Preface. This is a fantastic, must read book if you’re at all interested in how we evolved to be who we are today, how we domesticated ourselves, gossip, conformity, violence and more.  It reminds me of why I don’t read fiction — you can read all the best selling fiction in the world and not have a clue about how human behavior and psychology actually work. Below are excerpts from the book.

Alice Friedemann  www.energyskeptic.com  Author of Life After Fossil Fuels: A Reality Check on Alternative Energy; When Trucks Stop Running: Energy and the Future of Transportation”, Barriers to Making Algal Biofuels, & “Crunch! Whole Grain Artisan Chips and Crackers”.  Women in ecology  Podcasts: WGBH, Planet: Critical, Crazy Town, Collapse Chronicles, Derrick Jensen, Practical Prepping, Kunstler 253 &278, Peak Prosperity,  Index of best energyskeptic posts

***

Wrangham R (2019) The Goodness Paradox: The Strange Relationship Between Virtue and Violence in Human Evolution

The key fact about humans is that within our social communities we have a low propensity to fight: compared to most wild mammals we are very tolerant.

Before governments intervened, even among people with continuous war, there was a huge distinction between “peace at home” and “war abroad.  Despite a high rate of lethal violence in interactions among villages, within villages—family lives were “very tranquil” and episodes of aggression were largely regulated into formal duels.

Because seriously evil men can have a gentle side, we hesitate to empathize with their kindness for fear of seeming to rationalize or excuse their crimes. Such men remind us, however, of a curious fact about our species. We are not merely the most intelligent of animals. We also have a rare and perplexing combination of moral tendencies. We can be the nastiest of species and also the nicest.

A great oddity about humanity is our moral range, from unspeakable viciousness to heartbreaking generosity. From a biological perspective, such diversity presents an unsolved problem. If we evolved to be good, why are we also so vile? Or if we evolved to be wicked, how come we can also be so benign?

Our biology determines the contradictory aspects of our personalities, and society modifies both tendencies. Our goodness can be intensified or corrupted, just as our selfishness can be exaggerated or reduced. Once we acknowledge that we are at once innately good and innately bad, the sterile old argument gives way to fascinating new problems. If Rousseauians and Hobbesians are both partly right, then what is the source of our strange combination of behavioral tendencies?

On the one hand, human variation is limited. We really do have characteristic forms of society. Nowhere do people live in troops, as baboons do, or in isolated harems, like gorillas, or in entirely promiscuous communities, like chimpanzees or bonobos. Human societies consist of families within groups that are part of larger communities, an arrangement that is characteristic of our species and distinctive from other species.

Hunter Gatherers

The Ituri people in the Democratic Republic of Congo were, at heart, hauntingly like the villagers in my native England—loving their children, quarreling over their lovers, worrying about gossip, looking for allies, jockeying for power, trading information, fearing strangers, planning parties, embracing ritual, ranting at fate—and very, very rarely getting into fights.  There was a police force, for instance. The police were mostly male kinsmen of the chef de la localité. They used their status less to uphold the law than to exploit the villagers. On the rare occasions when they toured the neighborhood, a few policemen would arrive in a village after a walk of a few hours. They never brought food. So they would find a trivial excuse for fining an unfortunate host a chicken, eat it that night, and then stay on for as many days as they could continue extracting meals. The mundane corruption was of course resented, so the police were not much respected.

New Guinea is one of the few places where small-scale societies have been documented while living in true political anarchy, free from even the remotest interference of a state.  Its cultures are particularly informative because they show how people behave when living with a constant threat of being attacked by neighboring groups.  The Dani had one of the highest killing rates ever recorded. Sometimes small groups of men set off on raids to ambush an unsuspecting victim. Occasionally, there were battles: in the no-man’s-lands between villages, minor skirmishes could dissolve into larger chaos, with up to 125 villagers killed at a time. In a macabre index of the bloodletting, fallen warriors were commemorated by the removal of a finger from girls as young as three years old; among the Dani there were hardly any women with intact hands.

If the rest of the world had been like the Dani, the 20^th century’s sickening 100 million war deaths would have ballooned to an unthinkable 2 billion.

Beyond the intermittent mayhem, in the calm of ordinary life, Shangri-La really was a fair name for the Grand Valley. The Dani raised their pigs and tubers with a farmer’s typical steadiness. Heider wrote of the people’s low-key temperament, gentle demeanor, and rarity of anger. These were pacific, caring individuals who lived in systems of mutual dependence and support. When Dani households were not lulled by easy conversation, he said, they rang with song and laughter. Restraint and respect marked their daily interactions. As long as they had no war, the Dani were in many ways ordinary rural folks leading calm, thoroughly unaggressive lives.

The Dani proved typical of the remote New Guinea highlands in combining peace within the group with homicide of outsiders. Another New Guinea group, the Baktaman, occupied the headwaters of the Fly River. Every Baktaman community resisted trespass, often with violence. Territorial conflicts were so severe that they caused a third of the community’s deaths. Yet, within the villages, violence was severely controlled and “killing denied to be conceivable.” It was the same in the basin of the Tagari River, in west central Papua New Guinea, where the Huli terrorized their enemies but had no violence within their own villages.

Anthropologists documented intergroup fighting among the Aché hunter-gatherers of Paraguay shortly after an Aché group settled into a mission station. The Aché reported having previously used their hunting weapons of bows and arrows to shoot at strangers on sight. The result was a significant death rate.

In 1929, the anthropologist Maurice Davie summarized a consensus understanding that remains true today: people were as good to members of their own society as they were harsh to others. There are two codes of morals, two sets of mores, one for comrades inside and another for strangers outside, and both arise from the same interests. Against outsiders it is meritorious to kill, plunder, practice blood revenge, and steal women and slaves, but inside the group none of these things can be allowed because they would produce discord and weakness. The Sioux must kill a man before he can be a brave, and the Dyak before he can marry. Yet, as Tylor has said, “these Sioux among themselves hold manslaughter to be a crime unless in blood revenge; and the Dyaks punish murder. Not only is slaying an enemy in open war looked on as righteous but ancient law goes on the doctrine that slaying one’s own tribesmen and slaying a foreigner are crimes of quite different order.

Still, what about domestic violence? Even within a famously pacific group such as Botswana’s !Kung San hunter-gatherers, domestic violence has been recorded often. Furthermore, this form of aggression may have been systematically underreported. Early voyagers and anthropologists tended to be men from patriarchal societies. Wife beating tends to happen in private and can therefore escape the attention of anthropologists.

In the Aché, a settled group of people in Paraguay who had recently been hunter-gatherers, some individuals had a reputation for being generous, regardless of how much they were normally able to give. When those who were renowned for their generosity ran into difficulties, they were helped more than those with a reputation for stinginess. They were given more food, for instance. As predicted, reputations mattered.

Cooperation among the weaker individuals is needed. Among hunter-gatherers, women are not known to support one another in physical fights against violent men, so that can’t be how this trait evolved, as it did among female bonobos. In the Pleistocene, males were stronger and more robust than they are today, so it would have been even riskier for females to confront them in a fight.

Among hunter-gatherers, men are important as providers of food and as protectors, and women compete with one another to become the wives of the best husbands. There are no such dividing forces among female bonobos.

Among hunter-gatherers, as we will see, aggressors are stopped not by repeated coalitionary chases, and not by females acting on their own. When teasing and pleading and ostracizing and moving camp all fail to change a man’s violent behavior, the last resort of the coalition is execution, as Darwin foresaw.

Much like the combination of legal executions and mob lynchings in America, among hunter-gatherers communal support for killing was achieved in various ways. Sometimes executions were preapproved. Everyone might then want to be involved. In the Ju/’hoansi (or !Kung San), the famously peaceful hunter-gatherers of Botswana’s Kalahari desert, the community came together to solve the problem of a member of their group who had killed three men. The murderer’s name was /Twi. In a rare move of unanimity, the community…ambushed and fatally wounded /Twi in full daylight. As he lay dying, all the men fired at him with poisoned arrows until, after he was dead, all the women as well as all the men approached his body and stabbed him with spears, symbolically sharing the responsibility for his death….It is as if for one brief moment, this egalitarian society constituted itself a state and took upon itself the powers of life and death.

/Twi’s death was like Julius Caesar’s, whose body was said to be stabbed 35 times by the 20 or more senators who had conspired against him. They all wanted to show they were part of the coalition.

As in many societies, women were required not to intrude on the men’s secret goings-on on pain of death. Some years ago the Liagomir clan was holding a totemic ceremony, using its carpet snake totemic emblems (painted wooden trumpets). Two women stole up to the ceremonial ground and watched the men blowing the trumpet, went back to the women’s camp and told them what they had seen. When the men came back to camp and heard of their behavior, Yanindja, the leader, said, “When will we kill them?” Everyone replied, “Immediately.” The two women were instantly put to death by members of their own clan with the help of men from another group.

Rules were so important that an individual could kill for the sake of them regardless of his own personal preference. In Western Australia at the beginning of the 20^th century, men were required to behave themselves in their sexual liaisons. They were killed for having sex with a betrothed girl, or for taking a woman away when she was supposedly secluded during her menstruation, or for having intercourse before he was initiated into manhood. When a young woman fell in love with a youth during the segregation period of his initiation, the adoring woman followed him to express her feelings. He knew the penalty. Rather than face being killed himself, the young man killed her.

A romantic view holds that life in a small-scale society is delightful. In many ways, this is true. Unlike centralized tyrannies, the leaderless bands of nomadic hunter-gatherers or villages of slash-and-burn farmers are genuinely pluralistic societies. Disputes are solved communally. Anyone’s voice can be heard. No one is allowed to be sad. The sense of social support is enormous, and, as we have seen, the tenor of daily life is very peaceful. To add to the pleasures, the bands or camps are not subjects of any polity. They are members of a larger network of groups in which everyone speaks the same dialect or language and shares the same culture. Neighboring groups may sometimes have disputes and can even turn to violence against one another, but they all operate at the same political level, without hierarchy among groups, or group subjugation.

But liberty has its limits. In the absence of domineering leaders, a social cage of tradition demands claustrophobic adherence to group norms. Gellner called it a “tyranny of the cousins.” The cultural rules are paramount. Individuals have limited personal freedom; they live or die by their willingness to conform. Gellner’s “cousins” did not have to be literally relatives. “Cousins,” in the context of a small-scale society, is a metaphor for the group of adults whose decisions held sway. Their power was absolute. If you did not conform to their dictates, you were in danger.

The selk’nam on the island of Tierra del Fuego would kill any men who revealed secrets to women or uninitiated males. A tattletale could be killed by his father or brother. The male coalition was more important than kinship.

Gossip among men could lead to someone’s being pushed out of the group. The Gebusi people live in a remote lowland rain forest of New Guinea. They are hunter-gardeners or slash-and-burn farmers who had first been briefly contacted in 1940, then brought under government control in the 1960s. Their society was small, about 450 people. Like mobile hunter-gatherers, they lived in small villages, averaging fewer than 30 people, with no leaders holding authority within the group. Social relations within the society were in the classically gentle mold: serene conversations, rich humor, no boasting. Their political system was egalitarian, but, as always, not everyone was equal. Among the Gebusi, sorcerers, like witches, could be accused of bringing misfortune to others.

If blamed, the sorcerer was in peril. If he gets angry and denies any responsibility for the illness, he may be seen as unrepentant. His best chance is to admit to a minor crime—in other words, confess to being a sorcerer, and agree to stop making the patient sick. That is what he does. Straining to maintain composure, he pleads for his life: “I don’t know either. I don’t know anything about it. He’s my own relative, too, I couldn’t make him sick. I don’t know….I was so sorry for him when I heard he was sick. I only heard a few days ago. We had been out in the bush and I didn’t even know. When I finally heard, I said to my wife right then, ‘We’ll have to go and make sure he’s all right. I wonder, what sickness could he have?’…I don’t know about it, but I’m sure that now the leaves are pulled…if you throw them away he’ll get better. I might have been a little angry from not eating enough fish lately, but I certainly wouldn’t make my own relative sick like that.

The accused man would likely leave the longhouse, right to be nervous about what might happen. The fate of the accused would depend on divinations following a death. Unless the outcome was clearly helpful to his case, he would almost certainly be executed within a few months. The accuser would quietly ensure the support of the community. He and his trusted friends might convene a séance at a time when they knew the guilty party had few male kin at home. During a sleepless night of joking and shouting in the longhouse, the men would grow more and more enthusiastic about the idea that the alleged sorcerer was responsible for the death. Consensus is reached. All decide the accused is guilty.

At dawn, they stage an ambush. They kill with clubs or arrows. Sometimes there is torture first. Then they butcher and cook him. In precontact times, cannibalism.

Hunter-gatherers offer a chilling example. If there was a conflict of interest between men and women, moral rules would typically favor the men at the expense of the women. Male hunter-gatherers throughout Australia used their women as political pawns. Wives could be required to have sex with multiple men at special ceremonies. They could also be lent to a visiting man, or given sexually by a husband to a man with whom the husband had quarreled, in order to erase a debt or make peace. Women could be sent on sexual missions into dangerous situations. When potential attackers were seen approaching a group, one response was to send women out to greet them. If the strange men were willing to forgo their attack, they signaled their intent by having sexual intercourse with the female emissaries. If not, they sent the women back and then attacked. The final stage of peacemaking between two tribes almost always involved an exchange of wives. Apparently, women did not enjoy these coercive encounters. In 1938, the anthropologist A. P. Elkin reported that Australian Aboriginal women would live in terror of the use that was made of them at ceremonial times. In those cultures, all this was the moral practice of ordinary men. The men behaved prosocially toward one another while exploiting their wives and female kin. To say that these behaviors were good for the group implies a very restricted definition of “the group.” The behaviors were designed to be good for the group of married men who made the rules, but not for the women.

Coercive practices and unequal expectations of self-sacrifice cast doubt on the idea that moral practices are necessarily good for the group.

In the initial phase of generating a more peaceful species, the moral emotions would be little affected. The target of the newfound coalitions was only the hyperaggressive males. The next stage would be critical for the evolution of moral sensibilities. In developing the ability to kill the physically imposing alpha, the subordinate males had discovered an irresistible coalitionary power. They could now coordinate to kill anyone. So all kinds of troublemakers became at risk. Any kind of noncompliance with the interests of the killing coalition could in theory provoke an intimidating threat. Women and young men were as vulnerable as domineering bullies to the power of the male elders.

In nomadic hunter-gatherers, as in small-scale societies in general, aspiring alphas are not the only victims of the tyranny of cousins. Young men can be executed for messing with the old men’s wives. Women can be executed for breaking a seemingly trivial cultural norm, such as seeing the magic trumpets, or treading on the men’s secret path, or for having sex with the wrong men. Anyone who flouts the rules that the men impose can be executed.

The result is a society in which the male coalition not only holds the power but uses it. The anthropologist Adamson Hoebel documented the legal systems of small-scale societies. He found that belief systems were typically founded on a religious statement such as “Man is subordinate to supernatural forces and spirit beings, which are benevolent in nature.” That kind of idea legitimizes the belief system by referring it to forces beyond human control. A series of postulates would follow. For the Inuit, women were socially inferior to men but essential in economic production and childrearing. In no society were men socially inferior to women.

