Sex and Violence: Male Warriors Revisited

Is there really an evolutionary link between sex and violence?

Posted Feb 13, 2012

A recent scientific paper advances the Male Warrior hypothesis, according to which men are evolved to seek out violent conflicts in order to get women. In a blog here on Psychology Today, I challenged this rape-and-pillage model of human evolution. I think the evidence given for the male warrior hypothesis can be better explained by appealing to some widely accepted assumptions about human history. In a spirited and thoughtful reply, fellow bloggers Mark van Vugt and Anjana Ahuja have come to the defense of the Male Warrior hypothesis. Professor van Vugt was an author on the study that I challenged, and his reply provides a welcome opportunity for further discussion. I am grateful for that, and I will here attempt to clarify why I resist the evolutionary explanation of male violence.

Male Violence: Nature or Nuture?

First, a brief recap. According to the male warrior hypothesis, men evolved to form violent coalitions that would wage wars in order to capture women or collect resources that would make them more attractive to women. This is supposed to explain why, today, men wage all the wars and commit much more violent crime. I summarized an alternative account, which is commonly defended by anthropologists. This alternative explains male violence by appealing to historically obtained male dominance. I will call this the HOMDom hypothesis, and it can be summarized as follows: with the advent of agriculture, men became the primary food suppliers because their strength made them more efficient farmers; this allowed them to attain economic power over women and political control; once men achieve political dominance over women, they are the only ones who are in a position to wage war; and once men dominate women economically, they become the ones who must compete for property and resources.

Both of the Male Warrior hypothesis and the HOMdom hypothesis account for male violence. In my earlier blog, I suggested that the HOMdom hypothesis does a better job. I think the HOMdom hypothesis is also a simpler explanation, since all parties to this debate must accept that the agricultural revolution occured.  There is evidence that aggriculture helped men achieve power over women, and, together with gender differences in physical strength, this can account for sex differences in violence.  The male warrior hypothesis postulates innate machinery that we don't need to explain the data.

Objections and Replies

Here I will address the three objections raised by van Vugt and Ahuja.  First, though, I want to mention one more scientific finding, which they offer in supoort of the male warrior hypothesis.  They note that men, but not women, mention favorite athletic teams when providing reasons for preferring their favorite colors.  Against this, it must women have long depended on men econically, so the most obvious role that colors play in women's lives are in make-up and wardrobe, which are used to attact men.  Men are socialized to wear neutral hues and to blend in.  Thus, athletics is one of the few socially sanctioned opportunities for men to express color preferences.  In other cultures, such as the Wodaabe, men are socialized to use color for courtship, and the experimental results would probably look very different.  So, I don't find the color study compelling evidence that men are more tribal than women.  

Okay, now let's look at the three objections that van Vught and Ahuja raise against my original blog.

Objection 1:

Intergroup aggression predates agriculture, so agriculture cannot be the cause of male violence. They mention archeological evidence dating back 30,000 years.


First, let me offer an important concession and clarification. I should not have implied that agriculture is the sole source of male violence. Indeed, the fact that men are stronger than women also implies that warring groups will tend to assign men the primary role in combat. There are some cultural groups with female warriors (such as the Dahomey), but men tend to take on this role, just as men tend to dominate in any domain where physical strength matters. The advent of farming gives men an even more complete monopoly in violence: they obtain the sole political power to wage, and they become the property owners, which increases their role in competition for resources.

Second, even if men have always been the primary warriors, it doesn't follow that men are wired to wage war. Our ancestors may have lived in sparse populations, reducing chances and motives for violence. Modern human beings have been around for 200,000 years, so archeological evidence for violence comes far into our history. As far as we know, the human capacity to kill may have evolved for hunting, and only subsequently been used for intergroup violence.

Against this, van Vugt and Ahuja state that chimpanzees engage in coalitional violence, but, they also concede that bonobos, an even closer relative, may not. They also fail to note that chimps do not conduct raids to capture females.  Moreover, we should not assume that chimps are innately programmed to be warriors. Their violence may result from the fact that their territories are diminishing. In any case, it is risky drawing inferences from other species. It is more illuminating to look at human cultures, where we find remarkable variation in the frequency of violence, suggesting that there is no biological instinct forcing men onto the battlefield.

Objection 2:

Evolution can affect psychology, and there is little reason to deny that man and women have evolved to have different psychology traits. Parents can see this in their male and female children.


