail in many general works.44  Suffice it to say that a Muslim community ideally is ruled by the Sharîca, God’s law as revealed in the Qur’ân and Sunna, which covers all aspects of life. In a society where this does not prevail Muslims have the choice of either hijra, moving to another land where Sharîca is the law, or jihâd, struggle by all realistic means, even shedding blood if necessary, to institute the rule of Sharîca.  This divine imperative was the explicit compelling motive behind the jihâd of cUthmân dan Fodiye and his companions as far away as Adamawa and Ilorin. Their fight was mainly against compromising Muslim rulers,45 and once these were overthrown there was not much urgency about imposing Islam on non-Muslim Hausa subjects, but only an annual dry-season jihâd against unsubjugated pagan peoples outside Hausaland.

The Sokoto jihâd leaders were not blind religious fanatics, but were well acquainted with classical works about the constitution of an Islamic state.  They expressed their ideas in hundreds of written works, both long documented treatises and popular tracts and letters.  We can agree with H.F.C. Smith and T. Hodgkin that the jihâd leaders were extremely well read in Islamic literature,46 but Trimingham’s opinion that these writings “show no originality” is also valid, in spite of Hodgkin’s criticism, as far as general Islamic thought is concerned.47  The jihâd leaders wrote in the car al-jumûd, the age when Islamic thought was fixed and frozen.  It could be rephrased and applied to new situations, but not developed in any real way.48   Nevertheless it is to the jihâd leaders’ credit that they reflected carefully on Islamic theory and applied it to the situation of Hausaland in their time.

We can accept the religious model for the jihâd as basically correct, at least during the lifetime of cUthmân dan Fodiye.  Nevertheless this model does not exclude other models based on economic or social factors, but rather presupposes them, since Islam purports to be not only the right way to God but also the right formula for economic and social success.  The all-inclusiveness of Islam makes Muslims expect that its establishment will bring a better and more prosperous society and a greater enjoyment of the goods of this life as well as those of the next.  We may draw a comparison with the hajj, where along with its spiritual benefits, a pilgrim can also legitimately aspire to come home richer, either by trading on the way or by God’s special favour.

The question we need to ask is what economic or social advantages could be realized by cUthmân dan Fodiye’s jihâd.  The following models are some hypotheses that have been advanced in an attempt to provide an answer.49

      The ethnic model

One of the earliest socio-economic analysis of the jihâd made it an ethnic struggle of newly arrived nomadic and virile Fulani against the softened sedentary Hausa.  The jihâd was simply a question of the stronger taking over from the weaker to exploit the agricultural commodities the Hausa had to offer.  As alien herdsmen, the Fulani would naturally take any form of oppression as an attack on themselves as a community.  Even before the jihâd, separate Fulani risings took place in the future Adamawa with the appearance of an ethnic revolt.50

This view has the merit of recognizing that strong Fulani solidarity made possible coordinated and successful jihâd action over a very wide area, but it fails to take into account the participation of Hausa peasantry and also some settled Fulani in the jihâd.  Also the Zabarma people supported the jihâd to settle scores with the Gobir.51

We must also recognize the power of religious solidarity that bound jamâcas together over a very wide area.  Students traveled far to seek out teachers of repute, and the bonds of loyalty thus established made possible the political and military cooperation such as the jihâd of cUthmân dan Fodiye evinced.

      Marxist models

A Marxist theory proposed by the Russian Africanist D. Olderogge sees the jihâd as a rising of oppressed Fulani nomads, later joined by Hausa peasantry, to overthrow a feudal Hausa aristocracy.  cUthmân dan Fodiye used the discontent of the stable social stratum of farmers and pastoralists to put himself and his collaborators in the place of the Hausa aristocratic superstructure, continuing the same feudal system.52

This theory was attacked by Jack Goody, who points out that real feudalism hardly existed in Africa, outside Ethiopia, because of the abundant land and low technology.  The aristocracy were not landlords, but lords of people, and their power lay in the military means to protect and tax production and trade in their states and to wage war for defence and for booty.53  Moreover, as Marilyn Waldman points out,54 the discontented groups who supported the jihâd were not merely the Fulani pastoralists and Hausa peasants, but also many clerics, both poor ones and some court appointees, and many of the Fulanin gida, the settled Fulani businessmen within the Hausa cities. The clerics had a professional religious interest in the jihâd, while the businessmen saw fantastic economic opportunities in the success of the jihâd, as will be explained later.

M. Godelier55 and J. Suret-Canale56 proposed a revised Marxist model, taking as a starting point Marx’s “Asiatic model” of feudal society, whereby the despotic ruling class exact tribute and conscript some peasant labour for the execution of great public works, while leaving undisturbed the basic structure of the patriarchal village society with its subsistent economy.  Eliminating the features of great works and despotism, these authors transfer Marx’s model to African societies where a centralized state coexists with a village patriarchal society having a communal economy.  The latter institution is more stable and it remains should one aristocratic structure replace another, as was the case with the Islamic jihâds of the 18th and 19th century West Africa.57

This neo-Marxist model was criticized by Catherine Coquery-Vidrovitch,58 who pointed out that by eliminating from the Asiatic model the feature of great public works, such as monumental buildings or complex irrigation works, the state superstructure is deprived of its dynamic and reason for existing.  The ruling class may well draw tribute from the peasantry, but this is not the force behind the existence and organization of African empires.  For her, trade takes the place of great public works in a specifically African model of political economy.

