The aim of this article is to illustrate the process whereby certain Berber tribes during the eighth century A.D. substituted slaves from the Bilād al-Sūdān for Berber slaves from North Africa. From the outset, this conversion was influenced strongly, if not instigated, by Ibāḍī merchants until the slave trade became a predominantly Ibāḍī monopoly from the mid-eighth century onwards. The slave trade along the central Sudan route in particular provided the increase in the community's wealth and security, as well as the means for its establishment and expansion as a Muslim sect among diverse Berber tribes and, finally, as the origins for the subsequent, far-flung network of trans-Saharan trade.

Extending the approach of D. Eltis and L. C. Jennings to the seventeenth century, the author takes estimates for the decades 1623–32 and 1680–90 as the starting points for his discussion of trends in the composition and value of the Atlantic imports and exports of Western Africa. Contrary to prevalent opinion, he argues that at least from 1600 onwards the value of slave exports was two to three times higher than that of commodity exports, as measured according to the prices in America and Europe. However, during most of the century more imports were bartered in the commodity trade than in the slave trade, since the trading margin in the latter sector was considerably higher than in the former. The different margins go some way to explaining why the Portuguese concentrated on the slave trade from Angola between 1600 and 1635, which they could carry on with fewer European imports and more effectively protect, while the more efficient Dutch merchants achieved primacy in the competitive commodity trade of West Africa. The different margins also meant a very uneven distribution of imports over coastal regions. Owing to the predominance of Akan gold in the commodity trade, the Gold Coast drew an estimated fifty per cent of all imports at the beginning of the century and still accounted for 34 per cent at the end. Owing to its predominance in the slave trade, West-Central Africa drew 25 per cent of all imports throughout the century. The few available data on the composition of imports suggest that there may have been a shift from metal goods to textiles and a marked increase of Asian textiles and cowries. From 1593 on the Dutch may have initiated a shift in the gross barter terms of trade in favour of the African merchants which spread from the Gold Coast to other areas when the North-west Europeans obtained the major share in the Atlantic slave trade.

The article analyses the determination of prices in the pre-colonial West African kingdom of Dahomey, principally with relation to the domestic economy, on the basis of detailed analysis of contemporary documentation. The influential study of Dahomey by Karl Polanyi (1966) posited pervasive state control of both overseas trade and the domestic economy, including prices, which Polanyi argued were set according to traditional notions of equity or equivalence rather than responding to supply and demand. These hypothesized stable prices were held to be reflected in the long-term stability of the exchange value of the local currency of cowry shells, at least prior to the ‘Great Inflation’ caused by excessive European imports of these shells in the second half of the nineteenth century. Although his account of the state's control and management of economic matters was in many respects inaccurate or exaggerated, Polanyi was correct in asserting that prices were subject to state regulation. However, it is shown that prices were nevertheless liable to short-term fluctuations which reflected market conditions, and overall suffered a massive increase during the seventeenth and early eighteenth centuries, comparable in scale to the nineteenth-century ‘Great Inflation’, and likewise reflecting the uncontrolled increase in the money supply through European imports of cowries. Although these two periods of inflation were separated by a period of relative price stability between the mid-eighteenth and mid-nineteenth centuries, this was not restricted to Dahomey and should also be attributed to the operation of market forces rather than to the effectiveness of Dahomian economic administration. State intervention was probably more successful in holding down the wages of some groups of workers than in managing prices. Polanyi's insistence on the ‘non-market’ character of the Dahomian domestic economy is clearly untenable.

This article examines the impact of the introduction of the new technology of steam power into the West African trade in the second half of the nineteenth century. One of the changes that the introduction of steampower was expected to lead to was the opening up of the trade to small-scale African traders such as the Krios. Many Krios did make use of the steamships to extend their trading activities and entered areas previously ignored. Many used the steamship services to develop a coastwise trade; others, particularly in the Niger Delta, used them to enter the export trade to Britain. Yet others pioneered the use of steam launches, particularly on the River Niger and along the Slave Coast. In time however, such Krios found their ability to utilize the opportunities provided by the steamships under assault, partly from the European traders' counter-attack and partly from the general depression in the West African trade – itself indirectly caused by the introduction of the steamship – that set in by the 1870s. By the end of the century the position of Krios in the export-import trade of West Africa was being severely squeezed, just as it was in other areas of West African life. For them, steam power did not prove to be the boon it had been anticipated as being.

Although South African women's history has been growing in volume and sophistication over the past decade, the impact of gender analysis has yet to be felt in mainstream or radical historiography. One reason for this neglect is the way in which the categories of both ‘gender’ and ‘women’ have been conceived – with ‘women’ assumed to have a stable referent and ‘gender’ treated as synonymous with women. Those areas of social life where women are not immediately present have thus remained unreconstructed by the theoretical implications of gender. This is particularly the case with the history of ‘the state’.

The article identifies and looks critically at the major paradigms of South African women's and gender history in terms of how the relationship between ‘the state’ and ‘women’ is implicitly or explicitly represented. It argues that the understanding of the category ‘women’ as socially and historically constructed (as evident in more recently published gender history) provides a way of moving beyond the more static or abstractly posed state-versus-women relationship. This requires too that ‘the South African state’ be understood not as unitary or coherent but as institutionally diverse with different objectives being taken up and produced as policy and practice. The project then becomes one of understanding South African state formation as a gendered and gendering process, of exploring the different institutional sites and ruling discourses in which gender identities and categories are constructed.