In the target article, Archer argues persuasively that an evolutionary psychological perspective drawing on sexual selection theory can account for observed sex differences in aggression and violence more parsimoniously than social constructionist theories. In our view, however, the case for sexual selection's role in the evolution of these sex differences is even stronger than Archer's treatment suggests, and he concedes too much to advocates of the discredited null hypothesis that female and male psyches are undifferentiated.
According to Archer (sect. 2.2.1), the “main alternative” to a selectionist explanation of the origins of the sex differences of interest is the “social role” theory of Eagly and Wood (Reference Eagly and Wood1999; Wood & Eagly Reference Wood and Eagly2002). At best, these authors can be read as offering a partial account of ontogenetic processes in sexual differentiation, which, if upheld, would complement an evolutionary account. At worst, they can be read as proposing that the only evolved differences between women and men are “physical” (i.e., non-neural anatomical differences), and if this is indeed their meaning, they are simply uninformed (see, e.g., Kimura Reference Kimura1999). In neither case have they provided a viable “alternative” to an account that gives centre stage to sexual selection.
In explaining why sexual selection should have made men more intensely competitive than women, Archer (sect. 3.2) aptly cites anatomical, demographic, and behavioural evidence that Homo sapiens evolved as an effectively polygynous species in which male fitness variance exceeded female fitness variance. Recent genetic evidence (Wilder et al. Reference Wilder, Mobasher and Hammer2004) reinforces this conclusion: In both our species as a whole and in discrete subpopulations thereof, the most recent common ancestor (MRCA) of mitochondrial DNA, inherited matrilineally, lived about twice as long ago as the MRCA of Y chromosomes, inherited patrilineally. These results provide strong evidence that individual men have consistently faced a higher risk of reproductive failure than individual women.
Archer notes (sect. 2.8) that, consistent with sexual selection theory, there is evidence that males with limited access to reproductive opportunities, or the resources required to obtain such opportunities, are more likely to resort to violence. However, there is accumulating evidence to support a more specific prediction derived from the same theoretical perspective, namely that the prevalence of dangerous confrontations should vary predictably according to variations in the local intensity of intrasexual competition and that cues to higher fitness variance should lead males to modulate their willingness to engage in risky and violent interactions with other men (for review, see Wilson et al. Reference Wilson, Daly, Pound, Pfaff, Arnold, Etgen, Fahrbach and Rubin2002; Reference Wilson, Daly, Pound, Pfaff, Arnold, Etgen, Fahrbach and Rubin2009).
In an effectively polygynous species, the intensity of male-male competition will in part depend on the extent to which the resources required to obtain reproductive opportunities are distributed equitably. Extreme inequity creates a situation where those at the bottom have little to lose if they escalate their tactics of competition, and much to gain. Consequently, cues to inequity should lead to facultative adjustments in men's willingness to employ violent and risky tactics to gain status and resources, which are the means to fitness. Consistent with this, evidence indicates that relative deprivation (as indexed by income inequality) is typically a more powerful predictor of variation in male violence than other socioeconomic measures such as percent below the poverty line or average income (Daly & Wilson Reference Daly, Wilson, French, Kamil and Leger2001). In both cross-national and more local comparisons, the Gini index of income inequality consistently outperforms most other socioeconomic predictors of homicide rates (e.g. Blau & Blau Reference Blau and Blau1982; Daly et al. Reference Daly, Wilson and Vasdev2001).
Increased willingness to resort to violence where resources are distributed inequitably is not uniquely predicted by sexual selection theory. However, in contrast to social constructionist accounts, an evolutionary psychological perspective treats such increased risk-proneness as a facultative adaptive response to situations where the distribution of resources is such that excessive risk-aversion will lead to substantially reduced expected fitness (Wilson et al. Reference Wilson, Daly, Pound, Pfaff, Arnold, Etgen, Fahrbach and Rubin2002; Reference Wilson, Daly, Pound, Pfaff, Arnold, Etgen, Fahrbach and Rubin2009).
An evolutionary psychological approach based on sexual selection and life history theory also predicts that individuals should modulate their willingness to engage in risky and violent confrontations according to cues of future survival and hence reproductive prospects – in other words, when prospects are poor, organisms may be expected to discount the future steeply in the pursuit of more immediate goals (Daly & Wilson Reference Daly and Wilson2005). Archer notes (sect. 3.1) that greater male than female mortality is characteristic of the sexually selected “adaptive complex” generated by intense inter-male competition, but the target article could perhaps have examined the implications of this in more detail. It is not just that males are likely to discount the future more heavily than females as a consequence of the sex difference in mortality; moreover, future discounting and willingness to engage in risky escalation of social conflicts are expected to vary predictably in relation to future survival prospects. Where these are poor, men should become particularly risk prone and willing to risk death in violent altercations as they compete for the resources required to obtain reproductive opportunities, and directly for the opportunities themselves. Consistent with this, Wilson and Daly (Reference Wilson and Daly1997) found that across neighbourhoods in a major U.S. city (Chicago), the best statistical predictor of homicide rates was low male life expectancy (even with homicide as a cause of death removed).
Finally, we do not understand Archer's rationale for suggesting that a sexual selection approach warrants the “prediction” that the sexes will not differ in “aggression” or “anger,” but only in how they manifest these things. Sell's research and theorizing (e.g., Sell et al. Reference Sell, Tooby and Cosmides2005), which Archer cites, clearly suggests that insofar as becoming angry entails an elevated risk of violent confrontation, we may expect people to differ adaptively in their readiness to anger. Why should this proposition not apply to sex differences? More fundamentally, what does it even mean to suggest that men and women do not differ in “aggression” or “anger,” but only in the manifestations thereof? We lack both consensual definitions and good metrics of these states, and finding that the sexes give the same mean answer on a self-report scale of “aggression” or “anger” is uninformative. Consequently, evidence for the popular claim that men and women are equally aggressive, but that the former manifest their aggression “directly” (e.g., as violence) and the latter “indirectly” (e.g., as innuendo) is not convincing.
