December 7, 1998

The Influence of Anxiety

By Ann Hulbert

Parents, peers, and the rearing of children.

The Nurture Assumption:
Why Children Turn Out the Way They Do

by Judith Rich Harris
(Free Press, 462 pp., $26)

Three Seductive Ideas
by Jerome Kagan
(Harvard University Press, 232 pp., $27.50)


Beware of grandmothers, warned the pediatricians and the psychologists who set up shop as child-rearing experts a century ago. Doctors joked nervously among themselves about the gray-haired authorities whom they aimed to edge out of the business of advising parents. "I would suggest as a heading for a chapter `The elimination of the grandmother,'" opined one pediatric adviser at a meeting in 1917. "The grandmother is the great trial of my life. She probably has had eight children and I have not had any."

The experts were not kidding about counting her experience against her. The age of science had dawned, and grandmotherly wisdom was based on out-of-date anecdotes rather than on objective, extensive data. Hers was the voice of sentimentalism and superstition, full of homespun lore that was barely serviceable back in the nineteenth century. To prepare children to thrive in the intimidating twentieth, parents needed primers packed with advice straight from the new psychology labs. If grandmother balks or butts in, the experts counseled modern mothers, just throw the book at her.

Now a grandmother has thrown all the books back in the experts' faces. Judith Rich Harris is full of impertinence grise as she, a housebound non-academic, challenges roughly a hundred years' worth of psychological research on children's development that has purported (she says) to show what enormous power parents have to shape their offspring's futures. Turning the tables on the experts, she struts her grandmotherly status and argues that it is they who have succumbed to sentimentalism and superstition. The experts, Harris maintains, have never actually proved their case. Instead they have called upon science to prop up a hearth-warming myth that has turned out to be anxiety-inducing for all concerned.

Harris prides herself on clear-headed distance from the academic establishment. She was "kicked out of Harvard's Department of Psychology with only a master's degree." After having one child, she adopted another. Later, "stuck at home for many years due to chronic health problems," she read on her own all those books that she hadn't gotten to as a graduate student or, evidently, as a nervous parent. Harris turned to writing developmental psychology textbooks, freelance work that suited a woman who carried water for no one in the field and who was confined to bed--until, after years of growing uneasiness, "it suddenly occurred to me that many of the things I had been telling those credulous college students were wrong." Being shut in, Harris believes, allowed her to keep an unusually open mind. She was the informed outsider who could recognize scientific data as the wishful dogma that it has often been.

Yet Harris is not content to be merely a village skeptic, exposing the shoddy science inspired by what she calls "the nurture assumption--the notion that parents are the most important part of the child's environment and can determine, to a large extent, how the child turns out." She is a grandmother who aspires to be a scientific guru herself. Harris has come up with an alternative theory, with nothing less than "a new way of explaining why children turn out the way they do." She contends that it is the peer group, not individual parents, that determines how the genetic predispositions with which we are born develop into the socialized personality we are blessed, or burdened, with by the time we reach adulthood, "somewhere between the ages of 17 and 35." Harris presents this theory as a homespun patchwork of various schools of academic research that have rarely intersected--personality development, social psychology, cultural anthropology, and child development. Notions from the newer field of evolutionary psychology help her stitch it together, and further backing is provided by the comparably recent enterprise of behavioral genetics.

This grandmother's fate has been swift elevation rather than elimination. A glowing profile of Harris in The New Yorker and an award from the American Psychological Association preceded the appearance of her book. The Nurture Assumption immediately stirred up controversy. In his foreword, Steven Pinker pronounces it "a turning point in the history of psychology." Others in the field have lined up to say that "she's all wrong" and "I am embarrassed for psychology." Like many successful child-rearing gurus before her, Harris appears to be telling non-experts what we most, and least, want to hear. Her book has been celebrated as a long-awaited liberation for guilt-ridden super-parents. It has also been denounced as an ill-timed absolution for "self-absorbed parents with a track record of putting their own desires ahead of their children's needs." In the working-parent era, a book like The Nurture Assumption is certain to touch a nerve. Perhaps that is why it has not been read very carefully.


In any period of the twentieth century, a book such as The Nurture Assumption would have touched a nerve. For Harris has zeroed in on the dilemma that has kept popular child-rearing experts in business, and has tied parents in knots, for a hundred years. In the mysterious process of shaping personalities in and for a fast-changing modern society of group allegiances, how much power can parents at home really have? This is hardly the heretical question that Harris and her critics and her champions seem to think. It has a long historical pedigree. To understand just what seeds of sedition Harris might be sowing, it helps to understand the tradition of confusion behind her.

