Arthur Jensen’s generation of race hereditarians (Eysenck, Rushton, Shockley and perhaps even Charles Murray et al) were quite different in posture from many of their current young followers. Jensen, like most of his friends, apparently wished to be proved wrong about his genetic hypothesis of racial differences in IQ because he genuinely hoped that what he was observing was not true so that something could be done about black underperformance in school. This made him take almost every single published argument against his theory very seriously, and he patiently addressed their points even when they came from writers with limited knowledge of the field. He would at times actually even help strengthen the argument of his opponents, but his integrity obligated him to show that it still fell short, given all the statistical evidence at the time.
By contrast, some “neo-Jensenites” seem more concerned about being proven right and are terrified of the possibility that their anti-establishment Galileo complex — they have suffered much unfair personal and professional ostracism for their “scientific racism” — was actually never based on any vindicating ultimate truth. As such, they behave the same way that most environmentalist critics of Jensen have always behaved, out of a grave fear of being proved wrong: they gratuitously insult their strongest critics; they erect massive strawman arguments; they misrepresent the results of their debates with critics; and sometimes they even censor (delete) arguments that dismantle their weak conjectures.
Fortunately, there are exceptions on the other side, like Dr. James Thompson (plus a few pseudonymous commenters in our threads – Chuck, Szopen, Res, et al), who are willing to at least engage the critics of their favored racial genetic hypothesis in a robust but civil debate.
In his last article in response to my corrections of his statistical methods on testing the biological veracity of the low Nigerian (and African) IQ, Dr. Thompson responds by correcting my corrections to his calculations. Now, that’s the spirit: He won’t go down without a fight!
But — he’ll still go down!
Let’s start with the easiest ones.
1. The Chess Debate
Firstly, Dr. Thompson “corrects” my factual claim that the only chess grandmaster in South Africa is a black man (Kenny Solomon), and chides me for “confusing” a local African unofficial “grandmaster” title for an official FIDE title:
“However, I do not agree with his dismissal of the chess results as being due to a lack of training in Africa, particularly when Chisala then goes on to make a point about Kenny Solomon of South Africa being a Chess Grand Master. FIDE does not say that, because he has not reached their threshold requirement of 2500 points. …It shows how important it is to set thresholds to begin with, or one mixes local rankings with the ones that matter most, the international open competitions.” (my emphasis).
Solomon’s grandmaster title is not a “local ranking”; it is an official FIDE title (yes, FIDE does say that). Thompson could have checked his own link to the FIDE web site that ranks South Africans, and he would have noticed that FIDE itself is the one that gave Kenny Solomon the grandmaster title. He was apparently misled by (his reading of) Wikipedia. Forgiven.
Thompson gave us that link to the rating list of South Africans at FIDE because it gives a white person (Daniel Cawdery) as the highest rated player in South Africa. However, since one can accrue ratings by playing in more tournaments (especially if one can travel abroad for tournaments), higher ratings do not always reflect superior strength, particularly when matched against higher FIDE-titled players.
For example, as recently as June 2017, there was an open chess tournament for Southern African players. Grandmaster Kenny Solomon, who has apparently not played much in recent years, took part this time, with some of his South African teammates, including their highest rated player (Cawdery). Contrary to the ratings ranking preferred by Thompson, the actual results at the end of the tournament seemed to align with the official Fide titles more than the ratings:
Some of the other HBD bloggers also think that chess ratings are perfectly representative of innate chess talent (since Africans have low ratings). One such blogger observed that the highest rated players of European descent are over 1 standard deviation higher than the highest rated black players, and of course this was to force the idea that chess confirms the (at least) 1 SD difference in IQ scores between blacks and whites.
However, anyone who follows chess would know that things are not quite that simple. One of the highest ever rated American chess players (second only to Magnus Carlsen at his peak rating in 2015), Hikaru Nakamura (of mixed Japanese and white American descent), does indeed appear to be around 1 SD above Zambia’s grandmaster Amon Simutowe in rating, and this would make someone think that the two players are in completely different universes of chess talent. But it’s not hard to see that this is only because Simutowe has played in very few international tournaments in his career (he was mostly self sponsored to these international tournaments, contrary to claims made by the same blogger). The most obvious proof that he is probably not 1 SD below Nakamura in innate chess ability is the little fact that in his more active days, Simutowe did play against Nakamura in two very strong tournaments. In both of those games (2001 and 2007), Nakamura was beaten by Simutowe, with the latter defeat at the prestigious US Open being particularly spectacular.
Unfortunately, Simutowe stopped playing active chess not long after he achieved his childhood goal of officially becoming a grandmaster and instead decided to pursue his studies at Oxford (where he played the top board for the varsity chess team), while Nakamura, who has been professionally coached since very early childhood, continued playing chess full time and accruing even more points until he reached top two in the world. Simutowe, like the other African chess players, has never had a single chess coach in his life.
