THE recent debate about Iron Age Gao between John Hunwick and Dierk Lange within the pages of The Journal of African History has prompted the writing of this paper. Hunwick lamented the lack of archaeological research in the region, and hoped that the historical hypotheses which had been proposed would, one day, either be supported or refuted by archaeological work in conjunction with linguistic analysis. The limits of interpretation of the historical sources would appear to have been reached, and, as this author has recently conducted excavations and surveys at Gao, a contribution is offered to the debate based upon both the archaeological evidence recovered from these excavations, and from previous archaeological research which has been conducted in the Gao region.

In a recent article David Richardson revised estimates of the volume and African regional distribution of the eighteenth-century British slave trade. Richardson calculated that British slave vessels sailing between 1698 and 1807 – to the year of Abolition – embarked 3,052,509 slaves on the African coast. This figure falls at the midpoint between lower-bound estimates made by Curtin in 1969 and upper-bound estimates published by Inikori seven years later. Further, Richardson provided historians with the first extended year-by-year series of British slave exports; and he separated the trading data of the three principal British slaving ports of Liverpool, London and Bristol, which comprised 50, 28 and 18 per cent of the trade, respectively.

The work of historical and comparative linguists has long interested African historians. By classifying languages into families, linguists provide models of their historical development that may point to historical events and processes that occurred among peoples speaking those languages. Once classified, linguists can then reconstruct earlier forms of present languages, thus providing direct evidence of words, their meanings and historical influences in the past. Finally, linguists seek to explain innovations that are revealed in their reconstructions by pointing to a combination of internal linguistic developments and different forms of contact that occurred among speakers of different languages.

Simple classification, based largely on counting cognate words in related languages (a technique known as lexicostatistics), is still a very common activity, however, and thus the one most historians rely on, but lexicostatistics gives only a very limited, and often deceptive, view of language history. Historians should thus be aware of its limitations as well as the potential of a number of important techniques now employed by linguists, including the Comparative Method, reconstruction of ancestral languages, and contact models.

In 1858, after a reign of forty years, King Gezo of Dahomey died and was succeeded by one of his sons called Badahun, who took the royal name of Glele by which he is more generally known. Badahun had been Gezo's designated heir apparent for at least nine years prior to this but his accession to the throne was nevertheless challenged. The name Glele which he adopted alludes to these challenges, being according to Dahomian tradition abbreviated from the aphorism Glelile ma ñõn ze, ‘You cannot take away a farm [gle]’, meaning that he would not allow anyone to appropriate the fruits of his labours, which is explained as expressing ‘his contempt for the attacks to which he had been exposed as heir apparent’.

ALTHOUGH slaves were the most common merchandise in the Portuguese-dominated opening period of the seaborne trade between Europe and Africa, relatively little conclusive information is available on their overall numbers. Even less is known about the distribution of these early exports of slaves in space and time, although these are two of the key factors in assessing the much debated societal impact of the early Atlantic slave trade and the role of slavery in West and West-Central African economic, social and political life.

Since 1969 archaeological, paleobotanical and ethnographic research in Kagera region, north-western Tanzania (Map 1), have contributed to a deep-time understanding of the ecological history of that region. The primary insight that develops out of this interdisciplinary research is that changes in the ecology of the region can be linked directly to the productive economy and the cultural values of its inhabitants. The development of iron technology and settled agricultural life are the primary components in understanding the environmental changes that arose in this region over the last two millennia or more. This paper provides an overview of the most salient observations derived from this research program. I will begin with a short review of the history of Iron Age settlement, technology and major alterations made to the environment that form the backdrop to the specific reconstructions of forest exploitation gleaned from charcoal excavated from iron-smelting furnaces and from reconstructions of vegetational changes based on palynology.

