The races differ in intelligence, and it’s poorly explained by environmental factors
Dedicated to Lindsay, for helping me find myself
Since the inception of the Boas Cult (refer to MacDonald 2002) and the denial of race and race differences after World War 2, any comment that refers to race differences in a plethora of things are denounced as scientific racism and as being discredited without any examples how (e.g. Burmila 2008; Internationalist Socialist Review; Wolf 2019). Within the camp of race differences in intelligence, there are two sides to the debate: the hereditarians, people who believe that genetics are the primary influence on human behavior, intelligence and other characteristics, and environmentalists, people who believe that the environment is the primary influence. The hereditarians are often accused of being racist, spreading pseudoscience, and promoting a white supremacists agenda.
For example, a review from the site Amazon on Michael Levin’s book Why Race Matters accused Levin, a jew, of being a “Jew who fights in the same trenches of white supremacists.” Another hereditarian, J.P. Rushton, had his book Race, Evolution and Behavior criticized by having people call it works of “pseudoscience” and having his motives attacked as “evil” (Gottfredson 2012). Once a respected psychologist, Rushton was shunned simply for speaking about race differences. The same thing happened to Arthur Jensen and Charles Murray. Hereditarians are accused of implying that whites are superior and other minorities, like blacks, inferior due to race differences in intelligence – which are assessed by IQ tests.
“If it is ever documented conclusively, the genetic inferiority of a race on a trait as important as intelligence will rank with the atomic bomb as the most destructive scientific discovery in human history. The correct conclusion is to withhold judgment.“
In referring to the debate on race as a valid biological concept, Pigliucci (2013) states:
“Of course, anyone who has seriously looked into this endless debate knows very well that here is where the stakes really lie: it is not about small genetic differences that may or may not help build a more individualized medicine; it is not about forensic anthropologists and how well they do their work; it is about claims that one race has superior or inferior intellectual capabilities than other one.“
Either inferiority is implied – or attacks on IQ tests and their creators themselves are done, a tactic done by Stephen J. Gould in his work The Mismeasure of man (Gould 1980). An example of this was done by Harriot (2019):
“Some people say that even the idea of IQ tests are inherently racist because IQ testing derived from the eugenics movement—the idea that it is possible to improve the human race by preventing certain people from breeding. The man who invented the first IQ test, Alfred Binet, even argued that his own tests couldn’t adequately measure intelligence. The Scholastic Aptitude Test (SAT) and the Advanced Placement Exam have similar origins in racism and eugenics.”
I defend the hereditarian position for two main reasons.
(i) Just because someone claims that whites are more intelligent than blacks, this doesn’t mean that they think that intelligence equals superiority. By saying that “whites have higher IQ’s than blacks,” critics automatically associate higher intelligence with racial superiority. It is not the hereditarians who are associating intelligence with superiority, it is the critics who are because they automatically associate higher intelligence with superiority. There is no cosmic scale by which differences in intelligence make one race superior to another. A gap separates the facts from judgement of value.
(ii) The evidence on race differences in intelligence should not be shushed away simply because some denounce it as “scientific racism” (e.g. Evans 2018). In a society where some blacks are asking for reparations from the white majority in the United States (Craven 2019), whites are justified in bringing up race differences that absolve them from guilt. By blacks claiming that they are handicapped due to the historical oppression made by whites, they open up the door for race differences. Wade (2014) put it best in his book A Troublesome Inheritance:
“By referring to anyone who explores the biological basis of race as a ‘scientific racist’ and thus in essence demonizing them as racist, the academic left has managed to suppress almost all discussions of human differentiation.”
An Emergence of Controversy
Since the 18th century, European philosophers and scientist proposed that there may be race differences in intelligence. During the 19th and early 20th century, it was though that race differences in the brain were responsible for the race differences in intelligence.
During this time there were no IQ test, so people like Francis Galton estimated the intelligence of various racial groups from traveler observations, intellectual achievements, and using the percentage of eminent men in their respected racial groups. According to Galton, intelligence was distributed normally in all racial and ethnic groups, but that distribution varied between groups.
The first IQ test, named the Binet-Simon scale, was developed by Alfred Binet and Theodore Simon in 1904. The French Ministry of Education asked these researchers to develop a test that would allow them to distinguish mentally retarded kids from normally intelligent, but lazy kids. Binet and Simon began working on a test that wasn’t specifically about things taught in school, but that also focused on attention, memory, and problem solving skills.
Binet himself, however, believed that IQ tests couldn’t be used to measure a single permanent level of intelligence. He also said that intelligence was far too broad of a concept to express with a single number. He suggested that intelligence changed over time due to a number of factors and could only be compared to people from similar backgrounds.
After Binet’s original IQ test, Stanford psychologist Lewis Terman took Binet’s original IQ test and standardized it with a sample of American participants. This adapted test was published in 1916, and was called the Stanford-Binet Intelligence Scale. This wouldn’t be the first time that a new version of an intelligence test was created.
Building on the Stanford-Binet test, American psychologist David Wechsler created a new IQ test. Wechsler also believed that intelligence also involved different mental abilities. Since he wasn’t satisfied with the Stanford-Binet test, he published his new IQ test in 1955 called the Wechsler Adult Intelligence Scale (WAIS), the Wechsler Intelligence Scale for Children (WISC) and the Wechsler Preschool and Primary Scale of Intelligence (WPPSI).
Lewis Terman in the Stanford-Binet Intelligence Scale handbook noted the higher frequency of “morons” among non-white American groups, and that more research needed to be done on race differences in intelligence (Benjamin 2006).
In the 1920s – 1960s, psychologists started questioning the causes for race differences in IQ. During this time, race differences in intelligence were used to support the idea of racial eugenics. The hereditarian hypothesis also started to change due to eugenicist claims regarding race differences in behavior and morality.
From the 1960s – 1980s, the hereditarian position grew stronger thanks to the works of Arthur Jensen. In 1965, William Shockley made a public statement at the Nobel conference, stating that social programs helping the disadvantaged wouldn’t work, and that the most competent group in the U.S. were from the original European settlers. During this, Shockley also supported eugenics, and claimed that his arguments were backed by statistics. Shockley was supported by the Pioneer Fund, a group dedicated to the study of heritability and eugenics in the human race.
