This article reports over 250 new radiocarbon dates relevant to recent archaeological research in West Africa. Thanks to the continuing trend towards series of dates from either single sites or groups of related sites, some major blanks on the archaeological map of West Africa have been replaced by well-dated regional sequences. An example is the Malian Sahara, where palaeoenvironmental and archaeological investigations at a large number of sites have clarified the relationship between Holocene climatic change and Late Stone Age occupation. Other areas that were largely archaeological unknowns until the research reported in this article was undertaken include the middle Senegal valley, the Inland Niger Delta, and the Bassar region in Togo. Other research included here reinterprets previously studied, ‘classic’ Late Stone Age sequences, such as Adrar Bous, Kintampo and Tichitt. There are also new dates and details for early copper in Niger and Mauritania which prompt a reconsideration of the true nature of this proposed ‘Copper Age’. Of particular significance to general reconstructions of West African prehistory is the documentation of regional and long-distance trade accompanying the emergence of complex societies along the Middle Senegal and Middle Niger in the first millennium A.D.

The article begins with a brief commentary on calibration, in view of the recent publication of high-precision calibration curves. Several prevalent misconceptions of what calibration is and what it ought to do are addressed. We suggest that archaeologists and historians should routinely make reference to calibration in order to avoid misinterpreting radiocarbon results.

A substantial minority, perhaps 15 per cent of all Xhosa, refused to obey the prophetess Nongqawuse's orders to kill their cattle and destory their cornl. This divided Xhosaland into two parties, the amathamba (‘soft’ ones, or believers) and the amagogotya (‘hard’ ones, or unbelievers). The affiliation of individuals was partly determined by a number of factors – lungsickness in cattle, political attitude towards the Cape Colony, religious beliefs, kinship, age and gender – but a systematic analysis of each of these factors in turn suggests that none of them was sufficiently important to constitute the basis of either party.

The key to understanding the division lies in an analysis of the indigenous Xhosa terms ‘soft’ and ‘hard’. ‘Softness’ in Xhosa denotes the submissiveness of the individual to the common will of the community, whereas ‘hardness’ denotes the determination of the individual to pursue his own ends, even at communal expense. Translated into social terms, the ‘soft’ believers were those who remained committed to the mutual aid ethic of the declining precolonial society, whereas the ‘hard’ unbelievers were those who sought to seize advantage of the new opportunities offered by the colonial presence to increase their wealth and social prominence. The conflict between the social and personal imperatives was well expressed by Chief Smith Mhala, the unbelieving son of a believing father, when he said, ‘They say I am killing my father – so I would kill him before I would kill my cattle.’ Certainly, the division between amathamba and amagogotya ran much deeper than the division between belief and unbelief, and the Xhosa, in conferring these names, seem to have recognized the fact.

The protracted subjugation by the Portuguese of Guinea-Bissau was made possible by Abdul Njai and his army of auxiliary troops. Njai became an ally of the Portuguese in the mid-1890s and continued his support for the Portuguese conquest until about 1915. He provided logistical support, and served both as a commander in the Portuguese army and as a recruiter of African troops. Oral as well as written sources indicate that Njai was directly responsible for the successful campaigns fought against the strongholds of resistance to Portuguese authority. As a reward for his services, the Portuguese granted Njai political authority over Oio province. Thus Njai became a kind of ‘warrant chief’ in an area where his only legitimacy was based on force rather than traditional affiliation. Portuguese control remained limited even after 1915 and Njai governed his region as he pleased. African communities in Oio and elsewhere in Guinea-Bissau feared and respected this warlord more than the Portuguese. He thus became a threat to Portuguese colonial officials in Guinea-Bissau as well as their French counterparts in neighbouring Senegal. The Portuguese therefore turned on their erstwhile ally and, after unsuccessful attempts to bring him to heel through negotiation, mounted an expedition which resulted in his capture and deportation in 1919.

This article explores the origins and nature of the reformatory in Cape colonial society between 1882 and 1910. Born in a period of economic transition, its concern was with the reproduction of a labouring population precipitated by colonial conquest. Unlike the prison and compound, which gained their distinctive character from the way in which they were articulated to an emerging industrial capitalist society, the reformatory was shaped by the imperatives of merchant capital and commercial agriculture. Although based on the English model, local social realities quickly began to mould the particular nature of the reformatory in the Cape Colony. Firstly, classification for the purposes of control came to mean segregation in a colonial context. secondly, the needs of commercial agriculture meant that in Porter there was a much greater stress on the apprenticing of inmates than there was in the internal operations of the British reformatory.

The legal emancipation of slaves in Zanzibar and on the Swahili coast of Kenya was enacted in 1907, but the measure was not enforced on Lamu Island until 1910. The slave-owners of Lamu were already in dire straits from the decline of their plantations on the mainland and from the changes accompanying colonial rule which, by contrast with Mombasa, left Lamu Island as an economic backwater. They were little inclined to co-operate with the provisions of the legislation and were actively abetted in this by some their slaves. Emancipation was therefore a more protracted process than in those parts of the coast where alternative opportunities had opened up for ex-slaves and for landowners. Those who were gradually liberated either emigrated elsewhere or entered into new forms of dependent relationships with the Afro-Arab aristrocracy. Meanwhile, slavery lingered on under the noses of British officials.

The pervasive importance of gold mining in modern South Africa has become embedded in South African historiography. Despite this, little research has been done to ascertain its impact on the other major sector of the economy, agriculture.

The gold mines had a profound effect upon one particular branch of agriculture – beef farming. The mines purchased large amounts of beef and were able to use their buying power to confront beef farmers in the marketplace. In the recession following the First World War, the mines were caught in a profitability crisis that was to lead to the Rand Revolt in 1922. One of the ways in which mining attempted to ease its position was by cutting back on the cost of the meat it supplied to its African labour force. This initially involved co-operation with a powerful cold-storage company, big ranchers and a number of smaller farmers to form a Meat Producers Exchange. This fragile alliance fell apart when farmers, themselves on the verge of bankruptcy, attempted to take control of the Exchange and raise beef prices. The farmers failed and in 1923 the exchange collapsed.

The victory of the mining and cold-storage companies rested on a number of factors. Farmers were unable to organize effectively because of the defection of ranchers to the mines. Changing economic conditions in 1922 and 1923 permitted the mines to terminate their co-operation with beef farmers. Finally the mines were able to call upon the state for support. The state ensured the demise of the Exchange and the defeat of the beef farmers. In the process it showed itself capable of intervening decisively to protect the interests of certain sections of capital.

In 1926, the Governor-General of French West Africa issued a decree allowing local administrations to use a portion of the annual military draft as labourers on public works programmes. The only administrations to take full advantage of this decree was that of the French Soudan, where work had already begun on the first phase of the vast Niger irrigation scheme now known as the Office du Niger. During the next twenty-five years, more than fifty thousand so-called ‘second-portion’ workers from Soudan were assigned to the Office du Niger for a period of three years' service. Ironically, this new system of forced labour to exploit the irrigated land.