« The socialism of moralizing fools | Main | Power without conspiracies »

March 30, 2018


Matthew Moore

"Luck egalitarianism says that inequalities that are due to luck should be neutralized."

Does it give any indication as to how we might actually do this?

Handy Mike

"To see my point, let’s assume that the most provocative claims made by Charles Murray in The Bell Curve are correct. As summarized by Harris these are: 'that “general intelligence” is a scientifically valid concept and can be measured by IQ tests; that IQ is highly predictive of success in life; that mean IQ differs across populations; and that IQ is partly genetically determined."

Of these, only the claim that IQ differs across populations remains in the play. All of the rest are far better established in social and psychological science than the ragbag of cognitive biases you reach for to explaining why Tories are stupid and Marxism hasn't triumphed yet. There's been *a lot* of work and *a lot* of papers on this. I've criticized your argument-from-hyperlink fallacies before, but on a matter such as this, it's especially tendentious and dodgy.

All that aside, there's an important and rather obvious sense in which this stuff would, if true, be strikingly relevant to the remedy of inequality. At present it is common for people to point to uneven distributions of social outcomes - disproportionate representation in certain professional fields, say, - as unacceptable inequality caused by racism.

*If* the IQ-Success-Genetics-Population stuff is a significant part of the picture, then some of the measures commonly proposed are not going to work. If it's not mainly racism, then anti-racism, diversity awareness training etc, won't fix it.

Instead, we might need to simply set aside the gross incoherence of Luck Egalitarianism and push on with more coarse-grained redistributions of resource and opportunity.

Perhaps of the sort recommended by Toby Young.

Oh, and Charles Murray.


In the UK academic achievement gap between black and white students at the age of 16 used to be enormous, but it's now basically nonexistent. I don't know why American liberals don't cite this fact when arguing against Murray and his ilk. As far as I can judge, it completely falsifies his theory.



I agree. However, some people don't.

Recently there was this. https://www.nature.com/articles/s41539-018-0019-8. Some on Labour side went round touting this as evidence that selection in schools was a waste of time as it made no difference. However the abstract to the paper states "We created a genome-wide polygenic score (GPS) derived from a genome-wide association study of years of education (EduYears)"

I'm surprised there hasn't been more fuss made about this, but maybe because it is supporting a left-wing idea against selection it is considered acceptable.

Ralph Musgrave

Re Chris’s “four ways” point are nonsense. For a start, there are only three of them. Next…

1. Re “unjust economic systems” IQ is almost certainly required to get to top in unjust systems. Is it realistic to suppose any with a very low IQ would get to the top in Nazi Germany?

2. No one ever claimed IQ was of “moral worth”.

3. The fact that there are inequalities other than IQ is not evidence that IQ differences are not of importance.


«Joey Essex, Paris Hilton and the Kardashians probably do better in capitalism than they would under communism, but they drag down the IQ-success correlation.»

I suspect that they are also highly intelligent -- being somewhat vulgar and posturing as a shallow clown does not indicate low intelligence, as BN Johnson demonstrates amply.


«In the UK academic achievement gap between black and white students at the age of 16 used to be enormous, but it's now basically nonexistent.»

There is a much better and more revealing story: in the USA, *on average* (not in every case), black orphan boys given in care to a family with a white adoptive mother have the same social and school achievements as white boys; when they are passed in care to a black mother they switch to achieving the same as other black children, and if they are passed on again to a family with a white mother they switch again to achieving like white children.
I wish I had read something about white orphan boys and black adoptive mothers, or black boys and asian adoptive mothers, and the same for girls (but then in the USA black girls have much the same social and school achievements as white girls of comparable backgrounds, the big difference is with boys).

Anyhow the brief summary is that the achievements of boys depends largely not on race or even school quality but on whether they have a "tiger mom" in charge, and then the big question is why black mothers are not "tiger moms", for which there is a "lateral thinking" politically incorrect possibility.

«I don't know why American liberals don't cite this fact»

Identity politics is about foolish moralizing and distracting from neolib policies, not about solving problems.


«In the UK academic achievement gap between black and white students at the age of 16 used to be enormous, but it's now basically nonexistent.»

