How to Succeed in Childhood

by Judith Rich Harris


very day, tell your children that you love them. Hug them at least once every 24 hours. Never hit them. If they do something wrong, don't say, "You're bad!" Say, "What you did was bad." No, wait--even that might be too harsh. Say, instead, "What you did made me unhappy."

   The people who are in the business of giving out this sort of advice are very angry at me, and with good reason. I'm the author of The Nurture Assumption--the book that allegedly claims that "parents don't matter." Though that's not what the book actually says, the advice givers are nonetheless justified in their anger. I don't pull punches, and I'm not impressed by their air of benevolent omniscience. Their advice is based not on scientific evidence but on prevailing cultural myths.

   The advice isn't wrong; it's just ineffective. Whether parents do or don't follow it has no measurable effect on how their children turn out. There is a great deal of evidence that the differences in how parents rear their children are not responsible for the differences among the children. I've reviewed this evidence in my book; I will not do it again here.

   Let me, however, bring one thing to your attention: the advice given to parents in the early part of this century was almost the mirror image of the advice that is given today. In the early part of this century, parents were not warned against damaging their children's self-esteem; they were warned against "spoiling" them. Too much attention and affection were thought to be bad for kids. In those days, spanking was considered not just the parents' right but their duty.

   Partly as a result of the major retoolings in the advice industry, child-rearing styles have changed drastically over the course of this century. Although abusive parents have always existed, run-of-the-mill parents--the large majority of the population--administer more hugs and fewer spankings than they used to.

   Now ask yourself this: Are children turning out better? Are they happier and better adjusted than they were in the earlier part of the century? Less aggressive? Less anxious? Nicer?


t was Sigmund Freud who gave us the idea that parents are the be-all and end-all of the child's world. According to Freudian theory, children learn right from wrong--that is, they learn to behave in ways their parents and their society deem acceptable--by identifying with their parents. In the calm after the storm of the oedipal crisis, or the reduced-for-quick-sale female version of the oedipal crisis, the child supposedly identifies with the parent of the same sex.

Freud's name is no longer heard much in academic departments of psychology, but the theory that children learn how to behave by identifying with their parents is still accepted. Every textbook in developmental psychology (including, I confess, the one I co-authored) has its obligatory photo of a father shaving and a little boy pretending to shave. Little boys imitate their fathers, little girls imitate their mothers, and, according to the theory, that's how children learn to be grownups. It takes them a while, of course, to perfect the act.

It's a theory that could have been thought up only by a grownup. From the child's point of view, it makes no sense at all. What happens when children try to behave like grownups is that, more often than not, it gets them into trouble. Consider this story, told by Selma Fraiberg, a child psychologist whose book The Magic Years was popular in the 1960s:

Thirty-month-old Julia finds herself alone in the kitchen while her mother is on the telephone. A bowl of eggs is on the table. An urge is experienced by Julia to make scrambled eggs.... When Julia's mother returns to the kitchen, she finds her daughter cheerfully plopping eggs on the linoleum and scolding herself sharply for each plop, "NoNoNo. Mustn't dood it! NoNoNo. Mustn't dood it!"
Fraiberg attributed Julia's lapse to the fact that she had not yet acquired a superego, presumably because she had not yet identified with her mother. But look at what was Julia doing when her mother came back and caught her egg-handed: she was imitating her mother! And yet Mother was not pleased.

hildren cannot learn how to behave appropriately by imitating their parents. Parents do all sorts of things that children are not allowed to do--I don't have to list them, do I?--and many of them look like fun to people who are not allowed to do them. Such prohibitions are found not only in our own society but everywhere, and involve not only activities such as making scrambled eggs but patterns of social behavior as well. Around the world, children who behave too much like grownups are considered impertinent.

Sure, children sometimes pretend to be adults. They also pretend to be horses and monsters and babies, but that doesn't mean they aspire to be horses or monsters or babies. Freud jumped to the wrong conclusions, and so did several generations of developmental psychologists. A child's goal is not to become an adult; a child's goal is to be a successful child.