Incidentally, this mistake of taking chess rating as unqualified representation of innate chess ability, without considering the different player contexts, is very similar to the commonest fallacy made by the race hereditarians on IQ itself. They take the scholastic test scores of people in very different countries, for example, as near accurate representation of their true cognitive abilities and see nothing wrong with that. I guess they think that since twin studies in developed countries have shown that genes are ultimately much more decisive than home (or school) environment in explaining test performance variance, then this can be extrapolated to comparison of people in very different societies, including countries in which the average pupil has never seen a mathematics textbook and is probably taught math by a teacher who actually failed her high school math (because anyone who did not do that badly in math would be very unlikely to choose the scandalously low teacher salaries of African public primary schools).
One of the reasons Thompson gives for thinking chess would be a better proxy for cognitive comparison is that it appears to be more complex than Scrabble, (for example, he writes, “One would have to make allowances for the fact that Scrabble is simpler than chess…”). However, it is Thompson himself who also reported for us a study in his field that found performance on some video games to have a high correlation with intelligence. The authors of that study noted that
remarkable relationships [with general intelligence] are only achieved when video games comprise moderate levels of complexity, display low consistency across practice sessions and have no possibility of obtaining benefit from previously acquired skills…
I proposed moderately complex games (Scrabble and checkers) with high cognitive demands (at least at the super-elite level) but scarce advantage from educationally acquired skills. Chess, like math, has thousands of books (of long opening lines of play analyzed by masters) and also benefits highly from exposure to well-trained teachers (although it is likely that general intelligence still makes a difference among highly trained players).
The reason I used Kenny Solomon’s grandmaster achievement in South Africa is not to show that blacks are genetically smarter than whites in South Africa, but to show that even chess might not be as friendly to the racial hypothesis as our racial hereditarian friends presume, once the chess training/educational environment begins to get a bit more equalized. South African whites may have a longer history of playing competitive chess and probably have more access to chess books, but the country as a whole is not that developed in the chess education culture (coaches, chess schools, etc), so the training regime gap between the different socioeconomic classes is significantly reduced. The results of this self-teaching environment begins to approach the “miraculous” results of Scrabble and checkers: the group that has the borderline mental retardation level IQ is the one that has produced the only chess grandmaster in the most multiracial society of Africa, just as it has also produced the top professional checkers grandmaster of the country (Kondlo) who rose as high as first in the world of American checkers.
Incidentally, I’m sure Thompson will be tempted to abandon his confidence in chess as the better reflection of the IQs of (most) nations when he analyzes the names of the people in the world top 100 or even the names in some of the Western national chess teams. Here are the names of the top players of the innocent nation of Canada, for example:
Yes, Canada can probably beat any African country in chess and our HBD friends will obviously conclude that this is because of Canada’s superior IQ of 99, but how much of Canada is Canadian? I may be going out on a limb here, but I’m going to guess that Evgeny Bereev and Anton Kovalyov are not descendants of the first French settlers of Nova Scotia!
It is clear that the state-sponsored chess culture of the former Soviet Union has distorted international chess rankings to this day, and the distortion goes way beyond their former communist allies (note that Russian Canadians are less than 2 percent of the Canadian population, but dominate their top player list). So yes, differential training cultures and resources is too large a confound for the role of national IQ in international chess rankings. (Incidentally, most of the top Scrabble players in any nation do not share this ancestry problem with chess.)
I also looked briefly at the Wikipedia profiles of the top 5 chess players in the US (all of whom are also on the top 100 of the world that, as Thompson reminds us, has no African). Although the top 5 US players are not necessarily all of Russian descent, there was not one whose father was born in the US. Among the top ten, there was only one who was possibly American-descended beyond one generation. This is further reason to reduce your confidence in a meaningful correlation between national chess strength and national IQ.
It’s also interesting that Dr. Thompson seems to regard membership to the top 100 list of chess players as a valid estimate of chess strength of different nationalities. When he noticed that the list of the top 100 players in Scrabble contain more Nigerian than American or British players, he decided (without explanation) that that’s not the best way of judging Scrabble strength: he proposed that we look instead at which nations have won most world championships (an obviously weak metric since many factors could affect why a country did not win in the past: for example, some African countries, including Nigeria itself, have actually missed some of the past world championship tournaments due to failure to obtain visas.)
Finally on this point, it is not necessary that a game be played by all nations or many nations for us to test if Africans should be able to compete at top levels. We only need to establish that the top levels of performance in any higher-IQ country require cognitive strengths that should realistically be inhibiting for Africans, given their average IQs (just as it is inhibiting for children). A gross defiance of such statistical expectations would only be possible if their nominal IQs are grossly determined by non-biological factors. Environment can be highly plastic, biology can not.
