Sixty-five primate species out of a hundred kill infants of their own kind. Twenty-two kill adults from another group, nineteen kill adults from their own. The figures come from Bonaventura Majolo, Samantha Wakes and Marcello Ruta of the University of Lincoln, in England. Their study appeared in Evolution Letters on 3 March 2026. ScienceDaily picked it up on 6 October under a headline about humans, with no date for the paper.

The underlying question is old. Are daily squabbles and homicide rungs on the same ladder, one leading to the other? If they were, the most quarrelsome species should also kill most. The Lincoln team tested the idea on 98 species and two pairs of subspecies, from 43 genera and 11 families. Humans are in the sample. All the species live in groups, free-ranging and not fed by people.
The data come from literature published between 1950 and June 2022 and from a survey of 39 primatologists covering 28 species or subspecies. The authors separated five behaviours. Non-lethal fights count twice, between groups and within a group. So does adulticide, the killing of an adult of one’s own species. The fifth behaviour is infanticide.
Killings are rare and seldom witnessed, so they enter the analysis only as present or absent. Fights can be measured. On average 52 per cent of encounters between groups end in mild aggression. Within a group each animal is involved in 0.17 episodes an hour.

The comparison allows for kinship. Two related species resemble each other partly because they share an ancestor, and the Bayesian statistical model separates that inheritance from the rest. A second version of the model adds the size difference between males and females, their ratio, group size, the share of leaves in the diet and territorial defence.
The three kinds of killing travel together. Species that kill adults from other groups tend to kill within their own as well, with a correlation of 0.52 on a scale that tops out at 1. The link with infanticide is 0.44 for killings between groups and 0.48 for killings within them. Fights barely track killings. The correlations are weak and uncertain, and some are negative. Between fights with other groups and infanticide the value is minus 0.32, below the reliability threshold the authors set themselves. In their own words, “mild aggression is not closely linked to killing”. A species that fights often is not, for that reason, a species that kills.
Ancestry weighs more on killing than on fighting. In the model without ecological factors, the species family tree accounts for 42 per cent of the variation in infanticide. For fights between groups the share falls to 22 per cent, and for fights within groups to 9.

The picture shifts between the sexes. In males, killings between and within groups stay linked, at 0.47 in the simple model and 0.46 in the full one. In females the link stops at 0.24 and stays uncertain, rising to 0.33 only when the ecological factors come in. The authors argue that the different kinds of killing should not be lumped into a single violence score.
The limits are in the paper. Killings are coded yes or no because rates are not available. The authors think adulticide is under-recorded. Infanticide turns up about three times as often as adulticide. Part of the gap may reflect researchers’ attention, since infanticide has long been studied systematically. Another reason may be risk, which is lower for an attacker who goes after an infant than for one who takes on an adult.
The aggregator’s headline promises news about human nature. The result concerns primates as a whole, and humans appear as one species in a hundred. Among primates, fighting and killing follow separate evolutionary paths, and the step from one to the other follows rules more complex than current ecological models predict. The study had no specific funding and the authors declare no conflicts of interest. The data are deposited in the University of Lincoln repository.

References
Majolo B., Wakes S.J., Ruta M., "Origins of violence: evolutionary decoupling between mild and lethal conspecific aggression in primates", Evolution Letters, 10, 2, 207–216, published online 3 March 2026, open access (CC BY). doi: 10.1093/evlett/qrag002
Study data: University of Lincoln repository, hdl: 10779/lincoln.31026160
University of Lincoln, press release via ScienceDaily, "Humans May Not Be as Hardwired for Violence as We Thought", 6 October 2026 (used only as a lead).

*Board Member, SRSN (Roman Society of Natural Science)
Past Editor-in-Chief Italian Journal of Dermosurgery