HOMICIDE AND THE SOCIAL STRUCTURE: A CROSS-NATIONAL ANALYSIS OF LETHAL VIOLENCE RATES, 1950-1970

National statistics of homicide are considered more reliable when compared to those of other crimes because of the extreme seriousness with which the offense is viewed and its higher \u27reportability\u27. This study examines the socio-structural determinants of national homicide and lethal violence (homicide and suicide) rates using a comparative, longitudinal framework. While there are major problems involved, given certain precautions, it is concluded that national homicide rates can be studied in this manner. The literature in this area falls into three research traditions: One considers homicide and suicide as expressions of frustration-aggression and anomie; another relates homicide to inequality levels (which is associated with economic development); and another explains homicide in terms of a subculture which legitimates violence. Synthesizing these three traditions, broad propositions predicting the amount and direction (whether expressed as suicide or homicide) of lethal violence are presented. Structural determinants of these phenomena considered are level of inequality and economic development; systemic frustration-anomie, and the cultural legitimation of violence. The propositions were tested on data from 31 nations at 3 time points for the period 1950-1970, and an overall summary for the 21-year period. Cross-sectionally, at all 3 time points and in the overall summary, it was found that inequality and economic development were significant determinants of both the amount and nature of lethal violence. Cultural legitimation of violence was not a significant predictor. Systemic frustration-anomie was dropped from the analysis due to its collinearity with economic development. The stability of the dependent variables as well as their rank transformation measures over the 21-year period forced the abandoning of the longitudinal analysis. The implications of these findings for the three traditions of homicide research, and for comparative analysis are also discussed