Professor Diedrich Westermann, whose international reputation in the field of African studies needs no emphasis here, was intimately associated with the foundation and early development of our Institute. As a former missionary in Togoland and as Director of the Seminar für Orientalische Sprachen in the University of Berlin, he attended the Conference at High Leigh in 1924 when the proposal to found an International Institute for the study of African languages and cultures was first formulated. He served on the provisional committee which gave effect to this proposal and, at the first general meeting of the Institute in 1926, he and Professor Delafosse1 were appointed Directors of the Institute to serve with Harms Vischer as Secretary General and Lord Lugard as Chairman.

This article is intended primarily as a contribution to the study of the village headman in British Central Africa. While most reports from the area have paid attention to the headmanl's structural position, this has been specifically considered by Gluckman, Barnes, and Mitchell in ‘The Village Headman in British Central Africa’ (Africa, xix, no. 2, 1949) and, more recently, by Mitchell in his forthcoming analysis of The Yao Village. The closely related question of the values attached to village life has received attention notably from Prof. M. Wilson in Good Company.

Since the household contains members drawn from the larger minor lineage group, we may expect to find that it functions within the framework of the latter. This is indeed the way in which Konkomba think of the household—as part of something larger, as something that is now isolated from and now merged in the larger group. The phrase ‘Ti je mfum mba’ (we are one) may refer to a household, a minor lineage group, a major lineage group, a clan, a tribe or the whole Konkomba people. When speaking of co-operation they speak in terms of the minor lineage group rather than in terms of the household. We shall try to differentiate the role of the household as a unit of reciprocal help, of social control, and of instruction, &c.

This paper first seeks to show how far the actual forms of the Konkomba family and household coincide with their forms as conceived by the people themselves. The second part (which will appear in a later number of this journal) analyses some of the functions of these units of organization. By function, I mean the relation of the household to certain aspects of Konkomba life: namely, the household as a unit of production and consumption, as a unit of social control, as a ritual unit, and so on. The term ‘household’ refers to the total group of persons living together in one compound (letʃeni), which is a cluster of round houses distributed about a central space and linked by a low wall (see Fig. I). The head of a household (letʃendaa) is the senior man, the husband and the father of the family that is the nucleus of the household. This may be an elementary or a polygynous family; or it may be an expanded family consisting of a number of brothers and their wives, sons, and unmarried daughters; or it may be an extended family, consisting of a man, his wives, their sons, sons' wives and children, and their unmarried daughters. To this nucleus other kin are added and it will be shown that these additional members are always either members of the minor lineage group of the household head or wives or widows of members.

The fact that most of the studies recently carried out on the Ibo have been initiated in response to an urgent political situation, has resulted in an emphasis on those aspects of the culture of most immediate interest to administrators; and although a considerable body of data on religion has been included in the results, this sphere of activity has perhaps received rather less attention than others.

In this article an attempt is made to sketch some salient features of religious belief in a single Ibo village-group, which, it is hoped, may illuminate some facets of the problem not emphasized in previous work. My investigations were carried out in the Northern Ibo Nike group which lies to the immediate north-east of Enugu, capital of the Eastern Provinces of Nigeria. Rich land and low population density (48 per sq. mile) have here mitigated the usual economic pressures towards education for a life of migrant labour, commerce or clerkship in the towns, with the result that, despite their proximity to a very highly developed urban area, the Nike people tend to take little interest in mission activity. Ibagwa, the senior village of the group, contains some 260 adult males, of whom only a bare half-dozen practise a loose form of Roman Catholicism, so that mission influence is not at all developed. This of course is a distinct advantage for the present study.

La structure sociale des Dogon a fait l'objet d'enquêtes consignées, depuis 1931, dans un certain nombre d'ouvrages. Cette population observe dans le cadre de la famille indivise une filiation patrilinéaire et patrilocale, le totem étant hérité en ligne agnatique. Abstraction faite des manages avec les gens de caste (forgerons, griots, travailleurs du bois, du cuir), strictement prohibes, les Dogon pratiquent, au sein des quatre grandes tribus ou groupes qui les composent, l'échange généralisé.

Une synthèse des renseignements recueillis a été effectuée par M. D. Tait qui a procédé à ‘une analyse structurale des matériaux’ et conclu à la nécessité, pour comprendre les problèmes de la structure dogon, d'une ‘investigation plus poussée’.

The initial spread of Swahili from the coastal areas of Kenya and Tanganyika during the last century must, in large measure, be attributed to the Swahili traders, who, with their often unsavoury reputations and their caravans, took the language as far as the Congo. Where these traders settled in any numbers, as at Tabora, Mwanza, Ujiji, Usumbura, Lake Nyasa, and, to a smaller extent, at Mumias, small Islamic islands of coastal-Swahili-speakers have persisted: some flourishing like Usumbura, others, like Mumias, declining. In general, however, the language became more emasculated the farther it extended from the coast, so that in North Uganda and parts of the Eastern Congo it is essentially a highly simplified means of communication for individuals of widely differing groups; pruned of all refinements, its link with the coast is a tenuous one.

