A Refutation that wasn’t using Science that isn’t

The Facebook Link of Virtue © this week was Vox’s "Charles Murray is once again peddling junk science about race and IQ." It had everything for the IFLS Liberal who has everything but wants more. It was written by actual scientists – prominent ones, to boot – who actually study the subject, so it comes from an Authoritative Source. It denounced Official Enemies – Charles Murray and Sam Harris. It had a Shareable Manichean Headline, so your less thinky friends on Facebook can receive your signal without even the minimal effort of reading it. And, it was actually readable and superficially convincing. So, even if you’re one of the 5 percenters on FB who actually read the links they comment on, you could come away thinking that Charles Murray really had been put in his place, and that genetic science had been Made Safe for Liberalism.

Actually, I don’t want to come down too hard on this one. It really is written to a higher standard than most of the stuff on Vox, and it really does expose a lot of its target audience to ideas they’re otherwise uncomfortable with and likely to simply avoid. Moreover, it raises some good points about the sociology of science discussion. It’s not enough, though, because in distilling a complex subject down to internet-shareable terms, it consistently makes presentational choices that allow its readers to come away with their biases unchallenged.

Probably the best example of this is their general unwillingness to acknowledge that Murray actually is a victim.

In an episode that runs nearly two and a half hours, Harris, who is best known as the author of The End of Faith, presents Murray as a victim of “a politically correct moral panic” — and goes so far as to say that Murray has no intellectually honest academic critics.

Murray has intellectually honest academic critics, so if Sam Harris claimed or implied he doesn’t, it was deceptive. Only slightly more deceptive than the quoted sentence. Charles Murray very much is a "victim of a politically correct moral panic." He is regularly hounded off of campuses before he’s even had a chance to speak. The Southern Poverty Law Center lists him as a "white nationalist" and deliberately misrepresents both Murray’s position and the state of IQ science to do it. The fact that Sam Harris repeated this fact and you documented it in a sentence which lumps it together with a clearly false statement of Harris’ doesn’t actually succeed in laundering the fact away.

Later in the article the authors euphemize this as "Yes, Charles Murray has been treated badly on some college campuses." No. Charles Murray has been silenced on some campuses, forced to flee from threat of bodily injury before he could even take the stage, let alone speak and be refuted. That’s not simply bad manners, its anti-intellectual totalitarianism. It is dangerous to civilization and scientific inquiry, and you are all cowards for not saying that clearly.

They also misrepresent Murray’s position consistently throughout the article.

Our bottom line is that there is a responsible, scientifically informed alternative to Murrayism: a non-essentialist view of intelligence, a non-deterministic view of behavior genetics, and a view of group differences that avoids oversimplified biology.

Since Murray defends all of those things in his books1, including The Bell Curve, that shouldn’t be a problem. What you won’t be doing, if you do any of that, is presenting an "alternative to Murrayism."

But I think the thing that irks me most about this article is the way they hold the reader’s hand and nudge him away from conclusions they don’t want him to reach.

They give a good summary of the premises of Murray’s argument "in declining order of actual broad acceptance by the scientific community:"

  1. Intelligence, as measured by IQ tests, is a meaningful construct that describes differences in cognitive ability among humans.
  2. Individual differences in intelligence are moderately heritable.
  3. Racial groups differ in their mean scores on IQ tests.
  4. Discoveries about genetic ancestry have validated commonly used racial groupings.
  5. On the basis of points 1 through 4, it is natural to assume that the reasons for racial differences in IQ scores are themselves at least partly genetic.

There is, as far as I know, no serious controversy about (1)2.

