Early history of the central highlands of Madagascar, called Imerina, must be written from oral literature. The first European account dates only to the late eighteenth century, long after the Merina monarchy arose, and it is not until the nineteenth century with the establishment of the London Missionary Society that detailed written sources appear. Moreover, unlike his colleagues in African history, the historian of Imerina cannot refer to archeology to test conclusions derived from oral sources since archeology in Imerina has only a few years of work behind it. Thus oral literature alone holds the key to questions about the foundations and growth of the Merina monarchy which by the mid-nineteenth century ruled all of Madagascar.

In one respect, however, the historian of ancient Imerina is more fortunate. The Merina have produced perhaps the largest corpus of historical literature in any part of Africa. I intend first to describe this literature and to point out in very general terms the problems of using it. I will focus not only on the Malagasy milieu which gave rise to historical material, but also on two related aspects of European influence. First, I will show how European ideas of social evolution penetrated Merina thinking. Second, and perhaps more important because it has received such scant attention, is the effect of the introduction of writing and printing on Malagasy texts. I consider this the outstanding historiographical problem: how did writing, editing, and redacting change the Merina's own view of their distant past. Finally, to illustrate general points raised in the first section of this study, I will examine a particular historical problem, the founding of monarchy, and show how the transformation of Merina kinglists through years of editing created a new vision of the past. I see this transformation of Malagasy texts by writing as the most important European influence on the development of a Merina history.

The purpose of this discussion is to call attention to the presence at various times of large quantities of volcanic dust in the atmosphere. Remarkable visual effects from this dust may imply hitherto unconsidered problems in identifying alleged references to solar eclipses in African chronicles and oral traditions. Climatologist Hubert Lamb has compiled the data on volcanic eruptions and dust veils since 1500 AD and attempted to determine possible meteorological effects.

Contemporary notions of volcanic activity are predisposed to spectacular lava flows, which make more interesting journalism and greater popular interest. However, eruptions characterized by free-flowing lava –– conventionally termed Type A –– are relatively unimportant in the volcanic dust question. We are interested in so-called Type B events, usually violent explosions resulting from the inability of lava or steam to find convenient outlets. Such an explosion may inject huge quantities of dust and debris into the atmosphere, causing a wide variety of visual anomalies as they are carried around the world by high-altitude winds. If the dust veil is sufficiently dense it may even cause short-term alterations in weather patterns.

Under the influence of certain conclusions in comparative linguistics, historians, archeologists and ethnologists have been led to believe that the diffusion of the Bantu languages could be linked to that of iron metallurgy. Yet from a purely linguistic point of view, the only indications of metallurgical knowledge by the Proto-Bantus are the reconstruction of terminology directly related to metallurgical techniques, reconstructions made on the basis of words gathered from the ensemble of present-day Bantu languages.

It is thusly that Guthrie was able to reconstruct certain terms such as: “iron,” “forge,” “hammer,” and “bellows,” which led him to the conclusion, as expressed in his last article on the subject, that “the speakers of the proto–language probably knew how to forge iron before the Bantu dispersion began.”

A conclusion of such historic import, however, was based on only a few words, the reconstructions and the original meanings of which were often confused. It therefore becomes necessary to re-examine on a larger scale the vocabulary related to metallurgy in the Bantu languages.

In contrast to most other nonliterate African peoples, the Himba and Herero -- pastoral Bantu peoples of southwestern Africa -- have developed quasiannual chronologies based upon important episodes and occasional conditions of the region in which they live. For example, to establish the time of a personal or family event, a Himba may refer to the epoch of the disarming or the epoch of the locust invasion as readily as we refer to 1917 or 1923. Such a chronological system is naturally of interest to the student of African history, for it is a key to the events that are considered important by the Himba themselves. But beyond this, a system of ordered epochs is also of value for demographic and other sociological studies in which the ages of individuals are important for, while the Himba, like most non-literate peoples, do not enumerate the years of their lives or the years that have passed since crucial events, many are able to provide the names of the epochs of their birth, marriage, childrens' births, and so forth. It remains, then, for the investigator merely to correlate the epochal names with our numbered years in order to determine, with a fair degree of approximation, individuals' current ages and their ages at their life crises -- data which are important for certain kinds of detailed sociological research. In essence this procedure is simple enough, but establishing a sufficiently accurate correlation between the Himba epochs and the numbered years has proved to be a difficult task and one requiring a variety of approaches. The nature of the Himba epochal chronology and the method employed for correlating it with the years of the Christian calendar are the subjects of this paper.

