A study of 100 primate species finds everyday aggression and killing evolved separately

Killing a rival and squabbling over food turn out to be statistically distant behaviors across the primate family tree. A comparison of 100 free-ranging, group-living primate populations, including humans, found that species with frequent everyday fights were not the ones more likely to kill. The analysis, led by Bonaventura Majolo of the University of Lincoln, appeared in Evolution Letters in March 2026.

Majolo, a professor of social evolution, and Samantha Wakes contributed equally; Marcello Ruta is the third author, and all three work at Lincoln. The journal published the paper online on March 3, 2026, in volume 10, issue 2, and Lincoln announced it on March 19, which means the October coverage is a repost of a spring result. The paper is titled “Origins of violence: evolutionary decoupling between mild and lethal conspecific aggression in primates.”

Five forms of aggression across 100 taxa

The journal record specifies the dataset as 100 free-ranging, non-provisioned, group-living primate taxa, made up of 98 species plus two pairs of subspecies. Majolo’s team scored each against five forms of aggression: mild aggression between groups, mild aggression within groups, killing of adults between groups, killing of adults within groups, and infanticide.

Majolo said understanding the evolutionary roots of violence is important not only for biology, a framing the release ties to a long-running argument over whether violence is rooted mainly in deep evolutionary history or shaped by social and cultural forces. Separating the five matters because older socio-ecological models treated aggression as one trait on a single scale, running from minor disputes to lethal attacks. Under that view, a species with frequent mild conflict would be expected to produce more killing as well, and a species that rarely fights would be expected to rarely kill. The Lincoln team set out to test that expectation against all five measures at once rather than against a single composite score.

Decoupled traits in the comparison

The data did not follow that pattern. If aggression sat on one dial, species at the high end of everyday conflict would also sit at the high end of killing; instead, the ranking by mild aggression and the ranking by lethal aggression came apart. That is what the authors mean by decoupling. Phylogeny, the shared evolutionary history among species, explained less of the variation in mild aggression than in lethal aggression. Within-group and between-group killing of adults were positively correlated with each other, and both were positively correlated with infanticide. The two types of mild aggression, by contrast, were only weakly related to lethal aggression.

Majolo said in the University of Lincoln’s release, published on March 19, 2026, that the evolution of aggression follows more complex patterns than earlier models proposed. He called it biologically inaccurate to rank species on their general propensity for aggression, and he argued that debate over whether humans are inherently violent needs a more nuanced approach. The authors’ own conclusion is that mild aggression is not closely tied to killing, so escalation from one to the other may be governed by different pressures.

The journal record also reports that analyses restricted to male attackers mostly confirmed the same links. The finding remains correlational and comparative. It describes how traits are distributed across species rather than the cause of any single killing, and it leaves the explanation for why lethal forms cluster together to later work.

Chimpanzee killing as a contrast case

Separate earlier work illustrates why lethal behavior might follow its own logic, and both studies below sit outside the Lincoln dataset. A 2014 Nature analysis of 18 chimpanzee communities and four bonobo communities recorded 152 chimpanzee killings across 15 communities, of which 58 were observed, 41 inferred and 53 suspected. Males accounted for 92% of attackers and 73% of victims, about two-thirds of the killings were attacks between communities, and attackers typically outnumbered victims by a median of eight to one. The authors found that variation in killing rates was unrelated to measures of human impact and fit adaptive explanations in which killing gains resources or mates.

A separate 2016 Nature study of 1,024 mammal species found that lethal violence between members of a species is nearly absent in some groups, such as bats and whales, but is a notable feature of primates. Its phylogenetic prediction was that about 2% of human deaths would result from interpersonal violence, a figure the authors matched to prehistoric bands and tribes and found exceeded in most later periods.

The Lincoln release does not report human-specific figures within its 100-taxon comparison, and a syndicated retelling reached readers in October. For humans, the 2% estimate sits in a different study. Majolo, Wakes and Ruta add the narrower result that, across 100 primate taxa, mild conflict was only weakly related to the killing of adults and infants.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview