More than a million people occupy an area of roughly 68,000 km² in the semiarid savanna of northwest Kenya and practice mobile pastoralism by keeping cattle, camel, sheep, goats, and donkeys. They are divided into 18 distinct social groups each comprising several thousand people. Each group has its own territory within which group members freely graze their animals. Crucially, despite the ubiquity of firearms and the individual benefits that Glowacki emphasizes could be had by raiding, there is peace among these groups. As defined in the target article: They have ongoing interactions marked by infrequent violence with harmonious relationships not enforced by the threat of (intergroup) violence. In fact, there is rampant intermarriage and trade between these groups, negotiations frequently occur between elders of these groups to share grazing areas during the dry season, and most conflicts that arise between individuals from different groups are resolved without violence.

These groups are different ngitela, geographically defined social divisions of the Turkana. While there are persistent peaceful relations among the ngitela, there are also norms and institutions that promote warfare. In the part of Turkana where we do our research, about half of adult male mortality is due to intergroup combat (Mathew & Boyd, Reference Mathew and Boyd2011), and about 40% of male participants have visible bullet scars from combat (Zefferman & Mathew, Reference Zefferman and Mathew2021). Norms reward the killing of out-group members, especially men and boys (Zefferman & Mathew, Reference Zefferman and Mathew2020). What explains this seeming contradiction? While there is peace between Turkana ngitela, there is often war between Turkana and neighboring ethnolinguistic groups, such as the Toposa. There are norms and cultural institutions that promote peace at one level of organization and norms that promote war at another. It is hard to discuss war and peace without discussing the levels of organization at which these happen.

There is also a seeming contradiction at the heart of the target article. Glowacki proposes that peace is a prisoner's dilemma where warfare results in private gains to the participant, but the costs are shared widely by the participant's group. This is different from the usual assumption that war is the dilemma with costs borne by individuals and benefits accrued to the group. One gets the impression from the article that peace is the real dilemma or at least a coequal one to war. In Glowacki's framing, groups must invent cultural institutions and norms that promote peace as opposed to war to overcome war's private benefits. However, the private benefits to the individual, as Glowacki frames them, are themselves the result of cultural institutions and norms to promote warfare – specifically norms that reward warriors and punish cowards. Additionally, the group costs are themselves the result of norms that promote escalatory tit-for-tat revenge. In short, to the extent that peace is a collective action problem at all, it is because societies have culturally evolved norms that solve the collective action problem of war! To attain peace, why not just drop the war-promoting norms instead of inventing new contradictory norms? The target article does not explain this contradiction.

We do not disagree with much of the target article's thesis. However, we think that (1) it exogenizes the evolution of some norms (those that promote war) while endogenizing others (those that promote peace); (2) it is ambivalent about defining the social scale of groups and at what levels of organization we should expect more intragroup conflict or intergroup cooperation; and (3) it does not identify a mechanistic process for the patterning of war and peace.

We have argued (Zefferman & Mathew, Reference Zefferman and Mathew2015) that group-structured cultural selection can help clarify the diverse forms of intergroup relations in humans, and provides a mechanism for the evolution of both war and peace. In brief, rather than a war-oriented psychology or a peace-oriented psychology, humans have a norm psychology, a propensity to recognize and adopt the cultural norms of one's local group. Norms allow cultural adaptation to novel local circumstances much faster than genes alone would. These adaptive forces maintain between-group variation in norms at multiple social scales. Norms that increase the success of their adherents will spread at the expense of less successful ones. When group-level interactions are strong, norms that benefit the group over the individual will be favored.

Group-structured cultural selection can yield war or peace based on which norms lead to the success of a group under local circumstances. If a group's success increases more by trading and intermarrying with their neighbors than by looting or killing them, those norms spread, amalgamating different social groups under a normative umbrella that supports social order at a larger scale. If it increases more by raiding and warfare, then norms encouraging raiding and warfare will spread. Most societies have a mix of norms where war is context dependent – dictating when, how, and with whom it is appropriate or inappropriate to have war or peace. If so, when should we expect to see peace, and when should we expect war? Group-structured cultural selection predicts that warfare will tend to occur along social boundaries in which cultural differentiation occurs and peace will tend to occur along social boundaries where there is little cultural differentiation. This is exactly what Handley and Mathew (Reference Handley and Mathew2020) document in the pastoral populations of northwest Kenya. Norms promote helping across social boundaries that are culturally similar and promote raiding across social boundaries that are culturally dissimilar.

The recognition that norm psychology is crucial to understanding war and peace in humans also helps resolve Glowacki's question about when peace evolved. To the extent that peace is a collective action problem solved by social norm enforcement, it likely evolved at the same time as our norm psychology – and ironically – at the same time we developed the capacity to invent norms for war. Knowing the propensities of our distant common ancestor with chimpanzees and bonobos would not be too informative. As we have previously argued (Zefferman & Mathew, Reference Zefferman and Mathew2015), “if primordial propensities for war or peace exist, they seem to be quite readily overwhelmed by local cultural norms.”