      The international trade model

This model, such as proposed by Coquery-Vidrovitch, combines a tribute paying communal patriarchal economy with a system of long-distance trade controlled by a royal or bourgeois aristocracy.  Trade, in the broad sense of selling for profit what was procured cheaply, either by peaceful trade or military raids, was once of the motors of history in tropical Africa.  It was at the same time the most dynamic but also the least stable element in society, because it was easily disrupted by wars, blockades or changes in demand.

The history of the medieval West African empires confirms the theory that long-distance trade was the fundamental dynamic in the formation of large states in this part of the world.59  Whether the exchange was of salt for gold or between any other commodities, the need and opportunity was present for a powerful government to control the routes, eliminating both custom barriers along the way and the threat of highway robbers. Whether the Sudanese rulers engaged directly in trade or, as was more common, simply taxed private trade, their role was that of powerful middlemen, and this made them rich.  They did not usually control production, and when they tried to do so it led to disaster, as when the kings of Ghana took over the Bambuk gold fields and later the salt mines of Awdaghust.  As a general pattern, then, any expansion of political and military power took place to serve economic interests.  These included security for people’s lives and their production and exchange of goods, as well as ensuring a fair distribution of profits. This was an endemic source of conflict, but ideally every party to the trading enterprise had his proportionate share.

So far this model is quite satisfactory, but it requires some further refinement.60  Both Coquery-Vidrovitch and the Marxist theorists supposed that the patriarchal peasant level of African society had a “subsistent economy”, one which produced all and only its basic needs and did not require or engage in the exchange of goods.  J.E. Flint 61 and A.G. Hopkins62 have pointed out how fallacious this notion is and how diversified the local economy generally was, in that people specialized in many different industries besides food production, and this was the reason why local markets thrived, quite apart from goods exchanged in long-distance trade.63

Another modification of this model needs to be made to take account of the complexity of the slave trade, which was becoming increasingly important and formed the basis of the economy of some states.  We must differentiate between export and domestic use of slaves. In the case of 18th century Futa Jalon under the rule of the Fulani almamis, the aristocratic economy was based almost exclusively on exporting slaves to the Atlantic, while only a few surplus slaves were kept for rice production.64  In the Sokoto federation, however, where slave raiding was equally important, slaves were used in much greater numbers to augment the population of the emirates and thereby increase local production and revenue for the aristocracy.65  Export of slaves, nevertheless, was also important, both across the Sahara throughout the 19th century and to the Atlantic up to the middle of the 19th century when the Atlantic slave market ceased to function.66  The importance of the Atlantic trade to Sokoto will be clarified in the last section of this article.

      A composite model

From the preceding discussion of models we can describe a composite model embracing several dynamics, all of which, it could be argued, played an important part in the rise of the Sokoto federation.  The absence or weakening of any of these dynamics would conversely entail a decline or weakening of Sokoto power.

The first dynamic is the religious one.  To be effective any ideology must be religious in the sense that it relates man to something of ultimate value which will command his entire allegiance; it must also provide the basis for a community loyalty. Islam clearly does this.  African traditional religion does also to a certain extent, especially when it sacralizes authority, but it is limited to a narrow, usually village, society, whereas Islamic monotheism is better adapted to an international social order.67

The second dynamic is social, and that consists in an equitable distribution of the national cake.  If an ethnic group or economic class, such as the nomadic Fulani or Hausa peasantry, becomes disaffected at seeing the privileges of the few, they will readily contribute to a change of regime.  but to succeed they must have good leadership.  Sometimes a warlord will suffice, but intellectual and religious leaders, such as led the Sokoto jihâd, are more effective in the long run.

The third dynamic is economic.  The ruler of a small state can aspire to the wealth of tribute or tax from the surplus of the local economy in return for guaranteeing order and security.  Avoiding the pitfall of the concept of “subsistent economy”, we realize that this tribute can take as many forms as the variety of local production.  At a next stage, a ruler may desire to expand his domains in order to receive in tax some other product not available locally. Such expansion is equivalent to the creation of a wider common market.  In another development trading arrangements may be made with outlying areas without controlling their production of goods.  In this case the central state may act simply as a middleman, transmitting goods for re-export.  Many variations are possible within this model; so we should not see a sharp distinction between patriarchal local economies and those of large empires. The two are rather extreme points of a continuum, with the possibility of many intermediate forms in between. In any case, wherever an expansion of borders takes place we must ask what is the economic advantage of it. Empires do not come into being simply for prestige, but have something to gain by expanding.  Otherwise they stay within their inherited boundaries or even shrink.

