Subjects were administered a neuropsychological test battery and a comprehensive alcohol consumption index. It was hypothesized that subjects with poorer test performance and higher alcohol consumption would have violent arrest records. Results of main effect and interaction analyses failed to support this hypothesis. Results of regression analysis did show an additive effect for amount of alcohol consumed prior to the commission of the crime and the line judgment and rhythm measures of the neuropsychological battery. Both were significant predictors of violent criminal history. This suggests a need to scrutinize more closely the predictors of criminal violence, as well as the relationship among alcohol use, neuropsychological function, and violent crime. 5 tables and 12 references. (Author abstract modified)