by Bryan Caplan
One of the earliest studies of nature, nurture, and character used the Swedish Adoption/Twin Study of Aging. Researchers gave personality tests to 500 reared-apart twins and 600 reared-together twins and found small to moderate nurture effects. Suppose you’re higher on conscientiousness and agreeableness than 80 percent of the population. According to the Swedish twin study, you should expect your adopted sibling to be more conscientious than 54 percent of the population, and more agreeable than 57 percent. A survey article published in Science a year later reported similar nurture effects for conscientiousness but smaller effects for agreeableness.
More recent twin research finds even smaller effects of parenting on character. One team looked at almost 2,000 German twins. While most personality research simply asks people to describe themselves, the German twin study got three perspectives on each subject: a self-report, plus two independent reports by friends or other close acquaintances. When researchers combined all this information, they found unusually large effects of nature on conscientiousness and agreeableness, and no effect of nurture. Studies of 1,600 American twins and 600 Canadian twins also found that upbringing did not affect character.
If you’re skeptical about self-reporting, you might prefer the Swedish approach. Sweden has universal military service. After high school graduation, a professional psychologist reviews each recruit’s dossier and interviews him to rate the young man’s “emotional stability, persistence, his ability to function in a group and his ability to take initiatives.” Researchers often call these traits “non-cognitive skill,” but they sound a lot like character. Since Sweden keeps excellent twin records, we can use psychologists’ ratings to understand where character comes from. As usual, genes play a serious role: Identical twins are much more alike in character than fraternal twins. Upbringing, in contrast, only has a small effect. If you had more character than 80 percent of Swedish recruits, you should expect your adopted sibling to have more than 54 percent.
In 2005, leading psychologist John Loehlin took a comprehensive look at family studies of personality. Character does run in families: If your parents were higher in conscientiousness and agreeableness than 80 percent of the population, you’re typically more conscientious than 53 percent of us, and more agreeable than 54 percent. However, the resemblance is about equally strong when you compare adoptees to their biological parents, and virtually nonexistent when you compare adoptees to the parents who raised them. Heredity wins again.
Parents have little or no effect on criminal behavior. Faced with all this evidence, parents might protest, “I’m not trying to turn my kid into a saint; I’m just trying to raise a law-abiding member of society.” As Bart Simpson told Judge Constance Harm:Your Honor, it’s not easy being my parents. I’m always screwing up in school and getting in trouble with the law. But if I grow up to be a halfway decent person, I know it will be because of my mom and dad.
Bart’s not alone. Most of us think that good parenting can keep kids from turning into common criminals. But how much effect on criminality do parents really have? As usual, there are twin and adoption studies to enlighten us.
In 1984, Science published a study of almost 15,000 Danish adoptees age fifteen or older, their adoptive parents, and their birth parents. Thanks to Denmark’s careful record keeping, the researchers knew whether any of the people in their study had criminal convictions. Since few female adoptees had legal problems, the study focused on males—with striking results. As long as the adoptee’s biological parents were law abiding, their adoptive parents made little difference: 13.5 percent of adoptees with law-abiding biological and adoptive parents got convicted of something, versus 14.7 percent with law-abiding biological parents and criminal adoptive parents. If the adoptee’s biological parents were criminal, however, upbringing mattered: 20 percent of adoptees with law-breaking biological and law-abiding adoptive parents got convicted, versus 24.5 percent with law-breaking biological and adoptive parents. Criminal environments do bring out criminal tendencies. Still, as long as the biological parents were law abiding, family environment made little difference.
In 2002, a study of antisocial behavior in almost 7,000 Virginian twins born since 1918 found a small nurture effect for adult males and no nurture effect for adult females. The same year, a major review of fifty-one twin and adoption studies reported small nurture effects for antisocial attitudes and behavior. For outright criminality, however, heredity was the sole cause of family resemblance.
The lesson: Even if your standards are low, instilling character is hard. Genes are the main reason criminal behavior runs in families. Contrary to popular opinion, good upbringing is not enough to steer a child away from a life of crime. If that depresses you, there are two sources of comfort. First, as long as you and your spouse are law abiding, it’s good news, because noncriminality is hereditary, too. Second, most children of criminals don’t become criminals; in the Danish adoption study, over three-fourths of the boys born of and raised by people with criminal convictions weren’t convicted themselves. You don’t have to be hopeful about parenting to be hopeful about your children’s future.
WISH #6: VALUES
Babies, chum: tiny, dimpled, fleshy mirrors of our usness, that we parents hurl into the future, like leathery footballs of hope. And you’ve got to get a good spiral on that baby, or evil will make an interception.
—The Tick
Character isn’t controversial; almost all parents want their kids to be hardworking, honest, kind, and so on. Yet few parents stop there. Most of us also try to impart controversial values. We teach our children to see the world as we do—even if others say our lessons are wrong or wicked. When your side does it, it’s “education”; when other sides do it, it’s “brainwashing.”
