The Mutapa state occupied a triangle of land between the Zambezi river in the north, the Hunyani river and Umvukwe range on the southwest, and the Mazoe and Ruenya rivers on the southeast. It thus consisted of a small segment of the southern Zambezian plateau and an arc of the Zambezi valley lowlands. The state dated back to at least the fifteenth century, and some branches of its ruling dynasty continue to control fragments of the state under the governments of Rhodesia and Mozambique. These descendants of the Mutapa dynasty, like their fellow-members of the Korekore dialect cluster of the Shona-speaking peoples, retain traditions of their past that are passed on from generation to generation by an informal learning process. These traditions are almost all devoted to the ruling dynasties rather than to the mass of the people and are especially concerned with lines of descent and land rights. They range from myths to relatively accurate factual accounts, with a wide variety of traditions between these two extremes. It was at one time thought that the mediums of mhondoro ancestral spirits were equivalent to the professional tradition-keepers of states such as Rwanda, but this theory has not been adequately proven.

The Mutapa state is of especial interest because it is the only one of four known major Shona states—Zimbabwe, Torwa, Mutapa, and Changamire—to escape being uprooted entirely by new settlements of people, and the only one that was close to Portuguese centers (in which information was recorded). It has thus been possible to compare traditions and documents in a way that cannot be done for the other states. Because of the reluctance or inability of many researchers to work in Rhodesia and Mozambique in the last fifteen years, the history of the Mutapa state has been heavily dependent upon the work of D.P. Abraham, at least as far as traditions are concerned. Abraham collected traditions from ca. 1950 to 1971; but so far the only works of his that are readily available are eight papers, of which the most important were produced in the period 1959-1963. These have formed the basis of most of the secondary writing on the Shona states; the inter-relationship between them and a developing archaeology has been discussed in an earlier article, and only a few points of this discussion need be brought in here.

The earliest detailed account of Upper Guinea was written by Cadamosto, probably in 1463, and published in Italian in 1507. In the decade 1500-1509, a more detailed account was compiled by Valentim Fernandes, and a similar account of the whole Guinea coast was prepared by Pacheco Pereira—but neither of these Portuguese works was published in the sixteenth century, and even Portuguese writers of the period do not appear to have seen the manuscripts. Late fifteenth century roteiros (guides to nautical routes, “rutters” in sixteenth century English) also remained in manuscript, but being in regular use were more easily borrowed and copied: detailed roteiros for Guinea certainly contributed to Pacheco Pereira's text, and less detailed information from roteiros seems t o have trickled through to some of the non-Portuguese sources we are about to discuss. The position regarding maps was very similar; thus, non-Portuguese writers on Guinea during the first half of the sixteenth century had available the following limited sources of information: (a) references, mainly mythological, in ancient and classical accounts of Africa—e.g., Herodotus, Pliny, and Ptolemy; (b) Cadamosto's account, after its publication in 1507; (c) later experience of seamen in Guinea, passed on orally, or from notes and roteiros not now extant; (d) maps based on some or all of the above sources.

The golden rule of all scientific publication is that the reader must be in a position to be able to check the author's statements.

The scientific level of existing publications of oral traditions is, as a rule, not very high.

The ideal treatment of oral traditional materials by the historian may be considered as occurring in four main stages:

1. the collection of the traditions in the field

2. the creation of a primary souce with the pubhcation of the raw materials

3. the analysis of this raw material

4. the creation of a secondary source with the publication of the results of this analysis and interpretation

These various stages cannot easily be demarcated very clearly. Preliminary analyses influence the field work and often pass into the pubhcation of the raw material (that is, the nature of the published material is anticipated in the very recording of the data). In turn, the publication of the interpretative analysis frequently contains quotations from the raw field data and should yield insights into the analytical process.

The accumulated (and rather extensive) literature on this subject has heretofore been devoted mainly to the analysis of traditions, and quite rightly so: if anything, it has tended to discuss the methodological problems associated with the analysis, sometimes drawing attention to these problems for the first time. After Jan Vansina, it was Philip Curtin who particularly concerned himself with the first stage (the collecting and recording of traditions in the field), outlining the standard to be sought and offering a series of important practical suggestions. In my opinion, Curtin is the only author so far who has dealt in a systematic manner with the preliminary processing and publication of the raw material (stage 2) and established guidelines for it.