In the case of Sokoto, the first two dynamics have been studied and established.  The third dynamic was hinted at by Guy Nicolas, who pointed to the interest the business and caravan men of Gobir in the success of cUthmân dan Fodiye’s jihâd.68  R. Horton comments that “there is at least a suspicion that, in two of the jihâds commonly regarded as pastoralist-dominated, those of Malik Sy and Usuman dan Fodio, mercantile interests may also have been of some importance.”69  To explore how this third dynamic applies to the Sokoto jihâd we must look at how Sokoto stood to benefit from extending the jihâd in  Oyo’s direction.

Sokoto’s interest in the fall of Oyo

Well before the jihâd, Hausaland was tied into the slave trading network of the Oyo empire.  Robin Law observes that enslavement of Oyo citizens did not become common until after 1810, with a consequent aggravated instability of the kingdom. Before that time slaves were obtained from outside, with a great expansion in the 1780s of trade in slaves purchased from the north.  He concludes that “any political consequences should be sought in Hausaland rather than in Oyo. 70

There is evidence for considerable slave raiding within Hausaland in the late 18th and early 19th centuries.  Besides the testimony of Adams mentioned above, cUthmân dan Fodiye complains of the enslavement of Muslims in his Masâ’il muhimma, written in 1802, two years before the jihâd, and again in his Fulani poem Tabban hakikan.  His brother cAbdallâh makes a similar complaint in his Tazyîn al-waraqât.71poem” as found in Wakoin Hausa (Zaria, 1957).  In the Zaria edition, however, the title is Tabban hakikan.  This Hausa translation of a Fulani original is adapting an Arabic phrase.  Which one?  Hiskett corrects the text to Tabbat hakîkâ, evidently to represent thabbat haqîqa (Truth is established, Hiskett’s “forsooth”, p. 171, not Tabbat haqîqa “Truth has perished”).  Yet with no change in the Hausa spelling, Tabban hakikan, the phrase could be taken to represent tabcan, haqîqan, meaning “naturally, truly”.  Muhammad Bello in 1812 wrote explicitly in his Infâq al-maysûr that “the people of that land (Yoruba) import slaves from our land and sell them to the aforementioned Christians (on the coast).  I mention this problem so that you may not sell a Muslim slave to anyone who would bring him there.”72  Further evidence that the supply routes of the Atlantic market went right up to the north is the large numbers of Muslim slaves, both Hausa and Yoruba, in Brazil, many of whom later returned to Lagos and other places.73

Before the jihâd slaves were readily available because of the constant wars going on between the small Hausa states.  Katsina was at war with Gobir from 1750 up to the time of the jihâd, and Gobir was constantly eating away the territory of Zamfara during the same period.74  Kebbi suffered attacks from Gobir, Zamfara and the Tuareg of Asben.75  Also during most of the 18th century Kano was caught between attacks from Borno and Gobir.76  Zaria was more at peace but, like Katsina, at the price of having to pay a periodic tribute of slaves to Borno.77  These wars naturally yielded good numbers of captives, many of them Muslims.  Although many of these were kept locally and others sold to Arabs for export across the Sahara, many also were sold to dealers exporting them to the Atlantic coast.

The wars among the Hausa states, while contributing to the slave market, in reality prevented the development of this market to any considerable extent, because the fighting made the long-distance trade routes unsafe and prevented the tapping of richer slaving grounds further off, especially in Bauchi and Adamawa, where pagans could be enslaved without any religious scruple.  Any slave dealer could see the advantage or necessity of uniting Hausaland and the surrounding areas into a single commonwealth, so that traders and slavers could move freely.  We can assume that these traders would be prepared to support any jihâd that promised to effect the desired political unification. 78  How explicitly conscious the merchants were of the advantages the jihâd offered and how unified and organized was their support for the jihâd could be further researched.  To be sure, they operated more by pragmatic instinct and business sense rather than by formal study of marketing opportunities and planned strategy of operation.  Nevertheless, they had wide contacts which could keep them well informed about the commercial and political vicissitudes of the whole Nigerian scene under question.  They were possibly also aware of the 18th century jihâd in Futa Jalon and the lucrative trade in slaves the victorious Fulani secured as a result.79  The chances of a similar bonanza in Hausaland may not have been guaranteed, but the traders could scent a good opportunity in supporting the jihâd which was worth risking.

The jihâd of cUthmân dan Fodiye answered perfectly the aspirations of the slave dealers.   Except for occasional rebellions which the Sokoto authorities had to repress, security for trade movement was assured throughout the emirates.  Raids were systematically carried out against the “Middle Belt” pagan areas which provided all the centres of Fulani rule with a rich supply of slaves.  What was done with these slaves?

It seems the bulk of the slaves were absorbed locally, either in the cities for crafts or military service or in slave farm villages, increasing the population and prosperity of Hausaland.80  These slaves adopted the Hausa language and the Islamic religion, and are estimated to have constituted the majority of the population of Sokoto and Kano.  Exact statistics are not available, but a rough estimate of this high number is furnished by the literature written by the Fulani of the period, testimonies of European travellers, oral tradition and some studies of the city populations made at the beginning of the British period.

