Scientific study of the evolution of warfare has matured greatly over the decades. One of the most interesting and exciting developments in this regard is an increasing willingness by social scientists to turn to evolutionary theory as a lens with which to investigate the origins and psychological mechanisms of war. The application of evolutionary theory toward the study of the brain has become more sophisticated as well. For example, past scientific and popular debates have been unduly distracted with questions such as whether humans are naturally selfish or altruistic. Partly stemming from misapplication of metaphors such as “survival of the fittest” and the “struggle for survival,” it was assumed that Darwinian processes could only produce selfish individuals that care not for the welfare of others. In the case of coalitional violence, early ethologists argued that chimpanzees and humans naturally strive to dominate each other and that aggression is the inevitable consequence of competitive social environments. The lesson was: we are stuck with violence, aggression, and war, so we better just learn to deal with it. Proponents of this “killer ape” hypothesis viewed instinct and innateness as a source of inflexibility and inevitability.

Unfortunately, this view persists even today, and especially in the popular press. For example, John Horgan, a prominent blogger at Scientific American, essentially argues that warfare is not “innate” but is instead better understood as a response to environmental pressures (1). Taking the opposite and equally false view, David Brooks writes in the New York Times that humans are “natural-born killers and the real question is not what makes people kill but what prevents them from doing so” (2). Again, we see the futile tug-of-war between two positions, both of which are untenable: Are we compelled to violence by our inner passions and brute instincts, or are we born peaceful yet pulled tragically into conflict by the various pressures of social life? In part, this illusory dichotomy is made possible by a false understanding of adaptation and innateness.

Modern evolutionists recognize that although adaptations are innate, they are not inflexible; indeed, psychological adaptations that regulate social behavior must prove flexible if they are designed in response to the complex selection challenges that prevailed in ancestral social environments. Thus, humans are neither egoists nor altruists. It is more likely that we possess a range of psychological adaptations designed for a range of reproductively significant and evolutionarily recurrent social situations, and that these adaptations are designed to motivate forms of altruism in certain circumstances, and forms of egoism in others. Thus, adaptations for warfare, if they exist, must render coalitional aggression a conditional response to adaptively relevant environmental circumstances. Two recently published papers, which I discuss below, seek to explain warfare from this perspective, and they provide a useful and comprehensive evaluation of the state of the art on the evolutionary science of war.

In a paper published in Philosophical Transactions of the Royal Society, Melissa M. McDonald, Carlos D. Navarrete, and Mark Van Vugt build an evolutionary framework for understanding two important pillars of the psychology of intergroup conflict (3). McDonald and colleagues begin with sexual selection and parental investment theory to explain that ancestrally, engaging in coalitional aggression was a relatively low-cost, high-benefit reproductive strategy for males more so than for females. Briefly, in any species where a stable parental investment asymmetry exists between two sexes, the low investors tend to compete over access to the high investors, and natural selection tends to favor various “weapons” (e.g. aggressiveness) for use in these intra-sexual contests. In humans, as with most primates and mammals in general, males are the low investors and the more aggressive sex. In the context of coalitional aggression, this investment asymmetry renders warfare a reproductively beneficial activity even in the face of substantial mortality, assuming that the other coalition is defeated and the pre-battle within-group distribution of risk is roughly symmetrical (4).

Given an ancestral social environment characterized by inter-group conflict in which coalitional aggression was reproductively beneficial for males more than females, McDonald and her colleagues outline two sets of hypotheses regarding the adaptations that ought to exist for warfare and how they are expected to operate. The authors then survey existing research to explore the extent to which empirical findings corroborate theoretical expectations. The first is the Male Warrior Hypothesis. If coalitional aggression was primarily a male endeavor, it is likely that adaptations for warfare operate in distinct ways depending on the sex of the individual. Adaptations in males should render coalitional aggression conditional upon cues such as personal and coalitional formidability; in-group members should be more suspicious of out-group males than out-group females, and; out-group males should tend to provoke anger and aggression among in-group males, but should provoke fear and avoidance among in-group females. Indeed, the authors find that, compared to women, men are in general more xenophobic, and are more likely to dehumanize out-group members. Additionally, when primed with out-group threat, male – but not female – in-group identification, as well as willingness to sacrifice for the group, becomes amplified. The authors also cite an interesting study in which, “men, but not women, were more likely to endorse statements supporting war after they had been primed with attractive members of the opposite sex.” McDonald et al. explain that a range of adaptations ought to exist to defend against the threat of out-group male coalitions – a set of expectations that they refer to as the Outgroup Male Target Hypothesis. The authors find empirical support for theoretical expectations derived from this hypothesis as well.

