The establishment of a journal specifically concerned with method in African history is to be welcomed. However, the early issues of History in Africa have demonstrated that the content of the term ‘method’ is itself at stake. The great majority of contributions to date have seized on a narrow and limiting conception of method as the development of techniques of collecting and evaluating data. The necessity of such techniques is not in question, but they are subordinate to, and indeed partially determined by, a broader and more fundamental conception of method as the principles of investigation and explanation in scientific practice. There are historians who do not regard the production of historical knowledge as a scientific enterprise, hence subject to certain theoretical demands, and they would not want to. Accordingly, they need not read on, but we are confident that there are others who are interested in method in the second sense and who may also have noticed its virtual absence in the pages of this journal.

On the other hand, it would be disingenuous to imply that a common interest in method in the broader and more fundamental sense is sufficient ground for agreement. Our argument in what follows derives from an understanding of historical materialism that has nothing in common with the stereotyped views held by it bourgeois critics. Our central concern is with method as the principles of constructing scientific explanations. But what is to be explained? We attempt to show that method necessarily starts with the correct posing of questions, as well as bearing on their investigation.

For thirty-eight days in August and September of 1607, two ships of an English expedition to the East Indies lay in the Sierra Leone estuary, mainly in order that the sailors might recover from the scurvy developed in a four-months' voyage by drinking lime-water prepared from local fruits. When the ships returned from the East, journals of the expedition became available. Richard Hakluyt collected three, including that of the expedition's commander, William Keeling. These passed eventually to Samuel Purchas, who in 1625 published extracts from them and from other journals independently collected. In Purchas' publication the Sierra Leone stage of the voyage is referred to in extracts from two writings: Keeling's daily journal, and the “large Journall” of the merchant William Finch -- not in fact a daily journal but a detailed, accurate, and very valuable synoptic description. While Finch's “Remembrances touching Sierra Leone” appears to have been published in full, Purchas cut the relevant section of Keeling's journal. He stated “I have beene bold to so shorten as to express only the most necessary Observations for Sea or Land Affaires.” The original of Keeling's journal has disappeared, and therefore, with the exception of some entries to be shortly described, what Purchas cut cannot now be reinstated. However, four other journals of the voyage which contain material on Sierra Leone have survived. The one with the briefest reference to Sierra Leone was published in 1878, and extracts from the Sierra Leone sections of the other three appeared in print in 1877 and 1923. The present paper makes use of a transcription of the whole of the relevant sections in the three manuscripts, a transcription which I hope to publish soon elsewhere.

As heavily as historians must rely on orally-derived data for their study of the African past, historians of Oceania are far more in thrall to such materials in attempting to reconstruct the history of the various Pacific island groups. Although archeology and historical linguistics can sometimes help to provide broad sequences and interrelationships as well as evidence concerning origins, neither can, of course, provide circumstantial local detail or close dating. Oral traditions, often supported by genealogies of sometimes extraordinary length and complexity, have been collected in all parts of the Pacific almost since the time of Cook, but the latter part of the nineteenth century was a period of particularly feverish activity. The result is a vast body of material, much of it still in manuscript form. Of this corpus far more relates to the Maori people of New Zealand than to the inhabitants of any other island group.

In the course of the first half of this century a homogenized orthodox view of New Zealand's more remote past developed -- an interpretation based on three pivotal events, each of which came to be dated calendrically by means of Maori genealogies. The first was the arrival of the “discoverer” of New Zealand, one Kupe, who was dated to ca. 950. Then, two centuries later, came Toi and his companions. Finally, so this version goes, the so-called Great Fleet, comprising about seven large canoes (the number varies slightly) arrived in about 1350, and New Zealand began to be well and truly peopled by Maori.

Oral traditions are the common patrimony of a varied group of scholars. Although professional historians embraced African traditions only in the late 1950s, local -- and usually amateur -- historians had been writing from them throughout the colonial period and occasionally before. Meanwhile, students of anthropology and comparative religion from Frazer on have been studying the same traditions from very different perspectives. They have pointed out the common motifs recurring within the genre in widely scattered and disparate areas, and have related myth to its functions in present-day society, watching how it is molded to fit changing realities.