A century before Harris arrived on the scene, parents were on the defensive, worried that the family's socializing clout was waning in urban, industrial, mobile America. Where parents had once reared their children to follow in their footsteps, the daunting challenge had become to prepare children "to grapple successfully with the complex conditions and varied responsibilities which will be their lot," in the words of Dr. Luther Emmett Holt, the pioneering expert of the 1890s. No wonder grandmothers were told to sit quietly in their rockers. What could they possibly know about such things? But parents themselves were less than secure about their own credentials for the job. In an era that extolled the unknown wonders of the future, they were acutely aware of being the voice of the past or, at best, the present.

The experts tooling up in the numerous psychology labs across American university campuses did not have to plant doubt in parental minds. After all, adults inhabited the jangly modern world, and they daily inhaled, along with new industrial fumes, two mutually reinforcing fears: there was so much more that children needed to be prepared for (and protected against), and parents were so much less prepared to help them. The experts quickly acquired prominence because they told parents what they wanted to hear. Precisely because the world was full of so many competing pressures and lures, the various gurus agreed, parents had become more crucial than ever in guiding their children's growth. The right by-the-hearth intimacy guaranteed that a child would get ahead in a heartless world. The best by-the-book intimacy also guaranteed that a mother need not get left behind the scientific times, nor go along with the soul-deadening times. Busy at home, expertly equipping her child for the world, she was up-to-date. The nurture assumption, in short, was the ungainly brainchild of nurture panic.

After World War I, parents and experts could not help focusing more intently on what it was in the world that seemed most threatening, and most alluring: a "new generation" of gangly youths who took so long to grow up. In an emerging meritocratic system, where the middle-class route increasingly led through high school and on to college, adolescent peer groups had become a cultural force to contend with. Americans were fixated on youth. They were fascinated and appalled by the cohort that seemed to challenge so many pieties of the past. Sociologists and psychologists hurriedly got to work on socialization theories that could take account of this new element in family and social life--without letting it take over. The nurture assumption, unsettled by the peer principle, went through a decidedly awkward growth spurt.

Perhaps the best way to summarize the convoluted parental wisdom that emerged is: beat 'em by joining 'em. The peer-ification of the family proceeded apace, as Paula Fass has shown in The Damned and the Beautiful, a history of "American Youth in the 1920s." The "affectionate" modern family of expert lore, which had already elevated the wife to loving equal of her husband and expert guide of her children, regrouped on the level of supremely compatible, adaptable companions. Parents were becoming pals.

Depending on whom you listened to, these peer-parents were meant to cooperate with, or to complement, or to compete with, or to co-opt the crowd of youths knocking at the door, ready to turn mama's boy into the suitably conforming--yet also suitably competitive--personality for the next generation. One mother captured the ambivalent ethos with comic poignancy in Middletown, Robert and Helen Lynd's "Study in Contemporary American Culture," in 1929. "I put on roller skates with the boys and pass a football with them," she earnestly explained to the sociologists. "My mother back East thinks it's scandalous, but I tell her I don't think anything very bad can happen to boys when they're there with their mother and father."

Seventy years later, Harris still thinks that such a parental approach, while not scandalous, is pretty preposterous and certainly pointless as a personality-shaping ploy. (She would also note that it is lacking in the spirit of spontaneity that might make roller skating with mother much fun at all.) By contrast, many of the rest of us-- the by-the-book super-moms and those "self-absorbed" specimens of parenthood, too--have raised the ideal of the peer-parent to new heights. If we are home, squatting down to discuss the bad feelings that just erupted in a sibling brawl, we are self-consciously taking our children seriously as equals, and "modeling mediation skills" without which their lives at the playground might be a social disaster--and their future careers in jeopardy.

Or if we are out playing in our silly t-shirts, or working at our gratifying jobs, we see it as closing the gap, too. Don't parents, just like children, need creative outlets? Plus we are being role models for life in an on-the-go world. Whichever spot we find ourselves in, though, we are usually feeling nagged by anxiety and lacking in authority. We may even feel somewhat ridiculous, but we tell ourselves that our narcissistic, companionate nurturing is surely not pointless in the long run--even as we panic that something very bad might indeed happen to boys and girls. We panic, in fact, more than ever before.

Harris thinks that it is high time to do what parents and experts have spent a century studiously talking about doing for children and failing to do for themselves: draw lines, set limits, and adjust expectations about the relations between adults and children through life. She does not settle the question of who rules--peers or parents--nearly as starkly, or as scientifically, as she aims to. But why should she, and how could she? Harris's own zeal to replace one exaggerated determinism with another is strangely at odds with what seems, on a close reading of her book, her true mission. For the underlying message of The Nurture Assumption is not that parents should give up, it is that parents should grow up--which means acknowledging that much of our children's fates eludes anyone's control but their own.