The anthropologist Les Hiatt summarized the effects in Australian Aboriginal society. Women often had strong traditions of independence and cultural autonomy. In some places, they had secret societies. They could have the largest role in choosing whom their daughters married. But, even though women were not subservient, the genders were not equal. Sanctions against women for discovering male secrets included rape and death. By contrast, there were no physical reprisals for men who intruded into a women’s ceremony. Men could arrange for gatherings with neighboring societies; there was no equivalent for women. Men could insist that women provide food for the all-male secret ceremonies, or provide sexual services to whomever the men required. Religious knowledge, controlled by men, justified their dominance.

Among the Inuit: Threats and abuse may lead to the same end. The obnoxious person is first ostracized, then liquidated if he continues his bothersome behavior. The execution of liars was reported across the span of Inuit territory, from Greenland to Alaska. Everywhere the same was true. A coalition of men controlled life and death according to rules that they set. 

Once men dominate the society through their control of death, their word becomes law. Everyone knows the importance of conforming. People accept the inequities. Men get the best food, have the most freedom, and are the ultimate arbiters of group decisions.

The punishing of deviants occurs because people feel individually threatened or dispossessed by social predators, but also, in a larger sense, because socially disruptive wrongdoers so obviously lessen a group’s ability to flourish through cooperation.”

The suppression of theft, fighting, and antisocial play would tend to be good for everyone. It is also plausible that more selfish motivations applied. Men could have enforced patriarchal norms that allowed them to trade women, use them as sexual and political pawns, and beat them. So, although the coalition promoted prosociality by punishing deviants, it did not necessarily advance the welfare of the whole group.

The first rule of life in a dense web of gossip is: Be careful what you do. The second rule is: What you do matters less than what people think you did, so you’d better be able to frame your actions in a positive light. You’d better be a good ‘intuitive politician.’

Conscience is a mechanism of self-defense. Through natural selection humans became equipped with an increasingly sophisticated moral conscience for steering clear of moral mobs. These cognitive mechanisms would prospectively compare the individual’s potential actions against the set of moral wrongs in order to avoid actions that could trigger coordinated condemnation by third parties. Conscience protected our ancestors from the kind of behavior that could lead them to be accused of being deviants. Once again, self-defense explains our moral motivation.

The third puzzle is why humans are so sensitive to right and wrong that we not only try to do the right thing ourselves, but also monitor one another’s behavior and punish those who we judge to do wrong. The answer seems clear. Individuals need to protect themselves from being regarded as nonconformists.  Such sensitivity to moral values also became biologically embedded into novel emotional responses. Prominent human emotions not known to occur in animals include shame, embarrassment, guilt, and the pain of being ostracized, all of which are human universals.

Because shame signals acknowledgment of having violated a social norm, it offers the restorative power of an ingratiating apology. Shame thus seems designed to protect from the ostracism that can lead to social or physical death.

The same kind of argument applies to embarrassment, a social blunder. Emotionally it is felt as shame.

Within less than a second of having been offensive (which is normally unintentional), the embarrassed individual starts a signal that lasts up to three seconds. He looks downward, turns his head (mostly to the left), smiles, controls the smile such as by sucking his lips, gives furtive glances, and often touches his face. Meanwhile, a blush emerges over a longer period,

Embarrassment serves to restore social relationships that have gone awry. Individuals who do not show embarrassment after a social faux pas are more likely to be viewed negatively, whereas those who readily blush redeem their standing. The execution hypothesis offers an explanation of why social standing matters so much.

Guilt is another painful emotion that serves to mend social relationships. Defined as “a painful affect arising from the belief that one has hurt another,” guilt involves admission of a wrong. The acceptance of blame is supposed to inhibit self-assertive aggression toward others, turning it inward on oneself. The associated expressions of remorse again pave the way to being forgiven.

In a mere two or three minutes of play with strangers, followed by being excluded, people predictably became sad, angry, and had negative feelings of alienation, depression, helplessness, and even a reduced sense of meaning in life. The effects did not depend on the subject’s personality or whether he or she felt similar to the ostracizers. Experimental subjects experienced elevated activation of a part of the brain, the dorsal anterior cingulate cortex, that is also activated by physical pain. To be ostracized engages a swift and strong series of neurally encoded responses that are very unpleasant.

We have to use our thinking ability, because what a culture determines to be “right” and “wrong” can vary. Our ancestors had to learn what kind of behaviors the culture considered appropriate, and selection would have favored mentally internalizing the norm, recognizing norm violators and responding, such as ostracizing a violator.  Our moral psychology was forged during a period when being a social outcast was even more dangerous than it is for most people today.

***

From an evolutionary perspective, the human rate of physical aggression within social communities is already strikingly low. Because chimpanzees are one of humanity’s two closest relatives, they provide an illuminating comparison. Chimpanzees are not like people. A day spent with wild chimpanzees gives you a good chance of seeing chases, sometimes some hitting, while hearing fearful screams. Every month, you are likely to see bloody wounds. Primatologists quantified the difference between an ordinary group of chimpanzees and a particularly disturbed population of Australian Aboriginals who had recently stopped hunting and gathering. Among the Australians, social disintegration and alcohol were considered responsible for raising the likelihood of physical aggression to especially toxic levels. However, even in this comparison with an unusually violent group of humans, the chimpanzees were several hundred to a thousand times more aggressive. The difference in the frequency of fighting between humans and chimpanzees is enormous.

Bonobos are the other species most closely related to humans, a similar-looking ape with a well-deserved reputation for being much more peaceful than chimpanzees. They are not unaggressive, however. A recent long-term field study found that wild male bonobos were aggressive at about half the rate of chimpanzees, while female bonobos were more frequently aggressive than female chimpanzees. So, although male bonobos are less violent than male chimpanzees, the rates of aggression in both these species of great ape are far higher than the rates among humans.

Overall, physical aggression in humans happens at less than 1% of the frequency among either of our closest ape relatives. Compared to them we really are a dramatically peaceful species.

In 2005, the World Health Organization’s Multi-Country Study on Women’s Health and Domestic Violence produced detailed data from 24,000 women in ten countries. Physical violence by partners included slapping, shoving, punching, kicking, dragging, beating, choking, burning, and using or threatening the use of a weapon. In cities, the proportion of women who said that they had experienced physical violence by their partners averaged 31%, from 13% in Japan to 49% in Peru. In rural areas, the rates were higher, averaging 41%. Between 50 and 80% of the intimate-partner violence was considered “severe.” These rates appear to be slightly above those in the United States, where, in more than 9,000 interviews, 24% of women reported severe physical violence inflicted by an intimate partner.

A 2013 WHO study found that the proportion of women who had experienced either physical or sexual violence averaged 41% in cities and 51% in rural areas of their ten focal countries. The equivalent figure in the United States was 36%. Some 41 to 71% of women have been beaten by a man at some time during their lives. Yet this range is low compared with the incidence among our closest animal relatives. One hundred percent of wild adult female chimpanzees experience regular serious beatings from males. Even among bonobos, whose females are routinely higher-ranking than males, males attack females rather often. In subgroups averaging nine individuals, the primatologist Martin Surbeck saw a male bonobo physically attacking females every six days on average.

Overall tendencies are clear: compared with other primates, we practice exceptionally low levels of violence in our day-to-day lives, yet we achieve exceptionally high rates of death from violence in our wars. That discrepancy is the goodness paradox.

Delgado’s work was part of a wave of enthusiasm for the idea that violent tendencies might be controlled through biological science. A neurobiologist working with animal aggression, he thought that experiments like those that he conducted with the fighting bull might have wider implications. He fantasized about being able to “psychocivilize” people by “the use of implantable brain electrodes that could be modulated via remote control.

How people express their aggression also varies. Some are confrontational, some are passive-aggressive, some are gossips. Petty insults can flash to lethal aggression too quickly for the fighters to have second thoughts.

When the criminologist Marvin Wolfgang conducted the first major study of the reasons for murder in the United States, in 1958, he found that, during a four-year span in Philadelphia, character contests were responsible for 35 percent of the city’s homicides, the largest category of any type of murder. Similar frequencies have been found elsewhere.

Reactive aggression is variously described as hostile, angry, impulsive, affective, or “hot.” It always comes with anger, and often with a loss of control, such as losing one’s temper. It is a response to provocations such as a perceived insult, embarrassment, physical danger, or mere frustration. In the state of intense arousal that is typical of reactive aggression, a fighter easily lashes out at anyone around him or her. Reactive aggressors have no goal beyond getting rid of the provoking stimulus,

Reactive aggression is especially prominent among males fighting over status or mating rights.

During a study of rut fighting among male pronghorn antelope, 12% of conflicts over mating rights to estrous females led to the death of one or both males. In various populations of red deer, 13 to 29% of male deaths came from rut fighting.

Proactive aggression is characterized as premeditated, predatory, instrumental, or “cold.” Unlike reactive aggression, it involves a purposeful attack with an external or internal reward as a goal, rather than an effort to remove a source of fear or threat. It is the calculated act of the professional assassin.

The behavior is self-rewarding rather than serving to remove an aversive stimulus: the killer is pleased at achieving a goal. Proactive aggression can be triggered by a wide variety of motivating factors, including desire for money, power, control, or sadistic fantasy,

Individuals who have a greater tendency for proactive aggression than others have characteristic social emotions. They tend to have reduced emotional sensitivity, feel less empathy for their victims, and experience less remorse over their actions.

Whereas Marvin Wolfgang had found that 35% of Philadelphia murders resulted from character contests, he and criminologist Franco Ferracuti concluded that “probably less than 5% of all known killings are premeditated, planned and intentional.” Those percentages sum to only 40%, leaving it unclear how many of the remaining 60% of homicides were proactive or reactive.

Criminologist Fiona Brookman reported revenge as a motivation in 34% of British homicides in which both killers and victims were men, and since revenge always includes a component of planning, it can be considered proactive. A second reason for thinking that the 5 percent figure is too low is that because proactive murderers have had time to plan events carefully, they presumably get away with their crimes relatively often.

The U.S. Federal Bureau of Investigation suggest that at least 35% of murderers in the U.S. are never brought to justice. For these reasons proactive killing probably occurs at a higher frequency than 5% of murders.

Evolutionary psychologists Margo Wilson and Martin Daly argued that most killings that result from trivial altercations reflect a drive to maintain status, a drive that would have been adaptive in a world with no alcohol and less effective weapons, but that today is no longer adaptive because it leads to the aggressor’s becoming homicidal. Criminologists Kenneth Polk and Fiona Brookman argue that fights over status are especially frequent among lower-working-class and underclass men because their material resources are scarce, making honor all the more important.

The majority of spousal batterers are readily classified as either a predatory (proactive) or an impulsive (reactive) type. Predatory batterers are more violent in general, more concerned to dominate and control their partners, and more likely to be violent, particularly when the spouses talk back. In contrast, it is when spouses try to withdraw from an argument that impulsive batterers are most likely to lose control.

Compared with nonmurderers, all of the accused had high neural activity in subcortical parts of the brain, including the limbic system, which is a brain network that processes emotional responses. This finding suggests that all of the accused murderers tend to experience particularly strong emotions. Just as Raine had anticipated, however, the brains of the accused murderers differed, depending on whether their murderous acts were characterized as reactive or proactive. The reactive murderers had less activity in their prefrontal cortex, the inhibitory part of the brain. The difference contributes to an explanation of why some people are more vulnerable to committing a crime of impulsive violence: they find it hard to control themselves.

Psychopathy is found worldwide. According to a standard rating scale devised by criminal psychologist Robert Hare, psychopaths tend to exhibit 20 features, including superficial charm, frequent lying, sexual promiscuity, and a low threshold for boredom. They are insensitive both to what others think and to what others feel. This can work in their favor at least in the short term, because despite their arrogance, ambition, and readiness to be deceptive, their confidence can make them attractive. They show less evidence of empathy than ordinary people, and they tend to feel less guilt or remorse. This lack of concern makes them relatively more liable to be aggressive. Finally, psychopaths are prone to try to get what they want, regardless of the means necessary. In sum, psychopaths are self-centered and uncaring people with impaired moral judgments. Not surprisingly, they are likely to be criminally delinquent, and they are also predominantly male.

A survey in the U.K. found psychopathy in less than 1% of the household population, which is probably the approximate proportion worldwide. Psychopathy is more common among men and younger adults than among women and the middle-aged or elderly. Psychopaths show more violent behavior than others. In the U.K., psychopathy was also associated with suicide attempts, imprisonment, drug dependence, antisocial personality disorder, and homelessness. Of the many features contributing to psychopathy, a lack of conscience is found to be particularly important.

The limbic system is a subcortical series of small structures deep in the brain that are connected with one another and are heavily involved in the production of emotional responses such as anger, anxiety, fear, and pleasure. In keeping with their stronger emotional reactions, wild mammals tend to have a larger limbic system than domesticated mammals. A well-studied part of the limbic system is the amygdala, a pair of almond-sized regions. An amygdala bigger than normal is associated with more fearful and aggressive reactions across individuals, and larger amygdalae are more typical of a wild than a domestic animal.

Psychopaths appear particularly fearless, a trait that seems to be supported by their tendency to have an amygdala that brain imaging reveals as smaller, sometimes deformed, and less active than in other people.

A cat cannot fight and hunt simultaneously—a useful adaptation to avoid the confusion of trying to do two incompatible things (fighting and feeding) at the same time. In rats, by contrast, connections between the mediobasal hypothalamus and lateral hypothalamus are minimal, so there is less inhibition of one type of aggression by expression of the other type. This means that if a rat initiates a premeditated, proactive “quiet biting attack” but finds the victim fighting back, the attacker can instantly respond with reactive aggression. Thus in rats, the absence of reciprocal inhibition allows proactive and reactive aggression to occur at the same time without one inhibiting the other.

A 2015 survey of all available twin studies of aggressiveness, of which there had been 40 in the preceding five years alone, found that the genetic heritability of aggressive behavior was typically in the range of 39 to 60%, averaging 50%. This means that, in those environments, genetic and socialization influences were roughly equally important in shaping individuals’ aggression. Interestingly, the same does not apply to some closely similar behaviors, such as rule breaking. Differences among children in the tendency to break rules, as opposed to being aggressive to one another, have been found to come almost entirely from socialization.

Proactive aggression might therefore prove to be more strongly influenced by genes than reactive aggression, but for the moment all we can say is that both types show important genetic influences.

Given that humans, compared with other species, are low on reactive aggression and high on proactive aggression, the question is why we have this mixture.

In 2000, his friend Erich Klinghammer, director of Wolf Park, Indiana, invited him into the cage of a captive wolf. Coppinger hesitated. “I don’t know much about tamed wolves,” he said. Klinghammer reassured him. His wolves were progeny from several generations of captive breeding, far removed from the wild. They had been hand-raised by human “puppy parents” since they were ten days old. Even as adults, they were still handled every day. They were used to being on leashes, and were as tame as wolves could be. “Just treat them like dogs,” Klinghammer said. So Coppinger did. He joined Klinghammer and the wolves inside their pen. Saying something like “Good wolf” to an adult female called Cassi, he patted her on her side. In Coppinger’s words: That was when she became all teeth. Not a nip, but a full war—a test of my ability to stay on my feet and respond to Erich’s excited command, “Get out, get out! They’ll kill you!” Note the wording “They’ll kill you!” I had a blurred vision of a collection of wolves gathering, and a wolf tugging on my pants as Cassi focused on my left arm. “Why did you hit her?” Erich said later, almost too softly to be heard over my pounding heart. “It wasn’t hitting! I was patting her! You said treat them like dogs and I pat dogs and if I do some social misconduct with a dog, I don’t get my head bit off, and why is it that all you people who socialize wolves have those nasty scars!” I said in a single breath while applying a tourniquet to the mangled arm of my goose-down jacket. Never again did I think that tame wolves could be treated like dogs.

