One of the main purposes of much recent work in subnational politics in Africa has teen to find levels and categories of action that are both meaningful and persistent. The study of politics at the local level may avoid some analytical difficulties associated with earlier macro-level studies (for a critique of earlier work, see Zolberg 1966), hut the problem of nonreproducibility begins where macro-analysis leaves off. One result of this dilemma is that few of the increasing number of local studies in African politics either build on previous research or relate theoretically to other work. While this random quality in the literature may be partially overcome as more local research is undertaken, for the moment one of the safest generalizations we can make about local politics in Africa is that there is great variety in social formation and political process in local political arenas.

This suggests that the problem of focus in the study of contemporary African politics cannot easily be resolved merely by shifting the research effort from the national to the subnational level. Both macro- and micro-level studies present their own challenges of validity and interpretation. What both approaches have tended to share, however, is an almost exclusive orientation to politics as the influences on, and the processes of authoritative decision-making. In fact, this orientation largely characterizes political science as a discipline; it is not confined to the study of African politics. But it is equally as important for political scientists to develop categories to analyze the substantive performance of African political systems as it is for them to analyze how the systems operate. Such categories will certainly involve the notion of public policy, to which we now turn our attention.