REVENGE
The determination to hurt someone who has hurt you has long been exalted in purple prose. The Hebrew Bible is obsessed with revenge, giving us pithy expressions like “Whoso sheddeth blood will have his blood shed,” “An eye for an eye,” and “Vengeance is mine.” Homer’s Achilles describes it as sweeter than flowing honey welling up like smoke from the breasts of men. Shylock cites it as the climax in his listing of human universals, and when asked what he will do with his pound of flesh, replies, “To bait fish withal: if it will feed nothing else, it will feed my revenge.”
People in other cultures also wax poetic about the settling of scores. Milovan Djilas, born into a feuding clan of Montenegrins and later a vice president of communist Yugoslavia, called vengeance “the glow in our eyes, the flame in our cheeks, the pounding in our temples, the word that turned to stone in our throats on our hearing that our blood had been shed.”161 A New Guinean man, upon hearing that the killer of his uncle had been paralyzed by an arrow, said, “I feel as if I am developing wings, I feel as if I am about to fly off, and I am very happy.”162 The Apache chief Geronimo, savoring his massacre of four Mexican army companies, wrote:Still covered with the blood of my enemies, still holding my conquering weapon, still hot with the joy of battle, victory, and vengeance, I was surrounded by the Apache braves and made war chief of all the Apaches. Then I gave the orders for scalping the slain.
I could not call back my loved ones, I could not bring back the dead Apaches, but I could rejoice in this revenge.
Daly and Wilson comment: “Rejoice? Geronimo wrote these words in a prison cell, his Apache nation broken and nearly extinct. The urge for vengeance seems so futile: There’s no use crying over spilt milk, and spilt blood is equally irrevocable.”163
Yet for all its futility, the urge for vengeance is a major cause of violence. Blood revenge is explicitly endorsed in 95 percent of the world’s cultures, and wherever tribal warfare is found, it is one of the major motives.164 Revenge is the motive of 10 to 20 percent of homicides worldwide and a large percentage of school shootings and private bombings.165 When directed against groups rather than individuals, it is a major motive of urban riots, terrorist attacks, retaliation against terrorist attacks, and wars.166 Historians who examine the decisions that led to war in reprisal for an attack note that it is often befogged in a red mist of anger.167 After Pearl Harbor, for example, the American people were said to react “with a mind-staggering mixture of surprise, awe, mystification, grief, humiliation, and above all, cataclysmic fury.”168 No alternative to war (such as containment or harassment) was ever considered; the very thought would have been tantamount to treason. The reactions to the 9/11 attack were similar: the U.S. invasion of Afghanistan the following month was motivated as much by a sense that something had to be done in reprisal as by a strategic decision that it was the most effective long-term measure against terrorism.169 The three thousand killings on September 11 had themselves been motivated by revenge, as Osama bin Laden explained in his “Letter to America”:Allah, the almighty, legislated the permission and the option to take revenge. Thus, if we are attacked, then we have the right to attack back. Whoever has destroyed our villages and towns, then we have the right to destroy their villages and towns. Whoever has stolen our wealth, then we have the right to destroy their economy. And whoever has killed our civilians, then we have the right to kill theirs.170
Revenge is not confined to political and tribal hotheads but is an easily pushed button in everyone’s brains. The homicidal fantasies to which a large majority of university students confess are almost entirely revenge fantasies.171 And in laboratory studies, students can easily be induced to avenge a humiliation. The students write an essay and are provided with an insulting evaluation written by a fellow student (who is a confederate of the experimenters or entirely fictitious). At that point, Allah smiles: the student is asked to participate in a study that just happens to give him the opportunity to punish his critic by shocking him, blasting him with an air horn, or (in more recent experiments, vetted by violence-averse human subjects committees) forcing him to drink hot sauce in a bogus experiment on taste. It works like a charm.172
Revenge is, quite literally, an urge. In one of these experiments, just as a participant was about to deliver the retaliatory shock, the apparatus broke down (thanks to a bit of subterfuge by the experimenter), so he or she was unable to consummate the revenge. All participants then took part in a bogus wine-tasting study. The ones who had never gotten the chance to shock their insulters sampled a lot more of the wine, as if to drown their sorrows.173
The neurobiology of revenge begins with the Rage circuit in the midbrain-hypothalamus-amygdala pathway, which inclines an animal who has been hurt or frustrated to lash out at the nearest likely perpetrator.174 In humans the system is fed by information originating from anywhere in the brain, including the temporoparietal junction, which indicates whether the harm was intended or accidental. The Rage circuit then activates the insular cortex, which gives rise to sensations of pain, disgust, and anger. (Recall that the insula lights up when people feel they have been shortchanged by another person.)175 None of this is enjoyable, and we know that animals will work to turn off electrical stimulation of the Rage system.