Wolves are different from dogs. However much you tame a wolf, it will not become domesticated. After years of behaving well, a wolf can suddenly and unpredictably forget its training. You should not trust wild animals, because they are all too reactively aggressive. Domesticated animals, in contrast, have changed genetically from their wild ancestors; they are less easily stimulated into producing reactive aggression.

Humans’ place in this tamed/domesticated dichotomy is clear. We are calm compared with a typical wild animal—more like a dog than a wolf. We can look each other in the eye. We do not lose our tempers easily. We normally control our aggressive urges. In primates one of the most potent stimulus for aggression is the presence of a strange individual. But child psychologist Jerome Kagan reports that in hundreds of observations of two-year-olds meeting unfamiliar children, he has never seen one strike out at the other. That willingness to interact peacefully with others, even strangers, is inborn. Like domesticated animals, humans have a high threshold for producing reactive aggression. In this respect humans resemble domesticated animals much more closely than we resemble wild animals.

From 1713 until his death in 1740, King Frederick William I of Prussia, an arrogant and domineering man who was also a drunkard, wanted to make his Potsdam Guards regiment the most impressive in the world. To do so, he paid a thousand recruiters to roam 15 European countries to capture the tallest men and bring them to Prussia. The king committed vast sums to the effort. The tall soldiers were intended to be the pride of the army, but because they were there against their will they had to be imprisoned. The difficulties of building up his collection of “giants” were so great that the king turned to artificial selection as an alternative method of arranging for an army of tall soldiers. He decided that if giants could not be easily recruited he had better breed them.

Accordingly, his men searched peasant villages for tall women to marry his tall guardsmen, to whom they would be assigned as mating partners. No consent was sought, no inquiries were made as to previous marriage relations.  The Prussian king’s experiment is said to have led to some unusually tall people in Potsdam, but overall it was a failure. It was deeply resented by both husbands and wives, and when the king died the experiment ended. Obviously, artificial selection of humans was an aberration if even a powerful monarch could not make it happen.

How we domesticated ourselves

It was Darwin’s and Aristotle’s theory that different populations were domesticated to different extents. Darwin wrote, in 1811: “Man is a domesticated animal…born and appointed by nature the most completely domesticated animal…the most perfect of all sorts of domestic animals that have been created.”

Human domestication came to be seen not only as a cause of racial differences but also as an index of human value: some races or populations were thought to be better than others, depending on how domesticated they were. The divisive potential of this idea became explosive in the hands of Nazis and their associates.

Lorenz considered that, under the influences of civilization, humans had become overly domesticated, leading to unattractive, infantilized, and unviable people.  He considered more highly domesticated populations to be a degraded version of the natural ideal.

Humans became less reactively aggressive and increasingly docile around the time of our becoming Homo sapiens.  The list of traits associated with domestication, called the domestication syndrome, is useful, because it provides telling clues to the human past.

Critically, the domestication syndrome includes changes to bones. Leach listed four characteristics of the bones of domesticated, but not wild, animals that are found in contemporary humans.

First, domesticates mostly have smaller bodies than their wild ancestors.  The effect is so predictable in dogs or herd animals like sheep and cattle that archaeologists use it as one of their main criteria to recognize when domestication of different species has occurred.  Smaller human bodies in the past are also indicated by changes in the relative thickness of bones. The limb bones of our ancestors are thicker, both at the ends and in the midshaft, compared with their length. Cross sections also show that limb bones formerly had thicker walls surrounding the marrow cavity. The thicker the bones, the more weight they carried. Based on bone thickness, humans have been gradually weighing less since the time of Homo erectus, about two million years ago, including an especially pronounced weight loss since the appearance of modern-looking Homo sapiens. Such changes are often summarized as humans’ becoming less robust and more gracile.

Second, the faces of domesticated animals tend to be shorter, projecting relatively less forward, than those of their wild ancestors. Teeth also become smaller, and jaws smaller still.  Decline of tooth size has been noticed over the last hundred thousand years. Teeth diminished in size at a rate of approximately 1% per 2,000 years until 10,000 thousand years ago, when their decline accelerated to a loss of 1% of volume every 1,000 years.

Third, differences between males and females are less highly developed in domesticated than in wild animals, always for the same reason: males become less exaggeratedly male.  During the last 35,000 years, according to anthropologist David Frayer, males have become more like females not only with respect to stature but also to the size of the face, the length of the canine teeth, the area of the chewing teeth, and the size of the jaws. Further back in time, around 200,000 years ago, male faces were already becoming relatively feminized. An analysis by biologist Robert Cieri and colleagues showed that brow ridges above the eyes became less projecting in males, and the male face became shorter from the top of the nose to the upper teeth.

Finally, domesticates have a strong tendency to have smaller brains than their wild ancestors, whether mammals or birds. The average reduction in brain volume for a given body weight is around 10 to 15%, but some level of brain reduction is found in every domesticated mammal except for laboratory mice. Although human brain size, as measured by the volume inside a skull has steadily increased over the last two million years, the trajectory took a surprising turn around 30,000 years ago, when brains started to become smaller. In Europe, modern brains are some 10 to 30% smaller than those of people living 20,000 years earlier.

In domesticated animals the loss of brain size is not associated with any consistent reduction in cognitive ability. Indeed, the smaller-brained species sometimes outperform their bigger-brained ancestors

The differences between modern humans and our earlier ancestors have a clear pattern. They look like the differences between a dog and a wolf. Our ancestors were less docile than we are today. They would have had a greater propensity for reactive aggression, losing their tempers more easily, quicker to threaten and fight one another. But somehow we became domesticated.

What is the connection between reduced aggressiveness and smaller chewing teeth? No relationship is obvious. Animals do not fight with their chewing teeth, so there seems no reason for the teeth to become reduced in size.

When foxes were bred to be friendly, by the fourth generation, the experimenters were amazed to find a few pups approached by humans wagging their tails as if they were dogs. Unselected foxes had never wagged their tails. The sixth generation marked the appearance of a “domesticated elite.” Elite foxes not only wagged their tails, they even whimpered to attract attention, and they approached the experimenters to sniff and lick them.  By 1969, 40% of females were breeding three times per year, compared to once in the unselected lines. The change was evidently genetic, because it was concentrated in the same families. The shift from one to two annual mating cycles was not always beneficial: litters often failed. But even though there were no immediate practical benefits in terms of increased fur production, selection for docility alone had led to a loss of breeding seasonality. It also led to foxes’ reaching sexual maturity a month earlier than unselected foxes, having longer mating seasons, and producing bigger litters.  After 15 to 20 generations foxes had shorter tails and legs, and underbites or overbites. These are all features of the domestication syndrome.   Skull shape changed. The domesticated foxes had narrower skulls, with less cranial height, than the farm foxes, and the skulls of males were more like females.

This research evinced some immediate effects of selecting for tameness, but it did not show how long the domestication syndrome would last. This is critical. If features of the domestication syndrome are sufficiently maladaptive, natural selection would be expected to reverse them. So, in theory, although the domestication syndrome might be thrown up rapidly by selection for tameness, it could be lost equally quickly. Yet even in the wild, it turns out, lineages of originally domesticated animals can thrive for many generations without reverting to their ancestral type.

Europeans were so impressed by the commercial prospects of mink fur they imported the American domesticates for breeding. Unfortunately, many escaped into the wild, where they thrived. By 1920, these feral American minks were spreading fast. Hundreds of thousands of the invasive species established themselves throughout continental and archipelagic Europe, including Norway, Italy, Spain, Britain, Ireland, Iceland, Russia, and Belarus. In Belarus, the success of the domesticated American mink caused a substantial decline in the populations of two native species of carnivores, the European mink and polecats. The American mink outcompeted their wild cousins.  This raises a new puzzle. If traits of the domestication syndrome work well in the wild, in what way were the wild ancestors better adapted than the descendants of domesticates to the wild environment? If a small brain and short face are good enough for American mink on the loose in Belarus, why did the mink ancestors evolve a larger brain and longer face in the first place? The answer is unknown.

Only about 80 generations had been needed to produce aspects of the domestication syndrome; after 50 generations in the wilds of Belarus, the mink had shown no reversal to wild anatomy. Small brains and short faces appeared to be just as well adapted to the woods and waterways of Europe as to Canadian cages.

Dingoes are descended by thousands of generations from domesticated dogs, but even after at least 5,000 years back in the wild, they have brains no larger than those of dogs. Dingoes’ brains have not reverted to being large and wolflike.

Domesticated animals flourishing in the wild are found mostly in habitats where their wild ancestors are absent, such as the American mink in Europe, pigs on the Galapagos Islands, horses in the American Southwest, or dingoes in Australia.

In general, wolves are more aggressive toward humans than dogs are. Wolves are also significantly more violent toward members of other packs than are dogs. Wolves are so innately aggressive to wolf “strangers” that the leading cause of death in the wild is being killed by other wolves, accounting for as much as 40% of adult mortality. Among packs of feral dogs, by contrast, killing a “stranger” from another pack has been described only once. Within packs, too, dogs appear to be more tolerant of one another than are wolves, as indicated by their sharing breeding opportunities more equally.

That domesticated animals are restrained in their aggressiveness toward humans does not necessarily mean they are equally restrained in their aggressiveness toward one another. In general, however, animals that are managed by humans are doubtless often selected for being tolerant toward one another, since fights in the farmyard are costly to the farmer.

The domestication syndrome is produced by selection against reactive aggression, not merely by living with humans. The implication is remarkable. It means that a domestication syndrome should be produced any time that reactive aggression is selected against. Since aggressiveness must sometimes be selected against in the wild, there should be many cases of domestication syndromes in the wild.

If selection against reactive aggression were to produce the domestication syndrome in an animal living in nature without contact with humans, it would be a real-world example of domestication happening without a domesticator. It would strengthen the likelihood that humans, too, could have become domesticated.  And there is such a species: the bonobos. The feature that points most directly to bonobos’ being self-domesticated is their relatively muted tendency for aggression. Bonobos are much less aggressive toward one another, and much less fearful of one another, than chimpanzees are. Zookeepers find bonobos easy to accommodate, because groups readily accept new individuals without any serious tension.

Introducing chimpanzees to one another, by contrast, tends to be a painfully slow process, weeks or months of gradually familiarizing strangers with one another through wire mesh in order to minimize the risk of violence. Even after such cautious preparation, when chimpanzees who have not spent time together finally meet, they might easily fight. They regularly charge at one another in displays intended to demand clear expressions of subordinacy. If the target does not give any signal of submissiveness, fighting tends to erupt. Usually, the aggressor wins, the reluctant subordinate screams passionately,

Males also commonly beat up on females, often in surprise attacks. The male’s aim in such attacks is to intimidate a chosen female into readily acceding to his future demands for sex. For each female, one male distinguishes himself from other males by being the one who most frequently attacks her. The tactic is often successful. Over subsequent weeks, a female’s most frequent aggressor tends to be her most frequent sex partner, and eventually, even though she is likely to mate several times with every male in her community, he will be the most likely father of her next baby. This stomach-churning practice is part of the reason why, as males become adult, they go through a ritual of beating up on every female. A male’s ability to intimidate females is a vital component of his strategy for having as many offspring as possible.

Bonobos could hardly be more different from chimpanzees in terms of competition and aggression.

More important than their own fighting ability is the support given by mothers to their adult sons. The top-ranked males are mostly those with high-ranked mothers; when their mothers die, they are liable to fall in rank. Far less male intimidation of females occurs among bonobos than among chimpanzees. In one study bonobo females were more often aggressive toward males than vice versa. Male bonobos do not beat up on females, and in competition over food, females are more likely to win than males. No one has recorded any violent infanticides, nor any killing of adults, either within or between bonobo groups. Bonobos are certainly not free from disputes, and when groups from different communities meet, they sometimes fight to the point of inflicting bite wounds and scratches. But overall, the intensity of aggressive bonobo behavior is immensely less than among chimpanzees.

Brian Hare, Victoria Wobber, and I tested the prediction that bonobos would show the domestication syndrome in 2012, and found the first evidence for the domestication syndrome in a wild species. The cranial anatomy of bonobos turns out to fit the domestication syndrome extraordinarily closely. To start with, bonobo brains are smaller than those of chimpanzees. The reduction is especially marked in males, where it can reach as much as 20%. This echoes the decrease in brain size of almost every species of domesticated vertebrate compared with their wild ancestors. All of the other major cranial features of the domestication syndrome are present as well. The faces of bonobos are relatively short, projecting less than faces of chimpanzees. Bonobos have smaller jaws, and smaller chewing teeth.  Many of the characteristics that so richly distinguish bonobos from chimpanzees did not evolve as adaptations. Instead, they evolved as incidental side effects of selection against reactive aggression.

If bonobos could be self-domesticated, the bonobo case supports the idea that humans could be self-domesticated too. But what the evidence of self-domestication does not do is explain why bonobo aggression was reduced. An animal that fights too often, or too intensely, wastes energy and takes unnecessary risks. The trick is to get the balance right, which is  why, over evolutionary time, males with gentler, less aggressive proclivities tended to have higher reproductive success.

Female power is clearly an important part of the answer. A male bonobo who confronts an adult female might well win if she is the only female in earshot. But female bonobos are rarely far from other females. The challenging male must expect that, if he makes a female scream, within seconds he may be confronted by a coalition of females ready to attack him, and so effective in doing so that his best response will be to run away. Female support for one another explains why males give up easily when competing with females over food, or why males rarely try to bully females, or why males do not, on average, outrank females. Coalitionary attacks need not be common. Although females in the wild can use coalitions well, they do so rarely, mostly when males threatened their young. Despite their size disadvantage, females very effectively suppress bullying by males. Males seem to have learned where the ultimate power lies: numbers beat physical strength.  The reason that female bonobos can predictably present a united front appears mundane: they stay close to each other.

Chimpanzee parties, by contrast, are numerically dominated by males. Females tend to travel alone or in smaller subgroups. It seems likely that because of this relatively dispersed way of life, female chimpanzees fail to gain confidence in each other’s support against males.

The ability of gorillas to specialize in eating these herbaceous foods, and their willingness to do so once fruits are scarce, appear to create a problem for chimpanzees. Every morning chimpanzees eat newly ripened fruits as their first main meal within a few minutes of leaving their night beds. They continue eating fruit until ripe fruits become too scarce to be found easily, which might be around midday. Then they resort to finding a patch of leaves or stems to eat. But if gorillas have been there first, the herbaceous food patch will be inadequate for the chimpanzees.

