Human Diversity

by Charles Murray

  Once again, look ahead to 2030. By that time, scientists will be able to make predictions about personality characteristics, abilities, and social behavior of groups on the basis of polygenic scores. If the orthodoxy is right, such statistical predictions should be impossible. If everything is socialization, then DNA samples shouldn’t be able to tell you anything about differences in personality characteristics, abilities, or social behavior among adults.

  “But they can just soften their position,” it may be argued. The orthodoxy can shift to an “it’s almost but not entirely a matter of socialization” and carry on as before. Perhaps that’s right, but I doubt it. We didn’t need to sequence the genome to recognize that human beings have evolved underlying characteristics that are deeply biological. In that sense, the orthodoxy on campuses has been hanging on inexplicably, like a religious cult whose leader’s predictions have been contradicted by events time and again. That amounts to a fragile situation. I don’t think the orthodoxy can tolerate acknowledging openly that, for example, men and women are biologically different above the neck even a little bit.

  This explains my probably starry-eyed expectation: The ability to predict a wide variety of human differences solely from baseline genetic measures will puncture the center of the orthodoxy’s beleaguered defense. Evolutionary psychologists will be liberated to make strides in describing human nature, not just individual heritable traits, in an environment where the legitimacy and importance of their contributions is accepted.

  Reconciling Human Nature with Ideology

  The insistence that only the environment shapes cognitive repertoires has been yet another case, like “gender is a social construct” and “race is a social construct,” of taking partial truths and running them into the ground. In the centuries when Locke and Smith wrote, most people with the potential to be philosophers probably really were living out their lives as street porters—and farmers and housewives. People in all eras and places really are shaped to some degree by their environment. But the limits of shaping are governed by hundreds of millions of years of evolution.

  An apt example of the clash between utopian plans and human nature comes from Israeli kibbutzes founded in the 1930s and 1940s. Infants and children were cared for by full-time caretakers and lived in a centralized children’s quarter, spending only a few hours per day with their parents. The intention was to replace the nuclear family with the extended family of the community. The rationale for doing so was grounded in ideals about human flourishing. The members of the early kibbutzes were themselves idealistic and committed to their utopian goals. The kibbutzes were small enough (a few hundred people) to permit intimate communities to form. The members shared powerful bonds of a common ethnic and cultural heritage. If ever a socialist utopian community could succeed, the kibbutzes had the best chance. For a while, it seemed to work.23 Over the years, however, the nuclear family ineluctably reasserted itself.24 It was inevitable. Societies must be made to fit human nature, because human nature cannot be reshaped to fit theoretical utopias.

  This brings us to the intersection of scientific findings and political ideology. I have a problem finding the right words here, because the meanings of “conservative” and “liberal” have shifted so radically in recent years. I’m using them as they were popularly understood in the United States from the New Deal until 2016. With those meanings in mind, this much is undeniable: The belief in constraints and limits on government’s ability to change people is inherently conservative. The belief in open-ended potential for changing people through the right policies is inherently liberal.

  That doesn’t mean that liberals are forced into denying human nature; it just means that they need to take it into account. In the case of sex differences, this can often be done by adjusting policy to recognize differences in distributions between men and women. Custody of children after divorce is a case in point. Until the last quarter of the nineteenth century, men were automatically awarded custody of children. By the latter part of the nineteenth century, and largely as a result of first-wave feminism, the “tender years doctrine” automatically gave the mother custody of young children unless there was clear evidence that the mother was unfit. In the 1970s, one effect of second-wave feminism in many states was to remove the mother’s legal advantage in custody disputes, substituting “the best interests of the child” as the basis for custody decisions and encouraging joint custody.

  What is likely to be in “the best interests of the child” when the child is of “tender years”? By any measure of which sex is better at nurturing young children, there is a big effect size favoring females and an overwhelming evolutionary case that the female advantage is grounded in biology. If that is accepted as scientific reality, what is a principled liberal position toward child custody? One option is to follow Pinker’s principle that people should not be judged by the average properties of their group. That’s clearly the right call when we’re talking about fairness in treating job applicants or criminal defendants. But what about when a helpless third party’s “best interests” are involved? In a world where judges in custody disputes often are faced with no clear evidentiary basis for favoring one parent over another and a helpless third party’s welfare is at stake, a principled liberal position can acknowledge an important innate difference between men and women.

  Similarly, Title IX of the Education Amendments of 1972 had a good rationale (if you aren’t a libertarian) for requiring schools to provide equitable opportunity for female students to participate in sports.25 Some girls love competitive sports, and they weren’t being offered enough opportunities to play them. It’s also true that by any measure of interest in team sports there’s a big effect size favoring males and there is again an overwhelming evolutionary case that the greater male interest is grounded in biology.26 Title IX should be administered with that difference in mind (it often hasn’t been).