One of Shockley’s lobbying campaigns involved educational psychologist Arthur Jensen, from U.C. Berkeley. Jensen, surprisingly, was an environmentalist – but changed his mind in 1966 – 1967. Jensen stressed the importance of genes for intelligence, and that the black-white performance gap had a genetic component to it in his article “How Much Can We Boost IQ and Scholastic Achievement” (Jensen 1967). Jensen stated that “various lines of evidence, no one of which is definitive alone, but which, viewed all together, make it a not unreasonable hypothesis that genetic factors are strongly implicated in the average Negro-white intelligence difference. The preponderance of the evidence is, in my opinion, less consistent with a strictly environmental hypothesis than with a genetic hypothesis, which, of course, does not exclude the influence of environment or its interaction with genetic factors” (Sesardic 2005).
Thanks to Jensen, an academic interest in the hereditarian viewpoint on race differences in intelligence sprung up once again. In 1971, Richard Herrnstein, an American psychologist and sociologist, wrote an article on The Atlantic discussing IQ differences between social classes rather than races. Much like Jensen, he took a strong hereditarian stance on the issue of intelligence. Although the hereditarians simply tried to counter the environmentalist position, a storm of hate against them ensued.
Jensen and Herrnstein were charged with accusations of racism. Jensen was protested by Berkeley’s Students for a Democratic Society. They staged protest against Jensen at U.C. Berkeley, chanting to “Fight Racism. Fire Jensen!” (Jensen 1972). While teaching at Berkeley, Jensen required bodyguards even though he denied accusations of racism. In 1972, 50 academics – which included Jensen and Herrnstein, signed a statement entitled “Resolution on Scientific Freedom Regarding Human Behavior and Heredity.” The statement criticized the “suppression, punishment and defamation of scientists who emphasized the role of heredity in human behavior” (Badcock 2015).
Since Jensen’s publication of his article, rebuttals were made against his positions; many of these he has responded to (Jensen 1980). After his 1967 article, Jensen became more open on his views on race differences in intelligence. He stated that “something between one-half and three-fourths of the average IQ differences between American Negroes and whites is attributable to genetic factors” (Brody 2013).
Critiques against Jensen, for example from the American Psychological Society, American Anthropological Association and Genetics Society of America, all seem to be to simply label Jensen a racist and for simplifying such a complex issue.
In the 1980s – 2000s, hereditarian research continued. Richard Lynn and J. Philippe Rushton, with help from the Pioneer Fund, released their work on race differences in intelligence. Rushton’s 1994 book Race, Evolution and Behavior sparked controversy. He was under police investigation for complaints of promoting racism and white supremacy, something he denied (Rushton 1994).
In 1994, Richard Herrnstein and Charles Murray reignited the topic on race differences in intelligence with their book The Bell Curve. After its publication, countless articles and books were released in an effort to debunk the book (Gould 1980; 1994; Jacoby and Glauberman 1995; Fischer 1996). The criticisms against the book were mostly stock arguments: correlation doesn’t equal causation, misuse of statistics, and that it wasn’t submitted for peer review (even though books don’t get peer reviewed). Defenses of the Bell Curve and their use of their statistics have also been published (Goodnow 2014; Hu, 2015; Murray 1995; Winegard and Winegard 2017).
Since Herrnstein and Murray’s work, other hereditarian works arguing for a larger genetic influence on IQ have since been released (Levin 1997; Jensen 1998). The debate still goes on, and it seems that the totality of the evidence supports the hereditarian hypothesis.
The Black-White IQ Gap: The Data
When it comes to assessing the intelligence of whites and of Negroes, the main line of evidence comes from researchers who give both groups IQ tests. Different tests are used, such as Wechslers, Stanford-Binet, Raven’s Progressive Matrices, and others. The most thorough research on the literature was done by Shuey and Richard Lynn.
Shuey (1966) in The Testing of Negro Intelligence reported on 382 studies involving 80 different tests administered on hundreds and thousands of black and white children, high school and college students, military personnel, civilian adults, deviates, and criminals. The average black IQ score in these studies were a bit below 85, and the average for whites was also a bit above 100. The average black-white difference was always close to 1 SD.
Lynn (2011) reviewed the overall literature on the black-white IQ gap in the United States. In all the studies Lynn reviewed on American blacks, they averaged an IQ of 85. Whites, on the other hand, averaged an IQ of 100. The following graph has been adopted from Lynn’s book.
Negro Intelligence in the United States
Lynn also looked at the average IQ of whites in the United States.
White Intelligence in the United States
From this, we can see that whites average a higher IQ score than blacks.
Coleman et al. (1966) reached identical conclusions:
“…the Negroes’ averages tend to be about one standard deviation below those of whites.”
The National Academy of Science reported:
“Many studies have shown that members of some minority groups tend to score lower on a variety of commonly used ability tests than do members of the white majority in this country. The much publicized Coleman study provided comparisons of several racial and ethnic groups for a national sample of 3rd, 6th, 9th and 12th grade students on tests of verbal and nonverbal ability, reading comprehension, mathematics achievement, and general information. The largest difference in group averages usually existed between blacks and whites on all tests and at all grade levels. In terms of the distribution of scores for whites, the average score for blacks was roughly one standard deviation below the average for whites. Differences of approximately this magnitude were found for all given tests at 6th, 9th and 12th grades… The roughly one-standard deviation difference in average test scores between blacks and white students in this country found by Coleman et al. is typical of results of other studies” (Garner and Wigdor, 1982)
In order to assess what 85 and 100 means, we can use the metrics provided to see where the black IQ score falls into within the overall intelligence level, respectively.
IQ Score Metric
|< 50 – 79||Very Dull|
Using this metric provided to us in The Bell Curve, we can see that the average white is in the normal range, and that the average black is in the dull range. This means that the intelligence of the Negro is literally dull. Now, the question is: Is the level of black intelligence always dull? The answer to that is an astounding yes as low intelligence among blacks can be seen since the age of 3.
Among 2 year old blacks, they average an IQ of 92 (Kaufman and Kaufman 1983). This is higher than the IQ of 85, but this should be well excepted given the fact that blacks mature faster than whites (Rushton 1994; Lynn 1998). It isn’t until the age of 3 to 4 that their IQ declines to reach 85.
Montie and Fagan (1988) gave 3 year old pre-school children a standardized sample of the Stanford-Binet LM; the achieved an average IQ of 86. So black preschoolers have the same IQs as black adults. Jencks and Philips (1998) found an average IQ of 82 among 3-4 year old blacks. Peoples, Fagan, and Drotar (1995) got an IQ of 85 for infants aged 3.0 to 3.4 years old using the standardized sample of the Stanford-Binet-4.