Religion is the most obvious example. Whatever your religious view—Christian, Jewish, Muslim, Zoroastrian, or atheist—I bet you want your kids to believe as you do. In fact, I bet you want your kids to accept not just your religion, but your interpretation of it. If you’re a liberal Jew, you don’t want your child to join an ultra-Orthodox synagogue.
In politics, there’s a little more room for independent thought, but less than we care to admit. If you’re a moderate Democrat, you want your kids to be moderate Democrats, too. You don’t want Marxist sons who denounce your bourgeois compromises, or Republican daughters who tell you what Ann Coulter said. When controversy erupts, parents want their children on their side.
Religion and politics are only the beginning. Some parents teach traditional values; others call for open-mindedness and tolerance. Some tell their children to have faith; others praise the power of critical thinking. Some parents preach mercy; others demand an eye for an eye. Indeed, some parents raise their children to be fruitful and multiply; others tell them that we need to limit our numbers, lest we destroy the planet.
You know where I stand on the last issue, but I’m not bringing up controversies to take a side. My task isn’t to distinguish good education from evil brainwashing. I am simply observing that parents of all stripes want their kids to share their values and believe they can cause this to happen. Then I ask my standard question: “Is this widespread belief true?” When parents try to educate/brainwash their kids, do their lessons stick?
As usual, the fact that we often share our parents’ values does not prove that upbringing matters. The similarity could be hereditary. If genes affect health, intelligence, happiness, success, and character, why not values? To measure parents’ influence, we must once again defer to twin and adoption studies. They find that contrary to the Jesuit motto “Give me a child until he is seven and I will give you the man,” the effect of upbringing on values is superficial. If you want your children’s lip service, parenting can probably get it for you. But if you want to win your children’s hearts and minds, twin and adoption studies show the power of parenting to be greatly overrated.
Parents have a big effect on religious labels, but little on religious attitudes and behavi
or. Parents strongly affect which religion their children say they belong to. A major study of over 7,000 adult Australian twins finds that identical and fraternal twins are highly and equally likely to share a religion—precisely what you would expect if nurture mattered a lot and nature didn’t matter at all. Another study of almost 2,000 women from the Virginia Twin Registry reaches a similar conclusion: Family has a big effect on religious denomination, while genes have at most a small effect.
If the key to salvation is merely pronouncing the words “I belong to religion X,” these are nurture effects of infinite importance. If you’re raised by a family of believers, you’ll probably be saved; if you’re not, you’ll probably be damned. Whatever your religion, though, I bet you hold to higher standards. As the book of James asks, “What good is it, my brothers, if a man claims to have faith but has no deeds? . . . Faith by itself, if it is not accompanied by action, is dead.”
Almost all religions have detailed lists of doctrinal and behavioral do’s and don’ts that separate the faithful from the hypocrites. Catholics are supposed to attend mass on Sundays and holidays, fast on Ash Wednesday and Good Friday, delay sex until marriage, oppose abortion, believe in transubstantiation, and confess their sins at least once per year. When we call a person religious, we usually mean that he takes his religion’s do’s and don’ts seriously.
Twin and adoption research reveals surprisingly little parental influence on how truly religious children grow up to be. One early study of almost 2,000 adult Minnesota twins reared together and apart found little or no effect of parenting on religiosity. Researchers measured the twins’ interest in both religious activities (such as attending services, volunteering, and religious study) and religious occupations (such as being a minister, priest, rabbi, missionary, or religious writer). Identical twins were much more similar on both measures than fraternal twins. Nurture effects were small for religious activities and zero for religious occupations. A recent follow-up found similar results.
Another team of researchers looked at the religiosity of over 11,000 adult twins from Virginia. Parents have almost no effect on adult church attendance. Suppose your adopted parents attend church more than 80 percent of adults. When you grow up, we should expect you to attend church more often than 51 percent of us. The researchers also measured the Virginia twins’ affinity for the “Religious Right.” They found a very small nurture effect for men, and none for women. Another major study of American and Australian church attendance looked at over 5,000 families with twins in each country. Parents in both countries have zero effect on the church attendance of their adult sons, and at most a tiny effect on the church attendance of their adult daughters.
A different team of researchers created a test to measure “the centrality of religion in the individual’s life.” The test covers the frequency of church attendance, prayer, scripture reading, religious discussion, religious moralizing, and religious holiday observance; membership in religious groups; having friends with similar beliefs; and the importance of religious faith in daily life. When over 500 male Minnesota twins in their early thirties took the test, upbringing had only a modest effect. If you’re more religious than 80 percent of people, expect your adopted sibling to be more religious than 56 percent. When about 600 female Minnesota twins took the same test (once when they were twenty, then again when they were twenty-five), though, the nurture effect was larger. The source of the discrepancy was apparently age, not gender: The effect of upbringing was large for twenty-year-old women, weaker for twenty-five-year-old women, and weakest of all for thirty-something men.