2 is interesting for its phrasing. For one thing, the choice of the word "moderate" is accurate but misleading. On adoption and twin studies, intelligence generally shows up anywhere from 40-85% inheritable, so it would be better to say it ranges from moderate to high heritability, or that it’s anywhere from moderately to highly heritable. For another thing, this sidesteps the real substance of their disagreement with Murray. Their bone to pick with Murray has less to do with the heritability of IQ per se than with his belief in the biological reality of an underlying factor of general intelligence – often called simply g to emphasize its status as a statistical observation whose physical correlates remain undiscovered. My understanding is that g is a statistical reality – that is, it’s been repeatedly observed that changing the domain or content of intelligence assessments doesn’t matter much, people who score high on one kind of assessment typically score comparably high on other kinds of assessment as well. There’s no serious disagreement that it exists, at some level of abstraction, but there is, as the authors put it, "a vibrant ongoing debate about the biological reality of g." One way to present it might be to say that, as it impacts heritability, the debate about to what extent g is "biologically real" comes down to how diffuse you think the underlying genetic correlates of intelligence are. If a large number of not-particularly-well-correlated genetic factors contribute to overall intelligence, then it is obviously more difficult for "intelligence" as a category to be correlated with other genetically determined things. Like race, for example, which is point (3). But the controversy over "the biological reality of g" is strong enough, unresolved enough, and central enough to how Murray arrives at his policy proposals that it really merits a separate entry on this list. Indeed, given how much leverage the authors get in the rest of the article acting as though Murray conflates heritability with genetic determinism, it’s hard to believe the decision to leave it off was entierly innocent. The article makes it sound like Murray is wilfully ignorant of the debate surrounding g, and that it is this ignorance that allows him to follow "the faulty path by which he casually proceeds from a few basic premises to the inflammatory conclusion that IQ differences between groups are likely to be at least partly based on inborn genetic differences." Nothing about Murray’s actual writing backs that up, though. Even the introduction to the The Bell Curve, lo those many years ago when he first wandered into this minefield, makes clear that he is aware that g is not only controversial in academic circles but also that the concept is contrary to common sense. It’s fine to note that there are empirical and conceptual reasons to doubt g, but Murray is not outside the scientific norm in believing in it, nor is it fair to imply he is ignorant of the extent to which his policy recommendations are premised on it. The best that can be said is that he’s maybe too sanguine about the counterevidence, or that he maybe doesn’t appreciate the degree to which g is and always has been a statistical construct.

3 is also not particularly controversial as an empirical fact. There is, of course, a lot of controversy as to the cause. This is where the real policy debate starts, where it really starts to matter how much is environmental and, for the portion that is heritable, how well it is correlated with racial grouping.

4 is positively fascinating to me, not so much for the genetic classification angle as how conspicuously absent it is from policy activism. By that I mean that on an issue so politically fraught where the scientific consensus is so strong, there is a curious lack of any movement to atomize the racial classifications we use in everyday life. If they are not genetically grounded, then they must be mostly social constructs, suggesting that the solution to racial disparities would involve tearing down these constructs, and yet all the right-thinking policy proposals build them up instead. That’s an observation that’s admittedly somewhat tangential to the article, but it does make the authors’ fixation on Murray as a dangerous figure seem more political than scientific.

5 is the point where the authors start being deceptive. At least, they do in calling it "completely incorrect" and putting the section under a heading called "flawed logic." In fact, even by their own presentation, (5) is quite a natural conclusion to reach, if you bought everything in points (1)-(4). If you believe to some degree or another in every point on the way down to (5), then while you’re not logically compelled to believe that at least some part of measured racial IQ differences are genetic, it would still be the smart way to bet. It certainly wouldn’t be "junk science," as the hugely disingenuous title claims. And if you throw in the missing point about g, the conclusion that follows from the stated premeses becomes more likely still.

Additionally, it’s not even necessary to buy (4) to reach that conclusion. The conclusion is perfectly compatible with a state of affairs where, say, "black" refers to 15 wholly distinct racial groups, 12 of which have lower-than-average mean IQs and 3 of which higher-than-average. It will still be the case, if intelligence is "moderately to highly" heritable, as we know it is, that offspring from within these groups will be more likely to revert to their ancestors’ means – that is, will be more likely, on average to have lower-than-average IQs. This will also be true of offspring produced by mixing the groups, albeit probably (depending on the combination, and on how intelligence is actually inherited) to a lesser degree. Indeed, if you follow this thought experiment through, you’ll notice that the contrary view on (4) – that there is almost no biological reality to the racial classifications that the government uses for its reports and surveys, just as easily cuts the other way: government categorizations may just as easily be masking biologically identifiable groups that, if tested separately – i.e. in greater correspondence with biological reality – would show even more IQ heritability according to race than is currently observed. And again, it’s rather telling that the authors choose not to address this obvious point.

