According to G. K. Nukunya, the contemporary social organization of the Aŋlɔ Eve—a people located in the southeastern corner of the Republic of Ghana—is ‘based on a [segmentary] lineage system … generalized at a higher level in clanship.’ (Nukunya 1969:20). The clan or hlɔ is the largest social unit in the Aŋlɔ social system and is defined as a group of males and females who observe the same totems and taboos, worship the gods of the particular clans to which they belong, and who claim to be the descendants, through approximately eight to ten generations, of a common putative male ancestor. Fifteen clans, dispersed throughout the towns, villages and hamlets of Aŋlɔ are recognized by the traditional political authorities as part of the hlɔ system. They include the Lafe, Amlade, Adzovia, Bate, Like, Bame, Klevi, Tovi, Tsiame, Agave, Ame, Dzevi, Vifeme, Detsofe and Blu.

The emergence of small-scale cash crop producers throughout West Africa is of central importance to those historians, anthropologists and sociologists who are working on change in the political economies of various parts of West Africa.

This discussion grows out of an attempt to account for the presence among the Suku of Zaïre of a rather common but somewhat puzzling feature of African ‘ancestor cults.’ For all the importance of ancestors in African traditional life and thought, there is surprisingly little elaboration of the mode of their existence in the other world. In a masterly discussion of ‘ancestor worship’ in Africa, Fortes (1965: 126ff) notes this fact but does not attempt to explain it. To be sure, the issue is distinctly secondary to his main concern—the sociological matrix of ancestor cults. In sociological perspective, what matters about ancestors is their social role in this world, not their style of life in the next. This sociological perspective is in tune with African ethnographic fact: like Fortes, African informants are primarily interested in the social relationship of the ancestors with the living and they show little sympathy with the ethnographer's occasional curiosity about the details of the ancestors' own existence.

The Ewe-speaking people live in the southern part of the Volta Region (Ghana) and Togo. They dwell in nucleated settlements (villages) of substantial sizes (varying from a few hundred to 16,000 inhabitants in Keta, their largest town). A group of such villages (from as few as one—hardly a group!—to as many as 116 in Anlo) form a larger entity dubbed ‘traditional area’ or ‘Division’ by the British colonial administrators. These Divisions were defined as ‘the groups of villages acknowledging a common fiaga or otherwise-titled Paramount Chief', and there are over one hundred of them in the whole of Eweland. The prevailing ethnographic and administrative opinion has been to equate these Divisions or ‘traditional areas’ with the precolonial Ewe sovereign political groups.

There is a growing acceptance of the position that low resource farmers act and make decisions in a manner consistent with economic ‘rationality’ (Helleiner, 1975; Schultz, 1964). As a result, agriculturalists are becoming increasingly concerned to design technological innovations which are consistent with the needs and means of low resource farmers (CIMMYT, 1980; Zandstra et at., 1980). In particular, constraints due to the quantity or intensity with which a farmer can cultivate his land (land and labour scarcity), his low asset level and access to credit, other demands of the household on the farm-family resources, and the uncertain environment (both natural and man-made) within which the smallholder operates, all influence his capacity and willingness to adopt what are thought to be ‘improved’ agricultural technologies (Clark and Haswell, 1970; Reynolds, 1975; Wharton, 1968).

Students of West African rice agriculture (cf. Dresch 1949, Mohr 1969) often distinguish between the Upper Guinea coast, where wet rice has been grown for centuries in permanent swamp fields recovered from the mangrove, and a more extensive area further inland, where the predominant form has been dry or mountain rice grown by shifting agriculturalists (fig. 1). The Diola of Senegal (Pélissier 1966, Linares 1970), the Balanta of Guinea Bissau (Espírito Santo 1949), and the Baga of coastal Guinea (Paulme 1957), belong to the first category. They transplant rice in inundated fields that are desalinated, diked, ridged and irrigated. In contrast, the Mande-speaking peoples of Sierra Leone and Liberia are mostly ‘upland’ farmers. They broadcast rice on rain-fed fields that are rotated and fallowed.

The idea that gods are made by men, not men by gods, is a sociological truism. It belongs very obviously to a detached and critical tradition of thought incompatible with faith in those gods. But Yoruba traditional religion contains built into it a very similar notion, and here, far from indicating scepticism or decline of belief, it seems to be a central impulse to devotion. The òrìṣà (‘gods’) are, according to Yoruba traditional thought, maintained and kept in existence by the attention of humans. Without the collaboration of their devotees, the òrìṣà would be betrayed, exposed and reduced to nothing. This notion seems to have been intrinsic to the religion since the earliest times. How can such an awareness be part of a devotee's ‘belief’? Rather than speculate abstractly, as Rodney Needham does (Needham 1972), about whether people of other cultures can be said to ‘believe’ at all, it seems more interesting to take a concrete case like the Yorba one where there is an unexpected–even apparently paradoxical–configuration of ideas, and to ask how these ideas are constituted. Only by looking at them as part of a particular kind of society, with particular kinds of social relationships, can one see why such a configuration is so persuasive. The notion that men make gods is by no means unique to Yoruba thought. It is present to some degree in a number of traditional West African religions, and in some, such as the Kalahari one, it can be seen in an even more explicit form than in the Yoruba one. A comparison may help to show how it is the constitution of social relationships which makes such a notion not just acceptable but central to the religious thought of the society.