The same can not be said for whites. By the early age of 3 to 5, white children average an IQ score in the 98+ level, showing that very early on in their life white children are in the normal IQ range (Lynn 2011).
Using cross-cultural studies, race differences in intelligence are fully present in pre-school children outside of the U.S. also. This further verifies the fact that race differences in intelligence are present at an early age as African 3 year olds in Dominica average an IQ of 67 (Lynn, 2011), and black kids at the age 4 in St. Lucia average an IQ of 62 (Murray, 1983).
This means that since very early on, blacks are less intelligent than whites. One has to wonder if the black-white IQ gap ever diminishes as people get older, and the answer is no. Analyzing the PIAAC survey of adult skills with a sample size of 8,700 people, Dalliard (2017) found that there is a 0.98 (14.7 IQ point) standard deviation between blacks and whites from the age of 16-65.
The gaps in IQ shouldn’t be controversial as Roth et al. (2001), which was a large meta-analysis which included more than 6,000,000 individuals, found that blacks score 1 SD lower than whites. Chuck (2013) looked at 100 years of testing done on black intelligence, and every study looked at found lower intelligence among blacks.
One of the few studies to find no race differences in IQ results is Fagan and Holland (2006). They claim that race is not related to the g-factor, and that it’s rather culture differences in the provision of information that account for racial differences in IQ. The issue with this paper is that it attacks a straw man, namely that the black-white IQ gap is due solely to the g-factor. There are also a few methodological errors in their study (refer to Kirkegaard 2007). This study is not a problem for the black-white IQ gap debate.
From all the studies reviewed, there’s a 1 SD (15-point) gap between blacks and whites. Rushton and Jensen (2005) note:
“Currently, the 1.1 standard deviation difference in average IQ between Blacks and Whites in the United States is not in itself a matter of empirical dispute. A meta-analytic review by Roth, Bevier, Bobko, Switzer, and Tyler (2001) showed it also holds for college and university application tests such as the Scholastic Aptitude Test and the Graduate Record Examination, as well as for tests for job applicants in corporate settings and in the military.
Race differences in intelligence are also not new. In fact, contrary to some beliefs, it’s been recognized for a very long time that blacks are less intelligent. The historical view on race differences in IQ seem to have been consistent with modern day evidence.
Pieterse (1992) recorded the agreement of European cultures that blacks are less intelligent, brutal and highly sexed. Lewis (1990) describes how medieval Arab slave traders saw their black slaves as rhythmic, highly sexed, unintelligent, and prone to merriment. Even philosophers like Kant and Voltaire have given their input on the low intelligence of Negroes.
Kant (1764) wrote that
“The Negroes of Africa have received from nature no intelligence that rises above the foolish. Hume invites anyone to quote a single example of a negro who has exhibited talents. He asserts that among the hundred thousands of blacks who have been seduced away from their own countries, although very many of them have been set free, yet not a single one has ever been found that has performed anything great whether in art or science or in any other laudable subject; but among the whites, people constantly rabble and acquire esteem through their superior gifts. The difference between these two races of man is thus a substantial one: it appears to be just as great in respect of the faculties of the mind as in colour.”
Voltaire (1756) described the phenotypic features of Negroes and their intelligence:
“Their round eyes, their flat nose, their lips which are always thick, their differently shaped ears, the wool on their head, the measure even of their intelligence establishes between them and other species of men prodigious differences. If their understanding is not of a different nature from ours, it is at least greatly inferior. They are not capable of any great application of ideas, and seemed formed neither in the advantages nor the abuses of our philosophy.”
In Negroes and Negro Slavery, Van Evrie (1853) wrote:
“The negro mind, in essential respects, is always that of a child, the intelligence, as observed, is more rapidly developed in the negro child, those faculties more immediately connected with sensation, perception, and perhaps memory, are more energetic ; but when they reach twelve and fifteen, they diverge; the reflective faculties in the white are now called into action, the real Caucasian character now opens, the mental forces fairly evolved, while the negro remains stationary, a perpetual child. The negro of forty or fifty has more experience or knowledge, perhaps, as the white man of that age has a more extended knowledge than the man of twenty-five, but the intellectual caliber the actual mental capacity in the former case is no greater than it was at fifteen, when its utmost limits were reached.”
Some criticisms of these findings may emerge after seeing this, especially from informed egalitarians who are aware of the Flynn Effect and critiques against the use of IQ tests to measure intelligence.
Criticisms Against Intelligence Testing
A common tactic by some people is to first ask “what is intelligence?” From my personal experience, this is usually a tactic that leads to the argument that there is no universal agreement on what intelligence actually means – thus IQ tests don’t help us at all. This, of course, is absolutely true. There is no uniform agreement on what the word means, but we can look at what other researchers have defined intelligence – or cognitive ability – as and use that as a proxy.
Rindermann, Becker, and Coyle (2016) surveyed intelligence researchers to see what was the cause of international differences in IQ tests. In their introduction, they defined intelligence as
“Cognitive ability is understood as the ability to think (intelligence), the disposition of knowledge (the store of true and relevant knowledge), and the intelligent use of this knowledge.”
In looking at the heritability of intelligence, Plomin and Stumm (2018) defined intelligence as
“the ability to learn, reason and solve problem.”
The definition given by Plomin and Stumm is similar to the one given by the American Psychological Association in 1996. After the publication of The Bell Curve, the American Psychological Association made a Task Force to talk about the known and unknowns of IQ testing which was headed by Ulric Neisser. Neisser et al. (1996) reported that researchers regard intelligence as efforts to organize a complex set of phenomena, including the
“ability to understand complex ideas, to adapt effectively to the environment, to learn from experience, to engage in various forms of reasoning, to overcome obstacles by taking thought.”
Gottfredson (1994) was signed by 52 intelligence researchers and was also released on the same year as The Bell Curve. According to the paper, intelligence was defined as involving
“the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience. It is not merely book learning, a narrow academic skill, or test-taking smarts. Rather, it reflects a broader and deeper capability for comprehending our surroundings- ‘catching on,’ ‘making sense’ of things, or ‘figuring out’ what to do.”
