IQ is extremely important, more so than any other psychometric that has been constructed. The easiest way to explain why this is, is to simply look at the predictive ability of IQ scores. A meta analysis of longitudinal studies by Tarmo Strenze in 2006 with sample sizes of over 90,000 people found that IQ is the most powerful known predictor of education level, occupation level, and income level (1). The Black White IQ Gap (BWIG) is about a standard deviation, or 15 points, and has been detectable since at least the 1960’s. Despite a few studies in the 2000s that claimed the gap was shrinking, a 2012 meta-analysis conclusively found no such evidence for the claim. In other words, the BWIG is 15 points large and has remained stable for at least two generations, despite society becoming more supportive of blacks through desegregation and affirmative action (2).

This gap in IQ between the races explains most of the racial performance gaps (3 p320-340). After controlling for IQ, blacks are more likely than whites to graduate from college and to attain a high IQ occupation, or in other words, a good job. Blacks, after controlling for IQ, make just as much money as whites, and are only 5% more likely to be in poverty than whites, whereas before controlling for IQ blacks are 20% more likely to be in poverty than whites. In essence, the BWIG is the immediate source of a great deal of performance difference between the races. This means that discovering the ultimate source of the BWIG can basically determine the major source of these performance differences. Heritability is the proportion of the variance in IQ of a population that can be explained by the variance in the genetics of that population. Using quantitative genetic practices, heritability can only be directly measured within a population, which means it is impossible to directly measure the heritability of the BWIG, except through using molecular genetic techniques. Despite the lack of data in molecular genetics, decades of quantitative genetics research has provided more than enough data to allow for an inference of the heritability of the BWIG.

Studies on the heritability of IQ have been performed incessantly for at least 60 years to the point where 100 or more papers could be cited for just one sentence which gives an average heritability estimate from those studies. The heritability of IQ within first world nations is probably one of the best replicated findings of the science of psychology and of behavioral genetics. And statistical power tends to be enormous - as exemplified by one 2009 study that found, with a sample of 11,000 twin pairs from the USA, Australia, the UK, and the Netherlands, that the heritability of IQ is about 66% by the age of 17 (4).

Before the age of 17, heritability is measured to be lower, but this does not mean that the IQ of a person is more malleable when they are a child. Evidence indicates, for example, that intervention programs and other such tactics designed to raise the IQ of children tend to produce an effect in childhood that wears off by adulthood (7). This is likely due to the decreased complexity and g-loading of childhood IQ tests, meaning no real increase in intelligence is occuring, even temporarily. Furthermore, child IQ data is generally less reliable (8), which is why heritability estimates using children tend to be lower (since variance due to measurement error is defined as nonshared variance). Due to the decreased reliability of childhood IQ data, siblings correlate to a lesser extent, yielding a lower heritability estimate. The reasons for the lower reliability of childhood IQ data can range from developmental differences among children to test flaws that result from the constraints of measuring a still-developing trait. And the fact that the trait is still developing does not explain the lower heritability estimates from studies with children, despite sounding like it should, because heritability estimates using adults already capture the total variance in development-influencing factors.

Other studies have found that what is not heritable tends to be a function of unshared environment, as opposed to shared environment, which by definition includes factors that are not shared between children of the same family (5). Research on the relevant factors that compose unshared environment for IQ has yielded results that point to the differential treatment of children by parents (6), which itself can have a measure of heritability, meaning that the true heritability of IQ is likely higher than measured, at least from some study methods. Additionally, heritability estimated from identical twins reared apart tends to be in the 75% to 85% range, largely due to the fact that it is broad sense, meaning it includes similarities in IQ due to similarities in non-additive genetic effects, while heritability estimated using identical and fraternal twin pairs, as in the previously mentioned study, is narrow-sense, meaning it filters out most similarity due to non-additive effects due to the algebra used to estimate it and the fact that fraternal twins only share, on average, only a fourth of nonadditive effects, effects which can be due to gene-gene interaction or dominance effects. The “true” heritability of IQ in first world countries tends to be better estimated at about 80% than 66%, then, as estimates in the 50% to 70% range tend to be only narrow sense heritability, which does not give as clear of a view as broad sense heritability on the extent to which differences in IQ are dictated by genetic differences.