At the same time that academic historians at last began to overcome their doubts about the historical value of African oral traditions and began to use them, a French anthropologist was propounding a new approach to myth, a principle he called “structuralism.” Claude Lévi-Strauss argued that even the seemingly most direct and least stereotyped American Indian traditions he examined were actually philosophical expositions of world-view and religious belief encoded in symbolism. This article tests the structuralist approach through a detailed critique of one particular example of its application.

Luc de Heusch, a Belgian disciple of Lévi-Strauss, has recently studied the traditions of Lunda state formation as myth in Le roi ivre, ou l'origine de l'Etat. His aim is to discover the univers intellectuel of the Lunda and the other peoples of the Central African savanna, but his work touches the historian in two crucial respects. First, he highlights dangers in the path of the naive and helps paint parts of the larger human dimension in which historical events and historical records have existed.

Here Reindorf described the events leading to the defeat and death of MacCarthy. Unlike Ellis, his account differed substantially -- both in detail and emphasis -- from earlier (mostly British) ones. The panegyric verse with which he closed the chapter was, however, quite typical of the Fante-Wesleyan response to this Governor. With the exception of a few details (and, of course, the vernacular) this chapter is fairly and fully reproduced, including even the panegyric.

In a detailed chapter covering the period from July, 1824 to July, 1826, Reindorf described the siege of Cape Coast and the circumstances inducing the Asante to retire to Kumase. He then traced the recovery and reorganization of the Asante army under Osei Yaw Akoto and the efforts of Europeans and Accra businessmen to revive the southern alliance. In his own work Cruickshank was concerned only with the siege and withdrawal and on these parts Reindorf relied for some events of the siege. But, rather characteristically, he emphasized the Gã role in certain of these (197/190).

It has been argued that one of the more salient features of Reindorf's work is its palpable orientation toward the Gãs. The extent to which this is valid criticism will be discussed below but, rather ironically, the evidence in this chapter which might support this thesis has been largely omitted from History II. For example, in the second sentence of the chapter Reindorf had labeled the Gas as “bull-dogs of the European governments” for fighting in no fewer than five engagements between March of 1823 and June of 1824, but this was omitted in History II (196/[189]).

The following exploration of social scientific thought as applied to Africa is an exercise in ethnography, not in debunking. At one level the argument, referring mainly to social change in western Zaire in the last hundred years, offers a descriptive interpretation which I assert is better than others. At another level I seek to establish what kind of phenomenon, in an ethnographic sense, historiography really is. At this level, I apply the methods of social science to a body of work in “social science” which is, in fact, myth, and which itself deals, in part, with a body of “myth” which is, in fact, social science.

In “African Traditional Thought and Western Science,” Robin Horton first shows the ways in which the first is like the second in producing model-building explanatory schemes, and then argues that the difference between the two is essentially that “in traditional cultures there is no developed awareness of alternatives to the established body of theoretical tenets; whereas in scientifically oriented cultures, such an awareness is highly developed ” He elaborates this difference between “the open and closed predicaments” in terms of such factors as unreflective versus reflective thinking, mixed versus segregated motives, absence versus presence of experimental method; these factors he finds superior in explanatory power to “most of the well-worn dichotomies used to conceptualize the difference between scientific and tradition religious thought. Intellectual versus emotional; rational versus mystical; reality-oriented versus fantasy-oriented …”

Only occasionally does Horton refer to social science, however; he seems to assimilate it to natural science.

