‘When life hands you a lemon, just bite in’

Judith Rich Harris is a psychologist and author.

I first become aware of you when I read The Nurture Assumption in 1998. In it you proposed that a child's peer group has greater influence on development than her parents. Can we begin by outlining this theory?

Group socialisation theory was my attempt to solve a puzzle I had encountered while writing child development textbooks for college students. My textbooks endorsed the conventional view of child development – that what makes children turn out the way they do is 'nature' (their genes) and 'nurture' (the way their parents bring them up). But after a while it dawned on me that there just wasn't enough solid evidence to support that view, and there was a growing pile of evidence against it. The problem was not with the 'nature' part – genes were having their expected effect. But 'nurture' wasn't working the way it was supposed to. In studies that provided some way of controlling for or eliminating the effects of heredity, the environment provided by parents had little or no effect on how the children turned out.

And yet, genes accounted for only about 50 per cent of the variation in personality and social behaviour. The environment must be playing some role. But it wasn't the home environment. So I proposed that the environment that has lasting effects on personality and social behaviour is the one the child encounters outside the home. This makes sense if you think about the purpose of childhood. What do children have to accomplish while they're growing up? They have to learn how to behave in a way that is acceptable to the other members of their society. How do they do this? Not by imitating their parents! Parents are adults, and every society prescribes different behaviours for children and adults. A child who behaved like his or her parents (in any context other than a game) would be seen as impertinent, unruly or weird. So the first step in becoming socialised must be to figure out what sort of person you are. Are you a child or an adult? A male or a female? In complex societies there are more categories, but age and gender were probably enough for the small groups of hunter-gatherers of our ancestors.

Once a child had identified with a particular social category – let's say, female child – her next job would be to learn how to behave like the others in her category. A social category is an abstract concept, not necessarily an actual group of children. My use of the term 'peer group' turned out to be misleading. I should have said 'social category' or perhaps 'reference group'.

Why?
The problem with 'peer group' was that it made people think 'friends'. Group socialisation theory is not about the influence of friends. Friendships are relationships. Socialisation is not a product of relationships.

The expanded theory presented in my second book, No Two Alike, explains why. The theory is based on the idea, put forth by evolutionary psychologists such as Leda Cosmides and John Tooby, that the human mind is modular, a collection of specialised devices which each evolved as a solution to a specific problem or need. I proposed that there are three such devices involved in social development – the relationship system, the socialisation system and the status system. These systems work more or less independently; sometimes they even issue contradictory commands. They collect different kinds of information from the environment and process it in different ways. Friendships – like parent–child relationships – are in the purview of the relationship system, which collects data on specific individuals and makes fine distinctions among them. The socialisation system, in contrast, doesn't bother with individuals – it computes means. It forms a prototype for each social category. The child is influenced by the norms of the social category she identifies with, even if she never interacts personally with any of its members.

The Nurture Assumption completely split the field. People either said it was a serious step forward in our understanding of child development or they just weren't having any of it. I'm in the first camp – it changed my view of child development. But why do you think there was so much hostility?
Part of the problem was the media coverage, which was often headlined 'Do parents matter?'. Parents were understandably irked by the question. (My answer, by the way, is: Of course parents matter!) But the real opposition to my work came from the academic world – from professors of developmental psychology. Some of these people had spent their entire careers doing studies designed to support the traditional view of child development. Then some troublemaker pops up – a complete nobody, with no PhD and no academic affiliation – and announces that the professors are wrong and their studies are worthless. You wouldn't expect them to greet me with open arms, would you?

You were particularly critical of their correlational studies of development.
I still see those worthless studies all the time – they get a lot of publicity. I see them as a shameful waste of time and research money. I see them as reminders  that I failed in my goal of reforming the methodology of developmental psychology.

The studies are worthless because the results they produce are ambiguous, so the researchers can interpret them any way they please. Let's say they find a correlation between how often a family eats dinner together and how well their teenager manages to stay out of trouble. Such results are presented as evidence that eating dinner with their parents has 'protective' effects on teenagers. But the research method provides no way of controlling for, or estimating, the effects of inherited genes on the teenagers' behaviour. (Conscientious parents tend to have conscientious children.) No way of controlling for what I call 'child-to-parent effects.' (Parents are more likely to enjoy eating dinner with well-behaved teenagers.) No way of controlling for the teenagers' own willingness to show up at dinnertime. (Teenagers are less likely to enjoy eating dinner with their parents if they are doing things their parents don't approve of.) The researchers assume that, even though these other factors might play a role, some of the correlation must be due to the beneficial effects of family dinners. That is a logically indefensible assumption, not supported by studies that do provide the necessary controls.