The authors might respond here by saying that logically compelled is what they meant – that Murray’s conclusions don’t follow inexorably from his premises. But I don’t think he’s ever claimed that they do. He’s reasoning to the best explanation, which is all beliefs based on empirical science ever are at any given time. He’s never claimed, that I’m aware of, that his conclusions are beyond dispute, or that the field has exhausted all fruitful lines of inquiry and should pack it in. The claim about his beliefs is that they are what strike him as most consistent with the data. Now, I said "reasoning to the best explanation" is "all beliefs based on empirical science ever are," which, while true enough, skirts over the more important component of empirical science: that currently-held beliefs are testable. And on this dimension Murray comes off looking – to my eyes, anyway – somewhat better than the authors of this article. Unlike them, he is not advocating "protests" to discourage testing of hypotheses they don’t like3. He presents his view of the world and invites criticism. The authors are the ones doing the name-calling.

Most of the article from here rests on two arguments, one of which – that g is merely a statistical observation of dubious biological reality – has already been partly addressed. g is real, the question is how. While it’s entirely possible that Murray’s view of it is too crude (in fact, for what it’s worth, that’s what I personally believe, though I’m of course not an expert in this field and therefore not the best person to ask), there’s no reason to believe that Murray is ignorant of the issues.

The other main argument is that heritability is not destiny: however heritable IQ turns out to be, there’s clear evidence that it is nevertheless modifiable.

This argument obviously misses the point entirely. Again, there’s nothing in Murray to suggest that Murray is ignorant of this fact, nor that he ignores it in reaching his own policy recommendations, so to some degree the authors are burning a straw man. Their point seems to be that even if we accept that certain racial groups have lower average IQs due (in part) to heritable causes than the national mean at present, the fact that a lot of genetic traits only determine tendencies and not outcomes means we can still achieve … what? Actually, that matters a lot. Is the goal simple equality of outcome among the racial groups the authors ostensibly don’t believe in, or is the concern that social and environmental factors are prohibiting some groups from reaching their full potential, whatever that potential may be? Or is the argument rather meant to suggest that however heritable IQ may be, genetics are less of a constraint than they may seem?

The classic example is height, which is strongly heritable (80 to 90 percent), yet the average height of 11-year-old boys in Japan has increased by more than 5 inches in the past 50 years.

Yes, but here’s another fun fact about height: no matter where in the world you are, the average height difference between males and females is about 5in., and sex is biologically determined. Moreoever, whatever it is about sex that biologically determines height, it’s complex and not reduceable to any single proximate cause. There is, in other words, something like g for height, and that’s sex, and no one is under the illusion that there’s any single gene or genetic factor that produces this outcome directly, rather it’s a whole rag-tag fleet of proximate causes. All of which seriously waters down the argument from ignorance approach that they are taking with regard to g and intelligence. Height is a product of a whole bunch of structural factors that, while individually largely unrelated, all correlate rather well to sex, and the conspiracy of influences produces the extremely robust height differential between the sexes.

But what does this have to do with racial differences in intelligence?

There has been an 18-point gain in average IQ in the US from 1948 to 2002. One way to put that into perspective is to note that the IQ gap between black and white people today is only about half the gap between America as a whole now and America as a whole in 1948.

That’s true, and it’s relevant to the discussion in that it proves that genetics don’t necessarily impose ceilings on IQ maturation, but notice the point they haven’t addressed, which is whether IQ differences between the races are more or less like the height differences between sexes? Right out of the gate, we know that this is going to fall more on the side of "less" than "more," just because race has a lot less biological reality than sex, and the authors have already reminded us that traditional race groupings have less biological reality still. But that’s not the same as "none," and indeed, the IQ gap between whites and blacks today is not too different than it was in 1948 (though it has gotten smaller). Much the same way that, though Japanese people have gotten taller, they’re still shorter than white people. Or, to put a fine point on it, if height is heritable – and according to the authors it is roughly as heritable as intelligence – then if I inherit tall genes and you inherit short ones, and you raise us in Japan in 1912, I’m going to be my genetically-determined 3in. taller than you (or whatever) in 1912, even if I’m also 4in. shorter than I would be if you’d raised me in the US in the 1980s. I’m glossing over some complexity there, obviously, but the point is that very little about environmental changes is going to change the relative standing, when you run it all through the wash of large numbers of people, of racial groups on IQ if indeed IQ is highly genetically heritable and these genetic factors are correlated with race. Because even if you believe in very little biological reality for race, "race" is still a proxy for heritability: the same family lines end up in the same racial groupings, for the most part, and even the authors admit that geograpical ancestry is genetically real and traceable.

For their coup de grace, they cite one of their own studies that shows that "The heritability of intelligence, although never zero, is markedly lower among American children raised in poverty" … and conveniently fail to mention that that study only followed the children involved to the age of 7, when it is well known across a large number of studies that environmental effects fade with increasing age. This point has of course been made about this study in follow-up studies, but why let science get in the way of flattering your readers’ vanity? Just the "relevant" facts, I guess, where "relevant" is "relevant to demonizing Murray."

Now, I’ll go ahead and, in wrapping up, admit I’m out of my league here. I’m not an expert on genetics or heritability, much less as these things apply to IQ. And I really don’t have a horse in the race for how the racial groupings data ultimately come out specifically. Nor do I have any particular investment in Murray. Murray, to the extent I’ve read him, strikes me as a better-than-average popsci writer. That is, he gives a better introduction to his field to the lay public than you typically get in these books, but his treatment is still shallow, and I wouldn’t take it to be the final word on any of these issues. So, the purpose of this post really isn’t to say that Murray is right and the authors are wrong.

The purpose is to denounce this article as typifying a dangerous, anti-intellectual tendency on the Left. Our entire culture has a problem with anti-intellectualism, and all of us who believe in science and progress have a part to play in setting it right. On the Right, the response to science they don’t want to hear is either to sneer and ignore, or else to cook the books to get the conclusions they want. That’s all Very Bad. But consider how much worse it is on the Left, and how laughable an irony it is that somehow the Left is generally perceived – and certainly perceives itself – as being the more scientific half of the whole. Let’s remind the authors once again that Murray wasn’t merely "treated badly" on campus, he was hounded out of his speech by people who shouted obscenities and pulled fire alarms to force an evacuation and shut down the power to the livestream. He was able to continue online from a separate location, but once that location was leaked, his car was physically attacked, and the person who was invited by the school to rebut him was hospitalized as a result. And it wasn’t just that – but voices from the left defending all of this weren’t hard to find. The fact that the authors want to euphemize this as "treated badly," rather than calling it out for the violent fascist censorship it actually is, says a lot about their own commitment to free and open scientific discourse, none of it flattering to them. But what’s equally concerning, to my mind, is that this article presents itself as having refuted Murray when it does nothing of the kind. What it does is fairly raise objections to his conclusions. People who read Murray definitely need to know that g is a problematic category. They definitely need to know that heritiability is not coterminous with genetic essentialism. They definitely need to know that improvments in the general level of intelligence (or height, for that matter) put the mechanism by which genetic determinants operate in a context that suggests they are not destiny. They definitely need to know that heritability of intelligence is heavily correlated with social class. All of these things are important contributions to the debate, and all of them belong in the space of public discourse, and the authors do a service to the public by adding them to the newsstream. But none of them actually support the conclusion that Murray’s presentation is "junk science," nor do any of them end the debate. If the authors of this article were scientifically responsible, they would’ve framed this as a collection of useful pointers for people considering attending a Murray talk. They are all good questions to put to Murray, to see what his answers are, and it’s certainly fair to fault Harris for not having raised them during his podcast. But Murray’s public appearances go well beyond that podcast, and there is an entire literature of works by Murray that form his position. Focusing on a single podcast hosted by an attention-seeking fraud like Sam Harris is intellectually dishonest. What these authors are doing by publishing this article with this spin in a liberal zine for a liberal audience is not promoting science, or furthering any debate. Quite the contrary – it’s encouraging liberal readers to tune in, turn off and drop out by falsely flattering their vanity that science, somehow, always reaches conclusions they like.

