by Bryan Caplan
Parents have little or no effect on teen pregnancy. The fact that parents affect their daughters’ sex lives more than their sons’ is not surprising. Girls’ parents are more likely to take extreme measures due to fear of teen pregnancy. Their precautions largely fail. While the Australian study of almost 7,000 female twins found that parents had a moderate effect on the pregnancy of teens born between 1964 and 1971, parents had zero effect on the pregnancy of teens born between 1893 and 1964. A study of about 2,000 female Swedish twins born in the Fifties also found zero effect of upbringing on teen pregnancy.
Parents have little or no effect on adult sexual behavior. Few parents want their children to be forty-year-old virgins. Less than a third of Americans believe that premarital sex is always wrong. Still, this doesn’t mean that anything goes. Parents try to instill values about proper sexual behavior and hope these values last a lifetime.
They hope in vain. Two major twin studies find little effect of upbringing on adult sexuality. The first surveyed nearly 5,000 Australian twins about their “sociosexuality,” better known as promiscuity. The survey had questions about sexual attitudes (such as “Sex without love is okay” and “The thought of an illicit sex affair excites me”) as well as sexual behavior (such as “With how many partners of the opposite sex have you had sexual intercourse within the past year?” and “With how many partners of the opposite sex have you had sexual intercourse on one and only one occasion?”). Family environment had almost no effect on sociosexuality; if you were in the 80th percentile, you could expect your adopted sibling to stand in the 51st. A second study of 1,600 female twins in the United Kingdom focused on number of sexual partners and infidelity. It found a small effect of family environment on the total number of sex partners, a moderate effect on attitudes about infidelity—and zero effect on the actual practice of infidelity.
Parents may have a small effect on sexual orientation. Psychologists used to label homosexuality a mental illness caused by overprotective mothers and distant fathers. Now we tend to see sexual orientation as a preference inherent in our genes. When you look at the evidence, however, neither story quite works. Genes definitely play a strong role—every major twin study finds that identical twins are more alike in their sexual orientation than fraternal twins. Yet genes are far from the whole story—if you’re gay, your identical twin is usually still straight. Upbringing might make a difference, too. In surveys, adopted brothers of gay men and adopted sisters of gay women are about six times as likely to be gay as the general population. This would normally be a smoking gun, but sexual orientation remains a touchy issue. The adoption results might merely show that gays with gay adopted siblings were six times as likely to mail in their surveys. A large Swedish twin study with an exceptionally high response rate—nearly 60 percent—confirms a small nurture effect for women’s orientation, but none for men’s.
Parents have little or no effect on marriage, marital satisfaction, or divorce. In Tolstoy’s novels, parents decide who their children should marry, and when. In War and Peace, set during the early 1800s, parents arrange and forbid marriages without apology. In Anna Karenina, set sixty years later, parents preach respect for their children’s choices but still practice the old ways:The princess realized that in the process of getting to know each other, her daughter might fall in love, and fall in love with someone who did not care to marry her or who was quite unfit to be her husband. And, however much it was instilled into the princess that in our times young people ought to arrange their lives for themselves, she was unable to believe it, just as she would have been unable to believe that, at any time whatever, the most suitable playthings for children five years old ought to be loaded pistols.
Modern parents like to think they’re light years from nineteenth-century Russia. Still, we have views about who and when our children should marry, and few of us keep these views to ourselves. Even if we held our tongues, children might learn by example. Growing up in a broken home has to make kids less likely to have a happy marriage later on, right?
Twin research says otherwise. A study of over 4,000 Minnesota twins, most in their thirties and forties, found zero effect of parenting on marital status. A long-running study of almost 6,000 men from the World War II Twin Registry found moderate nurture effects for early marriage, but none for marital status at thirty, forty, or fifty. Upbringing also has little or no effect on marital satisfaction. A research team asked 1,000 female Swedish twins and their spouses about the quality of their marriages. The women’s parents had no effect on the marital satisfaction of their daughters, but they did have a small effect on the marital satisfaction of their sons-in-law. You might infer that some parents raise unusually good wives, but the simpler story is that some parents are unusually good in-laws.
What about divorce? An early study using the Minnesota Twin Registry found large effects of genes and no effect of family environment. Individuals with a divorced identical twin are almost six times as likely to be divorced; for individuals with a divorced fraternal twin, the risk of divorce less than doubles. A later study expands the sample and reaches the same conclusion: Heredity matters a lot, upbringing doesn’t matter at all. One of the main reasons why divorce is heritable, the authors learned, is that marital stability depends upon personality and values, which in turn depend upon genes.
Parents have little or no effect on childbearing. I often half jokingly tell my three sons that they’re required to have three kids each, but twin studies say I’m wasting my breath. While fertility runs in families, the reason nowadays is almost entirely genetic. A major study of Danish twins born in 1870–1910 found moderate nurture effects on family size. Half a century later, though, these nurture effects had disappeared. Upbringing had a tiny influence on when Danes tried to start a family, but none on the total number of children produced by those thirty-five to forty-one years old. A different team of researchers looked at about 2,000 American twins, siblings, half siblings, and cousins born between 1958 and 1965 and found minimal nurture effects on fertility.
