In recent years a number of studies have observed empirical associations between the occurrence of key life events such as marriage, employment, and military service, and desistance from crime. The relationships between these life-course transitions and changes in criminal behaviour have been cited as evidence in support of social control and social learning theories of delinquency and in contradiction to alternative theoretical perspectives that downplay the significance of life events in the development of criminal behaviour over the lifespan. In this paper we develop and test an alternative explanation for the apparent impact of marriage on criminal and delinquent behaviour. We argue that transitions such as marriage might also promote desistance, in part, by enabling offenders to develop and exercise increased self-control. We then test this hypothesis using data from the National Longitudinal Survey of Youth (NLSY) and explore the implications of our findings for the study of desistance and for self-control theory.