The Atlantic slave trade in its various manifestations has never lacked scholarly attention, be it disinterested or selfish. The major focus has often been on the motivations and roles of those who participated in the trade other than as victims. In the eighteenth and nineteenth centuries, interest tended to be confined either to the apologists for the trade or to its critics; but in recent years, the matter has not failed to engage the attention of more serious enquiry.

As a major center of the trade throughout the period, Dahomey has been studied extensively from the very beginning. Much of the work has regarded Dahomey as the slave trading state par excellence. Recently, however, I.A. Akinjogbin has advanced the stimulating and appealing argument that the Dahomey state was created partially, but explicitly, in defensive reaction to early signs of European interest in slaves on the Guinea coast. Akinjogbin further argues that–although Dahomey did in fact eventually develop into an important slave trading polity–it did so reluctantly and only because the Europeans trading along the coast demanded slaves–and only slaves–for their own goods.

Needless to say, attractive arguments rather have a way of being more readily (and less discriminatingly) accepted, and Akinjogbin's interpretation of early Dahomey history has already re-appeared in several important recent works on the history of west Africa. With this in mind, the present paper has two purposes. First, it proposes to examine the validity of Akinjogbin's thesis by examining one particular aspect of his argument: the motives of the Dahomey ruler Agaja (ca. 1708 to 1740) in conquering the coastal states of Allada and Whydah between 1724 and 1727. In discussing Akinjogbin's elucidation of Agaja's motives, we propose to concentrate not so much on the logic of his argumentation, but on his use of the sources on which any assessment of Agaja's motives must be based. With a single exception the material examined here is the same used by Akinjogbin, and in this sense the first part of the paper should be seen as a study in the use of evidence and inference.

The second part of this paper will be an examination of European and Dahomean commercial activities in the first few years after the conquest of the two coastal states. The sources describing these activities suggest that the motives and mechanisms of all parties were more complex than generally assumed.

The pre-colonial history of the Yoruba has attracted considerable attention from academically trained historians in recent years. This academic historiography–in Yorubaland as elsewhere in sub-Saharan Africa–does not antedate the 1950s, but it was preceded by a tradition of historical writing by local amateur historians which stretched back well into the nineteenth century. The modern academic historians owe a great deal to these amateur predecessors: much of the “oral tradition” utilized by the academic historians comes in fact at second hand from the writings of the amateurs, and the current generation of local historians has figured prominently among the informants from whom the academics have collected their oral evidence.

This fusion of academic and amateur historiography was, indeed, given some institutional recognition in the Yoruba Historical Research Scheme launched by the government of the Western Region of Nigeria in 1956, in which both academics and local historians were employed as research associates to collect traditional material. Despite their importance, however, little serious work has been done on the early historians of Yorubaland. The existence of a local tradition of historiography in Yorubaland has been mentioned in general surveys of historical writing on Africa, and attention has been drawn to it as constituting an aspect of the development of “cultural nationalism” among western-educated Africans in the late nineteenth and early twentieth centuries. These discussions of early Yoruba historiography, however, have dealt with only a few of the better known works and have given little idea of the wealth of the published material or of its character—nor is there any comprehensive bibliography of the writings of the early Yoruba historians. The present article, therefore, attempts to present as complete a survey as possible of historical writing on Yorubaland in the period before c. 1950 and seeks to make some contribution towards assessment of the value of the Yoruba local histories as sources by giving some information on the context in which the local historians wrote and the way in which they went about the task of reconstructing Yoruba history.

[This continues the comparison of the texts of the Dutch and English versions of Bosman which was begun in History in Africa 2 (1975): 185-216, and which will be continued in future numbers. Procedural matters are discussed in the first installment, to which the reader is referred.]

Stretching across North-central India from Kāthiāvaṛa to Orissa lies a great geographical and cultural shatterbelt formed by the Vindhyan mountains and their associated tracts, an area traditionally characterized by high internal subdivision and political fragmentation. The northwestern extension of this belt comprises the frontier zone today known as Rājasthān (“the land of the Princes”). Strategically situated between the rich Gangetic plains of Hindustān to the northwest and the fertile regions of Mālvā and Gujarāt to the south and southwest, it forms an area of marginal agricultural importance whose historical significance lay primarily in its position as a key transitional zone between larger cultural centers, criss-crossed and intersected at a number of points by major caraven routes.