Regardless, the point of discussing within group heritability measures is not to just flatly transfer those estimates to the BWIG. Rather, it is simply to establish that it is possible, and probable, for genes to have a large say in determining IQ both for individuals and for populations. For these heritability estimates cannot reveal the extent to which blacks and whites have different IQ relevant environments. It is this question whose answer determines the heritability of the BWIG, regardless of the actual within group heritabilities. If blacks and whites have, on average, the same IQ relevant environments, then the heritability of the BWIG is 100% because the BWIG could not be explained by anything other than genes in this case. From there, the greater the difference in their IQ relevant environments, the lower the heritability of the BWIG can be inferred to be, in general. Of course, the exact heritability of the BWIG is proportional to the magnitude of the IQ-relevant genetic differences between the races weighed against the magnitude of the IQ-relevant environmental differences. Since the magnitude of the genetic differences cannot be directly measured except with molecular genetics techniques, it will be the most fruitful to assess the IQ-relevant environmental differences. The previously mentioned heritability estimates show that differences in genes between people can explain large portions of the differences in IQ between people, no matter how large those portions are exactly. Applying this knowledge to the BWIG gives the impression, that if the two races are in fact not just skin color and politics, but rather semi-discrete groups which evolved under separate selective pressures, then it is for all intents and purposes a certainty that some fraction of the BWIG is explained by genetics, depending on the extent of IQ relevant environmental differences between the races.

The evidence, of course, shows this to be the case. Computers can classify ancestry groups by DNA and match self-reported race with 99% accuracy (9). A 2014 study graphed racial clusters as identified by a computer sequencing DNA and found that self reported race groups semi-discreetly. Furthermore, US Blacks were found to be the racial group with the most distance from the others, including whites. Both this data and evolutionary theory, then, predict that there is some measurable degree of difference between the frequencies of IQ relevant genes in whites and blacks. This, however, is not necessarily sufficient evidence to infer the extent to which genes are responsible for the BWIG. Examining the differences between black and white IQ relevant environments can, however, yield a good estimate. Quite frankly the null hypothesis here is that the BWIG is largely genetic, considering the fact that these are two genetically distinguishable groups living in the same first world country. It therefore comes down to the failure of every environmental hypothesis to provide evidence for the heritability of the BWIG being high. These failures encompass such broad environmental factors that in essence there is nothing reasonable for the environmentalist to fall back on after demonstrating these failures.

In general, there are three waves of environmental factors that environmentalists like the point to: test bias, physical factors, and psychological factors. The first one contends that IQ tests are not measuring a real difference in g between blacks and whites; rather, IQ tests are culturally biased, and the BWIG can be explained by differences in culture. This has been readily disprovable since the mid- 20th century, when if anything tests were surely more biased against black people than today. In 1986, it was found that the difference between black and white scores on questions independently deemed to be culturally biased was actually less than the black and white score differences on less biased test questions (11), leading the authors of the study to conclude that the BWIG likely reflects real differences in g, the reasoning ability that IQ aims to test. Don’t make the mistake of thinking that this finding rests on one study - an entire book on the topic was published in 1980 titled Bias in Mental Testing which argued for the same conclusion as the 1986 study.

Next, environmentalists, after being forced to accept that there is no evidence for cultural bias on IQ tests, like to hold that blacks are disadvantaged when it comes to various environmental factors, generally because of systemic racism (for if it were their fault, then the supposed environmental differences would be as good as genetic). The first line of factors tend to be the physical ones - the narrative goes that blacks lack intergenerational wealth, money for proper nutrition, and good schools, all of which depress IQ scores. There are, of course, multiple ways to attack this reasoning, including attacking the notion that blacks differ in factors like nutrition, but the simplest is to just demonstrate that, when controlled for, these factors do not explain the BWIG.

The most powerful piece of evidence against this environmentalist hypothesis is the Minnesota Transracial Adoption Study, which tracked blacks, 50/50 black and white mixes, and white adoptees who were all adopted by upper middle class white parents from an age young enough to where the correlation between IQ and age of adoption was .13 at the age of 17. Unless there exists systemic racism so pervasive as to make adopters of black orphans give them a significantly impoverished environment due to their skin color, this study effectively controls for socio-economic status, nutrition, school resources and quality, and a whole host of other factors that could be hypothesized like a generally IQ depressing black culture. This study basically found a BWIG between adoptees equal to the general societal BWIG (10). It also found a gap between whites and mixes about half that size, which is effective evidence against the systemic racism narrative (Drake and Obama are “black” ) and excellent evidence for the genetic hypothesis.

