Using new evidence on the British, French and North American slave-carrying trades, this article seeks to revise Lovejoy's recently published estimates of the levels of slave exports from West and West-Central Africa in the eighteenth century. The figures suggest that Lovejoy's estimate of the total volume of slave exports from the west coast of Africa to America between 1700 and 1810 was probably reasonably accurate, being only 8 per cent lower than the total indicated here. However, the new data reveal temporal and coastal distributions of slave exports that differ substantially from those proposed by Lovejoy. In particular, they suggest that previous work significantly understated levels of slave exports between 1713 and 1740, and again in the 1760s and 1770s. Contrary to earlier findings, in fact, it appears that slave exports from the west coast of Africa to America in the decade prior to the War of American Independence were very similar to levels attained after 1783. Furthermore, in terms of coastal distributions, it seems that the Bight of Biafra and West-Central Africa, particularly the Loango coast, contributed much more substantially to the slave traffic to America during the early decades of the century than was previously assumed. These revisions of Lovejoy's figures have important implications for movements in slave prices in Africa and for assessing the demographic effects of the trade on the slave-supplying regions. In addition, they help to improve our understanding of the relationship between the slave trade and changes in sugar and other commodity production in America during the eighteenth century.

This article examines the age and sex composition of the Atlantic slave trade in the belief it was of considerable significance in shaping black society in both Africa and the Americas. Focusing on the French slave trade, two main samples are analysed. One is composed of 177,000 slaves transported in French ships during the years 1714–92, which is taken from the Répertoire des expéditions négrières of Jean Mettas and Serge Daget. The other, derived from nearly 400 estate inventories, consists of more than 13,300 Africans who lived on Saint Domingue plantations in the period 1721–97. The results are compared with existing knowledge of the demo-graphic composition of the Atlantic slave trade to show the range of variation that existed through time between different importing and exporting regions, and to shed light on the forces of supply and demand that determined the proportions of men, women and children who were sold as slaves across the ocean. Significant and consistent contrasts are found between different ethnic groups in Africa and different slaveholding societies in the New World, many of them thus far unnoticed in the scholarly literature.

This article, which extends and modifies the analysis offered in an earlier article in this journal (1977), examines what is known of the organization of the supply of slaves for the trans-Atlantic trade in Dahomey, with particular emphasis on the relative importance of local slave-raiding and the purchase of slaves from the interior, and on the evolution of a group of private merchants within Dahomey. It is argued that initially the kings of Dahomey sought to operate the slave trade as a royal monopoly, and relied exclusively upon slave-raiding rather than purchasing slaves from the interior. From the mid-eighteenth century, however, Dahomey did seek to operate as a ‘ middleman’ in the supply of slaves from the interior, and since its kings did not normally attempt to control this aspect of the trade this involved the emergence of a private sector in the slave trade. Although merchants in Dahomey were in origin state officials, licensed to trade on behalf of the king or ‘caboceers’ (chiefs), they subsequently acquired the right to trade on their own account also and thus became in some measure independent of the state structure. The autonomy and wealth of the merchant community in Dahomey were further enhanced by the transition from slave to palm oil exports in the nineteenth century, when leading merchants moved into large-scale oil production, using slave labour supplied by the king. There were recurrent tensions between the monarchy and the merchants over commercial policy and over the monarchy's expropriatory fiscal practices, and the conflict of interests between the two was exacerbated by the development of the oil trade, undermining the solidarity of Dahomey in the face of the European imperialism of the late nineteenth century.

The Yoruba toll system has not been studied, in spite of its important place in Yoruba economy and politics. This essay fills the gap by examining toll collection among the Yoruba-speaking states of south-western Nigeria. It is divided into two parts, the first on the practice of toll collection during the pre-colonial era and the second on the changes introduced by the colonial administration. For the pre-colonial, it emphasizes the dominant aspects of the system, most notably the significance of toll revenue in relation to other sources of income; the control of toll gates by chiefs in order to appropriate the revenues; the character and privileges of collectors; and the features of collection at the toll gates, especially the duties imposed and their implications for trade.

The second part explains the steps taken by the new colonial administration to regulate toll collection after 1893, notably by the reduction of customs houses and the printing of tariffs. These reforms failed to solve the problems of corruption by toll clerks and evasions and smuggling by traders, or allay the fear that the imposition of tolls constituted an obstacle to modern commerce. Consequently, the colonial administration decided to abolish the system, and was able to achieve this between 1904 and 1908. Both reforms and abolition were possible because of the gradual approach adopted, the administrative and military power available to the administration, and its ability to generate alternative sources of revenue to maintain itself and pay the chiefs. There can be no doubt that abolition was a major step towards the constitution of the colonial economy.

