Note: Although he will probably disagree with its conclusions, this article is dedicated to Professor Thomas Sowell, whose astonishing breadth and depth of research are humanly unmatchable.
I have held the main idea of this article for almost a year now, and I recently noticed that some of the concepts that form the building blocks of my hypothesis are being slowly blogged about, which should hopefully make it easier for many people to understand the thrust of my argument.
Before I heard about the term ‘canalization’, I speculated on this theory about racial groups and intelligence: The different racial groups have faced different levels of environmental challenge and have survived by evolving different levels of a protective mechanism (probably through natural selection) against environmental insult. The East Asians, who faced the most severe climatic and other environmental challenges, have evolved the highest level of environmental protection, and sub-Saharan Africans have the lowest. In modern times, this has meant that East Asians can survive strong environmental intellectual deprivations (low Socioeconomic Status, etc) more than the other groups, which ensures a higher average IQ by having very few people phenotypically falling into intellectual depression at the bottom end. I didn’t quite know how this worked at first (or what other kinds of protection could be involved), but I thought there was a lot of evidence to justify the abstraction (and even more evidence against the proposition that such higher protection simply correlates with higher innate intelligence).
When I looked at the school performance data in the UK, this pattern was clear. There was practically no difference between Chinese students from low income households and those from higher income households.
Ron Unz had also found this East Asian IQ robustness in his earlier research on global ethnic IQs. He called it the East Asian Exception Hypothesis (we could, alternatively, yield to the pun and call it the Unzian Asian Conjecture!). Professor Richard Lynn dismissed the observation by suggesting that similar robustness equally applies to Europeans (for example, East Europeans did not suffer a very large IQ reduction from poverty-inducing Communist policies compared to West Europeans). However, Lynn’s counter-argument only works by focusing perhaps too literally on the word “exception.” In fact, there is much evidence that the East Asian average IQ is at least (significantly) more robust than any other ethnic group’s average IQ; under most circumstances in modern human history, it can quite rightly be described as absolutely exceptional.
The Europeans also do apparently have a (lower) level of protection against environmental perturbations, so all Lynn had to do was pick an example of a situation where they too largely survived a certain degree of environmental deprivation. He evaded a strong example (early poor immigrants to the US) in which European IQ seemed to have been much more depressed in the same conditions that did not similarly affect East Asian IQ. Our UK data above appears to also confirm that East Asian IQ is certainly much more robust than European IQ in the UK when we look at relative academic performance of children from different socioeconomic conditions.
Looking at the UK GCSE table, something else jumps out at you immediately: the girls perform above the boys in every group, but even more spectacularly among the Chinese since even the girls from the poorer environments are outscoring the boys from “rich” backgrounds. At the other (lower) end, of course, we also see the Caribbean girls beating the Caribbean boys significantly, although the environmental impact (within gender) is much more significant than in the Chinese group. In general, there is nothing very strange about girls beating boys on such tests and at that age, as this is common in many countries, but when the girls from poor backgrounds are also beating boys from higher income homes, there is some extra explanandum there that cannot be ignored.
Is there a theory that can explain all this? Would that theory help us explain racial and gender intellectual performance differences in general? To answer that will require a concatenation of a few key concepts gathered from different fields and sources.
THE SOWELL EFFECT
Thomas Sowell has been one of the strongest opponents of the genetic hypothesis for black-white IQ differences in the US and globally. Sowell has always used two arguments to cast doubt on the genetic hypothesis: the first one is the Flynn Effect or prior versions of it that he had noted himself, which shows that IQs have been rising with time for blacks and other people all over the world. The second very unique and original argument he has used is the differential IQ performance of black males and females, which seems to favor the females. He charges that the genetic hypothesis can not explain this, but it is explainable under an environmental hypothesis.
Sowell’s second argument is much stronger than the Flynn Effect argument because it is very difficult for hereditarians to explain why there should be a gender difference in African American IQ, especially one favoring females (let’s call this the “Sowell Effect,” to avoid repetition). This is very problematic for hereditarians, particularly since the trend is normally for male IQ to exceed female IQ, especially at the higher levels of the IQ distribution curve. We can see this unique trend among blacks even in the applications to medical school, a field that is considered a good metric for group intellectual comparisons.
