Human warfare is shocking and an evolutionary puzzle, via Getty Images.

The most atrocious acts of violence humans commit have been in warfare. Through the course of human history we have left countless children orphaned and violently raped millions of women. We have found untold means to torture enemy combatants deliberately inflicting pain beyond what most living organisms may have experienced. We have displayed the skulls of our enemies as trophies in our homes, or worse, used them as cups to consume our beverages. It seems that few things we do are as morally depraved as our behavior in warfare.

Yet, it is not the egregious violence and moral depravity that makes human warfare stand out. Deliberately torturing others may be a special human quality, but there is ample violence, injury and pain endured by animals in the struggle to obtain resources, reproduce and avoid death.

What is truly shocking about human warfare is that large numbers of reproductively capable, unrelated, and unfamiliar individuals die in combat for benefits that are widely shared. From our closest living relative in the animal kingdom, to the highly cooperative eusocial insects—no animal cooperates in war in this manner.

Sign up for our newsletters

I wish to receive updates from:
Newsletter



Chimps raid neighboring communities, but in the several decades of observing them, no chimp in the attacking party has been killed. They only attack when they outnumber the opponent sufficiently so that the attackers are unscathed. And the chimps that gang up for a raid know each other well, as they hail from the same community.

Ants readily sacrifice their lives in inter-colony battles, but the ants that do so are sterile individuals. They are giving up their lives to increase the fitness of the reproductively capable queen they are genetically related to.

Reciprocity and relatedness suffice to explain chimp and ant wars. Human warfare calls for a novel explanation.

But does human warfare stand out in the animal kingdom if kings, states, and other centralized political institutions are taken out of the picture. Perhaps our weird behavior is a result of powerful rulers who can coerce us to do anything, including give up our lives.

Answering this question has taken me a place in East Africa where different pastoral societies wage wars for cattle, pastures and water. The Turkana, the people I work with, are egalitarian herders. They make a living in the semi-arid savanna of northwest Kenya by keeping cattle, camel, goat, and sheep, and seasonally moving to find pastures and water. Periodically they mobilize and raid other settlements to acquire cattle and pastures, and to take revenge for previous attacks.

These attacks give the impression that human warfare does indeed require a novel explanation. Turkana warriors are not coerced by any authority. Yet in some areas of the Turkana one out of five males die in warfare. Of the males who survive to adulthood, one out of two die in warfare.

You may be tempted to think that in an egalitarian small-scale society everyone is either a friend or relative, and so this is simply cooperation with one’s kith and kin. But this is not the case. The Turkana number a million people, and are divided into about two-dozen different sub-territories. On Turkana raids hundreds of men from different territories come together. For a typical warrior most of his fellow combatants are neither kin nor close associates. Many are strangers.

So, really, why do these men go on raids, trusting that the strangers they are fighting with will do their part?

Some may say it is obvious why these men participate in warfare. After all, cattle are food, wealth, and the path to marriage. And cattle have feet—drive them away and you can make a fortune overnight. Not only so, without a fight they would lose their territory, and what is life for a herder without good pastures? And lets not forget, it is reproductive-aged men wielding AK-47s who go on these raids. The mix of youth, testosterone, and firearms—how can war not transpire?

Yet, acknowledging these motives—cows, pastures, and firearms—gets us only so far. AK-47-wielding, young, unmarried men have plenty of reasons to have a dustup with other men of their community. They share pastures and water, and vie for the same women. Yet, in quarrels with each other, they put aside their AK-47s, and hash out disputes with their herding sticks and wrist blades.

If you think it is the desire for cows, then consider that there are cows everywhere. The neighboring family has cows, the settlement across the river has cows, and herders in distant Turkana settlements have cows. Yet, Turkana men pass up on these hundreds of thousands of cows, and instead will travel large distances until they reach the settlement of people who do not consider themselves Turkana, before they raid cattle.

And yes, territory is precious. But, remarkably, Turkana from one territory typically allow Turkana from other territories to graze in their pastures, and such sharing is especially common in the dry season when grass and water are scarce. Yet, if the Toposa encroach, the Turkana of the area will mobilize a retaliatory raid.

