It gets worse. In many studies, the same parties (in some studies the parents, in others the children) supply the data on both the parents’ behavior and the child’s. Parents tell the experimenter how they treat their children and what their children are like, or adolescents tell the experimenter what they are like and how their parents treat them. Those studies—suspiciously—show much stronger correlations than ones in which a third party assesses the parents and the child.33 The problem is not just that people tend to look at themselves and at their families through the same rose-colored or jaundiced lenses, but also that the relationship between parents and adolescents is a two-way street. Harris sums up the problems when commenting on a widely publicized 1997 study. The authors claimed, solely on the basis of teenagers’ responses to a questionnaire about themselves and their families, that “parent-family connectedness”—close bonds, high expectations, lots of affection—is “protective” against adolescent ills such as drugs, cigarettes, and unsafe sex. Harris notes:

  A happy person tends to check off upbeat answers to all the questions: Yes, my parents are good to me; yes, I’m doing fine. A person who cares about presenting a socially acceptable face to the world checks off socially acceptable responses: Yes, my parents are good to me; no, I haven’t been in any fights or smoked anything illegal. A person who is angry or depressed checks off angry or depressed responses: My parents are jerks and I flunked the algebra test and to hell with your questionnaire….

  … Perhaps what misled those eighteen federal agencies into thinking they were getting their 25 million dollars worth was the positive way the researchers phrased their findings: good relationships with parents exert a protective effect. Expressed in a different (but equally accurate) way, the results sound less interesting: adolescents who don’t get along well with their parents are more likely to use drugs or engage in risky sex. The results sound still less interesting expressed this way: adolescents who use drugs or engage in risky sex don’t get along well with their parents.34

  Yet another problem crops up when researchers direct all their questions to the parents rather than to the offspring. People behave differently in different settings. That includes children, who tend to behave differently inside and outside the home. So even if parents’ behavior does affect how their children behave with them, it may not affect how their children behave with other people. When parents describe their children’s behavior, they describe the behavior they see in the home. To show that parents shape their children, then, a study would have to control for genes (by testing twins or adoptees), distinguish between parents affecting children and children affecting parents, measure the parents and the children independently, look at how children behave outside the home rather than inside, and test older children and young adults to see whether any effects are transient or permanent. No study that has claimed to show effects of parenting has met these standards.35

  If behavioral genetic studies show no lasting effects of the home, and studies of parenting practices are uninformative, what about studies that compare radically different childhood milieus? The results, again, are bracing. Decades of studies have shown that, all things being equal, children turn out pretty much the same way whether their mothers work or stay at home, whether they are placed in daycare or not, whether they have siblings or are only children, whether their parents have a conventional or an open marriage, whether they grow up in an Ozzie-and-Harriet home or a hippie commune, whether their conceptions were planned, were accidental, or took place in a test tube, and whether they have two parents of the same sex or one of each.36

  Even growing up without a father in the house, which does correlate with troubles such as dropping out of school, remaining idle, and having babies while a teenager, may not cause the troubles directly.37 Children with experiences that should make up for the missing father, such as having a stepfather, a live-in grandmother, or frequent contact with the birth father, are no better off. The number of years that the father was in the house before leaving makes no difference. And children whose fathers died do not have the poor outcomes of children whose fathers walked out or were never there. The absence of a father may not be a cause of adolescent problems but a correlate of the true causes, which may include poverty, neighborhoods with lots of unattached men (who live in de facto polygyny and hence compete violently for status), frequent moves (which force children to start from the bottom of the pecking order in new peer groups), and genes that make both fathers and children more impulsive and quarrelsome.

  The 1990s was the Decade of the Brain and the decade in which parents were told they were in charge of their babies’ brains. The first three years of life was described as a critical window of opportunity in which the child’s brain had to be constantly stimulated to keep it growing properly. Parents of late-talking children were blamed for not blanketing them in enough verbiage; the ills of the inner city were blamed on children’s having to stare at empty walls. Bill and Hillary Clinton convened a conference at the White House to learn about the research, at which Mrs. Clinton said that the experiences of the first three years “can determine whether children will grow up to be peaceful or violent citizens, focused or undisciplined workers, attentive or detached parents themselves.”38 The governors of Georgia and Missouri asked their legislators for millions of dollars to issue every new mother with a Mozart CD. (They had confused experiments on infant brain development with experiments—since discredited—alleging that adults benefit from listening to a few minutes of Mozart.)39 The pediatrician and childcare guru T. Berry Brazelton had the most hopeful suggestion of all: that nurturance during the first three years will protect children from the lure of tobacco when they become adolescents.40

  In his book The Myth of the First Three Years, the cognitive neuroscience expert Jon Bruer showed that there was no science behind these astonishing claims.41 No psychologist has ever documented a critical period for cognitive or language development that ends at three. And though depriving an animal of stimulation (by sewing an eye shut or keeping it in a barren cage) may hurt its brain growth, there is no evidence that providing extra stimulation (beyond what the organism would encounter in its normal habitat) enhances its brain growth.