There is another funny story about a similar problem: decades ago USA universities started funding many students, based on exam results, and what happened is that they rapidly filled with immigrant students of mediterranean and middle-eastern religious heritage (to use an euphemism) out of all proportion, because their legendarily brutal "tiger moms" made them cram for exams, and also those people of mediterranean and middle-eastern religious heritage are the only case where there is clear quantitative evidence of racial superiority, quite ironically.

So many prestigious USA universities changed their admissions to value both exam results and "community and personal achievements", because the latter are not easily within reach of immigrants who cram full time for exams, and affluent WASP students who spent a year in the Peace Corps etc. instead of cramming went back to their previous percentage.

So called "meritocratic" admissions policies in some universities have also resulted in the USA in a disproportionate number of immigrant asian heritage students with ferocious "tiger moms", displacing large numbers of high-fee paying WASPs and darker-skinned ones...


Can "race" ever be defined in a way which leads to accurate academic study? IMHO, it's an inherently fuzzy distinction, highly dependent on the history of the area where one grew up. West of the Pond (USA), our history of Slavery & Jim Crow leaves us with a very black & white (sorry) view of race. Legal definitions of race which were once important now seem ridiculous.

Studies of race can only make sense in the Social Sciences, where "race" can be defined as a social distinction, based how people define themselves and each other. The study of "IQ" pretends to be more medical/mathematical than social.


«in the USA black girls have much the same social and school achievements as white girls of comparable backgrounds»

Haha! No.

This "Tiger Mom" lollery also accepts, uncritically, the idea that a minority parent's role should be organized around raising a kid to prop up the hierarchy which is responsible the kid's own marginalization in the first place. That some Asian mothers ruthlessly subscribe to this worldview is their personal tragedy, but I don't understand why anybody would want to export the disorder. Maybe poor black mothers are teaching their kids skills other than "be as rapacious and competitive a capitalist as possible" — which means their kids might not do as well in school, but have other abilities that might stand them in good stead in a less awful economic regime.

«Is it realistic to suppose any with a very low IQ would get to the top in Nazi Germany?»

Why are people still associating Nazis with excellence? They were (and are) white trash. They carefully cleaned their country of legitimate intellectuals when they came to power. The fact that anybody would suggest that intelligence is a prerequisite for the performance of racist genocide is further evidence that 'racism' is the primary, and only significant, reason black people have trouble succeeding in the US and the UK.

I think, in a lot of important ways, that this post is fundamentally incorrect; it doesn't address the arguments about 'racial IQ' as they actually exist. Handy Mike kindly articulates these racist arghblarghings for you — the real idea is to prevent any kind of social justice from occurring in any formal setting. No diversity training, no hate crime laws, no affirmative action. These things cannot and will not work, say the world's scientific racists, because they are incapable of addressing the true source of inequality — innate racial/gender inferiority. Charles Murray and Jordan Peterson and Stephen Pinker are not just attempting to dismantle equality as a governmental undertaking; they are attempting to dismantle the idea that racism exists, that it has effects, and that it should be remedied. They are defending white, well-off, heterosexual men as an institution; of course such people deserve to occupy the top of the social hierarchy! They are fitted for success and ownership not just by the perfectly fair processes of social selection, but by evolution itself. Any attempt to dethrone them is at its heart a rebellion against the true, natural order of things.

If there's a more despicable set of opinions available in the marketplace of ideas, I haven't heard them. I would rather a man spit in my face and call me a n*gger or a c*nt or a dyke than gently explain to me why my genes prevent me from legitimately participating in society, or from having a claim to the same sovereignty he expects as a birthright.

I do appreciate your attempts to argue with facts and evidence, though. You're only failing because the argument is predicated on scientized racist/sexist hysteria, not because you haven't already won.

Although I do think that if you're going to continue to refuse to write a book so that I don't have to do any work, you could at least do your commenters the courtesy of letting us use HTML.

David Friedman

You start with:

"Many of you will find this assumption heroic."

You then give a list of qualifications, only two of which are relevant to the assumption. "IQ is partly genetically determined," for example, is not inconsistent with the claim that it is influenced by other things as well, and "it does not follow that anti-poverty programs are ineffective" does not contradict anything in your description of the assumption. You seem to be sliding from a clearly stated set of assumptions to a conclusion that you are implicitly attributing to those assumptions, and then pointing at things inconsistent with that conclusion but not with any of the assumptions.

