By the mid-first millennium a.d., Middle Niger cities took the form of many separate mounds clustered together. Many of these mounds may have been settlements of specialists. This distinctive city form may have had its origin in segmented, but articulated, Late Stone Age communities in the southern Sahara. The Pulse Model is an attempt to reconstruct the circumstances of environmental change and interactions among these communities that encouraged occupational specialization. The model predicts the best locations to search for evidence of early specialization, namely the several north–south trending palaeochannels of the southern Sahara. There, groups increasingly concerned with intensification of production within separate microenvironments would nevertheless have been in close contact. Climate shifts over the past several millennia create a ‘pulse’ of population movements, or shifts of ecological adaptations, along these long corridors. However, adaptation to climate change and stress incompletely explains the emergence of specialization. Tradition, myths, legends and material reinforcements of divisions between present-day ethnic and artisan groups in the Middle Niger suggest the ways in which corporate identity may have been constructed and maintained in the very distant past. If corporate identity can emerge in a form that discourages conflict between groups, the result might be increasingly specialized responses to climate change and to the economic and social opportunities of early urbanism. There should be no sharp discontinuity between the emerging specialization of the last millennia b.c. and the earlier clustered urbanism of cities such as Jenne-jeno. Middle Niger urbanism is an intensification of prehistoric social dynamics, not a revolutionary process.

Following the late fifteenth- and sixteenth-century cavalry revolution in Senegambia, the horse and slave trade became a major sector of the desert-edge political economy. Black African states imported horses from North Africa and the western Sahara in exchange for slaves. Over time, under conditions of increasing aridity, the zone of desert horse-breeding was pushed south, and through crossbreeding with the small disease-resistant indigenous horses of the savanna, new breeds were created. Although the savanna remained an epidemiologically hostile environment for the larger and more desirable horses bred in North Africa, in the high desert and along the desert fringe, Black African states continued to import horses in exchange for slaves into the period of French colonial rule.

The evidence assembled on the horse trade into northern Senegambia raises the difficult issue of the relative quantitative importance of the Atlantic and Saharan/North African slave trades and calls into question the assumption that the Atlantic slave trade was the larger of the two. Most available evidence concerns the Wolof kingdoms of Waalo and Kajoor. It suggests that the volume of slaves exported north into the desert from Waalo in the late seventeenth century was probably at least ten times as great as the volume of slaves exported into the Atlantic slave trade. For both Waalo and Kajoor, this ratio declined during the first half of the eighteenth century as slave exports into the Atlantic markets increased. The second half of the eighteenth century saw an increase in predatory raiding from the desert which produced an additional flow of north-bound slaves. For Waalo and Kajoor – and probably for the other Black African states of northern Senegambia – the flow of slaves north to Saharan and North African markets probably remained the larger of the two export volumes over the eighteenth century. This northward flow of slaves continued strong after the abolition of the Atlantic slave trade and was only shut down with the imposition of French colonial authority.

The Muslim Wandala state controlled large areas of the plains south of Lake Chad between the seventeenth and twentieth centuries a.d. The Wandala also engaged in an extremely complex, and often hostile, set of relations with the inhabitants of the Mandara Mountains, which bordered their state to the south and closely adjoined successive Wandala capitals. These Wandala – montagnard relationships had diverse economic, ritual, political and military aspects. Their complexity appears to be due in large part to the fact that the Wandala and many of the montagnard groups share ethnic origins, and to the violent processes by which differentiation took place and the Wandala gained hegemony on the plains. These processes probably began with a Wandala use of trading advantages to gain access to the Kanuri market system and the subsequent use of products so obtained to expand their dominion.

In 1903 the South African mining industry began recruiting African labor from Central Africa in order to shore up their labor supplies. From the outset, Central African recruitment was problematic, for Central African mine workers died at very high rates. The primary source of Central African mortality was pneumonia. In response to this high mortality the Union government threatened to close down Central African recruitment, a threat which they carried out in 1913. From 1911 to 1933, the mining industry fought to maintain, and then after 1913 to regain access to Central African labor. Of central importance in this struggle were efforts to develop a vaccine against pneumonia. While the mine medical community failed to produce an effective vaccine against pneumonia, the Chamber of Mines successfully employed the promise of a vaccine eventually to regain access to Central African Labor in 1934. The mines achieved this goal by controlling the terrain of discourse on the health of Central African workers, directing attention away from the unhealthy conditions of mine labor and toward the imagined cultural and biological peculiarities of these workers. In doing so the mines constructed a new social category, ‘tropical workers’ or ‘tropicals’. The paper explores the political, economic and intellectual environment within which this cultural construction was created and employed.

Sebatakgomo — a migrant worker-based movement – was founded in 1954 and went on to play a central role in the Sekhukhuneland Revolt of 1958. It was launched from within the ANC, and a number of its leaders were also members of the Communist Party. This article explores the roles played by these wider political movements in the formation of Sebatakgomo. It argues that, while ANC networks and individuals within its central leadership made an important contribution, the rural presence of the ANC was fragmentary in this period and that its central organizational strategies had been effectively checkmated by an increasingly authoritarian state. It suggests that the crucial initial impetus and strategy behind Sebatakgomo came from Communist Party members living in a migrant world and trained in the Party's history and methods of organization. In particular Alpheus Maliba, who led the Zoutpansberg Balemi Association in the northern Transvaal in the early 1940s, provided a mentor and model for Flag Boshielo, who was the driving force in the establishment of Sebatakgomo. The article also suggests that the history of Sebatakgomo provides an example of the impact of Communist Party activists in transforming the ANC into a mass organization in the early 1950s.