Van Lange et al. present a model wherein ecological variations (climate) predict cross-societal variation in aggression. We agree with Van Lange et al. that an ecological framework can provide ultimate explanations for such variation in aggression, as well as a wide variety of other human behaviors. However, any approach seeking to understand cross-cultural variation in human behavior via an ecological framework would do well to consider a wider array of theoretically relevant ecological variables (e.g., Grossmann & Varnum Reference Grossmann and Varnum2015; Varnum & Grossmann Reference Varnum and Grossmann2016a; Reference Varnum and Grossmann2016b).

To illustrate our point, we draw on established work in behavioral ecology, evolutionary biology, and animal behavior to identify two ecological features that might play a role equal to if not larger than that of climate in explaining cross-cultural variation in the frequency of two types of aggression (e.g., Clutton-Brock & Parker Reference Clutton-Brock and Parker1995; Daly & Wilson Reference Daly and Wilson2001; Emlen & Oring Reference Emlen and Oring1977; Fisher Reference Fisher1930; Griskevicius et al. Reference Griskevicius, Tybur, Ackerman, Delton, Robertson and White2012). Here we focus on income inequality (a marker of the distribution of resources in an ecology) and adult sex ratio (the adult male-to-female ratio in an ecology) as predictors of homicide (typically male-male violence) and rape (typically male-female violence).

Income inequality has been touted as arguably the best predictor of variability in violence across nations, with greater inequality linked to greater (typically male-perpetrated) violence (e.g., Daly & Wilson Reference Daly and Wilson2001; Kenrick & Gomez Jacinto Reference Kenrick, Gomez Jacinto, Gelfand, Chiu and Hong2013; Ouimet Reference Ouimet2012; Wilson & Daly Reference Wilson and Daly1997). Because males experience greater fitness variance, they engage in more risky, frequent, and intense intrasexual competition (e.g., violence) for status and related mating opportunities. Income inequality is thought to exacerbate male mating competition, and thus male violence, by increasing the perceived benefits of high-risk competition, perhaps especially when lower-risk routes to status are unavailable and/or yield unsubstantial gains (e.g., Daly & Wilson Reference Daly and Wilson2001).

Compared with income inequality, the link between biased sex ratios and violence is equivocal; previous research has found variously that male-biased sex ratios are associated with more, less, or no differences in violence (e.g., see Schacht & Mulder Reference Schacht and Mulder2015). There may be a stronger case, however, for linking male-biased sex ratios to higher rates of sexual violence against women. Previous work has argued that more males means more male mating competition, increasing the likelihood that some males resort to rape to obtain mating opportunities and/or to intimate partner violence to prevent the loss of existing relationships (e.g., D'Alessio & Stolzenberg Reference D'Alessio and Stolzenberg2010; Messner & Blau Reference Messner and Blau1987; Thornhill & Palmer Reference Thornhill and Palmer2000; Trent & South Reference Trent and South2012).

We gathered archival data on 2009–2010 income inequality (World Bank 2015), 2010 adult sex ratios (ratio of males to female, aged 15–64 [Central Intelligence Agency 2016a]), and climate (total climatic demands [Van de Vliert Reference Van de Vliert2013a]). Zero-order correlations are presented in Table 1. We tested the relative contribution of these ecological factors to cross-societal variations in rates of intentional homicide (World Bank 2016b) and rape (United Nations Office on Drugs and Crime 2011) in 2010, using multiple regression analyses in which all three predictors were simultaneously entered and only countries with data on all three predictors and the dependent variable were included. Data for rates of intentional homicide and all three predictors were available for 87 countries, and data for rates of rape and all three predictors were available for 39 countries (all data available at https://dataverse.harvard.edu/dataverse/Krems_Varnum_2016_BBS_Commentary).

Table 1.

Income inequality, adult sex ratio, climatic demands, intentional homicide rates, and rape rates: Correlations

^† p < .12, *p < .05, **p ≤ .01, ***p ≤ .001.

Multiple regression analysis revealed that the strongest predictor of homicide rates was income inequality, β = 0.402, p = .001. Male-biased sex ratios, β = −.176, p = .070, were marginally associated with homicide rates, but climatic demands were not, β = −0.139, ns.Footnote ¹ Income inequality was also the strongest predictor of rape rates, β = 0.571, p = .003. Male-biased sex ratios were also a significant predictor of rape rates, β = 0.363, p = .018, but climatic demands were not, β = 0.219, ns.

Drawing on the cross-disciplinary ecology literatures, we identified two ecological features with established theoretical and empirical links to violence. These features were comparatively better predictors of cross-societal variation in homicide and rape rates than was climate. Although these findings are consistent with the idea that we can use ecology to understand cross-societal variations in aggression, they also illustrate the importance of considering multiple ecological dimensions in such models.