The following study evaluates the complex association between legal involvement and mental illness. It describes a population of consumers of community mental health programs, comparing those with legal involvement to those without legal involvement, on a number of demographic, clinical and social indicators. It is a secondary analysis of data collected in studies making up the Community Mental Health Evaluation Initiative (CMHEI) in the province of Ontario, Canada. Legal involvement was a significant issue among community mental health program consumers; about one in five consumers had at least some contact with the legal system in the preceding nine months. Legally involved consumers were more likely to be in receipt of social assistance and be unstably housed than those legally uninvolved. However, there were no significant differences between legally involved and uninvolved consumers with respect to severity of symptomatology, current medication use or number of hospitalization days in the past 9 months. A predictive model compared the differential impact of clinical and social determinants upon legal involvement. Analyses failed to uncover a significant relationship between severity of psychiatric symptomatology and legal involvement. Significant predictors of legal involvement included gender, race, drug use as well as housing instability, and receipt of social assistance. Legal involvement was attributable to factors other than the severity of mental illness; these results challenge assumptions that the most symptomatically severe consumers are most at risk of legal involvement. Accordingly, the rate of legal involvement in a sample of community mental health program users must be considered in a broad context, with particular emphasis on social disadvantage.