For an environmental case, Sowell says there has been a pattern among other environmentally-affected low IQ groups in favor of females and cites some research indicating that female intelligence has been more robust against strong environmental insult. He gives an example of some study that showed that American Jews apparently had the same pattern when they also had low IQ scores as recent poor immigrants from Eastern Europe, before they radically improved after assimilating into mainstream American culture and raising their income levels. Sowell believes that this refutes a genetic hypothesis since there is no genetic explanation for such gender dynamics. The explanation from an environmental perspective is not unequivocal either: Sowell says the gender reversal among blacks could be because the same backward culture that depresses their IQ also emphasizes macho roles for men (as it does among the redneck whites too). He also says it might just be something that happens to all groups intellectually depressed by environmental factors, since it reportedly affected Jewish immigrants too.
I actually disagree with Sowell. There is indeed no racial genetic explanation for this Sowell Effect per se, but there is a genetic explanation, even if it has never been given by hereditarians, which is actually much stronger than the more ambiguous environmental explanations. It is genetic but it is not racial, which makes it problematic both for standard hereditarian racial genetic hypotheses and standard environmentalist social hypotheses for group intellectual differences.
The Economist (2014) published an article that reported on a study about numerical gender differences in neurological conditions like autism. This can give us an important clue for understanding the abnormal black female-male gap in IQ, which can in turn lead us to answering the question of the racial IQ gap in America and some other multiracial societies. The article, which was based on a study published in the American Journal of Human Genetics, discloses that this trend is not just common in autism patients but is true in all kinds of cognitive conditions:
Boys are four times more likely to be diagnosed with autism than girls are. For high-functioning autism, the ratio is seven to one. Moreover, what is true of autism is true, to a lesser extent, of a lot of other neurological and cognitive disorders.
The article also tells us that the mutations involved in such conditions are the same in men and women even though the manifestation of the condition is much wider in men. What this suggested to the researchers is that there could be a protective mechanism for females against these deleterious mutations; I propose that this is the canalization phenomenon that is well-known in genetics (which works by limiting phenotypic variability), although it has apparently not been assumed to vary by race or gender.
The milder IQ depressing mutations (causing relatively mild learning disorders compared to clear Intellectual Disability patients, but still significant compared to the unaffected population) are probably much higher in the human population than these other stronger neurological conditions just because they are much less severe and probably pay their dues to the evolutionary system with the higher desire for mating that they seem to induce.
So, the black IQ in America may be depressed by either excessive mutations that the black women are handling better than the males through the gender protective mechanism or it could be depressed by environmental conditions, as Sowell suggests, which the women are handling better than the males through a similar (or the same) protective mechanism; or both.
Using this relative black female IQ as a barometer, we can check what happens to blacks when the environment is radically improved for them. I believe that the gender gap will still remain (since I think that the low black American IQ ultimately has a recent mutational cause that affects women less, phenotypically); Sowell believes that the gender gap will disappear (or even reverse), as it reportedly did with American Jews and others when their conditions changed, since he believes the cause is purely environmental (specifically cultural in the case of modern blacks).
Sowell (2013) claims this empirical victory in Intellectuals and Race (page 79):
Further evidence that the male-female difference in IQs among blacks is cultural is that black orphans raised by white families show no such female superiority in IQs, in addition to both sexes having higher average IQs than other black children.
Now, I was skeptical when I read this because I have other strong empirical reasons to believe that changing the environment should not make the Sowell Effect among blacks disappear. Arthur Jensen’s observation, confirmed by anthropologist John Ogbu and the Journal of Blacks in Higher Education and others, that children of blacks do not do very well even when they come from good high income black families, suggests a likelihood that the Sowell Effect does not disappear even when the environment changes. I doubted that it would make a difference if they were adopted by whites.
I therefore decided to double-check Sowell’s cited source for his claim of its disappearance. It was a paper by Sandra Scarr (1976), who was involved in the famous Minnesotta Transracial Adoption study. The cited page indeed confirms that Sowell had made a small reporting mistake (which is extremely rare for the hyper-meticulous professor). The results he references on that page do not disaggregate gender differences by race, so we can not know for certain that the black gender gap was closed (or reversed) as he asserts.*
There are other studies that could possibly back Sowell up if he is right and we should check those too. For example, there is the well-known Eyferth Study in Germany which monitored the IQs of illegitimate children of black and white American soldiers who were stationed there at the end of the Second World War. These children should have no gender gap between the black males and black females in IQ since they are brought up very far from the black ghetto cultural influence. In fact, Sowell himself has several times cited this study as a cultural control experiment.