The most comprehensive data on what defines intelligence, that I could find, comes from Legg and Hutter (2007). They looked at definitions of intelligence from organizations, psychologists, and AI researchers. While they do say that there is no uniform agreement on what intelligence means, the definitions they got could all be summed up:
“Intelligence measures an agent’s ability to achieve goals in a wide range of environments.”
Intelligence can also be seen as something known as the g-factor, something that the races primarily differ in. This is why a column entitled simply as “g” can be seen in the analysis from Lynn.
General intelligence, also known as the g-factor, refers to the existence of a broad mental capacity that influences performance on cognitive ability measures. In 1904, Charles Spearman first described the existence of general intelligence. According to Spearman, g was responsible for someone’s overall performance on IQ test. Spearman noted that while some people could and often excel in certain things, most people who did well in one thing tended to do well in other things, too. A person who did well in a test on verbal intelligence, for example, did well on a test of spatial intelligence.
IQ tests measure cognitive abilities that are thought to make up general intelligence. These include things like:
- Quantitative Reasoning: ability to solve numerical problems
- Fluid Reasoning: flexible thinking to solve problems
- Visual-Spatial Processing: ability to put together puzzles and solve complex shape; tell where objects are in space, how far something is, read maps etc.
- Knowledge: an individual’s knowledge about a vast array of topics
- Working Memory: capacity of short term memory, like repeating a set of numbers
All these things are supposed to correlate with g, according to Spearman, and they do. Jensen (1998) found that all these sub tests correlate with g, which vindicate the theory put forth by Charles Spearman.
To extract g, people performed a technique called factor analysis. To quote Gottfredson:
“factor analysis determines the minimum number of underlying dimensions necessary to explain a pattern of correlations among measurements. A general factor suffusing all tests is not, as is sometimes argued, a necessary outcome of factor analysis. No general factor has been found in the analysis of personality tests, for example; instead the method usually yields at least five dimensions (neuroticism, extroversion, conscientiousness, agreeableness and openness to ideas), each relating to different subsets of tests. But, as Spearman observed, a general factor does emerge from analysis of mental ability tests, and leading psychologists, such as Arthur R. Jensen of the University of California at Berkeley and John B. Carroll of the University of North Carolina at Chapel Hill, have confirmed his findings in the decades since. Partly because of this research, most intelligence experts now use g as the working definition of intelligence.”
In respects to factor analysis, Dolan (2000) and Dolan and Hamaker (2001) claim that the structure of intelligence can not be inferred. The problem with both of these studies is that they lacked statistical power. Other studies, with better statistical power, have looked at the structure of intelligence and have been able to outline it properly (Frisby & Beaujean 2015; Reynolds et al. 2013; Major, Johnson & Deary 2012; Carnivez, Watkins & Dombrowski 2017; Reynolds & Zeith 2017; Dombrowski et al. 2015; Reverte et al. 2014; Chen & Zhu 2012; Carnivez 2014; Carroll 2003; Kaufman et al. 2012).
It’s recognized that a general cognitive factor appears in data from every human culture (Warne and Burningham 2018). Some people have tried to deny the reality of g by claiming that it’s only a statistical artifact (Gould 1996), especially Hampshire et al. (2012) . Hampshire et al. is especially noteworthy as articles have ran with the headline that, according to Hampshire et al.’s findings, “IQ test are fundamentally flawed” (Connor 2012). The claims Hampshire et al. make are specially similar to Gould’s claims and Bowles and Ginitis (1973). In the Connor article for The Guardian, Hampshire et al. claim that:
“The results disprove once and for all the idea that a single measure of intelligence, such as IQ, is enough to capture all of the differences in cognitive ability that we see between people,”
“Instead, several different circuits contribute to intelligence, each with its own unique capacity. A person may well be good in one of these areas, but they are just as likely to be bad in the other two.”
In their paper, they cite Stephen J. Gould, a popular Marxist egalitarian who wrote the book The Mismeasure of Man. Gould’s work has been used to discredit the book The Bell Curve – and it seems that the authors are aware of Gould’s work on the topic of race and intelligence as they cite his work.
“It remains unclear, however, whether population differences in intelligence test scores are driven by heritable factors or by other correlated demographic variables such as socioeconomic status, education level, and motivation” (Gould, 1981).
“More relevantly, it is questionable whether they relate to a unitary intelligence factor, as opposed to a bias in testing paradigms toward particular components of a more complex intelligence construct”.
Gould makes the argument in his book that g is just a statistical artifact. Focusing on Gould, though, isn’t worthy for this article as refutations to his attack on g have already been done (Carroll 1995; Hunt 1995; Jensen and Weng 1994).
“The authors issued a press release from their university (The University of Western Ontario in Canada) the day before the Neuron publication on December 20th. This press release is in Appendix C. The title is: “Western University-led research debunks the IQ myth.” The press release received some attention mostly in non-science media outlets and hyped the study as demonstrating definitively that IQ was a meaningless concept. For example, the senior author, Adrian Owen, was quoted as saying: “When we looked at the data, the bottom line is the whole concept of IQ — or of you having a higher IQ than me — is a myth… There is no such thing as a single measure of IQ or a measure of general intelligence.” Of course, most psychologists understand that this is a classic “straw man” argument since no one claims that an IQ score (which is a composite of a test battery) measures the whole of human intelligence. It is also widely understood that the g-factor is not synonymous with IQ.”
“The study’s design appears to have suffered from a conceptual confusion that sometimes appears in studies using factor analysis. Hampshire et al. used averaged voxel activations (in a limited set of brain areas) across all 12 tests and all 16 subjects as the basis for identifying the two brain networks but then related these average across-subject voxel activation differences to individual differences in task performance in the much larger internet sample. The interpretation of these associations as indicating that neurological factors underlie variation in task performance is questionable.”
In laymen’s terms: it’s recognized that IQ test do not measure all of human intelligence, and Hampshire et al. start off with this straw man regardless. They used average voxel activations (voxels are 3-dimensional images in fMRI that represent a tidy cube of brain tissue, and each voxel can represent a million brain cells) in a limited set of areas on the brain on across all the 12 test used and 16 people for identifying two brain networks. The flaw is that they used these two brain network averages across-subject voxel activation in task performance for the much larger internet sample, largely making a blanket conclusion.
