Children of immigrants soak up not just the language of their adopted homeland but the culture as well. For their entire lives, my shtetl-born grandparents were strangers in a strange land. Cars, banks, doctors, schools, and the urban concept of time left them baffled, and if the term “dysfunctional family” had been around in the 1930s and 1940s it would surely have applied to them. Nevertheless, my father, growing up in a community of immigrants who had arrived in different decades, gravitated to other children and families who knew the ropes, and ended up happy and successful. Such stories are common in chronicles of the immigrant experience.52 So why do we insist that children’s parents are the key to how they turn out?
Studies also confirm what every parent knows but what no one bothers to reconcile with theories of child development: that whether adolescents smoke, get into scrapes with the law, or commit serious crimes depends far more on what their peers do than on what their parents do.53 Harris comments on a popular theory that children become delinquents to achieve “mature status,” that is, adult power and privilege: “If teenagers wanted to be like adults they wouldn’t be shoplifting nailpolish from drugstores or hanging off overpasses to spray I LOVE YOU LIsA on the arch. If they really aspired to ‘mature status’ they would be doing boring adult things like sorting the laundry and figuring out their income taxes.”54
Even the rare finding of an effect of the shared environment, and the equally elusive finding of an interaction between genes and the environment, emerge only when we substitute peers for parents in the “environment” part of the equation. Children who grow up in the same home tend to resemble each other in their vulnerability to delinquency, regardless of how closely related they are. But that similarity only holds if they are close in age and spend time together outside the home—which suggests they belong to the same peer group.55 And in a large Danish adoption study, the biological children of convicts were somewhat more likely to get into trouble than the biological children of law-abiding citizens, which suggests a small across-the-board effect of the genes. But the susceptibility to crime was multiplied if they were adopted by parents who were criminals themselves and who lived in a large city, which suggests that the genetically at-risk children grew up in a high-crime neighborhood.56
It’s not that parents “don’t matter.” In many ways parents matter a great deal. For most of human existence, the most important thing parents did for their children was keep them alive. Parents can certainly harm their children by abusing or neglecting them. Children appear to need some kind of nurturing figure in their early years, though it needn’t be a parent, and possibly not even an adult: young orphans and refugees often turn out relatively well if they had the comfort of other children, even if they had no parents or other adults around them.57 (This does not mean that the children were happy, but contrary to popular belief, unhappy children do not necessarily turn into dysfunctional adults.) Parents select an environment for their children and thereby select a peer group. They provide their children with skills and knowledge, such as reading and playing a musical instrument. And they certainly may affect their children’s behavior in the home, just as any powerful people can affect behavior within their fiefdom. But parents’ behavior does not seem to shape their children’s intelligence or personality over the long term. Upon hearing this, many people ask, “So you’re saying it doesn’t matter how I treat my child?” It is a revealing question, and I will consider it at the end of the chapter. But first, the public reaction to Harris’s theory, and my own assessment.
The Nurture Assumption was, by any standard, a major contribution to modern intellectual life. Though the main idea is at first counterintuitive, the book has the ring of truth, with real children running through it, not compliant little theoretical constructs that no one ever meets in real life. Harris backed up her hypothesis with voluminous data from many fields, interpreted with a keen analytical eye, and with a rarity in the social sciences: proposals for new empirical tests that might falsify it. The book also contains original policy suggestions on tough problems for which we sorely need new ideas, such as failing schools, teenage smoking, and juvenile delinquency. Even if major parts turn out to be wrong, the book forces one to think about childhood, and therefore what makes us what we are, in a fresh and insightful way.
So what was the public reaction? The first popular presentation of the theory was in a few pages of my book How the Mind Works, in which I presented the research behind the three laws of behavioral genetics and Harris’s 1995 paper explaining them. Many reviews singled out those pages for discussion, such as the following analysis by Margaret Wertheim:
Never in my fifteen years as a science writer have I seen the subject I love so dearly abused so greatly…. What is so appalling here—quite aside from the laughable grasp of family dynamics—is the misrepresentation of science. Science can never prove what percentage of personality is caused by upbringing…. By suggesting that it can and does, he invites us to see scientists as at best naïve and at worst fascistic. It is precisely this kind of claim that, in my opinion, is giving science a bad name and is helping to fuel a significant backlash against it.58
Wertheim, of course, confused “the percentage of personality that is caused by upbringing,” which is indeed meaningless, with the percentage of variance in personality that is caused by variation in upbringing, which behavioral geneticists study all the time. And scientists can show, and have shown, that siblings are as similar when reared apart as when reared together and that adoptive siblings are not similar at all, which means that the conventional wisdom about “family dynamics” is simply wrong.
