The natural selection paper, part 3: reflections

People's genes are changing. Does it matter?

This is my third post about my new paper with Abdel Abdellaoui on contemporary natural selection. Here are the first and second posts. Today I want to talk about the big picture.

First, a little meta. In earlier posts, I had my scientist hat on. I have some claim to expertise about the work we did… I hope! This post is more about my thoughts as an ordinary person, as to how society should react to these findings, if they hold up. I don’t claim special expertise on that. The opinions expressed here are quite personal: they are my own, and do not reflect the views of any institution, or even of my co-authors.

Okay, natural selection is happening in the population of Great Britain. Why does that matter? Does it have practical implications?

One answer is “we have no practical concerns, it’s simply interesting to understand how natural selection works”. But that seems a touch disingenuous. I suspect that research on this topic is not just motivated by theoretical curiosity. If natural selection is making people less healthy and smart, then that could be a practical problem. Without prejudging whether it is a serious problem, let’s admit that the question is important.

At this point I must acknowledge what is sometimes referred to, rather abstractly, as the dark history of 20th century eugenics. It may help to be concrete. Below is a picture of Ernst Lossa. Contemporary accounts say he was a naughty boy. For his naughtiness, he was taken away by the Nazis and murdered. That is what happened the last time people started to worry about dysgenic selection.

Yet, despite this terrible history, it could still be true that natural selection was imposing serious health burdens on modern societies, and if it were true, we would have to acknowledge it and decide what to do. So, is it true?

Up front, I should say that our research does not prove that. You can read our paper and go away believing that the genetic changes it details are theoretically interesting, but practically no big deal. I give you license to do that, if you prefer.

The reason is that the effect sizes we describe are small. For example, our correlations between the “EA3” polygenic score for educational attainment, and realized fertility, predict that the next generation will have an average EA3 score of –0.04 to –0.08, compared to this generation’s average of 0. The standard deviation of the score is 1, so this is a bit less than one-twentieth to one-tenth of a standard deviation. That is not a big change. Especially, in the context of the big environmental changes over the same time period — many more people going to university, or leaving school at 18 instead of 14 or 15 — it seems trivial.

But that is not quite the whole picture. Our effect sizes are small, but our measures are noisy. The EA3 score captures only a small proportion of the total contribution of genes to educational attainment, as estimated from twin studies. That is because polygenic scores are still very inaccurate, compared to where they will eventually be. If we had more accurate polygenic scores, two things might change about our results. First, clearly, a given shift in the score would make a bigger difference to the predicted phenotype of education, ADHD or whatever. Second, perhaps, our effect sizes — correlations with fertility — might get larger. That is, since we are using an inaccurate score, we may underestimate the true effect of the “real” score. (The technical term is “errors in variables”.)

Another aspect to consider is the trade-off between environmental and genetic ways to improve human performance. Imagine when, far back in prehistory, humans started to wear clothes. Conceivably, this reduced the selection pressure for body hair. Some traditionalist might have complained that humans were getting soft and losing their fur. Well, so what? Clothes are better: only my dog disagrees. Similarly, suppose the invention of spectacles helped myopia, which is very heritable, to spread among the population. Again, that probably wouldn’t be a big deal, because we have spectacles to compensate for it.

Is intelligence like that? People might have genes that make their brains a bit slower, but it doesn’t matter because we have mass education to compensate for it. Or, people are less healthy, but we have better hospitals.

Maybe. But a competing intuition, which has some evidence for it (PDF), is that raw intelligence and education are complements, not substitutes. That is, education gets more valuable if you have more raw intelligence. We certainly act that way: we try to choose the smartest kids to receive extra years of education at university. If this is so, then a decline in raw intelligence will also make education less productive. Similarly, it seems unlikely that better healthcare can perfectly compensate for worse health.

So, in my view, it is too early to tell whether contemporary natural selection is important enough to worry about in practical terms. It might not be. But it might be.