The Rev. Carl Christian Reindorf began his research into the history of the Gold Coast and Asante in the early 1860s. His decision to collect and record traditions appears to have been stimulated by the death in 1860 of his grandmother and by the Basel Mission which in 1864 under the direction of J.G. Christaller entered an important phase in its study of the language, traditions, and folklore of the Gold Coast peoples. Some thirty years later Reindorf's History of the Gold Coast and Asante based on Traditions and Historical Facts Comprising a Period of more than Three Centuries from about 1500 to 1860 was published.

The process of revising History I began as early as 1907 and culminated in 1951 with the publication of a revised edition in Basel and later (in 1966) in Ghana. Not surprisingly History II has become much more widely available and used than History I -- in no small measure because of the increasing demands from a new generation of historians for ‘a new look’ at the history of Ghana and the part played in it by Africans. As a result Reindorfs History II has been used increasingly in the reconstruction and reinterpretation of Ghanaian history. In his recent study of Asante Ivor Wilks recognized this development by deciding to use only History II on the grounds that it was more readily available to the reader. Earlier Michael Kwamena-Poh had noted the difficulties in consulting History I when, after quoting at second-hand a passage from it he cautiously added that

On 3 November 1863 the first issue of Leselinyana la Lesotho, datelined Morija, was published by the Revd. Adolphe Mabille of the Paris Evangelical Mission Society. Leselinyana continues to be published today. Its contents have been recently used by two biographers of Moshoeshoe and in the present paper I hope to suggest some other ways in which the materials in this publication can be of value to historians of the Sotho.

Mabille had arrived among the Sotho in 1860 and had already published several small pamphlets written by himself on a small primitive printing press. The first issue of Leselinyana comprised just a single sheet containing two pieces of writing -- an editorial signed “Leselinyana la Lesutho” in which, inter alia, the newspaper introduced itself in the first person to the Basotho. The other piece was the first of what came to be a series of articles with the general title “Mogaogeli oa Moetsalibe” (“The Sinner's Pardoner”). From the beginning Leselinyana emphasized articles of a religious and catechetical nature, clearly indicating its basic purpose as a vehicle for religious instruction and propaganda.

According to Gérard the publication of Leselinyana was interrupted only twice. The first hiatus was from 1865 to 1869 when the Boers of the Orange Free State invaded and temporarily occupied areas west of the Drakensberg mountains, evicting the French missionaries who had supported the Africans there. The second occurred during the so-called Gun War of 1880/81 when the Sotho successfully resisted efforts by the Cape government to disarm them and to open Sotho lands to white settlement. In the following discussion I will touch briefly on a few representative materials in Leselinyana which suggest the richness of the paper as a source for historians, not only of the Sotho themselves but of their relations with the whites of the area and with the missionaries.

In the early eighteenth century a Venda kingdom stretched from the Limpopo in the north to the Olifants or even Ngwenya (Crocodile) in the south and included the much smaller area now inhabited by the Venda. The main evidence for the existence of this polity comes from data on the Venda gathered between 1723 and 1730 by the Dutch at Delagoa Bay. The most substantial of these sources, part of which has been transcribed and translated here, is the so called ‘report of Mahumane’, an African from the chiefdom of Mpfumo near the Dutch trading factory who had visited the Venda king in 1727/28 and who named a number of rivers which can still be identified. Mahumane's account supports Venda traditions of an earlier and larger state, a tradition whose validity has sometimes been questioned in the last decades. The report shows that from the early eighteenth to the second half of the nineteenth centuries cultural and political changes occurred which influenced the identification of groups by outsiders and to some extent also their own self-identification. In turn this suggests that it is not always safe to project groups and polities as they existed in the second half of the nineteenth century into the remoter past.

The report also contains some information on the Lemba and on the northern and eastern neighbors of the Venda. In order not to raise false hopes I might add that Mahumane's account contains no information on the identity of the royal lineage of the Venda in the 1720's and earlier.

Origin stories are an important genre of central African oral traditions. Historians have long been intrigued by these stories, for their plots tell of the beginnings of societies and of the founding of ruling dynasties. It has been possible to cross-check the information in the oral traditions of many of the societies of west central Africa against data in Portuguese written records dating to the sixteenth and seventeenth centuries. These documents have helped to direct investigators towards the best assumptions to make when analyzing the oral traditions of west central African societies. However, writing the early history of the savanna societies in the very middle of central Africa (in southern Zaire and northern Zambia) has always been hampered by the absence of written documents which describe the area much before the early nineteenth century. Historians studying the early political history of these societies have been forced to link the events and characters of origin stories to each other without any anchor in written documentation.