Recently Edward Steinhart has discussed the assassination of the British colonial officer Harry St. G. Galt in Uganda in 1905. Steinhart's account is based almost entirely on the Entebbe Secretariat Archives and on interviews he conducted. It focuses on the political intrigue brought to light by the murder enquiry, but much of the interest derives from the mystery surrounding Galt's death. In our opinion, Steinhart comes very close to the truth when he writes of his discovery that a conclave of chiefs had been held on the night of the murder, and he begins to suspect that Kahaya, ruler of Ankole, as well as Igumira, a reactionary chief, were involved in a plot which was somehow intended to discredit Nuwa Mbaguta, enganzi (Prime Minister) of Ankole, and Igumira's rival. Steinhart is then deflected because he thinks that the evidence for a conspiracy rests only on this event, and on statements from Sir Charles Gasyonga, Kahaya's successor. We think we can come much closer to a solution to this mystery by using archival material which has hitherto escaped notice. The Diary of the White Fathers Mission at Mbarara, the capital of Ankole, refers to the events of this year, and for the first time offers a plausible motive for the assassination. The White Fathers Archives in Rome have been open to scholars for some years, but it seems that not all researchers are aware of the wealth of material which they contain. Mission archives assume particular importance in the case of Uganda, which is now virtually closed to research, and we will try to illustrate their value by reference to this incident.

Historical studies of Kivu are still in their very infancy. Recent work has been carried out in Bufulero, Bushi, Buhavu, and Bunande, but lacking the results of these studies, historians working from published materials have very few sources at their disposal. Existing sources include works by Colle, Moeller, Willame, and Cuypers, with the latter two based primarily on the former, at least in their historical dimensions. Because the sources are so few and are essentially similar, little critical attention has been given them; by constant citation and repetition they have become hallowed as truth and used as a basis for teaching and university theses. By this process such essentially colonial interpretations have become entrenched in the historical ontology of the region. This paper proposes to review some of the written sources in light of current research in the region, by first presenting certain themes which appear to have guided earlier historical inquiry and then discussing the works of these four influential authors in light of these themes.

The first attempts to record historical traditions in the Kivu area were influenced by earlier studies of Rwanda which emphasized the centralized and hierarchical nature of the Rwandan state. Many of the early missionaries and Zairean priests in Kivu, men to whom contemporary researchers owe much for their accumulated sources, had close contacts with the seminaries and published work in Rwanda. In most historical works, Rwanda was seen as the end development for other states in the region, and prominence was given to those historical factors which were assumed to have had a common impact throughout the area.

Work on volume 7 of the Cambridge History of Africa, covering the years 1905-1940, has obliged me to reflect on the nature of the literature for this period in African history. The past twenty years or so have of course seen some notable research on the subject, and much of this is now published. A common theme in this work, as in the study of earlier African history, has been an emphasis on African initiative and enterprise, whether among clerks and teachers, cocoa farmers or mineworkers. More recently, there has been a renewal of interest (often related to theories about “underdevelopment”) in the external forces which set limits to the scope and success of African initiatives. It has once again become acceptable in Africanist circles to study white farmers, trading companies, mine corporations, finance houses, administrative hierarchies, legislative councils, or mission headquarters. Thus the Cambridge History should now be able to achieve a more balanced appraisal of both local and alien factors in the colonial order than would have been possible even five years ago.

The question I want to raise here is whether, for the purposes of such a volume, it is enough to think in terms of synthesizing the work of professional historians of Africa over the past two decades, or whether recent shifts of emphasis have exposed topics and questions on which we must still look for guidance to works written when colonial rule was still a living reality and African history, as an academic subject, had not yet been invented.

During the last decade the question of counterfactual arguments has attracted a good deal of attention both in philosophy and history. A recent example, which concentrates on the logics of counterfactual analysis in history, is T.O. Climo and P.G.A. Howells' “Possible worlds in historical explanation.” Further progress in this field probably depends both on a development of the purely logical issues involved and an analysis of the actual usage of counterfactuals in the language of practicing historians. My own approach belongs primarily to the latter category. I shall consider examples of counterfactual arguments in two hotly debated fields, first the partition of the African continent and, second, the effects of colonial rule. This approach will provide examples of the function of counterfactuals both in causal analysis and in historical evaluations. My primary aim is to establish a categorization of different usages, but the opportunity will also be taken to discuss in a general manner criteria for the legitimate use of counterfactual argument in history. In this connection I should emphasize my lack of knowledge in formal logic, except that provided by my two Norwegian collegues, Ottar Dahl, and Jon Elster, on whom I rely heavily for the more theoretical parts of this paper.

