Reply To Lance Welton: Why Do Blacks Outperform Whites in UK Schools?
Print Friendly and PDF Editor Peter Brimelow writes: And now for something completely different…I was deeply impressed to receive this contribution from Zambian immigrant Chanda Chisala, who has emerged as one of the most ingenious opponents of the race-IQ hypothesis. Our Ruling Class’s position, as shown by the Stephen Miller and “Judge” Ashley Tabaddor farces, is that should simply be shunned, not engaged in rational debate. Naturally, we think this is because Ruling Class positions cannot stand up to rational debate. So we have a soft spot for mainstream figures who have nevertheless engaged us in rational debate, for exampleUniversity of London Professor Eric Kaufmann, more recently the author of Whiteshift: Populism, Immigration, and the Future of White Majorities,and the blogger Razib Khan, hired and fired as a New York Times columnist on the same day because he had written us a letter back in 2000 disagreeing—disagreeing!—with something we’d posted. I very much hope nothing nasty happens to Chanda Chisala as a result of his commitment to the West’s scientific method. recently published an interesting article,Race Denier Chanda Chisala Does Not Deserve His 15 Seconds of Twitter Fame by Lance Welton, responding to my Unz Review articles on the superior scholastic performance of black African high school kids in the UK, who now consistently outperform British whites, thereby putting into serious question the race-genetics-IQ promoted by the Human Biodiversity (HBD) intellectual movement (henceforth, HBDers).

Welton’s article contained a number of errors and misunderstandings that I would like to correct, and hopefully help the reader to get a better grasp of the very real challenge posed to the race hypothesis by the UK’s test score data. I will end by demonstrating that the only “logical” HBD response to this unexpected data—the claim that these high-scoring Africans must be just children of highly selected Africans—is itself untenable, which should make the race hypothesis implausible, and thus falsified.

Is Race Really Real?

Welton’s errors begin on a personal level with the very first words in the title: ‘Race denier Chanda Chisala...’ The idea that anyone who opposes the conclusions of HBDers is somehow some radical “woke” progressive liberal who denies the very existence of race or the possibility of any gene-based individual (or population-level) differences is rather presumptuous. I’m not and I don’t.

I guess Welton did not bother to read anything I’ve written on the subject in the past since he could not spare more than “15 seconds” of his time. I am as far from a “race denier” as you can get, and I believe I made this clear in at least one of those articles at UNZ that he lambasts:

“I am not arguing against the heredity of intelligence in families or tribes or (theoretically) even races. Neither am I arguing that all ethnicities and races on earth necessarily have the same average intelligence, presently or potentially, or that races do not even exist – all straw men that so many commenters have been ascribing to me so that they could enjoy the illusion of making an argument. The average (genetic) potentials of intelligence could indeed be as varied as the heights of different populations. What I have contested is the empirical evidence for the specific genotypic estimates and rankings of the racial cognitive heights, if you will.” [emphasis added.]

Closing the Black-White IQ Gap Debate, Part 3, December 9, 2015  

In short, I have never thought that race is nothing more than a mere “social construct”; my humble academic background is in biology, not sociology. Populations that have lived separately for a very long time are probably bound to have some differences that are a little more than skin deep.

But that does not automatically mean that the way these populations rank now, under strongly variable environmental pressures, is necessarily the way they would be ranked under optimal conditions, when no one faces any significant detrimental assaults.

Similarly, genotypic heights are bound to be different among different individuals, families, tribes and races, but this does not mean that what we see now among these groups reflects their ultimate genetic height rankings. (For example, Nigerians are reportedly shorter than Koreans under current height rankings—should we assume this is how it will be even when all environments are optimized for full genetic height development?

Welton also suggests that I’m also “environmental determinist”  and, even more puzzlingly, a “science denier.” And just in case his reader hasn’t yet despised this race-and-science-denying environmental determinist  enough he throws in a bit of slander for good measure:

But to reach this conclusion, Chisala has manipulated and misrepresented his data.