WISH #7: APPRECIATION
To me, it’s more important to deserve my children’s appreciation than to get it. I would raise my children with kindness and respect even if I knew they wouldn’t reciprocate. Still, I hope they will. I want them to feel good about me and look back fondly on their childhood.
The odds that my efforts pay off are pretty good. Both twin and adoption studies confirm that parents affect how their children perceive and remember them. When adoptees in the Colorado Adoption Project were ten to twelve years old, researchers asked both biological and adopted children questions about their families. How loving, communicative, and conflict-prone were they? Siblings gave fairly similar answers whether or not they were biologically related. Suppose you rated your family more positively than 80 percent of kids. Expect your adopted sibling to rate your family more positively than 59 percent of kids.
Two research teams using the Minnesota Twin Registry reached similar conclusions. The first questioned about 1,200 Minnesota twins. How much did they like their parents? How much did their parents like them? How involved were their parents in their lives? How much did they fight with their parents? The twins answered twice—once when they were eleven years old, once when they were fourteen. At both ages, parents had moderate influence over how much their children liked them, and small to moderate influence over perceived involvement and conflict. (Parents had less influence, though, over how much kids thought their parents liked them.) A later research team doubled the number of twins in the sample to almost 2,500 and interviewed them when they were sixteen to twenty years old. They confirmed that parents continue to have moderate effects on how their young adult children feel about and perceive them.
When they ignore their children’s wishes, parents often protest, “You’ll thank me later,” suggesting a disconnect between how we perceive our childhood and how we remember it. But researchers find that upbringing matters for long-term memories as well as im
mediate perceptions. An early Swedish study of 1,400 middle-aged and elderly twins asked them how their parents raised and treated them. Most respondents hadn’t been children for fifty years—but nurture mattered. Identical twins’ portraits of their parents were only moderately more alike than fraternal twins’, and twins raised together gave much more similar answers than twins raised apart. The Swedish study discovered moderate nurture effects for seven out of eight measures of parenting style. Suppose you remember your family as more loving and harmonious than eight percent of adults remember theirs. You should expect your adopted sibling to remember more familial love and harmony than 59 percent of adults. Other studies of German, Canadian, and Swedish twins find that parents have comparable or larger effects on how their grown children remember them.
Many parents have found inspiration in the words of Forest Witcraft:A hundred years from now it will not matter what my bank account was, the sort of house I lived in, or the kind of car I drove . . . but the world may be different because I was important in the life of a child.
If you halve the number of years, Witcraft has science on his side. Half a century from now, your children will remember how you treated them. If you showed them kindness, they probably won’t forget. If you habitually lost your temper, they probably won’t forget that, either. Out of all the wishes on the Parental Wish List, “good memories” are one of the few that clearly depend upon how you raise your child. Don’t forget it.
BEYOND THE EXTRAORDINARY
“Parents have little long-run effect on their kids.” It’s an extraordinary claim—and as astronomer Carl Sagan wisely insisted, “Extraordinary claims require extraordinary evidence.” But behavioral genetics comfortably passes Sagan’s test. Over the last few decades, twin and adoption researchers cast a wide net. They looked at every major trait—and most of the minor traits—that parents seek to cultivate. They approached each of the main questions from multiple angles: Different subjects, different times, different measures. The researchers come from disparate fields: medicine, psychology, economics, sociology, and more. Despite their intellectual diversity and the ambition of their project, twin and adoption researchers have built an impressively consistent body of knowledge about the causes of family resemblance.
If the only lesson you take away from this book is that scientists have solid but shocking answers to the nature-nurture puzzle, I’ll be happy. Behavioral geneticists have done so much great work that it’s an honor just to be their messenger. But I don’t want to merely deliver their message and go home. Now that we’ve got solid but shocking answers to the nature-nurture puzzle, it’s time to tackle a deeper question: How should these answers change the way we raise our children and live our lives?
APPENDIX TO CHAPTER 2: WHERE THE NATURE-NURTURE EFFECT SIZES COME FROM
The final products of most twin and adoption studies are their estimates of three variables:1. the fraction of variance explained by heredity—usually called h2 ;
2. the fraction of variance explained by shared family environment—usually called c2;
3. the fraction of variance explained by non-shared family environment—usually called e2.
To explain these effect sizes in layman’s terms, I take advantage of two mathematical facts implied by standard behavioral genetic models. First, a trait’s h2 equals the expected correlation between identical twins raised apart. Second, a trait’s c2 equals the expected correlation between unrelated individuals raised together.
On the plausible assumption that traits are normally distributed, we can then calculate the expected performance of your separated identical twin or adopted sibling. If you are s standard deviations (SDs) above average on a trait with h2=.3 and c2=.1, we should expect your separated identical twin to be .3s SDs above average, and your adopted sibling to be .1s SDs above average.
