There are signs that the familiar bias towards research on the Later Stone Age of the southern Sahara and Sahel and on the Iron Age of other parts of West Africa is beginning to be redressed, although Nigeria and Ghana still furnish the vast majority of dates from the southern coastal states. Historical and protohistorical archaeology is becoming increasingly popular, and some areas, for example southern Mali, are receiving serious attention for the first time. Dates are now appearing in large numbers; this paper reports on 355 new radiocarbon and thermoluminescence dates.

Multi-disciplinary research has helped to clarify the Holocene environmental and cultural-stratigraphic sequence in several areas, including the Aïr and Ténéré in Niger and coastal Mauretania and Senegal. There is now strong evidence of a centre of copper metallurgy in the Azelik region of Niger in the 1st millennium b.c. which may well have facilitated the spread of iron working to Nigeria. At the same time, new dates from Mali may indicate a second diffusion route through the western Sahara. In Nigeria, new and early dates for the Nok culture are supported by thermoluminescence dating. Recent work on the later Metal Age of Senegal has permitted the description of four partially overlapping zones, and should soon lead to understanding of their internal and external relationships. Similarly, a substantial body of data has been accumulated, though not yet published, on the Iron Age of the Inland Niger delta. In the more southerly parts of West Africa, several new dates from just north of the forest refer to the emergence of the earliest Akan groups in Ghana and of Oyo in Nigeria. Previously reported dates from Ife and Benin have now been supplemented by a series of thermoluminescence dates.

Recent archaeological research conducted west of Lake Turkana, Kenya has shed new light on the prehistory of eastern Cushitic and Nilotic speakers in East Africa. The Namoratunga cemetery and rock art sites, dated to about 300 B.C., are clearly related to the prehistory of Eastern Cushitic speakers. The newly defined Turkwell cultural tradition, dated to the first millennium a.d., is associated with eastern Nilotic prehistory. Lopoy, a large lakeside fishing and pastoralist settlement, is discussed in terms of eastern Nilotic prehistory. The archaeological data agrees with the independent findings of historical linguistics.

The articles in this special issue are revised versions of papers delivered in September 1978 at the African Studies Association of the United Kingdom conference on ‘Whites in Africa – Past, Present and Future’. The choice of conference topic aroused some controversy. Addressing himself to the ‘Pink ASAUK’, one correspondent enclosed a cutting from West Africa in which news of the forthcoming conference was ominously included on the same page as a report on ‘Freetown's Rubbish Plan’. The writer made two points: ‘Human beings are citizens, not colours or numbers as your hollow minds suggest’; ‘Now that Africa is nearly free from imperialism and exploitation…dishonest intellectuals from British universities are conspiring to draw a new map of Africa.’

From various kinds of evidence it can now be argued that agriculture in Ethiopia and the Horn was quite ancient, originating as much as 7,000 or more years ago, and that its development owed nothing to South Arabian inspiration. Moreover, the inventions of grain cultivation in particular, both in Ethiopia and separately in the Near East, seem rooted in a single, still earlier subsistence invention of North-east Africa, the intensive utilization of wild grains, beginning probably by or before 13,000 b.c. The correlation of linguistic evidence with archaeology suggests that this food-collecting innovation may have been the work of early Afroasiatic-speaking communities and may have constituted the particular economic advantage which gave impetus to the first stages of Afroasiatic expansion into Ethiopia and the Horn, the Sahara and North Africa, and parts of the Near East.

The historiography of Hausaland has laboured under a strong tradition of orthodoxy which recent secondary works have inherited from the more-or-less primary oral-cum-written sources. General cultural evidence (linguistic, ethnographic and archaeological) has been regarded as subsidiary, so that its potential for reconceptualization and for critical reevaluation of the conventional sources and orthodox interpretations has been missed. Instead, antiquarian approaches have been encouraged. Thus the view has persisted that Hausa as a cultural and linguistic entity has an antiquity running to several millennia, and also that it originated in the Sahara or around Aïr, whence it was pushed southward by desiccation or by Tuareg nomads. Contrarily, the clear message of linguistic geography and of Hausa's place within the Chadic family is that Hausa. expanded from east to west across the savanna belt of northern Nigeria. And the relative homogeneity of the language and culture within this vast zone indicates that the spread is quite recent (within the present millennium, say). It would have involved some assimilation, of previously settled peoples of the northern Nigerian plains, most of whom wouldl have spoken languages of the ‘Plateau’ division of Greenberg's Benue–Congo subfamily, of Niger-Congo.