Recently, doubts have been raised about the nature and function of origin stories. It is no longer clear that the first step in studying the early history of central African savanna societies is to compare different origin stories with each other, as one would a group of written documents, in order to establish a consistent historical story-line or narrative. Rather, questions raised about the nature of origin stories have brought out the point that the first step in writing the early history of this area is to resolve methodological and historiographical issues, before the historical essence can be distilled from these tales.

Hartwig and O'Barr state in The Student Africanist's Handbook that

This paper is the result of a study carried out in late summer 1976, to determine how well that statement stands. My approach was to go directly to the indexes (in the social sciences and humanities) to learn how many Africanist journals are actually indexed, and how well Africana materials are covered.

For this study “index” is interpreted in its broadest sense: any bibliographic list arranged in some order to allow a subject approach. Therefore, abstracts and bibliographies were included, as well as current contents and advanced contents indexes, in addition to the more “traditional” indexes, but machine-readable data banks were excluded. Also, I sought indexes that are current and that appear on a regular basis, at least annually. 1975 volumes (or the most recent) were used for comparison.

In a previous article in this journal I warned about the dangers of facile generalization and misinterpretation of traditional origin myths leading to the elaboration of historians' myths which bear little resemblance to the traditional, linguistic, cultural, or documentary data. I then proceeded to analyze the Singwaya (Shungwaya) myth pertaining to the origins of the peoples of the Kenya coast to show how earlier interpretations cannot be sustained by the evidence, before demonstrating that the myth is nevertheless valid for the Mijikenda, Pokomo, Swahili, Taita, and Segeju, where such evidence supports its basic veracity.

As chance would have it, while the above article was in press I delivered a related paper on the same panel as Thomas Hinnebusch who, on the basis of extensive linguistic fieldwork and analysis of the same area, had found a number of fallacies in the published linguistic data on which I had relied. While confirming the exclusion of the highlanders (Thagicũ-speakers) from the same language group as the coastal peoples and hence presumably from common origins with the others at Singwaya, Hinnebusch also excludes both the Taita and Segeju from that group while adding the Swahili to it, thus calling into doubt the integrity of the Singwaya group as I had reconstructed it. Hinnebusch's data and his analysis of it, then, obviously require a reexamination of the role of the Taita, Segeju, and Swahili peoples in the migration from Singwaya while, at the same time, casting doubt anew on the validity of origin myths or legends. But the linguistic data also extends our view enormously.

[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, ans 3(1976), pp. 91-126. Procedural matters are discussed in the first instalment, to which the reader is referred]

The historiography of continental Europe has been revolutionized since the Second World War by a group of historians generally known as Annales School, from the name of the periodical which has served as their principal forum. Originally a somewhat eccentric group, they have come in the eyes of many to incarnate historical orthodoxy. And yet Africa has scarcely been touched by these profound currents which have so powerfully affected Europe. It therefore seems necessary to attempt a brief sketch of the nature of the Annales movement and to consider how African historiography might benefit from this approach.

Annales historiography derives from a very broad and relatively recent tradition in Europe, which may be characterized as involving the rejection of the centrality of events in history in favor of more unobtrusive but more basic structures. This approach is perhaps most clearly apparent in the work of Marxist historians, but it can be seen more generally in the growing body of social and economic history written from a wide variety of ideological perspectives. Annales thus owes much both to Marxism and to such inter-war historians as Marc Bloch and Lucien Febvre. In addition, Annales is rooted in the peculiar French academic tradition of treating history and geography as a single indivisible subject.

In the spring of 1976 I returned from almost three months of research in archives in Lubumbashi, Zaire, the present-day capital of Shaba Region (ex-Katanga). My interest was in any and all records having to do with historical developments affecting Katanga's peoples from the late 19th century through the First World War. As some important changes concerning archival research seem to have taken place since Thomas Reefe's experiences in Shaba in 1972-73 (History in Africa, III, 1976, pp. 187-192) I thought it appropriate to mention some of them and include a few additional thoughts of my own.