What does it take to be a successful child? The child's first job is to learn how to get along with her parents and siblings and to do the things that are expected of her at home. This is a very important job--no question about it. But it is only the first of the child's jobs, and in the long run it is overshadowed in importance by the child's second job: to learn how to get along with the members of her own generation and to do the things that are expected of her outside the home.

Almost every psychologist, Freudian or not, believes that what the child learns (or doesn't learn) in job 1 helps her to succeed (or fail) in job 2. But this belief is based on an obsolete idea of how the child's mind works, and there is good evidence that it is wrong.

Consider the experiments of developmental psychologist Carolyn Rovee-Collier. A young baby lies on its back in a crib. A mobile with dangling doodads hangs overhead. A ribbon runs from the baby's right ankle to the mobile in such a way that whenever the baby kicks its right leg, the doodads jiggle. Babies are delighted to discover that they can make something happen; they quickly learn how to make the mobile move. Two weeks later, if you show them the mobile again, they will immediately start kicking that right leg.

But only if you haven't changed anything. If the doodads hanging from the mobile are blue instead of red, or if the liner surrounding the crib has a pattern of squares instead of circles, or if the crib is placed in a different room, they will gape at the mobile cluelessly, as if they've never seen such a thing in their lives.


t's not that they're stupid. Babies enter the world with a mind designed for learning and they start using it right away. But the learning device comes with a warning label: what you learn in one situation might not work in another. Babies do not assume that what they learned about the mobile with the red doodads will work for the mobile with the blue doodads. They do not assume that what worked in the bedroom will work in the den. And they do not assume that what worked with their mother will work with their father or the babysitter or their jealous big sister or the kids at the daycare center.

Fortunately, the child's mind is equipped with plenty of storage capacity. As the cognitive scientist Steven Pinker put it in his foreword to my book, "Relationships with parents, with siblings, with peers, and with strangers could not be more different, and the trillion-synapse human brain is hardly short of the computational power it would take to keep each one in a separate mental account."

That's exactly what the child does: keeps each one in a separate mental account. Studies have shown that a baby with a depressed mother behaves in a subdued fashion in the presence of its mother, but behaves normally with a caregiver who is not depressed. A toddler taught by his mother to play elaborate fantasy games does not play these games when he's with his playmates--he and his playmates devise their own games. A preschooler who has perfected the delicate art of getting along with a bossy older sibling is no more likely than a first-born to allow her peers in nursery school to dominate her. A school-age child who says she hates her younger brother --they fight like cats and dogs, their mother complains --is as likely as any other child to have warm and serene peer relationships. Most telling, the child who follows the rules at home, even when no one is watching, may lie or cheat in the schoolroom or on the playground, and vice versa.

Children learn separately how to behave at home and how to behave outside the home, and parents can influence only the way they behave at home. Children behave differently in different social settings because different behaviors are required. Displays of emotion that are acceptable at home are not acceptable outside the home. A clever remark that would be rewarded with a laugh at home will land a child in the principal's office at school. Parents are often surprised to discover that the child they see at home is not the child the teacher sees. I imagine teachers get tired of hearing parents exclaim, "Really? Are you sure you're talking about my child?"

The compartmentalized world of childhood is vividly illustrated by the child of immigrant parents. When immigrants settle in a neighborhood of native-born Americans, their children become bicultural, at least for a while. At home they practice their parents' culture and language, outside the home they adopt the culture and language of their peers. But though their two worlds are separate, they are not equal. Little by little, the outside world takes precedence: the children adopt the language and culture of their peers and bring that language and culture home. Their parents go on addressing them in Russian or Korean or Portuguese, but the children reply in English. What the children of immigrants end up with is not a compromise, not a blend. They end up, pure and simple, with the language and culture of their peers. The only aspects of their parents' culture they retain are things that are carried out at home, such as cooking.

ate-20th-century native-born Americans of European descent are as ethnocentric as the members of any other culture. They think there is only one way to raise children--the way they do it. But that is not the way children are reared in the kinds of cultures studied by anthropologists and ethologists. The German ethologist Iren´┐Żus Eibl-Eibesfeldt has described what childhood is like in the hunter-gatherer and tribal societies he spent many years observing.

