No one theory or model can explain the totality of the observed statistical associations between alcohol use and crime. Further, the existence and distribution of initial conditions for the applicability of any model or theory explaining part of the statistical association will have to be predicted or explained by other theories. Integration can be attained through joint analyses of theories on alcohol use and its concomitants and effects and theories of crime and deviance. Although studies suggest that alcohol use is sometimes causally relevant to family violence, little can be confidently said about how alcohol exerts its influence on such behavior. An analysis of the relationship between alcohol use and criminal behavior among blacks indicates that when social class is accounted for, empirical data do not support the proposition that a disproportionate amount of black crime results from drinking, with the exception of data from tavern studies. Although data show that drinking and crime problems both vary by age, with alcohol consumption and crime being greater among juveniles and young adults, few inferences about the causal processes associated with this covariation are possible. A review of the research indicates that little systematic work has been done on the setting-alcohol-violence nexus, and there is no research which uses another drug besides alcohol as a control, along with the placebo. A major weakness of most of the research is its largely atheoretical nature or its failure to specify underlying theoretical positions and assumptions. Future research must focus on the development of complex theoretical models based in a multidisciplinary perspective. About 550 references are listed along with author and subject indexes. For individual entries, see NCJ 87750-55.