Bonobos, in contrast, without feeding competition from gorillas, have free rein to eat all of the ape foods that flourish in the environment. No other animal in the habitat of bonobos offers serious competition for eating these choice herbs, so bonobos can take the best of them. And that makes all the difference. An ape that can rely on cropping the herbaceous meadows can travel in relatively stable subgroups, slowly working their way from plant to plant. That is what gorillas do. The bonobos’ daily access to “gorilla foods” appears responsible for subgroups’ relative stability (as is also the case with gorillas) when compared with the shifting and smaller subgroups of chimpanzees.

Females greatly extended their periods of sexual receptivity,

The males became much less sure of when to compete with each other, so intimidation of females no longer paid off as it does for male chimpanzees. As selection increasingly favored the less aggressive males as mates, the self-domestication syndrome emerged. Homosexual behavior emerged spontaneously and was then woven into the bonobo social system as a means to strengthen bonds and reduce tensions.

The Island Rule applies not just to size, but also to many aspects of species’ growth and reproduction. Island animals tend to have delayed sexual maturity, to have fewer offspring in their litters, to live a longer time, and to have reduced sexual dimorphism—in other words, males are physically more similar to females than on the continent.

Island animals tend to be less reactively aggressive than their ancestral kin. Lizards, birds, and mammals all show the trend. Some animals abandon any efforts to defend territories, even though the continental relatives are fully territorial.

The reduction in reactive aggression is explained by an island’s being too small to hold a full complement of predators, which means that the risk of being killed is less than on the continent. As a result, animals survive longer on islands and live at higher population densities. Island populations are therefore relatively crowded, which means that being too aggressive can be overly exhausting. For example, defending a territory may not be an effective strategy when, as soon as the territory holder has chased one invader out, three more appear. If aggression does not pay, it is better not to waste time and energy and incur high risks by fighting. Under these conditions selection favors the less aggressive.

Select against reactive aggression, and the domestication syndrome emerges.

Helen Leach has already identified humans as having a domestication syndrome, based on our skull and skeleton. According to Belyaev’s rule, the implication is clear. In the course of evolution, humans experienced selection against reactive aggression. The fossil record becomes highly instructive, because a domestication syndrome is indicated as having been present during just one phase and species of the genus Homo. That phase is the last 300,000 years, and the species is Homo sapiens. To greatly simplify a rich and complex story: two kinds of Homo dominated our evolution for at least the last 250,000 years.

Neanderthals and their ancestors were all members of the archaic type of Homo, whereas the last species to emerge from Africa was the less hefty, more gracile form, Homo sapiens.

To be unambiguously Homo sapiens, skulls must be markedly round (globular) in profile with a clearly flexed base, and have such a small face that it is tucked mostly under the cranium. The earliest examples with these features come from the Omo River in Southern Ethiopia, dated to 195,000 years ago.

Based on genetic differences among living peoples, the ancestors that gave rise to everyone alive today are estimated to have lived between 350,000 and 260,000 years ago.

The people are heavily muscled, both men and women, more like wrestlers than runners. Their faces are strikingly broad and strong, particularly those of the males. Heads are somewhat slanted, sloping forward from the peak down to a great big brow ridge, and lacking a prominent forehead. That brow ridge is wide and thick, giving a daunting look to the eyes. Large mouths surmount a heavy, chinless jaw.

Fossils from elsewhere in Africa show that, sometime after 200,000 years ago, there were further reductions in the size of the face and brow ridge. Sex differences also diminished, as male faces became more feminine. Much later, during the Upper Paleolithic from 40,000 years ago onwards, the whole body also became lighter, to judge by a reduction in the diameter of the femur (the thighbone). The limbs became less robust in a further way too: they were less bony.

In the last 35,000 years, sex differences in height and tooth size have likewise been reduced. In all these ways, modern Homo sapiens is a less forcefully male species than our ancestors were 300,000 years ago. Our ancestors became feminized.

For the greater part of Homo sapiens’s existence, brains did not reduce in size. Instead, they grew larger. There are not enough preserved skulls from the Middle Pleistocene to be sure how big the brains of the earliest Homo sapiens were, but by 300,000 years ago they were probably about 1,200–1,300 cc, a little less than the average of about 1,330 cc in living people. For the next quarter of a million years the brains of Homo sapiens continued to increase in size, up to an average of a little more than 1,500 cc.13 Meanwhile, as the brains of Homo sapiens were slightly increasing in size, their shapes changed also. By 200,000 years ago their skulls had become increasingly round, or globular.

While the continuing increase in brain size during the Middle and Late Pleistocene shows that humans did not completely conform to the domestication syndrome, in one respect the brain growth in Homo sapiens does echo a pattern found in dogs.  By the time that Homo sapiens skulls stop growing, their shape resembles the shape of a Neanderthal skull that is in its penultimate phase of growth. Essentially, Neanderthal skulls (and by inference, their brains) continued to grow beyond the final point reached by Homo sapiens.  Some scientists regard the brain reduction as a further instance of the domestication syndrome. The drop in brain size is remarkable considering that for the last 2 million years the lineages leading to Homo sapiens have experienced a steady increase in brain size.

If self-domestication was indeed responsible for the changes associated with the origin of Homo sapiens, then the selection pressures that caused it must already have started before 315,000 years ago. The process appears to have accelerated over time, suggesting that the selection pressure against reactive aggression became increasingly strong since then and up to the present time.  How long before 315,000 years ago is a matter of conjecture. The self-domestication process could have begun, very slowly, around 400,000 years ago.  Or 500,000 to 600,000 years ago.

Neanderthals show no evidence of reduction in aggressive anatomy or a domestication syndrome. Their skulls and faces remained robust in Europe and Asia.

A key unresolved question is: What were the selection pressures that favored the evolution of modern humans in Africa around 200,000 years ago?

The main adaptation of Homo sapiens as our ability to accumulate cultural adaptations. We cannot live without the cultural knowledge that enables each new generation to re-create its society’s way of life. Naïve animals dropped into a new environment can often work out for themselves how to find food and survive. By contrast, humans mostly have to learn from others how to make a living by digging for edible food, cooking, fashioning tools, building houses, making boats, irrigating farmland, taming horses, making clothes, and so on. Without the learned skills passed down to us by previous generations, we are in trouble. With them, we dominate the planet.  Three features of Homo sapiens enabled us to accumulate those kinds of cultural skill: we are highly intelligent, we are highly cooperative, and we excel at learning from others—so-called social learning.

And it may have resulted from a crucial advance in food production. One population, which might have lived on the southern African coast, developed an ability to gather and hunt so well that their food resources became far more productive. The population naturally grew to the point where there was competition over the food supply, and soon groups were fighting over the best territories. Success in war became imperative. Groups accordingly allied with one another, giving rise to large societies of the type that hunter-gatherers form today. Cooperation among warriors within groups was so vital for winning conflicts that it evolved to become the basis of humans’ exceptional propensity for mutual aid. Sociality became more complex, learning became more vital, and culture became richer.

This theory identifies success in war as a plausible explanation for Homo sapiens’s outcompeting other species of Homo. It stresses that the process of producing Homo sapiens was not a single momentous event but rather a continual development, which fits the fact that our species has never stopped evolving, culturally or biologically. The paleontological evidence of brain size and cultural flourishing points are widely accepted, and the hypothesis that intergroup competition and warfare promoted sociality are constructively tied together

Two important problems concerning the evolution of Homo sapiens are not addressed—or any other theory of our origins, for that matter. First, none of the alternative theories accounts for the self-domestication syndrome evident in Homo sapiens. The importance of cooperation is an important ability, but ignores the fact that cooperation depends on a very low propensity for reactive aggression. Blumenbach, Darwin, and many subsequent thinkers would surely have seen this as a critical omission.

Given how much less emotionally reactive humans are than chimpanzees, bonobos, or most group-living primates, a low propensity for reactive aggression cannot be taken for granted in our Mid-Pleistocene ancestors. Reduced reactive aggression must feature alongside intelligence, cooperation, and social learning as a key contributor to the emergence and success of our species.

Docility should be considered as foundational of humankind, not just because it is unusual, but because it seems likely to be a vital precondition for advanced cooperation and social learning. The importance of tolerance can be seen in that wild, chimpanzees cooperate in territorial patrols and alliances against others, but in captivity, they often show little interest in working together. Among captive chimpanzees, the less reactivity within a pair, the greater their ability to work together.

A study of spotted hyenas focused specifically on aggression and showed that cooperation was better in pairs that were not merely more tolerant but also less aggressive. Similar links between tolerance and cooperation have been seen in a number of mammals and birds, including macaques, marmosets, ravens, and keas (a terrestrial parrot species).  Considerable evidence supports the idea that the evolution of cooperation depends on tolerance.

Another factor is that cooking may have led to decreased jaw and tooth size. But those one-by-one ideas do not solve the problem totally. Why are the characteristic features of Homo sapiens congruent with the domestication syndrome found in animals? Perhaps the reason that human features are described by the domestication syndrome is simple: humans are a domesticated species.

Selection against aggressiveness and in favor of greater docility came from execution of the most antisocial individuals.

In Darwin’s 1871 discussion of human evolution, The Descent of Man, and Selection in Relation to Sex, he sketched a simple version of the execution hypothesis as a way to explain the evolution of two important features—reduced aggressiveness and increased social tolerance—that nowadays we regard as central to domestication. The reason Darwin wanted to explain how aggressive tendencies had been reduced, was that he considered the evolutionary reduction of aggressiveness to be a problem of morality and was anxious to provide an evolutionary explanation for positive moral behavior.  The idea that morality was God-given posed a challenge to Darwin’s evolutionary theory, because Darwin proposed that all of life’s features had evolved without the intervention of a deity. If evolutionary theory was to be as complete as Darwin hoped, he had to explain morality without invoking the influence of religious beings.

Darwin had an answer to his question about the fate of exceptionally aggressive men. “In regard to moral qualities,” he wrote, “some elimination of the worst dispositions is always progress even in the most civilized nations. Malefactors are executed, or imprisoned for long periods, so that they cannot freely transmit their bad qualities….Violent and quarrelsome men often come to a bloody end.”

Nowadays, he was saying, criminals and aggressive wrongdoers are punished by the law. If “they cannot freely transmit their bad qualities,” their traits are less likely to be inherited by the next generation. If comparable kinds of punishment had been applicable throughout human evolution, genes promoting aggressive behavior would have been steadily selected against. Generation by generation, less aggressive, more positively moral behavior would tend to spread.

Punishment of criminals in Darwin’s time, the Victorian era of nineteenth-century Britain, was made possible by features of contemporary society not found among nomadic hunter-gatherers. Police, written laws, trials, and prisons all contributed to sanctioning the violent. Until recently, our ancestors had none of these institutions. But Darwin recognized that even if prehistoric human societies were different from today, they might still have found ways to deal harshly with “violent and quarrelsome men.” If exceptionally aggressive men were always routinely punished in ways that reduced their reproductive success, there would have been eons of prehistory in which the culling of violent men could lead to evolutionary change. Darwin’s conclusion was forthright. The morality problem could be solved by an ancient system of execution leading to the eradication of selfishly immoral individuals, which would lead to selection against selfish tendencies and in favor of social tolerance. Through this kind of natural selection, he wrote, “the fundamental social instincts were originally thus gained.

Selection favored moral behavior thanks to the “bloody end” of “violent and quarrelsome men.  The parochial altruism hypothesis for the evolution of cooperation appeared in The Descent of Man alongside his execution hypothesis. It was based on the benefits of cooperation rather than the costs of aggression. Although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, yet that an advancement in the standard of morality and an increase in the number of well-endowed men will certainly give an immense advantage to one tribe over another. There can be no doubt that a tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage and sympathy, were always ready to give aid to each another and to sacrifice themselves for the common good, would be victorious over most other tribes….At all times throughout the world tribes have supplanted other tribes; and as morality is one element in their success, the standard of morality and the number of well-endowed men will thus everywhere tend to rise and increase.

This type of explanation—that ingroup solidarity fosters success in between-group competition—continues to attract scholars up to the present day.

International solidarity is often imagined as our earthly response to the arrival of space aliens. In fiction and reality, war can promote cooperation within groups.  In intergroup conflicts chimpanzees have very similar rates of death as in hunter-gatherer war.  Theoretical problems aside, a fundamental issue regarding parochial altruism is that it has not been shown to occur among human hunter-gatherers.

Hunter-gatherers in face-to-face battles tended to stop fighting after the first injuries. Until evidence is produced that hunter-gatherers exhibit risky self-sacrifice in war, parochial altruism should be regarded as a culturally induced behavior rather than an evolutionary product of selection. Japanese kamikaze pilots who flew their planes into enemy ships during World War II, or Islamist suicide bombers, responded to intense cultural pressures more than innate inclinations.

Darwin also decided that the effects of war for promoting solidarity within groups is cultural. The social instincts could not have evolved as a result of intergroup fighting, he said, because, even within the most cooperative and morally virtuous tribes, some people would be more selfish than others; and the more selfish people would have more babies than the moral. “He who was ready to sacrifice his life…would often leave no offspring to inherit his noble nature….Therefore it seems scarcely possible…that the number of men gifted with such virtues, or that the standard of their excellence, could be increased through natural selection, that is, by the survival of the fittest.” Parochial altruism, or self-sacrifice in war, might be explicable in specific societies by their military cultures or promotion of risk-taking ideals, but not by evolution.

The theory has been important, because it has dominated efforts to explain the exceptionally positive aspects of human sociality. The focus on self-sacrifice distracted from the question of why humans are so docile and eclipsed Darwin’s speculation about the “bloody end” of “violent and quarrelsome men.” For a century, the question of reduced aggression was forgotten, and the execution hypothesis ignored.

Explaining humans’ extremely high cooperative tendencies has garnered more attention than explaining our strikingly low reactive aggression.

At some unknown point in our evolution, language skills developed to the point where gossip became possible. Once that happened, reputations would become important. Being known as a helpful individual would be expected to have a big effect on someone’s success in life. Good behavior would be rewarded. Virtue would become adaptive.

Chimpanzees can demonstrate negative feelings toward others, but they cannot explain why they feel that way. They cannot gossip about the guy who bit a rival, or slapped a female, or stole some food, or whether a particular chimpanzee is trustworthy, generous, or kind. Their communication skills are simply inadequate. Obviously, this implies that chimpanzees would not care about their reputation.

In a setup where a chimpanzee could steal food from another chimpanzee, at times a potential thief was watched by a third individual. If being observed matters to chimpanzees, they should be less likely to steal when a third individual is watching. As expected, the presence of a third individual made no difference. The same was true when the experiment was about helping rather than stealing. Chimpanzees behaved selfishly or helpfully according to their own proclivities, and did not modify their behavior according to whether they were observed. Reputation did not appear to matter.

Individuals choose whom to interact with, depending on how they have been treated in the past. Good cooperators tend to be favored. Bad cooperators tend to be shunned. The same is true in many species. So the absence of concern about reputations among chimpanzees does not stem from any lack of individual differences, nor from an inability to evaluate others. Chimpanzees know that partners vary in quality, but they have to use that information on their own. They cannot talk about it.