Only newly captured slaves or refractory old ones were normally sold for export.  Yet the number exported was considerable enough for slaves to become a standard currency in Hausaland and the surrounding areas, replacing cowries for any larger transactions.81  Many slaves were exported through the Atlantic market.  No doubt the profits and value of this trade in the late 18th century were well known in commercial circles of the north, since northern Muslim traders constantly frequented the routes to the coast.  The same traders could report the advantage of securing control of these trade routes and integrating them into the Sokoto caliphate once this was established.  Thus, in spite of their proximity, Sokoto and Gwandu made no serious attempt to conquer Argungu or the remnants of Gobir north of Illela.82  On the contrary, all efforts were expended for a thrust to the sea, by a Muslim-Fulani take-over of Nupe and Oyo.  Successful coups were engineered in both these states, as we have seen, mainly through the agency of local Muslim forces, but also with some assistance to Ilorin from Gwandu, the headquarters of the western part of the caliphate. The jihâd thrust to the south certainly catered to the demand for a workforce in the Caliphate heartland. But Samuel Johnson’s phrase, “to dip the Qur’ân into the sea”,83 summarized the attraction of the Atlantic market.

The mercantile party of the Sokoto confederation had every interest in pushing the conquest to the coast, since from the beginning of the 19th century until 1830 the Atlantic slave business actually increased.  This was the result of increased demand from Brazil, Cuba and Southern United States, in spite of the stop of the slave trade by the British in 1810 and the French in 1815.84  The Atlantic slave trade dipped sharply after 1840, until it was completely stopped by 1860.  So the thrust of Sokoto to the sea was frustrated not merely by the intervening independent states and later British interference, but more importantly by the evaporation of the major economic motive.  Expansion to the coast was also blunted by the rise of independent states in the former southern dependencies of Oyo, such as Ibadan, Abokuta and the coastal city-states, so that the southern end of the trade routes was still subject to the exactions of middlemen and periodic instability and insecurity.

Slaves were also exported across the Sahara.  Once the jihâd was consolidated, the Sokoto confederation rivaled Borno for the trans-Saharan trade.  Trans-Saharan trade increased throughout the 19th century, in spite of the growth of European trade on the coast.85   Trans-Saharan slave exporting thrived, especially after the Atlantic slave market vanished.

Yet the attraction of the Atlantic was later renewed because other advantages became apparent.  One of these was the supply of firearms, which in the 19th century became not only more common but also indispensable for survival.  Even after the British imposed their rule on the whole of modern Nigeria, the heirs to the Sokoto establishment continued to show concern for the route to Lagos and took every available means to assure that it remained open and secure for their trading operations.


In conclusion, we can present a composite model for the Sokoto jihâd which includes three dynamics: the religious ideology espoused by the jihâd leaders, the aspirations for social equity of the Fulani and Hausa peasantry, and finally the economic interest of the northern long-distance traders, particularly those specializing in slave trade.

The dynamic of large state formation is long distance trade.  As I have argued elsewhere,86 the trans-Saharan trade gave impetus to the formation of empires in the West African savanna from the 8th to the 15th century (and longer in Borno).  The Turkish occupation of North Africa in the 16th century put a damper on trade heading for the Maghrib.  So the Songhay empire died and had no immediate successor.  The Atlantic trade, especially from the 18th century, gave rise first to coastal empires and then, as its impact was felt further inland, to the jihâd states.  Oyo was the most important empire to rise in response to demand for slaves on the coast. The Alafins controlled the routes from the northern supply zones to the coastal ports.  The Hausa traders, after a risky collection of slaves in war torn Hausaland, had to surrender their profits in tolls passing through Oyo territory.  So they supported the Ilorin rebellion.

The ongoing jihâd conducted by the emirates of the Sokoto federation was a systematic slave-gathering operation.  Most slaves were absorbed internally.  But it is no accident that the Sokoto jihâd began at a time when the promise of a rich slave-exporting economy looked brightest, and that the extremities of the jihâd farthest from Sokoto were on the one hand the slave grounds of Adamawa and on the other hand the outlet routes controlled by Oyo.

Unfortunately for Sokoto ambitions, just as its hold on the North was consolidated and Oyo destroyed, leaving no major obstacle to the sea, European demand for slaves ended.  So the Sokoto confederation confined its attention to the North and the Sahara until trade in other commodities sparked its interest in the Atlantic once more.

19th century developments in Borno

      The al-Kanemi dynasty in Borno

Fulani flag bearers of Uthmân dan Fodiye attacked the western provinces of Borno in 1805 and conquered Hadija and Katagum.  In 1808 they destroyed the Borno capital Nagazargamu. Within Borno the Mai had all but lost his power and the country was ruled in his name by Muhammad al-Amîn al-Kanemi.  The latter corresponded with Muhammad Bello of Sokoto protesting the invasion of Borno.  Muhammad Bello justified the invasion on the grounds that Borno cooperated with the Hausa leaders whom the Fulani were fighting and that the Borno people engaged in pagan rites and customs.