Historical analysis is preoccupied with causes, and in causal analysis there is a particular urge to identify socalled “sufficient” and “necessary” causes. In propositions about such causes some counterfactual assumptions are logically implicit. The clearest example of a necessary cause or precondition for European expansion in Africa is that of technology. To take an early and typical example, Holland Rose maintained that it was an “essential condition” of colonization that “mechanical appliances should be available for the overcoming of natural obstacles.” The implicit counterfactual is that without technology, never colonization. Counterfactuals of this kind scarcely attract attention precisely because of their obvious legitimacy.

For the people of the Third World, nationhood in the twentieth century frequently demands the solution of a dual complexity: on the one hand, a search for an identity long suppressed by colonialism; and on the other hand, an effort to come to terms with the problems engendered by often violent entry into modernity. The challenge presented by this quest for self-identity and self-determination is at once paradoxical and parallel. Under foreign domination the colonized is denied his past, his real history; in addition he is forbidden any role in the making of his future. No wonder then that modern revolutionary movements stress the necessity for recovering the colonized's indigenous background — his roots — as a necessary corollary to independence and the eradication of colonialism.

In a sense the student of history also shares in the problem. Particularly is this true of students in North African history for in the pursuit of knowledge about the Maghrib's past and present, where does one turn historiographically? For purposes of organization only, three prospective “schools” of historical analysis are considered here: colonialist, nationalist, and Western. In suggesting these three “schools,” I make no attempt to be inclusive; many other variations are possible. The model used here is presented simply as a guide to complement the discussion which follows.

The aspiring historian may first seek truth in history as written by the colonialist. In most instances, however, this will prove inadequate because the colonizer usually relegates the pre-colonial past to obscurity; history under dynamic colonialism or protectionism is inevitably seen as forever enlightened whereas independence is chaotic and despotic, if the colonizer bothers to write about the reborn nation at all.

The Jos Plateau comprises an area of approximately 2500 square miles in the north-central part of Nigeria. It includes a high plain, interspersed with granite hills and is bounded by a broken scarp some 1500 to 3000 feet in height. Historically as well as geographically, the Plateau has stood apart from the neighboring lower plains region. In the nineteenth century, the Muslim emirate of Bauchi controlled nearly all the territory surrounding the Plateau, but inhabitants of the Plateau successfully repelled intermittent Bauchi Incursions and maintained their independence until the arrival of the British at the beginning of the twentieth century.

Most of the literature dealing with the Plateau has focused on this theme of resistance and a number of interrelated topics. As it happens, such studies nearly always emphasize the isolated nature of the region, whether cultural or economic. As a result, a stereotyped view of the Plateau has emerged, a view which characterizes the area as a “hill refuge,” settled by small groups who lived in the most inaccessible reaches, were economically self-sufficient, and who at best maintained minimal links with each other. The isolated nature of these groups is seen as a major cause of their intrinsically conservative nature and predisposition to resist change. According to the stereotype, the people of the Plateau remain a useful subject of research since they continue to exhibit many of the “archaic” forms of social, political and economic organization that may have been prevalent throughout Africa thousands of years ago.

[This continues the comparison of texts of the English and Dutch versions of Bosman. For earlier instalments see History in Africa 2(1975, pp. 185-216; 3(1976), pp. 91-126; 4(1977), pp. 247-273. Procedural matters are discussed in the first instalment, to which the reader is referred.]

Traditions of origin of most of the prominent Gambian Mandinka lineages tend to be bodies of myth containing small, but often recognizable, skeletons of historical truth. Members of these lineages typically think of one of several mythical, sometimes composite, figures as their ancestors. These figures are invariably great leaders, if not of western migrations with large followings, then of successful military expeditions in the west for the Mali empire. Usually, these traditions serve a particular purpose. The ancestral figures symbolize the lineages' ties, real or imagined, to the Mandinka homelands and the center of Mandinka culture on the upper Niger River. The stories of their ancestors' early arrival or successful conquests in the western Mandinka region justify historically the positions of political and social prominence these lineages held throughout the last four centuries or more of the pre-colonial period.