Unlike the chimpanzees, for preschoolers the presence of an observer mattered. When they were being watched, they stole less and helped more.  So the social pressures wrought by the effects of a good or bad reputation may underlie the evolution of morality. In the short term, individuals with bad reputations might mend their ways and become conforming members of society. But over the long term, the effects of a bad reputation would have genetic evolutionary consequences. Individuals who were too feisty, short-tempered, or selfish ever to adapt successfully to the criticisms of their peers would have poor chances of surviving and reproducing well.  Selection would thereby have favored the kind, cooperative, tolerant types: the morally positive ones, less aggressive than their ancestors. Our ancestors would have evolved to be a nicer species. Language begat reputation, and reputation begat morality.

We can probably all remember bullies from our school years who were so powerful that they did not care what some of the less well-connected children thought of them. They were big and bold, and if others resented them, so what? The bullies got what they wanted without the approval of the mutterers. They were not stopped simply by gossip. They had to be stopped by someone fighting back, or by adults who whisked them off to detention. Translate those types into the past and we are faced with a vital question unsolved by the reputation hypothesis. Why would a bad reputation matter to a male who was bold and strong enough to take matters into his own violent hands? How would his bad reputation have inhibited him if, like an alpha chimpanzee, he did not care what others thought?

A female might have preferred a kinder, gentler male as a mate. But how was she to stop a domineering male from coercing her? More tolerant men might provide more meat. But what would prevent a powerful individual who was willing to throw his weight around from refusing to be denied, aggressively seizing more than his share? Tyrants who did not care about their reputations could bully their way to getting more than others—whether more food, more matings, the best sleeping-spot, or more social support. That is what happens among chimpanzees.

Even among today’s men (but not women), since 2008 facial breadth has been discovered to be correlated with a propensity for reactive aggression. The time when male faces become relatively broad compared with female faces is in puberty, apparently under the influence of testosterone. In professional hockey games the number of minutes spent in the penalty box tends to be greater for broader-faced men than for men with narrower faces. In general, among European whites, wider-faced men have been found to have a higher propensity not only for aggression and retaliation but also for more self-centered and deceptive behavior, less cooperative negotiation, a higher score on the psychopathic trait “fearless dominance,” and a higher score for self-centered impulsivity.

Subjects who are unaware of these findings tend to treat broader-faced men warily, as if recognizing that a relatively broad face is a signal of aggression. This unconscious sensitivity to a man’s facial breadth suggests that during human evolution, broader-faced men would have demonstrated more socially undesirable behavior, and that our broad-faced Pleistocene male ancestors would have been relatively impulsive, fearless, uncooperative partners who were swift to act aggressively in defense of their selfish needs.

The difficulty with the reputation hypothesis is that it does not explain why such tendencies would have been reduced. Physical aggressors could have succeeded in bullying their way to the top.  The same challenge faces all ideas for explaining the evolution of human goodness that do not address the control of aggression.

The only proposal sufficient to explain how our ancestors overcame the problem of determined bullies, which was surely a vital first step to the elaboration of cooperative behavior, is the elaboration of Darwin’s idea that “violent and quarrelsome men” came to a “bloody end.” The execution hypothesis claims that, during the Pleistocene, a new kind of ability crystallized. For the first time, coalitions of males became effective at deliberately killing any member of their social group who was prepared to use violence on his own behalf and simply did not care what others thought about him. In the end, execution was the only way to stop such a male from being a tyrant.

In the 17^th century, hundreds of felonies in America had been capital crimes. In New England, you could have been executed for witchcraft, idolatry, blasphemy, rape, adultery, bestiality, sodomy, and, in New Haven, masturbation. You could be put to death for being “a child of 16 or older who was a ‘stubborn’ or ‘rebellious’ son, or who ‘smote’ or ‘cursed’ a parent.” These penalties were far from theoretical. From 1622 to 1692, Essex and Suffolk Counties of Massachusetts recorded the executions of 11 murderers, 23 witches, 6 pirates, 4 rapists, four Quakers, two adulterers, two arsonists, and two each charged with bestiality and treason.

The principle of execution was popular. It was not unusual for criminals to be hunted down by citizens, tried, convicted, sentenced, and executed in less time than the four days that were supposed to elapse after the pronouncement of a death penalty.  Spontaneous mob killings reflected the community’s enthusiasm for savage justice. Fischer describes how women in Marblehead whose husbands had been taken by Indians “seized two Indian captives and literally tore them limb from limb.

From the first recorded capital sentence in America in 1622 (for theft) to 1900, there were perhaps 11 to 13,000 legally determined executions; mob violence or lynchings were thought to have killed an additional 10,000 during the same period. Occasional lynchings and unofficial executions still bedevil the powerless.  It was not until the late 18^th century that capital punishment started to fall in popularity. Before then, if you challenged the rules, you risked death. The counterpoint was impressive. Trouble was rare. Home owners could sleep with their doors open. They did not have to lock up their valuables. As long as you followed the rules, New England was a place of peace.

Executions and the reasons for them have been described in every society with written records. Capital punishment was present in all the earliest civilizations, from Egyptian, Babylonian, Assyrian, Persian, Greek, and Roman to Indian, Chinese, Inca, and Aztec. It happened not only for violent crimes but also for nonconformism (as in Socrates’s case), for minor felonies, and even some heartbreakingly trivial matters such as malpractice in selling beer (according to the Code of Hammurabi), or stealing the keys to one’s husband’s wine cellar (according to the laws of the early Roman Republic). Executions were an accepted part of life, often drawing great public crowds to watch spectacles of unequivocal cruelty. They continued in all historically known societies until the 1764 publication of On Crimes and Punishment, by the Italian jurist Cesare Beccaria. Beccaria’s arguments against the death penalty helped launch the change that continues today in community attitudes toward capital punishment. Prisons then increasingly took over the responsibility for social control.

So you can see how dangerous the social dynamics of a small anarchic group can be.  To most people growing up in comfortable state societies, the idea that a group of closely connected people who know one another well can agree to kill one of their own is disturbingly strange. Yet, during times of crisis, the same practice has sprung up in the richest nations. It happened in the Second World War concentration camps, where prisoners were desperate for food. Food sharing was common despite the suffering, a mark of humanity in dark times. But stealing was common, too, and it was reviled. Prisoners worked out how to stop it. Rudolf Vrba wrote about the “bread law” in Auschwitz: “If a man stole your food, you killed him.” The bread law sustained the community of sufferers. “If hunger so demoralized a man that he stole another’s bread, no one reported him to the SS or even to the Block Leader. The room attendants themselves took care of him….If he did not die of the beating, they so incapacitated him that he was fit only for the crematorium.”

The purposes of capital punishment are in some ways different in state and small-scale societies. In state societies, capital punishment has often served to remove individuals who challenge the leader. In keeping with those aims, states often execute ostentatiously in public, especially during the early years of a state’s existence, when power structures are still unstable. Kim Jong Un, the North Korean president, inherited his position from his father in 2011. In the first four years of his reign, he reportedly executed at least 70 of his subjects, including his vice-premier, his minister of defense, and others.

The ironic and disturbing conclusion is that egalitarianism, a system that appeals because of its lack of domineering behavior, is made possible by the most domineering behavior in the human arsenal. To understand what it means to be egalitarian, consider the structure of hunter-gatherer society. A typical society averages close to a thousand people, sharing the same unique language (or dialect) and cultural practices, such as the rituals that they practice at funeral ceremonies. Societies are too large for everyone to live together, because the environment does not afford enough resources to support hundreds of people in a small area. People typically live in bands averaging fewer than 50 people. Each band occupies its own subregion within the society’s territory and tends to stay in one place for a few weeks at a time. When foraging becomes too arduous because nearby resources are exhausted, the band moves on, normally reoccupying a previous campsite.

The egalitarianism that is such a special feature of relationships among hunter-gatherer men is centered on the five to ten married men within a band. Those few husbands are the “elders,” or “cousins”. The elders behave like a boardroom without a chairman. Everyone has a voice, but they all show considerable reluctance to use it. Men are so averse to grandiosity that self-deprecation is a highly regarded part of public behavior. It was important to show shame and embarrassment because they demonstrate to others that one does not have a conceited view of oneself.

Men and women were not equal in hunter-gatherer societies

Equality among the Ju/’hoansi has been claimed to apply to all adults, but if a woman is beaten by a man, his punishment is minimal at most. Tanzania’s Hadza hunter-gatherers are said to be egalitarian, but if a hot area has few shade trees, the men get the shade while the women sit in the sun. Still, even if some individuals’ voices hold little sway, there is no such thing as a group leader who can demand obedience.

Most group-living primates, by contrast to humans, have a clear dominance hierarchy enforced by brute fighting ability. To the extent that there is leadership in hunter-gatherer bands, such as in taking initiative for group decisions, prestige is the important criterion. People compete for influence mostly by producing good arguments, creating good plans, being the best mediators, telling the best stories, or seeing the future most convincingly. A person who is skilled in these ways might be recognized as a leader or headman, but that role would be earned by his or her being wise and persuasive rather than assertive, pushy, or a good wrestler. Although leaders can be admired and respected, they cannot enforce their ideas, nor can they use their position to take anything from other members of the band. An inability to dictate to others means that, among mobile hunter-gatherers, there is no alpha position.

Hadza men recognize the hazard of being shot when asleep in camp at night or being ambushed when out hunting alone in the bush. The absence of alphas among nomadic hunter-gatherers was caused by killing: The means to kill secretly anyone perceived as a threat…acts directly as a powerful leveling mechanism. Inequalities of wealth, power and prestige…can be dangerous for holders where means of effective protection are lacking.  A potential bully always seems to be waiting in the wings. For all their famous egalitarianism, hunter-gatherer men can be unpleasantly competitive.

The anthropologist Richard Lee described his own experience of being humbled.  He surprised the tribe with a magnificent ox.  Their response to the splendid gift of meat shocked Lee. The men were insulting. They said the ox had no meat on it. They called it a bag of bones. They said the ox was so thin they would have to eat the horns. Eventually, an elder called Tomazo explained what was going on: the gift had made Lee look arrogant. “When a young man kills much meat,” said Tomazo, “he comes to think of himself as a chief or a big man, and he thinks of the rest of us as his servants or inferiors. We can’t accept this. We refuse one who boasts, for someday his pride will make him kill somebody. So we always speak of his meat as worthless. This way we cool his heart and make him gentle.”

If an offender did not respond to subtle disapproval, his or her resistance was clearly deliberate. The protests then became more obvious. Shaming, ridicule, and cutting references occur. A song of derision might be sung in the offender’s face while he stood before the assembled community. If more drastic measures are needed, shunning or ostracism is often effective because being avoided and excluded is deeply painful to most people living in small groups, and will normally do the trick.

When the anthropologist Jean Briggs was living with a remote Arctic group, the Utku, she made the mistake of losing her temper. Slushy ice had dropped from the roof of the igloo onto her typewriter. She threw a knife into a pile of fish and railed against the endless fish diet. The igloo emptied fast, and over the ensuing weeks, Briggs found herself left alone in her tent, with no one calling on her anymore. She found the experience desperately painful until she finally found a way to explain herself.

One way in which the practice of capital punishment can reduce societal aggressiveness is by encouraging conformity. In the 1960s the rural Bang Chan people of central Thailand, a largely Buddhist population of rice growers, became renowned as one of the world’s most peaceful societies, a place where fighting, domestic violence, and child abuse were virtually unknown. This owes much to parenting and social control. Children and adults are regularly reminded not to be too pushy. In conversations criticism was eight times more frequent than praise. Every culture seems to have learned the trick of taming the next generation through socialization, using social control more than reward. The fear of capital punishment can doubtless contribute to encouraging a spirit of conformity and restraint.

The idea of the execution hypothesis is that during thousands of prehistorical generations, the victims of capital punishment were disproportionately those with a high propensity for reactive aggression. Killing or repression of such individuals is supposed to have happened so often that our species evolved a calmer, less aggressive temperament.

Even though capital punishment has been widespread, the idea that it would have been sufficiently frequent to have evolutionary effects on aggression may seem surprising at first. But considering the length of time, and the number of generations within that period, the rate of evolution for self-domestication appears relatively slow. As we have seen, the process of self-domestication could have begun at least 300,000 years ago. Three hundred thousand years is equivalent to about 12,000 generations; the domestication syndrome apparently intensified in the later times. The number of generations is considerably greater than the number for any mammals to become domesticated, such as the roughly 15,000-year evolution of dogs from wolves.  The average generation time for wolves is between 4 and 5 years, suggesting that dogs have been separated from wolves by fewer than 4,000 generations.

In highland New Guinea, Ray Kelly was documenting witchcraft killings among the Etoro; the victims were typically people known for their “archetypical selfishness and lack of responsiveness to others.” Kelly found that 9% of 55 adult deaths were by execution. Rumor, fear, competition, and ignorance can combine into a surprisingly deadly brew.

Language seems necessary for planned killings of specific individuals. The criticisms and ridicule given in response to an aggressor depend on gossip for their conspiratorial power. When those gentler methods of social control do not work, people start floating the idea of killing the offender. For that, linguistic ability is vital, and considerable skill is needed. In Boehm’s words, “The tactical problem is obvious: whoever speaks up first may be putting his life in danger. The danger may involve a physical attack, or in bands that have shamans it may also incorporate sorcery.

Gossip solves the coordination problem by allowing individuals to test their feelings cautiously and to generate shared plans. Julius Caesar’s more than twenty assassins talked in small secret groups for weeks to gain confidence in one another. After individuals sound one another out sufficiently, a consensus builds to the point where, if the troublemaker does not respond appropriately, he or she can be killed by collaborative decision. Even then it can be dangerous for the initiators.

Neanderthals do not show the signs of self-domestication that Homo sapiens do. Because Neanderthals did not achieve the same degree of cultural complexity as Homo sapiens, and particularly because their symbolic culture was much more limited, most experts think that their linguistic ability was less than ours. There are also some tentative biological indications supporting this claim, such as the temporal lobes of the brain, which are involved in language functions (as well as memory and social behavior), are relatively larger in Homo sapiens than in Neanderthals. The evidence of skull and skeleton, cultural complexity, and brain all suggest that Neanderthals did not have language as we know it.

Linguistic ability improved substantially in the Homo sapiens lineage compared to all other Homo. With that improvement came the ability for individuals within a group to form coalitions that excluded or ostracized a member of the group who had become a domineering aggressor. Those coalitions enabled human selection against excessively aggressive men. The result was continuing change in the direction of a more cranially gracile, paedomorphic, and tolerant species—or self-domestication.  Language was the key feature of Homo sapiens that allowed many tools of social control, from gossip to killing.

But language is not the only feature that has been proposed as the key to unlocking the domestication syndrome. Various scholars have focused on the use of weapons as a way to explain the transition from a primate system of alpha males to a human system of egalitarianism and cooperation. Weapons would have been useful in organizing and launching the move against an alpha male, because thrown rocks or spears could make a proactive attack more overwhelming and easier to carry out safely. Several points seem to me to undermine the importance of weapons, however, compared with the development of language. Weapons are not needed for capital punishment. Animals such as wolves, lions, and chimpanzees kill using collaboration, not weapons. Humans can also kill without weapons. In his worldwide survey of small-scale human societies, Otterbein admittedly found that stoning, spearing, and shooting were frequent execution techniques, but also noted hanging, burning, drowning, beating, throwing off a precipice, enforcing suicide by requiring a guilty party to jump off a tall tree, and (for rape) forcing a briar into the penis.