So far as what the political implications of the assumption are, I think there is only one important one—that racial inequalities in outcome need not demonstrate discrimination. That's important—more important for the gender case, for which the argument is stronger, than for the racial case. The reason it is important is that the contrary is routinely taken for granted in discussions of differential outcomes.

The Gonch

The Sam Harris, Andrew Sullivan, Charles Murray, Toby Young well-rehearsed script is, unfortunately, a highly naïve – indeed outmoded – view of the gene and biological systems (see https://thepsychologist.bps.org.uk/so-what-gene).

Moreover, even if it were accurate (it is not) it does not link to future success. Pre-University educational attainment at age sixteen may create a route to higher status but it is a very poor predictor of success either in future education or in the world of work. That applies even to A levels. A 2012 review notes that “in U.K. data, a small correlation was observed between A level points and university GPA (r = 0.25), again reflecting previous findings.” As the authors say, that could be due to non-cognitive factors: “self-efficacy was the strongest correlate (of 50 measures)”(1).

It is ironic that just as psychologists, and online commentators who wish to deny social and political racism, are claiming to have found genes for predicting success the wider world is eschewing their criteria of success as useless. In an interview with the New York Times (June 13, 2013) Laszlo Bock, a vice-president of human resources at Google said, “One of the things we’ve seen from all our data crunching is that GPAs [Grade Point Averages] are worthless as a criteria for hiring, and test scores are worthless – no correlation at all except for brand-new college grads, where there’s a slight correlation”. In the UK, as reported by the BBC (January 18, 2016), major employers are abandoning reliance on educational attainments because there is no clear link between holding a degree and performance on a job.

Also the science is a prime case of correlation outweighing causation. The method entails a formidable battery of assumptions, data corrections, and statistical maneuvers. But the most fatal assumption is that human societies can be treated as random breeding populations in randomly distributed environments with equally random distributions of genes.

On the contrary, human populations over significant periods of time reflect continuous emigration and immigration. Immigrants with related genetic backgrounds tend not to disperse randomly in the target society. In their flow to jobs they concentrate in different social strata. This creates (entirely coincidental, non-causal) correlation between social class and genetic background persisting across many generations. For example, the Wellcome Trust’s “genetic map of Britain” shows strikingly different genetic admixtures among residents of different geographic regions of the United Kingdom. Evan Charney notes this is “omnipresent in all populations and it wreaks havoc with assumptions about ‘relatedness’ and ‘unrelatedness’ that cannot be ‘corrected for’ by the statistical methods [devised]” (2).

All this correlation is simply a sophisticated re-description of the class structure of British society, its deeper history and all its corruptions, including differential effects on races and communities.

1. M. Richardson, C. Abraham, & R. Bond, “Psychological Correlates of University Students’ Academic Performance: A Systematic Review and Meta-Analysis,” Psychological Bulletin 138 (March 2012): 353–387. 

3. E. Charney, “Still Chasing Ghosts: A New Genetic Methodology Will Not Find the ‘Missing Heritability,’ ” Independent Science News (September 19, 2013).


@ Gonch

So for clarity, do you disagree with this paper? https://www.nature.com/articles/s41539-018-0019-8

And if so, do you think Nature was wrong to publish a paper that ascribed differences in exam performance to genes?

The Gonch

@ Dipper

I can provide a long-winded reply with citations if you wish, (it would start with the literature on the failure of of GWAS and similar methodologies) but this is the short version!

