The target article examines sex differences in aggression, arguing that differences in aggressive behaviour may be better explained by sexual selection theory rather than social role theory. In this commentary I examine the related question of sex differences in the developmental antecedents of aggression, and show that these too may be better explained by sexual selection theory, rather than social role theory.
Archer argues that the magnitude and nature of sex differences in aggression, which he defines as differences in aggression between same-sex individuals as well as between opposite-sex individuals, are better explained in terms of sexual selection than of social role. One of the key underpinnings of this argument is that evidence concerning the development of aggression suggests that physical aggression emerges early in life and tends to decline thereafter, suggesting that aggression is not a learned response.
The developmental perspective on aggression has also underpinned studies examining the causal influence of early experience on later aggression. Several studies have examined the extent to which certain developmental factors, such as family functioning, socio-economic conditions, exposure to abuse, and other factors can account for aggression and violence later in life (Daigle et al. Reference Daigle, Cullen and Wright2007; Fergusson et al. Reference Fergusson, Boden and Horwood2008; Howells & Rosenbaum Reference Howells and Rosenbaum2008). One of the key findings common to these studies is that there may be reliable sex differences in the extent to which certain environmental or behavioural factors may be related to later aggression. While this is a somewhat different issue than that addressed by Archer, an examination of the pattern of sex differences in the developmental antecedents of aggression shows that these differences can also be better explained by sexual selection theory than by social role theory.
Although a wide range of studies have examined the developmental processes that predispose individuals to aggression and violence (for reviews see, e.g., Emery & Billings-Laumann Reference Emery and Billings-Laumann1998; Loeber & Hay Reference Loeber and Hay1997; Tolan et al. Reference Tolan, Gorman-Smith and Henry2006), relatively few studies have identified sex differences in the extent to which certain risk factors may have differential effects on males and females in terms of predicting later violence. One such study was conducted by Fergusson et al. (Reference Fergusson, Boden and Horwood2008), using data from a longitudinal birth cohort. These researchers found that several factors predicted both perpetration of and victimization by intimate partner violence (IPV) in adulthood, including childhood conduct problems, exposure to family adversity, abuse exposure, and adolescent alcohol abuse/dependence. Importantly, however, they found that exposure to family adversity was more strongly predictive of later IPV involvement for males, whereas childhood conduct problems were more strongly predictive of later IPV for females. Fergusson et al. concluded that the data suggested a varied developmental pathway to IPV for males and females, although the precise mechanisms behind this pathway were unclear. Comparable findings were reported by Howells and Rosenbaum (Reference Howells and Rosenbaum2008) and by Daigle et al. (Reference Daigle, Cullen and Wright2007).
Social role theory (e.g., Eagly Reference Eagly1997; Eagly & Steffen Reference Eagly and Steffen1986) would predict that sex differences in the developmental antecedents of aggression and violence should reflect the differential sex-role socialization experienced by males and females. For example, under such an explanation we will expect males to be more influenced by exposure to violence or by affiliation with violent and aggressive peers (both features of the male sex role under social role theory). On the other hand, on the assumption that the socialization of females tends to move individuals away from violence and aggression, it may be expected that females will be more influenced by the weakening of social bonds via family dysfunction.
The data on sex differences in the developmental antecedents of aggression do not seem to be congruent with this position, however. For example, Fergusson et al. (Reference Fergusson, Boden and Horwood2008) found that a broad measure of family dysfunction predicted later IPV for males more strongly than females. This finding suggests that the weakening of social bonds caused by dysfunctional family processes increases the risks of violence among males relative to females, counter to what would be expected under social role theory. Furthermore, conduct-disordered behaviour in childhood predicts adult IPV involvement more strongly for females than males, suggesting that there are lower levels of continuity of aggressive behaviour across the lifespan among males than females, again counter to what would be expected under social role theory.
The question then arises: Can sexual selection theory better explain the sex differences observed in the developmental antecedents of violence and aggression? Archer argues that sexual selection theory would view variability in aggression as reflecting resources important for reproduction. In the cohort studied by Fergusson et al. (Reference Fergusson, Boden and Horwood2008), males at greater risk of later aggression were more likely to have come from dysfunctional homes in which they were at greater risk of exposure to a wide range of environmental stressors, including material deprivation. It could therefore be argued that exposure to family adversity increases violence and aggression in males by making salient resource limitations, engaging adaptive modules that serve the purpose of increasing access to resources (via aggression). Furthermore, in the study by Fergusson et al. males were less likely than females to show continuity of aggression, in terms of childhood conduct disorder being linked to adult IPV involvement. Again, this may be linked to Archer's general argument that, under sexual selection theory, there should be greater variability among males than females in terms of the effect of local environmental conditions. If we extend “local environmental conditions” to include environmental conditions across the lifespan of the individual, the greater discontinuity in males relative to females might reflect adaptation to variable environmental influences on aggressive behaviour.
In summary, I agree with Archer that an initial examination of the evidence pertaining to sex differences in the developmental antecedents of aggression appears to support the notion that aggression is primarily a product of sexual selection, rather than social role. Further research is needed to shed light on this question.