Another method was to give the victim to a vengeance party from a hostile neighboring group and let them deal with the problem, as happened in Australia. One day, a group of them encouraged the bully to climb high into a tree to extract honey. Unsuspecting, he laid down his weapons before climbing up. His assassins were now safe: they gathered the victim’s weapons and merely had to wait for him to come down before easily killing him. The importance of ensuring that the victim is helpless at the time of the attack dramatizes the importance of attackers’ sharing their intentions with one another—making plans by using language.

The vital human novelty that propelled the origin of a new kind of political system was plotting. The ability to plot together, rather than the ability to make weapons, surely determined when the change occurred in the balance of power between the classic alpha-male type and the new coalition of subordinates. The ability to plot together is an example of what psychologist Michael Tomasello calls “shared intentionality,” defined as “collaborative interactions in which participants share psychological states with one another.” Humans excel at shared intentionality, which emerges in children at about the age of one year, whereas chimpanzees show barely any evidence for it. Tomasello considers that the uniquely human development of shared intentionality explains why humans can do many special human things, from using math and building skyscrapers to playing symphonies and forming governments. If the hypothesis that selection against reactive aggression led to the domestication syndrome is correct, however, none of those human abilities was as special as the one that enabled conspirators to trust one another sufficiently to collaborate to kill a bully. That ability could both have domesticated us and made possible many kinds of human cooperation.

Even if reasons for the origins of language remain mysterious, the magnitude of its impact points to the origin of Homo sapiens as the time when linguistic ability took a major step forward.  The development of increasingly skilled language thus provides the best basis for the ultimate explanation of human domestication.

To explain anatomical changes in the evolution of Homo sapiens, conventional scholarly wisdom presents an idea different from self-domestication. Paleoanthropologists have traditionally interpreted the special anatomical features of Homo sapiens as a series of parallel adaptations, rather than as incidental by-products. To explain the lighter build, shorter faces, and feminization of our lineage, they have invoked forces such as changed climate, better diets, or increasing sophistication in tool use. Those proposals are very reasonable if one assumes, as scientists often do, that biological features always evolve by the direct action of natural selection.

If our docility was an incidental by-product rather than an adaptation, it might have evolved without ever being positively selected. The evolutionary anthropologist Brian Hare raised this possibility based on the finding that species with larger brains are especially good at inhibiting their responses. Lower reactive aggression would thus be a by-product of a more general tendency toward lower emotional reactivity.

Looking across 36 species, what feature is associated with success at inhibiting initial reactions? The MacLean-Hare team found that the probability of success varied systematically with brain size. Species with bigger brains did better. The reason that brain size leads to better inhibition, or self-control, is most likely that bigger brains contain more cortical neurons. The cortex is the source of willpower and voluntary control over emotions. In species with bigger brains, a higher proportion of the brain is cortex, and more cortex means more neurons. Humans have more neurons in the cortex than any other species, about 16 billion. The great apes and elephants follow, with a mere 6 billion or so. More neurons in the cortex—especially the prefrontal cortex, at the front of the brain—allow an animal to inhibit its emotional reactions better.

The inhibition experiment raises the possibility that our low propensity for reactive aggression results from our large neural networks, which enable us to think before we act. But in practice, this explanation seems unlikely to contribute very significantly to our low aggressiveness, because our emotional responses tend to be too quick to be controlled completely by the cortex. Still, the idea offers a second alternative theory for Pleistocene self-domestication.

Domestication leads to changes in the age at which fear responses become stronger, in playfulness, in sexual behavior, in the speed and effectiveness of learning, and in the ability to understand human signals there are also physiological changes in the production of hormones and neurotransmitters, and in the size of the brain and its component regions.

Dog breed-specific features have evolved by various different mechanisms. They are not consistently paedomorphic (child-like). Selective breeding has pulled dog crania in all sorts of different directions.

But in addition to breed-specific features, some uniform species-wide aspects are found in the skulls of all dogs regardless of breed. These features of the shapes of dog skulls are what make a dog a dog.

In every mammal, juveniles tend to be friendly. Compared with adults, they are strikingly unafraid and inquisitive. That is why petting zoos, where children can meet animals close up, use animals that are not merely domesticated but also juvenile. The evolutionary reason for juveniles’ friendliness is that they must learn whom to trust when they are older and on their own. Youth is the perfect time for them to learn, because their mothers protect them from the wrong kind of social interaction. As long as they are helpless wards of the mother, the juveniles can trust her judgment and do not need to be fearful. They can let their guard down and be open to developing trusting relationships.

But in any mammal’s life, there is a predictable change as it grows older, more mobile, and less likely to be protected by its mother. It becomes more easily frightened and then, as a response, more aggressive. At an age that varies across species, a fear response kicks in.

Unselected silver foxes become fearful at about 45 days old (six and a half weeks). From that time onward, the fox puppies show fear and aggression toward unfamiliar individuals, whether other foxes or humans. Their socialization window, as it is called, has closed. The bonds they have formed up to that time can last a lifetime, but youthful innocence has gone. Henceforward, they will find it hard to trust unfamiliar individuals. After the socialization window closes, even dogs can be difficult to train. The socialization window closes later in dog puppies, 8 to 12 weeks, than in wolf puppies (six weeks).

The development of the fear response is likewise less acute in dogs than in wolves. At the end of the socialization window, fear crystallizes sharply in wolves, whereas in dog puppies fear of new individuals rises more slowly and is only fully expressed by around three months.

Some genes that are functional in the prefrontal cortex are identical in chimpanzees and humans, but the time when these genes actively make proteins and other products (in other words, the time when the genes are expressed) is several years later in humans than in chimpanzees. The largest differences are found in genes that help form junctions between nerve cells, called synapses. In chimpanzees, the peak expression of these synapse-forming genes occur when the youngster is less than one year old. In humans, by contrast, peak expression is extended to five years old. Human brain development is thus greatly delayed.

Nerve myelination is similarly delayed in humans. Myelination is a process that makes nerve impulses travel faster by coating neurons in a protective layer of fatty myelin (and thereby making them into the familiar “white matter” of the brain). The downside of myelination is that myelinated neurons lose their ability to grow and form new synaptic connections. In chimpanzees, myelination ends at around 10 years of age, whereas in humans it continues until individuals are as much as 30 years old.

The skulls of Neanderthals and Homo sapiens change in parallel as they grow, but Neanderthal skulls continue to grow after reaching the equivalent of the end-stage for Homo sapiens. In that final period, which occurs in Neanderthals but not in Homo sapiens, Neanderthal faces grow bigger compared with their braincases. Neanderthal skulls are not identical to those of the less well-known Mid-Pleistocene Homo, which spawned our lineage, but they are sufficiently similar to offer a helpful model of our ancestor.

Developmental delays have been found in the behavior of many domesticated animals compared with their wild ancestors, including in social relations, play, learning, sexual activity, and vocalization. Weaning (at around two or three years) is two to three years earlier in hunter-gatherers than in apes, but after weaning, children continue to rely on food supplied by their parents for longer than is seen in any other animal. Humans are strikingly playful as young adults, and even later in life. Humans, of course, are the ultimate learners, both as children and as adults. Human sexual behavior is frequent and prolonged, and strikingly emancipated from its purely reproductive function. Language makes Homo sapiens a complete outlier with regard to the elaboration of communication.

Neanderthal material culture is now known to resemble that of Homo sapiens closely. Neanderthals apparently lived much like recent mobile hunter-gatherers. They controlled fire, cooked their meals of hunted meat, and gathered plant foods or shellfish. Their prey ranged from doves to woolly rhinoceroses. They had cozy home bases, where they slept on furs and likely treated themselves with medicinal herbs. They had status markers such as bird-wing cloaks, they sometimes buried their dead, and they used deep caves for activities presumed to be rituals. After 300,000 years ago they used the sophisticated Levallois method for making stone tools. They could make fine blades, pigments, ornamental beads, and engraved artwork. As early as 200,000 years ago, they synthesized pitch from birch bark “through a multi-step process that relied on strict control of temperature and required a dry distillation excluding oxygen. Neanderthals expressed their creativity at a lower rate than Homo sapiens. Fewer than ten ornamental beads worked by Neanderthals have been found from 200,000 years of rich archaeology, compared with thousands wrought by Homo sapiens before and after their swift takeover in Europe. Similar comparisons apply to stone blades, stylized figurines, ritualized burials, and engraved symbols. There were also some abilities that Neanderthals appeared not to have had at all. They seem to have lacked facilities for the long-term storage of food. There is no evidence that they built sleds, despite living through harsh winters. Nor do they appear to have produced or used boats. Homo sapiens colonized Australia about 60,000 years ago, in a journey that requires multiple legs of sea travel, so they clearly negotiated water crossings somehow. Homo sapiens in South Africa made bows and arrows (evidenced by arrowheads), spear-throwers, and fine bone points by 71,000 years ago, but none of these important tools have yet been proved for Neanderthals. Although Neanderthals used fire, they did not use it to make better stone tools, as South African Homo sapiens did, or to heat water by using hot rocks.

Based on the idea that Homo sapiens self-domesticated and Neanderthals did not, Homo sapiens may have been better at cooperating. Evidence of regular inbreeding in Neanderthals points to their social networks’ being small. The genome of a Neanderthal woman from Siberia showed that her parents were close relatives, such as half-siblings or an uncle and niece, and that similarly inbred matings had occurred regularly among her recent ancestors. In small-scale societies of humans, groups break up into smaller units when social tensions erupt. Groups so small that matings were often inbred could have resulted from a tendency among Neanderthals to react to one another overly quickly with aggression, and consequently breaking up into ever-smaller groups.  Some have argued that Neanderthals differed from Homo sapiens in the construction of items that required coordinated labor. The more elaborate the cultural skill, the more cooperation is needed. Thus the reason for the poorer manifestation of Neanderthal culture was that they were less adept socially, rather than intellectually.

Differences between Homo sapiens and Neanderthals may have been due more to emotion than to intellect.

Group benefits are not the only reason that morality might have evolved. Even when moral behavior leads to group benefits at an apparent expense to the individual, the individual’s behavior might actually serve his or her own selfish ends. There are various forms of this idea. A benign version is that group-directed moral responses lead to the individual’s being able to form useful alliances for cooperation, as the philosopher Nicolas Baumard and others propose. A darker form is that morality is all about self-protection. I have suggested that capital punishment emerged with language in the Mid-Pleistocene. Afterward, an individual who challenged the dominant culture could be in mortal danger. Sensitivity to social disapproval would have been favored as never before. As a result, individuals might behave in morally correct ways in order to promote their own survival. Group benefits would then be incidental.

The evolution of human morality was largely a response to the same power plays that self-domesticated us: we evolved to fear the killing power of the men in the group. The idea explains why group-directed moral emotions have been elaborated far more strongly in humans than in any other species.

Evolutionary theorists from Charles Darwin onward have commonly treated morality as being concerned only with altruism and fairness. However, being moral can include not only acts of kindness but also deeds of conformity and violence. Being morally good might mean restraining yourself. You can be good, according to some societies’ moral prescriptions, if you refrain from various individual activities that are deemed “wrong,” such as committing suicide, masturbating, or burning your national flag.

Why have we evolved to be nicer to one another than other mammals are?  What selective pressures caused our emotions to evolve as moral guides? The last is about the interfering aspect of morality. Why did we evolve to monitor not only our own behavior, but others’ behavior, too?

Moral behavior could result entirely from cultural indoctrination. Frans de Waal names this the “Veneer Theory” of morality, the notion that human morality is a purely rule-bound system lying atop a foundation of ancient animal-derived behavior that lacks morality, like a veneer of fine lacquer on a wooden box. The Veneer Theory is a nonstarter, however, because moral actions are produced partly by evolved moral emotions. Untutored children have prosocial inclinations that are not explained by kin selection or mutualism. The developmental psychologist Felix Warneken showed that 18-month-old toddlers will help a random adult who asks for help. The infants will pick up accidentally dropped objects, for instance, or hold a door open to allow the grown-up to put toys away. Significantly, experiments show that these kinds of helpful behavior are not explicable by the infants’ merely trying to interact, or being intimidated, or wanting to be stimulated. They simply want to help, even when it costs them. They will give up their own food to an experimenter when the experimenter’s bowl is empty, or sacrifice their own toys for the sake of others.

People often expect religious believers to act in especially prosocial ways, which they often do, but religiosity is not always a predictor of moral kindness. A study of sharing among 1,170 children in six countries and four continents found a tendency for those brought up in more religious families to be less altruistic than those from nonreligious families.

Since 1982, the Ultimatum Game has provided a standardized context for studying moral choices. The game allows investigators to study people’s choices about sharing a resource with a stranger. Conventional economic theory predicted that decisions would depend on self-interest. However, in worldwide tests in more than 30 countries, from hunter-gatherers to the Harvard Business School, both adults and children are spontaneously and routinely more generous than expected by theories of economic maximization. This result makes humans very different from chimpanzees—and probably any other nonhumans.

Captive chimpanzees behave like the imagined Homo economicus, a species in which individuals always try to maximize their personal economic gains. Decider chimpanzees accept even the smallest rewards offered by Donor chimpanzees; unlike humans, they never reject “unfair” offers. The stark difference draws attention to the uniqueness of the human moral senses.

A popular kind of solution is group selection. Group selection theory suggests that self-sacrifice by an individual can be favored over evolutionary time if it provides sufficiently large benefits to the individual’s group, which normally means a social breeding unit such as a hunter-gatherer band. Very often, however, the group that benefits from an individual’s generosity is not a social breeding unit. As Robert Graves’s recollection of his school days reminds us, the beneficiaries might be only a subgroup of a given social network. In the group as a whole, moral behavior might benefit some individuals at the expense of others.

The second major problem is how we come to classify some actions as “right” and others as “wrong.” Scholars looking for the consistent application of moral rules have traditionally considered two main ideas. They are the “utilitarian” and the “deontological” principles. Both work sometimes, but neither is followed all the time, which means they fail as general explanations.

People do not follow any general moral principle. Instead, moral decisions are influenced by a series of unconscious and unexplained biases. Three are particularly well studied. The “Inaction Bias” pushes us to do nothing rather than something. Suppose you are caring for a terminally ill patient. Most of us would rather deliberately withhold life-prolonging treatment than give the patient a lethal injection. We prefer an act of omission to an act of commission.

Would you prefer to order your bombers to kill civilians so as to break the enemy’s will, or to attack an army base so as to reduce the enemy’s military capacity? Given that the number of civilian deaths is expected to be the same either way, most people would prefer to bomb a military target.

The third major puzzle about our moral psychology is why, as a species, we have evolved to be so sensitive to the abstract notions of right and wrong that we monitor one another’s behavior and even sometimes intervene to punish someone whose behavior we disapprove of.

