Abstract
Previous studies have explained the transition from criminal propensity in youth to criminal behavior in adulthood with several hypotheses: (1) enduring criminal propensity, (2) unique social causation, and (3) cumulative social disadvantage. In this article we develop an additional hypothesis derived from the life-course concept of interdependence: (4) that the effect of social ties on crime vary as a function of individuals' levels of criminal propensity. Prosocial ties, such as ties to education , should deter criminal behavior most strongly among individuals prone to crime. We term this a "social-protection" effect. Antisocial ties, such as delinquent peers, should promote criminal behavior most strongly among the same, criminally-prone individuals - a "social-amplification" effect. We tested these four hypotheses with data from the Dunedin Study. In support of previous hypotheses, low self-control predicted more criminal behavior, prosocial ties predicted less crime, and low self-control predicted weaker social ties that led to more crime. In support of life-course interdependence, low self-control significantly interacted with social ties. Prosocial ties, such as education, employment, family ties, and partnerships deterred crime, and antisocial ties, such as delinquent peers, promoted crime, most strongly among individuals displaying low self-control. Our findings bear upon the generalizability of standard psychological and sociological theories of crime and on practical intervention for youthful offenders.