At least when conservatives plug up their ears and shout "lalalalala!" to anything they don’t want to hear, they’re aware that expert opinion disagrees with them. The problem with liberals runs much deeper, because it involves so much more denial. Because the media leans left, and because the academy leans left, left-leaners are used to science reporting always telling them things that align with their policy preferences. When someone like Murray comes along, someone who doesn’t share those preferences, it induces a lot of cognitive dissonance. The response is exactly what we’ve seen. Sometimes, there’s a honest-to-god book burning of the kind we saw at Middlebury, and which the authors of this article can’t even call by its real name. But more often, there are articles like this that exaggerate the scientific objections to a theory to the point where they sound like a thorough debunking. Notice that nowhere does this article encourage readers to engage with the subject. Nowhere does it acknolwedge that this issue is deeper and more complex than a Vox article can cover. Nowhere does it point to any primers on the literature, nor does it even acknowledge that there are objections to the points it raises. The points are presented as though they were the whole of the story, that Murray were soundly refuted by this, when even a layperson who has done no background reading can see that controversies remain and that he isn’t, and it’s all wrapped up in a rank name-calling title that actively discourages readers from exploring further. This is not food for intelligent people. It is a soothing balm for mental weaklings who do not like their opinions challenged, and who believe in Science as a kind of mystical cargocult force that always affirms their righteousness.

People who "fucking love science" need to get in the habit of going through some simple exercises. For example: name one time you read a scientific conclusion you didn’t like but were forced to accept on the strength of the evidence. If you can’t do it, you don’t "fucking love science." For another: name one time you presented a scientific argument whose political implications you disliked without editorializing. If you can’t do it, you don’t "fucking love science." Another: name one time you read a science article on Vox (or HuffPo or wherever), and you liked the conclusions, but you weren’t satisfied that the authors had given you the whole picture, and so you went googling for counterarguments, just to check. If you can’t do it, you don’t "fucking love science."

I get why the authors took the tack they took. There’s a fear, confronted with the blatant anti-intellectual bent of the modern left, that if you don’t tell them what they want to hear, they will simply ignore you. In this case, that would mean a continual abdication of IQ science to the social darwinist right, to hear these authors tell it.

The left has another lesson to learn as well. If people with progressive political values, who reject claims of genetic determinism and pseudoscientific racialist speculation, abdicate their responsibility to engage with the science of human abilities and the genetics of human behavior, the field will come to be dominated by those who do not share those values.

One can’t help but notice that the authors are exhorting people to get invovlved with genetic science so that they can push it to political conclusions they like. This is, of course, not how science is supposed to work. The right reason to engage with science is to get to The Truth. They try to hedge this in the closing:

Liberals make a mistake when they try to prevent scholars from being heard — even those whose methods and logic are as slipshod as Murray’s. That would be true even if there were not scientific views of intelligence and genetics that progressives would likely find acceptable. But given that there is such a view, it is foolish indeed to try to prevent public discussion.

But it’s easy to see through the ploy. Yes, it’s foolish to drop out of something that might end up going your way. But you know what’s even more foolish? Deciding you know what’s what before asking reality what really is. Nothing in this article discourages the anti-intellectual tendency on the Left to assume that Science always ends up where they want it to go. Quite the contrary, it enables it.


  1. Although I have to put an asterisk by the weasel word "essentialist" since that can mean a lot of things, and the authors aren’t entirely clear what they mean by it here. They surely can’t mean what it would typically mean in casual speech, since that would imply they are denying any heritability to IQ at all, and they’re not. So what they probably mean by "non-essentialist" instead is that both environmental and genetic factors play a role in formation of IQ. Since that is something Murray believes as well, determining the truth value of any claims would require them to specify just how heritable one has to believe IQ is to be an "essentialist" and then justify their choice.

  2. "No serious controversy" doesn’t mean "no one disputes it;" it means that the scientific consensus around it is overwhelming.

  3. I am basing this on this statement near the end: "We are absolute supporters of free speech in general and an open marketplace of ideas on campus in particular, but poorly informed scientific speculation should nevertheless be called out for what it is. Protest, when founded on genuine scientific understanding, is appropriate; silencing people is not." Nothing about Murray merits "protest," never mind that it’s quite clear that the students protesting against Murray are, by and large, not doing so out of any "genuine academic understanding." The authors are saying the bare minimum that is required to avoid association with fascists. But it is not enough. People like Murray invite discussion and counterargument. They do not invite protest – not in any genuine academic community.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>