I looked at the Eyferth Study data at Wikipedia and indeed it seemed to confirm not only the elevation of black IQ, but even the slight reversing of the black gender gap, as Sowell would predict. There are also many papers published by many scholars that have used this same data, which made it much harder to be skeptical. But it was still worth checking as our theory does not expect this gender correction here since the white women involved in the Eyferth Study were poor whites (the significance of which will become clear later.)
Wikipedia got its data from The g Factor, a book by Arthur Jensen (1998) that is probably the most cited in the racial intelligence debate. I went to the cited page and indeed found that Wikipedia had correctly reported Jensen’s data. The Sowell Effect had apparently disappeared among the black children born in Germany and the strong culture hypothesis seemed to be vindicated.
Our last option was more improbable but worth pursuing: what if Jensen himself is wrong? It’s of course inconceivable that there is a mistake in his book since it has been cited and scrutinized extensively for almost two decades now by friends and foes alike. However, if Sowell can be slightly human, there’s a chance that Jensen might just also turn out to be occasionally human (especially since the Eyferth study was in German, which means he probably had to depend on secondary sources.)
One of Jensen’s sources turns out to be another top psychologist in the field, Professor John Loehlin, whose treatment sounds like he actually analyzed the primary data. Checking Loehlin’s comments elsewhere on the same Eyferth study shows a discrepancy with Jensen, which suggests that Jensen may have indeed slightly mis-reported the data (or perhaps there is some other explanation). Loehlin (2000) sounds like there is still no gender correction for blacks:
Interpretation of this study is complicated by the presence of a race x sex interaction. Among the boys, those with Black fathers averaged below those with White fathers; but among the girls it was the other way around – those with White fathers did worse. Looked at differently, the boys and girls with Black fathers were approximately equal in mean IQ (96 and 97, respectively).
Now, one might argue that a one point IQ advantage is insignificant and we should grant that the gender gap is practically extinguished. In fact, it is far from extinguished, when you get a bit more context. The test itself apparently favored males in general at the time, so the black males were supposed to have scored at least 4 points above the black females, but they scored a point below, which means that the reversed black gender gap was as alive in Germany as it is in America. Let’s hear from the horse’s mouth again (Loehlin, 2000):
The boys and girls with White fathers were markedly unequal (101 and 93, respectively). These last two samples were fairly small (37 boys, 33 girls), and thus perhaps this 8-point difference is a statistical fluke. In some sense, it must be at least partly such. After all, the standardization population of this test consisted of children of German women whose males had genes of European origin! The test does show a difference in IQ that tends to favor boys, but less about 4 IQ points, judging from Eyferth’s graphs. The samples of children with black fathers are larger, 81 boys and 90 girls, and so should be somewhat more stable. But that leaves us with the question of why the offspring of black fathers do not show the sex difference that the population seems to. Back to “more research is necessary.” (My emphasis).
This means that we can also resolve the debate about whether the black soldiers in this experiment were more selected than the white soldiers. It appears that the hereditarians were probably right on this point: the black soldiers had to have been significantly more intelligent than the white soldiers because the presence of a Sowell Effect indicates that the IQ of the black children has received extra depression (through an abnormal lowering of the male IQ, as usual.) However, it’s another Pyrrhic victory for hereditarians: the continued existence of apparent extra depression for black male IQ makes their simple models impotent, just as it does for standard environmentalist models.
This also applies to another favorite study of the environmentalists, Willerman (1974), which is used to argue against any genetic explanation of the black-white IQ gap. The idea is that if the gap was genetic, it would not matter if the mother was white or black. Since Willerman shows a significant difference in the IQs of the mixed race children depending on the race of the mother, the environmental suggestion is that there is something in the nurturing environment provided by the white mothers that causes the IQ difference with the mixed race children brought up by black mothers, thus refuting a genetic hypothesis.
The environmentalists are once again logically right if their argument is against a racial genetic hypothesis. However, they also face a problem because a disaggregation of the IQ scores by gender shows large gaps that they can not explain since the children are growing up in the same environments; the gender of the child should not matter. Thus, racial hereditarians can not explain why the race of the mother matters and environmentalists can not explain why the gender of the child matters.
