Comments on "A Cross-Cultural Theory of Political Conflict and Violence"

Abstract

Ross's article attempts to assess numerous hypotheses, theories and frameworks for explaining internal and external violence through aggregate, cross-societal correlations using half of the Murdock ethnographic sample. It does not present or test a cross-cultural theory of political conflict and violence. The author has translated a number of rather sophisticated and context-sensitive theories into simplistic hypotheses linking single societal traits or behaviors to violence, which itself is simplistically aggregated as if the different aspects of either internal or external violence have no significance and are not affected differently. The expectable result is that all but the partly tautological and possibly spurious relationships wash out. Thus the author finds that harsh socialization (defined so as to include child abuse) correlates with internal violence, but it is internal violence; the author finds that factionalism correlates with internal violence, but factionalism (fraternal interest groups and lack of cross-cutting ties) could as well be outgrowths of prior conflict as potential sources of conflict. For external warfare, the links between dispositional factors and violence are less tautological, but the structural factors could well be related for reasons other than their contributions to greater or lesser external violence. For example, is endogamy a cause of external violence or a result? Is socioeconomic complexity correlated with external violence because societies in less isolated areas are exposed to both technology and more potentially bellicose neighbors? The author tries to dismiss the factors that have very low coefficients; even this is invalid inasmuch as the lack of apparent