The shortest answer to your question is (i) I don’t disagree with the paper, based on its methodology (not least because it is an argument against Grammar Schools and Private Schools) (ii) But the paper does not support the actually conclusions of Harris and Young et al, for reasons I outlined above (there arguments are not just about genetics and genes, but about politics and education policy as well) moreover, (iii) It can easily be argued the paper presents a correlation (not causation). Again, this does not make the paper *wrong*; a proper understanding of the method is that it is *only* designed to find correlation, but the abstract and conclusion oversells the conclusion for public consumption - a common occurrence in Nature (iv) Nature IS wrong to sell the paper with lines such as “However, the possible role of DNA differences between students of different schools types has not yet been considered.” This sets the paper up as arguing against a straw man. If you like, I can happily provide other academic papers demonstrating the statistical traps in this kind of work (for fun, here is one: )

My point was not to say one cannot find correlations. It was three fold (i) Even if you could find a genetic reason for pre-16 exam success, it would miss the point, because pre-16 exam success does not collelate with future life success (or value), a lot of work has looked at the connection between A-Levels and University success (for example) and their isn’t one (ii) This is reflected in employment markets that are putting far less weight on things like A-Level’s and SAT’s (iii) There is not a simple dichotomy between “it’s all genes” vs. “It’s all the environment”; that’s the argument of simplistic opinion writers (like Harris and Young).

Here is a thought experiment. Three brothers/twins are born are born within five minutes of each other. Do they all get the same level of intelligence and therefore future exam success. If so, why? If not, why not?

The problem with all this is that *no* seriously informed geneticists argue that people’s traits are set in stone from birth. Just look at height; it is clearly strongly heritable (that is, its variation in Western society is largely due to genes, hundreds of which have been found; Wood et al., 2014), and yet it can be changed by a better or poorer diet; indeed, it’s average level has increased substantially over the years (that is, its mean can be influenced by non-genetic causes).


@ The Gonch.

Thanks for the reply. I think you are splitting hairs. "Correlation not causation" seems to be having your cake and eating it. Surely the point of the paper was that the genetic disparities caused the observed differences in performance? And you say "that *no* seriously informed geneticists argue that people’s traits are set in stone from birth" but the argument of the paper is precisely that; that the different kinds of schools add no value to exam performance beyond the genetically pre-determined capabilities of the children.

Toby young;s argument was that were there to be an intelligence gene then working class parents should be allowed to screen their foetuses and choose the one with the intelligence gene. The paper seems to be saying that such screening would be possible, and hence Toby Young's opinions have some scientific basis.

Personally I don't agree with a genetic basis of intelligence apart from one or two highly mathematical things. But that isn't the point. The point is that a genetic basis to "intelligence" is being used when it suits a particular argument and then denounced by the same people when it promotes an unwelcome argument.

From Arse To Elbow


Just to be clear, Toby Young (yes, that one) is a co-author of the paper you cite.

An "expert reaction" can be found here: http://www.sciencemediacentre.org/expert-reaction-to-study-investigating-genetic-differences-in-selective-and-non-selective-school-students/

I particularly liked this: "It is also worth noting that we have almost no understanding of the mechanism by which genetics affects educational attainment. We do know that genetic contributions to educational years overlaps with a vast number of other traits, ranging from height to obesity, bipolar disorder to rheumatoid arthritis, depression to snoring. In short, I don’t think this paper advances our understanding of the role of selective schools or the biological mechanisms that can impact on educational attainment."

Handy Mike

Guys, give it up.

You will, of course, be able to find papers here and there setting their faces against the prevailing wind of research. But in doing so, you're ferreting around in exactly the same motivated way as climate skeptics.

The simple fact is that the reality of IQ/g, its the link to life outcomes and heritability (around 0.8) are much better established findings in psychology than the toolkit of cognitive bloopers with which Mr Dillow tries to diagnose our continuing inability to spot, let alone shrug off, capitalist oppression.

You're doing one or both of two things. Displaying simple ignorance of how things are in the field - an ignorance that around twenty seconds on Google could release you from. And/Or allowing your discomfort with what these findings might be taken to mean, to misrepresent what you know the findings to say.

Neither is tenable. But, if you really think you're going to be able to wish this stuff away, then stick with it by all means.

Only, here's what you'll also have to do, on pain of inconsistency.

Chris hangs a lot of his posts off research papers in psychology and social science. There are going to be other papers claiming that these papers are wrong, or flawed, or skewed, or present an incomplete picture.

Do the same for those.

Or if not, why not?