  So nothing in the research on family environments contradicts the behavioral geneticists’ Second Law, which says that growing up in a particular family has little or no systematic effect on one’s intellect and personality. And this leaves us with a maddening puzzle. No, it’s not all in the genes; around half the variation in personality, intelligence, and behavior comes from something in the environment. But whatever that something is, it cannot be shared by two children growing up in the same home with the same parents. And that rules out all the obvious somethings. What is the elusive Mister Jones factor?

  REFUSING TO GIVE up on parents, some developmental psychologists have trained their sights on the only remaining possibility that gives parents a starring role. The impotence of the shared environment says only that what parents do to all their children is powerless to shape them. But obviously parents don’t treat their children alike. Perhaps the individualized parenting that mothers and fathers adapt to each child does have the power to shape them. It is the interaction between parents and children that affects them, not a one-size-fits-all parenting philosophy.42

  At first this looks reasonable. But when you think it through, it does not restore a shaping role for parents, or for parenting advice, after all.43

  What would individualized parenting look like? Presumably parents would tailor their parenting to the needs and talents of each child. A headstrong child would elicit firmer discipline than a compliant one; a fearful child would elicit more protectiveness than a bold one. The problem, as we saw in an earlier section, is that the differences in parenting cannot be separated from the preexisting differences in the children. If the fearful child turns into a fearful adult, we don’t know whether it was an effect of the overprotective parent or a continuation of the fearfulness th
e child was born with.

  And surprisingly, if children do elicit systematic differences in parenting it would show up as an effect of the genes: it would go into the heritability term, not the unique-environment term. The reason is that heritability is a measure of correlation and cannot distinguish direct effects of the genes (proteins that help wire the brain or trigger hormones) from indirect effects that operate many links away. Earlier I mentioned that attractive people are more assertive, presumably because they get accustomed to other people’s kissing up to them. That is a highly indirect effect of the genes and would make assertiveness heritable even if there were no genes for assertive brains, just genes for violet eyes to die for. Similarly, if children with certain innate traits make their parents more patient, or encouraging, or strict, then parental patience, encouragement, and strictness would also count as “heritable.” Now, if such individualized parenting does affect the way children turn out, a critic could legitimately say that the direct effects of the genes had been overestimated, because some of them would really be indirect effects of the children’s genes on traits of the children that affect their parents’ behavior, which in turn affects the children. (The hypothesis is baroque, and I will soon show why it is unlikely to be true, but let’s assume it is true for argument’s sake.) But at best, the effects of parenting would be fighting with other genetic effects (direct and indirect) for some portion of the 40 to 50 percent of the variation attributed to the genes. The 50 percent attributable to the unique environment would still be up for grabs.

  Here is what would have to happen if the effects of the unique environment are to be explained by an interaction between parents and children (using the statistician’s technical sense of the word “interaction,” which is the one relevant to our puzzle). A given practice would have to affect some children one way, and other children another way, and the two effects would have to cancel out. For example, sparing the rod would have to spoil some children (making them more violent) and teach others that violence is not a solution (making them less violent). Displays of affection would have to make some children more affectionate (because they identify with their parents) and others less affectionate (because they react against their parents). The reason the effects have to go in opposite directions is that if a parenting practice had a consistent effect, on average, across all children, it would turn up as an effect of the shared environment. Adopted siblings would be similar, sibs growing up together would be more similar than sibs growing up apart—neither of which happens. And if it was applied successfully to some kinds of children and was avoided, or was ineffective, with other kinds, that would turn up as an effect of the genes.

  The problems with the parent-child interaction idea now become obvious. It is implausible that any parenting process would have such radically different effects on different children that the sum of the effects (the shared environment) would add up to zero. If hugging merely makes some children more confident and has no effect on others, then the huggers should still have more confident children on average (some becoming more confident, others showing no change) than the cold fish. But, holding genes constant, they don’t. (To put it in technical terms familiar to psychologists: it is rare to find a perfect crossover interaction, that is, an interaction with no main effects.) This is also, by the way, one of the reasons that heritability itself almost certainly cannot be reduced to child-specific parenting. Unless parents’ behavior is completely determined by their child’s inborn traits, some parents will behave somewhat differently from others across the board, and that would turn up in effects of the shared environment—which in fact are negligible.

  But let’s say that these parent-child interactions (in the technical sense) really do exist, and really do shape the child. The moral would be that across-the-board parenting advice is useless. Anything that parents do to make some children better will make an equal number of children worse.