Earlier in this post I noted that warfare is where moral depravity seems to abound. But perhaps the question to ask is why we have moral concerns at all, and why they extend to an arbitrary set of people who are neither relatives nor friends. Why does a Turkana herder pass up on the cows of some distant stranger, to go and raid the cows of some other distant stranger? Why use sticks to fight with some people, and AK-47s to fight with others? Why let some strangers graze in your scarce pastures and kill others for venturing too close? And is that set of people we have moral concerns towards just arbitrary, or is there some logic to our moral inclusivity?

Answering this can help make sense of a lot of the violence that we want to understand and limit. It would be a place for evolutionary thinking to make a useful contribution. And it has. Over the last couple decades, the field of cultural evolution has developed a game-changing idea—the theory of cultural group selection. Posited originally by Peter Richerson and Robert Boyd 1, and honed further by Joseph Henrich 2, the theory reveals that the cultural capacity of humans creates conditions for group selection to occur. Not genetic group selection, but selection among culturally distinct groups. Peter Turchin has applied this theory to answer questions of human history such as why empires rise and fall 3, and how cooperative states emerged 4. My work on Turkana warfare provides empirical support for cultural group selection in a non-state society 5. Together with Matthew Zefferman I’ve posited that cultural group selection can subsume existing evolutionary theories of warfare and account for many of the bizarre features of human warfare 6.

There is more to be done to evaluate the theory of cultural group selection…but as of now the theory tells us that the moral sphere of humans readily extends to include culturally similar people. This is useful because it implies that we could possibly expand the moral sphere by creating perceptions of cultural similarity. Finding the common thread that connects disparate cultures may not be just a cliché, but an evolutionarily backed-up path to peace.

Works Cited:

  1. Boyd, R. & Richerson, P. J. Culture and the evolutionary process. (University of Chicago Press, 1985).
  2. Henrich, J. Cultural group selection, coevolutionary processes and large-scale cooperation. J. Econ. Behav. Organ. 53, 3–35 (2004).
  3. Turchin, P. War and Peace and War: The Rise and Fall of Empires. (Plume, 2006).
  4. Turchin, P. Ultrasociety: How 10,000 Years of War Made Humans the Greatest Cooperators on Earth. (Beresta Books, 2015).
  5. Mathew, S. & Boyd, R. Punishment sustains large-scale cooperation in prestate warfare. Proc. Natl. Acad. Sci. 108, 1091–6490 (2011).
  6. Zefferman, M. R. & Mathew, S. An evolutionary theory of large-scale human warfare: Group-structured cultural selection. Evol. Anthropol. 24, 50–61 (2015).

Published On: January 22, 2016

Sarah Mathew

Sarah Mathew

Sarah Mathew is an Assistant Professor in the School of Human Evolution and Social Change at Arizona State University. She is an evolutionary anthropologist who combines formal models and field studies to explore how humans evolved the capacity to cooperate in large groups of genetically unrelated individuals. She has conducted in depth field studies among Turkana pastoralists in Kenya examining why warriors risk their life in warfare, which has illuminated the role of moral sanctioning and cultural evolutionary processes in maintaining cooperation in small-scale societies. Currently she is spearheading an empirical study among the Turkana and other small-scale societies to examine what is the scale of cooperation, group-beneficial norms and moral sanctioning mechanisms. This research program will help provide a more detailed picture of the extent to which the cultural boundary is also the moral boundary, data that will be critical for evaluating theories of how culture enabled human cooperation. She was among thirty one scholars nationwide who were recently awarded the Andrew Carnegie fellowship in recognition of work that would contribute towards a better understanding of the challenges facing democracy and international order.

5 Comments

  • anon says:

    I am surprised to see the Turkana called “egalitarian” considering they have large inequality in wealth (livestock) and wives, and elders wield disproportionate social influence. Pastoralists are generally considered to have “significant inequality” (Kaplan, Hooper, and Gurven. 2009. Phil Trans). Not every group can be egalitarian but that doesn’t lessen the importance of the questions being asked.

