Background: Antisocial personality disorder (ASPD) identifies adults with persistent offending behaviour and social dysfunction. However, it lacks discrimination within high-risk and criminal populations and gives little indication of an individual's history of violence. Existing measures of violence have significant limitations. The Liverpool Violence Assessment (LiVA) is an investigator-based standardised interview for measuring patterns of violence. Method: A total of 61 male prisoners who had been sentenced for serious violent offences were interviewed using the LiVA and the Structured Clinical Interview for DSM IV antisocial personality disorder and alcohol and drug dependence. Official records of offending were examined. Results: The inter-rater reliability for the LiVA was high. There were significant correlations between histories of violence assessed by the LiVA and official records, but the frequency of self-reported violence was much higher than in the official records. Antisocial personality disorder was associated with increased violence. However, analyses revealed marked variability of the levels of violence among those with antisocial personality disorder and contrasting patterns of association of violence with antisocial personality disorder depending on the context. Conclusion: The LiVA is a reliable and valid measure of the patterns and characteristics of violence. The findings suggest that the causes of violence should be studied in their own right and not only as a feature of ASPD.