But then the brain can slip into a different mode of information processing. Proverbs like “Revenge is sweet,” “Don’t get mad; get even,” and “Revenge is a dish best served cold” are hypotheses in affective neuroscience. They predict that patterns of activity in the brain can shift from an aversive anger to a cool and pleasurable seeking, the kind that guides the pursuit of delectable food. And as so often happens, the folk neuroscience is correct. Dominique de Quervain and his collaborators gave a sample of men the opportunity to entrust a sum of money to another participant who would invest it for a profit, and then either share the total with the investor or keep it for himself.176 (The scenario is sometimes called a Trust game.) Participants who had been cheated out of their money were then given the chance to levy a punitive fine on the faithless trustee, though sometimes they would have to pay for the privilege. As they were pondering the opportunity, their brains were scanned, and the scientists found that a part of the striatum (the core of the Seeking system) lit up—the same region that lights up when a person craves nicotine, cocaine, or chocolate. Revenge is sweet, indeed. The more a person’s striatum lit up, the more he was willing to pay to punish the crooked trustee, which shows that the activation reflected a genuine desire, something that the person would pay to have consummated. When the participant did choose to pay, his orbital and ventromedial frontal cortex lit up—the part of the brain that weighs the pleasure and pain of different courses of action, in this case presumably the cost of the revenge and the satisfaction it afforded.
Revenge requires the disabling of empathy, and that too can be seen in the brain. Tania Singer and her collaborators ran a similar experiment in which men and women had their trust rewarded or betrayed by a fellow participant.177 Then they either experienced a mild shock to their fingers, watched a trustworthy partner get shocked, or watched their double-crosser get shocked. When a trustworthy partner got shocked, the participants literally felt their pain: the same part of the insula that lit up when they were shocked lit up when they saw the nice guy (or gal) get shocked. When the double-crosser got shocked, the women could not turn off their empathy: their insula still lit up in sympathy. But the men hardened their hearts: their own insula stayed dark, while their striatum and orbital cortex lit up, a sign of a goal sought and consummated. Indeed, those circuits lit up in proportion to the men’s stated desire for revenge. The results are in line with the claim by difference feminists such as Carol Gilligan that men are more inclined toward retributive justice and women more toward mercy.178 The authors of the study, though, caution that the women may have recoiled from the physical nature of the punishment and might have been just as retributive if it had taken the form of a fine, criticism, or ostracism.179
r />
There is no gainsaying the cool, sweet pleasure of revenge. A villain getting his comeuppance is a recurring archetype in fiction, and it’s not just Dirty Harry Callahan whose day is made when a bad guy is brought to violent justice. One of my most enjoyable moments as a moviegoer was a scene in Peter Weir’s award- winning Witness. Harrison Ford plays an undercover detective who is assigned to live with an Amish family in rural Pennsylvania. One day, in full Amish drag, he accompanies them to town in their horse-drawn buggy, and they are stopped and harassed by some rural punks. True to their pacifism, the family turns the other cheek, even as one of the punks taunts and bullies their dignified father. The straw-hatted Ford does a slow burn, turns toward the punk, and to the astonishment of the gang and the delight of the arthouse audience, coldcocks him a good one.
What is this madness called revenge? Though our psychotherapeutic culture portrays vengeance as a disease and forgiveness as the cure, the drive for revenge has a thoroughly intelligible function: deterrence.180 As Daly and Wilson explain, “Effective deterrence is a matter of convincing our rivals that any attempt to advance their interests at our expense will lead to such severe penalties that the competitive gambit will end up a net loss which should never have been undertaken.”181 The necessity of vengeful punishment as a deterrent is not a just-so story but has been demonstrated repeatedly in mathematical and computer models of the evolution of cooperation.182
Some forms of cooperation are easy to explain: two people are related, or married, or are teammates or bosom buddies with the same interests, so what is good for one is good for the other, and a kind of symbiotic cooperation comes naturally. Harder to explain is cooperation when people’s interests at least partly diverge, and each may be tempted to exploit the other’s willingness to cooperate. The simplest way to model this quandary is a positive-sum game called the Prisoner’s Dilemma. Imagine a Law and Order episode in which two partners in crime are held in separate jail cells and the evidence against them is marginal, so each is offered a deal by the assistant district attorney. If he testifies against his partner (“defects” against him) while the partner stays true (“cooperates”—with the partner, that is), he will go free while his partner is sent away for ten years. If each of them defects and testifies against the other, they will both go to jail, but their sentences will be reduced to six years. If each stays loyal to the other, the prosecutor can only convict them of a lesser crime, and they will be free in six months. Figure 8–5 shows the payoff matrix for their dilemma; the choices and payoffs for the first prisoner (Lefty) are printed in black; those for his partner (Brutus) are printed in gray.
FIGURE 8–5. The Prisoner’s Dilemma
Their tragedy is that both ought to cooperate and settle for the reward of a six-month sentence, which gives the game its positive sum. But each will defect, figuring that he’s better off either way: if the partner cooperates, he goes free; if the partner defects, he only gets six years rather than the ten he would get if he had cooperated. So he defects, and his partner, following the same reasoning, also defects, and they end up serving six years rather than the six months that they could have served if they had only acted altruistically rather than selfishly.