Un énorme travail a été réalisé sur les populations Pygmées du Zaïre. Dans ce domaine, l'ethnologie, l'anthropologie, la théologie africaine doivent beaucoup à J.-P. Hallet, à P. Schebesta, à C. M. Turnbull et á bien d'autres. Bon nombre de ces études ont diffusé et privilgié l'image prédominante du Pygmée nomadisant á travers la forêt, son milieu naturel. Cependant, elles ont occulté la connaissance et empêché l'investigation sur de nombreux autres groupes de Pygmées vivant loin de la forêt. En effet, pour Schebesta, qui fait de l'anthropométrie un critère privilégié, les seuls vrais Pygmées sont les Bambuti de la forêt de l'Ituri (1933: 63). Tous les autres ne seraient que des ‘Pygmoformes’ ou des ‘Pygmoïdes’, des ‘peuples métissés dont un croisement aura abâtardi les caractères de race et de culture’ (1940: 20). De nos jours, les critéres ont changé. Certains chercheurs se fondent sur le caractère polyphonique de la musique comme élément primordial dans la définition du Pygmée.

The term ‘Dorobo’ denotes an ethnic category embracing small hunting-and-gathering groups residing on the fringes of various agricultural and pastoral peoples in East Africa. The essence of the Dorobo's position is that they engage in economically symbiotic activities with regard to local farmers and herders, while retaining their social marginality as people of the bush. Much is known of them through the constructs of their neighbours, who assign them attributes commensurate with their marginal social position; the Dorobo are amalgamated with wild amoral creatures, and their ancestors are thought to have been in attendance at the birth of the present world-order. Their marginality therefore has economic, spatial, and temporal dimensions.

The purpose of this study is to draw attention to the large volume of monetary transactions and profits within the traditional sector of a farming community in West Africa. More specifically it deals with the production, distribution and pricing of local brews, points to the effect of beer making on farming, and relates this trade to contemporary developments in other spheres of social and economic life.

It is thought that African rice (Oryza glaberrima), domesticated from the annual wild rice Oryza barthii, a plant of savanna river valleys and water holes, may have been first cultivated along the valley of the upper and middle Niger (Portères, 1970; Harlan, 1975). Cultivation under high rainfall conditions in the forest zone followed the development of ‘upland’ cultivars. Today, farmers in the drier savannas continue to depend on the seasonal rise and fall of flood water in valley bottom land for rice cultivation, and on occasion, this has developed into fully-fledged ‘irrigation’ agriculture (Linares de Sapir, 1970; Loquay, 1980), but elsewhere in the West African rice zone rain-fed ‘upland’ cultivation techniques now predominate. Where rainfall is more reliable ‘upland’ cultivation tends to be preferred because of its greater efficiency of labour use, but, as in ancient times, the occasional supplementary use of swamp to cope with bad years is not uncommon. This paper focuses attention on rice farming systems in Sierra Leone. Rice cultivation probably reached this area several thousand years ago (Atherton, 1979). Although in a few instances regular use of swamp land dates back to the later part of the nineteenth century (e.g. in the Bumban valley and the tidal swamps of the Scarcies estuary: Fyle, 1979, Carpenter, 1978) ‘upland’ cultivation still retains its former importance over large parts of the country.

This paper has three related themes. First, the hunting peoples of the northern East African coast are defined; second, a period of origin of each group is proposed; and, third, the processes which resulted in their creation and persistence to the present are hypothesised and discussed.

Kwaara banda daarey, hala ga kano, yeow s'a gar. The stranger (no matter how long he/she has lived in a town) will never possess the choicest fruit of the daarey tree behind the village.

The daarey tree (Ziziphus jujuba) is found in and around Songhay towns in the Republic of Niger. Since its sweet red fruit is coveted, the whereabouts of the finest daarey are never divulged to foreigners, lest they steal from the Songhay that which is most delicious in their communities. The proverb underscores just how careful the Songhay are in protecting from non-Songhay that which is their own. It also suggests that just as non-Songhay will never possess the choicest daarey fruits, so too they will neither marry the most beautiful Songhay maidens nor become influential in community affairs. The most beautiful maidens, like the best daarey trees, are kept from the sight of non-Songhay; the sociocultural knowledge which enables a person to be perceived as influential is never taught directly to those who have migrated to the lands of the Songhay.

Evans-Pritchard's assessment of what contemporary social anthropologists might call ‘contextual ethnicity’ has offered many with no first-hand research experience in this region of Africa a prerogative in which some apparently revel: to opine on the bizarre and the obvious. Examples of the former persuasion are typified by articles which, with a passion for ‘theoretical purity’, openly declare an indifference to ethnographic fact (Newcomer 1972). Others attempt to account for the present distribution of pastoral Nilotes in the Southern Sudan as though ‘tribal’ groups and their tract of migration possessed the physical properties of pool balls pelleted across the swampland of the Upper Nile Basin (e.g. Southall 1976). Some must surely have had moment to wonder if the Nuer really exist at all. Of course they do, as even the uninterested tourist could note by observing the ethnic composition of labour groups in Khartoum, but only because so do the Dinka, who live north, south and west of Nuerland.

Les études de cas ont été effectuées par les auteurs de cet article en pays malinke, guro, gbã (ou gagu) et bete à l'ouest du Bandama, baule à l'est. Ces études ne sont pas toujours exhaustives et homogènes eu égard au sujet du présent article, et, de plus, elles concernent essentiellement les parties forestière et préforestière de la vallée du Bandama.