This Hausaization, as it proceeded from its old bases in eastern Hausaland, would have been both a cultural and an ecological process, through which woodland would have been converted into more open and continuous savanna to support grain-cultivation and a denser peasant population. This process would have reached western Hausaland (Zamfara and Kebbi) around the middle of this millennium. Cattle – and Fulani herdsmen – would in time have played an important role in this cultural ecology (and in restricting the tsetse zones).

The old theory of a northern origin for the Hausa is bound up with the problem of Gobir in north-western Hausaland. Gobir's claim to be one of the original seven kingdoms (Hausa bakwai) is probably a late invention. Moreover, the common assumption that Gobirawa Hausa migrated from Aïr seems to derive from a misinterpretation of the written sources.

Finally the bakwai legends are reconsidered. Despite the scepticism of some modern critics, the legends appear to reflect, albeit in idealized form, a real historical development. They represent a foundation charter for the Hausa as a multi-state ethnicity, and enshrine the vague memory of how Hausaland and ‘Hausaness’ began from a series of small centres and hill-bases on its eastern side. Thus the interesting argument of Abdullahi Smith, that the Hausa people emerged long before state systems arose among them, is disputed. Rather, these should be seen as two facets of a single process during the present millennium.

In earlier publications the present writer was at pains to emphasize that the state of Sinnār was not a mere confederation of nomadic Arab tribes; rather, it rested upon a firm agricultural base and was governed bureaucratically, while incorporating an ingeniously conceived system of nobility. This interpretation, though valid as far as it went, rested largely on evidence from riverain Sinnār and left little room for nomads. The present study, based primarily upon sources from the Sudanese rainlands, proposes that such a view of the place of herdsmen in the society of Sinnār is not well founded, and indeed that the habit of thought which perceives a sharp and enduring distinction between the peoples of ‘ the steppe and the sown’ – however appropriate in other contexts – does not pertain to Sinnār. The vision of rainland life which emerges from the sources here examined reveals a single society of herdsmen and cultivators, a society in perpetual metamorphosis within a framework of possibilities limited by ecology and custom. Ruling houses came and went; tribes grew, sundered and re-formed into new polities. Individuals and groups migrated freely, occasionally over vast distances, and changed their mode of livelihood whenever the opportunity for herding or the necessity for cultivating presented itself. All were subordinate to the state, whose continuity of authority in the long run overshadowed the more ephemeral corporate realities of the moment.

For a brief period between about 1878 and 1886, European traders in Eastern Pondoland played an important role in the politics of the chieftaincy. In these years the formerly amicable relationship between the Mpondo chiefs and the Cape Colony broke down. An expansionist colonial government confiscated the Mpondo port at St Johns, took control of an area claimed by the paramountcy, and threatened the political independence of the chiefs. The traders, especially a few larger wholesale firms, allied themselves with the paramount in his struggle to regain his land and port. Some traders feared that colonial encroachments would affect the concessions they had obtained from the paramount and that customs duties charged by the Cape at Port St Johns would undermine their competitiveness. Others, faced with an acute trading depression in 1882–3, sought to increase their share of the Pondoland market and hoped that conflict between the Cape and the Mpondo would stimulate a new boom. The chiefs, on their side, were receptive to the political overtures of the traders. As traders penetrated the area, they began to lose control over production and became increasingly dependent on taxing trade and granting concessions. They had lost confidence in missionary scribes and looked to the traders to provide secretarial skills. Various groups of traders could also give aid in the prosecution of Mpondo diplomacy by supplying firearms, opening negotiations with the imperial authorities, and establishing a new port. The alliance was formed at a specific moment in the penetration of merchant capital into an independent African polity. It serves to illustrate that some traders were prepared to act independently of the colonial powers in order to protect their interests.

This article deals with climatic reconstruction over a period of centuries, on the basis of indirect evidence found in historical and geographical sources. Histories, archives, local chronicles and journals of travellers and settlers contain references to lakes, landscapes, famines, droughts and floods, as well as occasional descriptions of climate and meteorological measurements. Such information can be combined with evidence from geology, palynology or the study of tree-rings to support hypotheses regarding climate and environment several centuries ago.

