Why should organisms ever evolve to seek to harm other organisms? The answer is not as straightforward as the phrase "survival of the fittest" would suggest. In his book The Selfish Gene, which explained the modern synthesis of evolutionary biology with genetics and game theory, Richard Dawkins tried to pull his readers out of their unreflective familiarity with the living world. He asked them to imagine animals as "survival machines" designed by their genes (the only entities that are faithfully propagated over the course of evolution), and then to consider how those survival machines would evolve.

Darwin gave us a theory of why living things have the traits they have, not just their bodily traits but the basic mindsets and motives that drive their behavior. A hundred and fifty years after the Origin of Species was published, the theory of natural selection has been amply verified in the lab and field, and has been augmented with ideas from new fields of science and mathematics to yield a coherent understanding of the living world. These fields include genetics, which explains the replicators that make natural selection possible, and game theory, which illuminates the fates of goal-seeking agents in a world that contains other goal-seeking agents.

Anyone who has ever seen a hawk tear apart a starling, a swarm of biting insects torment a horse, or the AIDS virus slowly kill a man has firsthand acquaintance with the ways that survival machines callously exploit other survival machines. In much of the living world, violence is simply the default, something that needs no further explanation. When the victims are members of other species, we call the aggressors predators or parasites. But the victims can also be members of the same species. Infanticide, siblicide, cannibalism, rape, and lethal combat have been documented in many kinds of animals.

To a survival machine, another survival machine (which is not its own child or another close relative) is part of its environment, like a rock or a river or a lump of food. It is something that gets in the way, or something that can be exploited. It differs from a rock or a river in one important respect: it is inclined to hit back. This is because it too is a machine that holds its immortal genes in trust for the future, and it too will stop at nothing to preserve them. Natural selection favors genes that control their survival machines in such a way that they make the best use of their environment. This includes making the best use of other survival machines, both of the same and of different species.

Dawkins's carefully worded passage also explains why nature does not consist of one big bloody melee. For one thing, animals are less inclined to harm their close relatives, because any gene that would nudge an animal to harm a relative would have a good chance of harming a copy of itself sitting inside that relative, and natural selection would tend to weed it out. More important, Dawkins points out that another organism differs from a rock or a river because it is inclined to hit back. Any organism that has evolved to be violent is a member of a species whose other members, on average, have evolved to be just as violent. If you attack one of your own kind, your adversary may be as strong and pugnacious as you are, and armed with the same weapons and defenses. The likelihood that, in attacking a member of your own species, you will get hurt is a powerful selection pressure that disfavors indiscriminate pouncing or lashing out. It also rules out the hydraulic metaphor and most folk theories of violence, such as a thirst for blood, a death wish, a killer instinct, and other destructive itches, urges, and impulses. When a tendency toward violence evolves, it is always strategic. Organisms are selected to deploy violence only in circumstances where the expected benefits outweigh the expected costs. That discernment is especially true of intelligent species, whose large brains make them sensitive to the expected benefits and costs in a particular situation, rather than just to the odds averaged over evolutionary time.

Hobbes considered competition to be an unavoidable consequence of agents' pursuing their interests. Today we see that it is built into the evolutionary process. Survival machines that can elbow their competitors away from finite resources like food, water, and desirable territory will out-reproduce those competitors, leaving the world with the survival machines that are best suited for such competition.

The logic of violence as it applies to members of an intelligent species facing other members of that species brings us to Hobbes. In a remarkable passage in Leviathan (1651), he used fewer than a hundred words to lay out an analysis of the incentives for violence that is as good as any today:

So that in the nature of man, we find three principal causes of quarrel. First, competition; secondly, diffidence; thirdly, glory. The first maketh men invade for gain; the second, for safety; and the third, for reputation. The first use violence, to make themselves masters of other men's persons, wives, children, and cattle; the second, to defend them; the third, for trifles, as a word, a smile, a different opinion, and any other sign of undervalue, either direct in their persons or by reflection in their kindred, their friends, their nation, their profession, or their name.

The second cause of quarrel is diffidence, a word that in Hobbes's time meant "fear" rather than "shyness." The second cause is a consequence of the first: competition breeds fear. If you have reason to suspect that your neighbor is inclined to eliminate you from the competition by, say, killing you, then you will be inclined to protect yourself by eliminating him first in a preemptive strike. You might have this temptation even if you otherwise wouldn't hurt a fly, as long as you are not willing to lie down and be killed. The tragedy is that your competitor has every reason to crank through the same calculation, even if he is the kind of person who wouldn't hurt a fly. In fact, even if he knew that you started out with no aggressive designs on him, he might legitimately worry that you are tempted to neutralize him out of fear that he will neutralize you first, which gives you an incentive to neutralize him before that, ad infinitum. The political scientist Thomas Schelling offers the analogy of an armed homeowner who surprises an armed burglar, each being tempted to shoot the other to avoid being shot first. This paradox is sometimes called the Hobbesian trap or, in the arena of international relations, the security dilemma.

We also know today why "wives" would be one of the resources over which men should compete. In most animal species, the female makes a greater investment in offspring than the male. This is especially true of mammals, where the mother gestates her offspring inside her body and nurses them after they are born. A male can multiply the number of his offspring by mating with several females -- which will leave other males childless -- while a female cannot multiply the number of her offspring by mating with several males. This makes female reproductive capacity a scarce resource over which the males of many species, including humans, compete. None of this, by the way, implies that men are robots controlled by their genes, that they may be morally excused for raping or fighting, that women are passive sexual prizes, that people try to have as many babies as possible, or that people are impervious to influences from their culture, to take some of the common misunderstandings of the theory of sexual selection.

