What’s the point of parenting?

As an aside in a recent blog, I made the statement that, “shared environmental effects like parenting have no effect on adult’s behaviour, characteristics, values or beliefs.” This excited quite a bit of handbag clutching so I’ve decided to delve a little more deeply into the evidence supporting this contentious claim.

It is, I hope, unlikely to upset anyone to point out that identical twin share (virtually) 100% of their genetic material*. It’s also uncontroversial to note that despite this, there are often observable differences in the behaviour and personalities of identical twins. What accounts for these differences is referred to as ‘the environment’. Behaviour geneticists have identified two broad categories of environmental effects: shared and non-shared. Shared environmental effects are those that siblings have in common and, by definition, these cannot account for the differences between identical twins because, well, they share them. Therefore, all differences must be non-shared in origin. Essentially these are environmental factors that are unique to each twin. Possibly they have different friends, maybe they’re in different classes at school and so have different teachers, and so on.

Still with me? In his appendix to What If Everything You Knew About Education Was Wrong, Andrew Sabisky writes the following:

Surprisingly, it would appear that environmental influences on traits are, by adulthood, almost entirely non-shared in origin. Or to put it another way, while the nurturing effects of family has some importance in childhood, it wears off almost completely by adulthood. In the long run, we resemble our families because of our shared genes, not our shared experiences. This pattern holds true not just for IQ but almost every trait you can imagine. As shared environmental influence reduces to zero, heritability – the statistic that gives us the percentage of genetic variance on the trait in question – usually rises with age. For intelligence we find heritability coefficients of well under 50% in early childhood, but by adolescence and adulthood these rise to 70-80%, and even higher figures have been reported for the g factor. It is common to find heritabilities of about 50% or higher for most behavioural traits. [my emphasis]

He also provides this handy table of heritability estimates for different traits:

Heritability is not a fixed value and does not apply to individuals. It is not an attempt to sum up the percentage to which we can attribute the effects of your genes on your behaviour, rather it is a probabilistic statement about the chances that genes and environments contribute to the differences between one person to another within a population. Universal education, the Welfare State and other organs of modern society act to raise heritability estimates because the environmental differences between people are less pronounced than in the past. The better a school’s behaviour policy, the more likely it is that all pupils will behave well, regardless of their genetic predisposition. In a school with no effective structure for managing behaviour, some children will still do the right thing, despite the chaos unfurling around them.

The exact same argument can be made about the quality of teaching. In a school with a well-designed curriculum and where children are well taught, all pupils will learn more (although some will, of course, learn more than others) whereas in a school with little in the way of curricular provision and where most of the teachers are bumbling loons, all pupils  will learn less (although some will, of course, learn less than others).

So, what has all this got to do with parenting? It doesn’t take much of a stretch of the imagination to guess that if shared environmental effects have no lasting affect on identical twins, there’s little likelihood that parenting will have much more impact on anyone else. But, this runs counter to every parent’s lived experience. We see our children ape our mannerisms and take on our beliefs and values; how could this not be the result of parenting?

It turns out that while shared environmental effects are often strong during childhood, they wear off by adulthood. This makes some sense. After all, children have little control over their environments when they’re at home, but, as soon as they start school, alien mannerisms and foreign values begin to enter a child’s repertoire, until, by the time we reach adulthood we are free to behave exactly as we please (except perhaps when we return to visit our parents.)

This paper by Thomas J. Bouchard, Jr. summarises much of the research into the role of genes on human behaviour and finds that, “Shared environmental influences are often, but not always, of less importance than genetic factors, and often decrease to near zero after adolescence.”

As you can see in the above table, shared environment effects are very low or non-existent for most traits in adults. Aha! – you might be thinking – it looks like parenting can affect ‘religiousness’ to a moderate degree, and look, various other traits show some small effects. Clearly the claim that parenting has no effect is obvious bunkum.

Well, that could be the case, but these small estimates are often equally attributed to measurement errors and unaddressed confounding factors. As I’ve previously made clear, extreme neglect or abuse – especially before the age of 5 – is likely to do lasting damage and often results in permanent cognitive impairment. You could chalk this up for a win for parenting, but personally, I can only see this as a bad thing. Perhaps it’s better to say that there’s no evidence for positive effects of parenting on adult behaviour.

In her book The Nurture Assumption, Judith Rich Harris presents three key propositions:

  1. Parents have little or no power to shape their children’s personalities
  2. Children are socialised, and their personalities shaped, by the experiences they have outside the home, in the environment they share with peers.
  3. The tendency for an individual to behave in a somewhat similar fashion in different social contexts is due, in most cases, to a genetic predisposition to behave in that manner.

The evidence she amasses in support of these elegantly constructed propositions is truly impressive and, for those who remain unconvinced by my poor arguments, you really should give this eminently readable book a go. For those who unwilling to undertake such a chore, just read Appendix 2 in the link above (p 357 – 369). If you’re short on time, it might even be worth just reading Chapter 2, The Nature (and Nurture of the Evidence.