@Handy Mike

I'm not close to being an expert in genetics. My take on this topic begins with the view that all reasoning is motivated (and that even an absence of a motivation is a motivation of a different type - in the sense that it will still have an effect on action). So far as I can tell, both sides in this debate conclude that given the variance within any large demographic, the only rational way to interact with individuals is to treat them as just that - individuals - and not as representatives of the mean traits of the demographics to which they belong. This, however, is not how humans actually think. It's pretty uncontentious that much of our thought is the result of 'heuristics and biases' and that we do in fact - consciously or otherwise - ascribe to people traits based on presuppositions. Race and gender are preeminent sources of these ascriptions as they are among the most readily conspicuous traits. Moreover, the effects of advantages and disadvantages of presuppositions are likely to compound due to in-group bias. This is the root of my understanding of 'white' and 'male' privilege.

Given this disjuncture between how we would have people behave (i.e. reasonably) and how they actually do, it seems sensible to me that the starting motivation for our thinking on this topic should be that, other things equal, we favour accounts of reality that facilitate reasonable behaviour - in this case, treating people as individuals. As such, those who claim that genes (really 'race' or 'gender') are more-or-less immutable causal factors in people's life outcomes need to demonstrate not only that this is so, but that the effect is sufficiently significant that it warrants modifying our a priori preference for treating people as individuals. If this bar is not attained, then the research is practically inconsequntial. I can't find fault with Sam Harris' stated position that he's uninterested in racial differences (but that he is interested in preserving the freedom of intellectual inquiry). Pragmatically speaking, if we're reasonable people seeking to treat each other as individuals then racial differences are uninteresting.

Therefore, as a lay-person unqualified to make an informed analysis of the science on this topic (although I haven't seen a compelling counter to Simon's point about the improvement in IQ of black British school children), I take the pragmatic view that racial differences in IQ, while extant, are not significant.

Jordan Peterson is interesting on this point as an advocate of the notion that epistemological truths are subordinate to ethical ones (in contrast to Sam Harris). Even if Murray's claims were true in the strongest sense, under the liberal paradigm they would still not be true enough to be useful. The severity of the backlash that Murray has faced is, in my view, unfortunate, but given that people understand that an implication of his position is to undermine the paradigm in which they live, I don't think they're surprising. If you think that race (say) is a significant enough causal factor to account for a large part of life outcomes, then you will need to have an unusually high ability to tolerate cognitive dissonance to avoid being racially prejudiced. If we think that it is unreasonable (and maybe unjust) to be racially prejudiced and don't have heroic assumptions about our ability to sustain contradicting views simultaneously then it seems sensible, while there is a controversy, to give the benefit of the doubt to the side of the controversy that counters our prejudice.

Handy Mike


By far the most sensible, best-argued, nicely-written and humane comment so far. I agree with almost everything you say, and where I don't quite agree, I certainly hope that what you say ends up being true.


Moyenda is just committing the reverse naturalistic fallacy. It would be pleasant to believe this is true, so "let's give the benefit of the doubt to the side of the controversy that counters our prejudice". Basically a convoluted reasoning which simultaneously elevates Moyenda's position to moral superiority and assumes corroboration. Why be right when you can be morally superior AND right?

Actually the correct response is to refuse to give the benefit of the doubt to any side, and instead continue to vigorously research the issue using the scientific method. From everything I have read, the vast majority of experts in the field currently believe that genetics explains at least some of the variation between IQ across populations. Let’s open the floodgates and continue to research the topic. Perhaps they will be proved wrong. Perhaps the opposite. Anyone disagreeing with this approach is burying their head in sand (and yes, there appears to be a lot of this on this topic)

As to the different results in the UK, the hypotheses potentially explaining the disparity from the US, is a) different selection criteria between the two countries (one came involuntarily, the other voluntarily), and b) cultural differences (pretty much all experts agree culture matters too, and there is no assumption that black British culture is similar to black American culture). I could come up with two or three other hypotheses, but the point is that I have no idea what the answer is until someone researches the admittedly complex issue.

As for the main post, I have no idea why he adds the line about invalidating anti poverty programs, or what IQ has to do with the justice of society, or the correlation of IQ and moral worth. To the best of my knowledge, no reasonable person on any side of the debate is making any of these arguments, so adding them are just disingenuous distractions from the issue. I assume they are rhetorical tricks to disparage the other side of the debate for those not following closely. If not, please correct me.