Wertheim is sympathetic to radical science and social constructionism. Her reaction is a sign of how behavioral genetics—and Harris’s theory, which aims to explain its findings—touches a nerve on the political left, with its traditional emphasis on the malleability of children. The psychologist Oliver James wrote, “Harris’s book can be safely ignored as yet another application of Friedmanite economics to the social realm” (an allusion to the economist who, according to James, stands for the idea that individuals should assume responsibility for their own lives). He suggested that Harris was downplaying research on parenting because it “would indirectly pose a real challenge to the theories of advanced consumer capitalism: if what parents do is critical, it calls into question the low priority given to it, compared with the pursuit of profit.”59 Actually, this fanciful diagnosis has it backwards. The most vehement propagandists for the importance of parents are the beer and tobacco companies, which sponsor ad campaigns such as “Family Talk About Drinking” and “Parents Should Talk to Kids About Not Smoking.” (A sample ad: “Daughter speaks to the camera, as if it were her mother, reassuring her that her words about not smoking are with her, even when her mother is not with her.”)60 By putting the onus on parents to keep teens sober and smoke-free, these advanced consumer capitalists can divert attention from their own massive influence on adolescent peer culture.
In any case, Harris drew even more venom from the political right. The columnist John Leo called her theory “stupid,” ridiculed her lack of a Ph.D. and a university affiliation, and compared her to deniers of the Holocaust. He ended his column, “It’s not time to celebrate a foolish book that justifies self-absorption and makes non-parenting a respectable, mainstream activity.”61
Why do conservatives hate the theory too? An axiom of the contemporary American right is that the traditional family is under assault from feminists, a licentious popular culture, and left-wing social analysts. The root of social ills, conservatives believe, is the failure of parents to teach their children discipline and values, a failure that can be traced to working mothers, absent fathers, easy divorce, and a welfare system that rewards young women for having babies out of wedlock. When the unmarried sitcom character Murphy Brown had a child, Vice President Dan Quayle denounced her for setting a bad example for American women (a headline of the time: “Murphy Has a Baby; Quayle Has a Cow”
). Harris’s review showing that Murphy’s baby would probably have turned out fine was not welcome. (To be fair, concerns about fatherlessness may not be ill founded, but the problem may be the absence of fathers from all the families in a neighborhood rather than the absence of a father from an individual family. These fatherless children lack access to other families in which an adult male is present, and worse, they have access to packs of single men, whose values trickle down to their own peer groups.) Also, the Great Satan, Hillary Clinton, had written a book on childhood called It Takes a Village, based on the African saying “It takes a village to raise a child.” Conservatives despised it because they thought the whole idea was a pretext for social engineers to take childrearing out of the hands of parents and give it to the government. But Harris quoted the saying too, and her theory implies there is some truth to it.
And then there were the experts. Brazelton called the thesis “absurd.”62 Jerome Kagan, one of the deans of scholarly research on children, said, “I’m embarrassed for psychology.”63 Another developmental psychologist, Frank Farley, told Newsweek:
She’s all wrong. She’s taking an extreme position based on a limited set of data. Her thesis is absurd on its face, but consider what might happen if parents believe this stuff! Will it free some to mistreat their kids, since “it doesn’t matter”? Will it tell parents who are tired after a long day that they needn’t bother even paying any attention to their kid since “it doesn’t matter”?64
Kagan and other developmentalists told reporters about the “many, many good studies that show parents can affect how children turn out.”
What were these “many, many good studies”? In the Boston Globe, Kagan laid out what he called the “ample evidence.”65 He mentioned the usual see-no-genetics studies showing that smart parents have smart children, verbal parents have verbal children, and so on. He observed that “a 6-year-old raised in New England will be very different from a 6-year-old raised in Malaysia, Uganda, or the southern tip of Argentina. The reason is that they experience different child-rearing practices by their parents.” But of course a child growing up in Malaysia has both Malaysian parents and Malaysian peers. If Kagan had considered what would happen to a six-year-old child of Malaysian parents who grew up in a New England town, he might have thought twice before using the example to illustrate the power of parenting. The other “evidence” was that when authors write their memoirs, they credit their parents, never their childhood friends, with making them what they are. An irony in these feeble arguments is that Kagan himself, in the course of a distinguished career, often chided his fellow psychologists for overlooking genetics and for accepting their culture’s folk theories on childhood instead of holding them up to scientific scrutiny. I can only imagine that on this occasion he felt compelled to defend his field against an exposé by a grandmother from New Jersey. In any case, the other “good studies” produced by defensive psychologists were no more informative.66
So HAS HARRIS solved the mystery of the Third Law, the unique environment that comes neither from the genes nor from the family? Not exactly. I am convinced that children are socialized—that they acquire the values and skills of the culture—in their peer groups, not their families. But I am not convinced, at least not yet, that peer groups explain how children develop their personalities: why they turn out shy or bold, anxious or confident, open-minded or old-school. Socialization and the development of personality are not the same thing, and peers may explain the first without necessarily explaining the second.
