Along with a renewed interest in the philosophy of place and region and the return of a political economy focused on regions as the settings for economic development, addressed in two previous reports (Agnew, 1999; 2000a), a major thrust in recent geographical studies has involved examining the political uses to which regions are increasingly put by political movements and parties. Although many autonomist and secessionist political movements are based in specific regions within existing states, it is their putative ethnic rather than regional (or other) attributes that garner attention from most of those who study them. Ethnicity, defined in religious and/or linguistic terms, allied to collective identity as a distinctive population, is seen as the most important cause of demands for increased political autonomy or 'home rule'. Increasingly, in fact, all types of rebellion and political fragmentation, from the Hutu genocide against the Tutsi in Rwanda in 1994 through the tribal break-up of Somalia in the early 1990s to the flowering of anti-state movements in Europe and throughout the former USSR, are swept into the category of 'ethnic conflict', notwithstanding the range of factors involved in the genesis and direction of the various intergroup situations to which the term is applied. In this essay I report on the emergence of a literature taking seriously the regional dimension of anti-state revolts, investigating why such revolts are flourishing, and examining the virtues and possibilities of partition and 'multiregionalism' as solutions.
II Regional resentments are not always simply ethnicAmong writers on autonomist and secessionist movements the regional or territorial element is typically played down at the expense of the ethnic or cultural differences that movements supposedly rely on to mobilize local populations. Even publication in a geography journal does not guarantee that geography is taken very seriously (see, e.g.,