Science

Babies Want Revenge

… As long as someone else levies the sentence. The science of right and wrong in infancy.

Babies Playing
Silently plotting revenge

Photo by Fuse/Thinkstock

Excerpted from Just Babies: The Origins of Good and Evil by Paul Bloom, out now from Crown.

The comedian Louis CK has a routine in which he talks about his daughter’s understanding of fairness. He begins, “My 5-year-old, the other day, one of her toys broke, and she demanded that I break her sister’s toy to make it fair.” This would make the sisters equal but the joke here is that something here doesn’t feel right: “And I did. I was like crying. And I look at her. She’s got this creepy smile on her face.”

Other intuitions about fairness are simpler. Imagine you have two toys and two children, and you give both toys to one child. If the other child is old enough to speak, she will object. She might say “That’s not fair!” and she’d be correct. An even split would maximize the overall happiness of the children—give each child one toy and they’re both happy; divide them unevenly, and the child who gets nothing is miserable, her sadness outweighing the extra pleasure of the child who gets two. More to the point, it’s just wrong to establish an inequity when you don’t have to.

But fairness is more than deciding the best way to distribute the positive. We also have to determine how to allocate the negative.

Revenge—the personal form of punishment, directed against those who have wronged us personally or who have harmed  our  family or friends—has certain distinctive features. Adam Smith describes our feelings toward a man who has murdered someone we love: “Resentment would prompt us to desire, not only that he should be punished, but that he should be punished by our means, and upon account of that particular injury which he had done to us. Resentment cannot be fully gratified, unless the offender is not only made to grieve in his turn, but to grieve for that particular wrong which we have suffered from him.”

Inigo Montoya, the character in The Princess Bride who seeks to avenge his father’s death,  echoes this sentiment. Montoya tells the man in black his plan: He will approach the murderer and say, “Hello. My name is Inigo Montoya. You killed my father. Prepare to die!” The killer must know precisely why he is being punished and by whom. Then, and only then, can Montoya kill him. (And when he does, it is deeply satisfying.)

In our modern Western societies, first-person revenge plays a less prominent role than it does in the so-called cultures of honor—the  Bedouin, criminal subcultures such as the Mafia, and the cowboy culture of the American West, for example. Individuals living in such cultures cannot rely on external authority to mete out justice, so it’s up to each individual to defend himself and those he cares about. A reputation for violence matters in these societies; this is what deters others from attacking or abusing you. Consistent with this theory, psychologists find that individuals in such societies tend to be disapproving of acts of disrespect and forgiving of acts of retribution.

The psychologist Steven Pinker argues that one reason for the drop in violence over history is the decline of such cultures. We’ve managed, in many parts of the world, to check our appetite for personal retribution. First-party revenge has been largely replaced with third-party punishment, enforced by the government. When my car window was smashed and my belongings were stolen a few months ago, I felt a flash of anger, but really, the problem was best addressed through a police report and a helpful insurance company. If Inigo Montoya were around now, he wouldn’t need to storm the castle to bring his father’s murderer to justice; the police would do it for him, and fewer people would have to die.

Most toddlers do not live in a culture of honor.  There is usually a Leviathan that will resolve conflicts and punish wrongdoers—such as a parent, babysitter, or teacher. Things do change in middle childhood, when children often find themselves in societies where tattling is discouraged and one is expected to fight one’s own battles. Many middle schools and high schools are much like the Wild West. But 2-year-olds are permitted to cry or run away or find an adult when someone smacks them; they aren’t required to retaliate.

This doesn’t mean that children are innocent of retributive desires. They are hardly pacifists, after all. Young children are highly aggressive; indeed, if you measure the rate of physical violence through  the life span, it peaks at about age 2. Families survive the Terrible Twos because toddlers aren’t strong enough to kill with their hands and aren’t capable of using lethal weapons. A 2-year-old with the physical capacities of an adult would be terrifying.

Children’s moralizing impulses are sometimes reflected in violence but are also expressed in a more subtle way. Children tattle. When they see wrongdoing, they are apt to complain about it to an authority figure, and they don’t need to be prompted to do so. In one study, 2- and 3-year-olds were taught a new game to play with a puppet; when the puppet started to break the rules, the children would spontaneously complain to adults. In studies of siblings between the ages of 2 and 6, investigators found that most of what the children said to their parents about their brothers or sisters counted as tattling. And their reports tended to be accurate. They were ratting their sibs out, but they were not making things up.

