In the late sixties and early seventies Africanist trends elsewhere on the continent, backed by the path-breaking work of Jan Vansina18 on the usefulness of oral traditions as historical sources, prompted a new interest in precolonial southern Africa. As Saunders shows, the work of Omer-Cooper and Leonard Thompson initiated this development. Their endeavours were soon overtaken by a new generation of students working within a materialist paradigm, and trained mostly, though not exclusively, at two institutions, the University of California, Los Angeles, and the School of Oriental and African Studies, University of London. The work of these scholars featured in forums such as Shula Marks' interdisciplinary seminar on the societies of southern Africa, at the Institute of Commonwealth Studies.
In July 1976, this interest led to the holding of a workshop on 'Pre-capitalist Social Formations and Colonial Penetration in Southern Africa' at the National University of Lesotho, Roma.19 As the workshop title makes clear, the participants were concerned to investigate the nature of pre-capitalist economies, and were particularly interested in questions of social stratification and forms of exploitation, as well as how these aspects of precolonial societies were transformed by the penetration of mercantile, and later industrial, capitalism. The publication in 1975 of Barry Hindess and Paul Hirst's Pre-capitalist Modes of Production20 powerfully influenced the papers and discussion at the conference. The Lesotho conference set in place many of the major theoretical ideas which informed a crop of doctoral theses which were completed in the middle and late 1970s dealing with the economies and histories of a range of precolonial polities such as the Zulu and Swazi kingdoms, the Pedi paramountcy, and so on.21
A follow-up conference was held in Lesotho in 1977, and, in the same year the University of Natal, Pietermaritzburg, hosted a regionally specific conference. The prime purpose of this latter meeting was 'to provide a forum for students of Zulu history working in southern Africa to discuss the nature of the Zulu political economy before 1879'.22 The Pietermaritzburg workshop also drew inspiration from the latest developments within Marxist thinking. Earlier that year, Claude Meillassoux had given a seminar in Pietermaritzburg, and the workshop title, 'Production and Reproduction in the Zulu Kingdom', drew attention to the central concepts of the French Marxist anthropologists. The tight regional focus of the Pietermaritzburg conference, and the conscious effort by the organizers to solicit papers from outside the discipline of history, reflected an awareness of a need for new sources of evidence within the field.
A further conference, also concerned with themes of differentiation, production and exchange, focused on the precolonial history of Nguni-speakers more broadly. It was held at Rhodes University, Grahamstown in 1979, and the conference proceedings were edited by Jeff Peires and published under the title Before and After Shaka.23 These papers reflected a growing concern about the inability of the available evidence to show contradictions, conflict and social cleavages, the dominant themes of a Marxist conception of history. Evidence difficulties, coupled with an increasing sense of the irrelevance of precolonial history in apartheid South Africa, led a number of the scholars hitherto involved in the study of pre-capitalist societies to turn their attention to the history of twentieth-century southern Africa. The emergence of the Wits History Workshop – with its focus on industrialisation and the impact of capitalism – as the pre-eminent historical forum of the period, was an important marker of this trend.
As a result, little new research into the precolonial past was undertaken in the early 1980s. The major publications of the period drew on the seminars and doctoral theses of the previous decade. Marks and Atmore reproduced as Economy and Society in Pre-industrial South Africa a selection of the seminar papers delivered in the 1970s, and a number of the doctoral theses appeared as published monographs.24 It was to be seven years before the next meeting of scholars working on the precolonial past and the presentation of fresh perspectives and research.
This happened at the 1986 'Workshop on Precolonial History' organised by the Centre for African Studies at the University of Cape Town, which was attended by historians, anthropologists, linguists and archaeologists.25 The papers and the conference debates revealed a shift away from the strong political economy approach of the materialist scholars of the 1970s to an examination of questions of ideology, and a concern with methodology. Increasingly influenced by literary theory, a new approach was emerging at the conference and in other work produced at this time, which raised fundamental questions about historians' reading of their evidence, and about the genealogies of their sources.26 In addition a number of papers took up questions of the public presentation of the precolonial past in textbooks and popular histories.
This brief review of the development of precolonial studies in the 1970s and 1980s reveals that Cobbing's early papers were presented in a hiatus between the Rhodes and Cape Town conferences, at a time when interest among historians had swung away from the precolonial past to a concern with the making of the apartheid state. However, the new directions which emerged at the Cape Town conference interlocked with certain of the concerns of the Cobbing critique, notably an interest in how received versions of the past came into place. This congruence of interests underlay the subsequent attention paid to Cobbing's arguments and ultimately, the holding of the Wits colloquium in 1991.
The colloquium itself was preceded by an upsurge of public interest in Zulu history and invocations of the mfecane (widely reported in the South African press) in terms of which contemporary political transformations and upheavals were likened to a modern mfecane.27 The methodological concerns of the academy were thus further complemented by the pressing question in the public domain of how the role of the history of southern Africa before European domination, and narratives of first contact, are drawn on as metaphors for current political developments. Interest in the debate over the mfecane in 1991 reflected the intersection of the Cobbing critique with, on the one hand, specific developments within precolonial historiography, and, on the other hand, contemporary political developments.
This convergence allowed the set of debates which characterised the colloquium to take place when they did, and in the form that they did, even though Cobbing's critique had first been made much earlier, and even though he had sought a different form for his intervention. I use the phrase 'set of debates' for not all the contributors would necessarily agree what those debates are, nor is there any consensus on their ranking in terms of importance.
Generally, it is assumed that in order to have an historical debate, the participants and adversaries must agree on what is at issue. They have to agree broadly on how arguments are properly made, on how sources should be read, on what constitutes evidence or proof, and on what might be persuasive. Usually, in a debate about events in temporally distant times, the participants must agree to share the same imagined idea of what societies were like in remote times, before an argument can be made about the causes of changes in such societies, or about the reconstruction of those causes later in time.28
Readers of the various essays in the book are likely to find such agreements elusive, and occasionally even absent. For this reason, it is not possible to see the book as centred on a single debate, or to regard essays that say different things about particular events as necessarily opposed. This Introduction, and the contextualising essays which precede each of the three sections into which the book is divided, point up some of the areas of debate and difference. Doubtless, readers will find others, and will be able to explore for themselves the significance of points of abrasion between the various studies.
Julian Cobbing argues that the debate is whether the notion of the mfecane should be retained within, or 'jettisoned' from, southern African historiography. Other scholars argue about whether the concept of the mfecane carries the meanings within the historiography which Cobbing ascribes to it. There are debates amongst the essays in this volume about the very issues which are usually agreed upon in the setting of the parameters of an historical debate. The 'rules' of historical reconstruction are themselves contested. Where, for one contributor, a dissembling colonial official is an unreliable or 'bad' source who cannot be used, for another scholar, the same source – indeed, the official's