In traditional cultures, the baby is coddled for two or three years--carried about by its mother and nursed whenever it whimpers. Then, when the next baby comes along, the child is sent off to play in the local play group, usually in the care of an older sibling. In his 1989 book Human Ethology, Eibl-Eibesfeldt describes how children are socialized in these societies:

Three-year-old children are able to join in a play group, and it is in such play groups that children are truly raised. The older ones explain the rules of play and will admonish those who do not adhere to them, such as by taking something away from another or otherwise being aggressive. Thus the child's socialization occurs mainly within the play group. . . . By playing together in the children's group the members learn what aggravates others and which rules they must obey. This occurs in most cultures in which people live in small communities.
Once their tenure in their mothers' arms has ended, children in traditional cultures become members of a group. This is the way human children were designed to be reared. They were designed by evolution to become members of a group, because that's the way our ancestors lived for millions of years. Throughout the evolution of our species, the individual's survival depended upon the survival of his or her group, and the one who became a valued member of that group had an edge over the one who was merely tolerated.

uman groups started out small: in a hunter-gatherer band, everyone knows everyone else and most are blood relatives. But once agriculture began to provide our ancestors with a more or less dependable supply of food, groups got bigger. Eventually they became large enough that not everyone in them knew everyone else. As long ago as 1500 b.c. they were sometimes that large. There is a story in the Old Testament about a conversation Joshua had with a stranger, shortly before the Battle of Jericho. They met outside the walls of the beleaguered town, and Joshua's first question to the stranger was, "Are you for us or for our adversaries?"

Are you one of us or one of them? The group had become an idea, a concept, and the concept was defined as much by what you weren't as by what you were. And the answer to the question could be a matter of life or death. When the walls came tumbling down, Joshua and his troops killed every man, woman, and child in Jericho. Even in Joshua's time, genocide was not a novelty: fighting between groups, and wholesale slaughter of the losers, had been going on for ages. According to the evolutionary biologist Jared Diamond, it is "part of our human and prehuman heritage."

Are you one of us or one of them? It was the question African Americans asked of Colin Powell. It was the question deaf people asked of a Miss America who couldn't hear very well but who preferred to communicate in a spoken language. I once saw a six-year-old go up to a 14-year-old and ask him, "Are you a kid or a grownup?"

The human mind likes to categorize. It is not deterred by the fact that nature often fails to arrange things in convenient clumps but instead provides a continuum. We have no difficulty splitting up continua. Night and day are as different as, well, night and day, even though you can't tell where one leaves off and the other begins. The mind constructs categories for people -- male or female, kid or grownup, white or black, deaf or hearing -- and does not hesitate to draw the lines, even if it's sometimes hard to decide whether a particular individual goes on one side or the other.

Babies only a few months old can categorize. By the time they reach their first birthday, they are capable of dividing up the members of their social world into categories based on age and sex: they distinguish between men and women, between adults and children. A preference for the members of their own social category also shows up early. One-year-olds are wary of strange adults but are attracted to other children, even ones they've never met before. By the age of two, children are beginning to show a preference for members of their own sex. This preference grows steadily stronger over the next few years. School-age girls and boys will play together in places where there aren't many children, but when they have a choice of playmates, they tend to form all-girl and all-boy groups. This is true the world around.

he brain we won in the evolutionary lottery gave us the ability to categorize, and we use that skill on people as well as things. Our long evolutionary history of fighting with other groups predisposes us to identify with one social category, to like our own category best, and to feel wary of (or hostile toward) members of other categories. The emotions and motivations that were originally applied to real physical groups are now applied to groups that are only concepts: "Americans" or "Democrats" or "the class of 2001." You don't have to like the other members of your group in order to consider yourself one of them; you don't even have to know who they are. The British social psychologist Henri Tajfel asked his subjects--a bunch of Bristol schoolboys--to estimate the number of dots flashed on a screen. Then half the boys were privately told that they were "overestimators," the others that they were "underestimators." That was all it took to make them favor their own group. They didn't even know which of their schoolmates were in their group and which were in the other.