  • Carmi Turchick says:

    The main problem with your ideas here is that cultural groups are not replicators, and only replicators are units which can be selected in an evolutionary process. A further large difficulty is that you are talking about a behavior which evolved when our ancestors, as far as anyone has ever suggested, lived in groups where everyone was related or knew each other well. So the adaptations did not arise in a group of a million and how it functions in a group of a million today is only informative of how it was adaptive then. Further, cultural variation between groups which would make a significant fitness difference would have been exceedingly rare during the course of our evolution.

    The critical issue here in terms of how these behaviors evolved is group selection, because they really cannot be adaptive by individual level selection. Cultural group selection is a suggested solution to what is actually a simple confusion about group selection. While much effort has been dedicated to the meaning of the term “group” in fact the confusion has been with the term “selection.” Selection can be two distinct questions, what is selected and how it is selected (the source of the force of selection). So, what is selected? Only replicators can be selected, so the answer is genes and/or individuals. Not groups, not cultural groups. But how is not the same question or the same answer. If I asked how flight evolved in birds and you told me “by selection at the gene level” you would not have answered my question. But the anti-group selection scholars have been giving us just this answer to our questions of how altruism and warfare and conformity and other pro-social behaviors evolved for decades now.

    Humans compete for fitness limiting resources, mating opportunities, and survival at the group level. The theory of evolution tells us that competition for any of these generates the force of selection, the how of selection. There must then be force of selection generated at the group level in humans, or the theory of evolution is wrong. The genes are selected by the force of selection, not by the unit of selection. The result is AS IF the group was what was selected. Groups are not what is selected, but they are how these behaviors are selected.

    Once we properly see the difference between the two questions we no longer need cultural group selection or memes or claims that warfare is not an evolved behavioral predisposition in humans. Observation matches theory.

    Finally, the difficulty with culturally defined groups is that the purpose of defining a group that way is to allow for groups larger than kin defined groups, but the process entails both positive inclusion and automatic negative exclusion. We define both who is included and at least one group who is absolutely excluded. Often these groups can be incredibly similar, nearly identical culturally. The Crips and the Bloods come to mind. But for whatever reason they were rival groups and distinguished themselves by either a red or a blue bandanna. Similarly, there seems to have been no actual difference between the groups that had bloody riots over chariot teams in ancient Byzantium. A central evolved role of culture is to function as a means for establishing group boundaries. Emphasizing how similar two groups in conflict are will generate more cultural variation on both sides.

  • Tim Tyler says:

    OK, so briefly: manipulation via cultural kin selection – i.e. kin selection based on shared memes rather than shared DNA – explains group solidarity in warfare. It is not necessary to invoke group selection at all. Most theorists these days agree that group selection produces no new predictions over what are provided by kin selection theory. Kin selection is 40-year old orthodoxy, while group selection is mired in controversy. I see no good reason for using the group selection terminology – except perhaps to create noise by courting controversy. There’s a reason why group selection models and terminology were rejected by mainstream biology: they cause confusion. Kin selection effects decrease rapidly with decreased relatedness. Group selection terminology obscures this important fact; and instead promotes group-level functional approaches – which are rarely appropriate.

    The human sciences are a bastion for group selection. The most plausible reason for this in my opinion is cultural evolution’s scientific lag. In many respects, cultural evolution lags by decades behind mainstream evolutionary thinking. Cultural evolution has yet to experience its eqivalent of the kin selection revolution of the 1970s. When it does so, naive group selection will most likely be wiped off the map – all over again.

  • Chuck Willer says:

    How does group selection theory work with the research strategy of Cultural Materialism?

  • Bryan Atkins says:

    Props Dr. Mathew. Nice work.
    Think selection operates on the variations of possible physical transformations of aggregate information structures. Groups, genes, words, equations, trees, bacteria, legal codes, tech, nation-states, etc., are aggregate information structures. They have relationship interactions. The interactions generate variations of physical transformation, i.e., novel variations of aggregate structures of information, whether genes, software, trees, nation-states, tech, atmospheres, oceans, constitutions, etc. Some of these aggregate structures of information get selected. Processing the information of relationship interaction is “The Replication” that is selected.
    Relationship interaction is a form motion, or information processing:
    “Any final state contains information about the system’s initial state and about what has happened to it since. So, the motion of any physical system, because it obeys definite laws, can be regarded as information processing.” physicist David Deutsch

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.