Common chimpanzees live in communities of up to 150 individuals who occupy a distinct territory. As chimpanzees forage for the fruit and nuts that are unevenly distributed through the forest, they frequently split and coalesce into smaller groups ranging in size from one to fifteen. If one group encounters another group from a different community at the border between their territories, the interaction is always hostile. When the groups are evenly matched, they dispute the boundary in a noisy battle. The two sides bark, hoot, shake branches, throw objects, and charge at each other for half an hour or more, until one side, usually the smaller one, skulks away.

How far back can we trace the history of violence? Though the primate ancestors of the human lineage have long been extinct, they left us with at least one kind of evidence about what they might have been like: their other descendants, chimpanzees. We did not, of course, evolve from chimps, and as we shall see it's an open question whether chimpanzees preserved the traits of our common ancestor or veered off in a uniquely chimp direction. But either way, chimpanzee aggression holds a lesson for us, because it shows how violence can evolve in a primate species with certain traits we share. And it tests the evolutionary prediction that violent tendencies are not hydraulic but strategic, deployed only in circumstances in which the potential gains are high and the risks are low.

These battles are examples of the aggressive displays that are common among animals. Once thought to be rituals that settle disputes without bloodshed for the good of the species, they are now understood as displays of strength and resolve that allow the weaker side to concede when the outcome of a fight is a foregone conclusion and going through with it would only risk injury to both. When two animals are evenly matched, the show of force may escalate to serious fighting, and one or both can get injured or killed. Battles between groups of chimpanzees, however, do not escalate into serious fighting, and anthropologists once believed that the species was essentially peaceful.

Jane Goodall, the primatologist who first observed chimpanzees in the wild for extended periods of time, eventually made a shocking discovery. When a group of male chimpanzees encounters a smaller group or a solitary individual from another community, they don't hoot and bristle, but take advantage of their numbers. If the stranger is a sexually receptive adolescent female, they may groom her and try to mate. If she is carrying an infant, they will often attack her and kill and eat the baby. And if they encounter a solitary male, or isolate one from a small group, they will go after him with murderous savagery. Two attackers will hold down the victim, and the others will beat him, bite off his toes and genitals, tear flesh from his body, twist his limbs, drink his blood, or rip out his trachea. In one community, the chimpanzees picked off every male in a neighboring one, an event that if it occurred among humans we would call genocide. Many of the attacks aren't triggered by chance encounters but are the outcome of border patrols in which a group of males quietly seek out and target any solitary male they spot. Killings can also occur within a community. A gang of males may kill a rival, and a strong female, aided by a male or another female, may kill a weaker one's offspring.

Does chimpicide have a Darwinian rationale? The primatologist Richard Wrangham, a former student of Goodall's, has tested various hypotheses with the extensive data that have been amassed on the demography and ecology of chimpanzees. He was able to document one large Darwinian advantage and one smaller one. When chimpanzees eliminate rival males and their offspring, they expand their territory, either by moving into it immediately or by winning subsequent battles with the help of their enhanced numerical advantage. This allows them to monopolize access to the territory's food for themselves, their offspring, and the females they mate with, which in turn results in a greater rate of births among the females. The community will also sometimes absorb the females of the vanquished community, bringing the males a second reproductive advantage. It's not that the chimps fight directly over food or females. All they care about is dominating their territory and eliminating rivals if they can do so at minimal risk to themselves. The evolutionary benefits happen indirectly and over the long run.

When Goodall first wrote about these killings, other scientists wondered whether they might be freak outbursts, symptoms of pathology, or artifacts of the primatologists' provisioning the chimps with food to make them easier to observe. Three decades later little doubt remains that lethal aggression is a part of chimpanzees' normal behavioral repertoire. Primatologists have observed or inferred the killings of almost fifty individuals in attacks between communities, and more than twenty-five in attacks within them. The reports have come from at least nine communities, including ones that have never been provisioned. In some communities, more than a third of the males die from violence.

What does this have to do with violence in humans? It raises the possibility that the human lineage has been engaged in lethal raiding since the time of its common root with chimpanzees around six million years ago. There is, however, an alternative possibility. The shared ancestor of humans and common chimpanzees (Pan troglodytes) bequeathed the world a third species, bonobos or pygmy chimps (Pan paniscus), which split from their common cousins around two million years ago. We are as closely related to bonobos as we are to common chimps, and bonobos never engage in lethal raiding. Indeed, the difference between bonobos and common chimpanzees is one of the best-known facts in popular primatology. Bonobos have become famous as the peaceable, matriarchal, concupiscent, herbivorous "hippie chimps." They are the namesake of a vegetarian restaurant in New York, the inspiration for the sexologist Dr. Suzy's "Bonobo Way of Peace Through Pleasure," and if the New York Times columnist Maureen Dowd had her way, a role model for men today.