I don't think it's an exaggeration to state that The Nurture Assumption pretty much made you famous almost overnight. It's not only a radical alternative to traditional ideas, but also a real 'page-turner'. Was it your intention to write in that style?
Actually, I started out by writing a traditional article and publishing it in a traditional journal, the Psychological Review. No one called it a page-turner. In fact, though it did get some favourable responses from people in other areas of psychology, it was completely ignored by the audience I was hoping to reach – those professors of developmental psychology.

So I decided to go over their heads, as it were, and take my message directly to the general public. If you're writing a book on a complex topic and you want people to read it, you have to make it interesting. It also helps if you can give your readers an occasional laugh. My model for how to write a book for the general public was Steven Pinker's The Language Instinct.

Pinker of course went on to write several more books for the public – all page turners, and in many cases game changers. I noticed that he dedicated The Blank Slate to 'Don, Judy, Leda and John'. I would assume that three of these are Don Symons, Leda Cosmides and John Tooby. Would I be right in thinking you are the Judy?
Yes. Steve and I became e-mail friends after I sent him a copy of my Psych Review paper and some comments on The Language Instinct. After we had exchanged a few e-mails, he asked, 'Have you ever thought of writing a book?' It wasn't exactly a new idea to me, but it was nice to have the encouragement.

I think Steve was particularly receptive to my arguments because he's a psycholinguist. I often use examples from psycholinguistics in explaining my theory, for two reasons. First, the outcome is usually obvious. You don't need fancy statistical tests to decide whether or not someone has a foreign accent. Second, language and accent are among the very few social behaviours in which genetic differences play no role at all. Whether you speak Japanese or Swahili, whether your accent is Oxbridge or Liverpudlian, has nothing to do with heredity. But it does have a great deal to do with social context. The children of immigrants have the same accent as the other kids in the neighbourhood, even if they use their parents' native language at home. Around the world, it is quite common for children to use one language at home and a different one outside the home, or one language with Mummy and a different one with Daddy.  

A central tenet of my theory is that social behaviours are tightly linked to the context in which they were acquired. It's a mistake – one that's incorporated into all the major theories of child development – to assume that children automatically generalise what they learn, from one context or person to another: Mummy is nice to them so they expect everyone to be nice to them. But discrimination, not generalisation, is the default setting of the baby's mind.

Many of the behaviours that children acquire at home would be counterproductive elsewhere. Children who dominate their younger siblings at home would be making a mistake if they tried to treat their schoolmates the same way, especially if they happen to be small for their age. Fortunately, children don't make that mistake. Firstborns are no more likely than laterborns to try to dominate their peers.

Of course, some of the things children learn at home are useful elsewhere. Those who learn to speak the local language, or to read, or to play a musical instrument, don't have to acquire these skills all over again when they step outside. But they don't trot them out automatically. They are tentative at first, until they're sure that the behaviour or skill they learned at home will also work in the new setting.

For a young child, it's safer to discriminate than to generalise. The child's mind is not short of storage space. A child can store different rules of behaviour for every setting, and different expectations for every individual he or she encounters.

Your goal in No Two Alike was to explain why individuals differ so much, even if they grew up in the same family, right?
Right. I realised a couple of years after The Nurture Assumption was published that I had done only half the job: I had explained only how children get socialised. Socialisation is a process that causes children to become more similar in behaviour to their same-sex peers. And yet, despite being socialised, children continue to differ from one another in personality and social behaviour. If anything, the differences widen during childhood and adolescence. I made some ineffectual efforts to deal with that problem in The Nurture Assumption, but I didn't have a theory to account for it till I wrote the second book. The improved version of the theory presented in No Two Alike explains how children can, at the same time, become more similar to their peers in some ways and more different in other ways.

There was a fair bit of replying to arguments put forward by critics of The Nurture Assumption. Was that one of the aims?
It was. I was tired of journalists telling me that my theory must be wrong because some expert at some big university had told them that there were plenty of studies that disproved it. I searched diligently for the studies they cited. In some cases they were nowhere to be found; at any rate, they had never been published in a peer-reviewed journal. In other cases a study had been published but the results didn't do what the experts claimed – they didn't disprove my theory. In one case, a study they cited actually did the opposite – it supported my theory!

That 1995 Psychological Review piece you mentioned won the George A. Miller award for an outstanding article in general psychology. There was a certain irony about that?
In 1960 I was a graduate student in the Department of Psychology at Harvard. One day I got a letter saying that the Department had decided to kick me out of their PhD programme. They doubted I would ever make a worthwhile contribution to psychology, the letter said, due to my lack of 'originality and independence'. The letter was signed by the acting chairman of the Department, George A. Miller!

Sometimes, when life hands you a lemon, you should just bite in. Getting kicked out of Harvard was a devastating blow at the time, but in retrospect, it was the best thing that Harvard ever did for me. It freed me from the influence of 'experts'. It kept me from being indoctrinated. Many years later, it enabled me to write The Nurture Assumption.