The research upon which this paper is based was carried out over a period of about 18 months in the Luapula Province of Zambia. I first spent a month in Ntoto fishing camp, which is located about 15 km north of Kashikishi village on the slopes of Lake Mweru. For a year after that I conducted research in Kashikishi, a large village (about 1,500 population) adjacent to the southeastern shore of Lake Mweru. I then spentfour months in Lukwesa, which is situated about 130 km south of Kashikishi on the Luapula River, which flows into the lake. Finally, I spent an additional month in other smaller villages along the river. By these changes of locale I attempted to cover the area researched by Cunnison in 1949 (published in 1959), and to record some of the variability in kinship, economic, and religious activities within the river and lake region.

In January to March 1975 I visited certain Fante food-growing villages near Cape Coast, in southern Ghana, with the intention of investigating the organisation of food-farming, with special reference to the position of women farmers. In the event, as I shall relate, much of my time was spent studying the outward migration which involved nearly all young men and women. In this short article I shall deal with the main sociological factors which account for what I can only assume to be the extreme inefficiency of most food-farmers in these villages in terms of crop yields per cultivator.

Many strategies of rural development in Africa propose greater reliance on local institutions (Girard 1963; Seibel and Massing 1974; The World Bank 1975: 33–40; Tall 1976). Sometimes these proposals are primarily political rhetoric (Migot-Adholle 1970; Cliffe 1970); but they may, on the contrary, genuinely take account of rural social organization and history (Hamer 1967; Magid 1972; 1976: 154-69). Despite their merits, such proposals often fail to consider sufficiently the changing social and economic bases of rural institutions. One factor which deserves more emphasis is the dependence of rural dwellers on wider economic forces. As producers of cash crops, purchasers of imported goods, and suppliers of wage labour, rural Africans are integrated into the national and ultimately international economy. Integration and dependence are facilitated and encouraged by numerous government policies. Peasants are increasingly drawn to the cities by the symbiotic, subordinate relation of rural villages to the urban capitalist export sector, whose interests require a supply of cheap labour (Amin 1974a; 1974b; Meillassoux 1975; Seidman 1974) and predominate in the making of policy. The consequent dislocation and disintegration of rural communities limit severely the possibilities of delegating new tasks to existing rural institutions or of using them to decentralize authority, foster mass participation, or promote self-reliance.

In the protection of their kaya home deep in the forest of the Mombasa hinterland, the Giriama of Kenya developed a non-centralized government based on a council of elders (kambi), and derived from age-sets. They were supported by a special secret society within the kambi whose oath and select membership were used to maintain order and determine guilt in difficult situations. In the period from approximately 1700 until sometime in the nineteenth century, the Giriama developed this form of government by drawing upon their previous experience of life in Singwaya and subsequent southward migration, upon their identification or assimilation of members into six original clans, and upon their unique environment of the kaya neighborhood to become successful cultivators, keeping cattle when circumstances allowed and emerging as the prominent traders of the Mombasa hinterland.

Public masquerades constitute extremely important symbolic forms among the Mende of Sierra Leone. Masquerades are a means of mediating between the secret Societies which dominate Mende social life and the general community. Through masking performances the public is kept informed of important events which occur in the secret domain and is allowed carefully limited participation in the experience of Society members. The masquerades make visible the powerful ‘medicines’ (haleisia) of the secret Societies without revealing their essential mystery. Maskers personify and dramatize the powers of the ‘medicines’ and exact respect and tribute from the spectators. At the same time, participants and audience are drawn into a common experience, which is aesthetically heightened by techniques of theatre, dance, music, and the plastic arts. The resultant feeling of unity and harmony helps to overcome the threat of disunity implicit in the division of the community into separate secret factions.

This paper attempts to clarify a little further certain of the symbolic categories of the Banyoro of western Uganda, and at the same time to refocus attention on some of the methodological difficulties involved in the analysis of ethnographic material through the framework of the binary mode of symbolic classification. This entails some consideration of a recent contribution to this journal by Professor Rodney Needham, in which he attempts to rebut my detailed criticisms of his version of some aspects of Nyoro symbolism (Needham 1976; Beattie 1968; see also Needham 1967; 1973; and Beattie 1976: 232 n. 4). In doing so he repeats certain of his earlier misrepresentations, as well as introducing some new ones. He also asserts, of a number of my own—and some other—publications on Nyoro ritual and symbolism, to which I had drawn attention, that ‘they were either not relevant to [his] train of argument or, in general, were redundant to [his] primary sources’ (Needham 1976: 237). These assertions must be very briefly considered before I go on to deal with the further arguments he now advances.