The Prisoner’s Dilemma has been called one of the great ideas of the 20th century, because it distills the tragedy of social life into such a succinct formula.183 The dilemma arises in any situation in which the best individual payoff is to defect while the partner cooperates, the worst individual payoff is to cooperate while the other defects, the highest total payoff is when both cooperate, and the lowest total payoff is when both defect. Many of life’s predicaments have this structure, not least predatory violence, where being the aggressor against a pacifist provides all the benefits of exploitation, but being an aggressor against a fellow aggressor bloodies the two of you, so you should both be pacifists, and you would be but for the fear that the other will be an aggressor. We have come across related tragedies, such as the War of Attrition, the Public Goods game, and the Trust game, wherein individual selfishness is tempting but mutual selfishness is ruinous.
Though a one-shot Prisoner’s Dilemma is tragic, an Iterated Prisoner’s Dilemma, in which players interact repeatedly and accumulate a payoff over many rounds, is truer to life. It can even be a good model of the evolution of cooperation, if the payoffs are doled out, not in years of jail time averted or in dollars and cents, but in number of descendants. Virtual organisms play rounds of the Prisoner’s Dilemma, which may be interpreted as opportunities to help each other, say, via mutual grooming, or to refrain from helping; the benefits in health and costs in time translate into the number of surviving offspring. Repeated runs of the game are like generations of organisms evolving by natural selection, and an observer can ask which of several competing strategies will eventually take over the population with its descendants. The combinatorial possibilities are too numerous to allow for a mathematical proof, but strategies can be written into little computer apps that compete in roundrobin tournaments, and the theorists can see how they fare in the virtual evolutionary struggle.
In the first such tournament, hosted by the political scientist Robert Axelrod, the winner was a simple strategy of Tit for Tat: cooperate on the first move, then continue to cooperate if your partner cooperates, but defect if he defects.184 Since cooperation is rewarded and defection punished, defectors will switch to cooperation, and in the long run everyone wins. The idea is identical to Robert Trivers’s theory of the evolution of reciprocal altruism, which he had proposed a few years earlier without the mathematical paraphernalia.185 The positive-sum reward arises from gains in exchange (each can confer a large benefit to the other at a small cost to itself), and the temptation is to exploit the other, taking the benefit without paying the cost. Trivers’s theory that the moral emotions are adaptations to cooperation can be translated directly into the Tit for Tat algorithm. Sympathy is cooperating on the first move. Gratitude is cooperating with a cooperator. And anger is defecting against a defector—in other words, punishing in revenge. The punishment can consist in refusing to help, but it can also consist in causing a harm. Vengeance is no disease: it is necessary for cooperation, preventing a nice guy from being exploited.
Hundreds of Iterated Prisoner’s Dilemma tournaments have been studied since then, and a few new lessons have emerged.186 One is that Tit for Tat, simple though it is, can be dissected into features that account for its success and can be recombined into other strategies. These features have been named for personality traits, and the labels may be more than mnemonics; the dynamics of cooperation may explain why we evolved those traits. The first feature behind Tit for Tat’s success is that it is nice: it cooperates on the first move, thereby tapping opportunities for mutually beneficial cooperation, and it does not defect unless defected upon. The second is that it is clear: if a strategy’s rules of engagement are so complicated that the other players cannot discern how it is reacting to what they do, then its moves are effectively arbitrary, and if they are arbitrary, the best response is the strategy Always Defect. Tit for Tat is easy for other strategies to cotton on to, and they can adjust their choices in response. Third, Tit for Tat is retaliatory: it responds to defection with defection, the simplest form of revenge. And it is forgiving: it leaves the gates of repentance open, so if its adversary switches to cooperation after a history of defection, Tit for Tat immediately cooperates in return.187
The last feature, forgivingness, turns out to be more important than everyone first appreciated. A weakness of Tit for Tat is that it is vulnerable to error and misunderstanding. Suppose one of the players intends to cooperate but defects by mistake. Or suppose it misperceives another player’s cooperation as defection, and defects in retaliation. Then its opponent will defect in retaliation, which forces it to retaliate in turn, and so on, dooming the players to an endless cycle of defection—the software equivalent of a feud. In a noisy world in which misunderstanding and error are possible, Tit for Tat is bes
ted by an even more forgiving strategy called Generous Tit for Tat. Every once in a while Generous Tit for Tat will randomly grant forgiveness to a defector and resume cooperating. The act of unconditional forgiveness can flick a duo that has been trapped in a cycle of mutual defection back onto the path of cooperation.
A problem for overly forgiving strategies, though, is that they can be undone if the population contains a few psychopaths who play Always Defect and a few suckers who play Always Cooperate. The psychopaths proliferate by preying on the suckers, and then become numerous enough to exploit everyone else. One successful contender in such a world is Contrite Tit for Tat, which is more discriminating in its forgiveness. It remembers its own behavior, and if a round of mutual defection had been its fault because of a random error or misunderstanding, it allows its opponent one free defection and then switches to cooperation. But if the defection had been triggered by its opponent, it shows no mercy and retaliates. If the opponent is a Contrite Tit for Tatter as well, then it will excuse the justified retaliation, and the pair will settle back into cooperation. So not just vengeance, but also forgiveness and contrition, are necessary for social organisms to reap the benefits of cooperation.