One way that peers could explain personality is that children in the same family may join different peer groups—the jocks, the brains, the preppies, the punks, the Goths—and assimilate their values. But then how do children get sorted into peer groups? If it is by their inborn traits—smart kids join the brains, aggressive kids join the punks, and so on—then effects of the peer group would show up as indirect effects of the genes, not as effects of the unique environment. If it is their parents’ choice of neighborhoods, it would turn up as effects of the shared environment, because siblings growing up together share a neighborhood as well as a set of parents. In some cases, as with delinquency and smoking, the missing variance might be explained as an interaction between genes and peers: violence-prone adolescents become violent only in dangerous neighborhoods, addiction-prone children become smokers only in the company of peers who think smoking is cool. But those interactions are unlikely to explain most of the differences among children. Let’s return to our touchstone: identical twins growing up together. They share their genes, they share their family environments, and they share their peer groups, at least on average. But the correlations between them are only around 50 percent. Ergo, neither genes nor families nor peer groups can explain what makes them different.
Harris is forthcoming about this limitation, and suggests that children differentiate themselves within a peer group, not by their choice of a peer group. Within each group, some become leaders, others foot soldiers, still others jesters, loose cannons, punching bags, or peacemakers, depending on what niche is available, how suited a child is to filling it, and chance. Once a child acquires a role, it is hard to shake it off, both because other children force the child to stay in the niche and because the child specializes in the skills necessary to prosper in it. This part of the theory, Harris notes, is untested, and difficult to test, because the crucial first step—which child fills which niche in which group—is so capricious.
The filling of niches in peer groups, then, is largely a matter of chance. But once we allow Lady Luck into the picture, she can act at other stages in life. When reminiscing on how we got to where we are, we all can think of forks in the road where we could have gone on very different life paths. If I hadn’t gone to that party, I wouldn’t have met my spouse. If I hadn’t picked up that brochure, I wouldn’t have known about the field that would become my life’s calling. If I hadn’t answered the phone, if I hadn’t missed that flight, if only I had caught that ball. Life is a pinball game in which we bounce and graze through a gantlet of chutes and bumpers. Perhaps our history of collisions and near misses explains what made us what we are. One twin was once beaten up by a bully, the other was home sick that day. One inhaled a virus, the other didn’t. One twin got the top bunk bed, the other got the bottom bunk bed.
We still don’t know whether these unique experiences leave their fingerprints on our intellects and personalities. But an even earlier pinball game certainly could do so, the one that wires up our brain in the womb and early childhood. As I have mentioned, the human genome cannot possibly specify every last connection among neurons. But the “environment,” in the sense of information encoded by the sense organs, isn’t the only other option. Chance is another. One twin lies one way in the womb and stakes out her share of the placenta, the other has to squeeze around her. A cosmic ray mutates a stretch of DNA, a neurotransmitter zigs instead of zags, the growth cone of an axon goes left instead of right, and one identical twin’s brain might gel into a slightly different configuration from the other’s.67
We know this happens in the development of other organisms. Even genetically homogeneous strains of flies, mice, and worms, raised in monotonously controlled laboratories, can differ from one another. A fruit fly may have more or fewer bristles under one wing than its bottlemates. One mouse may have three times as many oocytes (cells destined to become eggs) as her genetically identical sister reared in the same lab. One roundworm may live three times as long as its virtual clone in the next dish. The biologist Steven Austad commented on the roundworms’ lifespans: “Astonishingly, the degree of variability they exhibit in longevity is not much less than that of a genetically mixed population of humans, who eat a variety of diets, attend to or abuse their health, and are subject to all the vagaries of circumstance—car crashes, tainted beef, enraged postal workers—of modern industrialized life.”68 And a roundworm is composed of only 959 cells! A human brain, with its hundred billion neurons, has even
more opportunities to be buffeted by the outcomes of molecular coin flips.
If chance in development is to explain the less-than-perfect similarity of identical twins, it says something interesting about development in general. One can imagine a developmental process in which millions of small chance events cancel one another out, leaving no difference in the end product. One can imagine a different process in which a chance event could derail development entirely, or send it on a chaotic developmental path resulting in a freak or a monster. Neither of these happens to identical twins. They are distinct enough that our crude instruments can pick up the differences, yet both are healthy instances of that staggeringly improbable, exquisitely engineered system we call a human being. The development of organisms must use complex feedback loops rather than prespecified blueprints. Random events can divert the trajectory of growth, but the trajectories are confined within an envelope of functioning designs for the species. Biologists refer to such developmental dynamics as robustness, buffering, or canalization.69