The only way out for hereditarians has been to dismiss the whole experiment on the basis of the extremely low scores of some children. However, the pattern of the children who score too low is predicted from our hypothesis of differential group and gender canalization.
Notice firstly that the male children of black mothers (married or unmarried) have the lowest IQs in the whole sample: the precise scores that caused hereditarians to dismiss the experiment as unreliable. However, if some of the black mothers have some deleterious mutations that are not expressed phenotypically because of the female protective mechanism, these will be expressed in their male children, but the canalization will continue to protect their female children. That’s exactly what we see here.
The (black) male children of the white mothers are not as affected because their mothers are not concealing as many deleterious mental mutations as the black mothers, for historical reasons that we will make clear.
Doesn’t that last point slightly contradict the Eyferth experiment where we claimed that the black male disadvantage showed up even though the mothers were white? Not necessarily, because the mothers in the Eyferth are said to have been poor, which increases the chances that they came from families with the relatively strong IQ-lowering mutations. These mutations were not expressed in their white children, including males, because those highly selected American soldiers raised the offspring genotype sufficiently high to make the racial canalization effective against these mutations (canalization works across a range of genotypes). For their black male children, on the other hand, the canalization (from race or gender) is largely absent.
IN THE BEGINNING
The question of how races and genders became differently canalized is not crucial for our hypothesis; more relevant is the process of how black Americans received these deleterious mutations that apparently differentiates them from Africans. But we will tentatively speculate on both questions.
The evolutionary origin of differential racial canalization is probably related to the Out of Africa theory of racial origins. The severe environmental challenges faced by the Out of Africa migrants may have ultimately produced strong protective mechanisms through natural selection, racially differentiated by the degree and severity of such challenges, which may have also led to phenotypic resistance (buffering) against many deleterious mutations. Whatever the process, the East Asians, who faced the strongest levels of environmental and mutational challenge, became the most canalized against these forces, and the sub-Saharan Africans remained the least protected.
The same explanation may go for differential gender canalization. The female gender apparently developed higher canalization even before the Out of Africa migrations, perhaps because they were much more vulnerable than men to certain strong environmental challenges, as intuitively expected. Their protective mechanism apparently also gave them protection against some deleterious mutations.
There is only one problem with all this. If the Africans had no need for strong racial canalization because they organically faced much fewer deleterious mutations, where did these mutations that cognitively affect native black Americans so strongly come from? That’s the more crucial question.
This is where we explain the differential IQs of the black groups themselves: why modern black immigrants to the West show a relatively high IQ, especially when they come from Africa. Why are they apparently coming with less cognitively deleterious mutations?
REDNECK CULTURE OR REDNECK GENES?
Thomas Sowell’s incredible intuition offers us another clue, although we will again need to slightly fix his idea after stealing it. In Black Rednecks and White Liberals, Sowell (2006) theorizes that the modern ghetto culture of black Americans came from their association with white rednecks during the time of slavery and he believes it is the preservation of this detrimental culture – preserved with the intellectual help of “white liberals” – that keeps the black IQ low due to its anti-educational, anti-intellectual disposition. Sowell convincingly demonstrates some very uncanny similarities between ghetto black culture today and some aspects of white redneck culture that was more dominant in the South in the past than it is today, as more and more whites have decided to abandon it.
Although I agree that the case for a cultural transfer from some groups of Southern whites is very strong, I think it is more likely that this “culture” was actually passed to blacks genetically rather than through mere influence and imitation. If that is the case, then it was in fact the presence of relatively strong mutations in that sub-population of whites that was affecting the stranger aspects of their behavior and intelligence, and they passed on the same genetic condition to blacks through mating with the black women.
The best argument against Sowell’s cultural influence actually comes from his own analysis. When he is arguing against scholars who claim that slavery is to blame for current black problems like the breakdown of the black family, he gives statistical trends that show that for a long time before the middle of the twentieth century, blacks in fact had more stable families and even had less out-of-wedlock children than whites. He uses this to show that if slavery was the root of these problems, they could have started much earlier.
He is right, of course, but the same argument can be made against his black redneck cultural transfer theory. If the whites that the blacks lived with in slavery are the ones who influenced those negative aspects of their culture, then it should have been more evident in an earlier time instead of showing up much later in time.
