A valuable study of what the term ""genocide"" means--and what the concept should include. To solve the problem of precision, UCLA emeritus professor Kuper starts with the UN Convention on Genocide (1948). There, genocide was defined as any of the following acts ""committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such"": killing members of the group, or causing them serious physical or mental harm; prevention of birth within the group, or transfer of children out of the group; infliction of conditions calculated to bring death or destruction to the group. By tracing the path of this document through the UN, Kuper shows the political processes that helped narrow the concept of genocide. For example, the West wanted a provision labeling the destruction of political groups as genocide, but the Soviets blocked it; conversely, the Soviets wanted cultural genocide included, and the West blocked that. Kuper concludes that nations don't want to block their own possible future actions--and argues that political groups should be included in the definition. Then, accepting the UN definition as a starting point, he turns his attention (via historical instances of genocide) to the 20th century and the nature of genocide. It is a feature of plural societies, he contends--of societies that contain ""persistent and pervasive cleavages"" between the racial, ethnic, and/or religious groups that compose them. In cases of extreme pluralism, the divisions between these groups are overlaid with divisions between power, and genocide is one possible means by which an elite can maintain itself. The classic case of this is the Turkish massacre of its subject Armenians, the first of two cases Kuper explores in some depth. The second, the German extermination of Europe's Jews, differed in degree; it knew no national boundary, employed more sophisticated technology, had a more developed ideology, and involved, in the case of the German Jews, a group that appeared to be relatively assimilated with the dominant group. In a separate chapter, Kuper explores what he calls ""related atrocities,"" which include the assaults on political groups that he thinks rightly belong under the heading of genocide. Prominent in this category are Stalin's massacre of ""kulaks"" and the Indonesian massacre of Communists after the aborted 1965 coup, as well as the selective Poi Pot slaughter of Cambodians. In a final chapter, Northern Ireland (where the British government has intervened) and South Africa (where international opinion has an effect) serve as examples of potentially genocidal situations where genocide has been effectively averted. Kuper concludes, consequently, that genocide is a function of structural conditions rather than innate human capacities. If that is a modest conclusion, it is not meaningless; and the discussion generally breaks new ground to serious effect.