Al-Kanemi built a new capital for Borno in 1814 at Kukawa.  Mai Dunama was desperate to regain his authority from this upstart general, but in a battle in 1817 the Baghirmi who had come to help Dunama killed him.  Al-Kanemi installed Dunama’s brother Ibrahim as the new Mai, but gave him no power. In 1824 al-Kanemi took Katagum back from the Fulani and prepared to invade Kano, but in 1826 Yakubu, the Sarki of Bauchi, defeated the Borno forces and drove them out of Katagum. Thereafter there were no conflicts between the two powers.

Muhammad al-Kanemi died in 1835 and was succeeded by his son Umar.  The latter was fighting a losing war aganst Wadai in 1846 when Mai Ibrahim made a bid to seize power.  Umar arrested and executed Ibrahim, putting an end to the thousand year old Sayf dynasty.

      Râbih ibn-Fadallâh

Born near Khartûm about 1840, Râbih ibn-Fadallâh became a soldier in the Egyptian army.  Returning to the Sudan around 1862, he took up service with Zubayr Rahma al-Mansûr Pasha al-Abbâsî, a trader in ivory and slaves who had become lord of Bahr al-Ghazâl and maintained an army of 12,000 men.  Râbih played an important role in solidifying Zubayr’s power in Bahr al-Ghazâl, which was recognized by Egypt in 1872, and in conquering Darfur in 1873. In 1874 Zubayr went to Cairo to get recognition of his authortiy over Darfur and was put in detention.  His son Sulaymân inherited Zubayr’s realms but later was denounced as a rebel against Egypt.  He surrendered to an Egyptian force in 1879 but was trecherously killed.

Râbih had refused to take part in the surrender and took over Bahr al-Ghazâl with his 1,000 troops.   From 1879 to 1882 he used his control over the ivory and slave trade to gather more men and arms and conquer small states to the south and west.  In 1883 he began overrunning the southern dependencis of Wadai. He could not defeat Wadai itself and therefore continued westward, coming into conflict with Baghirmi, south of Lake Chad.  In 1893 he had conquered Bahgirmi and was on the edge of Borno.

Borno had declined rapidly after the death of Umar al-Kanemi in 1880 and was weak when Râbih appeared.  Besides, Râbih identified himself with the Mahdist movement which had swept over the Sudan and thus found many sympathisers from Borno and Sokoto ready to help him.  Râbih’s men plundered Kukawa and defeated the main Borno forces in October 1893 and in the next year subdued all the rest of Borno.  Râbih’s conquest was so destructive that the whole country was in a state of economic collapse.  There was not enough food, and trade with North Africa and elsewhere was completely interrupted.  Although Borno’s neighbours attacked anyone going there to trade, the land recovered enough for Râbih to plan the conquest of Wadai in 1900.  The French estimated that he would succeed, but before he could try the French from Niger intervened and killed him in battle.  The reestablished the Kanemi dynasty in Kukawa, but handed over the territory to the British in 1902.  The Shehus (Kanemis) of Bornu thereafter became instruments of the British system of indirect rule.

«— Chapter 13

The Spread of Islam
Through North to West Africa

Chapter 15 —»

1See for example J.S. Trimingham, A history of Islam in West Africa (London, 1962), M. Hiskett, “The nineteenth-century jihads in West Africa”, ch. 4 in J. Flint (ed.), The Cambridge history of Africa, vol. 5 (Cambridge, 1976), and D. Murray Last, The Sokoto Caliphate (London, 1967).

2The best study is that of D. Murray Last, op. cit.

3See in particular M. Hiskett, The sword of truth, the life and times of the Shehu Usuman Dan Fodio (New York, 1973), and above all F.H. El-Masri, A critical edition of Dan Fodio’s Bayân wujûb al-hijra calâ l-cibâd, with introduction, English translation and commentary (University of Ibadan Ph.D. thesis, 1968).

4See particularly M. Waldman, “The Fulani jihâd: a reassessment”, Journal of African History, 6 (1965), pp. 333-355, and “A note on the ethnic interpretation of the Fulani jihâd”, Africa, 36 (1966), pp. 286-291.

5Notably by P. Morton-Williams, “The Oyo Yoruba and the Atlantic trade, 1670-1830", Journal of the Historical Society of Nigeria, 3 (1964), pp. 25-45, and “The Yoruba kingdom of y“, in D. Forde & P. Kaberry (eds.), West African kingdoms in the nineteenth century (London, 1967), pp. 36-69, R. Smith, Kingdoms of the Yoruba (London, 1969), J.A. Atanda, “The fall of the old Oyo empire: a re-consideration of its causes”, JHSN 5 (1971), 477-490, and I.A. Akinjogbin, “The prelude to the Yoruba civil wars of the nineteenth century”, Odu, 1:2 (1965), 24-46, “A chronology of Yoruba history, 1789-1840", Odu, 2:2 (1966), 81-86, “The expansion of Oyo and the rise of Dahomey 1600-1800", ch. 10 in J.F.A. Ajayi & M. Crowder (eds.), History of West Africa, I (London, 1976), 2nd ed.), and R. Law, The Oyo empire, c. 1600- c.1836 (Oxford, 1977).