[Note: Some hereditarians, including Thompson, also keep reminding me that their models already consider environment as part of explanation of variance, but this is an irrelevant distinction since they (should) believe that the only environmental factors that could have a significant effect on IQ work through biological means: eg making the brain smaller through lower nutrition etc, as Lynn et al suggests. They normally reject the large significance of things like training/teaching resources, given twin studies in America. Since I am concentrating on samples that were born -- and mostly live -- in Africa, it matters not whether the variance is 100 percent genetic or 50 percent, since it is all still supposed to be mostly biological. A defiance of predicted performance refutes the significance of all biological limitations, whether the proposed source is environmental or genetic, and leaves only very artificial differences as significant sources of variance.]
2. Thompson challenges my Nigerian calculations.
“If Scrabble IQ 140 is required, then dividing by half three times brings the smart fraction down from 1336 Nigerians to 167. Chisala gets the number down to 33 so he must have used a higher discounting… He has discounted in a different way, and can explain the precise reductions in our later exchanges…”
This was about estimating how many Nigerians should be able to competently play international elite Scrabble if it requires high IQs (above 140) as it apparently does in the US.
No, I did not use any higher discounting to arrive at my numbers. Dr. Thompson’s calculation here assumes a Nigerian IQ of 75, and yet he begun the same article by informing us of new research that confirms that Nigerian IQ is just 70, which is almost exactly what Richard Lynn calculated many years prior. Dr. Thompson will arrive at my numbers if he uses that same “confirmed” IQ of 70 for Nigeria.
But even if we used his IQ of 75, Dr. Thompson should help us understand why Nigeria, which has a lower average IQ than a Western hospital of mental patients, is able to produce challengers to “normal” Putnam-Prize-level mathematicians from the West in Scrabble. Could you raise a team of world class Scrabble players from such mental hospitals if perhaps you gave them enough government sponsored training? Why not? Isn’t *the whole point* of IQ to tell us about real world potential performance? (Watch how the other race hereditarian bloggers transform themselves into a convoluted hybrid of hereditarian-environmentalists to try to answer such questions using endless self-contradicting conjectures and epicycles!)
3. Thompson rejects my correction of his calculation assumptions.
“Chisala suggests that my estimate of the number of bright Nigerians be reduced … I think these reductions are excessive, and in excluding women from the general calculation of the smart fraction (because they are not in the top ranks of Scrabble players) mistaken.”
Not mistaken at all. You do have to cut down your base population quite “excessively.” I will just point Dr. Thompson to the link he recommended himself in this same article in which he was calling my method “mistaken,” which is an article by a stats legend within the racial hereditarian community, the pseudonymous LaGriffe Du Lion:
“Raw population statistics are not best suited for the computation of the required population fractions. At this level of chess women are simply not competitive. Neither are children or elderly. Consequently, a base population for this calculation is better characterized by the total population minus women, children and geezers. A good approximation to this number is obtained by reducing a population to 25% of its original value.” (My emphasis).
Du Lion’s “excessive” reductions even include the old people (“geezers”), which should probably be also applied to our base population of these Nigerian Scrabble players, thus making the total number of Nigerian men capable of playing elite international Scrabble even smaller than 30 people (in a country of 186 million!) — which means that even if all Nigerian men at that level of intelligence have chosen dedication to Scrabble as the best way to take advantage of their supposedly extremely rare intelligence in an extremely poor country, there would *still* not be enough of them to reach the number that they’ve actually reached on the elite Scrabble list. That’s a reductio ad absurdum, if there ever was one.
The only logical reason for this dramatic defiance of statistical predictions is that the IQ estimate of Nigeria was gathered from tests in which performance development resources play a much larger role than they do in Scrabble. For the exact same reason, Nigerians (and other Africans) will also perform way above expected in many other cognitive areas (including education) when they grow up in a country with such abundant resources, as has happened in the UK and the US in recent years.
Which takes us to Thompson’s next attempt at correcting my corrections.
4. African Self-Selection.
Thompson incidentally dismisses my usual argument that immigrant black Africans in the US should not be so dominant in academic achievements compared to native black Americans if the latter have a genetic cognitive advantage arising from their (25%) white admixture. He suggests that the black Africans are so conspicuously dominant simply because of extremely high immigrant selection from African countries (like Nigeria.)
This has been the most baffling rejoinder I have constantly faced from so many race hereditarian commenters from the time I started writing these articles. I really can’t understand what they fail to get here, since their fallacy is too obvious: high immigrant selection, *even if it was true*, should not affect relative performance *at the top*, especially given the size of the IQ gap between black Americans and continental Africans (15 points).
