To The Nurture Assumption home page
Posted with permission on judithrichharris/tna.
there lived a thoughtful man who sought solitude in order to probe into
the problems of how things come to be. One day he was walking by
the bank of the river, absorbed in contemplation. Suddenly, he
lifted his eyes and saw a man standing near by. The man had dug a
small hole on the bank and was pouring water from a jug into it.
Surprised, the philosopher asked the stranger, "What are you doing?'
The stranger answered, "I am going to empty the river by pouring all the water from it into this hole."
"This is utter madness!" exclaimed the philosopher. "It's an impossible task!"
The stranger retorted, "Sillier and even more impossible are the questions you're trying to answer."
the admonition of the stranger in this 13th Century tale,
today we are going to explore the question of what determines personality
and character according to Judith Rich Harris in a book called
The Nurture Assumption. Harris is a little known, but soon to be
famous, psychology textbook writer. In her book she summarized
evidence that our personality and character are fifty percent inherited
with the genes we get from our parents. The other fifty percent,
she speculates, has very little to do with our parents or our home life,
but is determined, in large part, by the peer groups children associate
with from nursery school through high school.
Harris' book is an expanded version of a paper she wrote in 1995. This paper received a prize from the American Psychological Association. The prize she got was named in honor of Harvard psychologist George A. Miller, who told Harris thirty years ago that she was not good enough to get a Harvard Ph.D. Thirty years ago, Miller wrote Harris, then a graduate student at Harvard, "I hesitate to say that you lack originality and independence, because in many areas of life you obviously posses both of those traits in abundance. But for some reason you have not been able to bring them to bear on the kind of problems in psychology to which this department is dedicated. We are in considerable doubt that you will develop into our professional stereotype of what an experimental psychologist should be." With those remarks, Judith was given a terminal masters degree.
Because Harris has no professorial title, or university affiliation, many psychologists refuse to take her ideas seriously. But it is precisely this outsider view that has facilitated her lack of investment in the theories of the day. It recalls the contribution of Gregor Mendel, who failed out of the University of Vienna, and went on to become not only an Augustinian Monk, but also The Father of Transmission Genetics. Mendel, the man who discovered the laws of inheritance was told by his teacher, the famous Nägeli, to repeat the work Mendel had done with domestic peas with, Nägeli's favorite plant, Hawkweed, a plant that turned out to be asexual and hence impossible to breed. Mendel's monumental work was not accepted until it was repeated by three established scientists, thirty-five years after his original publication. These pirates did not even cite Mendel's original work. (And then of course there was Nicolaus Copernicus; the 15th century Polish astronomer who showed the Earth was not the center of the "solar" system. And there was also Galileo.) Perhaps Harris is not another Mendel, Copernicus, or Galileo but she is causing a stir that is likely to change the way we look at our world, and if you read the popular press reviews of her book, like Copernicus and Galileo, she is accused of heresy . Her heresy is that she does not believe in the "nurture assumption" and she thinks that parenting does not determine the personality and character of children.
Because of who we are and what we value, we are inclined to believe that it is essential "to cuddle our babies as often as we can right from the moment of birth", "to be cautious during toilet training", and "to worry that a false parenting move at any time from infancy through adolescence may ruin our children for life". These assumptions grow in part out of our negative reaction to the genocidal thinking of the Third Reich and the nativist bias that came with it. And, in part from the particulars of the psychology that developed among our allies. John Watson said he could make any baby into any kind of professional. B.F. Skinner said he could train any behavior that muscles were capable of. Meanwhile on the other side of the war, Konrad Lorenz said all behavior could be divided into learned or innate and if it was innate nothing could be done to modify it. If Lorenz was right Jews will be Jews, Blacks will be Blacks, and neither are Aryans like real countrymen, so we might as well exterminate them because we cannot expect to make them into the kind of citizens we want.
As a society we reacted to this genocidal thinking with resolutions not with science. For example the American Anthropological Association passed a resolution in the 1950s (not based on any evidence whatsoever) that any person from any culture could bear the behavior of any other culture equally well provided they were raised in the culture they needed to bear. So for political and not scientific we became wedded to the notion that differences in environments make all of the differences among people. We denied that any important differences among people could be inherited. (They couldn't be or else Hitler may have been right.) Of course this is not only wrong, but it is a misunderstanding of inheritance. However,as we have learn more about the nature of genes it becomes clear that what is inherited are chemicals not behavior and chemicals don't behave. They react. These reactions are modifiable. Our genes do not contain our destiny. There is no fixed relationship between genes and behavior.
