Introduction

Of all child psychiatric disorders, conduct problems show perhaps the strongest associations with psychosocial adversity. Both within the family and beyond, decades of research has documented links between disruptive behaviour problems and adverse environments: poverty and social disadvantage, disorganized neighbourhoods, poor schools, family breakdown, parental psychopathology, harsh and ineffective parenting and inadequate supervision all occur at higher than expected rates in conduct-disordered samples (Earls, 1994; Loeber & Stouthamer-Loeber, 1986).

The consistency of these associations is in no doubt; interpreting their meaning has proved more challenging. First, especially in relation to family and peer-based correlates, there are quite basic questions of the direction of effects: children influence, as well as being influenced by, those around them (Lytton, 1990), so no simple assumptions can be made about the direction of the causal arrow. Second, behaviour genetic studies have shown that many ostensibly ‘environmental’ measures involve genetic mediation (Plomin & Bergeman, 1991), and that risks for conduct disorder are likely to involve a complex interplay between nature and nurture. Third, the environmental factors measured in many epidemiological studies – low SES, parental discord, harsh parenting, and so forth – are cast at too general a level to be more than broad brush indicators of the processes that are actually likely to put children at risk. To understand their effects, we need to ‘unpack’ their meaning.