6The Sokoto Caliphate (London: Longman, 1967).

7The Oyo empire, c.1600- c.1836.

8For this section see R. Law, The Oyo empire, c. 1600-c.1836 (Oxford, 1977); I. Akinjogbin, “The expansion of y“ (in Ajayi & Crowder), and “The economic foundations of the Oyo empire”, ch 3 in I.A. Akinjogbin & S. Osoba, Topics in Nigerian economic and social history (University of Ife Press, 1980), pp. 35-54.

9Cf. R. Smith, “The Alafin in exile”, Journal of African History, 6 (1965), pp. 57-77.

10Cf. R. Law, op. cit., pp. 201-205.

11Cf. B.A. Agiri, “The Yoruba and the pre-colonial trade”, Odu, 12 (1975), 55-68, and “The introduction of Nitida Kola into Nigerian agriculture, 1880-1920", African Economic History, 3 (1977), 1-14.

12Ibid., p. 198.

13Ibid., pp. 92-93 & 238; cf. also Kola Folayan, “Egbado to 1832: the birth of a dilemma”,  Journal of the Historical Society of Nigeria, 4 (1967), pp. 15-34.

14Cf. Law, pp. 93 & 238.

15Cf. Akinjogbin, “The expansion of Oyo“, p. 391.

16Cf. Law, The Oyo empire, pp. 219-220.

17Cf. Akinjogbin, “The expansion of Oyo“, p. 397.  For a general picture see R. Law, “Dahomey and the slave trade: reflections on the historiography of the rise of Dahomey”, Journal of African History, 27 (1986), pp. 237-267.

18P. Morton-Williams, “The Yoruba kingdom of Oyo“, p. 37; cf. his “The Oyo Yoruba and the Atlantic trade”.

19R. Smith, Kingdoms of the Yoruba, p. 45.

20Akinjogbin, “The expansion of Oyo“, pp. 386-387.  J.F.A. Ajayi & R. Smith, in Yoruba warfare in the nineteenth century (Ibadan, 1971), pp. 124-125, also try to minimize the fact and role of slave trade in Yoruba country.

21Ibid., p. 409.

22I.A. Akinjogbin, “The prelude to the Yoruba civil wars”, Odu, 1:2 (1965), p. 28; cf. his “The Oyo empire in the eighteenth century - a reassessment”, Journal of the Historical Society of Nigeria, 3:3 (1966), p. 457.

23Cf. R. Law, The Oyo empire, p. 219 and note 127.

24Ibid., p. 221.

25Ibid., p. 221, note 137.

26Ibid., p. 222.

27Ibid., p. 223; R. Law, “Trade and politics behind the Slave Coast: the Lagoon traffic and the rise of Lagos”, Journal of African History, 24 (1983), pp. 321-348.

28These were from  Oyo’s northern and western neighbours, according to Law, who disputes the assertion of Morton-Williams that they were from Ekitiland. Morton-William’s opinion is repeated by J. Smaldone, Warfare in the Sokoto Caliphate (Cambridge, 1977), p. 41.

29John Adams, Remarks on the country extending from Cape Palmas to the River Congo (London, 1823), pp. 221-222; cf. also pp. 78 & 82-95; cf. Law, The Oyo empire, pp. 225-227.

30Cf. Law, The Oyo empire, pp. 225, 227-228.

31Ibid., p. 188.

32J.F.A. Ajayi & R. Smith, Yoruba warfare in the nineteenth century (Ibadan, 1971), pp. 124-125; see also Paul E. Lovejoy, “Interregional monetary flows in the precolonial trade of Nigeria”, Journal of African History, 15 (1974), pp. 563-585.

33See the discussion of evidence and presumptive arguments in R. Law, The Oyo empire, pp. 234-236.

34Cf. Law, The Oyo empire, pp. 246-248.

35“The expansion of Oyo pp. 407-408; cf. R. Law, The Oyo empire, p. 234.

36Cf. R. Law, The Oyo empire, pp. 246-248.

37For debates about the details, see J.A. Atanda, “The fall of the old Oyo empire: a re-consideration of its cause”, Journal of the Historical Society of Nigeria, 5 (1971), pp. 477-490.

38Cf. R. Law, The Oyo empire, p. 294.

39For this section see Last, The Sokoto Caliphate, S.J. Hogben & A.H.M. Kirk-Greene, The emirates of Northern Nigeria (London, 1966), and R.A. Adely, Power and diplomacy in Northern Nigeria 1804-1906 (London, 1971).

40Cf. Michael Mason, The foundations of the Bida kingdom (Zaria: Ahmadu Bello U.P., 1981), ch.2; Ade Obayemi, “The Sokoto jihad and the ‘O-kun’ Yoruba: a review”, Journal of the Historical Society of Nigeria, 9:2 (1978), pp. 61-88, “Kakanda: a people, a history, an identity”, Journal of the Historical Society of Nigeria, 9:3 (1978), pp. 1-22.

41Cf. Michael Mason, The foundations of the Bida kingdom (Zaria: Ahmadu Bello U.P., 1981).