This methodology is here described and used to reconstruct the trend of rainfall fluctuation in Africa over the past millennium. Two approaches are considered: the one seeks to determine absolute variation (thus assessing whether particular episodes were wetter or drier than today); the other focuses on short-term climatic anomalies (e.g. droughts) in which rainfall differed from the mean prevailing at the time, without seeking to relate them to present conditions.

The results obtained from this study suggest that during the past millennium there have been two periods of relatively wet conditions in the semi-arid regions south of the Sahara: between the eighth and fourteenth centuries, and between the sixteenth and eighteenth centuries. Evidence for these episodes, and for synchronous fluctuations elsewhere in Africa is presented in the text.

This article attempts to assess the importance of Asante influence in the consolidation of the Fante in the eighteenth century. It shows at the outset that mis-translation of Willem Bosman's early eighteenth-century work has led to an erroneous conception of the Fante polities as being incapable of acting on common interests. Between 1700 and 1724 the Fante conquered surrounding states such as Asebu, Fetu, Cabesterra, Acron (Gomoa), and pockets of Etsi, and established a system of domination over trade on the coast thereafter. It is suggested that this development was not the result of Asante activities on the coast, but rather was linked with the presence of the English, who supplied firearms to the Fante. The Fante were constrained to maintain good relations with Asante because they were determined to be the exclusive middlemen for the slaves which came down from Asante and beyond. At the same time, however, the Fante, like other coastal states (e.g. Aowin, Wasa, Denkyera and Akyem), at times feared invasion by Asante and therefore protected themselves through a system of local political alliances.

The sudden expulsion of the Xhosa across the Fish River in 1811–12 created a practical and conceptual crisis which the traditional political authorities were unable to resolve. Two commoners, Nxele and Ntsikana, emerged in this vacuum, each proposing his own solution to the problems posed by the white irruption. Although these responses were religious responses, they were neither irrational nor incomprehensible. Xhosa religion had long functioned as an instrument for the control of the material world. By incorporating selected Christian concepts with the Xhosa world-view, Nxele and Ntsikana were able to provide the Xhosa with acceptable explanations of past events and prescriptions for future action.

Nxele urged resistance and Ntsikana preached submission, but an examination of their personal histories shows that these final conclusions were more the product of exterior pressure than interior revelation. It may be suggested that the future reputations of the two men, like their past actions, will be determined more by the popular mood than by anything they themselves did or said.

All the many definitions and descriptions of the akori bead published in the last five hundred years have proved inadequate. They appear either pedantically narrow, so that only a hundred years later or a hundred miles away beads also unquestionably called akori by West Africans by their very existence disprove such ‘precise’ characteristics, or, the other extreme, so broad as to be historically useless and logically unsatisfactory. Since all the akori records must be viewed as historical sources more or less reflecting the real situation in the past, we can hardly blame their authors for the akori puzzle; there must have been something wrong in the naming mechanism, in the factual relationship between the word and the thing. It is the main point of this article that the akori phenomenon originated in the existence of various types of natural glass in North and West Africa. An archetypal human attraction to this strange and rare substance can be observed in several culturally very different areas of the world. Understandably, at a later stage people made talismans and beads of this presumably protective material, yet the primeval veneration paid only to the mysterious matter, now under a more sophisticated but basically accidental form, remained the same. It was primarily this matter alone which West Africans tried to discover, test and obtain in European-imported artificial glass after 1500. The complex akori ‘problem’ is a telling and fascinating historical and logical expression of contact between different cultures.

By drawing on the current Marxist debate about the nature of the capitalist state, this article argues that the colonial state was obliged to be more interventionist than the mature capitalist state in its attempts to manage the economy, since colonies were distinguished by the way in which they articulated capitalism to local modes of production. This posed severe problems of social control, since the capitalist sector required the preservation of indigenous social institutions while also extracting resources from them. In early colonial Kenya this problem was mitigated by a rough compatibility between the needs of settler capital and the patronage exercised by African chiefs within a peasant sector which was expanded to solve the colonial administration's initial need for peace and revenue. The peasant sector was not destroyed, rather it was represented in the state, which never ceased thereafter to be plagued by the conflicts between the two modes of production over which it presided.