C'est à bon droit que Vansina considère les anciens dictionnaires comme des sources historiques, qui « conceal an ample description of a society and its culture at a given moment. » (HIA, 1 (1974), pp. 149-52). Pourtant il nous semble qu'il n'a pas toujours tenu compte de sa propre recommandation, très importante: « it must be constantly in mind who wrote them and for what purposes. »

En effet, à titre d'example, Vansina s'applique à extraire des informations précises sur l'ancien royaume de Congo, d'un dictionnaire trilingue du milieu du XVIIe siècle, copié par le capucin flamand Georges de Geel, arrivé au Congo en juin 1651 et mort le 27 décembre de l'année suivante. Cette copie autographe est conservée à la Biblioteca Nazionale Vittorio-Emmanuele de Rome et se présente sous forme d'un lexique latin-espagnol, augmenté de la traduction kikongo de chaque vocable. Vansina ne semble pas avoir eu recours à cette copie, la seule parvenue jusqu'à nous: il s'est fié à une soi-disant édition de manuscrit de G. de Geel, éditée par les jésuites J. van Wing et C. Penders, Le plus ancien dictionnaire bantou: Vocabularium P. Georgii Gelensis (Louvain, 1928). Cette édition a laissé tomber les vocables latins et espagnols et a regroupé les termes kikongo, récrits selon la graphie « moderne », par ordre alphabétique en y ajoutant la traduction française et flamande.

Ne tenant pas compte du caractère particulier du dictionnaire de G. de Geel. Vansina, utilisant uniquement l'édition Van Wing/Penders, semble avoir tiré de la présence de certains termes kikongo, des conclusion très hâtives. Ainsi de la mention de certains termes ou circonlocutions kikongo, il déduit l'existence dans le société Kongo de « wine merchants, butchers, fishmongers, booksellers, shop¬keepers, clothes sellers, perfume dealers, and pharmacists. » Il y aurait eu des bou¬langers, et même des restaurants.

One of the more striking recent trends in the study of African history has been the mushrooming of studies focusing on the “underdevelopment” of Africa (or some part of it) following the publication of Walter Rodney's How Europe Underdeveloped Africa in 1972. Part of the appeal of the “underdevelopment” literature is that it stands the apologists for colonialism on their heads. Barely more than a decade earlier, as the 1950s were coming to a close, one could readily find settlers, shopkeepers, colonial bureaucrats, and others arguing that the mounting pressure for independence in sub-Saharan Africa should be resisted because European rule had brought “development” to the continent. It is not surprising, therefore, that after a decade of independence there should be a ready audience for the argument that colonial rule did not “develop” African economies but, to the contrary, actually “underdeveloped” them—and perhaps we should also not be surprised that “underdevelopment” is often not defined oris defined inconsistently.

The central focus of this essay will be the main difficulties of studying either “development” or “underdevelopment” and, in particular, the problems of determining the causes of either of these states at any given period in Africa's history. All the examples from the “underdevelopment” literature will be drawn from Rodney's book, primarily because the work seems representative of the growing body of literature on this subject and because it has been by far the most influential of the studies dealing with the topic in Africa. I would like to emphasize that in the following discussion my approach is more that of the economist than of the historian, a point which will be clearly reflected in the few points I have chosen for illustration.

L'Oeuvre de la Propagation de la Foi a été fondée à Lyon le 3 mai 1822 par quelques laïcs organisant et mettant en oeuvre une initiative d'une jeune lyonnaise, Pauline Marie Jaricot, pour venir en aide aux Missions catholiques, par la prière et par les secours financiers. Ces derniers, indispensables dans l'abandon où la période révolutionnaire avait plongé les missionnaires, étaient réunis au moyen du « sou par semaine » que versait chaque associé, puis répartis entre les missions selon leurs besoins.