BY the twentieth century, the southern Yoruba town of Abeokuta was integrated into the international economy both as an exporter of cash crops and an importer of manufactured products. It became one of the primary producing areas of cocoa and kola nuts in western Nigeria. Abeokuta's integration into the international economy had a profound effect on its local textile industry. As weavers gained access to European threads and dyers gained access to European cloth, relations of production were transformed. Both sets of producers became dependent on European trading firms for their raw materials and were thus brought squarely into the nexus of international trade. Dyeing, which was predominantly a women's industry, benefited substantially from this economic relationship. Dyers' access to cloth as well as credit from the European firms allowed them to become autonomous producers of tie-dyed cloth, adire, that was in great demand across Nigeria, the Gold Coast, Senegal and the Belgian Congo.

The ancient Egyptians tended to consider both their immediate and more remote neighbours either as excellent sources of luxury trade items and slaves, at times of political power and strength, or as uncivilized forces threatening to destroy and overwhelm the Egyptian Nile Valley. This was true not only of cultures within Africa; those in the Aegean, the Levant, Cyprus and Mesopotamia also were viewed from one or the other perspective, occasionally both at the same time. All these, together with Nubia and Libya, have received much scholarly attention and, whilst we probably will never fully understand their ancient relationship to Egypt, we have a fairly good idea, ‘hearing’ the story from both sides when surviving evidence allows, what their attitudes towards each other were at various times in their history.

This article deals with what is surely the least investigated aspect of ancient Egyptian relations with its neighbours, simply because we know comparatively little about the two major successive cultures concerned – Punt and Aksum – in the ancient world.

Late in 1907 a missionary from Kisserawe in German East Africa complained of a spate of ngoma ritual dances among the Zaramo people. In particular he singled out an ngoma conducted by women to ameliorate a drought that was threatening that year's maize crop. As the women danced around a well, dressed as men and brandishing muskets, they appealed for rain from ‘their god’. Several aspects of this ngoma make it remarkable. It occurred following the Majimaji uprising in German East Africa, which the Germans put down with such violence as to make war as a tactic of resistance unpopular if not untenable. The ngoma was attended by Christian and non-Christian African women alike, suggesting a purpose whose expediency cut across competing belief systems. Finally, although cross-dressing was an aspect of certain Zaramo rituals, the symbolic appropriation of men's social roles by dress and wielding of weapons made this ngoma anomalous and suggests that the participants were consciously and purposefully reshaping gender roles at this time. The timing and symbolism of the ngoma make it clear that it was a reaction to the threat of famine, which had become a recurrent aspect of Zaramo life by 1907 and a symptom of ongoing rural social change ushered in by colonial rule. The larger question is whether changing perceptions of gender roles intersected with the Majimaji war (1905–7), and whether Majimaji had an underlying meaning for rural Tanzanian societies that has escaped the attention of historians. If so, it suggests that the prevailing conception of Majimaji needs to be questioned and re-examined.

This paper examines the role of the imperial and colonial governments in the formulation of policy towards the Nigerian palm oil export industry between 1939 and 1949. It argues that for most of the war years colonial officials in Nigeria accepted that metropolitan needs and conditions should dictate policy in the oil palm produce industry. However, towards the end of the war, they began to question whether policies centred around the requirements of the metropole would preserve the future competitiveness of the industry. Thereafter, they pressed for measures which gave priority to the problems and necessities of the local industry and the colonial economy. While colonial policy was sensitive to the concerns of imperial and local government officials, for most of the period under review it was reluctant, and on occasions, unable to accommodate the measures necessary to harmonize imperial and colonial goals. Consequently, the anticipated expansion in palm oil exports failed to materialize and the future competitiveness of the industry remained in doubt.

This article fills an important void in the current literature on the Nigerian palm oil export industry. To date insufficient attention has been paid to the thinking within imperial and colonial government circles which underpinned the policies adopted in the industry during World War II and the early post-war years, and which led to the failure of policy makers to achieve their objectives. Moreover, the current literature ignores the vigorous debate between the Colonial Office and the Nigerian colonial government, and among colonial government officials, over the best means by which the needs of the local palm oil industry could be reconciled with the demands of the metropole, especially between 1942 and 1949.