One of the advantages of looking at white society in Southern Rhodesia is that it was a very simple society by contrast to South Africa and that its essential structures show up very clearly. This was certainly true of the business of producing models of African societies, customs and conduct. In South Africa there grew up a cluster of intellectual ‘experts’ on Africans – including a distinguished group of anthropologists. In South Africa, therefore, the connexion between the requirements of the white economy and the most influential models of African societies was complex and indirect. In Rhodesia there was no such tradition of indigenous social science. As a contributor to the Native Affairs Department Annual admitted in 1956, ‘of all areas in Africa Southern Rhodesia was the most backward in anthropological knowledge of its own indigenous peoples’. In Southern Rhodesia the men who administered Africans, mobilized them for employment and kept them working were also the men who produced the authorized versions of the African past, of African customs and of African ‘personality’.

The much-researched history of Bulozi in the late nineteenth century is re-assessed here from a marxist viewpoint. A hereditary class of landlords owned the principal means of production and extracted rent in labour services and in kind from the direct producers. Tribute was also paid to the king and royal family, in recognition of the ultimate royal ownership of the means of production. Surplus was employed to increase leisure time, to indulge in conspicuous consumption, to raise the level of the productive forces through investment, and to maintain repressive political and ideological apparatuses. The main weight of oppression fell on the slaves, with free commoners in an ambivalent position. Slaves were both economically exploited and socially discriminated against, and their position was analogous but by no means identical to that of European serfs. The evolution of class struggle was greatly affected by articulation with capitalism and colonialism. The first contacts stimulated exploitation by providing a larger market and more effective means of repression, but the colonial state later intervened to abolish slavery in order to intensify the flow of migrant labour to the capitalist heartlands of southern Africa.

Successive Europeans in south-east Tanzania looked for an ethnically based political authority under whom to live or with whom to work. Bishop Edward Steere of the Universities' Mission to Central Africa predicted the existence of very large tribal and linguistic ‘nations’ when this turned out not to be so, the UMCA missionaries who had settled at Masasi sought anxiously for some influential chief who could be represented as heading an ethnic polity; first German and then British administrators over-readily assumed that the chiefs whom they installed as akidas did in fact represent such ethnicities; finally, in the late 1920s, the British instituted historical research prior to the establishment of Indirect Rule, which was intended to reflect the ethnic and political complexity of the region. This European preoccupation with ethnicity bore little relation to the actualities of the region, which from the nineteenth-century incursions of the Yao, Makua and Makonde had constituted a mosaic of small, autonomous and ethnically mixed groupings. Nevertheless, certain African adventurers were able to take advantage of the European need for allies to build up their power, to become recognised as ‘chiefs’, and ultimately to become regarded as leaders of ethnicities. This was the case with Matola I and Matola II of Newala who between them developed their polity from a very small scattering of huts to a large and prosperous paramountcy. Within the Matola polity various social and cultural processes were at work to produce a common sense of identity, but these processes had not fully eroded the marks of the varying ethnic identities of those who belonged or submitted to the polity. The Indirect Rule inquiries, therefore, with their fanatical emphasis upon ethnicity as the only legitimate base for political authority had the result of dismantling the Matola polity and thereby destroying the only effective local nucleus of political consolidation.

Cupido Kakkerlak's story provides a concrete example of Khoi experience under the impact of colonization at the beginning of the nineteenth century. During his first forty years or so he lived on Boer farms, learned a sawyer's skills, accumulated a little property and reared a family. In 1800, probably as a result of frontier disturbances at the time, he went to the village of Graaff-Reinet. There, in 1801, he met missionaries of the London Missionary Society and was converted. Casting his lot with the mission, he moved with his family to Algoa Bay and was based at Bethelsdorp until 1815. During this period he practised his trade as a sawyer at the same time as he gained prominence in mission work. In 1813 he served as John Campbell's ‘travelling director’ during a trip to the interior that lasted almost nine months. Campbell's proposals – that a number of new stations be established – made heavy demands on mission personnel and other resources. Six ‘native assistants’ were appointed, one of whom was Cupido. In 1817, after a short sojourn among the Griqua, he undertook a mission to the still nomadic Kora near the Harts River. Six years later, when difficulties both in and outside the mission society had multiplied, his services were abruptly terminated. He was then over sixty years of age. For frontier Khoi, hopeful of a new dispensation in the wake of the 1799–1802 war, the L.M.S. missionaries had provided an undreamt-of opportunity. In the interaction between missionary and Khoi, in the first stages of the mission project, Cupido played a leading part.