So how do we study behavior? At first we did it by case studies. The trouble with this approach is that it gives too much attention to the squeaky wheel. That is we only study and report on unusual cases and we have little or no idea of the relative frequency of what we see. We convince ourselves by examination of the weird. To counter the sampling error of the case technique, and cover a wider range of cases, surveys are done with questionnaires and the answers to questions are correlated. For example, suppose we hypothesize that parents who provide an intellectually stimulating environment for their children have smarter children. To measure the "environment" parents provide we count the number of children's books a home contains and to measure the smarts of the children we use IQ. If our hypothesis is correct we should find homes with more books have children with higher IQ's. This is a correlation. A perfect correlation score is 1. That means we can predict IQ perfectly from the number of books. If there is no correlation the score is 0 meaning the number of books in no way predicts IQ. Real studies come up with correlations between 1 and 0 like .1 .2 .5 etc.
What usually happens in studies of the impact of parenting is that correlations that show up in one study are not corroborated in subsequent studies. Nonetheless two weak correlations do recur in lots of studies. The first frequently corroborated correlation is--Parents who do a good job managing their lives, who get along with others, have children who manage well and get along well. And vice versa for parents and children who don't manage well or get along well. The second frequently corroborated correlation is--Children who are treated with affection and respect tend to manage their lives and their personal relations better than children not so treated. Note these are weak correlations of .2 or .3 but they are statistically significant.
Moreover, we like these results because they support our convictions. But the interpretation that these correlations are the result of parenting is flawed because the studies fail to separate the effects of inheritance from the effects of the environment. All of the parents in these studies are the biological parents of their children and as such the parents and children have half of their genes in common. One of the ways we can separate the effects of inheritance from the effects of environment is by comparing unrelated adopted children to biological siblings. Analysis is done as follows: correlations of adopted unrelated children reared in the same home are compared to correlations of siblings or twins reared in the same and different homes. The approach is to compare the same home environments with different genes to measure the effects of genes and to compare the same genes in different environments to measure the effects of environments. The conclusion of these studies is that roughly 50% of the variation in children is attributable to differences in genes and 50% to differences in environments.
Despite repeated measures that show the impact of inheritance, many developmental psychologists have continued to assume the nurture assumption that parenting is causing the difference and genes have nothing to do with it. So they make no attempt to consider or control for heredity and assume they are measuring only the environmental effect of parenting. This "nurture assumption" is rooted in Freud, in Spock, and in our revulsion to the atrocities that occasioned WW II. And it is popular. There are loads of books about how to parent that reinforce the assumption that if you treat your kids a certain way they will turn out a certain way or perhaps if you don't they won't. Few popularizers acknowledge what we all observe in our experiences with our children. Namely, that they are intrinsically different and when we treat them the same they don't react in the same way. Perhaps the first and most obvious disillusionment came to us when we tried to raise our boys and girls to be the same and they kept reverting to being boys and girls to the point where most of us gave up and rejoiced in the differences. But the same thing happens repeatedly with children of the same sex. One is introverted and needs to be encouraged the other is extroverted and needs to be constrained. And so forth. In fact in the same household we never raise our children precisely the same way.
Fortunately, not all developmental psychologists have ignored genetic factors. And many have attempted to determine what aspects of the environment actually have the most impact on the adult personality and character of a child. The first hypothesis was that parenting styles would be the most formative aspects of a child's environment. Harris reviews several studies that have failed to show the long term impact of parenting. Here are the highlights of her review of this literature.
Maccoby and Martin (1983) concluded that "That correlations found between parent's behavior and the children's characteristics were neither strong nor consistent." They draw attention to the puzzling findings that adopted children growing up in the same home are not at all alike in personality and that even biological siblings show little similarity. Finally they say, "The implications are either that parental behaviors have no effect, or that the only effective aspects of parenting must vary greatly from one child to the other within the same family." [Harris p38]
A 1990 study of birth order concludes: "There are no systematic differences in personality attributable to birth order, so any differences in parental behavior that are associated with birth order cannot be very significant for later developmental outcomes."
A 1997 study of the impact of daycare concludes: "Recent studies have demonstrated that variation in quality of care outside the home by persons other than parents, measured by experts, proves to have little or no impact on most children's development." So working moms need not worry that they are ruining their children by assigning their care to others. 
Researchers in California who have been studying unconventional families since the 1970s conclude: That the children of hippies or others who live in communes or have "open marriages" or single mothers are as bright, healthy and well adjusted as control groups from more conventional families. 
A study of impact of the smoking habits of parents on their children concludes: It is interesting to note that the biological children of cigarette smokers are more inclined to smoke than the children on non-smokers, but the adopted children of smokers are no more likely to smoke than the children of non-smokers. This is a case of all the variation being attributable to inheritance and none to the environmental influence of parents.