Two of the three traditionally dominant lineages in the region of the Gambia known as Niumi have traditions of origin that are no exceptions to this rule. Niumi was one of more than a dozen Mandinka states that existed along the banks of the navigable Gambia River from perhaps as early as the fourteenth century until colonial times. Occupying forty miles of the river's north bank at its estuary, Niumi was long a focal point for the exchange of slaves and other commodities among Africans and between Africans and Europeans. By what was likely the early sixteenth century, seven extended families of three larger lineages -- themselves segments of still larger, more widely dispersed groups of lineages having the same patronymics -- emerged to rotate political leadership in Niumi and to dominate social and economic aspects of life in the state.

Each of the following three papers is devoted to one or more aspect of conducting field research in Zambia. In this sense they form a complementary set, although this result was more fortuitous than planned. The intent had been to collect together several short papers which would address themselves to the manifold facets of fieldwork in African historical research. It is hoped that it will still be possible to do this, and it may be useful to discuss some (though by no means all) of the considerations that might merit attention in this regard:

1] With the increasing difficulties in securing research clearance in Africa, it is important -- perhaps even imperative -- that intending field workers develop viable back-up research proposals. The ways in which such alternatives can be applied could probably be illustrated best by one or two case experiences; for, if anything, it appears that the ability to pursue first intentions will diminish in the future.

2] Does Professor Kashoki's paper describe views which are representative of general opinion? Is there widespread disenchantment within Africa -- with the research attitudes and behavior of field researchers; with their commitments to the concerns of host countries; and with their care in assuring that the fruits of their labors are made easily available for local consumption? The views of other African historians, archivists, and librarians can help to reinforce or modify the arguments noted by Professor Kashoki, both by focusing on issues he has raised and by introducing new ones.

3] Graduate students (not to mention other researchers) cannot always function intellectually as freely as they might wish. They have always had to defer to the opinions and interests of their supervisors and their graduate faculty, as well as to the suspected attitudes of relevant funding agencies. Now it appears that the research priorities of host governments must be added to this litany, may even come to dominate it, as we begin to hear of “research brigades” and similar expedients. Whether this should be seen as good or bad will depend on a number of specific variables. In either instance, however, the short- and long-term ramifications of this new phenomenon need to be discussed.

In returning once again to an old theme and a familiar and much discussed topic one is keenly conscious of the danger of harping on, as one observer has labelled it, a “tired refrain.” The present revisit is predicated on the simple premise that, though no new ground is broken or fundamentally fresh issues raised, the problem is an intractable one, recurrent and pervasive, and consequently, if for no other reason, deserves one more round of attention. There is also the point, as Colson has aptly observed, that “the world is not an easy place” and abounds in social scientists and others who “share in the questioning of purposes and values.” Indeed, while what is raised in this paper might bear a distressing resemblance to arguments we have heard before, it is important to remember that to the communities who bear much of the brunt of the research conducted by foreign scholars, ethical, moral, and related issues surrounding foreigner-dominated research continue to constitute a critical area for concern and therefore a legitimate topic for sustained criticism and inquiry. Many African universities, for instance, have established ‘research affiliate schemes’ as an important, indeed integral, component of their academic functions, but in the course of operating these schemes philosophical and practical issues have arisen, many of which await satisfactory answers.

It would, however, be wrong to suggest that these concerns remain the preoccupation of academics alone, because in practical terms what a university does inevitably affects the wider public and in consequence arouses their interest and curiosity. A good example of this interest by the wider public in the affairs of the university, with specific reference to research and foreign scholars, is provided vividly and colorfully by an article written by Peter Mwaura which appeared in The Sunday Post, an East African newspaper.