Evolution can certainly affect psychology, and I have not denied that some sex differences have an evolutionary origin. I've only denied that the men are evolved to be warriors. Biology is often used to imply that extant gender differences are natural. For example, women were once said to be incapable of excelling in universities, and that they were too delicate to responsibly vote.  I discuss other examples in my new book.

As for the remark about parents, it has been well documented that gender socialization begins in the first days of life, and that adults will perceive gender differences even where they don't exist. In clever experiments, adults have been asked to describe an infant who is either presented with a male or a female name. The name dictates what people see! Parents perceive their boys as more aggressive, even when they are not. This can influence subsequent aggressive behavior. A leading developmental psychologist found that parents tend to punish boys for bad conduct, whereas when girls misbehave, parents explain to them why their actions were wrong. This could increase male violence. And, of course, girls may quickly learn that they are less powerful than boys, which may lead them to adopt other more social forms of aggression. When physical and social aggression are both considered, gender differences tend to diminish. Indeed there may even be cultures where girls are as physically aggressive as boys, though this is admittedly rare.

Objection 3:

The historical explanation sets up a false dichotomy between biology and culture, but in reality culture is a product of biology because cultural learning is innate.


The HOMdom account is actually a biocultural account. It begins with the biological fact that men are stronger. The idea is that a simple physical difference like this can have downstream consequences under certain social conditions. I think nature and nurture are both important, and evolutionary psychologists agree. There are, however, two issues that divide us. We differ on what biology is contributing in the specific case of male violence (a warrior instinct or a simple difference in strength), and we would probably place different bets about the relative size of biological and cultural differences. I suspect that if we looked across societies, variation in violence would correlate more strongly with social variables than with biological ones.

I also think we should strongly resist the claim that "culture is a product of biology." It is true that we wouldn't have culture without a capacity for social learning, but it doesn't follow that biology produces culture. Culture would also be impossible without oxygen and sunlight, but it is not a product of these things.  Saying that culture is a product of biology is like saying that Shakespeare's genes wrote Hamlet.  

Evolution clearly matters, but the focus on evolutionary explanations may lead us to overlook ways in which culture influences behavior.  This is regrettable, since cultural factors are often larger than biological factors in explaining differences across groups.  In my book, I discribe many ways in which culture influences psychology, even when we keep biology fixed.  

Why This Matters: Decoupling Sex and Violence

In my original blog, I said that it is important to understand the social and historical factors that lead to violence if we want to make the world more peaceful. There is also something unsavory about the male warrior hypothesis. It echoes the highly controversial thesis that man are naturally evolved to commit rape under certain circumstances. Indeed, the male warrior hypothesis is closely linked to the hypothesis that men have a rape instinct. Its defenders claim that men commit acts of violence to secure female partners. But this is unlikely. There is little anthropological evidence for thinking that men in early human societies tended to find partners by fighting wars. In fact, in the one society most famous for this practice—the Yanomami of the Amazon—fewer than 1 percent of wives are captives from enemy tribes. Tragically, men do sometimes rape women in the context of war, but describing this as an instinct may give the impression that it is natural or even inevitable. Either claim would be mistaken. Rape is not always central aspect of warfare, and, when war rape occurs, it is often a deployed as intentional strategy of humiliation and control.  The claim that rape is natural should be regarded with skepticism.  There is little evidence that the propagation of our species ever relied heavily on coercive intercourse.

The defenders of the male warrior hypothesis might agree that men are not evolved to be rapists.  In their original paper, they suggest an alternative to the war-for-rape hypothesis.  They say that men may be evolved to wage war in order to aquire resources that can be used to attract women in their own tribes. This is the war-for-spoils hypothesis.  But, in our ancestral societies, there was no stored food and few artifacts to obtain, so there would not have been many spoils of war.  Furthermore, in such societies, women probably contributed equally to subsistence, so women did not depend man to bring home valuable goods.  Nor was war necessary for acquire territory, because the African savannah was large and fertile.  It is hard to see why waging war would have been an evolutionary advantage, much less a tool for courtship.

The hypothesis that men are warriors yokes sex to violence in a way that is both implausible and unsettling. One shudders to think that it might be used to get men off the hook for their most vile behavior. If, in reality, male violence is the result of a historically obtained male dominance, then we can combat violence by working to change the male sense of entitlement.