Switching the parents around

In a 1995 paper in Psychological Review, I proposed a new theory of child development, based on the idea that children's personalities are shaped, not by their parents, but by the environment they encounter outside the home. This proposition, I said, doesn't imply that children can get along without parents. What it does imply is "that children would develop into the same sort of adults if we left them in their homes, their schools, their neighbourhoods, and their cultural or subcultural groups, but switched all the parents around."

It was a thought experiment – I wasn't suggesting that parents should actually be switched around. What I was saying was that, given a child's genetic makeup, and given the child's environment outside the home, the environment provided by the parents inside the home would not have any noticeable impact on the child's adult personality.

The experiment is an important one but it cannot be done, and not only for practical and ethical reasons. For one thing, there's no control group. We'd need two identical universes so that we could switch the parents around in one and leave them in place in the other. Then we could compare the children in the two universes. We'd have to compare them one by one, because my prediction wasn't about group averages – it was about individual differences. But that wouldn't work either, because we already know that two children with identical genes and essentially identical outside-the-home environments – namely, reared-together identical twins – don't end up with the same adult personalities. (The personality differences between reared-together identical twins is a mystery I address in my 2006 book, No Two Alike.)

There are ways to work around these problems and show that, given a child's genetic makeup, and given the child's outside-the-home environment, the environment provided by the parents inside the home makes no noticeable difference in the long run. But it involves putting together evidence from many different sources. This evidence already exists. For example, evidence exists that identical twins reared by different parents are (on average) as similar in personality as those reared by the same parents, and that adoptive siblings reared by the same parents are as dissimilar as those reared by different parents. Evidence exists that children reared by immigrant parents have the personality characteristics of the country they were reared in, rather than those of their parents' native land. Evidence exists that environmental differences within the family, such as those associated with birth order, leave no long-term marks on children's personalities. Even in childhood, firstborns do not behave differently from laterborns when they are outside the home, playing with their agemates.

Is it less convincing to put together many little bits of evidence (as I did in The Nurture Assumption and No Two Alike) than to point to a single grand experiment that proves one's thesis conclusively? It certainly requires more patience from one's audience. But sometimes the piecemeal approach is all that is possible.

Judith Rich Harris is the author of The Nurture Assumption and No Two Alike. She is an independent investigator and theoretician.