Parents have a big effect on political labels, but little on political attitudes and behavior. Modern American politics looks a lot like religion: People split into “right-thinking” camps and demonize rivals as stupid and/or evil. The splits usually run along family lines: Democratic parents tend to have Democratic children, Republican parents tend to have Republican children. Religion probably remains touchier than politics; your parents are more likely to disown you for rejecting their church than rejecting their party. Still, parents want their kids to share their political values and try to pass them on.
Twin studies confirm that politics is a lot like religion. Parents have a large effect on your political label. In the Virginia 30,000 study, both identical and fraternal twins were highly and almost equally likely to belong to the same political party, implying large effects of nurture and weak effects of nature. Party identification works the same way in Australia: Fraternal twins are highly similar, and identical twins are only slightly more alike.
In politics as in religion, however, the biggest nurture effects are also the most superficial. The Virginia 30,000 study found that parents have little effect on the strength of your partisan commitment —whether you always vote for the same party or straddle the fence. A national survey of young American adult twins found that parents have little influence over whether people bother to vote or participate in other political activities. If you’re more politically active than 80 percent of people, expect your adopted sibling to be more active than 56 percent. The same goes for overall political philosophy and positions on specific issues. Parents may slightly affect how conservative you are, or your views on immigration, abortion, socialism, or crime. But not much.
Parents have little effect on traditionalism and modernism. “Old-fashioned” parents raise children to be practical, respect authority, and value tradition. “New-age” parents raise children to be creative, question authority, and value diversity. Of course, even bohemians overstate their openness; the parent who says “Question authority” usually means “Question their authority,” not “Question my authority.” Whether you’re old-school or progressive, you have a mental picture of how your kids ought to turn out, and try to make it happen.
When people compare traditional to modern parenting, they usually argue that one is better. I’m asking a separate question: Do parents’ efforts to instill a “traditional” or “modern” outlook succeed ? To get an answer, we need to measure people’s traditionalism. Once again, personality psychologists are here to help.
Personality psychologists have spent decades studying the trait they call “openness.” Open is the opposite of old-fashioned. People high in openness see themselves as creative, interesting, tolerant, curious, and artistic; people low in openness see themselves as practical, normal, upright, respectful, and down-to-earth. When people are less open than us, they seem square; when they’re more open than us, they seem weird. The long-running cartoon King of the Hill is a fine depiction of the conflict: Hank Hill is a practical, old-fashioned Texas propane salesman who endures habitual embarrassment at the hands of his artistic, unconventional son, Bobby. In the classic episode “Rodeo Days,” Hank encourages his son to ride in the local rodeo, but Bobby decides—to his father’s horror—to be a rodeo clown instead.
Like other personality traits, openness is largely unresponsive to upbringing. Diverse twin studies find little or no effect of nurture on openness, including studies of over 1,000 Swedes raised apart and together, almost 2,000 Germans, about 1,600 American high school juniors, and 500 Canadians. The average adoption study finds a small but reliable effect of parenting on openness. Suppose two parents have a mixed family with both biological and adopted children. If the parents are more open than 80 percent of the population, we should expect their biological children to be more open than 56 percent of their peers, versus 52 percent for adoptees.
Family values. You probably think your love life is a private matter. But even if your parents officially believe in minding their own business, they probably involved themselves every step of the way. When children are young, parents frantically try to control the flow of sexual information. Some practice blanket censorship, some tell lies about storks, others make sure their kids hear it from them first. Once they’ve got teenagers, almost all parents want their kids to go slow. But some virtually lock them up, while others hand out condoms. By the time
they’re in their twenties, parents are pressuring their kids to marry—and vetting the candidates. Soon after the wedding, parents start asking about grandchildren. If you contemplate divorce, your parents will probably eagerly share their opinion on that, too.
Parents clearly try to control their children’s love lives, but how much influence do they really have? Counterexamples are easy to spot: Boys raised in puritanical homes sneak Internet porn; girls with purity rings get pregnant out of wedlock. Yet despite these counterexamples, our sexual, marital, and reproductive behavior tends to resemble our parents’—or at least the behavior of our parents when they were our age. The question is not whether people resemble their parents, but why. How much is due to upbringing, and how much to heredity?
Parents have moderate influence over when their daughters start having sex, but little over their sons. There are two major Australian twin studies of sexual initiation. The first included over 3,000 women born between 1922 and 1965. A follow-up roughly doubled the sample size by adding older and younger female twins. Both studies found moderate to large nurture effects. If you waited longer than 80 percent of girls, the first study found that you could expect your adopted sister to wait longer than 58 percent of girls; the second study, longer than 65 percent. Another research team used the National Longitudinal Survey of Youth to study the sexual initiation of about 2,000 American twins, siblings, half siblings, and cousins. For women, upbringing had as much effect in the United States as it did in Australia.
In contrast, parental influence on sons’ sexual initiation is weak. The first of the Australian twin studies included about 1,800 males. While it found a moderate nurture effect for the older men (born 1922–1952), it found zero parental influence for the younger men (born 1952–1965). The NLSY study similarly found only a small nurture effect for American men born between 1958 and 1965. If you waited longer than 80 percent of boys, the NLSY study estimated that your adopted brother would wait longer than 53 percent.