Some of Sowell’s strongest critics on this theory also suffer from the same progressional problem. Scholar and investigative journalist, Steve Sailer, for example, argued that much of the negative behavioral tendencies in black ghetto culture must have come with them from Africa. His theory is also unlikely to be true if the statistics about marriage and out-of-wedlock births etc are true. If their culture came with them from Africa they would not have had a long period where that culture seems to have been almost absent only to forcefully show up much later, in generations that had the least connection to or memory of Africa.
Our theory does not suffer from the same time gap problem; it predicts it. If this negative culture was transmitted genetically through mutations inherited from some affected white males who had mated with black women, then there has to be a period when such affected “genes” were not high enough in the population to cause the impact that they would later cause. Our model is consistent with an increasingly degenerative culture caused by a genetic condition that was once minimal but rose quickly in the population due to the hyper-sexuality that it also apparently induces in some of those suffering the condition at its strongest levels (as also seen in Irish Travellers, for example). This is compounded by the fact that its last victims (blacks) were the least canalized racial group in the world. Its spread among whites was further constrained by the fact that they already had a socially stratified society that consciously avoided mating with the lower class whites; after all, if one could rise from poverty, it was less likely that they carried such relatively strong mutations.
Sowell’s mode of cultural transfer also fails to explain why blacks would have just started imitating people that they had recognized to be morally debased, and whom they are said to have harshly branded as “white trash.” If the culture was really transferred from whites, then there was a point in time when the black slaves had a better culture; it is difficult to see why they would copy a culture that was obviously despicable to them. Historian F.N. Boney (1984) actually explained that it was not even all poor whites who were referred to as “white trash,” let alone all rednecks; it was just a small but conspicuously notorious and hyperactive subsection of the poor white segment – which was also a subdivision of the wider rural redneck community (pp.38,39):
At the top of the agricultural scale, a few [redneck] agrarians held sufficient land, slaves, and other resources to become “aristocrats.”… A larger group achieved more moderate success with hundreds of acres of land and some slaves – perhaps as many as twenty or thirty or even more…
At the other end of the redneck spectrum, a few fell into “poor white” and “poor white trash” categories. These two groups were far from the same; indeed, the unwary observer could get into real trouble by not knowing the difference between these tough, sensitive people at the bottom of the white heap. Poor whites simply lacked economic resources, but they lived in a land of great white mobility and might surge ahead at any time. On the other hand, poor white trash lacked more than just money…[they] were looked down upon by blacks and whites alike and considered hopelessly deficient in character as well as resources. Many ridiculed and denounced them, but seldom to their faces, for their pride had distilled into meanness and they could be very dangerous people to cross.
One could strongly speculate that the label given to them by blacks (“white trash”), which sounds extremely emotive, may have been earned by the fact that this small segment were probably raping young black women (thus injecting a genetic condition into the black community that would eventually manifest in similar behavioral and learning difficulties on a wider scale.)
When the eugenicists were trying to rid society of such elements in white society because they believed they were incorrigibly (and genetically) immoral, they accused them of not only mating with their own daughters, but also uncontrollably mating with blacks (Wray, 2013). Although the eugenicists were probably wrong to force sterilizations on many such whites in America (through a Supreme Court ruling), they may end up being proven to be right on the specific expectation that their condition could genetically spread to wider society through their offspring.
Our theory thus explains a paradox that is difficult to explain by present environmental or hereditarian models: when blacks from Africa, the Caribbean and the US are compared, it is the least white-admixed black group that apparently performs best (the Africans), followed by Caribbean blacks who are in between; the most white-admixed group, the native black Americans, do worst. And yet within these communities, it is not necessarily true that the more white-admixed individuals perform worse; they may actually be over-represented on the highest levels of academic or social performance. This is because on the inter-black group level, white admixture may be a good indicator of how much association between blacks and (poor) whites has historically occurred, leading to transfer of deleterious mutations. But within these societies, some of the lighter blacks may perform better in general because they descend from those “mulattoes” who were a mixture of higher class slave masters and their slaves, and who ensured that they kept their complexion and culture higher through self selection, as recounted in books like “Our Kind of People” (Graham, 2000), thus largely avoiding mating with the larger community of blacks who had received the deleterious white mutations. This separation has however not been perfect even among them because they have sometimes married educated black women who did not show the behavioral phenotype that they were avoiding, but who may have still carried the genotype that would be passed to some of their male children.