This leaves me with two conflicting intuitions. I have a strong belief that the Nazi eugenics programme was an unmitigatedly evil violation of basic human dignity. I also have the worry that maybe natural selection really is making our population less healthy, happy and smart.

One way out of this is simple pessimism. This was Charles Darwin’s point of view. Typically, he got to the heart of the matter very early on, in his work The Descent of Man. He accepted eugenic arguments that civilization was allowing less fit members of society to breed. But, he went on:

The aid which we feel impelled to give to the helpless is mainly an incidental result of the instinct of sympathy…. Nor could we check our sympathy, even at the urging of hard reason, without deterioration in the noblest part of our nature.

He concluded that dysgenic selection just had to be accepted.

The few contemporary researchers who are concerned about natural selection also seem to embrace pessimism. The authors of Modernity and Cultural Decline: a Biobehavioural Perspective suggest that the Fermi paradox (why haven’t we been contacted by alien species?) could be explained by dysgenic natural selection among ET:

It may… be biologically unavoidable that wherever intelligent life develops to a certain level of sophistication, it entropically degrades to a simpler state rapidly… countless extraterrestrial civilizations may have come and gone in the blink of a cosmological eye….

I don’t find these deep, dark perspectives congenial. Problems are for fixing. Otherwise we end up in the Land of Empty Talk. So, if in future it turns out that dysgenic selection is a serious problem, what do we do about that?

Here is one thought. As our paper shows, natural selection does not operate uniformly across society. Its effects are stronger among lower-income and less educated people, young parents, parents with more lifetime sexual partners and parents who are not currently living with a partner. There are many ways to interpret this, but one possibility is that it is linked to the collapse of traditional family structures, particularly among poor people. Remember that substitution effects, which lead people with higher incomes to have fewer children, may be stronger among single parents, or if people face a relatively high probability of becoming a single parent. That is certainly more common now than it was in previous generations.

If so, then we could improve natural selection by doing things that we ought to do anyway — things we should do, even if genetics had no effects on anything. We already have abundant evidence about the harm done to the poorest by the decline of two-parent families. This is something that conservative Charles Murray and liberal Robert Putnam agree on. Nobody seriously thinks that massive out-of-wedlock childbirth is a good thing. Decades of research point out the harm that divorce does to children. If by working harder to restore family structures, we could weaken substitution effects and dysgenic selection, that would be nice, but we have pressing reasons to do that in any case.

While in America this is a widely canvassed argument, for some reason it has not yet got acceptance in Europe. I too often hear arguments that family structure change is an inevitability that public policy should just accommodate, sometimes even described with the label “diversity”.[1] I believe these arguments are lazy, cowardly and/or smug, and harmful to society’s weakest. They would be so even if genetics did not matter a damn.

[1] In fact, in the UK 2011 census, more than 90 per cent of households with children fell into one of three categories: a married couple, a cohabiting couple, or a lone parent. Of the 25 per cent who were lone parents, more than 90 per cent were women. The simplest description of this situation is: very many women are, day-to-day, bringing up kids on their own.

In fact, our results may just reflect the genetic side of a broader phenomenon. Children get their genetics from their parents — but parents also provide them with much of their environment. Whatever process leads to parents with genes associated with lower income having more children, it is highly unlikely to involve more children growing up in a great environment. A basic and sad rule of social science is that good and bad things go together. The underlying concern, then, is that our society is not good enough at allowing children to be raised in happy families. Our results are simply one aspect of that. At least, that is how it seems to me.

---

If you liked this content, then I would love you to do three things:

1. Subscribe to this newsletter. It’s free, posts are occasional, and subscribers make me happy.

2. Share this post on social media. This newsletter is a new venture for me, so by telling your friends and/or followers, you’ll be doing me a huge favour.

3. Read about the book I’m writing. It’s called Wyclif’s Dust, too. You can download a sample chapter.