So, what is the point of parenting? Well, as the proud father of two wonderful daughters if you really need to ask, then I worry for you. Surely there doesn’t need to be a point for us to do the best for our children? The point, if there is one, is that being nice to your kids improves the quality of their childhoods. And, if you need further incentive, the one enduring effect of parenting is the quality of the relationship you will have with your adult children – the more effort you make to be kind, fair and supportive, the more likely your children are to like you and the less likely they are to dump you in a retirement home.

Harris ends her book with these words:

The bond between parent and child lasts a lifetime. We kiss our parents goodbye not once but many times; we do not lose track of them. Each visit home gives us opportunities to take out family memories and look at them again. Meanwhile, our childhood friends have scattered to the winds and we’ve forgotten what happened on the playground.

When you think about childhood you think about your parents. Blame it on the relationship department of your mind, which has usurped more than its rightful share of your thoughts and memories.

As for what’s wrong with you: don’t blame it on your parents.

I understand that all this is hard to accept. I understand that sharing this sort of research directly contradicts people’s cherished beliefs about the way the world works and I get that the power of cognitive dissonance makes some people want to lash out, nit pick and hurl insults. Fine. But please try to bear in mind that I’m not trying to upset you or devalue your experience. If you want to get angry, read Harris’s book rather than attacking this blog. If you’re unconvinced, so be it. This tweet sums up the problem beautifully:

Those looking for more might enjoy this post does a great job of bringing together huge swathes of research to address the claims of Tiger Mom, Amy Chua.

* Except for a few minute mutations.

20 Responses to What’s the point of parenting?

  1. @davowillz says:

    This seems far more measured and sensible. How do they determine whether parenting effects have disappeared by adulthood or not?

  2. While I more or less agree with this position myself when I read Pinker’s work. I do have a couple of unresolved thoughts.

    Why do we see certain cultures doing much better (or worse) than others within the same education system. Mark McCourt showed me a charts which separated British maths students by ethnicity and there were some huge differences. Is this not in large down to the culture being supplied by the parent? How would these large differences be explained if not down to parenting like the Chinese tiger mom.

    While on average parents have little affect (I think Pinker says it’s about 1-10%) Awful parents could have a huge difference which wouldn’t show up in the average. If we look at some of the worst behaved students then we might find they have the worst parents?

    Now 1-10% sounds small, almost insignificant, hence the handbag clutching. I’m watching the athletics as I type this. In a very tight field 1-10% can make the difference being last and first. Given that, at least once a child is born, we can’t change any genetic differences (yet) then maybe parenting is the most important thing to focus on? Maybe that parenting of 1-10% is the difference between a bottom set straggler and a flourishing A* student?

  3. […] Een ander Nature-Nurture debat: hoe groot is de invloed van opvoeding door ouders? […]

  4. Anne Constable says:

    Really interesting and thought provoking read. Particularly interested in the impact of quality school provision/systems on all children. Where can I read more about the nature of damage and possible cognitive impairment likely to result from ‘extreme neglect or abuse – especially before the age of 5’?

  5. This is a well-explained, easy to understand read. I don’t doubt the evidence on heritability and the tables are useful illustrations of this.

    One thing I’m wondering about is this:
    “Children are socialised, and their personalities shaped, by the experiences they have outside the home, in the environment they share with peers.”

    Do parents not have some control over the experiences they provide for their children? Eg the school choice, neighbourhood, activities and clubs joined, holiday destinations, play dates with other children arranged, other cultural experiences. Or comversely, the lack of all of those deliberate choices by some parents, often due to socio-economic background.

    Why are your thoughts?

    • David Didau says:

      Yes, parent do have *some* control over these sorts of things, but less and less as children get older. By the time they’re teenagers, parents have very little say in their children’s peer groups. That said, there’s always the nuclear option: if your child falls in with a bad crowd the only effective step you can take is move house. Few teenagers have the determination to maintain peer relationships over a long distance and may (hopefully) gravitate to a more conducive friendship group. Then again, they may not.

  6. […] In my last post I showed that shared environmental factors, like parenting, have pretty much no longterm effect on adults and that pretty much every human trait can be attributed fairly equally between heritability and non-shared environment. […]

  7. As Bryan Caplan has pointed out, the small net effect of parenting could be the result of different effects cancelling each other out, so that being strict about, for example, not smoking might make some children into lifelong non-smokers, while others begin smoking just to rebel.

  8. willorrewing says:

    Sam Harris and Charles Murray talk about this issue well on a recent podcast, citing Judith Harris’ work: https://www.samharris.org/podcast/item/forbidden-knowledge

  9. […] that allows people to refute the existence of things they don’t like so, when presented with evidence about the surprising lack of parenting effects, they will say something along the lines of, “Well, I think parenting must have more effect […]

  10. […] uproar on Twitter following David Didau’s recent blog posts that question the effect that families and schools have on children. I suspect that many teachers go into teaching in order to make a […]

Constructive feedback is always appreciated

%d bloggers like this: