Following up on the Vox article to which he contributed, Eric Turkheimer has written a new piece over at his Genetics and Human Agency blog (which also hosted Paige Harden’s essay on the “hereditarian left”, which led to the first of two blog posts on these issues here). Turkheimer’s new piece is really a venture into the philosophy of science: as the title puts it, he argues that the “origin of race differences in intelligence is not a scientific question.” This is wrong and Turkheimer does not write as though he believes it in the slightest. It is just a new attempt at burden-shifting, and a particularly dishonest one, as can be seen from the fact that the piece begins by pointing up similarities between the “hereditarian” and “anti-hereditarian” positions and ends by suggesting differences.

Here are just a few points readers should consider. And let me say this: regardless of the politics of the question at hand, a question about the substance of which I know not a jot, the slippery, scrabbling nature of Turkheimer’s piece demonstrates why everyone needs a shop philosopher to clarify their concepts and tell them when they are simply not making sense. I’m happy to offer my services for just 2% of GHA’s $3,500,000 grant.

If this isn’t science, there’s a whole lot more that isn’t science, too. Turkheimer writes that he is first unconvinced by genetic explanations and then, on the other side, unconvinced by environmental explanations:

[M]ost of the scientific designs that might be most informative aren’t available. You can’t breed people, you can’t mess with their DNA, you can’t raise them under controlled conditions. So the discussion is necessarily based on quasi-experimental science that is by definition fundamentally flawed. In research of this kind the grounds for objection are almost always greater than the valid scientific signal . . . Many if not most social scientific arguments are tiresome in exactly this way.

This seems like a fully general argument against doing social science on virtually any topic. Plus: most social scientists are firmly anti-hereditarian on pretty much any topic and, in fact, do nothing to account for genetic confounding in their research. Isn’t it convenient that Turkheimer only sees fit to make this critique in a piece that inveighs against genetic explanations – which form so little of the social-scientific corpus? I wonder if he has read anything at all in the philosophy of social science. The question of how something can be scientific without being experimental is so central there, I think.

Indeed, we should consider: what other social-science conclusions does Turkheimer doubt? Perhaps he doubts that race is a factor in police killings. After all, all we have is correlative statistical (“ecological”, right?) data on that. I wonder why he hasn’t spoken up? Or perhaps he thinks that, next year, women will probably commit just as many violent crimes as men. We can’t experiment with actual murders, right?

Beyond that: perhaps Turkheimer doubts the entire literature on social inequality. In fact, his own positions demand that he must. This is because he thinks it is impossible to conclusively prove that there is or isn’t a genetic basis for any group difference. Based on this, every theory suggesting a non-genetic cause for a social phenomenon is suspect – “fundamentally flawed”. But such theories are the foundation for a great deal of social policy and both popular and academic rhetoric about society. Affirmative action, disparate impact, claims about the continuing prevalence of racism and so on: why hasn’t Turkheimer come out against any of these things?


If a position is meaningless, it can’t be a strong one. Well, Turkheimer tells us why, but it doesn’t make much sense. He writes:

I should be clear that I am not making a “both sides do it” argument. It is the hereditarians who are trying to reach a strong and potentially destructive conclusion, and the burden is absolutely on them to demonstrate that they have a well-grounded empirical and quantitative theory to work with. So, if you are out there and think that group differences t are at least partially genetic, please explain exactly what you mean, in empirical terms.

There are two problems with this. First of all, in formal terms, “at least partially genetic” is not nearly as “strong” a conclusion as its negation, “not at all genetic” – at least if both sides were equally unconvincing, as Turkheimer led us to believe. This is especially the case if individual differences with respect to the trait in question are best explained by genetics. If Turkheimer has never actually found an alternative environmental explanation convincing, one might expect him to fall back on the genetic explanation almost by default!

But the second problem is even worse. If Turkheimer is to be believed, then he doesn’t even know what is meant by the statement that group differences are at least partially genetic in origin. How then did he come to the conclusion that this was a “strong” statement? This seems to be the entire reason that he has what he calls a “strong allegiance to [the anti-hereditarian] side” in the debate. But if the debate isn’t “actually about empirical data” then neither side can be said to be “stronger” (i.e., less probable prima facie or relative to some background theory) than the other.

So Turkheimer must have meant “stronger” in some other sense. I think you can probably guess what sense I have in mind: “more offensive”, “more outrageous”, “more ethically dangerous”. Well, whatever. Philosophy of science is not ethics, and biology is not philosophy of science. If you want to write something about values, write it about values, not about the demarcation question.

Unscientific and untrue are mutually exclusive. Read your Popper. Turkheimer’s spittle-flecked invective near the end just confuses the issue more.

At least in the social sciences, legitimate empirical research paradigms rarely come to all or none conclusions, so it becomes natural for people to conclude, with Murray and Harris, that the whole long argument is bound to settle eventually on the idea that group differences are a little environmental, a little genetic. But in fact, that is not where we are headed. . . .

Declaring something to be a science doesn’t make it so. The hereditarians want all the good things that come from being thought of as scientists. They want academic respect, they want protection from charges of racism, they want clear separation from the very recent history of “race science” that led directly to the Holocaust and Jim Crow. They have to earn it, by doing the hard work of developing the quantitative and empirical theories that transform intuitions about stereotypes into real science.

Now: we have just seen that Turkheimer doesn’t particularly believe in the social sciences to begin with. He owes it to us to demonstrate that his approach leaves room for “legitimate empirical research paradigms” “in the social sciences” at all. As far as I can tell, replacing “hereditarians” with “sociologists” does not affect the truth-value of the second paragraph’s second sentence, if we buy everything Turkheimer’s sold us so far. Again: if we take Turkheimer at his word, massive social policies are being enacted based on completely unscientific research done by sociologists, anthropologists, political scientists, even economists. He chooses, as far as I can tell, to pick on a bunch of socially ostracized bloggers instead.

But things get even worse:

I predict that in a relatively short period of time, contemporary race science will seem just as transparently unscientific and empirically untrue as the race science of the early 20th Century now appears from our modern perspective.

This is complete gibberish both analytically and considering what’s come before. Something cannot both be “transparently unscientific” (in the sense Turkheimer emphasizes above, of being untestable) and “empirically untrue”. If “contemporary race science” will be demonstrated to be “empirically untrue” “in a relatively short period of time”, that’s a victory for contemporary race science as a research programme, at least as concerns the demarcation question Turkheimer covers here. If it weren’t scientific, it wouldn’t be falsifiable! He has gone from saying he finds the entire literature on the question unscientific and unconvincing to saying that a major stance within it will be disproven in short order. How is that supposed to work?

So: Turkheimer doesn’t believe what he’s saying when it comes to whether or not the question of a genetic or non-genetic basis for group differences is a scientific question. His critique is too selective to us to trust that it’s in good faith, and it’s joined by statements that cease making sense if we’re convinced by it. Philosophy of science is a job for philosophers; ethics is also a job for philosophers. Pay me and I can do both. Sorry for this slapdash post; I wanted to get something up tonight. I hope to come back to it this weekend.