On 29 March 1978 the Federal Military Government of Nigeria promulgated the Land Use Decree, which, with immediate effect, vested all land in the territory of each state of the Federation in the Governor of that state. The land was to be ‘held in trust and administered for the use and common benefit of all Nigerians’ in accordance with the provisions of the Decree. This paper considers the origins and effects of this policy of land nationalization.

In a recent paper on ‘The state and social processes in Africa’, Lonsdale (1981) convincingly demonstrates the poverty of assuming that any one ‘grand theory’ is able to explain the complex relations among state, social process and class struggle and other equally complex social phenomena. Images of the state as parasitic (as in formulations of the ‘Asiatic mode of production’), as an epiphenomenon of property relations or as a neutral mediator, individually fail to grasp the complexity of states' organization and action (Jessop, 1977), yet each clearly evokes an important dimension of that complexity. The fact of political and economic domination has to be dealt with but equally the often contradictory quality of state power reflects the fact that the mediating institutions and ideologies between dominating and dominated are the objects and arenas for competition and struggle among different social groups and categories. The recognition that one has to deal analytically not with a monolithic entity but a set of institutions that are subject to being ‘redefined, added to and patched’ (Lonsdale, 1981: 2) is important. The challenge, then, is to devise analytical means that capture such processes and allow us to unravel the dynamic of contradictions rather than merely note their pervasive character.

Where peasant production predominates in the national economy, the state is faced with a difficult dilemma. It must derive its revenues from cash crops, but if it appropriates too much, it will drive peasants out of the cash market and into subsistence (Hyden, 1980; Bunker, 1983b). Many African states have attempted to resolve this dilemma through direct contrôl of crop markets (Bates, 1981). Market control in peasant economies, however, usually offers the major means of wealth and upward mobility at the local level, so different power groups there may challenge the state's hegemony (Saul, 1969; Hyden, 1970). I have already shown how such groups achieved significant control of markets in Bugisu, Uganda (Bunker, 1983a), and how the state responded with periodic interventions to limit their power and autonomy (Bunker, 1983b). In this article, I examine how the Ugandan state has justified its continued intervention in the local economy and how power groups among the Bagisu have legitimated their claims against the state.

The literature on general peasant–state relations in history, in Africa today, or in Kenya in particular, tells us that the peasants have little systematic political leverage over state policy. While I broadly agree with this assessment, I would like to comment on an exception. The changing nature of self-help in Kenya is analysed in order to argue, first, that peasants, in collaboration with other classes in self-help, have had some leverage on state policy and budgets in the realm of basic needs; second, that the fact of peasant initiative and political strength over time has served to alter subtly some aspects of everyday peasant–state relations to peasant advantage; and third, that self-help has historically been a doubleedged political sword. But, before proceeding to the Kenya context, it is essential to review theoretical and empirical aspects of peasant political weakness in contemporary Africa.

During the 1970s the Tanzanian state moved a very substantial part of the rural population into nucleated settlements. This was followed by the setting up of new village institutions. It was villagization in a double sense: the creation of physical ‘villages’ from dispersed households and the establishment of corporate ‘villages’ (more precisely, village councils) intended to regulate the activities of those households. It was an intervention officially justified as a means of increasing collective consumption (through the provision of health care, schools and other services) and simultaneously of transforming an essentially traditional hoebased agricultural system (by facilitating the use of modern inputs). External commentators have seen it either as a large scale example of misguided bureaucratic practice (Coulson, 1977) or as part of a process of state expansion intended to increase control over peasant production (Bernstein, 1981; Hyden, 1980). My objective here is not to provide another theoretical commentary, but to show that ultimately any overall assessment of the impact of villagization must rest upon detailed analyses of particular local economies. I have restricted my discussion of this impact to household production.