Chimpanzees kill other females’ babies.  In one situation, a kidnapped baby clung to Passion, who calmly dispatched her by deliberately biting into her skull. While Gilka watched, Passion, Pom, and Prof all cannibalized the victim. The proactive attack proved part of a pattern. Over the next three years, Passion and Pom killed at least three more very young infants and possibly as many as six. Other females have since been observed carrying out similar attacks. Eerily, the killers and their victims had often relaxed together with no hint of animosity, apparently unconcerned about any possibility of violence. Tiny babies are vulnerable, however. A particularly helpless infant in the arms of a competitor appears to stir something dark in the mind of a female chimpanzee. In Goodall’s words, it is as if a switch is thrown. Out of nowhere, a familiar companion turns into an enemy, without any seeming provocation. The grisly behavior does more than provide meat. The mothers whose infants Passion and Pom killed had spent much of their time in the same area as their tormentors, competing for access to the choicest fruit trees. Fear of lethal attack would likely keep their competitors away. Over many months, the attacks could be expected to lead to extra food for the killer mothers, which meant that the infanticides were selfish acts conducted at the expense of everyone except the killers’ family. They were not the novel acts of broken minds. They were adaptive behaviors to which others might be expected to respond.

The mothers of the victims tended to avoid the killing pair. Sometimes when Passion and Pom attacked, alert males intervened. Males tend to protect the weaker females. Males likewise support new female immigrants against longtime residents, apparently to encourage them not to leave the community: male policing of conflicts among females seems to be a selfish behavior. Immediate protection was the most anyone did. Too often, Passion and Pom prevailed. The suffering was widespread: tensions rose, infants died, mothers were bereaved, adult males lost offspring.

In the long run, the community was weakened by loss of numbers and reduced cooperation among mothers. If adult males had acted in concert, they could certainly have stopped Passion and Pom, because the collaborative power of males is enormous: they kill prime adult males without themselves getting a scratch. But though the males had the means to punish or kill Passion and Pom, they did not have the mind-set.

The contrast with a human community is obvious. The equivalents of Passion and Pom would never have got away with it. Only humans have community standards that decide the crucial difference between right and wrong. So the third question involves understanding not only why humans are sensitive to what is right or wrong, but also why humans punish those who do wrong, whereas chimpanzees hardly do so at all.

The three moral questions that I have outlined concern why humans are exceptionally prosocial, why we are guided as we are in deciding right from wrong, and why we care so much that we are prepared to intervene when we see wrong done. Christopher Boehm suggests that a solution to these problems lies in the claustrophobic setting of small groups in which capital punishment was a realistic threat for troublemakers.

States use coalitionary proactive aggression to round up criminals, terrorists, gangs, or rivals for power. The power of the state is social oil. Without it, a state grinds quickly into chaos of competing militias, as Libya reminded us after the 2011 death of President Qaddafi, or Yugoslavia after the 1980 death of President Tito, or the eastern Congo after the 1997 death of President Mobutu.

In Uganda’s Kibale National Park, there are few greater pleasures than to stop for a minute, close your eyes, and simply listen. At almost any time of day, you may hear the trills of warblers and insects, persistent cuckoos and tinker-birds. After dark, the calls of frogs, bats, and nightjars provide a backdrop for cicadas, bush babies, and owls. Peace seems to reign.

The sounds that soothe are mostly male. Overwhelmingly, they tell of typical male actions: showing off, defending territories, threatening neighbors, calling allies, attracting females. They speak of color, weapons, and readiness for aggression. The human listener may be relaxed, but the callers are not. Jacked on testosterone, the males are loud, rough, and pushy. The sweet harmony is a testament to the pervasiveness of reactive aggression.

Yet hardly any of those melodious species exhibit proactive aggression. Proactive aggression is so comparatively rare that at one time it seemed to be entirely absent in animals other than humans.

The percentage of infant deaths due to infanticide varied widely, and rose as high as 37% in a population of mountain gorillas, 44% in chacma baboons, 47% in blue monkeys, and a remarkable 71% in red howler monkeys. In 2014, the behavioral ecologists Dieter Lukas and Elise Huchard surveyed 260 species of mammals studied in the wild and reported that infanticide had been found in almost half of them, mainly in species where males have something to gain from the killing. It is usually a selfish reproductive strategy used by males to bring females into breeding condition as soon as possible. With regard to primates, among 89 wild species, they found infanticide in 60 (67%), including chimpanzees and gorillas.

Female primates can also kill infants. Among marmosets and tamarins, groups contain up to four females. Normally only the alpha female breeds. If a lower-ranking female gives birth, her offspring are likely to be killed by the alpha female. The killing is adaptive for the alpha female because additional infants compete for care by adults, and thereby jeopardize the survival of her own young.

In Kibale’s Ngogo community there were instances when males killed or fatally wounded 18 members of neighboring communities during a period of 10 years. The Ngogo community then expanded their territory into the area where most of the kills had occurred. In Gombe, when the territory occupied by a community increases in size, community members are better fed, breed faster, and survive better. Kill some neighbors, expand the territory, get more food, have more babies—and be safer at the same time, since there are fewer neighbors who might be able to attack you.

Yellowstone National Park in Montana and Wyoming have been studied closely. In a 12-year sample of 155 wolf corpses found inside the park, an estimated 37% were killed by other wolves. Shortage of space, rather than shortage of food, predicted aggression between packs. Likewise, in Denali National Park in Alaska, where wild wolves were monitored mostly by helicopter, about 40% of 50 adult deaths were caused by adults from other packs. How many of the kills were due to proactive aggression as opposed to reactive aggression getting out of control is unknown, but direct observations of fights between packs show that they were sometimes proactive.

In most species, the costs of attacking members of your own species are too high because you might get hurt. Only a few species happen to live in societies in which gangs of allied individuals can form, and the gangs can regularly find vulnerable loners of another group to beat up on with minimal risk of being hurt. Among mammals, these coalitions are so far known to occur only among social carnivores and primates. Coalitionary proactive aggression that kills adults in other groups of the same species is rare, but where it occurs it appears to be a natural and adaptive behavior that benefits the killers. It is known in wolves, lions, spotted hyenas, chimpanzees, white-faced capuchins, various ants, and other species.

Some people think of hunter-gatherers as so peaceful that coalitionary fighting would hardly be part of their lives at all. That idea is generally right for a specific type of hunter-gatherer—namely, those who lived alongside farmers or pastoralists (herders of mobile animals such as cattle, sheep, and goats). Classic examples are the Hadza of Tanzania and the Ju/’hoansi of southern Africa (who have also been called !Kung Bushmen, San, or Basarwa). They both live in the same area as pastoralists, with whom they also intermarry, and have done for hundreds of years. The pastoralists are militarily superior to the hunter-gatherers. Although there are historical records of warfare among those peoples, in recent years peace has reigned in both groups of foragers. If fights arose between members of the two cultures, as they sometimes did, the hunter-gatherers would be soundly defeated, as they were.

A 1940 estimate indicated there were almost 600 different linguistic groups, or societies, in Australia. Intergroup conflict was found to occur in every climatic zone throughout the continent, from the lush regions of the north and southeast to the harsh central deserts.

A common procedure in such warfare is to steal up to the enemy’s camp in the dead of night, and encircle it in the earliest dawn. With a shout, the carnage then begins. Night attacks appear to have been universal, since the aggressors know the most radical method to extinguish the enemy is to take them unawares, and to slaughter them before they can retaliate. For this purpose, it is best to either steal on them in the earliest hours of morning, or to lie in ambush at a place where the enemy is sure to call.

Intense violence within groups calls for a difficult decision, because an actor must decide whether to join a coalition against a victim who could, at a different time, be a useful ally.  Then there is the problem of planning. How is a target decided, and how does any one individual know who is going to join a coalition? There seems to be no way for chimpanzees to share in advance their intention to attack a particular individual. Chance brought an opportunity for the killers of Pimu, and even then, four were against him but two were not. Humans solve this coordination problem through language. Plotters gossip with increasing confidence that someone should be killed, yet uncertainty remains up to the moment of a joint attack.

The assassins check one another’s commitment, lest someone turn traitor. In short, complex calculation and refined communication are needed to make within-group killing a successful adaptive strategy. By contrast, in intergroup interactions the formula is simple: side with your friends against the enemy.

In the authoritarian courts of medieval European monarchs, Chinese emperors, twentieth-century fascist regimes, or Mafia families, a signal from the leader could be enough to cause the execution of a disrespectful member. The obedience of courtiers, slaves, prisoners, or unwilling soldiers shows us the consequence of hierarchical power in its rawest form. Those who attempt to challenge, escape, disobey, or desert are subject to being killed.

Hierarchical relationships among families began during the Upper Paleolithic, a few thousand years before agriculture was invented, when humans found a way to produce surplus food. Those who owned the surplus could use it to buy labor or goods, creating an advantage to producing as much food as possible.

Coalitionary proactive aggression is responsible for execution, war, massacre, slavery, hazing, ritual sacrifice, torture, lynchings, gang wars, political purges, and similar abuses of power. It permits sovereignty as a right over life, caste as a system of casual domination, and guards who make prisoners dig their own graves. It makes kings of wimps, underlies fidelity to groups, and gives us long-term tyrannies. It has battered our species since the Pleistocene. Along the evolutionary way, it brought us our great capacity for good, but it has also brought enormous harm.

It is therefore cheering to remember that in sane individuals proactive aggression is a highly selective behavior that is delicately attuned to context. Male langurs do not kill at random.  Chimpanzees reserve their proactive attacks for when they have overwhelming force. Hunter-gatherers living with pastoralists do not get into wars. Proactive aggression is not produced by individuals in a fit of rage, or in an alcoholic haze, or out of a testosterone-induced failure of cortical control. It is a considered act by an individual or coalition that takes into account the likely costs. It has a strong tendency to disappear when it does not pay.

In theory, the path to resistance is obvious. “When bad men combine,” the British parliamentarian Edmund Burke wrote in 1770, “the good must associate; else they will fall one by one.” But of course the “bad men” arrange matters to stop “the good” from “associating” in any meaningful manner.

Biological determinism is important, because ever since Darwin, and right up to today, it has cast a shadow on discussions of the past, present and future of war. A key question is whether, if our ancestors were adapted for war in the Pleistocene, we are biologically driven to conduct war today. As I will explain, my answer is that while war is not inevitable, conscious effort is needed to prevent it.

Why should the idea of war as natural mean that “there is little point in trying to prevent, reduce, or abolish it”? We do not apply that formula to other unpleasant natural things. We try to stop diseases even though they are clearly biological in nature. We try to intervene when men harass women, or when bullies throw their weight around, or when children fight one another. The fact that we think such behaviors have evolved does not inhibit us from trying to reduce their effects.

The simple war that occurs in small-scale prestate societies and is more similar to intergroup aggression in some animals. Its skirmishes are so brief, and it is so relatively unmilitary in its organization, that some anthropologists prefer not to apply the word “war” to this style of violence. It consists mainly of brief surprise attacks. Simple war is the only type of war in societies where men (ordinarily, the adult married men, or what Gellner calls “the cousins”) have egalitarian relationships, and no men work for or hold authority over others. All except the infirm are warriors, without military hierarchy. Each man makes up his own mind about whether to join an attack or stay at home.  After wounding or killing one or more victims, ideally the attackers leave so rapidly that they escape without being drawn into an escalated encounter. Battles are accordingly rare. When opposing groups of warriors find themselves confronting each other, men on both sides tend to disperse.

The raiders’ approach might be detected, the victim group might be unexpectedly well prepared, or there could be traps such as spikes embedded in the ground around the enemy village. The raiders therefore need courage, and in many cases a willingness to tolerate intense physical exertion. To help overcome their nerves, warriors often work themselves into a state of excitement before leaving home. Ritual practices may be used in prepartion for the attack. Raids might accrue benefits such as capturing women or taking heads; anticipated rewards may take the form of prestige or goods. In some societies, cowards may be punished.

Similar accounts, in which warriors perceive no benefits other than the thrill of making a kill, are rife.

Why do they kill? The unnerving answer that makes biological sense is that they enjoy it. Evolution has made the killing of strangers pleasurable, because those that liked to kill tended to receive adaptive benefits.  The rewards do not have to be anticipated consciously. All that is needed is enjoyment of the kill. Sexual reproduction works in a parallel way. A chimpanzee, or wolf, or any other animal, cannot be expected to know that an act of mating will lead to babies. Why do they mate? They enjoy it.

Where the social divide is sufficient, however, occasional outbursts of killing still seem to manifest a deep joy of killing. The historian Joanna Bourke wrote of atrocities on all sides in the Second World War. A Japanese solder remembered Nanking. “When we were bored, we had some fun killing Chinese. Buried them alive, or pushed them into a fire, or beat them to death with clubs, or killed them by other cruel means.” In Europe in the 1940s, members of the collaborationist Croatian Ustaše movement enthusiastically killed Jews, Serbs, and Gypsies, “often hacking them to death with primitive implements.” Slaughter was estimated at 40,000 Gypsies and 400,000 Serbs in Croatia and Bosnia.

There are probably very few wars in which equivalent histories could not be found.

Revenge motivations and moral pressures are only two of many unique features of our species that influence the practice of simple warfare. Others include advanced weaponry, language, social norms, docile psychology, training of warriors, and the ability to devise a shared plan. But the mere occurrence of simple warfare depends on none of those explanations. The traits that decorate human warfare beyond these elements are rococo additions, not necessary features.

We should therefore not find it surprising that small groups of men with disregard for national laws easily form gangs that take advantage of local imbalances of power to kill. Similar behaviors emerge among chimpanzees and hunter-gatherers as among freedom fighters, street gangs, or the underworld. Selection favors safe killing whenever it can be carried out at sufficiently low risk to the attacker, sometimes even when there is no obvious benefit is at stake. In an anarchic world, the satisfaction of killing an enemy can have its own rewards.

Whether the much shorter history of complex warfare also affected our evolutionary psychology is unknown, but 10,000 years is certainly enough for biological adaptation to occur.

The combination of commanders and soldiers makes complex warfare so much better organized for combat that it is described as being a truly military system. In battles of complex warfare, soldiers have no choice about whether to participate and may be thoroughly unenthusiastic about doing so. Their situation can be emotionally traumatic and often when the soldier is coerced into action, highly maladaptive for the individual. In some battles, soldiers are required to approach a body of armed opponents in an action exposing themselves deliberately to a high risk of injury or death. The question is what makes them do it.

All men on a battlefield are afraid, according to U.S. General S. L. A. Marshall’s study of battle behavior in the Second World War. The military scholar Ardant du Picq noted examples of how soldiers responded to being ordered into battle. Sometimes the whole army turns and runs. This occurs, in the words of the military historian John Keegan, “not because [the army] has been physically shaken but because its nerve has given.  At other times, “fainthearts” dribble away in small numbers until, by the time the two sides engage, hardly anyone is left. Or the attackers on both sides come to a halt before they are in range of their enemies.

Another outcome is for the opposing armies each to edge leftward as they approach each other, eventually sliding past each other without contact. The soldiers’ fear of engagement is so strong that the main task of officers is suppressing it by being ready to kill deserters.  Frederick the Great said that the common soldier must fear his officer more than he fears the enemy. The commanding officer not only must bring the soldiers to the killing zone at the front, he must make then stay there.

Clearly, the battle behavior of soldiers in complex warfare bears little relation to the relative eagerness of a hunter-gatherer warrior attacking an enemy camp. Soldiers in complex war have to be trained to reduce their fear, so as to perceive a face of battle which, if not familiar, and not friendly be totally petrifying. So although commanders tend to think that soldiers go into battle because they are obedient, whereas in reality two other reasons are more important. One is found when fighting can actually improve the chance of survival; in some circumstances, being left behind can be the worst outcome.