  In any case, the parent-child interaction theory can be tested directly. Psychologists can measure how parents treat the different children within a family, and see if the treatments correlate with how the children turn out, holding genes constant. The answer is that in almost every case they don’t. Virtually all the differences in parenting within a family can be explained as reactions to genetic differences that the children were born with. And parental behavior that does differ among children for nongenetic reasons, such as marital conflict triggered by some siblings but not by others, or more parenting effort directed at one sibling than at another, has no effect.44 The leader of a recent heroic study, who had hoped to prove that differences in parenting do affect how children turn out, confessed that he was “shocked” by his own results.45

  There is another way that a home environment could differ among children in the same family for reasons having nothing to do with their genes: birth order. A firstborn usually has several years of undivided parental attention with no annoying siblings around. Laterborns have to compete with their siblings for parental attention and other family resources, and have to figure out how to hold their own against stronger and more entrenched competitors.

  In Born to Rebel Sulloway predicted that firstborns should parlay their advantages into a more assertive personality.46 And because they identify with their parents, and by extension with the status quo, they should grow up to be more conservative and conscientious. Laterborns, in contrast, should be more conciliatory and open to new ideas and experiences. Though family therapists and laypeople have had these impressions for a long time, Sulloway tried to explain them in terms of Trivers’s theory of parent-offspring conflict and its corollary, sibling rivalry. He found some support for these ideas in a meta-analysis (a quantitative literature review) of studies of birth order and personality.47

  Sulloway’s theory, however, also requires that children use the same strategies outside the home—with their peers and colleagues—as the ones that served them well inside the home. That does not follow from Trivers’s theory; indeed, it contradicts the larger theory from evolutionary psychology that relationships with blood relatives should be very different from relationships with nonrelatives. Tactics that work on a sibling or parent may not work so well on a colleague or stranger. And in fact subsequent analyses have shown that any effects of birth order on personality turn up in the studies that ask siblings or parents to rate one another, or to rate themselves with respect to a sibling, which of course can assess only their family relationships. When personality is measured by neutral parties outside the family, birth-order effects diminish or disappear.48 Any differences in the parenting of firstborns and laterborns—novice or experienced parents, divided or undivided attention, pressure to carry on the family legacy or indulgent babying—seem to have little or no effect on personality outside the home.

  Similarities within a home don’t shape children; differences within a home don’t shape children. Perhaps, Harris says, we should look outside the home.

  IF YOU GREW up in a different part of the world from where your parents grew up, consider this question: Do you sound like your parents, or like the people you grew up with? What about the way you dress, or the music you listen to, or the way you spend your free time? Consider the same question about your children if they grew up in a different part of the world from where you grew up—or for that matter, even if they didn’t. In almost every case, people model themselves after their peers, not their parents.

  This is Harris’s explanation of the elusive environmental shaper of personality, which she calls Group Socialization theory. It’s not all in the genes, but what isn’t in the genes isn’t from the parents either. Socialization—acquiring the norms and skills necessary to function in society—takes place in the peer group. Children have cultures, too, which absorb parts of the adult culture and also develop values and norms of their own. Children do not spend their waking hours trying to become better and better approximations of adults. They strive to be better and better children, ones that function well in thei
r own society. It is in this crucible that our personalities are formed.

  Multidecade, child-obsessed parenting, Harris points out, is an evolutionarily recent practice. In foraging societies, mothers carry their children on their hips or backs and nurse them on demand until the next child arrives two to four years later.49 The child is then dumped into a play group with his older siblings and cousins, switching from being the beneficiary of almost all of the mother’s attention to almost none of it. Children sink or swim in the milieu of other children.

  Children are not just attracted to the norms of their peers; to some degree they are immune to the expectations of their parents. The theory of parent-offspring conflict predicts that parents do not always socialize a child in the child’s best interests. So even if children acquiesce to their parents’ rewards, punishments, examples, and naggings for the time being—because they are smaller and have no choice—they should not, according to the theory, allow their personalities to be shaped by these tactics. Children must learn what it takes to gain status among their peers, because status at one age gives them a leg up in the struggle for status at the next, including the young-adult stages in which they first compete for the attention of the opposite sex.50

  What first attracted me to Harris’s theory was its ability to explain a half-dozen puzzling facts in the part of psychology I work in the most, language.51 Psycholinguists argue a lot about heredity and environment, but they all equate “the environment” with “parents.” But many phenomena of children’s language development just don’t fit that equation. In traditional cultures, mothers don’t say much to their children until they are old enough to hold up their end of the conversation; the children pick up language from other children. People’s accents almost always resemble the accents of their childhood peers, not the accents of their parents. Children of immigrants acquire the language of their adopted homeland perfectly, without a foreign accent, as long as they have access to native speaking peers. They then try to force their parents to switch to the new language, and if they succeed, they may forget the mother tongue entirely. The same is true of hearing children of deaf parents, who learn the spoken language of their community without a hitch. Children thrown together without a common language from the grownups will quickly invent one; that is how creole languages, and the signed languages of the deaf, came into being. Now, a particular language like English or Japanese (as opposed to the instinct for language in general) is an example of learned social behavior par excellence. If children cultivate a fine ear for the nuances of their peers’ speech, and if they cast their lot with their peers’ language over their parents’, it suggests that their social antennae are aimed peerward.