How can intelligent agents extricate themselves from a Hobbesian trap? The most obvious way is through a policy of deterrence: Don't strike first; be strong enough to survive a first strike; and retaliate against any aggressor in kind. A credible deterrence policy can remove a competitor's incentive to invade for gain, since the cost imposed on him by retaliation would cancel out the anticipated spoils. And it removes his incentive to invade from fear, because of your commitment not to strike first and, more importantly, because of your reduced incentive to strike first, since deterrence reduces the need for preemption. The key to the deterrence policy, though, is the credibility of the threat that you will retaliate. If your adversary thinks that you're vulnerable to being wiped out in a first strike, he has no reason to fear retaliation. And if he thinks that once attacked you may rationally hold back from retaliation, because at that point it's too late to do any good, he might exploit that rationality and attack you with impunity. Only if you are committed to disprove any suspicion of weakness, to avenge all trespasses and settle all scores, will your policy of deterrence be credible. Thus we have an explanation of the incentive to invade for trifles: a word, a smile, and any other sign of undervalue. Hobbes called it "glory"; more commonly it is called "honor"; the most accurate descriptor is "credibility."

The policy of deterrence is also known as the balance of terror and, during the Cold War, was called mutual assured destruction (MAD). Whatever peace a policy of deterrence may promise is fragile, because deterrence reduces violence only by a threat of violence. Each side must react to any nonviolent sign of disrespect with a violent demonstration of mettle, whereupon one act of violence can lead to another in an endless cycle of retaliation. As we shall see in chapter 8, a major design feature in human nature, self-serving biases, can make each side believe that its own violence was an act of justified retaliation while the other's was an act of unprovoked aggression.

Hobbes's analysis pertains to life in a state of anarchy. The title of his masterwork identified a way to escape it: the Leviathan, a monarchy or other government authority that embodies the will of the people and has a monopoly on the use of force. By inflicting penalties on aggressors, the Leviathan can eliminate their incentive for aggression, in turn defusing general anxieties about preemptive attack and obviating everyone's need to maintain a hair trigger for retaliation to prove their resolve. And because the Leviathan is a disinterested third party, it is not biased by the chauvinism that makes each side think its opponent has a heart of darkness while it is as pure as the driven snow.

The logic of the Leviathan can be summed up in a triangle (figure 2-1). In every act of violence, there are three interested parties: the aggressor, the victim, and a bystander. Each has a motive for violence: the aggressor to prey upon the victim, the victim to retaliate, the bystander to minimize collateral damage from their fight. Violence between the combatants may be called war; violence by the bystander against the combatants may be called law. The Leviathan theory, in a nutshell, is that law is better than war. Hobbes's theory makes a testable prediction about the history of violence. The Leviathan made its first appearance in a late act in the human pageant. Archaeologists tell us that humans lived in a state of anarchy until the emergence of civilization some five thousand years ago, when sedentary farmers first coalesced into cities and states and developed the first governments. If Hobbes's theory is right, this transition should also have ushered in the first major historical decline in violence. Before the advent of civilization, when men lived without "a common power to keep them all in awe," their lives should have been nastier, more brutish, and shorter than when peace was imposed on them by armed authorities, a development I will call the Pacification Process. Hobbes claimed that "savage people in many places in America" lived in a state of violent anarchy, but he gave no specifics as to whom he had in mind.

In this data vacuum, anyone could have a go at speculating about primitive people, and it did not take long for a contrary theory to turn up. Hobbes's opposite number was the Swiss-born philosopher Jean-Jacques Rousseau (1712-78), who opined that "nothing can be more gentle than [man] in his primitive state… The example of the savages… seems to confirm that mankind was formed ever to remain in it,… and that all ulterior improvements have been so many steps… towards the decrepitness of the species."

Though the philosophies of Hobbes and Rousseau were far more sophisticated than "nasty brutish and short" versus "the noble savage," their competing stereotypes of life in a state of nature fueled a controversy that remains with us today. In The Blank Slate, I discussed how the issue has accumulated a heavy burden of emotional, moral, and political baggage. In the second half of the 20th century, Rousseau's romantic theory became the politically correct doctrine of human nature, both in reaction to earlier, racist doctrines about "primitive" people and out of a conviction that it was a more uplifting view of the human condition. Many anthropologists believe that if Hobbes was right, war would be inevitable or even desirable; therefore anyone who favors peace must insist that Hobbes was wrong. These "anthropologists of peace" (who in fact are rather aggressive academics -- the ethologist Johan van der Dennen calls them the Peace and Harmony Mafia) have maintained that humans and other animals are strongly inhibited from killing their own kind, that war is a recent invention, and that fighting among native peoples was ritualistic and harmless until they encountered European colonists.

FIGURE 2-1: The violence triangle

As I mentioned in the preface, I think the idea that biological theories of violence are fatalistic and romantic theories optimistic gets everything backwards, but that isn't the point of this chapter. When it came to violence in pre-state peoples, Hobbes and Rousseau were talking through their hats: neither knew a thing about life before civilization. Today we can do better. This chapter reviews the facts about violence in the earliest stages of the human career. The story begins before we were human, and we will look at aggression in our primate cousins to see what it reveals about the emergence of violence in our evolutionary lineage. When we reach our own species, I will zero in on the contrast between foraging bands and tribes who live in a state of anarchy and peoples who live in settled states with some form of governance. We will also look at how foragers fight and what they fight over. This leads to the pivotal question: Is the warring of anarchic tribes more or less destructive than that of people living in settled states? The answer requires a switch from narratives to numbers: the per capita rates of violent death, to the best we can estimate them, in societies that live under a Leviathan and in those that live in anarchy. Finally we will take a look at the upsides and downsides of civilized life.