“Spearman’s hypothesis can now be considered to be an empirical fact. Mean differences in intelligence between ethnic groups can be largely explained by the complexity of the subtests in an IQ battery. So, the present study shows clearly that there is simply no support for cultural bias as an explanation of these ethnic group differences. Apart from subtests with a strong language component, IQ batteries appear to be excellent measures of intelligence for all groups studied in our meta-analysis.”
Since the primary difference between groups in intelligence has to do with g, that explains why Rushton and Jensen (2005) found a positive correlation between how well a question measures general intelligence and how the races differ in it. This means that the racial gaps in intelligence are primarily gaps in general intelligence, although the gap is not due solely to the g-factor (Jensen 1985).
“Intelligence Means Different Things in Different Cultures”
An interesting argument made by some commentators is that intelligence can mean many different things in different cultures, so how Americans see intelligence may not be how, for example, Africans see intelligence. Sternberg (2004) makes the argument that intelligence can not be understood without understanding its cultural context. He has a small section arguing that a child’s talent may go unnoticed in an academic test. Benson (2003) writes in the American Psychological Association that
“…Taiwanese-Chinese conceptions of intelligence emphasize understanding and relating to others–including knowing when to show and when not to show one’s intelligence.”
Researchers also looked at what intelligence is in Africa. According to Benson,
“When rural parents in Africa talk about the intelligence of children, they prefer not to separate the cognitive speed aspect of intelligence from the social responsibility aspect.”
The main evidence that Sternberg and others use is tests that are based on specific culture contexts. When giving a speech on his theory on intelligence, Sternberg cited a study done on Kenyan children in a village called Luo (Geissler et al. 2002). Within this study, the researchers collected 91 plant remedies from mothers and found that it was shared knowledge that the elders also knew, as well as the children in the village.
The children were able to memorize many natural remedies to combat diseases. Sternberg uses this study to demonstrate that intelligence relies on cultural context, and therefore it can’t be measured between cultures. One of the questions on this test went as follows:
A small child in your family has Homa. She has a sore throat, headache and fever. She has been sick for three days. Which of the following Yadh Nyaluo (Luo herbal medicines) can treat Homa?
i. Chamama. Take the lead and sniff fito (medicine up to nose to sneeze out the illness.)
ii. Kaladali. Take the leaves, drink and fito.
iii. Obuo. Take the leaves and fito.
iv. Ogaka. Take the roots, pound and drink.
v. Ahundo. Take the leaves and fito.
This question sounds strikingly similar to one found in the B.I.T.C.H IQ test, which stands for Black Intelligence Test of Cultural Homogeneity and was designed by a black psychologist (Williams 1972). In this test, serious questions like what is a “handkerchief head” and “how much does a short dog cost” are asked. The main point is that test can be biased due to culture that differs by racial and ethnic groups. These questions, both in the B.I.T.C.H test and in the Luo test, have nothing to do with intelligence. Ironically, though, the only “IQ test” that blacks outscore whites is in the B.I.T.C.H test.
What this argument tries to do, much like the theory of multiple intelligence, is to stretch out what the word “intelligence” means. For example, “even though Bill has a PhD in astrophysics, that doesn’t mean that Bob is less intelligent than him since Bob has social skills.” Sternberg and others loosen the definition of intelligence so that it can fit in things like talents and skills. No person would make the argument that someone who understands music theory is as intelligent as Newton or Einstein. It is generally recognized that skills and talents can not be substituted for someone’s intelligence, but it seems that as you loosen a word up to fit multiple things, it loses its original meaning. Because someone may have social skills or can fish, this doesn’t tell us anything about their intelligence. This line of reasoning is just an ad-hoc as it to tries to lump talents, skills and other things to try to invalidate what intelligence is (Levin, 1997).
Chooi, Long, and Thompson (2014) looked at Sternberg’s theory of intelligence via his Triarchic Abilities Test which measures three types of intelligence: analytical, practical, and creative. It was found that g is the best model that explains his data.
“There Are Multiple Types of Intelligence”
Much like the argument taken up above, the theory of multiple intelligence also tries to expand the word intelligence to fit things like talents and skills. The proponent of multiple IQ theory, Howard Gardner, claims that IQ could be broken down into multiple things (Gardner 1983):
- Logical – Mathematical
The first thing that should be obvious is how musical talent is taken to be a type of intelligence of its own, rather than just that – a talent. There are problems proving Gardner’s theory in the first place as he’s opposed to psychometric testing. This argument can be shot down fairly easily as Waterhouse (2010) found no evidence to support the theory of multiple intelligence. Gardner has also failed to provide evidence that supports his position:
“Although Gardner and Moran…… claimed that there was a wealth of empirical support for their theories, Gardner and Moran offered no research evidence to validate MI.”
Again, all this tries to do is stretch out the word intelligence to fit stuff that it doesn’t even align with. Talents do not equal intelligence.
The Flynn Effect
The Flynn Effect is the substantial and long sustained increase of crystallized and fluid intelligence test scores over time. Since IQ has been rising for different countries over time, it may mean that race differences in intelligence across different countries may soon diminish. With this, the hereditarian position on the genetic influence on IQ may become obsolete since it shows that environmental changes can cause almost a 15 point increase in IQ scores. Even Levin (1997), who’s a philosopher of science and released one of the most thorough reviews on race differences in intelligence, had a hard time responding to the Flynn Effect.
Sowell (2002) uses the Flynn Effect as an attack on genetic influence on IQ scores between blacks and whites.
“Since the black-white difference in IQ is 15 points, this means that an even larger IQ difference has existed between different generations of the same race, making it no longer necessary to attribute IQ differences of this magnitude to genetics. In the half century between 1945 and 1995, black Americans’ raw test scores rose by the equivalent of 16 IQ points.”
Despite the Flynn Effect being a real thing, it doesn’t explain IQ differences between groups. Why? It’s because the Flynn Effect is not on g, the primary area of where the races differ in. In a review on the literature on the black-white IQ gap, Rushton and Jensen (2005) looked at the Flynn Effect:
“One culture-only hypothesis currently enjoying much support is based on the secular increase in test scores, known as the Flynn effect because of the repeated demonstration by James Flynn (1984, 1987, 1999) that the average IQ in several countries has increased by about 3 points a decade over the last 50 years. Some have suggested that the Flynn effect implies that the 1 standard deviation difference in the mean Black–White IQ difference in the United States will gradually disappear over time (Flynn, 1999). However, one statistical analysis shows that the Flynn effect is not on the g factor, the principal source of the mean Black–White group difference.”