To accuse someone of manipulating data is of course an attack on his integrity. The simplest evidence that I have “manipulated” nothing is the fact that this UK cognitive data was troubling to many race hereditarians themselves long before I was involved in the debate. [A Troublesome Intelligence? by Chanda Chisala,, September 26, 2018]

And that was when they thought the British children were scoring only an equivalent of 8 IQ points above the black African kids, despite the black children being drawn from a population that is supposed to be about 30 IQ points lower. When I entered the debate by showing that their “troubling” data was in fact misleading (those children who were from English-speaking African countries like Nigeria, Ghana and Zimbabwe were already outscoring the British whites), it was not long before some of the more mathematically adept HBD bloggers started publicly declaring that they were “now agnostic as to the real cause of the intelligence gap”! [Changing My Mind About Race and IQ, Jason Bayz, September 18, 2017]

Can Blacks Do Math?

When Welton finally leaves the ad hominem mode and decides to make some actual arguments against my observations, it quickly becomes clear why he needed to first poison the well so viciously: he actually does not have any good argument.

He begins by rehashing the arguments that I’ve most easily refuted already. He of course calls my refutations mere “attempts” at refutation, but fails to show why those “attempts” fail:

Clearly, GCSE [General Certificate of Secondary Education, a U.K, standardized achievement test]  Math is a better proxy for IQ than GCSE Art or GCSE Physical Education. Yet Chisala’s data relates to students' best 8 GCSE scores and the students' mean pass rate, meaning it does not distinguish between “hard” and “soft” GCSEs.

My easy refutation of this claim was just to present the group scores in GCSE Math (and GCSE English) separately. The Africans in question outscored the British whites even in those “hard” GCSEs. So why would anyone continue to insist that eight subjects are invalid because they contain some easier subjects?

I was curious to see the big counterargument that he promises to diffuse my “attempt” at refuting this point. But it only betrayed Welton’s unfamiliarity with both the history of the IQ debate in general, and the UK GCSEs in particular.

Welton’s big counterargument to the higher math scores of Africans, right after saying they only score well because the eight subjects are too easy, is this: it doesn’t matter if those blacks are outscoring British whites in math because math is not an IQ test anyway!

Unbelievable? Let’s see it in his own words:

Chisala attempts to get around this problem by showing that Africans in the UK—as distinct from those of Caribbean heritage—score higher than whites even on GCSE Mathematics. The obvious counter-argument is that GCSE Mathematics is still not the same as an IQ test. It includes coursework, for example. This is not done against the clock, so it will favor those who are high in certain personality traits, such as Conscientiousness (‘rule-following’ and ‘impulse control’). These traits are themselves associated quite strongly with educational success ...and, according to some studies, migration.

This has been the modus operandi of many HBDers in this debate. When a goal is scored against them (or even an own-goal by themselves), they simply shift the goal posts and declare it a non-goal.

First it was “Blacks can’t outperform whites in school anywhere in the world - just find one country where they do and it’s game over.”

When it was shown that black Africans do perform better than whites in the UK, it changed to “well, those are just pass rates, they can’t outperform whites when you look at actual scores.”

When actual scores were found and presented, it changed to, “Well, that’s in eight subjects; blacks can only beat whites in eight subjects because most of those are easy subjects, unlike hard subjects like math.”

When actual scores in math were found and presented, it has apparently now changed to “Oh, GCSE math has a lot of coursework, and blacks only beat UK whites in math because coursework requires conscientiousness rather than intelligence; if there was no coursework, whites would obviously beat blacks.”

In short, those high-scoring Africans are actually not that smart, they are just more serious about work, like all immigrants, and they would never beat whites if coursework was removed and it became a time-controlled hard test like the SATs in America.

On such a timed test, the greater brains will prevail, and that’s the British whites. Makes sense, right?

Well, it seems they might have to shift those tired goal posts once again!

It turns out that Welton does not quite know the GCSE system very well, which is why he imagines that he has the final “obvious” counterargument in this endless chain of shifting goal posts. GCSE mathematics does not actually have any coursework. It has not had coursework since 2009. Yes, it is indeed done “against the clock,” as he desires, just like the SATs.