La Propagation de la Foi est la plus ancienne et la plus importante des oeuvres qui se sont créées au cours du XIX° siècle pour des motifs semblables (Oeuvre de la Sainte-Enfance, 1843, Oeuvre de Saint-Pierre Apôtre, 1889). Peu après celui de Lyon, un Conseil Central fut aussi créé à Paris, puis un Conseil Supérieur dont dépendaient les deux Conseils Centraux de Lyon et de Paris. L'Oeuvre fut dirigée conjointement à partir de 1830, lorsque disparut le Conseil Supérieur, par les deux Conseils Centraux de Paris et de Lyon, ce dernier gardant une certaine prééminence, en tant que Conseil fondateur. Pendant un siècle environ, l'expansion européenne, puis mondiale de la Propagation de la Foi n'entraina aucune modification. Mais, en 1922, à la demande de la Papauté, les laïes qui composaient les deux Conseils acceptèrent que l'Oeuvre ait désormais son centre à Rome (création d'un Conseil Supérieur dépendant de la Sacrée Congrégation romaine de la Propagande et de Conseils nationaux).

Ces données rendent compte de deux caractéristiques des Archives de la Propagation de la Foi qui se trouvent in France:

1. Elles sont surtout intéressantes pour la période allant des origines (1822) au transfert de la direction de l'Oeuvre à Rome.

2. Ces Archives sont réparties entre deux fonds, celui de Lyon et celui de Paris où se trouvaient avant 1922 les deux Conseils responsables de l'Oeuvre.

Nous ne parlerons ici que des Archives de Lyon, les plus importantes il est vrai. Le fonds parisien peut être consulté dans le local des Oeuvres Pontificales Missionnaires (5, rue Monsieur, 75007 Paris).

Gaining access to the Portuguese archives in Goa, now part of the Indian state of Goa, is (or, at least, was in my case) relatively easy. About two months in advance I wrote to Dr. V.T. Gune, the Director, asking permission to carry out research. After clearance had been obtained from New Delhi—I had the impression that no research on Africa would be regarded as “sensitive” by the Indian government—Dr. Gune very quickly granted me access.

The archives are located in the town of Panjim in Goa, which can be reached by bus from neighboring Maharashtra or by the daily plane from Bombay and Trivandrum. Intending visitors should bear in mind that the monsoon can (and often does) sever all of these routes at various times. Panjim supplies most human needs, but microfilm has rarely been seen there since the Indian occupation of Goa in 1961, and researchers are warned to bring their own, although the actual work will be done by the photographic section. The section prefers extra fine grain Recordak Microfile Pancromatic 35mm double-perforated film, although similar films are acceptable. The main archives building houses most of the files, and researchers are provided space in the Archivist's office. Photography, however, is carried out at a separate depot in town. Unless they are unusually busy, the photographic staff can process material very quickly; in my case, they were able to film 863 exposures in little more than a week.

Before departing for Mali, researchers should write a formal request for permit to conduct research to: Ministre de l'Enseignement Supérieur et de la Recherche Scientifique, Bamako. A second letter should be sent to M. Mamadou Sarr, Chef de la Division de Recherche Scientifique, B.P. 159, Bamako, advising him that the first letter has been written and stating both the research topic and the anticipated duration of the research trip.

Once in Bamako (if permission is granted), the “Autorisation de Recherche” can be obtained from M. Sarr, whose office is in the Ministry of Education Building on the Place de la Liberté. If this can be done within the first few days after arrival and before expiration of the temporary visa, the research authorization can simplify the visa renewal process at the Direction des Services de Sécurité. Authorization for access to the Archives Nationales du Mali requires a separate application, which also is issued by M. Sarr and can be applied for at the same time as the general authorization.

Every researcher is required to work under the aegis of the Institut des Sciences Humaines, of which M. Sarr is director. Those intending to do field work are assigned an I.S.H.M. “technician” who may have some background in the subject and can also be helpful in contacting informants. In principle, the technician is supposed to accompany the investigator on all field trips, and the investigator is expected to pay the technician's transportation costs, food, and lodging, as well as 2,000 Malien francs (about $5.00) per diem.

The Regional Archival Center at Tamale contains extensive documentation on the former Northern Territories of the Gold Coast, which today make up the Northern and Upper Regions of Ghana. The bulk of the records relating to the colonial period are held in the main repository in Accra, but any scholar working on a northern topic should plan to visit Tamale. While there is a tendancy for “older” (i.e., pre-1940 materials) to be in Accra and for more recent items and Native Authority records to be in Tamale, there is no hard and fast rule, and many exceptions and duplications exist.