Beyond the genes they pass on, Harris maintains, parents do not have as much power over their children's futures as they hope and fear. This is not a particularly novel idea. And what Harris trumpets as her truly original position is also quite familiar: the unsettling notion that peers have inordinate sway has been around for a long time. Still, Harris does have a jarring idea for our times, which you would never guess from her press, or from her penchant for pronouncements such as "parents count zilch." Not altogether wittingly, but quite wisely, she ends up making a case that what influence parents do have with their children is perhaps more than they (or the experts) actually realize, and that they have it by virtue of not being their children's peers. They have it, rather, by virtue of being adults, with their own peers. It would be a pity, though hardly unprecedented in the dizzying advice industry of the so-called "century of the child," if this piece of mature realism were overwhelmed by Harris's own extremism, and by good old American alarmism.


Harris draws firm lines and sets up high expectations for herself. First, she wants to show that there has never been anything like definitive scientific evidence for the nurture assumption, and by doing so, to disprove it. Second, she proposes to prove that her peer principle can take over as the omnipotent--and scientific--explanation of why children turn out the way they do. Such an audacious agenda inspires rigor. It also sets her up for trouble. After all, as she herself admits, "child rearing is not physics." Harris very soon exposes the limits of her enterprise, and this turns out to be the most instructive lesson of all. Science cannot begin to guide us definitively in raising children, but perhaps it can goad us into seeing how much might be--and just as important, how much might not be--up to us and our own best guesswork.

Harris's case against the nurture assumption reveals what has been obvious from its inception: the myth's strength has never been that it rests on scientific certainty, but that it has been so resilient in the face of scientific (and social) uncertainty. Thus it is hopeless to try to demolish the nurture assumption by exposing the dubious data behind it. The best you can do is deploy the motley evidence to define parental influence more realistically--and this ends up being Harris's accomplishment. Nor can Harris prove that peer power is as decisive as she claims. But she doesn't need to: by failing to demonstrate peer determinism at work, she provokes a clarity about peer clout that has been lacking for decades.

Much of the case that Harris builds for the scientific shakiness of the nurture assumption has been laid out by academic specialists themselves. Already in the 1920s, when child development first emerged as a field in psychology, worries about making and keeping it "scientifically respectable" were much on researchers' minds. Social philanthropists were pounding on laboratory doors, eager for data to support the crusade of "parent education" that was sweeping America. Psychologists were apprehensive about rushing their research on children's growth and parental influence. They condescended to the brash colleague who had no such qualms: John B. Watson, the founder of behaviorism, who boldly extrapolated from a few colorful lab experiments (among them, conditioning a stolid eleven-month-old named Albert to be terrified of a white rat) to highly popular lessons for parents about their formidable power to mold fearless personalities. (And even his avid lay audience was dubious about just how deep his science went. "I swallow Watson whole because I want to believe I can do something about fears," one mother confessed to her parents' group, "not because I am convinced that he is sound.")

In 1927, the mood in the headquarters of child development research was anything but hubristic about the promise of science to equip parents and social planners with the secrets of successful childrearing. The report of a meeting of the Child Development Committee of the influential National Research Council sounded a scientific alarm that has been echoed since. "The fruits of present research, it is to be feared, are not sufficient to maintain in scientific health the rapidly growing movement," the secretary warned. The "very success of the childward movement" held a "grave danger," which was that the impatience of parents would overwhelm the patience of scientists,

that the demand from various quarters--legislative, humanitarian, educational, journalistic--will so far exceed the supply provided by the slow processes of research that the movement will escape the bounds of fact and wander off into the alluring jungle of easy generalization and over-confident dogmatism. Science is asked to point the way where as yet there is no way--but the movement proceeds nonetheless.

The committee's recommendation? "Illimitable research": not to confirm the crusaders' easy dogmas about more perfect parental and social control, but to put complicated data in their way. The secretary emphasized just how difficult the necessary studies were going to be to design. The limitations of the "artificial experimental set-up" (Watson's bailiwick) must be overcome through work in "the normal child environment." It was a daunting task. So much time was required for home investigations, and so many mundane variables had to be accounted for--"the slamming of a door, the sudden thunder shower, one angry temper display of a parent."

The research has indeed gone on and on, doggedly, awkwardly, ingeniously, inadequately attempting to overcome the obstacles posed by specimens who are, alas, neither as quiet nor as manipulable as mice. In the seventy years since the Child Development Committee's meeting, academic researchers have become only more aware of how often the mysteries of child-rearing and development elude the mastery of science. How do you create a control group of squirmy children and parents? Why should you believe the observations and the memories of family members? Who on earth has a consistent "parenting style," the same for all children or even for one child all the time? Can you trust personality studies, given the notorious problem of context--the inconvenient fact that tests of the same person turn out very different depending on the setting, among other things? And what can you really say about results when, as Harris says, "correlations come with no built-in arrows to distinguish causes from effects"?