Even Flynn (2011), an egalitarian in the IQ debate, and who the Flynn Effect is based on, has said that it is not on g.
“The magnitude of white/ black IQ differences on Wechsler subtests at any given time is correlated with the g loadings of the subtests; the magnitude of IQ gains over time on subtests is not usually so correlated; the causes of the two phenomena are not the same.”
Since the Flynn Effect is not on g, that means that there should be a negative correlation between score gains in IQ and g-loadings. Nijenhuis and Flier (2012) found a negative correlation between score gains and g-loading, and that the Flynn Effect and group differences have different causes – this suggests that the Flynn Effect is on item specific cognitive abilities and not g. This holds true even when assessing the Flynn Effect on scholastic achievement (Jensen, 1998).
The Validity of IQ Tests
Some commentators, like the internet personalities on The Majority Report with Sam Seder, argue that IQ tests only measure how well you do on IQ tests. IQ tests simply don’t measure how well someone does on IQ tests, but it also predicts real world outcomes. Some IQ correlates can be seen below.
When looking at this chart, especially for those who may not understand statistics, r refers to the correlation between two variables. In this case, it’ll be the correlation between IQ and something else. Anything with an “N” next to it signifies that it’s on the national level, showing that IQ simply doesn’t correlate with individual items only.
|Educational Attainment||0.53||Strenze (2006)|
| Becoming a leader in|
| Popularity among |
|Number of children||-0.11||ibid|
in primary education
| Occupational |
|0.5 – 0.7||ibid|
|Socio-economic Status||0.4|| Herrnstein and Murray |
|Dropping out of HS||0.58||ibid|
|Rape, 1990’s *n||-0.29||Rushton and Templer (2009)|
|Assault, 1990’s *n||-0.21||ibid|
|Homicide, 1990’s *n||-0.25||ibid|
|Economic Freedom *n||0.52 – 0.76||Lynn and Vanhanen (2012)|
|Income Inequality *n||-0.51 – -0.60||ibid|
| National School |
|0.87 – 0.92|| Rinderman (2007); Lynn et al. |
Since IQ scores on IQ tests measure outcomes in the real world, they are not simply tests that only measure how well someone does on IQ tests. More IQ correlates can be found in Herrnstein and Murray (1994) and Levin (1997). A common argument against IQ test is the reduction of intelligence to a single number. Despite this criticism against a reductionist take on intelligence, IQ test have proven their validity with the use of a single number to predict a groups outcome. Most intelligence researchers also agree, contrary to layman beliefs, that intelligence can be measured as a single ability or a cluster of abilities (Synderman and Rothman 1988).
IQ test scores strongly correlate with life outcomes, including socio-economic status and cognitive ability, even when measured early on in a child’s life (Foverskov et al. 2017). In a society where things are becoming more technology based, having to use maps and transportation schedules, banking, interpreting news articles and other stuff, higher intelligence may offer a small advantage to these things. Although the advantage of higher intelligence may not be extremely large, they accumulate to affect overall chances in an individuals life – as g is required for everyday life (Gottfredson 1997).
So, IQ tests do measure real world outcomes rather than how one simply does on an IQ test.
“IQ Tests Are Biased”
A popular belief in the social sciences is that IQ test are biased against minorities. Sternberg (1986) states that
“We must also recognize the limitations of present day intelligence tests. Largely developed and standardized on white middle-class children, these tests tend to be biased against black children to an unknown degree.”
Some researchers claim that since IQ tests were developed in white societies, they’d have culture bias ingrained into them (Martschenko 2017).
There are three ways to test if there is bias in IQ tests: seeing if cultural bias leads to lower IQ scores, if tests under predict black performance, and if internal bias can lead to lower scores.
When it comes to culture bias in IQ tests, Jensen (1980) reviewed the literature on bias in IQ tests and he found that IQ test are not biased. Besides Jensen, other sources have also found that IQ test are biased.
Reeve and Charles (2008) found that 73.3% of psychometric experts surveyed said that there is no cultural-bias in IQ test.
The National Research Council and The National Academy of Science concluded that cultural bias was not an issue in cognitive testing among racial groups in the US:
“The Committee further concluded that ability tests predict
equally well for all groups of test takers. Research evidence does not support the notion that tests systematically underpredict the performance of minority group members.”
Gottfredson (1994) noted that IQ test aren’t biased against blacks and people of different social classes:
“Intelligence is a very general mental capability that…can be measured, and intelligence tests measure it well. They are among the most accurate (in technical terms, reliable and valid) of all psychological tests and assessments….While there are different types of intelligence tests, they all measure the same intelligence….Intelligence tests are not culturally biased against American blacks or other native-born, English-speaking peoples in the U.S. Rather, IQ scores predict equally accurately for all such Americans, regardless of race and social class.“
Neisser at al. (1996) found that IQ test aren’t “biased against African Americans.” It’s worth remembering that Neisser et al. was also released after the book The Bell Curve. Referring to bias, Neisser et al. states that
“none…contributes substantially to the Black/White differential in intelligence test scores.”
Dyck (1995) states that
“The evidence indicates that cognitive test are equally reliable across races, are of equivalent item difficulty across races, yield similar sub-test correlations…and factor analyses yield similar results. The question of biased has been answered: they are not.“
Brown and Whitaker (1999) reviewed the literature on test bias in IQ test after Arthur Jensen’s 1980 book Bias in Mental Testing (Jensen 1980); they looked at claims that they were biased, not biased, and critiques against Jensen. They state that
“The major conclusion of this article can be stated confidently: Empirical evidence overwhelmingly supports the conclusion that well-developed, currently-used mental tests are of equivalent predictive validity for American-born, English-speaking individuals regardless of their subgroup membership.”
Based off of all this, we can safely say that IQ tests are not biased against blacks or other minorities, especially among people of differing social classes. Now we can see if IQ tests actually under predict the performance of blacks; the answer is no – they actually over predict.
Coleman et al. (1966) looked at more than 645,000 students and their survey included a standardized test of verbal and non-verbal IQ. The study also included educational achievement measures of reading and math levels that are thought to be straight forward measures of what the student has learned. The results were that black IQ scores actually over represented black academic achievement by about .26 standard deviations.