There is no catalog for the collection. Accession lists for the seventeen classifi¬cations employed are quite helpful, but they are often somewhat sketchy and sometimes give little indication of the actual contents of a file or the time period which it covers. Following is a list of the classifications, together with titles of selected files, to illustrate their extent and the types of material they contain.

It is impossible to write a comprehensive report on the condition of archives throughout the Republic of Zaire, for documents are not stored in national or centralized archives. For example, in Shaba Region (ex-Katanga Province), documents are kept in various bureaux at Lubumbashi. At subordinate territorial levels like the sous-region (ex-district) and zone (ex-territoire), documents are tucked away in cabinets, work rooms, and closets at their administrative centers. It is difficult to plan the time to be spent in archives in advance, for there seems to be no middle ground in their condition. Either they are in good order with documents neatly bundled in labeled file folders, or they are a complete shambles with most documents missing or stored in unclassified heaps on some shelves or on the floor. It is difficult to anticipate which archive will be in good order. There is no reason to assume that an archive in an urban center will be in better shape than an archive at an infrequently visited rural poste or vice versa. The only generalization that can be made is that an archive in Shaba Region which has been properly maintained will be a rich depository of documents. The Belgian colonial administration contained more European personnel than the administration of any other colonial power, and the paperwork produced by this bureaucracy was substantial. When the Belgians departed, they left a legacy of correspondence, reports, procès-verbaux, inquiries, etc., dating from as early as 1907, which can still be found in the little-used archives of Shaba.

In the course of my research into the social history of nineteenth century immigrants to Liberia from America, I had the occasion to consult the Liberian Archives in Monrovia. Although the focus of my own research was limited in scope, I was exposed to the full range of materials available in this repository. In 1966, Svend Holsoe was instrumental in helping to organize the archival documents, and he prepared a catalog listing of the same. The present effort should be considered a “second edition” of Holsoe's original work, which only circulated privately. Wherever possible I have additionally tried to include some idea of the extent of each volume or bundle by noting page content where known or estimating the number of pieces included in each bundle. The need for this new listing has arisen from the impact of the course of time on the collection. Over the period of almost a decade since Professor Holsoe's catalog was prepared, factors such as climate conditions and reorganization efforts necessitated by space limitations have altered the content of the collection considerably.

The government archival material comes under the administrative authority of the Ministry of Foreign Affairs. It has been organized and housed in a separate facility directly across from the Ministry of Foreign Affairs on Ashmun Street in Monrovia. Mr. Augustine D. Jallah is the Director of Archives and the person immediately responsible for the care of the documents. Permission to use the material must be received directly from the Ministry of Foreign Affairs.

The Ethiopian Manuscript Microfilm Library (EMML) is a joint project of the Ethiopian Orthodox Church, the Hill Monastic Manuscript Library (HMML) (formerly Monastic Manuscript Microfilm Library) of St. John's Abbey and University, Collegeville, Minnesota, U.S.A. Professor Walter Harrelson, former Dean of the Divinity School at Vanderbilt University, is Chairman; Dr. Julian G. Plante, Director of HMML and Research Professor of Classics in St. John's University, is Vice-Chairman; and Dr. Sergew Hable Selassie, Professor of History in Addis Ababa University, is Project Director. Grateful acknowledgement for the funding of this project must be made to the National Endowment for the Humanities, Washington, D.C.

EMML was established at the urging of His Holiness Abuna Tewoflos, Patriarch of the Ethiopian Orthodox Church, who was much concerned about the dangers of irreparable damage and loss to the manuscript treasures of his Church. The purpose of the project is twofold: To preserve on microfilm the precious treasures of manuscripts and to make those source materials available for study by scholars both within and outside Ethiopia. Efforts are also being directed toward obtaining copies of Ethiopian manuscripts outside of Ethiopia.

Actual photographic operations are being carried out by an Ethiopian team in conjunction with Xerox University Microfilms, Ann Arbor, Michigan, U.S.A., where the film is processed. Positive and negative copies are sent to the Ethiopian Orthodox Church, which has founded a center for the purpose at Addis Ababa, and positive copy is sent to HMML in Collegeville, where interested students and scholars may come for consultation and study. HMML is a research center with a focus on medieval and renaissance manuscripts on microfilm.