1. The material on which the present findings are based was collected during the period January 1950 to June 1951 near Kasama (Central Bemba) and the period July 1951 to June 1952 from Central Bemba informants in Lusaka, N. Rhodesia.

2. If we disregard all subject and object prefixes, locative suffixes, and the existence of high- and low-toned radicals (all of which affect tense signs, but only at a phonological and tonological level), we find that formally speaking the Bemba verb has at least 48 positive and 31 negative single-word main-sentence tenses. (Object and subject relative sentence and sub-relative sentence tenses also exist, but are essentially mainsentence tenses with regular tonal modifications.) In order to study and evaluate a verb of such complexity, it is necessary first to find some method of tabulation. There are two possible points of departure: form, and meaning. If we use the latter, we arrive at a matrix involving at least the categories of order (positive or negative), time, aspect, mood, and emphasis. But then we find that certain ‘pigeon-holes’ in the total table are filled by tense forms that also occur elsewhere, given the right context and the right radical. (‘Radical’ we may define as that element in a Bemba verb carrying that meaning which is independent of tense signs and prefixes and infixes.) It is convenient for demonstration and other purposes to construct our tables showing these dual- and triple-function forms occupying their complete range of pigeon-holes, even though this may mean a sacrifice of formal clarity. (It should be remembered, however, that even such compromise tables—in a ‘mere’ order-time-aspectemphasis-mood matrix—can only expose basic meanings; the precise applications of some of the tenses are so subtle as to defy even five-dimensional tabulation.)

The Oron, who call themselves ɔrɔ, number approximately 50,000 individuals, and inhabit 90 villages in Calabar Province, Nigeria. Lexicon and morphology establish the affiliation of Oron with the Ibibio-Efik dialect cluster recognized by Westermann and Bryan, and with Efik in the Central Branch (Cross River languages) of Greenberg's Niger-Congo family.

The material is based on the speech of Mr. Okon Ukam of Eweme village. Since the informant did not speak English the writer communicated with him through Mr. Dimo Udoh Eyo, an Ibibio from Etobodom, Uyo District, who became conversant with the Oron language in his youth. Between March and May 1953, 10,000 words of text and a vocabulary list of approximately 1,700 morphemes were transcribed. Although it is not assumed that all possible tenses or tone permutations have been obtained, it is believed that most of the tenses are described. The writer acknowledges his indebtedness to Ward's study of Efik, and Lounsbury's study of Oneida verb morphology since each has served as a model for the presentation of certain aspects of Oron morphology.

Towards the end of 1947 I carried out a public-opinion survey among the Cewa tribe of Fort Jameson district, Northern Rhodesia. Although the survey was a failure, judged by the rigorous standards of public-opinion polling, it nevertheless threw light on some of the problems that arise when public-opinion-polling techniques are adapted for use among preliterate peoples. Research of this kind has a place in assessing general morale, in gauging people's reactions to administrative and development policies, and in supplementing the more intensive, but highly selective, observations of the social anthropologist. Because this is an important but almost untouched field, I am recording the lessons that are to be learned from my experiment.

This paper presents a study of what is sometimes called detribalization—the process by which tribal people, especially those who have left their homeland and obtained paid employment in towns, are separated from the social and cultural heritage of their tribe. But this is too superficial a statement of the matter. It is necessary to define the problem in sociological terms before attempting a systematic analysis of the process. Accordingly I shall start by describing the system of social relations prevailing among Temne in Freetown, and shall examine the forces which, over the past fifty years, have influenced its character. At the beginning of this period relationships among the Temne immigrants appear to have been relatively close and stable, but, from the 1920's, disintegrative tendencies became progressively more marked until, at the end of the 1930's, the young men carried out a series of swift changes which resulted in a more successful adaptation of the system and its closer integration.

In this essay I propose to discuss three recent studies of African legal systems—Anderson's survey of Mohammedan law in British African territories, Gluckman's book on the Barotse (or Lozi, to mention the tribe mainly considered), and Howell's book on Nuer law. To begin with, this seemed merely a convenient way of combining three reviews; nor did the three books seem to have very much more in common than the date of their publication and the fact that they are all concerned with Africa. The types of law described and the scope of the descriptions differ widely. So does even the background of the authors; for Professor Anderson is a jurist, Professor Gluckman an anthropologist with legal training and interests, and Dr. Howell an administrator-cum-magistrate with anthropological training and interests.