Furthermore, when the data is corrected for the Flynn effect, the gaps slightly widen (12). Now the Flynn effect is sometimes babbled about by people who don’t know what they’re talking about as evidence against the genetic hypothesis. The Flynn effect is simply the phenomenon whereby IQ test scores tend to nationally rise by about 3 points each decade since the 1920s. They say that this simple phenomenon means IQ could not possibly “be genetic”. This is wrong on multiple levels, but the most simple way to refute this is to show that the Flynn effect is hollow for g, meaning that it does not reflect a true intelligence increase. Researchers have found this to be the case, with the correlation between Flynn increases and g loadings of different tests ranging from -.40 to -1.00 (14). Researchers have also found that renorming tests, meaning to make their averages 100 again, does not interfere with their predictive validity. And even if the Flynn effect is not completely hollow for g, it would still need to be demonstrated that it impacts blacks more than whites for it to be used as evidence against the genetic hypothesis, something which has not been done considering that the BWIG has not narrowed in 50 years.

More strong evidence against the environmentalist hypothesis is found when looking at the IQ gap between blacks and whites that have the same amount of some environmental factor, such as income or school quality. In the case of income, the gap never significantly changes at any income level and actually slightly rises as the blacks and whites being compared get richer (3 p288). This data likely controls for more factors than income, considering rich black children are probably not malnourished, nor do they likely go to poor schools. In regards to schooling, there is good evidence that, in general, school quality does not affect IQ, meaning that differences between black and white school funding or quality could not be responsible, in any significant way, for the BWIG. As previously mentioned, compensatory programs meant to raise IQ have, time and time again, been shown to not work (7). Even more damning are the well replicated results of voucher studies, which demonstrate longitudinally that school quality does not produce changes in school performance, which is much less heritable than IQ (15).

As for the spectre of slavery and segregation, all that needs to be shown is that black IQs around the globe tend to be depressed. Unlike the Irish, whose IQs were depressed and whose IQs have risen as Ireland has urbanized, and whose IQs rose in the USA when they moved to the cities, urbanized blacks still feature depressed IQs, no matter the country. In Africa, the average IQ tends to be about 80. In South Africa, the same racial IQ gap exists as in the United States, as well as in the UK, a country with no history of segregation or slavery (18).

More money and better schools fail to narrow the BWIG. Once this has become evident to an environmentalist, the argument tends to shift to Freud-like, hard to prove mental differences which are a supposed result of environment. People, to avoid being racist, are supposed to believe that “stereotype threat” explains a significant portion of the 15 point IQ gap between the races. Not only is stereotype threat a priming effect, there is no evidence that it explains racial gaps on actual IQ tests (16). Black culture is another whimsical psychological effect that is hypothesized by some environmentalists to make blacks less motivated to do well on IQ tests, but motivation measuring items are the only items blacks tend to do better on than whites (8 p111-113). Finally, the differences between mixed race blacks and full-blooded blacks, as reported in the Minnesota Transracial Adoption Study, greatly harms the credulity of this narrative.

The systemic racism narrative in general is discredited by previously shown evidence that shows that, once IQ is controlled for, most racial performance gaps disappear. This means that any real systemic racism in play must practically only impact IQ relevant environmental variables, and not directly act on income or graduation rates. Blacks must be treated so poorly by whites, that after controlling for income, nutrition, and school quality, blacks experience racism to such a high degree as to still depress their average IQ by a full standard deviation. And in every country on Earth! If systemic racism really is this severe and this unconsciously pervasive, in a country where being black is worth 230 SAT points (19), its supposed existence just ends up supporting the right-wing project of separating the races almost as much as a high heritability of the BWIG does. Aside from evidence that serves to negate environmentalism, there are a few lines of evidence that seems to imply a genetic difference being responsible for most of the BWIG, given that there is no large, detectable IQ relevant environmental disparity between the races. First, sibling regression toward the mean: a black sibling of somebody with a 120 IQ will regress toward the black mean and have an IQ of about 100, while a white sibling will regress toward the white mean and have an IQ of 110 (8 p118). This is extremely hard to explain without differential nonadditive genetic effects, given the fragility of the evidence for there being large IQ relevant environmental disparities between the races. At the least, this phenomenon shows either an extremely pervasive depressed environment or a depressed pool of additive genetic effects for the black population. Given that the BWIG persists when controlling for various hypothesized environmental variables that could be depressing the black IQ, the depressed pool of additive genetic effects is far more likely, which means that it is likely that the BWIG is mostly explained by genetics. Second, blacks have a smaller standard deviation for IQ than whites (8 p212). If blacks and whites have the same or a very similar pool of IQ genetics, then black heritability will be higher than white heritability for IQ because black variance due to the environment must be smaller than that of whites. This is not the case (8 p212), suggesting that the black IQ gene pool is dissimilar to the white IQ gene pool by a significant degree, more directly than simple evolutionary history can.