42Cf. Sa’ad Abubakar, “The emirate-type of government in the Sokoto Caliphate”, Journal of the Historical Society of Nigeria, 7 (1974), pp. 211-230; C.N. Ubah, “Administrative principles and practices in some Nigerian emirates 1800-1918", Journal of the Historical Society of Nigeria, 8:3 (1976), pp. 37-54.

43For example Hogben & Kirk-Greene, op. cit., Hiskett, “The nineteenth-century jihads”, and J. Boyd, The flame of Islam (Zaria, 1969).  On cUthmân danFodiye’s own views, see El-Masri, op. cit.; Muhammad al-Hajj, “The Fulani concept of jihâd - Shehu Uthmân dan Fodio”, Odu, 1:1 (1964), pp. 45-58.

44L. Gardet, La cité musulmane, vie sociale et politique (Paris, 1969), is especially excellent on this subject.

45This is the view of D. Murray Last; for the added dimension of the union of Hausa kingship with the traditional priesthood, see Finn Fuglestad, “Hausa history before the jihad”, Journal of African History, 19 (1978), pp. 319-339.

46T. Hodgkin, “Islam, history and politics”, The Journal of Modern African Studies, 1 (1963), pp. 91-97.

47J. Trimingham, History, p. 83.

48Cf. D.P. Ayagere, The life and works of cAbdullahi b. Fudi (University of Ibadan Ph.D. thesis, 1971).

49A summary of some of these hypotheses was made by P. Waterman, “The jihad in Hausaland as an episode in African history, some concepts, theories and hypotheses”, Kroniek van Afrika, 2 (1975), pp. 141-152, and by M. Waldman, “The Fulani jihâd”.

50This theory, proposed by Ch. Meek, The northern tribes of Nigeria (London, 1925), I, p. 100, reflects Ibn-Khaldûn’s views on the relationship between nomadic and sedentary peoples; see his Muqaddima, ch. 2 (tr. F. Rosenthal, abridged by N. Dawood, London, 1967). The ethnic theory is also discussed by Adely, Power and diplomacy, p. 19.

51Cf. H. Barth, Travels and discoveries in North and Central Africa (London: Cass, 1965), vol. 3, p. 101.

52Cf. D. Olderogge, “Feudalism in the Western Sudan in the 16th to 19th centuries”, summarized in African Abstracts, 10 (1959), pp. 11-12.

53J. Goody, “Feudalism in Africa”, Journal of African History, 4 (1963), pp. 1-18; “Economy and feudalism in Africa”, Economic History Review, 22 (1969), pp. 393-405.

54M. Waldman, “The Fulani jihâd”, pp. 341-345.  H. Barth, in Travels and discoveries (London, 1965), III, p. 101, relates that the Zamfara Hausa supported cUthmân dan Fodiye to settle an old score against their Gobir enemies.

55“La notion de ‘mode de production asiatique’ et les schemas marxistes d’évolution des sociétés”, in CERM (ed.), Sur le “mode de production asiatique” (Paris, 1969).

56“Les sociétés traditionelles en Afrique tropicale et le concept de mode de production asiatique”, ibid.

57For more on the Marxist interpretations, see Thea Buttner, “The economic and social character of pre-colonial states in tropical Africa”, Journal of the Historical Society of Nigeria, 5 (1970), 275-290.

58“Recherches sur les modes de production africaines”, La pensée, n. 144 (1969), pp. 61-78.

59Besides Coquery-Vidrovitch, see my chapter “The economic dimension of Islam in West African History”, in AECAWA, Christianity and Islam in dialogue (Cape Coast: Association of Episcopal Conferences of Anglophone West Africa, 1987), pp. 84-90.  See also N. Levtzion, Ancient Ghana and Mali (London, 1973), ch. 1, and “The Sahara and the Sudan from the Arab conquest of the Maghrib to the rise of the Almoravids”, ch. 11 in J. Fage (ed.), The Cambridge History of Africa, II (London, 1978).

60P. Waterman, “The jihad in Hausaland”, does not propose any such modification of this model.

61“Economic change in West Africa in the nineteenth century”, ch. 10 in J.F.A. Ajayi & M. Crowder (eds.), History of West Africa, II (London, 1974).

62An economic history of West Africa (London, 1973).

63For more on theories of the economy of pre-colonial Africa, see Ralph A. Austen, “The abolition of the overseas slave trade: a distorted theme in West African history”, Journal of the Historical Society of Nigeria, 5 (1970), pp. 257-274.

64Cf. W. Rodney, “Jihad and social revolution in Futa Djalon in the eighteenth century”, Journal of the Historical Society of Nigeria, 4 (1968), pp. 269-284.

65Cf. Paul E. Lovejoy, “Plantations in the economy of the Sokoto caliphate”, Journal of African History, 19 (1978), pp. 341-368.

66Cf. J. Smaldone, Warfare in the Sokoto Caliphate, pp. 12 & 147-149.

