Book Review: The Genetic Lottery: Why DNA matters for social equality

The Genetic Lottery: Why DNA Matters for Social Equality

By Professor Kathryn Paige Harden

Published by Princeton University Press

ISBN-10: ‎0691190801, ISBN-13: ‎978-0691190808

Buy this book from Amazon UK

'Genetic differences between people create differences between them in their likelihood of having speech and language problems. Genetic differences between people create differences between them in their likelihood of being homeless. The first sentence is not particularly controversial; the second one almost definitely is. But why?'

In this quote, from Professor Kathryn Paige Harden's book 'The Genetic Lottery: Why DNA Matters for Social Equality' the reader is reminded that behaviour(al) genetics research finds that pretty much every human trait, and life outcomes, are subject to wide variation, some of which is associated with genetic variation. Is this in itself controversial, or is the causal implication ('create') the controversial part?

'People need not stake their claims to equity and inclusion on genetic sameness, or on the irrelevance of genetic difference for human psychology. Rather, the problem to be fixed is society's recalcitrant unwillingness to arrange itself in ways that allow everyone, regardless of which genetic variants they inherit, to participate fully in the social and economic life of the country.'

Here Professor Harden outlines what she calls her 'anti-eugenic approach'. For the eugenicists, genetic differences implied fixed hierarchies not just for trait scores but also for moral worth. For many on the left today, as she sees it, the best approach is thought to be a 'genome-blind' one. For Harden, in contrast to both, the best approach is to separate difference from hierarchy and moral worth, but more than that, to positively use knowledge of difference to promote more equitable outcomes. She calls this third approach an anti-eugenic one.

The philosopher John Rawls asked us to think about how we would like the world to be if we did not know how we would place in it ('the veil of ignorance'). Professor Harden asks us: 'What sort of society would you want if you didn't know what the outcome of the genetic lottery was going to be?

The book is divided into two parts, 'Taking Genetics Seriously' and 'Taking Equality Seriously', addressing the issues raised in the two quotes at the start of this review.

In the first part, Professor Harden is at pains to explain that any genetic causes of individual differences within ancestral groups for a trait such as educational attainment cannot be taken to explain observed average differences between ancestral groups. She is also at pains to explain the connections and differences between ancestral groups and socially defined racial groups. A number of reviewers have argued that she fails to do this convincingly or properly. It seems to me that, on the contrary, she is careful and rigorous about this. At the same time, it is clear that individual differences within ancestral groups is her main interest, which is where I turn my focus.

Back in 2013, Professor Harden agreed with her former supervisor, Professor Eric Turkheimer, who argued then, and pretty much still argues now, that 'conducting genome-wide association studies (GWAS) was like studying the pits on a music CD to try to understand whether a song was any good... Human behaviour was just too complex' (Harden's summary, not a direct quote). But now, she says: 'we were wrong'.

A decade or more ago geneticists attempted to find individual genes with non-trivial influence, often based on a plausible biological model. This was the candidate gene approach. It was a dismal failure.

GWAS, in contrast, use large numbers of human subjects to build up a picture of the cumulative contribution of lots of very small variations, typically without attempting to relate these to a plausible biological model. This has been far more successful, in part because it is almost certainly a truer picture of the underlying genetic architecture, and in part because technology and computing power has advanced.

Twin and other studies suggest that around half the variation in traits such as educational attainment are associated with genetic variation. GWAS have, for defined populations, found many genetic variants that, cumulatively, account for more than 20 percent of that 50 percent, ie, over ten percent of the total variation. This, as Professor Harden points out, is a higher proportion than is accounted for by most environmental factors. What does this mean for an individual? Using this information, an individual can be genetically tested and given a score or a ranking on the known genetic variants. Proponents of this idea argue that this has predictive power: as an odds ratio, the difference between the top ten percent on the GWAS score and the bottom ten percent was a five-fold multiple in the rate of college graduation.

None of this means that we can understand or appreciate songs at the level of GWAS, as Professor Harden points out using various analogies. But it does mean that variations in many things we do not think of as genetic are linked to genetic variations.

But wait, the observant reader will notice that in the second paragraph I wrote about genetic variation being 'associated' with variation in an outcome. I have just used the word 'linked'. In between I used the word 'cause'. The precise meaning of these different words in context is undoubtedly a source of confusion and controversy.

Professor Harden follows David Lewis in arguing that cause is an appropriate word; specifically a genetic cause is best understood counterfactually. It 'makes a difference, and the difference it makes must be a difference from what would have happened without it.' This is a reasonable idea. Various studies, with different forms of controls, have established causation of this kind, with appropriate caveats about findings being limited to place and time, and not actually knowing the complexities in the chains of causation.

I have given merely a taste of the issues covered in the first part of the book. There is a voluminous technical, semi-technical and popular literature on the issues Professor Harden addresses in Part I. In recent years Professor Robert Plomin, in Blueprint, and Dr Kevin Mitchell, in Innate, have covered some of the same ground in books pitched at roughly the same audience, the 'intelligent lay reader'. The busy lay reader may reasonably ask whether Professor Harden adds to this literature. I have no hesitation in saying that she does. At the very least, comparing and contrasting The Genetic Lottery with books such as Blueprint and Innate is a rewarding exercise. On the other hand, for anyone unfamiliar with other recently published books in this area, Professor Harden's book works as an introduction to this fascinating and troublesome topic.

When it comes to trying to use genetic information for progressive ends, the subject matter of the second half of the book, I'm less convinced, or less convinced in general.

Professor Harden is on strong ground when she argues that sociological analyses and interventions can sometimes be ill-conceived, and a waste of valuable time and money, if they do not address genetic confounding. 'Using nature to understand nurture', the title of Chapter 9, highlights some of these issues.

Further to this, knowledge of widespread genetic variation can motivate the design of systems to compensate. For example, knowing that we all vary in health, in part for genetic reasons, leads Professor Harden to conclude that we should compensate by providing healthcare for all based on need not health profile. This is in essence an American arguing for a more British/European system of healthcare.

However, Professor Harden goes further than merely suggesting genetic understanding can help sociological and political perspectives: 'failing to take genetics seriously is a scientific practice that pervasively undermines our stated goal of understanding society so that we can improve it.' On the final page of the book she brings this idea to a conclusion: 'A society that protects – nay, loves – its most vulnerable with its choices must be able to see who is most vulnerable, so that it can see how its choices affect them.'

But what exactly does 'see' who is most vulnerable mean? Professor Harden is opposed to the suggestion by Professor Plomin that testing might be used at an individual level in recruitment processes. Regarding her proposals on healthcare, there is no individual genetic testing involved. More than this, she explicitly notes that the source of the variation in need is irrelevant: 'a more fully anti-eugenic response to the rising tide of genetic discoveries is a commitment to truly universal health care, to which everyone has access, regardless of the outcome of the genetic lottery (or the environmental one).'

Similarly, would genotyping children who are doing well (or badly) in school inform our teaching of them? Professor Harden doesn't suggest this and I'm struggling to see how it would help (the 'phenotype' is all we need and can use, for now at least).

In summary, it seems to me that beyond using genetics to inform sociological study (chapter 9, a very important topic), Professor Harden is mainly using the genetic lottery to argue for a Rawlsian approach at a general level. There is nothing wrong with this of course, but it does also suggest that Professor Turkheimer had at least half a point when he put forward his 'gloomy prospect' back in 2013.

Buy The Genetic Lottery: Why DNA Matters for Social Equality from Amazon UK.