As for the risks, the chimpanzees minimize them by picking unfair fights, those in which they outnumber their victim by at least three to one. The foraging pattern of chimpanzees often delivers an unlucky victim into their clutches because fruiting trees are distributed patchily in the forest. Hungry chimps may have to forage in small groups or on their own and may sometimes venture into no-chimp's-land in pursuit of their dinner.

The primatologist Frans de Waal points out that in theory the common ancestor of humans, common chimpanzees, and bonobos could have been similar to bonobos rather than to common chimps. If so, violence between coalitions of males would have shallower roots in human evolutionary history. Common chimpanzees and humans would have developed their lethal raiding independently, and human raiding may have developed historically in particular cultures rather than evolutionarily in the species. If so, humans would have no innate proclivities toward coalitional violence and would not need a Leviathan, or any other institution, to keep them away from it.

The idea that humans evolved from a peaceful, bonobolike ancestor has two problems. One is that it is easy to get carried away with the hippie-chimp story. Bonobos are an endangered species that lives in inaccessible forests in dangerous parts of the Congo, and much of what we know about them comes from observations of small groups of well-fed juveniles or young adults in captivity. Many primatologists suspect that systematic studies of older, hungrier, more populous, and freer groups of bonobos would paint a darker picture. Bonobos in the wild, it turns out, engage in hunting, confront each other belligerently, and injure one another in fights, perhaps sometimes fatally. So while bonobos are unquestionably less aggressive than common chimpanzees -- they never raid one another, and communities can mingle peacefully -- they are certainly not peaceful across the board.

The second and more important problem is that the common ancestor of the two chimpanzee species and humans is far more likely to have been like a common chimpanzee than like a bonobo. Bonobos are very strange primates, not just in their behavior but in their anatomy. Their small, childlike heads, lighter bodies, reduced sex differences, and other juvenile traits make them different not only from common chimpanzees but from the other great apes (gorillas and orangutans) and different as well from fossil australopithecines, who were ancestral to humans. Their distinctive anatomy, when placed on the great ape family tree, suggests that bonobos were pulled away from the generic ape plan by neoteny, a process that retunes an animal's growth program to preserve certain juvenile features in adulthood (in the case of bonobos, features of the cranium and brain). Neoteny often occurs in species that have undergone domestication, as when dogs diverged from wolves, and it is a pathway by which selection can make animals less aggressive. Wrangham argues that the primary mover in bonobo evolution was selection for reduced aggression in males, perhaps because bonobos forage in large groups without vulnerable loners, so there are no opportunities for coalitional aggression to pay off. These considerations suggest that bonobos are the odd-ape-out, and we are descended from an animal that was closer to common chimpanzees.

It would be nice if the gap between the common ancestor and modern humans could be filled in by the fossil record. But chimpanzees' ancestors have left no fossils, and hominid fossils and artifacts are too scarce to provide direct evidence of aggression, such as preserved weapons or wounds. Some paleoanthropologists test for signs of a violent temperament in fossil species by measuring the size of the canine teeth in males (since daggerlike canines are found in aggressive species) and by looking for differences in the size of the males and the females (since males tend to be larger in polygynous species, the better to fight with other males). Unfortunately the small jaws of hominids, unlike the muzzles of other primates, don't open wide enough for large canines to be practical, regardless of how aggressive or peaceful these creatures were. And unless a species was considerate enough to have left behind a large number of complete skeletons, it's hard to sex them reliably and compare the size of the males and the females. (For these reasons many anthropologists are skeptical of the recent claim that Ardipithecus ramidus, a 4.4-million-year-old species that is probably ancestral to Homo, was unisex and small-canined and hence monogamous and peaceable.) The more recent and abundant Homo fossils show that the males have been larger than the females for at least two million years, by at least as great a ratio as in modern humans. This reinforces the suspicion that violent competition among men has a long history in our evolutionary lineage.

Even if common chimps and humans discovered coalitional violence independently, the coincidence would be informative. It would suggest that lethal raiding can be evolutionarily advantageous in an intelligent species that fissions into groups of various sizes, and in which related males form coalitions and can assess each other's relative strength. When we look at violence in humans later in the chapter, we will see that some of the parallels are a bit close for comfort.