Opinion-Policy Nexus

Following on Chris's theme of genes and politics, I want to remark on a contrast in his post between the approaches political scientists have adopted from behavioural and population genetics, on the one hand, and findings on the health benefits of vitamin D, on the other. Those findings have to do with gene regulation, and are part of the burgeoning field of epigenetics.

This field has longstanding roots in older work on transposons, noncoding RNA, and various mechanisms regulating genetic expression and development and evolution, the deeper implications of which are only now being fully realized.

There's a nice piece in a recent issue of the Economist that captures the excitement and importance of these developments, but also the humbling complexity unfolding in the life sciences around the big questions of genetics. The report riffs on the findings of David Kelley and John Rinn, who study so-called lincRNA, and have discovered, among many other fascinating things, curious relationships between these molecules and transposable elements (my wife is a geneticist whose early interests were in transposons, and she has reminded me that McClintock, in her pioneering maize research, presciently called these "controlling elements").

Why does this matter to political scientists interested in relationships between our genes and our political attitudes and behaviour?

In one respect it doesn't: if there are durable heritable roots to the psychological dispositions that inform political phenomena, then we want to be precise about those relationships, not least to know where to direct future research in both genetics and social sciences. We don't need to worry too much about the causal complexity that leads to those stable heritable dispositions.

On the other hand, science is ultimately about explaining that complexity, so if political scientists want to be players in that game, then simply pointing, as several of these studies do, to correlations and specific clusters of genes, is a good start, but not in itself a very satisfying explanation.

True, many studies of political behaviour proceed as if ideas and environment were everything, and so these 'genes and politics' studies are a useful corrective to that view. Still, the humbling possibility is that ongoing work on regulation and expression may force behavioural and population geneticists to rethink some of the durable assumptions behind their methods. While scholars involved in the 'genes and politics' field recognize the inordinate complexity of genetic regulation and organism development, they tend to fall back on the familiar old definitions (what a gene is) and assumptions (the stability of protein conformational states), and the methods they've inherited from the great innovations of the past in behavioural and population genetics. Those methods remain powerful, but the old assumptions and definitions are, alas, increasingly simplistic or outright mistaken.

Consider this gloss from that Economist article:

Once, and not so long ago, received wisdom was that most of the human genome—perhaps as much as 99% of it—was “junk”. If this junk had a role, it was just to space out the remaining 1%, the genes in which instructions about how to make proteins are encoded, in a useful way in the cell nucleus.

That, it now seems, was about as far from the truth as it is possible to be. The decade or so since the completion of the Human Genome Project has shown that lots of the junk must indeed have a function. The culmination of that demonstration was the publication, in September, of the results of the ENCODE project. This suggested that almost two-thirds of human DNA, rather than just 1% of it, is being copied into molecules of RNA, the chemical that carries protein-making instructions to the sub-cellular factories which turn those proteins out, and that as a consequence, rather than there being just 23,000 genes (namely, the bits of DNA that encode proteins), there may be millions of them.

I think applied behavioural geneticists wandering into the social sciences, and the political scientists adopting their methods, should be humbled by that complexity, even as they do increasingly good work identifying heritable dispositions associated with politics. There is a certain elegance to the statistical and experimental methods these scholars have inherited, but the wonderfully messy complexity of genetic regulation and expression may, as it becomes increasingly well-understood, force us in the social sciences to ask hard questions about just what we are actually explaining with the tools we inherited from an earlier generation.

But of course that's science.