The other reason for bravery is to avoid incurring the contempt of close companions. Military organizations intentionally foster close bonds. Soldiers are typically organized into small groups, often around five to seven men. They have bonded with one another through training, prior action, often ritual exercises such as hazing, possibly hostility toward their officers, and endless hours of boredom. Their desire to maintain one another’s respect is sometimes suggested to have come from developing a false sense of kinship. It seems to me more likely due to a moral sense that evolved in response to the threat of execution in the Pleistocene.  Respect also comes from being hazed, a practice that emphatically shows a recruit that he is subject to the coalitionary power of his group members. A soldier exposed as a coward can be in danger from his own unit.

Human psychology is not well adapted to being a soldier. That is why the most successful armies are those that have most completely worn down the self-interested tendencies in their troops, whether through discipline or through inspiration.

Alexander led an army that conquered most of the Middle East, including the Persian Empire and kingdoms as far as western India. His substantial military actions include nine sieges, ten battles, and a major campaign. He never lost a fight. He inspired his troops by repeatedly fighting at the front and sometimes personally leading a charge.  Niall MacKay showed that the history of warfare is dominated by engagements that are similarly asymmetric, meaning that as in Alexander’s battles, the attacking side is greatly superior to its opponent. Commanders who initiate aggression do so when their forces are overwhelmingly stronger than the enemies, and as a result, until recently those who begin battles (or wars) have tended to win them.

The political scientist Ivan Arreguín-Toft tallied war victories from 1800 to 1998 according to whether one opponent was “stronger” than the other. To be judged as relatively “strong,” a side had to have a material power at least ten times greater than the other. The chance of the stronger side’s winning declined steadily from 88% before 1850 to 45% after 1950. Commanders are thus apparently no longer very good at anticipating victory. According to Johnson and MacKay, among other difficulties, the possibilities for counterinsurgency and guerrilla tactics have become too numerous.

He found four main symptoms of incompetence governing the outcome of battles: overconfidence, underestimation of the enemy, the ignoring of intelligence reports, and wastage of manpower.

Groupthink exacerbated the problem by contributing six additional symptoms: a shared illusion of invulnerability, collective attempts to maintain shaky but cherished assumptions, an unquestioned belief in the group’s inherent morality, stereotyping the enemy as too evil for negotiation (or too weak to be a threat), a collective illusion of unanimity in a majority viewpoint (based on the false assumption that silence means consent), and self-appointed censors to protect the group from information that might weaken resolve (such as reports from spies).

Evidence of the strength of the Cuban armed forces had been abundant, with no support for the claim that 30,000 would make their way through the swamps to fight alongside U.S. liberators. “How could I have been so stupid as to let them proceed?” President Kennedy repeatedly asked later.Most of his team were similarly puzzled about their assessment failures.

According to Peter Wyden the answer was clear. It was arrogance, “egos so tall that the eyes and ears can shut out whatever one prefers not to see or hear.” Kennedy, the final decider, desperately wanted to avoid being called “chicken,” had unbounded confidence in his own luck, and was surrounded by people who echoed his feelings. “Everyone around him thought he had the Midas touch and could not lose,” wrote Arthur Schlesinger. Richard Bissell, the CIA deputy director of plans who pushed for the invasion, was such an ambitious and confident risk-taker, says Wyden, that he was unable to give up on his superman task in the face of increasingly clear evidence of risk. Even after the fiasco, Bissell clung to his view that they had done the right thing. The CIA’s secret internal report on the affair, published in 1998, “painted a picture of an agency shot through with deadly self-deception.

Wherever groups compete, even without fighting, the same positively biased judgments recur. Mark Twain got it right: “Nations do not think, they only feel…each nation knowing that it has the only true religion and the only sane system of government, each despising all the others, each an ass and not suspecting it, each proud of its presumed supremacy, each perfectly sure it is the pet of God, each with undoubting confidence summoning Him to take command in time of war, each surprised when He goes over to the enemy, but by habit able to excuse it and resume compliments—in a word, the whole human race content, always content, persistently content, indestructibly content, happy, thankful, proud, no matter what its religion is.”

First, the focus afforded by self-belief allows total commitment to winning.  In an even match, a rational opponent would see himself as having a 50% probability of defeat and think about protecting himself in the event of failure, whether by devising escape plans, avoiding damage, or attempting to reassess the opponents’ strength. Such attention to the possibility of loss would lead to anxiety (a sure predictor of defeat) and, more generally, to distraction. So, because in a contest between equals 100% effort beats 90% effort, arrogant blind confidence will predict the winner. “Championship thinking” is irrational and wasteful and half the time deluded, but in an even match, it brings more psychological resources to the fight and increases the chance of winning.  Self-deception is common in battles. In the eastern Congo in the 1990s, the Mai-Mai fighters believed that bullets hitting them would turn to water. In Uganda in the 1980s, Alice Lakwena’s rebel groups were ferocious fighters because they thought themselves safe from bullets. A belief that one is magically protected from harm is a wonderful illusion for inspiring unrestrained aggression against the enemy.

The second benefit of self-confidence is that it can create fear in the enemy; often a good bluff is sufficient. The Dayaks inspired fear in various ways. Tales of their cannibalism and parading of severed heads meant that “they were possessed, they weren’t acting normally,” and their enemies were duly scared.

In the grand scheme of life, it is an irony of nature that selection for the ability to win brings with it a failure of assessment, and resulting “wastage of manpower.”

The more challenging difficulty for our species is that, as resource distributions change, new coalitions are expected to form repeatedly to challenge existing sovereignties. All human societies are composed of competing subgroups. Some subgroups will ignore existing laws by attempting to carve their own territories out of prior nations, as ISIS did in Iraq in 2014. Handling such efforts nonviolently will surely always be a challenge. The global response to ISIS illustrates the intensity of violence that can readily emerge when a new ideology defies existing mores.

Our species swings between the desire for peace and the temptations of power, and faces the contradiction that although the risk of dying from violence has fallen, the risk of nuclear holocaust has risen. The great merit of proactive aggression, from the perspective of a nonviolent philosophy, is that a well-adapted animal does not attack if it expects to get hurt.

The idea that warfare has evolved, and that even today it is facilitated by adaptive features of our psychology, does not make it inevitable. It does mean, however, that we are a dangerous species. In the face of our tendency for positive illusions about the merits of war, we will always need strong institutions and alert engagement to temper the rise of militaristic philosophies, the spread of excessively optimistic pacificism, and the abuse of power.

In the human lineage, reactive aggression became suppressed, while proactive aggression stayed high.

Language-based conspiracy was the key, because it gave whispering beta males the power to join forces to kill alpha-male bullies. As happens in small-scale societies today, language allowed underdogs to agree on a plan, and thereby to make predictably safe murders out of confrontations that could otherwise have been dangerous. Genetic selection against the propensity for reactive aggression was an unforeseen result of eliminating the would-be despots. The selection against alpha personalities led to males becoming, for the first time, egalitarian.

There is no reason to regard our domestication as complete. How much more domesticated we could become if we were tamed for another 12,000 generations, say, is an open question. Given sufficient sanctions against reactive aggressors, in another 300,000 years humans could in theory become as hard to rile as lop-eared rabbits at a petting farm, which remain gentle even when stroked repeatedly by dozens of eager children. Equally, however, if would-be despots were to escape sanctions, the process could go into reverse.

Group selection is commonly invoked to explain our species interest in nonrelatives and our occasional willingness to sacrifice our own interests on behalf of a larger good. Group selection theory, however, has never quite been able to explain how benefits at a group level override those of individuals. The theory that the moral senses evolved to protect individuals from the socially powerful suggests that group selection may be entirely unnecessary for explaining why we are such a group-oriented species. Our deference to the coalitionary powers within our own groups leads to a reduced intensity of competition, enabling groups to thrive.

Human hunting of prey seems likely to have been associated with the ability to kill rivals in neighboring groups for two million years. Much as in chimpanzees and wolves seeking opportunities to attack strangers, once our ancestors had achieved the ability to kill safely a motivation to kill would probably have been present too. There seems no reason to excuse our ancestors from the links between hunting and violence found in other mammals.

Consider the relationship between men and women. In small-scale societies, as I discussed, egalitarianism is primarily a description of relationships among men, particularly married men. Just as happens in every society throughout the world, in the public sphere men dominate women. In the public domain, however, where coercive alliances regulate societal rules, conflicts between the interests of men and women consistently end in men’s favor. Patriarchy in this sense is currently a human universal.

Miscellaneous

In 1958, the playwright and songwriter Noël Coward captured the strangeness of our duality. He had lived through the Second World War, and the bad side of human nature was fully obvious to him. “It is hard to imagine,” he wrote, “considering the inherent silliness, cruelty and superstition of the human race, how it has contrived to last as long as it has. The witch-hunting, the torturing, the gullibility, the massacres, the intolerance, the wild futility of human behaviour over the centuries is hardly credible.

Large body size in humans is a risk factor for antisocial personality disorder. Boys who are big for their age learn as early as three years old that they can win in fights with smaller peers. Being rewarded for successful aggression, they end up being more aggressive throughout childhood.

Hitler, who ordered the execution of some eight million people and was responsible for the deaths of many millions more, was said by his secretary Traudl Junge to have had an agreeable, friendly, and paternal manner. He hated cruelty to animals: he was a vegetarian, and adored his dog Blondi.

Whether the amygdala is reduced in size in domesticated animals has not been much studied, but in one mammal (the rabbit) and one bird (the Bengalese finch), the expected reduction in amygdala size has been found. Behind the telencephalon lies an area called the diencephalon, where the hypothalamus develops. Like the amygdala, the hypothalamus is a core part of the neural network that underlies reactive aggression (as well as proactive aggression). The hypothalamus also strongly influences the activity of the adrenal glands, and is involved in the regulation of females’ estrous cycles and reproductive behavior. Thus, in the brain as in the rest of the body, the traits of the domestication syndrome are likely to be affected by changes to the activity of neural-crest cells. Neural-crest cells have clear input into systems regulating stress, fear, and aggression, including the sympathetic nervous system and a set of brain structures regulating emotional responses. As we saw, neural crest development was first associated with the white patches typical of domesticated horses, dogs, cows, and other animals. We followed up our hypothesis that neural-crest cells are involved in many of the features of domesticates, finding support for links to such superficially unrelated traits as smaller jaws, smaller teeth, floppy ears, and even brain changes related to decreased reactive aggression. A further critical test of our hypothesis is whether domesticated species show changes in genes that affect neural-crest migration. Since 2014, such effects have been found in six species (horses, rats, dogs, cats, silver foxes, and mink). Whether changes to neural-crest migration will be found in every domesticated species is uncertain, but at this early stage no exceptions have been found. Key to this growing understanding of the domestication syndrome, simply put, is juvenilization.

Homosexuality

If homosexual behavior is not adaptive, perhaps it evolved in humans as a paedomorphic by-product of selection against reactive aggression. Homosexual behavior is particularly prominent among bigger-brained species in which sexual behavior is emancipated from hormonal control. In primates, it occurs among apes and monkeys, but not in the smaller-brained lemurs and lorises. It is found among many whales and dolphins. Male gray whales romp together in an obvious state of sexual excitement. Male river dolphins use one another’s blowholes for sex. Studies have tended to find that a predominant or exclusive orientation toward the same sex occurs at lower rates, in around 1 to 2% of women and 2 to 5% of men.  A couple of possible explanations for how homosexuality might be adaptive for humans have been explored. One hypothesis is that same-sex relationships confer advantages in social competition. Homosexual men might be more intensely supportive of each other, for example, like female bonobos and Spartan warriors. Although such social bonds could presumably indeed be beneficial, the reproductive benefits seem too low to explain the occurrence of exclusive homosexuality. Among 600,000 domestic pairs of same-sex individuals in 2000 in the United States, for example, 34% of female couples and 22% of male couples were raising children, compared to 39% of 16 million opposite-sex couples.

If homosexuals tend to have few children of their own, their sexual orientation could in theory be adaptive if it leads them to give exceptional help to their genetic kin. In some cultures, such as Samoa, homosexual men do show a stronger-than-usual interest in helping their brothers and sisters. But even in Samoa, the kin effect is too weak to explain the evolution of the homosexual tendency. In Japan, there is no evidence that homosexuals show more interest in kin than heterosexuals do. The numbers do not add up.  The small family sizes of exclusive homosexuals, together with the lack of evidence that they provide large benefits to kin, suggest that homosexual behavior in humans is not biologically adaptive. This leaves the fascinating question of why homosexual attraction is as prevalent and persistent as it is in our species.

The evidence that exclusive homosexual preference is common but not adaptive makes it a prime candidate for being an evolutionary by-product. An association with selection against reactive aggression is indicated on several grounds. First, the only nonhuman animal in which exclusive homosexual preference is known is a domesticated species—namely, sheep. Male lambs that are reared in the exclusive company of other males fall into two groups. The division depends on how, once they become adult, they respond to an ewe who is introduced to them when ready to mate. One group is the heterosexuals. On meeting an estrous ewe, these rams experience a rise in testosterone and are fully sexually interested in her. The other group are the homosexuals, who interact with the estrous ewes without showing any hormonal increase or sexual interest. Females are just not their thing. About 8% of rams brought up in single-sex groups adopt this homosexual orientation.  In wild sheep, dominant rams have been seen to mount subordinate rams in apparent displays of dominance, but the behavior is rare (4% of all social interactions in all-male groups). The obvious implication is that homosexual preference is an incidental consequence of domestication.  Domestic rams who experience relatively low levels of testosterone before they are born are more likely to become homosexual.

The INAH3 is larger in heterosexual men than in women, and has been found to be intermediate-sized in homosexual men. As with rams, evidence in humans suggests that variation in exposure to testosterone before birth might influence homosexual tendencies.  Homosexual men also have somewhat feminized face shapes and shorter, lighter bodies than heterosexual men, most likely from relatively low exposure to testosterone in the womb. These conclusions are still somewhat tentative.

The elaboration of homosexual behavior in bonobos, which occurs in both sexes, clearly fits the hypothesis that selection against reactive aggression favors homosexual behavior. A potential explanation is again that, in bonobos, prenatal exposure to testosterone is relatively low. The ring finger is longer relative to the second finger in males than in females in both bonobos and chimpanzees, suggesting that, as in humans, relative ring-finger length provides an index of prenatal testosterone exposure. As expected from the domestication hypothesis, relative ring-finger length is shorter in bonobos than in chimpanzees. Interestingly, relative ring-finger length in humans is more similar to the ratio in bonobos than in chimpanzees.

The Neanderthals had significantly longer ring fingers (relative to second fingers) than modern people while the finger-length ratio of the 100,000-year-old Homo sapiens at Qafzeh is in-between the ratios for living humans and the five Neanderthals. These fascinating hints suggest that living Homo sapiens have indeed been subject to lower levels of prenatal testosterone than Neanderthals, in keeping with the self-domestication hypothesis.

The question of why selection has not eliminated homosexuality, if it is a by-product of self-domestication without biological advantages, is also fascinating. A possible answer is that self-domestication has been continuing too recently, up to historical times, for selection to operate strongly against incidental consequences.