Yes, immigrant self selection can affect the average, but not the top performance. I’ve given these examples repeatedly: if all the fastest running Indians migrate to Jamaica, they will indeed run faster than the average black Jamaican by far. But that doesn’t mean *any* of these fast Indians will make it to the top Jamaican team. For a population from such a slow average group to overtake the average performance of a faster group, self selection level of the immigrants could indeed be a plausible explanation. But for extremely high elite performance (*at the top*), the average performance of their source population should continue to impose the limitation on the absolute numbers, particularly IF it has a biological cause. This statistical prediction can only be defied if very artificial environmental conditions of the two source populations were in fact the real cause of the difference in the original population averages.
Not only do my opponents give immigrant selection to explain why black Africans seem to outperform black Americans at the top, they incredibly even give that as the explanation for incidents when a black African student outperforms all white students in the UK. No, that gap (30 IQ points) is too large for any top honors to ever be affected by self selection: you need some new rationalization to explain such common “miracles.” Self-selection may affect average performance, but not top performance, especially not in the first two generations (in case someone thinks this has to do with immigrant Africans “breeding” a new cognitive race, as one of the more inane HBD responses once suggested!)
If 50 percent (or any portion) of the smartest people of Europe migrated to America, they could have a higher average IQ than even the Ashkenazi Jews in America. However, if you try to form some sort of Manhattan Project of the absolute smartest Americans to quickly build a new quantum bomb to vaporize all North Korean nuclear facilities remotely, the team will still probably be virtually all Ashkenazim, which would at least not contradict a possible biological/genetic cause of the superior Ashkenazi IQ. Every hereditarian can understand that.
Well, the reported IQ gap between the black Americans and continental Africans, as I’ve stressed ad infinitum, is even larger than the gap between the Ashkenazi Jews and Europeans, and the proportion of the American Ashkenazi Jews (6 million) to Europeans (743 million) is much smaller than the proportion of black Americans (46 million) to Sub-Saharan Africans (800 million). In short, this would be like increasing the Ashkenazi population 7 times while also increasing their IQ advantage by another 3 extra IQ points, and instead of resulting in an even higher dominance of Ashkenazi Jews at super-elite levels, we actually see a sharp decline!
The more selective the cognitive requirements of membership are to any elite class, the less immigrant blacks should be in proportion to black Americans. The trend, it seems, is the exact opposite. And it is not just these supposedly super-selected African migrants, but even their children (including those born before the migration) – the same children that should be *additionally* disadvantaged by a sharp regression toward that “retardation level” African IQ – who conspicuously dominate these high academic honors compared to native black American children!
Even as far back as 1999, when African immigration was much lower, the trend at elite American colleges was already against the prediction of the racial hypothesis. The Harvard Crimson reported:
The trend should be the other way around (as you go from “universities” to “Ivy League schools”), just as it is with Ashkenazi Jews’ increasing numerical dominance as cognitive selection rises.
The *only* way this statistical prediction can be defied is if the IQ advantage of black Americans over black Africans is only caused by environmental factors that are highly artificial and completely dependent on (the cognitive resources of) location. And if it is completely environmental, then the white genes in black Americans (their 20-25% admixture) give them no biological advantage, a conclusion that explicitly contradicts the key piece of “commonsense” evidence given for the global racial hypothesis by every leading racial hereditarian, including Lynn, Rushton, and even Cochran:
Cochran even understands why *top* elite performance should most conspicuously reflect such average differences, at least if they are real and not artificial:
Through my writings here, I have highlighted several failed logical and statistical predictions of the racial genetic hypothesis, and I have seen no explanation from the other side that does not borrow elements contradicting their own theoretical framework. They can’t explain the black African vs white children cognitive performance anomaly; or the black African men versus white women cognitive games performance anomaly; or the black African versus black American academic performance anomaly; or even the black African versus white “mental patients” performance anomaly.
If you have reasons that can explain away any of these apparent anomalies for the racial hypothesis, remember that those reasons will automatically contradict the entire point of IQ scores: to predict performance in real world cognitive tasks. White mental patients and 12 year old white children should not only be able to win more world championships in Scrabble and checkers (given sufficient training) than the adult Africans, they should even produce more National Merit finalists (etc) if they attempted the selection tests! (Not even the most radical 10,000-hour-rule environmentalists would be that optimistic.)
In fact, this author, as a black African, should really be having these debates with some smart 11 year old white children, as the immense intellect of smart 13 year old whites should be rather overwhelming for his humble soul!
The reductio ends in comical absurdity.
For as long as you have not found any instantly confirmable explanations for real-world commonsense trends that explicitly contradict the logical expectations of your models, your hypothesis should stand as falsified by default, no matter how much supporting evidence you thought you had. Scientific propositions are not sustained by the fervency of hope and loyalty.