Looking at SAT data, Breland and Gaynor (1979) reviewed the literature on SAT scores by ethnic groups and found that SAT scores over predicted freshman grades for blacks in 14/15 studies looking at. The over representation was about .20 standard deviations, and in an additional 5 studies where ethnic classification was only “minority” rather than “black”, SAT scores over predicted college performance in all 5 studies by about .40 standard deviations.
Some arguments of internal bias are some you have probably heard before or have made before, such as motivation, anxiety, race of examiner and time limits, etc. These are not problems for IQ tests.
When it comes to the race of the examiner giving the IQ test, having a white examiner actually reduces overall black-white differences (Sattler and Gwynne 1982; Jensen 1980); examination of time pressure fails to show that blacks do better in untimed than timed tests, or that “personal tempo” of blacks is different from that of whites (Jensen 1980); test anxiety actually slightly helps, and of the studies that have looked at black-white differences in anxiety – they’ve shown either non-significant results or that whites are more anxious than blacks (Jensen 1980). When it comes to motivation, the data is contradicting.
Motivation would just refer to the test takers level of interest, striving, their effort to take the test, concentration, persistence etc. Jensen (1980) reported on a study that found that for every authors finding supporting the role of motivation in IQ testing, there is contradicting evidence from another author. Jensen notes that measuring motivation is
“usually by means of projective techniques and questionnaires, has not yet been fully standardized in the research literature, and often it is not clear that different investigators are measuring one and the same construct.”
Heckhausen (1967) also concludes that most studies do not show a significant correlation between measures of scholastic motivation and IQ scores.
From all this evidence, race differences in IQ scores can not be pinned onto bias in IQ tests or internal bias.
Since the data shows that there are race differences in intelligence, some people may concede. Nisbett (2009) in Intelligence and How to Get It concedes that whites have higher IQ scores than blacks, but Nisbett is an environmentalist and takes the culture-only approach to explain away race differences in intelligence. So the next passages that you will read will discuss the environmental effects on race differences in intelligence and see if they can hold up.
In explaining race differences in intelligence, it might be that socio-economic status leads to having a higher IQ, which will lead to higher IQ scores. There are two ways to frame the socio-economic status question: Does lower socio-economic status reduce IQ across racial lines and; does controlling for socio-economic status close the black-white IQ gap?
In respects to the first question, the evidence does suggest that smarter people come from higher SES homes. Sirin (2005) meta-analyzed data on roughly 100,000 students and found that the correlation between IQ and SES is .26, which is weak to moderate for r. The problem is that SES is also heritable at .42% (Hyytinen et al. 2013). Regardless, socio-economic status does not reduce IQ across racial lines.
The effect on poverty on race differences in IQ don’t seem to hold across racial lines as a 9-12 point gap remains when status is controlled for (Jensen 1980; Neisser et al. 1996); white children have higher Peabody Vocabulary Test scores at all income levels (Currie and Thomas 1995); lower class whites consistently do as well or better than middle and upper class blacks (Jensen 1980; Scarr 1981).
Now the second question can be answered: Does controlling for socio-economic status reduce the black-white IQ gap? The answer to this, much like the answer to the first question, is no. It is true that as SES goes up IQ does also, but this doesn’t mean that blacks in the same SES level as whites will be as equally intelligent or even smarter than whites.
Herrnstein and Murray (1994) found that as black IQ scores go up with SES, it doesn’t close the black-white IQ gap.
Studies that have controlled for SES have not found that it closes the black-white IQ gap. Shuey (1966) looked at 42 studies carried out between 1913-1966 in which blacks and whites had the same SES. 95% found that whites still had higher IQ’s; 2 found no race differences when SES was controlled for, but none found that blacks were smarter than whites. Loehlin, Lindzey, and Spuhler (1975) also looked at 7 studies between Shuey’s work and 1973. All 7 studies showed that whites had higher IQ’s than blacks even when comparing blacks and whites in the same SES.
Even using the SAT, which is a highly g-loaded test (Frey and Detterman 2004), it shows race differences in IQ between blacks and whites at all income levels (Black Journal of Higher Education 2009), which fits in line with the findings presented by Shuey and Currie and Thomas.
Cottrell (2015) claims that the black-white IQ gap is caused by family income, maternal education, maternal verbal ability/knowledge, learning materials in the home, parenting factors (maternal sensitivity, maternal warmth and acceptance, and safe physical environment), child birth order, and child birth weight. There are strong reasons to cast doubt on this study.
Beaver et al. (2014) found that parenting factors are not significant variables in IQ scores, family income doesn’t close the black-white IQ gap even after SES is controlled for, and the rest can be found in Rushton and Jensen (2005). For example, Rushton and Jensen note that
“Racial-group differences in IQ appear early. For example, the Black and the White 3 year-old children in the standardization sample of the Stanford–Binet IV show a 1 standard deviation mean difference after being matched on gender, birth order, and maternal education.“
The most recent study claiming that poverty reduces IQ is Mani et al. (2013). They claim that lack of resources, or scarcity in economic terms, triggers something that impedes cognitive function. Graves (2015) tried replicating what Mani found, but instead they said that “experimental results do not find that mental and financial scarcities significantly impact test performance.” Although in the literature results do suggest that scarcity may hinder problem solving abilities, more research is needed.
A personal critique of Mani et al. is that they introduced two studies: one looked at shoppers and the other at farmers. With the former, they separated New Jersey shoppers into two groups: poor and rich. They found that showing the poor thoughts on finances reduced cognitive function, but not for the rich. The problem is that they didn’t even measure the IQ of the poor shoppers before hand, so we have no idea if they were already less intelligent before hand. Regardless of my critique, Dang, Xiao, and Dewitte (2015) looked at Mani et al. and Vohs (2013) (which found similar results as Mani et al.).
