Pastoral development projects in Africa have done little to increase livestock productivity, or to improve the standard of living or food security of pastoral peoples. Several of the reasons for this consistently poor record of project interventions have been pinpointed by Goldschmidt (1981:53):

1. that they take no account of the knowledge of indigenous populations
2. that they are unaware and unconcerned with the goals and needs of pastoral people
3. that they ignore indigenous social organisation

In an attempt to develop a new approach to pastoral development based on the active participation of pastoralists, several non-governmental agencies have experimented in recent years with the organisation of pastoral associations based on indigenous institutions. One of the largest of such projects, and important especially in view of its potential influence on the future direction of World Bank support for the pastoral sector, is the Ethiopian/World Bank funded Pilot Project in the southern rangelands of Ethiopia (see map). The project is intended to test a low cost, participatory approach to pastoral development among Boran pastoralists.

An outline of Boran society is presented first, followed by a discussion of the history, performance and future development of pastoralist cooperatives in the southern rangelands, and the lessons of Pilot Project experience for other pastoral areas. In this paper, I argue that an understanding of Boran society and, in particular, resource management and territorial organisation, is crucial to the design and effective organisation of pastoralist cooperatives in the southern rangelands.