You cannot say much--or you can say anything, at least when the data concern hot child-rearing topics such as bonding and discipline and personality development, and are inevitably distorted on their way to the public. This revelation has led plenty of scientists themselves (hardly card-carrying social constructionists) to caution that "child development is a product of social needs that had little to do with science qua science." That was the verdict of Robert Sears, a prominent figure in the field who spent much of the 1940s and 1950s trying to test psychoanalytic hypotheses about the effects of parents' child-rearing practices (with very disappointing results). William Kessen, a revered dean in child development who trained an influential generation of psychologists at Yale, described child psychology as "a peculiar cultural invention that moves with the tidal sweeps of the larger culture in ways we understand at best dimly and often ignore." Much of what experts have preached to parents, he emphasized, has portrayed research as being far more solid than it is. "In the 1950s and 1960s," Kessen confessed, "we oversold ourselves or were oversold ... in all the child fields, but certainly in areas of cognitive and social development, and exquisitely in the field called learning."

Jerome Kagan, an eminent psychologist at Harvard, has been perhaps the most outspoken skeptic about the scientific underpinnings of the pieties of parenting. He has pointed out that our fondest beliefs about child-rearing do not rest on proofs, but on fears and hopes--which also means, as he cautions in his new book, that they are not really susceptible to disproof either. In Three Seductive Ideas, he holds up a central tenet of the nurture assumption for critical scrutiny: "the allure of infant determinism," or the conviction that experiences in infancy and toddlerhood make the man or woman.

The idea that the die is cast by age three or so, during the years of intensive parental care (in our society at any rate), has been peddled as scientific gospel by psychoanalysts, developmentalists, and behaviorists alike since the 1920s. It is as vigorous as ever in the 1990s, thanks not least to Hillary Clinton's enthusiasm for it. In an "immature" field such as psychology that rests on "weak evidence," Kagan suggests, such an idea can thrive even though it is not readily demonstrable, because it is not readily deniable either. Meanwhile, it conforms cozily to the Western faith in personal responsibility and in an unbroken continuity between past and present. (Kagan also notes another source of its popularity: telling mothers to kiss or not to kiss their babies is cheaper than overhauling social policies.)

The reason that the infant-determinist credo cannot be rocked right out of its cradle is more than methodological. It is definitional. Think of the obvious difficulties in getting "robust" measurements of any trait over the long haul from infancy to adulthood. Then consider how much room there is for disagreement about what traits ought to be measured in the first place. Kagan asks about "the derivatives of early experience that should be given the highest priority. Are they specific cognitive abilities, psychiatric symptoms, fame, wealth, a good marriage, parenting success, a satisfying sex life, capacity for love, meaningful work, loyal friends, or simply a subjective feeling of well-being?" This woolliness makes experimentation a nightmare, and promises a field day for cheap nostrums and popular prescriptions. "The lack of accord among clinicians, humanists, or scientists as to which quality--or qualities--has the tightest link to specific early experiences," Kagan observes, "makes either refuting the determinist's claim or proving it correct impossible." IV.

So the nurture assumption has thrived despite the indeterminacy of data. If anything, proclamations of parent power have grown stronger as the scientific evidence for them has become murkier. And for many decades now, you haven't had to wear a lab coat to be aware of the inconsistencies and the confusions. Any parent who follows the popularized accounts of the latest research is uneasily aware that findings have a way of shifting frequently, even on the most basic physical questions. Should a baby be put to sleep on her stomach, her side, or her back to ward off the danger of Sudden Infant Death Syndrome? Babyboomer parents have heard it all. On psychological matters, who is surprised when a headline in the Science Times proclaims, "Re-evaluating Significance of Baby's Bond with Mother"? As Dr. Spock acknowledged in his second edition, science has been a less than confidence-inspiring addition to parenting.

Harris's contribution is to impose lucid shape on evidence the very inconclusiveness of which is by now enough to make conscientious parents even more persuaded that pretty much everything they do, or do not do, is likely to make some difference (or not enough difference) in the way their children turn out--but don't ask them what difference exactly. (After all, who knows what tidbit of unnerving, if also indeterminate, evidence is likely to turn up next?) Harris acknowledges her debt to academics with "long and distinguished careers" in the child development field who have lately bridged the newer "nature" camp of behavioral geneticists and the traditional "nurture" camp of socialization researchers, taking skeptical stock of the cumulative findings. They have concluded that very little of the socialization researchers' work thus far holds up well, mainly because it has failed to appreciate the role of heredity. They have also suggested that behavioral geneticists still have a long way to go to take subtle enough account of what happens in the intricate environment of the family.

Mixing logic-chopping rigor and wise-cracking humor, Harris turns these academic overviews and her own sleuthing into a brisk tour of controversial data collection and interpretation--a "Stats and Spats" course aimed at the attentive non-scientist. She deftly leads her readers through the inadequacies of socialization research, which has relied on studies of one child and failed to recognize that parents provide not just a home environment for their children, but genes as well. Thus it is impossible to be sure whether the correlations that do show up between parents and children are the result of conscious parental shaping, as opposed to mere evidence of heredity at work. Moreover, it is impossible to be sure what is cause and what is effect. Does the temperamentally jolly baby shape the engaged parenting style, or vice versa? Certainly two very different children can be shown to elicit very different kinds of behavior from the same parent.