Combined, the evidence clearly points to there being a strong genetic factor in the BWIG. It can be inferred that this effect explains almost all BWIG, considering the data on mixed race adoptees from the Minnesota Transracial Adoption Study. This in line with the complete lack of evidence for a significant environmental factor provides reasoning that the BWIG is almost completely heritable. And it is a virtual certainty that the BWIG is at least 50% explained by genetics, considering the evidence. At this point, the only evidence that can cast doubt on this position is some evidence that demonstrates that an environmental factor, when controlled for, reduces a significant portion of the BWIG.

SOURCES

https://emilkirkegaard.dk/en/wp-content/uploads/Intelligence-and-socioeconomic-success-A-meta-analytic-review-of-longitudinal-research.pdf

https://psycnet.apa.org/record/2012-24333-012 ; http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.843.2060&rep=rep1&type=pdf

https://libgen.is/book/index.php?md5=3BDA707E5BD060F627628CD810B48A9A The Bell Curve

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2889158/

https://web.archive.org/web/20161114181852/http://isites.harvard.edu/fs/docs/icb.topic1169974.files/Bouchard%20McGue%202003.pdf ; https://psycnet.apa.org/record/1993-98367-003

Behavioral Genetics 6th Edition https://libgen.is/book/index.php?md5=9303B823B96B155DDC249D9267EAD8DC

https://hepgjournals.org/doi/10.17763/haer.39.1.l3u15956627424k7; https://www.sciencedirect.com/science/article/pii/S016028961500135X

Educability and Group Differences - ctrl f “simplex” https://emilkirkegaard.dk/arthurjensen/wp-content/uploads/2014/06/Educability-and-Group-Differences-1973-by-Arthur-Robert-Jensen.pdf

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3951706/ ; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1180234/

http://faktasiden.no/dokumenter/minnesota-transracial-adoption-study.pdf ; https://en.wikipedia.org/wiki/Minnesota_Transracial_Adoption_Study (7/23/19)

http://arthurjensen.net/wp-content/uploads/2014/06/Black-White-Bias-in-Cultural-and-Noncultural-Test-Items-1987-by-Arthur-Robert-Jensen-Frank-C.-J.-McGurk.pdf

https://libgen.is/book/index.php?md5=7173C0103012053229366179E9546CD3 p185

http://www.triplenine.org/portals/0/PublicDocs/Vidya/vidya_282-283_BW_reprint.pdf p16-19

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.694.8353&rep=rep1&type=pdf

http://legis.wisconsin.gov/lab/reports/12-14full.pdf ; https://ies.ed.gov/ncee/pubs/20104018/pdf/20104018.pdf

https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1559-1816.2004.tb02564.x ; https://web.archive.org/web/20161114181944/http://isites.harvard.edu/fs/docs/icb.topic630262.files/sthreat_naep.pdf

https://www.news-medical.net/news/2005/04/26/Race-differences-in-average-IQ-are-largely-genetic.aspx - Unused source: Lynn’s arguments

https://www.researchgate.net/publication/222062631_The_Average_IQ_of_Sub-Saharan_Africans_Comments_on_Wicherts_Dolan_and_Van_der_Maas ; http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.503.9118&rep=rep1&type=pdf ; https://www.intelligence-humaine.com/wp-content/uploads/2019/03/Race-Differences-in-Intelligence-second-edition-2015-1.pdf p63 table 4.7

https://www.princeton.edu/~tje/files/files/webAdmission%20Preferences%20Espenshade%20Chung%20Walling%20Dec%202004.pdf