Historians and social scientists concerned with African societies and cultures are well aware that the archives of Protestant and Catholic missionary groups that have worked in Africa contain valuable primary and secondary materials. In the case of Catholic missions, however, scholars are perhaps more aware of collections of documents of various congregations and orders of sisters, brothers and priests, deposited in their central archives in Rome, than they are of more local collections. There are specific historical and sociology of religion factors that determine how a particular order or congregation is organized, and especially to what extent it is responsible to the Vatican and the Papacy, and to what degree it has national and regional autonomy. With respect to official and personal documents of relevance to Africanists, there is a greater “decentralization” and dispersion of available materials than might be supposed.

Archives in the Republic of Zaire contain valuable documents for historical and ethnographic research, but finding these treasures often demands a great deal of perseverance and a certain amount of luck. Since the government archives in Kinshasa deal mainly with national administration, they generally have little to offer researchers interested in more localized topics. Although a small number of documents from the countryside have trickled into the Archives Nationales (housed in a wing of the fire station on Avenue de la Justice in Kinshasa), most of them remain dispersed throughout the network of regional, sub-regional, and zonal offices that form the core of the country's administrative system.

Since policies regarding the conservation of archives and the granting of access to them are not uniform, but made on an ad hoc basis by the officials on the spot, conditions vary widely from place to place and from time to time but I should point out that during my research in the Bandundu and Equateur Regions in 1975-76, the officials with whom I worked were unfailingly helpful, though the quality of the collections varied greatly. Some archives had disappeared or fallen into disarray during the troubles of the early 1960s; yet others remained remarkably intact, although even in the best-kept archives, many older documents had been partially eaten by insects.

Since China was initially the favorite field of activity of the C.I.C.M. founded at Scheut (a suburb of Brussels) in 1862, it is not surprising that the oldest documents in its archives refer to the China missions. In 1888, however, three years after the establishment of the Independent State of the Congo, the first C.I.C.M. missionaries were sent out there and their successors still continue their activities. Thus nowadays a very important part of the archives concerns the Congo, covering a period of nearly ninety years.

At first the Apostolic Vicariate entrusted to the C.I.C.M. missionaries covered the entire extent of the Congo except for parts of the Apostolic Vicariate of Tanganyika, created two years earlier and given to the White Fathers to administer. Gradually, however, other missionaries came to help the C.I.C.M. pioneers and were put in charge of various territories which had originally been part of the Apostolic Vicariate. Eventually the C.I.C.M. retained only Leopoldville and its surrounding area, and the regions of Lower Congo, Equator, Lake Leopold II, and Kasai which, from the religious point of view, gave rise to the present C.I.C.M. provinces of Kinshasa, Boma, Lisala, Budjala, Inongo, Kananga, and East Kasai. This dismemberment explains why the documents preserved in the archives become more and more concentrated on these regions.

Using archives is often rewarding but seldom very exciting, and even less often amusing. The historian is most often working with materials which, despite being ‘raw’, are generally at rather a high level of abstraction. This is particularly true, it seems, for archival matters on Africa; we all too seldom can get the real ‘flavor’ of a situation. As the title of this note suggests, the following items are of little historical significance -- maybe none at all. Even so, perhaps they can offer insights into European modes of thought and life on the Gold Coast which can supplement those few we can discern in the more typical archival materials.

[1] Thomas Bucknell, Sekondi, to Cape Coast Castle, 21 September 1686 o.s., Rawlinson C.745, #1020:

In February of 1976 the American Historical Association published the first issue of Recently Published Articles. RPA is a current, comprehensive bibliography of scholarly historical articles in all fields, and replaces the bibliographical sections previously published in the American Historical Review under the same title. RPA is issued three times a year and, at present, lists nearly 15,000 citations annually, of which about one-eighth relate to Africa -- a number second only to that for the United States.

The creation of RPA resulted from the re-organization of the American Historical Association and its various services. A survey of American Historical Review subscribers revealed a desire by the majority for a more extensive and balanced book review section and most of the space left in the Review by the departure of the bibliography has in fact been filled by additional book reviews, including more in African history. Surprisingly -- and perhaps distressingly -- only a minority of those surveyed indicated that they consulted the bibliographical sections with any regularity. Because of this, and because the growth of the bibliographies during the 1970s (reflecting both increased output and better coverage) was adding steadily to the cost of producing the Review, it was decided to transfer these costs directly to the users -- a practice the reader will recognize as becoming more common. At the same time it was decided to intensify efforts at better and more extensive coverage, in particular to include more articles outside history proper which nevertheless have interest to historians.