Harris is much more impressed by the findings of behavioral geneticists. Relying on adoption and twin studies, they have set out to tackle the slippery problem of separating hereditary and environmental effects on personality traits. Their results have been usefully disconcerting. Adopted children do not resemble their adoptive parents or adoptive siblings, even when they have shared a home since birth. Identical twins reared apart turn out to have a great deal in common. Conversely, identical twins reared together--sharing the same genes and the same home environment--turn out to be quite different. In fact, "they are no more alike than the ones reared in different homes!"

By the late 1970s, behavioral geneticists had demonstrated, not all that surprisingly (or subtly), that heredity and environment each contributed about half to explaining the variations among people. But there was also an unexpected result: siblings who shared genes and a home did not seem to be alike enough. Their "shared environment" appeared, if anything, to be making them less, rather than more, similar. Both the "nature" camp and the "nurture" camp had thus converged on a version of the same conundrum: insofar as a child's home life played a part in shaping his or her future personality, it did not seem to be in any predictable way that could be shown, through scientific studies, to reflect a parent's particular style of rearing.

Here Harris's patience grows thin. In the absence of firm data, she sees only dogma at work; and she is eager to impose dogma of her own. She is ready to denounce as a totally empty promise the assumption that parents have any impact. Her survey of both sides' undaunted efforts to delve more deeply into the recesses of the home certainly demonstrates what an invitation to scientific frustration the nurture assumption is: their results have been mixed at best. Still, what would truly seem surprising, and even suspect, would be if researchers had been able to distill out some pure extract of parental power, utterly untainted by genetic factors or by "child-driven" effects. How much would it really tell us anyway if they could, given that no parent operates in a vacuum? As a contributor to an anthology called The New Generation: The Intimate Problems of Modern Parents and Children, which appeared in 1930, could have reminded Harris, "clarity does not begin at home."

Harris's criticisms in fact point up how much modesty prevails in the laboratory. Attachment theorists, for instance, now take pains to acknowledge that an infant's temperament, not just a parent's affect, plays a large part in forging the bond between them. Nor do they argue any longer that that bond is a "working model" for all the other relationships in a baby's life. Finally, they acknowledge how evanescent future traces of infant attachment can be. Kagan cites examples of their more hedged claims. "Early experience can not be more important than later experience," a prominent researcher has admitted, "and life in a changing environment should alter the quality of a child's adaptation." It would be hard to get more attenuated than this: "What is incorporated from the caregiving experiences are not specific behavioral features, but the quality and patterning of relationships, mediated by affect." Harris is quite right to emphasize that this sort of scientific mumbo jumbo is a far cry from parental determinism; but it certainly is not proof of parental powerlessness either.

Harris is also impatient with the attempts of behavioral geneticists to probe the ways in which the family can operate as a "non-shared environment" that helps to differentiate the hereditarily similar. They have pursued the question of how "feedback loops" get started in which small genetic differences between children trigger much larger environmental differences for them. Add siblings to the soup, and figuring out the recipe behind the resulting flavors becomes a truly daunting challenge. Birth-order research, which Harris is not the first to survey skeptically, is murky. The existence of a "first-born personality" hardly seems as clear as proponents (most recently Frank Sulloway) claim, but that is partly--as her appendix proves--because measurements are tricky to compile and to compare. It is also because personality is so tricky to define. Kagan wagers that "no present concept in personality will survive the next half century."

Other hypotheses about sibling effects are "difficult to test," Harris herself admits in the article that led to her book. Perhaps that jolly baby enjoys a more responsive mother than her fussy older sister did thanks not just to her own smiley ways. Perhaps she owes it also to her cranky predecessor, who primed their mother to greet a Gerber specimen with special eagerness. It sounds plausible, but researchers who tried to tackle this one concluded that the cause-or-effect problem "is a notoriously intractable issue." Harris proceeds as though such methodological problems were in themselves decisive proof of conceptual problems. In effect she concludes that "the hypothesis that children's within-family micro-environments play a causal role in the shaping of their personalities"--because it has proved so inhospitable to good testing--has been disproved.

These twists and turns of recent research look like conscientious (if sometimes tortuous) science on an admittedly elusive mission. To Harris, however, they look like a tendentious refusal to relinquish a myth, the myth that when we say "environment," all we can mean is parents at home. Bring in another category--the non-shared environment outside the home, where peers and group behavior operate--and suddenly, she promises, all the qualifications and inconsistencies are no longer necessary.