  Women in combat? It’s not an issue of female courage. But from early childhood into adulthood, males are far more attracted than females to physical contests, including ones involving violence, and are more physically aggressive and risk-taking than women. Once again there is an overwhelming case that the sex difference is grounded in biology.27 Are women as enthusiastic as men about attacking and killing total strangers if the proper institutional framework is provided (i.e., the military in battle)? Are women as obsessively driven to win at all costs in contest situations as men? Some women are, but what proportion? Can male soldiers be trained out of their instinct to protect women? Probably some can—but what proportion? Men and women have different distributions on these traits, with biology playing a major role. The conclusion need not be that women shouldn’t be used in combat roles, but that the relevant sex differences need to be taken into account. Women in combat as part of a missile-firing team on a warship sidesteps the potential problems I just mentioned; women in combat as part of a frontline infantry platoon triggers all of them.

  Being realistic about human nature goes far beyond sex differences. An acceptance of the constraints imposed by human nature should guide the administration of the civil and criminal justice systems, the regulation of business, the powers granted to bureaucrats—the operations of just about every social, cultural, economic, and political institution. It can be done by conservatives and liberals alike without either side having to abandon core principles. The challenge for conservatives is to accommodate their historic advocacy of freedom and limited government with the role of the genetic lottery in determining success. The challenge for liberals is to acknowledge the constraints of human nature in ways they have historically resisted.

  The Role of the Environment in Explaining Human Differences Has Been Misconceived

  It’s Not the Shared Environment Versus the Nonshared Environment. It’s the Manipulable Environment Versus Happenstance and Milieu.

  I cannot prove that the role of the environment in explaining human differences has been misconceived, but I can describe why I t
hink so.

  Murray’s Conjecture

  Analogies have been a popular way to describe the relative roles of genes and the environment for 50 years. Richard Lewontin started it in 1970 with one that Richard Herrnstein and I adapted for The Bell Curve: All of the kernels in a strain of hybrid seed corn are genetically identical. But if two handfuls of that seed corn are planted in Iowa and the Mojave Desert, there will be a huge difference in yield.28 Christopher Jencks continued the tradition in 1979 with his analogy of red-haired children who are denied education (hair color is highly heritable and will be correlated with a difference in academic achievement, but it has nothing to do with genes).29 Eric Turkheimer used piles of building materials to describe the limits of the genetic contribution in predicting phenotypic traits. Graham Coop produced the most elaborate analogy, the one about differences in French and English tea drinking that I recounted in chapter 14.

  What these analogies have in common is that none of them would pose a problem to a real analysis. That’s understandable; analogies need to employ simplistic situations to make their point. But social scientists trying to figure out differences in corn yield, educational achievement, the appearance of finished buildings, and tea drinking are not going to spend a nanosecond puzzling over the environmental forces at work. They are blindingly obvious. I think this is true far more often than most social scientists concede, hence Murray’s Conjecture: When a difference really is environmental in origin, it’s easy to prove it; when it’s hard to prove an environmental cause, it’s because the role of the environment is minor.30

  Why were there so proportionally few female physicians through the first half of the twentieth century? Guess what: It was the environment. Proving that it was the environment is trivially easy. Why has there been so little change in the vocational choices of women for the last thirty years? It is now difficult to make an empirical case that residual sexism is the culprit. The reason it’s so difficult, the conjecture says, is because the environment no longer has much effect.

  The conjecture applies to personality traits and abilities, but I must specify that it often does not apply to much of social behavior. As I will explain subsequently, I think milieu everywhere and always has pervasive effects on social behavior.

  As an elaboration of Murray’s Conjecture, I propose a distinction among three types of environmental forces: legal compulsion, hard custom, and soft custom.

  Legal compulsion. In the hierarchy of ways in which some human beings can force others to do their bidding, physical coercion is at the top. In a civilized society, that means legal compulsion. Why was European Jewish achievement in the arts and sciences so rare from the Middle Ages until the nineteenth century? Because Jews were prohibited by law from entering universities or engaging in scientific professions. Within two generations of their legal emancipation, Jews were disproportionately represented among the leading figures in both the arts and sciences.31

  Hard custom refers to means other than laws that entail coercion. An example of physical coercion (and the threat of it) was the extralegal enforcement of segregation in the South to supplement the Jim Crow laws until the Civil Rights Act of 1964 and continuing to some degree thereafter. A nonviolent example of covert prohibition is the de facto ceiling that many medical schools put on the number of women they would admit until the law mandated equal access. Hard custom can persist for a long time even though it is not underwritten by laws. Under some circumstances, legal prohibition of discrimination can shorten its persistence.

  Soft custom has no element of covert prohibition or the threat of physical coercion. It consists instead of social incentives to refrain from engaging in a behavior—stigma or social isolation. No one will beat you up, you won’t lose your job, nor will any other tangible punishment be administered from on high. But you must bear social punishment—the disapproval of your parents, teachers, or other people who matter to you. Some of your friends become ex-friends. Your coworkers may shun you. For example, if you were one of the first women hired as police patrol officers in the 1960s, you had to put up with a lot, even if you were treated correctly professionally (which often didn’t happen). Maybe none of your male coworkers would eat lunch with you and you would be subjected to endless derogatory comments about women’s abilities.