Research on children of divorced parents produces an interesting exception that seems to prove the finding that parenting does not have predictable long term effects. Harris summarizes as follows: When the parents are asked about their children's behavior the children of divorce are significantly in worse shape than the children of parents who remain married. However, if the observations are made outside the home, and not by the parents, the differences, though persistent, get much smaller. And, finally, if changes in neighborhood and income are controlled for, there don't seem to be any differences at all.  It seems that most of the impact of divorce occurs to relationships within the family and is NOT carried by the children to relationships outside of the family. However, the impact of lowered income, poorer neighborhood, and disrupted peer contact that often occurs following a divorce does have measurable impact on the adult personalities of children so treated.
So this is the Harris Heresy. It is the idea that variations in parenting have no measurable impact a child's personality or character in the long run. The implications are scary but there are significant qualifications that help us out of the panic.
First, there is a way in which parenting does affect a child's personality. And it is important. Parenting definitely affects the way parents and children get along and the way children react to parents and the way children behave in the presence of their parents. Many studies have shown this. In fact this is a flaw in many studies because they measure children's behavior in the home and they often don't look children's behavior outside of their homes.
Second, Judith Rich Harris herself thinks parenting matters a great deal. When challenged to say how parenting matters on the PBS Nightly News Hour a few weeks ago, she replied: "You are essential to your children's happiness. You have the power in your hands to give your child a happy childhood, and I hope that you will give your children a happy childhood, because every child deserves that."
Third and finally, the most important reason that parenting matters is that as ethical people we don't treat our children, or others around us, well or badly because of the impact it might have on them later. We treat them well because they are human and because we love them. We think and we feel that love and care as opposed to indifference and neglect makes a better world whether it forms the personality and character of our children or whether it does not.
I call your attention to the quotation from Rabbi Dan Danson that appears at the beginning of the program:
Love is Redemptive,
The repeated failure of many studies to show impact of different parenting styles on the adult personality and character of children has led Judith Rich Harris to look outside the home for an environment that may be determinative.
Harris and her husband have two children, the first, a biological daughter, and the second , an adopted daughter. The older biological daughter was a National Merit Scholar who Harris says, "Was much like me and my husband. Didn't require much guidance and gave little trouble. She did not seem to want to do anything we didn't want her to do. And, she liked to play by herself. Her peers were the intellectual kids." But the adopted daughter was different. "As a toddler she was always at our heels and always wanted to be with people. We started getting bad reports about her as soon as she went to school. She wouldn't sit in her chair, she bothered other children. Eventually she became a drop out. Her peers were the delinquent children." Happily both daughters became well adjusted adults. But what Harris remembers is how differently they responded to the same parental treatments and how little impact they seemed to have on the adopted child. Before having children Harris accepted the nurture assumption and she wrote about it in her development psychology textbooks. But after the experience of raising her children she began to doubt it.
In her consideration that children are different outside of their home and away from their families, Harris takes a lead from the story of Cinderella. She notes that despite Cinderella's years of degradation at home she was able to charm a sophisticated prince and convince him she could fulfill the duties of a princess and ultimately a queen. Moreover, the prince doesn't even recognize her at home and her sisters don't even recognize her at the ball. This story works, for both children and their parents, according to Harris, because we are actually aware that children are likely to be very different away from home and different in unpredictable ways that have nothing to do with the way they were parented.
Harris hypothesis is that separate from their families, children learn to deal with their peers from nursery school groups on up through high school and that this peer coping experience is the most formative experience for the development of their personalties. There is some preliminary evidence to support her idea.
According to child psychologist, William Corsaro, the impact of peers can be seen in four year olds. He reports the following scene: In a sandbox Jenny and Betty play house putting sand in teapots and such. A third child, Debbie approaches the two playmates. After a few minutes of watching Debbie reaches for a teapot. But Jenny takes it away saying "No". Debbie then watches some more. Then Debbie approaches Betty, who is filling a cupcake pan with sand. Debbie reminds Betty, "We're friends, right Betty." Betty replies, "Right". Then Debbie picks up a teapot and says, "I'm making coffee". Betty says, "I'm making cupcakes". Betty turns to Jenny saying, "We're mothers, right, Jenny?" "Right" says Jenny. Jenny and Betty are defending their shared play from the outside disruption of Debbie until Debbie shows that she is ready to affiliate and play by their rules--a conformity readily accepted by the outsider who is drawn to her peers.
cites a study of picky eating among primary-school children. It
showed that some kids were picky eaters at home and some were picky
eaters at school, but almost none were picky eaters in both places.
She uses this study to show that children typically act different at home
with their families than they do away from home with their peers.
She also notes a study that asked high school students to write autobiographies describing events in their lives that made them most unhappy. While some children identified something their parents had done. Four times as many children reported how they were treated by peers. Peers had discriminated against them, laughed at them, bullied, sexually harassed, taunted, chased, or beaten them. Harris uses this study to show that what happens outside the home with peers has at least as profound and lasting an effect as what happens at home.