Harris is right that figuring in peer influence makes a lot more sense than ransacking the house for a hidden key that, even if it could be found, cannot possibly open all the locks of an adult's personality. But she is hardly the first to propose looking farther afield, or to find peers out there in the driver's seat. Moreover, her discovery does not mean the end of wrangling over keys. Kagan points to a more heterogeneous array of competing influences on character--sibling relationships and "emotional identification with family, class and ethnic group," along with major historical events (such as generational turmoil). He warns of how unpredictably they mingle. Harris's peer theory, too, cannot begin to deliver on its determinist promise. It is tempting to say to them both, so what else is new?


Yet inside the peer theory that Harris hypes is a lucidly modest idea struggling to get out--an idea at odds with her own claims about peer supremacy, and out of step with the prevailing pal-parent anxiety. The idea is that parents are their own peers, not their children's, and that they have plenty of power, thank you, without aspiring to be held accountable for every turn in their offspring's fates. As family traditionalists condemn Harris and the bustling believers in the post-nuclear family welcome her news, and as she and the child development experts play "my study tops yours," nobody seems to grasp the old-fashioned thrust of this new guru's arguments, including perhaps the new guru herself.

Harris has usefully shown what many researchers, and most parents in their heart of hearts, know: that it is next to impossible to prove that parents' conscious, scientifically measurable childrearing behavior at home has a predictable, scientifically measurable impact on their children's social futures (beyond, that is, what can be ascribed to the genetic bond between them). Still, as she proceeds to point out, that leaves parents plenty of room (and responsibility) to be forceful (which doesn't mean fierce) influences on their children as they are growing up. This is a fact that other experts have helped us to forget.

By the same token, as Harris of all skeptics should have predicted, her own evidence for the peer assumption turns out to be just as shaky as data for the nurture assumption. And that should be reassuring news to parents. Moreover, they might be heartened to realize that Harris grants them, as a peer group of adults, cultural power to shape indirectly the peer culture that their children soak up from the minute they sit down to their cereal. These are the unscientific but significant lessons that have been lost in all the commotion around this book, and in Harris's own promotion of it. Parents have clout that, in their childish dreams of greater control, they and their experts have underrated--and undercut.

Harris's "group socialization theory," as she calls her peer principle, may be just the jolt that parents need to get a grip on themselves. Its weaknesses as a monocausal account of the molding of personality quickly become apparent, but Harris deserves real credit for trying to describe carefully where others have been content to prescribe wildly. Her theory represents a rare effort to explain how children actually interact with their peers. The usual focus in the long history of American panic about "youth" has been on the overreaction of parents to those peers. Decade after decade, experts and parents have been in a dither about how "the place of the home tends to recede before a combination of other formative influences," children's peers being the main one, as the Lynds reported of Middletown in the late 1920s and 1930s. That particular mother was out wobbling on roller skates with the kids--rather than inside teaching them, say, bridge--because she was on a mission to ensure that her children conformed to the "approved ways of the group." Which meant that she had to do her bit to go along with those ways, too. This was, at any rate, what prevailing advice suggested she do--if she understood it correctly. As another Middletown resident confessed, the newly emerging wisdom for parents in a peer world was thoroughly bewildering. "I try to do just what you say," she told the Lynds, "but I am a nervous wreck just trying to stay calm."

It was, understandably, hard to tell how peers were influencing children's behavior in the emerging consumer culture of postwar America. A flood of defensive prescriptions about the ways in which the existence of those peers should be influencing parents' behavior did not tend to clarify matters--except, perhaps, by implying that adults were suddenly in much the same boat as children. Like the offspring whom they sent out into a newly bureaucratized society, parents were buffeted by conflicting norms for home life and outside-the-home life, and they felt under new pressure to conform. Some experts, such as Watson, urged them to be harder and cooler at home, and "treat [children] as though they were young adults" to firm them up to prosper in a flexible world. Gentler helping professionals urged them to become softer themselves, "grown up children" communing warmly with their children before sending them into a harsh world. A new catch-all ideal, for child-rearer and child alike, arose to finesse all the tensions: "the adjusted personality."

Harris bluntly challenges the strained harmony. She argues that learning to distinguish the differences between children and adults, between the home and the world, between "relational" behavior and group behavior, is the essence of socialization--which she, like her predecessors, equates with personality development. In place of the adjusted or "wholesome" personality, she recognizes what might be called, in the biological terms she favors, the adaptable personality. She divides personality into "two components," an "inborn" part that "goes with you wherever you go; it influences, to some extent, your behavior in every context," and an "environmental" part that is "specific to the context in which you acquired it" and thus changes, depending on where you are. She emphasizes two contexts in particular: the home, where relational experiences happen, and the peer group, where the child struggles to assimilate to--and differentiate herself from--others. (The brain, conveniently, also has two "departments": one for those one-to-one "dyadic" relations, and another for social interactions involving more people.) Harris stresses what every parent soon notes, that there can be a jarring disjunction between a child's "home" self and "social" self. (The first school conference is usually a parent's awakening. "What a cooperative child," the teacher exclaims, and you can barely keep from falling out of your chair.)