67Cf. R. Horton, “African conversion”, Africa, 41 (1971), pp. 85-108; “On the rationality of conversion”, Africa, 45 (1975), part 1, pp. 219-35, part 2, pp. 373-399.

68G. Nicolas, “Fondements magico-religieux du pouvoir politique au sein de la principauté Hausa du Gobir”, Journal de la Société des Africanistes, 39 (1969), pp. 199-231.

69“On the rationality of conversion”, p. 388.

70The Oyo empire, p. 308.  The general position that slave raiding did not hurt or disrupt the slaving states themselves but only their weaker neighbours is maintained also by J.D. Fage, “Slavery and the slave trade in the context of West African history”, Journal of African History, 10 (1969), pp. 551-564.

71Cf. M. Hiskett, Tazyîn al-waraqât by cAbdullâh ibn Muammad, edited with translation... (Ibadan, 1963), p. 122, The sword of truth, the life and times of the Shehu Usuman Dan Fodio (New York, 1973), p. 78.  In A history of Hausa Islamic verse (London, 1975), Hiskett refers to the poem Tabban hakikan under the title Tabbat hakîkâ (xv, 171, 177).  Yet in his bibliography (xxiv) he states that he is using the “unedited text of this

72Infâq al-maysûr fî ta’rîkh bilâd at-Takrûr (Cairo, 1964, written in 1812), p. 48.

73Cf. Pierre Verger, Trade relations between the Bight of Benin and Bahia, 17th -19th century (Ibadan U.P., 1976), chs. 9 & 16, and Rolf Reichert, “L’insurrection d’esclaves de 1835 à la lumière des documents arabes des Archives publiques de l’Etat de Bahia (Brésil)”, BIFAN, 29 (1967), 99-104.

74Cf. Hogben & Kirk-Green, Emirates, p. 166.

75Ibid., p. 427.

76Ibid., p. 196.

77Ibid., pp. 219 & 163.

78Cf. W. Rodney, “Jihad and social revolution”; “African slavery and other forms of social oppression on the Upper Guinea coast in the context of the Atlantic trade”, Journal of African History, 7 (1966), pp. 431-443; A. Bathily, “La traite atlantique des esclaves et ses effets économiques et sociaux en Afrique: le cas du Galam, royaume de l’hinterland Sénégambien au six-huitième siècle”, Journal of African History, 27 (1986), pp. 269-293.

79On the slave trade in Hausaland in the pre-jihâd and jihâd periods see Smaldone, Warfare, pp. 12-13; L. Colvin, “The commerce of Hausaland, 1780-1833", ch. 6 in D. McCall & N. Bennett (eds.), Aspects of West African Islam (Boston, 1971); A. Fisher & H. Fisher, Slavery and Muslim society in West Africa (London, 1970); M. Johnson, “The slaves of Salaga”, Journal of African History, 27 (1986), pp. 341-362; J.S. Hogendorn, “Slave acquisition and delivery in pre-colonial Hausaland”, in B.K. Swartz & R.A. Dumett (eds.), West African culture dynamics (The Hague: Mouton, 1980).

80Cf. Paul E. Lovejoy, “Plantations in the economy of the Sokoto caliphate”, Journal of African History, 19 (1978), pp. 341-368; J. Hogendorn, “The economics of slave use on two plantations in Zaria emirate”, The Int. J. of African Historical Studies 3 (1977), 369-383; D. Tambo, “The Sokoto caliphate slave trade in the 19th century”, The Int. J. of African Historical Studies 3 (1976), 187-217.

81Cf. Paul E. Lovejoy, “Interregional monetary flows in the precolonial trade of Nigeria”, Journal of African History, 15 (1974), pp. 563-585.  E.A. Ayandele, in Nigerian historical studies. London: Cass, 1979, ch. 4 “Observations on some social and economic aspects of slavery in pre-colonial northern Nigeria”, contests the importance of slave trade to the Sokoto economy.

82S.A. Balogun alleges Sokoto-Gwandu tension as a reason; cf. “The place of Argungu in Gwandu history”, Journal of the historical society of Nigeria, 7 (1974), pp. 403-416.

83The history of the Yorubas (Lagos, 1921), p. 288.

84Cf. Flint, “Economic change”, p. 392.

85Cf. J.E. Flint, “Economic change in West Africa in the nineteenth century”, p. 390; H.A. Gemery & J.S. Hogendorn, “The Atlantic slave trade: a tentative economic model”, Journal of African History, 15 (1974), pp. 223-246; L. Brenner, “The North African trading community in the nineteenth-century Central Sudan”, ch. 7 in D. McCall & N. Bennett (eds.), Aspects of West African Islam; Marion Johnson, “Calico caravans: the Tripoli-Kano trade after 1880", Journal of African History, 17 (1976), pp. 95-117, and “By ship or by camel: the struggle for the Cameroons ivory trade in the nineteeenth century”, Journal of African History, 19 (1978), pp. 539-549.

86“The economic dimension of Islam in West African history”, in Christianity and Islam in dialogue, a publication of the Association of Episcopal Conferences of Anglophone West Africa (Accra, 1987), pp. 84-90.