It’s not just siblings who enjoy telling on each other. The psychologists Gordon Ingram and Jesse Bering explored tattling by children in an inner-city school in Belfast, Ireland, and concluded, “The great majority of children’s talk about their peers’ behavior took the form of descriptions of norm violations.” They noted that it was rare for children to talk to their teachers about something good that someone else had done. As in the sibling study, most of the children’s reports about their peers were true. The children who lied were not the tattlers but the tattlees, who would often deny being responsible for their acts. Children also don’t tattle about insignificant things: One study found that 3-year-olds will tattle when someone destroys an artwork that someone else made, but not when the individual destroys an artwork that nobody cares about.

Part of the satisfaction of tattling surely comes from showing oneself to adults as a good moral agent, a responsible being who is sensitive to right and wrong. But I would bet that children would tattle even if they could do so only anonymously. They would do it just to have justice done. The love of tattling reveals an appetite for payback, a pleasure in seeing wrongdoers  (particularly those who harmed the child, or a friend of the child) being punished. Tattling is a way of off-loading the potential costs of revenge.

It is hard to tell whether babies also have an appetite for justice. Here is the experiment that I wish we could do to find out: show a baby a good character and a bad character, using our standard methods (such as having one character help someone up a hill and another block that individual’s path). Then, one at a time, put the good character and the bad character alone on a stage, facing the  baby. Next to the baby’s hand is a large red button, and the baby is gently shown how to press on it. When the button is touched, the character will act as though it’s been given an electric shock—it will scream and writhe in pain. How will babies respond to this? Will they snatch their hand back when the good guy screams? Will they continue to press it for the bad guy? What if it is a difficult button to press—will babies push down on it, their little faces red with exertion, so as to enact just punishment?

I doubt that we will ever do this study. My colleagues, more fastidious than I am, have ethical concerns. But we have done other studies that offer clues to babies’ punitive motivations.  In one  study  with  Kiley Hamlin, Karen Wynn, and Neha Mahajan, we did a variant of the good guy/bad guy experiments described in the first chapter. In one scenario, one puppet struggled to open a box, one puppet helped to lift the lid, and another slammed the lid shut. In the other scenario, a puppet rolled a ball to one puppet who rolled it right back, and to another who took the ball and ran away. Instead of asking whether infants preferred to interact with the good or the bad puppet, we asked 21-month-olds either to choose  which  of the two to reward by giving it a treat or to choose which of the two to punish by taking away a treat. As predicted, we found that when asked to give a treat, they chose the good character; and when asked to take away a treat, they chose the bad one.

One problem with this study, though, is that it was set up so that children were basically forced to choose a puppet to reward and a puppet to punish. We don’t know, then, whether toddlers have an urge to reward and an urge to punish, let alone whether they feel that rewarding and punishing are the right things to do. Also, given the physical demands of rewarding and punishing, we had to use toddlers instead of infants in this study, and they may well have learned  some  of  the  rewarding  and  punishing  behavior from watching other people.

To explore how babies think about reward and punishment at an earlier age, we decided to look at what 5- and 8-month-olds thought of other individuals who rewarded and punished. Would they prefer someone who rewarded a good guy to someone who punished a good guy? Would they prefer someone who punished a bad guy over someone who rewarded a bad guy? For each contrast, by adult lights at least, one individual is acting justly and the other is not.

We tested the babies by first showing them the scenarios with the box—one puppet would help open the box; the other would slam it shut.  Then we used either the good guy or the bad guy as the main character of an entirely new scene. This time the puppet rolled the ball to two new individuals in turn: one who rolled it back (nice) and the other who ran away with it (mean). We wanted to see which of these two new characters the babies preferred—the one who was nice to the good guy or the one who was mean to the good guy; the one who was nice to the bad guy or the one who was mean to the bad guy.

When the two characters were interacting with the good guy (the one who had helped open the box), babies preferred to reach for the character who was nice to it as opposed to the one who was mean to it—probably because babies tend to prefer nice puppets overall. Indeed, the 5-month-olds also preferred to reach for the character who was nice to the bad guy. Either these younger babies weren’t keeping track of the whole sequence of events, or they just preferred nice puppets, regardless of whom they were interacting with.

But the  8-month-olds were more sophisticated: They preferred the puppet who was mean to the bad guy over the one who was nice to it. So at some point after 5 months, babies begin to prefer punishers—when the punishment is just.

Excerpted from Just Babies: The Origins of Good and Evil by Paul Bloom, out now from Crown