In fact, when the coursework was abandoned for the very reasons expressed by Welton, the perennial UK trend of girls beating boys in GCSE math ended immediately. [Boys overtake girls in maths GCSE as coursework dropped, by Polly Curtis, Guardian, August 27, 2009 ] Boys started outperforming girls, just as they do on the math section of the American SATs, which probably confirms an increase in the g-loading (cognitive complexity) of the GCSE math sans coursework.

However, Africans from English-speaking countries still out-performed whites—which should indicate, by the hereditarians’ own theory, that this black African advantage is really on general intelligence (g) itself.

If anything, it appears that it is the British whites whose performance improves when you remove the harder subjects (math and English), while the African performance goes down drastically without their advantage on the math and English sections.

Although more data is probably needed to prove this more conclusively, it does seem like the diametric opposite of what Welton and other HBD enthusiasts predict:

Source: Back to Basics: Towards a Successful and Cost-Effective Integration Policy

Yes, math scores are not IQ tests, but the whole point of this debate—the entire reason that the debate exists in the academic world—is to resolve the claim that racial IQ differences are what ultimately explain things like math score differences between black and white Americans. Arthur Jensen’s famous article that relaunched this debate in modern times was precisely titled “How much can we boost IQ and Scholastic Achievement?” [Harvard Educational Review, Vol. 39, No. 1, Winter 1969]

Jensen’s controversial answer was that the wide test score gap cannot be closed between blacks and whites in America because of differing biological capacities that produce an intractable IQ gap.

Richard Lynn, Philippe Rushton and others later proclaimed that this large gap between American blacks and whites on scholastic achievements should be universal, because it is rooted in the evolution of biological race itself rather than something specific to the historical black experience in America.

Anyone who rejected this universal race-based conclusion was openly challenged to find just one multiracial society where this was not conspicuously true—where whites did not vastly outperform blacks in school—and this would be accepted as a simple falsification.

If a multiracial society is found where these race differences in intelligence are absent, the evolutionary and genetic theory of these differences would be falsified. Those who maintain that there are no genetic differences in intelligence between the races are urged to attempt this task. – Richard Lynn. (Emphasis added) [ Consistency of race differences in intelligence over millennia: A comment on Wicherts, Borsboom and Dolan. Personality and Individual Differences, January 2010]

Lynn only asked that such a society be found for the falsification to apply, no ifs or buts.

And it’s not an unreasonable standard in biology: I believe one can plausibly say “if you find one multiracial society in which East African runners are outperformed by any other regional or racial group in the long distance races, you would falsify the strong genetic hypothesis for the higher long distance running speeds of those Kenyans and Ethiopians.”

When I humbly heeded Lynn’s bold open challenge to “attempt this task” and found one such society where the gap did not appear to exist in favor of whites at all (in Lynn’s own “multiracial society” of Great Britain), I expected the HBDers to at least admit, by their own standard, that this simple unambiguous falsification standard had apparently been met.

Instead, by doing exactly what they had asked for, I only proved that I was a “science denier”!

Can you see why someone might suspect that this once-valid scientific hypothesis is now beginning to exude the obstinate fervor of a pseudoscience?

The Super-Selected Immigrants?

Welton eventually makes the single argument he should have made in the first place: that African blacks in the UK are merely the most cognitively selected segment from Africa. This supposedly makes it possible for them to score above whites on average, even if they come from a race that is cognitively inferior on average.

Remember, the more universal your claim, the more it can potentially be falsified by a simple singular unambiguous event, which is why Lynn was right to give such a simple falsifying standard. It is not the complex confirmatory data that is critical for a hypothesis; it is the potentially disconfirmatory data.

If you hypothesize that all swans are white, it doesn’t matter how “overwhelming” the number of white swans you’ve observed in the past is; it will take only one black swan—no pun intended—to falsify your universal hypothesis.

So does this African immigrant self-selection explanation survive under any level of serious scrutiny?

There are two main reasons why the super-selection argument fails to save the race hypothesis:

  1. Arthur Jensen and regression to(wards) the mean

If it is true that the African immigrants are a select group whose IQ is, say 2 standard deviations above the African mean, their children should still regress significantly downwards—regression to the mean.