On the standard normal distribution, the 80th percentile is approximately .84 SDs above average. If you are in the 80th percentile on a trait, we would therefore expect your separated identical twin to be .84h2 SDs above average and your adopted sibling to be .84c2 SDs above average. Suppose the trait in question is education, and we use Baker et al.’s initial estimates derived from Australian twins: h2= .57 and c2=.24. If you were in the 80th percentile, we would expect your separated identical twin to be .84*.57 SDs=.48 SDs above average and your adopted sibling to be .84*.24 SDs=.20 SDs above average. Using a standard normal table, this translates to the 68th and 58th percentiles, respectively.
When twin and adoption studies explicitly report estimates for h2 and c2, I use them. If they report only correlations for identical versus fraternal twins, I estimate h2 using Falconer’s formula: h2=2*(rMZ -rDZ); since e2=1-rMZ, c2=2rDZ -rMZ. When more complex kinship studies distinguish “shared environment” from “vertical transmission,” I use the latter as my measure of the effect of nurture. When studies directly report genetically informed regression coefficients, I use them without further ado.
3
BEHAVIORAL GENETICS: CAN IT BE TRUE—AND WHAT DOES IT MEAN?
We worry about what a child will become tomorrow, yet we forget that he is someone today.
—Stacia Tauscher
WHEN THEY LEARN ABOUT TWIN AND ADOPTION RESEARCH, SOME parents stonewall, and others despair. Those who stonewall say that behavioral genetics is too counterintuitive to believe. Those who despair say that behavioral genetics is too depressing to embrace. If you’re in either camp—or both—I want to change your mind. You don’t have to stonewall to protect common sense, and you don’t have to deny the facts to avoid despair.
From a distance, the findings of behavioral genetics are indeed counterintuitive. When you look closer at the evidence, however, crucial caveats come into view. Crucial caveats like: Parents’ short-run effects are much bigger than their long-run effects. Upbringing has lasting effects on appreciation, and religious and political identity. And: For every kid who yields to parental pressure, there’s probably another who rebels against it. Learning about these crucial caveats lets us reconcile science and common sense.
In any case, there’s no reason for the science of nature and nurture to drive you to despair. Yes, it deflates some kinds of parenting, like bending kids to your will and riding kids for their own good. But the science also shines a spotlight on exciting opportunities. Exciting opportunities like: Raising kids your own way, guilt free. Treating your kids with kindness and respect, without worrying that you’ll make them soft. And: Having more kids—while still having time for yourself. Learning about these exciting opportunities lets us reconcile science and hope.
Behavioral genetics offers parents a deal: Show more modesty, and get more happiness. You can have a better life and a bigger family if you admit that your kids’ future is not in your hands. The offer is more than fair. In fact, it’s the deal of a lifetime.
“WHO YOU GONNA BELIEVE—ME OR YOUR OWN EYES?” NATURE, NURTURE, AND FADE-OUT
I gotta say that I’m enjoying adulthood. For a lot of reasons. And, I’ll tell you reason number one: As an adult, if I want a cookie, I have a cookie, okay? I have three cookies or four cookies, or eleven cookies if I want. Many times I will intentionally ruin my entire appetite. Just ruin it. And then, I call my mother up right after to tell her that I did it. “Hello, Mom? Yeah, I just ruined my entire appetite.”
—Jerry Seinfeld
As a father of identical twins, I readily accept the power of nature but still struggle to deny the power of nurture. Twin and adoption studies of health, intelligence, happiness, success, character, and values seem to contradict firsthand observation. Haven’t you repeatedly seen parents change their children’s lives? I have. Don’t you recall many times when your parents changed your life? I do. It’s tempting to treat all of this research like a “proof” that 1+1=3: Clearly wrong, but life’s too short to figure out why.
If that’s what you’re thinking, please pause. There is a simple way to sync science and common sense: The short-run effects of parenting ar
e larger than the long-run effects. If you think you’ve changed your kids’ behavior, I bet you’re right. You could go and change it some more right now. Feel free. The catch is that your efforts won’t last. The immediate, visible effects of nurture tend to wear off or “fade out” as children grow up. Think about all the times in your childhood when you got in big trouble and vowed, “I’ll be good from now on.” How long did your change of heart last? A month? A week? Five minutes? Parents could conceivably compensate for fade-out with extra persistence. But time is against them. No matter how strict or encouraging parents are, their children eventually grow up, and fade-out eventually kicks in.
If you want to discredit twin and adoption research with eyewitness testimony, I have no objection. I only ask to cross-examine you. You’ve seen parents change their children’s lives. You recall times when your parents changed yours. But haven’t you also seen how temporary these changes tend to be? Think about how many parents try to inspire a love of books. At the time, they succeed; lots of kids won’t sleep without a bedtime story. Ten or twenty years later, however, only a small minority read for pleasure. That’s fade-out for you.
If the short-run effects of parenting are as large as I say, why don’t twin and adoption studies report them? The answer is that they rarely look for them. Researchers typically focus on long-run effects—whether your parents change the kind of adult you become. That is why they reach the weird conclusion that nurture doesn’t matter much. When twin and adoption studies focus on the short run—whether your parents change the kind of child you are—they typically support the obvious conclusion that nurture matters. The research is a little sparse, but studies of intelligence, income, crime, and religion all find that upbringing matters less as we grow up.