he most famous experiment in social psychology is the Robber's Cave study. Muzafer Sherif and his colleagues started with 22 eleven-year-old boys, carefully selected to be as alike as possible, and divided them into two equal groups. The groups--the "Rattlers" and the "Eagles"--were separately transported to the Robber's Cave summer camp in a wilderness area of Oklahoma. For a while, neither group knew of the other's existence. But the first time the Rattlers heard the Eagles playing in the distance, they reacted with hostility. They wanted to "run them off." When the boys were brought together in games arranged by researchers disguised as camp counselors, push quickly came to shove. Before long, the two groups were raiding each other's cabins and filling socks with stones in preparation for retaliatory raids.

When people are divided (or divide themselves) into two groups, hostility is one common result. The other, which happens more reliably though it is less well known, is called the "group contrast effect." The mere division into two groups tends to make each group see the other as different from itself in an unfavorable way, and that makes its members want to be different from the other group. The result is that any pre-existing differences between the groups tend to widen, and if there aren't any differences to begin with, the members create them. Groups develop contrasting norms, contrasting images of themselves.

In the Robber's Cave study, it happened very quickly. Within a few days of their first encounter, the Eagles had decided that the Rattlers used too many "cuss-words" and resolved to give up cussing; they began to say a prayer before every game. The Rattlers, who saw themselves as tough and manly, continued to favor scatology over eschatology. If an Eagle turned an ankle or skinned a knee, it was all right for him to cry. A Rattler who sustained a similar injury might cuss a bit, but he would bear up stoically.

he idea for group socialization theory came to me while I was reading an article on juvenile delinquency. The article reported that breaking the law is highly common among adolescents, even among those who were well behaved as children and who are destined to turn into law-abiding adults. This unendearing foible was attributed to the frustration teenagers experience at not being adults: they are longing for the power and privilege of adulthood.

"Wait a minute," I thought. "That's not right. If teenagers really wanted to be adults, they wouldn't be spraying graffiti on overpasses or swiping nail polish from drugstores. If they really wanted to emulate adults they would be doing boring adult things, like sorting the laundry or figuring out their taxes. Teenagers aren't trying to be like adults; they are trying to contrast themselves with adults! They are showing their loyalty to their own group and their disdain for adults' rules!"

I don't know what put the idea into my head; at the time, I didn't know beans about social psychology. It took eight months of reading to fill the gaps in my education. What I learned in those eight months was that there is a lot of good evidence to back up my hunch, and that it applies not only to teenagers but to young children as well.

Sociologist William Corsaro has spent many years observing nursery school children in the United States and Italy. Here is his description of four-year-olds in an Italian scuola materna, a government-sponsored nursery school:

In the process of resisting adult rules, the children develop a sense of community and a group identity. [I would have put it the other way around: I think group identity leads to the resistance.] The children's resistance to adult rules can be seen as a routine because it is a daily occurrence in the nursery school and is produced in a style that is easily recognizable to members of the peer culture. Such activity is often highly exaggerated (for instance, making faces behind the teacher's back or running around) or is prefaced by "calls for the attention" of other children (such as, "look what I got" in reference to possession of a forbidden object, or "look what I'm doing" to call attention to a restricted activity.
Group contrast effects show up most clearly when "groupness"--Henri Tajfel's term--is salient. Children see adults as serious and sedentary, so when the social categories kids and grownups are salient --as they might be, for instance, when the teacher is being particularly bossy--the children become sillier and more active. They demonstrate their fealty to their own age group by making faces and running around.

This has nothing to do with whether they like their teachers personally. You can like people even if they're members of a different group and even if you don't much like that group -- a conflict of interests summed up in the saying, "Some of my best friends are Jews." When groupness is salient, even young children contrast themselves with adults and collude with each other in defying them. And yet some of their best friends are grownups.

earning how to behave properly is complicated, because proper behavior depends on which social category you're in. In every society, the rules of behavior depend on whether you're a grownup or a kid, a female or a male, a prince or a peon. Children first have to figure out the social categories that are relevant in their society, and then decide which category they belong in, then tailor their behavior to the other members of their category.