No, I am not saying they should regress all the way down to average African IQ. They should regress TOWARDS that mean IQ. In other words, their IQ should be significantly lower than that of their “selected” migrating parents, even if higher than the average of their source population.

Jensen and others demonstrated that black American parents with relatively high IQ (and high income) still produce children with IQs that are so low that they score lower than poor whites in school tests. Note that these are black American parents—the ones coming from a higher IQ population than Africans, supposedly because they carry more white genes!

Think about that for a moment. The hereditarian reason that those elite black American parents (average IQ around 110) have their children being outscored by poor whites (not just average whites, but poor ones—household combined income $20,000 per year, and IQ less than 100) is because they come from a genetic  population that has a low average IQ of 85.

Keep that in mind, even if you don’t understand precisely how it works: Population IQ of 85 pulls down their children so way down that even low IQ poor whites outscore them.

So, what should happen to children of Africans whose parents are at 110 but are being pulled downwards towards IQ 70 (not 85)? They should score even lower than those children of high IQ black Americans who are outscored by poor whites—not to mention the average whites!

Hereditarian scholar Gregory Cochran has given us an idea of how smart black Americans have to be for them to produce children who can outscore whites in school:

Consider IQ. Imagine a set of parents with IQs of 120, drawn from a population with an average IQ of 100. Suppose that the narrow-sense heritability of IQ (in that population, in that environment) is 0.5. The average IQ of their children will be 110. That’s what is usually called regression to the mean. Do the same thing with a population whose average IQ is 85. We again choose parents with IQs of 120, and the narrow-sense heritability is still 0.5. The average IQ of their children will be 102.5—they regress to a lower mean.

[The Breeder's Equation,, January 1, 2017]

Notice how smart a select population of black parents have to be to just produce children that score slightly above average white children—50 IQ points above the African mean IQ!

If that doesn’t already sound ridiculous to you, consider that this calculation is in fact for black Americans, whose mean IQ is 85. The black African parents would have to be much smarter than 120 IQ to produce children who are at or just above white average IQ.

If a population of black American parents need to be IQ 120 to produce IQ102.5 children on average, black African parents (regressing toward IQ 70) would need to be at 135 IQ!

Yes, to migrate from Africa supposedly requires the cognitive endowment of a gifted American surgeon! [Modern IQ ranges for various occupations, IQ Comparison Site]

In fact, the occupation profile of these African immigrant parents certainly doesn’t reflect such ridiculously high surgeon-type IQs, but maybe the HBDers have another creative explanation:

This reductio ad absurdum should be enough to cause at least some modest level of consternation in an honest HBDer.

For the more math-demanding: you can simply estimate from the normal distribution how many Nigerians (for example) can have IQ 135—the IQ needed to produce children who can score above whites—and compare that with the number of Nigerian parents who actually migrated to the UK (or the US). At IQ 135, and mean of 70 with standard deviation of 15, you should find only 1403 Nigerians from the current population of 190.9 million. And since we are only talking about parents, there should be less than 700 such super-smart people.

Well, the number of Nigerian migrant parents in the UK is certainly nowhere near as low as 700. One estimate of the number of Nigerian-born people in the UK: 190,000 [Number of Nigerians living in the UK, Office for National Statistics, May 18, 2018].

Of course not all of these are parents, but no matter how much you want to quibble on some figure in this rough calculation, there is no way that the number of migrating parents is anywhere near the number required for an average IQ of 135.

This is even more dispositive when you try the same math on the migrants from Nigeria’s neighboring country, Ghana, with its much smaller home population.

  1. University Graduates’ IQ

HBDer Gregory Cochran, who (to his credit) has apparently now accepted the fact that at least some African immigrant do perform very well in the UK, has unfortunately also embraced this mistaken “super-immigrant” explanation. He writes:

“Nigerians in the UK seem to be a very non-random sample, highly selected: most of them have at least some college while maybe 5% of the general population in Nigeria does.”