Harris thinks children are not particularly interested in being adults like their parents. Instead they want to be good at being children like their peers. We note as parents time and again, it is the peer group that is imitated and the peer group that they want to be like. They even want their parents to conform to the norms of the peer group. Remember how concerned your children were about how you acted and dressed around their friends?
A couple of weeks ago while I was working on this sermon, I had dinner with Jim Rosenbaum, a sociologist at Northwestern University who is studying a problem that Harris would find interesting. For several years now welfare agencies in the Chicago area have been trying an experiment. They've taken small groups of welfare families selected at random and moved them to each of several Chicago suburbs where they live near middle class families and go to school with their neighbors. The children of these welfare families join peer groups with the children of their suburban neighbors. In fact their success as measured by employment and college attendance closely approaches that of their neighbors. However, if the moved welfare groups are large enough to allow the formation of peer groups like themselves the results are NOT significantly different from what happens to these children in the ghetto. We conclude that even if the home life does not change, if the peer group does change, the results can be phenomenal. Peers certainly do make a difference.
Harris reports similar findings among the children of immigrants. If immigrants move into neighborhoods with the same foreign background and language their children end up speaking with an accent and remain relatively unassimilated. But if they move to a neighborhood that is different than the background of their parents, who may still speak their mother language at home or may speak English with an accent, the children quickly learn to speak without an accent and to assimilate the customs and values of their peers as opposed to their parents. Although at home these "assimilated" kids can still speak the language of their parents and behave like their parents want them to. As you might expect from being around your kids with and without their friends, they adopt the peer group tastes in the presence of their friends although they may be tractable to your values when you have them alone.
From my evolutionary perspective as a behavioral biologist, Harris' idea makes a special kind of sense. Animals that live in groups all seem to have a sort of adolescence. For example, mountain sheep, instead of following their elders from one pasture to the next, moving with the seasons, two-year-olds, often strike out to new ground much to the annoyance of their elders. And, male adolescent lions leave the pride of their birth and seek their fortunes with their same age brothers in other parts of the savannah. What seems to be the case with social animals is that they inherit specific abilities but need to learn a lot of things on their own by experience. One of the most important things they have to learn is not to follow precisely in their parents footsteps without trying alternatives. This is adaptive because environments are likely to change between generations and those who persevere in their parents' mode will not always to be as successful as their parents. If it were adaptive for offspring to behave like their parents then much more precise details of behavior would be inherited. That would be a much more efficient way to procreate. But the fact is that being like parents is not as successful in the long evolution of a species as experimenting with change in concert with peers. This is especially true for our species now because the times are a-changing.
Harris has many other lines of evidence for her theory that she calls "Group Socialization". She looks across cultures, and history, and at numerous cases. But, alas, she is not a Ph.D. developmental psychologist or even a member of the academic establishment. And she is not saying something that most of us, who have obsessed over our parenting, want to hear. She is an iconoclast and she is going to change the face of the way we think about what we do with our children.
If Harris proves to be right in subsequent thorough research there are clear implications for what we ought to do. Divorce laws need to be reconsidered. Maybe the children should get the house and the parents should separate around them. The importance of neighborhoods will need to be re-evaluated. The importance of small communities will be looked at differently. Cities could even become towns? When it comes to children there may not be economies of scale. Diversity is of interest to many of us, but what should we do if our children's peers have intolerable values and customs? When do we intervene and how? We probably do not have to worry about working moms or single parent families, but we sure need to concern ourselves with the children our children associate with.
Whether Harris is proven right or not, her work should not be taken as evidence that parenting does not matter. Of course it matters and it affects the way we live and love in our families, the contentment of our children, and our capacity to live out our ethical commitment to love one another. As Dan Danson has said, "Love is redemptive. We can never love too early, or too often.
Jerry Woolpy (firstname.lastname@example.org) was born and raised in Chicago in the shadow of the Ivory Towers of the University of Chicago, where he went to kindergarten, grade school, high school, college, and graduate school. He has a Master's degree in Psychology and a Ph.D. in Genetics.
Woolpy was Professor of Biology and Psychology at Earlham College in Richmond, Indiana for twenty-nine years, retiring in 1996. He taught genetics, evolution, ecology, anatomy & physiology, motivation, animal behavior, and many interdisciplinary courses with teachers in the social sciences and humanities.
His research and publications include many works on the social behavior and socialization of timber wolves. He has also worked on menstrual synchrony and human pheromones, the genetics of aggression in rabbits, courtship in Drosophila, curiosity in the cockroach, and novelty and the element of surprise in Rhesus monkeys. He also developed a computer managed educational software package called DELPHI that has been used in colleges and universities in the United States and Canada. DELPHI provides the opportunity for essay exchange and commentary with the instructor and among students in a class.
Woolpy now lives in Minocqua, Wisconsin, where he serves on various community boards, teaches occasionally at the local junior college, writes, and gives public lectures.