"One Can Make Up an Evolutionary Explanation for Anything"

Evolutionary biologist Michael Eisen has just written a terrific response to the idea that cultural characteristics of a group are the result of "natural selection." No, he says, there's no evidence that the ability to create a democratic government — or to foster social unrest — is genetic.

Eisen is responding in part to journalist Nicholas Wade's recent book, A Troublesome Inheritance, which attempts to explain cultural differences between racial groups as the product of recent evolutionary developments. Wade is right about one thing: humans have been evolving rapidly over the past several thousand years. But, Eisen explains, he's wrong about everything else:

The reason that this is issue is bigger than Wade's book is that, while Wade's representation of modern genetics is simplistic and selective, and he veers sharply into racist claptrap, the points that he is trying to emphasize about human evolution are, in broad strokes, right. Human genetic diversity does mirror geography, as does much phenotypic diversity. While random drift likely accounts for most of both genetic and phenotypic variation, natural selection clearly has shaped recent human evolution, and there is the potential for cultural forces to impact genetic variation over relatively short timeframes.

The problem is that – for the moment at least – that's about all we can say. It turns out to be far easier to demonstrate that there has been a fair amount of recent natural selection acting on the human population, than it is to pinpoint specific examples, or to rigorously evaluate specific hypotheses. The reason is that different types of evolution (drift, positive selection, purifying selection) leave different fingerprints in the genome, and we can use these to estimate how prevalent each of these forces has been in human history, and, to a lesser extent, identify regions of the genome that have been subject to certain types of selection.

But the effect of specific examples of selection are almost always weak – especially the kinds of transient selection affecting relatively small groups of people on which Wade hangs his speculation. Furthermore, while natural selection leaves a signal behind in the genome, the signal is primarily that it happened – it's much more difficult to precisely identify what was being selected, let alone why or how.

Knowing that natural selection has occurred, in some cases recently, but being unable to be more specific leaves a huge void – and it is into this void that Wade has inserted himself. He spends the first half of his book summarizing (albeit it inaccurately and incompletely) a decade of huge advances in human genomics, but then shifts abruptly from science to speculation ...

My concern about this runs deeper than annoyance at someone for failing to use the tools of my trade, or for cleaving to our authority. The scientific method arose as a way to understand the world because the kind of just-so storytelling that Wade is engaging in is useless. Is it a surprise that Wade just happens to find evolutionary explanations for the most pervasive racist attitudes of the day? Of course not. Because unmoored from data and logical rigor, one can make up an evolutionary explanation for anything.

Read the rest of Eisen's post on his blog.