Harris has certainly learned her discriminating lessons well! Her zealous categorizing points up not just the unscientific fuzziness of so much expertise, but also the perils of over-schematizing personality development. Thus she sets up her own tidy theory for a fall. Children need intimacy with their parents and imitate them a lot, Harris maintains, but that "relational" personality only operates at home and doesn't count in toting up how a person "turns out." (In a similar spirit, she discounts the long-term influence of one-on-one intimacy outside the family, invoking just the kind of dubious studies you would think she would scoff at. Some research has evidently shown that grade-school friendships--or the lack of them--have no effect on "overall life status adjustment," whatever in the world that might be.) "The public personality ... the one that a child adopts when he or she is not at home" is the one that we carry with us into the world for life. And it, Harris claims, is shaped in the group maneuvering among peers, starting very early; and it counts for everything.

Surely her formulas are too formulaic. The public personality, almost by definition, is the one that gets submitted to the psychologists for measurements of social competence, but that hardly makes it the sum total of the "real" self. In her eagerness to give peers supreme control, Harris has manufactured an argument that verges on tautology. Indeed, she provides her own best criticism: "the human mind wants to categorize ... even when [things] fall along a continuum and not into convenient clumps." She also supplies ample evidence that muddies her extreme case for peer clout, and that helps to clarify the true contours of parental sway. (The really valuable revelations of this book are not what she touts or apparently thinks.) Harris does not spend much time discussing "home" nurture. But what she says is meant as a reminder to parents that they are not the "wimps" or the "wallpaper," in her words, that they often fear they are. They have an immediate, daily, "dyadic" power that they overlook--and can undermine--in trying too hard to be their children's equals and in worrying about how they might be influencing their social futures.

The unfashionable and useful lesson of Harris's book is that parents are not their children's peers at home. Sons and daughters depend every day on their greater power and experience, even if they are not defined for all time by the help and the guidance that their parents supply. What parents do with them and for them shapes a one-of-a-kind relationship, not a made-to-order social personality. More than a little of what parents impose and model at home will be overridden by their offspring's peers--but a lot will not be overridden. "Anything learned at home and kept at home--not scrutinized by the peer group--may be passed on from parents to their kids," Harris says. "Maybe even how to run a home." That sounds like child-rearing to me. To be sure, it is a rare parent who merely imitates her own parents' ways. Perhaps more often than not, mothers and fathers struggle to still those "home" voices in their heads--but that, too, is proof of how much lingers and how deeply.

Harris suggests in passing that religion, too, can be a durable legacy from home. On "values," she is vague and unconvincing--but then, who isn't? As Robert Wright has pointed out, Harris misreads what little evidence she considers, claiming that a classic study from the 1920s found no correlations between cheating and honesty inside and outside the home (at least during childhood: the experiment didn't even try to measure correlations between youth and adulthood). In fact, the study did yield some correlations--and a recent re-analysis of the data has found even more, for what it's worth. Genes, too, are passed on, of course, and they shape that "inborn component" of personality. All this is transmitted--and Harris says all this doesn't count. Yet add it up, and parents can hardly complain about being made to look marginal.

And Harris does not stop here. She slams the front door on parents, keeping them a housebound force, only to open the back door and let them out again. Her theory has a big loophole that restores parents as a socializing influence with power outside the home--not as individuals molding their own child, but as a peer group transmitting culture to their children's peer group. "The easiest way to tell who socialized a child--who gave the child her culture--is by listening to her," Harris sums up her theory, drawing on the linguistic evidence that serves as her strongest proof. "She got her language and accent in the same place where she got the other aspects of her culture. She got them from the children's peer group, which--in most cases but not all--got them from the parents' peer group." Deaf children of hearing parents and second-generation immigrants are the exceptions that prove her rule that the peer group comes first in the chain of influence: they end up talking like their fellow kids, not like their parents.

Yet Harris's evidence, much like the "proof" for the nurture assumption that she has ruthlessly scrutinized, actually makes a powerful case for the indeterminacy of influence under ordinary circumstances. Language acquisition is not the same as personality development, which is presumably a more amorphous process. The culture of the deaf is not ordinary culture, and the power of peers among such children is presumably abnormal. In more typical situations, Harris readily acknowledges, it is next to impossible for a scientist, much less a child, to filter out parent culture in our permeable world. And peer culture can be hard to pin down.

Consider a small study that Harris cites as a rare example of the kind of test that her theory requires. Students were interviewed about their school attitudes before and after they switched cliques, and it turned out that they were conformist chameleons. Kids switched cliques too fast, the researcher cleverly figured in designing the research, for any changes in attitude in the children to reflect their parents' views. Voila: pure peer shaping at work! But precisely because children--especially in middle childhood, when Harris says peer molding is at its peak--do move among groups so quickly, and because the groups themselves so often melt and shift, how could personality be as decisively shaped by the process as she suggests, "children ... typecast into roles that might last them the rest of their lives"?

