This article examines the social and political background of escalating violence between ethnic groups in south‐western Ethiopia who until recently had customary and ritually sanctioned ways of resolving conflict. It highlights the impact of the emerging state hegemony in a local setting on ethnic groups not yet involved in the global political economy. The account also indicates the changing arenas of ‘ethnic’ self‐definition and economic opportunity for local groups in post‐1991 Ethiopia. As the report of a big reconciliation meeting held between the government and the groups involved (and discussed here) makes clear, in the efforts of state agents to mediate emerging conflicts in conditions of increasing resource scarcity and identity struggle, the use of customary mediation mechanisms and their cultural symbolism was rhetorically recognised. But at the same time efficient mediation was structurally impeded by the very nature of the exercise of authority by the agents of the state and by their incapacity to implement practical measures to establish local peace. This failure to reconstitute a new political arena of conflict resolution was matched by the inability of the (representatives of the) ethnic groups concerned to redefine their relationship in a constructive and culturally acceptable manner.

This article briefly discusses the various factors that gave rise to tension in northern Ghana between Fulbe pastoralists coming in from Burkina Faso and the local Kassena farmers (who have livestock of their own). The Ghanaian government's decision in 1988–89 to expel the Fulbe resulted in more problems, not least the dilemma of how to patrol effectively the permeable borderland and the gradual introduction there of modern firearms. In consequence some argue that it is not the policies of conservation that are being called into question but the government's sovereignty.

This article examines political oral traditions in the Sefwi (Akan) area of Ghana. Two types of narrative are studied: negotiations over the political status of stools within the kingdom and the claims to succession of matrilineal branches within stools. Narratives are analysed in relation to their claims to historicity, to the political conflicts in which they are generated and to their correspondence to legal criteria of attribution of ‘traditional’ political offices. It shows that pre‐colonial dynamic norms concerning stool status and succession turned into a fixed legal corpus in the twentieth century. Contenders’ histories have been used as evidence to judge ‘traditional’ stool disputes. Narrators have thus constructed narratives presenting ideal pasts considered worthy of legal attribution of ‘traditional’ political office. Narratives have consequently legalised narrators’ claims with reference to ancient history. The study of the context of the emergence of oral traditions—hostility between particular stool holders, national politics’ influence or conflicts over the sharing of stool revenue—shows that narratives and political conflicts have a history of their own which is carefully omitted from the narration.

This article revisits S.F. Nadel's study on land tenure in highland Eritrea from the mid‐1940s, and presents the current framework of the communal land tenure system of meret shehena. Under the meret shehena system, all land under the domain of the village is perceived as the common property of the village inhabitants. To restrict outsiders access to land, habitation rights (tisha) to the villages is guided by agnatic descent, and individuals obtain usufruct rights to land through residence in the village. The article proves that since Nadel's study almost sixty years ago, descent rules defining habitation rights have been changed, in order to restrict distant descendants returning to the village and claiming their land rights. Nevertheless, the overall workings of the system that redistributes all shehena land every seventh year to the village inhabitants, shows a remarkable resemblance to the observations made by Nadel. It appears that the customary operational guidelines of the system are virtually unaffected by wars and political turbulence during the last fifty years, following the core principle that each male adult individual with habitation rights (tisha) who marries and establishes a separate household, will be entitled to an equal share (gibri) of the village land, in order to secure the livelihood of all household members.

There are two general approaches to assessing what is known as ‘development’. First, there are classical accounts focusing on Europe's development during the industrial revolution. They describe how urban areas expanded at the expense of the social and economic resources of the rural areas, disrupting an independent subsistence peasantry. A major consequence is that today all Europeans are dependent socially, politically, and economically on the modern capitalist system. The second (more common) approach to development focuses on the modern Third World. This approach assumes that, as with Europe, the entire Third World is dependent on the modern capitalist system. Development studies focus on the assessment of how Third World countries can most effectively engage world capitalism. Discussion is typically reduced to comparisons between world systems theory and neoclassical economics. The Tanzanian government has used standard policies grounded in neoclassical and world‐system assumptions since independence. But both policies failed to produce the predicted economic growth. This article argues that both policies failed because the Tanzanian peasantry, like the early modern European peasantry, is not dependent on the operation of world capitalism for basic subsistence. In fact, as studies have shown, rural Tanzania is only weakly incorporated into the capitalist world system, and in consequence has not been an easy target for what world‐system theorists call ‘peripheral integration’. What makes Tanzania different is the fact that the rural peasantry do not use market mechanisms in the distribution of the ‘means of production’, especially arable land for swidden agriculture, or, for that matter, labour or cattle.

Earth‐eating is common among primary school children in Luoland, western Kenya. This article describes the social significance and meanings attributed to it. Earth‐eating is practised among children before puberty, irrespective of their sex, and among women of reproductive age, but not usually among adult men or old women. To eat earth signifies belonging to the female sphere within the household, which includes children up to adolescence. Through eating earth, or abandoning it, the children express their emerging gender identity. Discourses about earth‐eating, describing the practice as unhealthy and bad, draw on ‘modern’ notions of hygiene, which are imparted, for example, in school. They form part of the discursive strategies with which men especially maintain a dominant position in the community. Beyond the significance of earth‐eating in relation to age, gender and power, it relates to several larger cultural themes, namely fertility, belonging to a place, and the continuity of the lineage. Earth symbolises female, life‐bringing forces. Termite hills, earth from which is eaten by most of the children and women, can symbolise fertility, and represent the house and the home, and the graves of ancestors. Earth‐eating is a form of ‘communion’ with life‐giving forces and with the people with whom one shares land and origin. Earth‐eating is a social practice produced in complex interactions of body, mind and other people, through which children incorporate and embody social relations and cultural values.