In addition, the "dodging bullets" line is beneath contempt. The author just assumes racism was responsible for the shooting (rather than crappy, cowardly policing for example), or assumes blacks are more likely to be killed than whites when interacting with the police, another conjecture with no evidence once we adjust for disparate rates of violence and police interaction. Arguments by selective anecdotes.

Finally, to answer the question waved away that IQ has nothing to do with telling us what to do about inequality. This is absurd with even the slightest amount of contemplation. Does any reader of this blog NOT recognize how population level differences in genetics can influence the issue? To the extent Asians DO have higher IQs than whites, then disparities In admissions, results, incomes or such are correspondingly less likely to be due to privilege or culture. I can think of myriads of approaches to inequality (and more importantly unfairness) which change based upon how we answer thtese scientific questions. So let’s answer them and quit demonizing anyone on the side of the answer which differs from what we wish to hear.

The proper response to Murray and the vast majority of scientists with these beliefs is to encourage them to make their arguments and for anyone disagreeing to rationally attempt to refute it with (non ad hominem) evidence to the contrary.


@Roger - Thanks for saving me from the consequences of agreement-on-t'internet induced shock.

@Handy Mike - Very big of you to have space for a slightly different perspective.

@Roger again - I don't assume the benefit of the doubt for the sake of any abstract 'moral superiority', but because it accords with what I do (relatively uncontrovertially) know: that people within demographics vary so much as to make generalisations to any specific instance unreasonable.

I have no problem with people conducting as much research into demographic IQ differences as they like (so long as it's not unduly at the expense of other useful avenues of study), but I don't think the truths thereby uncovered will have transparent policy implications. Firstly, as I've stated, I doubt that any demographic IQ research will have results sufficient to outweigh the pragmatic social utility of seeking to avoid prejudice in our dealings with others. And second, as Chris points out, the justice of social arrangements is not a quality of the facts of those arrangements. Say, for example, it's established that black Americans have a reduced genetic endowment as a result of the systematic murder of intelligent, black slaves - what does this tell us about how society should treat present day black American populations who could be seen to be suffering as a result of this damaged genetic endowment? It seems to me that the answer is to this hypothetical example is not clear cut.

Furthermore, we do know that in-group bias and prejudice exist. Not only do they exist, they are - to greater or lesser degrees - normal psychological adaptions. What they're not, however, is reasonable. And this is why 'reasonable people' should prefer circumstances that counter our innate prejudices to circumstances that support them. (This assuming a broad view of the good towards which reason is directed.)

That said, I feel that finding the right social tools for a given job is a tricky business. I'd take a lot of persuading that a demographic quota is a net better criteria for determining selection than competence, for example. I also feel that resentment by marginalised groups, while understandable, is pragmatically unhelpful. The correct and pragmatic response from these groups, for me, has to be the assumption of responsibility by the individuals that constitute them for their own improvement. And for those people not marginalised? Well if they're interested in a broadly improved world (rather than the narrow improvement of their 'tribe' - perhaps another heroic assumption) then they should not unduely emphasise narratives that undermine peoples' efforts to take responsibility for improving themselves (like the notion that individuals are the captives of the demographics to which they belong). And all parties should seek to minimise their inevitable prejudices. Science and the scientific conception of truth can get you half way, but it is not - to my mind - sufficient alone.


@ Moyenda

I agree with quite a few of your points and warnings. Certainly nobody is suggesting we use population level statistics to affect how we treat any individual, indeed it would be wrong and silly to do so.

However as Haier and Sullivan make clear in their excellent and mature (yet devastating) takedowns of Vox (and this blog), there are risks in believing comfortable lies and there are dangerous precedents when people carefully expressing the dominant consensus of science are denigrated as racist, sexist supremecists.

Murray, Harris, Sullivan, and Haier have both the moral high ground (both in pursuit of truth and in refusal to attack and lie about their opponent's character) and the scientific consensus behind them.

Handy Mike


"here are risks in believing comfortable lies"


This is the real problem.

Every day liberals and 'progressives' spend lying to themselves and others about this stuff is another day some of the most awful people in the world spend getting more familiar with the detail of it and preparing their catastrophic construals of how it should inform policy and society.

The comments to this entry are closed.

blogs I like

Why S&M?

Blog powered by Typepad