An American Dilemma, West Hunter, October 24, 2019

Thus Cochran assumes that Nigerian migrants are highly selected for cognitive ability because they have much more college education than the general Nigerian population. If the IQ of college graduates is unknown, one might be justified in estimating that there is very high cognitive selection in a migrant group if most of them have “some college” education.

But J. Philippe Rushton reports that the average IQ of African university students is indeed not that high above what he reports as the average IQ of Africa: only in the high 70s!  [Thirty years of research on race differences in cognitive ability, by Rushton and Jensen,  Psychology, Public Policy, and Law, June 2005]

Rushton also tested the IQ of students at an elite African university in the most developed African country and reported that it was only around 85!

So even if you stipulated there was 100 percent university graduates in the African immigrant population, that would still make their IQ only in the high 70s.

And in fact the fraction of university graduates among these immigrants is never even 50 percent, which makes deriving high cognitive selection from these facts even more problematic.

The only way to try to get out of this trap is to suggest that these university graduates are only IQ 78 because of the harsh African environment:

...some people have argued that the hereditarian view on race and IQ is falsified by the fact that the offspring of immigrants don’t always regress back as far as one would expect given their parent’s IQ. The mistake in this reasoning is to calculate the predicted degree of regression based on the parent’s IQ as opposed to the IQ they would have had if they grew up in a first world Western nation.

 Regression to the Mean, by Ryan Faulk, The Alternative Hypothesis, January 16, 2017

But this is even more self-defeating.

If you say that these parents would have indeed been above 135 (instead of IQ in the 70s or 80s) had they grown up in the UK, then you are suggesting that the African environment depressed their IQs by around 50 IQ points or more—which more than defeats your hypothesis!

Thus, even if you reverse the procedure of the calculation, it doesn’t solve your problem. If the parents have IQ in the high-70s or even mid-80s, but their children are in the 100s (instead of regressing way lower than their super-selected parents) due to being brought up in a more advanced Western country, then you are still saying that the African environment depresses IQ by 30 points or more, which still eradicates and/or reverses the racial IQ gap.

Or Maybe It’s IQ 85?

When HBDers see that the math is just not working for them, they start offering that the real or genetic IQ of Africans should be significantly higher than 70 and concede that it would be so if Africa had a better environment. They usually assert that it should be 85, like the blacks in America.

When you ask them how the black Africans could be the same exact IQ as a group that has more “superior” white genes in them (black Americans have over 20 percent white admixture and IQ 85), they simply become more creative:

“The slaves who were captured in Africa and sent to America must have had very low IQ—because they allowed themselves to be captured—and only came back up to 85 after they got some white genes in them!”

How neatly convenient: the “dumb” African slaves just happened to have a cognitive deficit that also happened to be the exact amount that whites would generously restore to them through a random genetic contribution!

Never mind the fact that many of these “dumb” slaves were captured from a tribe called the Igbo, which the same HBDers started claiming is an exceptionally bright tribe when I first wrote my first article revealing how African tribes perform in the UK! [The IQ Gap Is No Longer a Black and White Issue,, June 25, 2015  ]

The biggest problem with accepting this IQ 85, however, is that it still doesn’t solve the problem of IQ 120 migrant parents. If they should be 120 to produce children who are slightly above the white mean, but are below IQ 90 due to being brought up in an African environment, then the African environment artificially depressed their IQ by 30 points or more.

Again, this means that environment explains the entire IQ gap, and so is a total refutation of the race hypothesis. The hypothesis still dies, even if it dies less.

Besides, the same HBDers who say that African genotypic IQ should be closer to 85 than 70, also tout the works of HBD researchers like Davide Piffer who claim to have found new genetic evidence that it should be closer to 70 (or less) than 85. They want to have their cake and eat it too.

Honest HBDers owe it to scientific integrity to admit that their hypothesis is now falsified by unmanipulated empirical data.

Chanda Chisala (email him | Tweet him) is a former Knight Fellow and former Visiting Scholar at the Hoover Institute, Stanford University. You can support his ongoing self-funded contributions to the IQ Debate through

Print Friendly and PDF