Caribbean blacks in the UK also clearly show the Sowell Effect even among adults, as can be seen in surveys that have investigated educational levels of parents of school children. Observe that it is only the Caribbean blacks who clearly have more unqualified fathers than mothers in this UK survey:
Caribbean blacks in the UK may have in fact increased this deleterious mutational load further in modern times by willingly mating with the lower classes of British whites, who are more likely to still carry such relatively strong mutations.
The evidence of such deleterious mutations still existing among modern day poor whites can be seen, not just from their low intellectual performance (going even lower than poor Caribbean boys), but even from their violent reactions against their fellow well-performing students, a culture that is also seen among ghetto black Americans, which is further evidence of a mutational rather than an imitational cause.
The advantages of higher canalization are obvious. However, strong canalization also has a slight but conspicuous downside, especially in conditions where the need for such protection is lower. This is because it logically comes with a limitation on what is called “phenotypic plasticity.” The higher protection of women that (fortunately) prevents too many of them from falling off the bottom end into extreme dullness also (unfortunately) “protects” them at the top: not too many can “fall” into the highest levels of (phenotypic) mental ability either, even if they have the genotype for it. Thus, the paucity of female “geniuses” has less to do with inferior innate ability than it does with the fact that they are basically paying for their gender’s genetic protection from extreme dullness. They are in a protective “canal” that is difficult to escape either at the bottom end or the top end, by its restriction on phenotypic variability.
Similarly, East Asians are protected against all kinds of environmental and mutational “attacks” that would normally destroy other people, intellectually, but this special blessing means that they will also tend to be under-represented among the most original human intellects. This solves one of the stronger challenges raised against the Unzian Asian Exception conjecture, asking why it was not East Asians who produced the greatest epochs of human intellectual achievements in history if it is true that their average IQs have consistently been stubbornly high for most of modern human history. It would be because the same canalization that protected them from low intelligence also “protected” them from producing the numbers of super-creative intellects that would be required for such revolutionary achievements in a concentrated period of time. They have a small creative smart fraction, in short.
This differential canalization theory is certainly more plausible than the existing models given by many hereditarians to explain why East Asians have a high average IQ. For example, the idea that East Asians were selected for intellectual novelty because they faced a very challenging environment has one obvious problem that somehow escapes the analyses of most hereditarians: if they were really selected for their ability to find novel solutions to problems, that should probably be the characteristic that distinguishes them the most even today. And yet the same hereditarians admit the conspicuous paucity of highly significant originators and innovators among East Asians, despite showing over-representation in high intellectual aptitude, sometimes very precociously so. East Asian women, who have the highest canalization coming from gender and race, are the most exemplary of this contrast. The shortage of such super-creative phenotypes can not be because they lack the numbers of people with the right genotype, but because the genotype is “buffered” from phenotypic expression by canalization.
Ashkenazi Jews, on the other hand, may be the most over-represented at the top of creative achievements in different intellectual fields (from chess to physics to literature, etc) simply because they happen to also be quite lowly canalized. Although there are so many theories that attempt to explain Ashkenazi Jewish achievement, it may come down to just two factors: high assimilation into Western culture (of quality secular education), combined with a level of canalization that is lower than their fellow Europeans (as descendants of the Middle East – which is why similarly grand achievements have been accomplished by Persians in periods where they were also more open to secular ideas). This explains why Ashkenazi Jews have sometimes had very low IQs (whenever they were too poor to be highly educated or were in environments with little such opportunities), something that hardly ever affects the East Asians.
Our model attributes it to the relatively low canalization of Jews, at least environmentally. Lower canalization also means that their improvement will be more rapid when such environmental conditions positively change (as can also be seen among recent black African immigrants, whose radical improvements begin even in children who were born under bad conditions in Africa, thus defying all kinds of hereditarian limitations.) Had the Ashkenazi Jews not strictly avoided mating with the poorest Europeans that they lived among in some parts of Europe, they would have inherited deleterious mutations that could have probably affected them as much as they have affected black Americans, and their creative over-representation would have also been limited to those highly abstract fields that can somehow escape the necessity of formal education, like music and stand-up comedy – where male American Jews are only out-over-represented by male black Americans at the top.