That brief description seems to imply that socialization makes children more alike, and so it does, in some ways. But groups also work to create or exaggerate differences among their members--differences in personality. Even identical twins reared in the same home do not have identical personalities. When groupness is not salient--when there is no other group around to serve as a foil--a group tends to fall apart into individuals, and differences among them emerge or increase. In boys' groups, for example, there is usually a dominance hierarchy, or "pecking order." I have found evidence that dominant boys develop different personalities from those at the bottom of the ladder.

Groups also typecast their members, pinning labels on them--joker, nerd, brain--that can have lifelong repercussions. And children find out about themselves by comparing themselves with their group mates. They come to think well or poorly of themselves by judging how they compare with the other members of their own group. It doesn't matter if they don't measure up to the standards of another group. A third-grade boy can think of himself as smart if he knows more than most of his fellow third-graders. He doesn't have to know more than a fourth-grader.

ccording to my theory, the culture acts upon children not through their parents but through the peer group. Children's groups have their own cultures, loosely based on the adult culture. They can pick and choose from the adult culture, and it's impossible to predict what they'll include. Anything that's common to the majority of the kids in the group may be incorporated into the children's culture, whether they learned it from their parents or from the television set. If most of the children learned to say "please" and "thank you" at home, they will probably continue to do so when they're with their peers. The child whose parents failed to teach her that custom will pick it up from the other children: it will be transmitted to her, via the peer group, from the parents of her peers. Similarly, if most of the children watch a particular TV show, the behaviors and attitudes depicted in the show may be incorporated into the norms of their group. The child whose parents do not permit him to watch that show will nonetheless be exposed to those behaviors and attitudes. They are transmitted to him via the peer group.

Thus, even though individual parents may have no lasting effects on their children's behavior, the larger culture does have an effect. Child-rearing practices common to most of the people in a culture, such as teaching children to say "please" and "thank you," can have an effect. And the media can have an effect.

In the hunter-gatherer or tribal society, there was no privacy: everybody knew what everybody else was doing. Nowadays children can't ordinarily watch their neighbors making love, having babies, fighting, and dying, but they can watch these things happening on the television screen. Television has become their window on society, their village square. They take what they see on the screen to be an indication of what life is like--what life is supposed to be--and they incorporate it into their children's cultures.

ne of my goals in writing The Nurture Assumption was to lighten some of the burdens of modern parenthood. Back in the 1940s, when I was young, the parents of a troublesome child--my parents, for instance--got sympathy, not blame. Nowadays parents are likely to be held culpable for anything that goes wrong with their child, even if they've done their best. The evidence I've assembled in my book indicates that there is a limit to what parents can do: how their child turns out is largely out of their hands. Their major contribution occurs at the moment of conception. This doesn't mean it's mostly genetic; it means that the environment that shapes the child's personality and social behavior is outside the home.

I am not advocating irresponsibility. Parents are in charge of how their children behave at home. They can decide where their children will grow up and, at least in the early years, who their peers will be. They are the chief determiners of whether their children's life at home will be happy or miserable, and they have a moral obligation to keep it from being miserable. My theory does not grant people the license to treat children in a cruel or negligent way.

Although individual parents have little power to influence the culture of children's peer groups, larger numbers of parents acting together have a great deal of power, and so does the society as a whole. Through the prevailing methods of child rearing it fosters, and through influences--especially the media--that act directly on peer-group norms and values, a society shapes the adults of the future. Are we shaping them the way we ought to?


Reprinted from the Winter 1999 Wilson Quarterly


This article may not be resold, reprinted, or redistributed for compensation of any kind without prior written permission from the author. For further reprint information, please contact Permissions, The Wilson Quarterly, One Woodrow Wilson Plaza, 1300 Pennsylvania Avenue, NW, Washington, D.C. 20004-3027 (202-691-4200).