concern. We next turn to examine circumstances that seriously threaten and often undermine the parenting process.

The challenges associated with parenting become abundantly evident when we take a look at parents who are struggling. There is an extensive literature on the effects of maternal depression, a relatively common, but potentially very serious, problem for parent-child interactions and child development. Child abuse and neglect represent a more unusual and far more extreme disruption in parenting. The long-standing literature on orphanage-reared children can inform questions about situations that undermine the basic and powerful tendency of infants to form relationships and the young child's capacity to recover from the effects of extremely aberrant care. These examples illustrate circumstances that place young children at risk of highly compromised development. In addition, the behaviors that are seen in young children exposed to these circumstances may be usefully considered as “canaries in the mineshaft”—that is, as early warning signs of serious problems in children who experience less extreme, but nonetheless harmful, early rearing experiences.

Research on the developmental consequences of disrupted parenting is part of a broader literature on environmental influences leading to psychopathology (see Rutter, in press; Rutter et al., in press). This is an exceedingly complex literature that is fraught with challenges regarding, for example, the need to elucidate how genetic factors interact with environmental factors to affect susceptibility to risk and the need to consider how children themselves contribute to parental behavior, including behavior that poses risks to their development. A good example of this latter point is provided by the discussion of the work by Ge and colleagues (1996) in Chapter 2, which links children's inherited antisocial tendencies to more harsh parenting by their adoptive mothers and fathers. Another example of how genetic and environmental influences interact is provided by a study of Scandinavian adoptees (Bohman, 1996). The children who were characterized by neither genetic (indexed by criminality or alcoholism in one or both biological parents) nor environmental risk (indexed by the same in adoptive parents, together with serious adversities in rearing) had a 3 percent rate of adult criminality, the children characterized by only one source of risk had rates of 6-12 percent, and those characterized by both genetic and environmental risk had a 40 percent rate of adult criminality. Environmental risk, in other words, led to negative outcomes primarily in the presence of genetic risk.

Sorting out genetic contributions and the direction of effects in research on parent-child relationships is a daunting task. Longitudinal designs,