L'etude des conséquences de l'émancipation des esclaves est d'une grande importance pour la comprehension de l'histoire du travail en Afrique occidentale. Le système social des Soninké de la haute vallée du Sénégal, comme pour beaucoup de peuples sahéliens à l'époque précoloniale, reposait largement sur l'esclavage. L'apparition des migrations de travail chez les Soninké, cependant, s'explique beaucoup moins par l'abolition de l'esclavage que par la disparition progressive du commerce esclavagiste en Sénégambie au dix-neuvième siècle. En effet, c'est alors que furent substitutés à la traite intérieure les migrations saisonnières des jeunes Soninké. Ces migrations, traditionnellement orientées vers le commerce en Gambie, furent détournées vers la production d'arachide dans cette même région, probablement sous l'influence des trafiquants d'esclaves Soninké. Quant à l'émancipation, elle ne créa nullement un exode chez les esclaves nouvellement libérés mais elle permit leur entrée dans la courant des migrations saisonnières. Dans ce sens, l'abolition de l'esclavage fut un phénomêne important dans l'histoire du travail en Afrique occidentale. De plus, elle suscita des transformations dans 1'organisation du travail familial chez les Soninké, qui résultèrent en un surcroît de migrants.

En conclusion, l'histoire des Soninké illustre l'importance de la question des migrations traditionnelles pour la compréhension des migrations modernes en Afrique occidentale, rappelant en cela l'histoire des migrations de travail en Europe, qui furent elles aussi les héritières de courants plus anciens de mobilité géographique.

During the formative years of capitalist agriculture, white farmers in the Orange Free State relied on both legal and extra-legal means to create a docile labour force of African workers. Coercive laws were a necessary component in the overall process of fashioning a rural working class out of quasi-independent squatter communities. Yet, no matter how repressive, the legal system alone was not sufficient to ensure work-discipline and docility on the white-owned farms. Frustrated with their inability to force African farm labourers to work diligently and to prevent them from deserting the farms, white farmers frequently turned to indiscriminate violence to sow terror amongst the dispossessed African rural population. White vigilantes formed a kind of paramilitary wing of the white farming class. While the state authorities never officially sanctioned vigilantism, those white farmers who used violence to intimidate their African labourers had little to fear with respect to prosecution, let alone conviction. Yet despite all their efforts to break the will of rural Africans, white farmers lived in continued fear of ‘native risings’. Rural Africans, too, remembered: the shooting incidents that occurred in the Orange Free State were indelibly etched in the collective memory of the emergent rural working class.

The western-educated Krio population of Sierra Leone participated in British imperial activity along the West African coast in the nineteenth century. Facing a far more complex ethnic configuration than their counterparts in Yorubaland, the Sierra Leoneans (Saro) in Port Harcourt, Nigeria, acquired much influence through the manipulation of class and ethnic relations. Though most Saro here had a modest education and were working-class, a few came to form the cream of the petty-bourgeoisie and were active in economic life and city administration. Potts-Johnson, arguably their most famous member, developed a flair for operating in his middle-class world, and also in the cultural orbit of the local and immigrant working-class. I. B. Johnson, another prominent Saro, lacked this quality. Though presenting a homogenous ethnic front, celebrated in the Sierra Leone Union and in church activity, Saro society was sharply polarized on class lines, a weakness not to be lost on the numerically superior and ambitious indigenous population. Faced with a choice, the indigenes opted for the avuncular Potts-Johnson, for whom they felt a greater social affinity than for the more distant I. B. Johnson. After Potts-Johnson, however, no Saro was to be allowed scope to develop a similar appeal.

The role of custom and tradition in the development of colonial rule in Africa has received little attention from scholars. Historians of colonial Kenya, particularly, have focused on the powerful transforming impact of the colonial state and economy and on the growth of opposition movements; they have had little to say about the processes through which previously autonomous societies negotiated their incorporation into the Kenya state. Yet by the 1920s and 1930s that state had acquired a substantial degree of popular legitimacy. ‘Customary’ institutions and rituals played an important part in the development of that legitimacy. This essay examines the institution of the genealogically defined ‘generation’ in the Embu-Mbeere area in colonial central Kenya and the ceremonies held in 1932 to mark the transition from one generation to the next. These ceremonies attracted considerable attention because they provided the occasion for the proclamation of rules, supported by the British administration, relating to the bitter issue of genital mutilation in female initiation. But this was not a crude case of the manipulation of custom. The attempt to reform female initiation was part of a larger process, of which the rituals of generation succession were elements, of building the ideological basis of a new ‘tribe’ in a society previously characterized by local autonomy and collective authority. As investigation of the succession ceremonies makes clear, the notion of a tribe dominated by appointed chiefs and identified with an exclusive territory lay at the centre of this ideology.