AMONG the most critical effects of labour migrancy on a labour reserve economy in southern Africa are the psychological and economic implications the dependence on migrants' earnings, and the prolonged absence of men at the labour centres, have for women. Lesotho is one such case. But Basotho women, for their part, developed various survival strategies in response to their social and economic predicament. Judy Kimble and Philip Bonner have pioneered the study of these responses, but they made no effort to look at the strategies of the women remaining at home. When scholarly attention has been given to the latter category of women, the focus has been on the period after 1940. This study, by focusing on the engagement of Basotho women in commercial beer-brewing and prostitution in Lesotho before the Second World War, intends to address this lacuna. An attempt will also be made to link the rise of these phenomena to social and economic changes in colonial Lesotho, particularly the deepening of dependency on the migrant labour system.

The fundamental preoccupation with race in later historical writing in South Africa has its origins in the Great Debate between liberals and their enemies in the early nineteenth century. Standard overviews of South African historiography date the emergence of racially structured histories to the second half of the nineteenth century. For Saunders, the making of the South African past and its thematic ordering in terms of race only began in the 1870s ‘when the first major historian [G. M. Theal] began to write his history’. Prior to Theal's monumental efforts, ‘only a few amateur historians had turned their hands to the writing of the history of particular areas or topics’. Likewise, in Smith's analysis, also published in 1988, the construction of South African history in terms of race is seen almost exclusively as the product of the late nineteenth and twentieth centuries. In a very brief introductory section, Smith suggests that what little historical writing there was before the middle of the nineteenth century is scarcely to be taken seriously, and his study offers no more than a bare outline of historiographical developments before Theal and his heirs.

Historical writing on the coloured community of South Africa has tended to accept coloured identity as given and to portray it as fixed. The failure to take cognizance of the fluidity of coloured self-definition and the ambiguities inherent to the process has resulted in South African historiography presenting an over-simplified image of the phenomenon. The problem stems partly from an almost exclusive focus on coloured protest politics which has had the effect of exaggerating the resistance of coloureds to white supremacism and largely ignoring their accommodation with the South African racial system. Furthermore, little consideration has been given to the role that coloured people themselves have played in the making of their own identity or to the manner in which this process of self-definition shaped political consciousness. This is particularly true of analyses of the period following the inauguration of the Union of South Africa in 1910, a time when the legitimacy of coloured identity was not in any way questioned within the coloured community and when coloured protest politics was dominated by one body, the African Political Organization (APO).

Ce livre est composé, outre d'une introduction originale qui tente de synthétiser l'approche générale, de sept articles de l'auteur parus entre 1988 et 1993.

The last twenty years have seen a series of studies dealing, at least in part, with the nineteenth-century history of slavery at the East African coast. Each has, in its own way, focused on transformations associated with changing patterns of accumulation in the nineteenth century. If there has been a general theme it is of the increasing constraints placed upon slaves and the increasing demands made on them, as owners sought to reorganize labour time and processes to take advantage of the new opportunities offered by the rapid expansion of commerce from the 1830s. While Morton has attacked Cooper's ‘hegemonic’ perspective and accused him of presenting slavery as benign and static, both are agreed on a basic trend: the increasingly commercial orientation of slave-based agriculture considerably diminished slave autonomy between 1820 and 1890. Recently, Glassman has offered a study which is decidedly non-hegemonic in perspective, and has revealed the ways in which marginal members of society appropriated and sought to reinterpret the ideology through which they were subordinated. Yet he too describes the increasing circumscription of slave autonomy in response to the demands of new kinds of production – in his case, the sugar plantations of the Pangani valley.

‘Writing history’, claims Paul Landau in a revealing comment in his epilogue, ‘is an attempt to pin and mount lives on the page of a book, like butterflies or rose petals’ (p. 219). The generalization is sweeping, but the tropes aptly symbolize the type of history written in this book. This is a subtle, nuanced account of the intertwining of Christianity and politics in central Botswana from the 1850s to the 1940s. It is deeply informed by Foucauldian notions of power. Landau, moreover, is troubled by the implications of exercising power, through the very act of ‘pinning’ down lives.