McDonald and her colleagues offer a broad theoretical framework for generating and testing adaptationist hypotheses regarding human warfare, and their paper is a unique and valuable contribution in this regard. However, there is another puzzle that most adaptationists who study warfare tend not to directly engage, which is the question of the overlap between chimpanzee and human coalitional aggression. Many evolutionists will (rightly) point to the existence of chimpanzee coalitional aggression as evidence that the origin of warfare extends back even before the hominid-chimpanzee split. Without disagreeing with this claim, of course, it is noteworthy that the predominant form of chimp coalitional violence is the lethal raid, initiated in the backdrop of inter-group competition, and always when a favorable size ratio (3:1) permits low-risk victory by the larger coalition. Humans, on the other hand, demonstrate a remarkable willingness to engage in coalitional violence even when the balance of power between coalitions is relatively symmetrical. This is a puzzle that requires explanation and is not well captured by extant models of chimpanzee coalitional violence.

A paper in Human Nature by primatologist Richard Wrangham and Luke Glowacki is the first to engage this puzzle directly (5). Wrangham has previously established the “imbalance of power” hypothesis, which effectively explains the distribution of coalitional violence among chimpanzees (6), and with Glowacki, Wrangham examines the pattern of human coalitional aggression in order to discover whether the chimpanzee model is sufficient to explain human warfare as well. Where group-based territoriality exists, and groups are subject to variation in size and stability, the chimpanzee model predicts that humans (like chimpanzees) evolved adaptations to conditionally engage in low-risk, high-probability of success coalitional aggression. However, to the extent that humans engage in riskier forms of coalitional aggression, alternative hypotheses to the chimpanzee model must be explored.

Wrangham and Glowacki find that humans (predominantly males) do engage in species-typical forms of low-cost/high probability of success lethal raiding, and that the fitness benefits associated with such endeavors have indeed been substantial. This suggests the existence of psychological adaptations in humans for this form of coalitional aggression, and is consistent with the chimpanzee model. However, the authors also find that humans engage in forms of high-risk aggression between coalitions of relatively symmetric size, which is not predicted or explained by the chimpanzee model. These contests between human coalitions are not only dangerous because of the power symmetry between coalitions, but also because of social dynamics such as revenge, which can turn random incursions into multi-generational battles of attrition. The authors therefore conclude that two classes of explanation are possible. First, it is possible that psychological adaptations exist that were designed in response to selection pressures unique to human evolutionary history and which conditionally motivate high-risk coalitional aggression, or; Second, that greater risk-taking on the battlefield is better explained as the product of cultural norms rather than biologically regulated decision-rules.

Wrangham and Glowacki argue that if the first possibility is correct, then adaptations in humans for greater risk-taking in battle (relative to chimpanzees) most likely emerged either through genetic group selection or cultural group selection during human (but not chimpanzee) evolution. Although the authors are not particularly explicit on this point, the suggestion here is that individual-level selection could not have favored adaptations that motivate the degree of risk-taking observed in human coalitional violence. Therefore, it may be necessary to appeal to “higher levels” of selection, such as genetic group selection, which can, under certain conditions, favor behavioral strategies (i.e. psychological adaptations) that operate to benefit the group (e.g. by motivating self-sacrifice in battle) even if such strategies are disadvantageous from the perspective of the individual. However, in another paper by Langergraber et al., in which Wrangham is co-author, cross-cultural and cross-species analysis indicates that genetic evidence of group selection in humans is relatively weak (7). Additionally, the analysis by McDonald and her colleagues above suggests that even in the face of significant mortality, it is still possible for individual level selection to favor great risk-taking in battle. From this perspective, the relevant challenge may not be to explain why humans engage in symmetric coalitional aggression, but rather, why chimpanzees do not.

Scientists continue to wrestle over these and other questions regarding the evolution of war in chimpanzees and humans, and there remains much fertile ground for future investigation. As indicated above, evolutionary theory promises a multitude of novel insights on this complex and central human phenomenon, and taken together, these two articles represent a nice cross-section of much of the leading theoretical and empirical evidence on the evolution of war.

1. See my full reply to Horgan here:

2. Link:

3. McDonald, M. M., Navarrete, C. D., & Van Vugt, M. 2012. Evolution and the psychology of intergroup conflict: The male warrior hypothesis. Philosophical Transactions of the Royal Society B – Biological Sciences 367: 670-679.

4. Importantly, the investment asymmetry is a necessary but not sufficient condition for the emergence of coalitional aggression. Even where parental investment dynamics would render coalitional aggression reproductively advantageous, coalitional activity of any kind requires a suite of adaptations that enable n-person coordination and tracking, which relatively few species possess.

5. Wrangham, R. W. and Glowacki, L. 2012. Intergroup aggression in chimpanzees and war in nomadic hunter-gatherers: Evaluating the chimpanzee model. Human Nature, March 3 [published ahead of print]

6. Wrangham, R. W. 1999. Evolution of coalitionary killing. American Journal of Physical Anthropology 110(29): 1-30.

7. Langergraber, K., Schubert, G., Rowney, C., Wrangham, R., Zommers, Z., and Vigilant, L. 2011. Genetic differentiation and the evolution of cooperation in chimpanzees and humans. Proceedings of the Royal Society B – Biological Sciences 278: 2546-2552.


Leave a Reply