  My speculative proposition here is that once legal prohibitions and hard custom are no longer an issue, soft custom has a short half-life. The arrival of even a second policewoman on the force is a big relief to the pioneer. Then there are three, five, and 10 women. Many male police officers who were initially hostile see that the women are pretty good cops after all and begin to accept them as equals. I am not saying that soft custom goes away entirely, but that its power to intimidate diminishes nonlinearly, just as the half-life of radioactivity diminishes nonlinearly. Further, I would argue that the half-life is often a matter of years, sometimes a decade or so, but seldom many decades.

  Translated into the way I see the role of the environment as it is usually construed, I think that much of the remaining discrimination against women, ethnic minorities, and LGBT people in the United States consists of soft custom and is well into or past its first half-life, albeit to varying degrees in different geographic and socioeconomic settings. That’s why proving a large independent role for the environment in differential outcomes is often becoming difficult.

  Happenstance, Milieu, and the Manipulable Environment

  None of the above means that the environment is unimportant in determining how people’s lives turn out. But in conceptualizing the environment, we shouldn’t divide it into the shared environment and the nonshared environment. Rather, we should divide it into happenstance, milieu, and the manipulable environment.

  Happenstance. Happenstance is equivalent to the nonshared environment seen from a different perspective.

  Surely everyone who has reached middle age occasionally stops to muse on how differently life would have turned out except for random events. Certainly it is true of people who are lucky enough to be in happy marriages. Meeting that right person is almost always the result of a series of capricious events, and it’s scary to think how your life would have been different if any one of the pieces hadn’t fallen into place. With regard to one’s children, happenstance amounts to one-in-millions chances that they turned out to be who they uniquely are. And of course, so was the union of egg and sperm that created you. The role of randomness is huge.

  But do you feel the same way about the kind of person you have become? Your abilities? Your personality strengths and weaknesses? Speaking for myself, I know that my circles of friends, the places I’ve lived, the books I have written, and all sorts of other ways in which my personality traits and abilities have been expressed in behavior have been hugely affected by happenstance. So has my happiness. I can easily imagine having reached old age sad and lonely because of happenstance. But I sense that my personality traits and abilities are close to what they would have been no matter what, short of some extreme psychological or physical trauma.

  I would argue that happenstance explains why people discover as they get older that they are becoming more like their parents—a common personal experience that is reflected in the technical literature demonstrating that heritability of many traits rises with age. My logic goes like this: In adolescence and young adulthood, people act inconsistently with their genetic predispositions for many reasons. Sometimes they are consciously rebelling against their parents. The broader generic reason is that adolescents and young adults are immature and prone to spectacularly poor judgment. That’s why most of us recall things we did in adolescence and into our 20s and say to ourselves, “What could I have been thinking?”

  THE REAL MEANING OF CLASS PRIVILEGE

  I am generally skeptical of claims about the power of privilege. Growing up in an upper-middle-class or wealthy home has a variety of potential downsides. An exception involves the stupid things that adolescents sometimes do. Upper-middle-class families a
re often able to rescue their children from adolescent mistakes that can have lifelong disastrous consequences for poor children.

  During those same years, we get bounced around by happenstance. Some random influences push us to do foolish things; other random influences push us toward maturity. For most people, this multitude of random events balances out, and we emerge into adulthood less likely to do things that run against genetic predisposition. Whether that’s good, bad, or indifferent depends on what genetic predispositions we’re talking about. But the nonshared environment has had its heyday. It may have played havoc with our lives for a time, but those effects were temporary (recall evidence to that effect from chapter 13).

  Milieu refers to the world into which we are born in its cultural, social, and economic totality—the water in which we swim. Milieu often exerts itself most powerfully on a national scale, but it has many variations. The Amish have managed to preserve a comprehensive milieu in the face of a competitive national milieu. The typical college or university has a milieu that transcends socioeconomic status. So does a black neighborhood in Harlem or a Korean neighborhood in Los Angeles. In America’s current polarized state, many of the politically committed of both extremes live in separate milieus. In my view, the current biggest American division is the milieu of small towns and small cities versus the milieu of the megalopolises and their suburbs.

  Milieu is pervasively causal. It doesn’t just influence the ways we behave. It entirely rules out many possibilities and makes others almost inevitable. A thought experiment may make the point. Imagine that Michelangelo, Mozart, and Shakespeare are all cloned in the twenty-first century as babies with exactly the same DNA as the originals. Let us suppose that as adults they respectively become an artist, a composer, and a writer. It is easily imaginable that they will rise to the top of their professions in our time—their talents would be prodigious in any century. But it is inconceivable that their masterpieces will be anything like the Pietà, the Jupiter Symphony, or King Lear. Expressions of genius are decisively shaped by the milieu into which geniuses are born.