FALSE ASSORTATIVE MATING
Finally, one of the strongest pieces of evidence used by hereditarians for a racial genetic explanation for intellectual differences between American blacks and whites is the differential regression to the mean. Elite black American parents have children who are not as smart as the children of equally elite white parents, on average, which seems to indicate that the former come from a population with lower genetic intelligence. Our model explains this phenomenon quite easily without suffering explanatory weakness for the Sowell Effect: an elite black man is relatively free of deleterious mutations that would have strongly depressed his phenotypic IQ. However, his equally elite black wife may not necessarily be free of these mutations (he might have to look at her brothers for a clue). She is only protected (or buffered) against their phenotypic expression. The truth will be seen in the average intellectual depression of their male children, which will go way below what is predicted by the phenotypic IQs of the parents. (For the same reasons, very smart black children will statistically have larger IQ gaps with their siblings compared to the smart white children gap with their siblings).
In short, there is basically false assortative mating among black elites on average. This also explains why the mixed black male children have lower IQ when their mother is black than when their mother is white, as we demonstrated above. It also explains why the gap gets worse at the higher levels of parental IQ since this is where such false selection would be most significant. (On the more severe level, The Economist article gives a similar explanation for why strong autism is normally genetically inherited from mothers rather than fathers in all races: had the mothers showed the phenotype of the level of deleterious mutations they carry, they would have likely not been married, which is the case for men with the expressed phenotype. This also means that they actually got married to men who would have otherwise not married them, but only did so through this process of false selection).
Our theory is therefore quite easily falsifiable: an achievement comparison of only the female children of high SES native black American families (especially ones who have no elder brothers to influence their behavior) with children of similarly elite white couples (with or without their boys excluded) should yield results that would be inconsistent with the present predictions of racial hereditarians. Similarly, elite black African immigrant couples’ children, even with boys included, should not show the steeper regression to the mean that has been observed with black Americans relative to whites; it’s quite possible that the Regression gap might in fact be reversed there.
This obviously would not mean that the usual theories of environmentalists are correct either, since it should also not make a difference to them if the boys are included or excluded from the black American samples, especially in elite families. However, as we have faithfully acknowledged, both environmentalists and hereditarians also have some empirically confirmed arguments. Our present hypothesis, taking account of differential gender and racial canalization in human populations, can hopefully help to unify the valid aspects of the environmental and hereditarian frameworks.
NOTE: *Even if we assumed that the gender gap is closed, given the large component of the black/mixed group in the adopted sample, the other problem is that in the black/mixed sample, almost all the mixed children have white mothers (Loehlin, 2000), which means that the mixed boys will not necessarily have depressed IQs, as seen in the Willerman (1974) data, but the black boys from the black mothers should still have a significantly depressed IQ.
Chanda Chisala, originally from Zambia, has been a John S. Knight Visiting Fellow at Stanford University, a Visiting Fellow at the Hoover Institution, and a Reagan-Fascell Fellowship at the National Endowment for Democracy.
 Boney, F. N. (1984). Southerners all. Mercer University Press.
 Graham, L. O. (2000). Our Kind of People: Inside America’s Black Upper Class. New York: Harper Collins.
 Jacquemont, S., Coe, B. P., Hersch, M., Duyzend, M. H., Krumm, N., Bergmann, S., & Eichler, E. E. (2014). A higher mutational burden in females supports a “female protective model” in neurodevelopmental disorders.The American Journal of Human Genetics, 94(3), 415-425.
 Jensen, A.R. (1998). The g factor: The Science on Mental Ability. Westport, CT: Praeger, p. 482
 Loehlin, J. (2000) Group Differences in Intelligence. Handbook of Intelligence. Edited by Robert J. Sternberg. P188.
 Scarr, S., & Weinberg, R. A. (1976). IQ test performance of Black children adopted by White families.American Psychologist,31(10), p.731
 Sowell, T. (2006). Black rednecks and white liberals. Encounter Books.
 Sowell, T. (2013). Intellectuals and Race. Basic Books, p. 79
 Unz, R. (2012). The East Asian Exception to Socio-Economic IQ Influences. The American Conservative.
 Willerman, L., Naylor, A. F., & Myrianthopoulos, N. C. (1974). Intellectual development of children from interracial matings: Performance in infancy and at 4 years. Behavioral Genetics, 4, 84–88.