This paper examines some recent developments in family and residential group structure in the village of Amedzofe-Avatime. The village is one of a group of seven which, with their surrounding farmlands, constitute the Avatime traditional area in the Volta Region of Ghana. The study of Amedzofe (and Avatime) family structure is singled out here as of interest for two reasons. In the first place, a study of current (1977) Avatime family structure indicates that new forms are emerging of patterns of child rearing. Esther Goody's work on both traditional and more recent patterns of fostering in Ghana is particularly relevant here (E. N. and J. R. Goody 1967; Goody 1970; 1975). Secondly such a study provides empirical evidence that Goode's (1963) pattern of change in family structure showing the increasing influence of ‘modernisation’, while it may be pertinent in the long term, is not being realised among Avatime. What is happening, rather, is a drift away from a stable conjugal pattern. Here I am concerned mainly with the first point. As it is elucidated, it will become clear that the incidence of Goode's nuclear family based units is not increasing in modern rural Avatime. A detailed examination of the structure and composition of residential groups and their significance for a modern society is forthcoming.

Wherever belief in witchcraft permeated an African society, fear prevailed and people demanded protection and control. Even though the degree of African concern about witchcraft was not always appreciated by outsiders, it was possible, at least in centralized societies, for such outsiders to discern the processes that were involved in its control. A king or a priest who failed to control the spread of witchcraft and to alleviate the fear was unlikely to maintain his authority for long. In non-centralized societies, the problem of witchcraft and the means of control were less clear-cut. Solutions were rarely obvious and easy.

In their initial interaction with the Colonial powers, several East African peoples such as the Maasai, the Turkana, the Sebei, the Karamojong, and the Nandi—all organized through some type of age-based institution—united around prophetic leaders, diviners, or ritual experts who mobilized men from several territorial sections to confront the intruders. This ad hoc military unity was necessarily short-lived, usually ending with the defeat of the people by the colonial power and see the imprisonment or death of the prophetic leader involved. (See Fosbrooke 1948: 12-19; Merker 1910: 67-105; Jacobs 1965: 20-108; Dyson-Hudson 1966: 15-16; Gulliver 1950: 229, 240; Meinertzhagen 1956: 222 ff; Weatherby 1962: 200-12; 1967: 133-44; Lamphear 1976: 225-43.) While ethnological studies of various age-organizations often mention that diviners or prophets provided professional services for the members of an age-group at their ceremonies, no one has examined the process by which a prophetic leader or diviner established his legitimacy during periods of peace so that he might lead the people during times of crisis. An examination of the prophetic institution among the Maasai and the relationship between the prophets and the members of the age-sets may provide some insight into the process, especially the manner in which prophets emerged as leaders of the people during two major crises in the history of the Purko-Kisongo Maasai: the Ilaikipiak war and the rinderpest pan-zootic.

The heroic Bacwezi, founders of kingship in Ankole and in other East African interlacustrine states, are said to have ‘disappeared’ because of their subjects' ungrateful insurrection (Oberg 1940: 124). The moral of the legend—that revolts against Ankole's kings, successors of the Bacwezi, are doomed!—has been curiously adhered to.

The Abutia Ewe form one of what the British colonial administrators dubbed the ‘traditional areas’, over one hundred of which are said to compose the Ewe people or Eweland. These Ewe traditional areas lie in the southern half of the Volta Region (in Ghana) and in southern Togo. Although much has been written about the Ewe, little is known about the political organisation of the inland areas, north of the coastal savanna. In fact, most authors have treated the Ewe as if they were thoroughly homogeneous and could be analysed as one ethnic group or one society, only acknowledging variations between the north and south, not considered significant enough to make them completely different groups (Spieth 1906; 1911; Westermann 1935; Ward 1949; Manoukian 1952; Nukunya 1969; Friedländer 1962; Asamoa 1971 amongst many others). And yet I contend that the northern areas are as distinct from the southern ones as they are from the Akan populations. As a result, none of the present available literature is particularly useful as a paradigmatic model of the northern areas' political organisation.

In her lucid and persuasive Henry Myers Lecture on ‘The power of rights’ (Man, 1977), Professor La Fontaine argues that since the Gisu themselves cite transition from one status to another as the purpose of their initiation rites, this transformation cannot logically explain them. Taking up Maurice Bloch's idea that religion is concerned with inequality and hierarchy (1974; 1977), La Fontaine argues convincingly that the explanation of Gisu initiation rites lies in their functions of validating traditional knowledge and of maintaining traditional authority. The relationship between religion and political power had been noticed before but it had not previously been applied to the field of rites of passage nor had the mechanism of the supportive role of traditional knowledge been so clearly analysed.