Readers need not be reminded that an abundant literature exists concerning the techniques for recording and interpreting field work data; indeed, oral sources have been subjected to the most rigorous textual and literary criticism and we are even beginning to see what one observer calls ‘schools’ of oral history. All of this has been to the benefit of African history, as the many fine monographs of the last decade attest, while the proliferation of oral history projects in other areas of history attest to a general acceptance of oral data (except in the very darkest corners of the discipline) as a valuable source for the historian. But the concern with interpretation has been carried on largely to the exclusion of other fieldwork related issues. I would like to take up a number of these here, with the cautionary note that it is obviously impossible in this format to discuss them in the detail and with the variety of views they deserve and that my motivation in raising them at all derives from an interest to stimulate some debate on the topic of field work rather than to arbitrate what is correct or incorrect procedure. A further point is that, although my observations are first hand and therefore obviously limited, I believe they represent problems which are more widespread than the examples which follow.

Since African history began to be produced in quantity one-and-a-half academic generations ago, there have rarely been shortages of new explanatory theory, though sometimes there has been paucity of data, more often of field than of archival materials. Usually there has been little open discussion of the kinds of methodological problems that both of the other circumstances pose. This contribution to that debate attempts to be deliberately simple, perhaps naive, in order to permit general points to peer through specific examples. It is about the intellectual, technical, and personal complications of field work generally and is illustrated from my own research on the last hundred years in Bulozi, the western part of Zambia. In topic as well as technique, I hope that these experiences have a wider relevance, for much attention is focused on the times of colonial impact.

I have in the title purposely set limits on the discussion. I look at the grist being brought to the mill rather than at what is done with it after it has been ground, in the belief that if the quantity and nature of adulteration can be judged -- for no grain is entirely pure -- one may hope to compensate for it in the baking and so produce reasonable bread. Also, extending the analogy a little, I shall identify types of grain, for no amount of baker's skill can produce a wheat loaf from rye flour.

The slave trade has been a major theme of historical writing on Africa for many years, but it is only recently that scholars have begun to look seriously at slavery within African societies themselves. With the recent publication of three books the trickle of studies on this subject has now become a flood. The first of these, L'esclavage en Afrique précoloniale, edited by Claude Meillassoux, contains seventeen studies of slavery, mostly from francophone West Africa. Slavery in Africa, edited by Suzanne Miers and Igor Kopytoff, includes sixteen studies, eleven of which focus on west Africa. Frederick Cooper's book, Plantation Slavery on the East Coast of Africa, presents a regionally comparative study of east African plantations. Each of these volumes opens with a general introduction to the subject of African slavery, and these will form the primary focus of this essay.

A salient impression that emerges from reading the three books is that African slave systems showed enormous diversity; plantation slaves in Zanzibar had little in common with the elite military tyeddo in Senegambia, or for that matter with Sena slaves in Mozambique, who were integrated into the acquiring lineages. Often, a single society would embrace several co-existing slave systems, and the picture is further complicated by the fact that slave systems changed over time. Miers and Kopytoff underline this diversity by suggesting no fewer than seventeen different (though sometimes overlapping) terms to describe different types of slavery (p. 77).

In this brief note I hope to draw effective attention to Bibliotheca Missionum, a bibliography which, in its scope, reliability, and accessibility stands unequalled among bibliographies of any kind. More important, though, than its superior technical attributes is the fact that Bibliotheca Missionum provides entrée to a vast but largely dormant body of source materials -- materials which are as little used as they are indispensable to the proper study of the African past. With all fairness, it can be said that Bibliotheca Missionum's superiority is rivaled only by our disregard of it.

Bibliotheca Missionum was conceived by Robert Streit, a missionary of the Oblates of Mary Immaculate. The first volume in the series appeared in 1917 and since then twenty-nine others have appeared. Initially the volumes appeared more or less under the auspices of the Oblates, but in the late 1920s the project was taken over by the newly-established Pontificia Biblioteca Missionaria in Rome. Streit was succeeded by Johannes Dindinger in 1930 and other Editors have since held office, but Bibliotheca Missionum continues to be referred to as “Streit and Dindinger.” Of the thirty volumes so far published, six (volumes 15 to 20) relate directly to Africa. All were published between 1951 and 1954 and include the following:

In addition, Vols. 1, 22, and 23 are devoted to work of general missiological import and naturally contain much that relates to Africa.