How, at any rate, could studies ever decisively prove it? By means of the media, cultural influences pervade all different groups. And as Harris herself emphasizes, membership in a peer group is not just about conforming to its norms, it is also about distinguishing oneself within it. So, for example, even if those students had not turned out to be chameleons, their views could be explained as peer-influenced anyway. Talk about slippery studies that could prove anything, and nothing. Harris's own daughters, whom she holds up as proof of parental powerlessness, work just as well as disproof of lock-step peer determinism. They hung out in similar elementary school crowds, and one proceeded to be a model high school student (who needed no guidance), the other a "burnout" (who heeded no counsel)--and both have turned out fine.

Moreover, as Harris herself shows, children (unlike scientists) are not focused on ingeniously trying to filter out parent culture. Their main order of business is trying to fit into their peer culture. Sure, they like to distinguish themselves as a group from adults (as from other cliques of kids), but until adolescence they do not have to work very hard at that: children are different from grown-ups, and they know it. They are also savvy imitators and creatures of habit. They bring their home "culture with them to the peer group, but they do it carefully, tentatively," eager to conform and wary about any weirdness they might betray, yet also primed to stick with what they know.

In short, children are not plotting a cultural revolution. They are committee members ready to compromise. What children discover they hold in common with other children, Harris emphasizes, they keep. Thus the more cohesive the parental culture--the more that the parents of a child's group of friends hold in common--the more will be passed on to the children's culture. Parents can cater to an adolescent common denominator in the attitudes and the interests that they cultivate or tolerate. The Disneyfied, commercialized, sexualized, media-dominated world of parental nightmare is not some deracinated "peer" realm. (Toys, and plenty else, R all of us. They are part of the vast cultural equipment that adults transmit--or try to resist transmitting--as parents, and that is conceived, manufactured, and marketed by adults.) Or parents can get in touch with their inner-adults and decide what standards, what personal and social qualities, they think are important, not just for their children but for themselves, in concert with the adults around them. Children won't buy all of it anyway--where would civilization be if they didn't bite back? Still, they will absorb plenty.

Harris's logic of group norms knows no ideological bias. It can justify gated communities as havens of carefully chosen parental homogeneity. It also helps to explain gangs, and might inspire some hope about them. Where blighted early family life has often been blamed for neighborhoods under siege, concerted effort to change peer culture, long before adolescence, might help. Harris would be the last to suggest that it would be easy, but probing for the secrets of children's culture rather than for the secret of parent nurture certainly looks like a sensible place to start. Still, in the end, a change of peer culture is really nothing less than a change in culture generally. It may be that cultural criticism is one of the requirements of good parenting. After all, one of the many rubrics under which American children are reared is America.

Amid millennial alarm that, in the words of Kiku Adatto, the director of Children's Studies at Harvard, "all the barriers between childhood and adulthood are breaking down," Harris's book--if you actually read it, and between its lines--strikes a rare note of reasonable calm. She is in league with children, demonstrating what it really means to pay respect to their autonomy. She is also on the side of parents, defending their maturity and reminding them of what that really means. She cannot quite decide about the experts--she wants to beat 'em and to join 'em--but she is not about to be bullied by them. If The Nurture Assumption becomes a benchmark in a century of childrearing lore, it should be for the right reason. Harris's book is valuable not because it stirs up a version of the same old, defensive controversy about parent power, but because it encourages parents to step back from determinist dogma--and from themselves. Understandably unsure of our own authority, we have freighted our hopes for our children with fears about our own power. A quest for harmony between generations has turned into a crusade for unity, equality, and ever greater control. And we have been surprised to find ourselves lost in anxiety instead.

In truth, this message, too, has been around before. Back in the 1920s, in an article called "Progressive Parents--Their Tragedy," a writer in The New Republic was struck by widespread parental unease. Conscientious companions to their children from cradle through college and beyond, mothers and fathers had found their reward was a rude surprise: a generation with mores and aspirations quite at odds with their own. This writer knew where to turn for the right seasoned perspective. Not to the experts with their primers and promises. "Lucky are we," she concluded instead,

if we attain the perspective of one grandmother who said to her daughter: "Always remember that mothers aren't important. They don't amount to much. They give their children life, keep them clothed and nourished, teach them manners and the fundamental moral things we have to have to get along at all. But the big things that count we do not give. What we give they keep forever, but they are unconscious of it, it is not a live force. The inspirations that will grip and shake them must come new and fresh. The ideas for which they will give their lives will not be ours. They will get them from teachers, books, friends, from those they fall in love with, not from us." Humility, unlike clarity, does begin at home.

(Copyright 1998, The New Republic)

Table of Contents:
The New Republic
The New Republic
to The New Republic
the Archives
to The New Republic