To quote Dang, Xiao, and Dewitte:
“Although these authors went through great efforts to safeguard external validity of their independent variable, we contend that they paid insufficient attention to that of the dependent variable. First, the cognitive tasks they used (e.g., IQ tests, the Stroop task) are irrelevant to participants’ daily life. It is possible that the poor do not have sufficient motivation to fully engage in these tasks while worrying about their financial situation.“
“Further, we suggest poverty does not necessarily lead to self-regulation failure. Previous studies demonstrated that engaging in a concurrent inhibitory task (e.g., retrenching expenditure within limited budget) would facilitate self-regulation through an inhibitory spillover mechanism (Tuk et al., 2015) or by blocking individuals from recognizing the tempting value of attractive stimuli (Van Dillen et al., 2013). From this perspective, the poor may excel in self-regulation under certain circumstances. Future studies are needed to specify.“
McClelland et al. (2015) note some statistical errors in Mani et al., such as median splits, and how critiques made by Wicherts and Scholten (2013) against Mani et al. were correct even after Mani et al. responded to them. When replying to Wicherts and Scholten, Mani et al. just made the argument that “other people did the same thing we did, so it’s alright.” According to Wicherts and Scholten (2013),
“the stronger cognitive impediment experienced by the poor could merely be the result of an inappropriate statistical test and an overly easy cognitive control measure. The latter could obscure an equally “threatening” effect in the rich, simply because they were unable to obtain higher scores when not threatened. With such methodological issues remaining to be addressed, the authors’ proposal of far-reaching policy changes, such as timing HIV educational campaigns to harvest cycles, seems premature.”
Beyond all this, it could be that low-SES among blacks could lead “black dialect.” This means that blacks from lower SES homes will speak a non-standard dialect distinctly different from standard English, and that black dialect could differ from the person who’s administering the test which would lead to a lower IQ score. Although Negro children do produce different speech, they understand standard English at least as well as they understand their type of dialect (Eisenberg et al. 1968).
The effect of Negro dialect on the IQ of lower class black children has been researched by Quay (1971; 1972; 1974). Given the Stanford-Binet test translated into black ghetto dialect by a linguistic specialist, no significant difference was found between non-standard English dialect and standard English forms on the Stanford-Binet. Thus, Negro dialect from lower SES can not be blamed for lower IQ scores.
So it seems that socio-economic status doesn’t close the black-white IQ gap. The next possible environmental factor is adoption. If blacks are adopted by whites, could their IQ be raised? Some studies have been cited by environmentalist in an effort to show that it can increase IQ by almost 15 points, but many of these studies do not show permanent gains.
Twin studies have shown that poor kids put into higher SES have had their IQ raised by x points. Schiff et al. (1978) found that poor french children put into higher SES homes had their IQ raised by 15.5 points. There was no follow-up on this study, so we can’t see if they’re permanent gains.
Moore (1986) compared the average IQ of 23 black children adopted into white homes. The children, aged 7 and 10, had an IQ score of 117 compared to an IQ of 103 for those in black homes. The sample size in this study is incredibly small, making it an unrepresentative sample. Since the IQ’s of those in black homes high were already high, it’s very likely likely that Moore was studying on a sample where there were no differences in IQ – making this study useless for the race and IQ debate.
Moore also noted the differences in behavior in how whites and blacks were treated. White adoptive mothers reduced stress by joking, laughing and grinning, where as black mothers reduced stress in less positive ways by coughing, scowling, and frowning. She also states that the white adoptive mothers gave more positive reinforcements than the black mothers, and attributes this to culture rather than genetics (Rushton and Jensen, 2005).
Eyferth (1961) looked at the IQ scores of out-of-wedlock children fathered by U.S. soldiers in Germany reared by white German mothers during WW2. The mean IQ for the white children was 97, and 96.6 for the mixed race children, respectively. The problem with this study, of course, is that the children were very young and there was no follow-up done. This matters due to Wilson Effect, dictating that the heritability of IQ increases with age. So only testing kids at a young age may mask genetic effects. Also, 20 – 25% of the black fathers in this study were French North Africans (Caucasian). This shows why the mixed kids had a higher IQ (Rushton and Jensen 2005).
Probably the only good adoption study comes from Scarr and Weinberg (1983). They looked at the IQ scores of 130 black and interracial children adopted by advantaged white families. They tested the IQ of the children at age 7 and 17. The results can be seen at the table below.
From very early on, the high IQ among blacks gave hope to the environmentalist who hoped that IQ increases were possible. By the age of 17, this was shown not to be the case. Scarr and Weinberg claimed that genetics were not the major cause of IQ differences, and that black and interracial children performed the same as other adopted children in similar families. However, the racial IQ gaps between the adopted children did not differ significantly from the general population (Scarr and Weinberg 1992).
The reason Scarr and Weinberg claimed that genetics was not the cause of the black-white IQ gap might be because of their stake in the race and IQ debate.
“The test performance of the Black/Black adoptees [in the study] was not different from that of ordinary Black children reared by their own families in the same area of the country. My colleagues and I reported the data accurately and as fully as possible, and then tried to make the results palatable to environmentally committed colleagues. In retrospect, this was a mistake. The results of the transracial adoption study can be used to support either a genetic difference hypothesis or an environmental difference one (because the children have visible African ancestry).Scarr (1998)
We should have been agnostic on the conclusions.”
It is possible to say that the Scarr and Weinberg study is old by now and there haven’t been other adoption studies like this done – which is valid critique – but there have been studies of Asians adopted by whites. If the IQ of Asians increased but dropped later on in life, then it lends to the support of the Scarr and Weinberg findings. (The following table was adopted from Lynn.)
IQ of Asians Adopted by Whites
The following sample from Winicks et al. (1975) has been divided into three groups: consisting of severely nourished children (row 1), poorly nourished (row 2), and well nourished (row 3). The IQ of these 3 groups were related to their nutritional history. Group one did not score differently than American whites, but the other two groups did score higher. The mean IQ of these 6 studies (Clark and Hanisee 1982; Frydman and Lynn 1989; Stams et al. 2000) is 109, and if Winicks is removed, the mean is 111. Applying the results of Scarr and Weinberg to the findings of adequately nourished children, it suggests that their adult IQ would’ve dropped by 6 points which matches up with the IQ of North East Asians in the US and in other places, according to Lynn who used evidence from Levin (1997).
So adoption studies do show an increase in IQ gains, but they are not permanent. If anything, they can raise IQ by < 5 IQ points, but it can not raise it to 15 points in which the black-white IQ gap will totally diminish.
Intervention would refer to intervention in a child’s life that, if the hypothesis holds true, would help increase their IQ. Granted, there does seem to be some support for.
According to Nisbett (2009), blacks have closed the IQ gap by about 5.5 points between 1970 – 1992. At the same time, blacks also closed the educational achievement gap by 35%. According to Nisbett, intervention can narrow the black-white IQ gap. However, it’s very unlikely this is the case since Chuck (2013) and, using Nisbett’s data, Rushton and Jensen have found that the black-white IQ gap has remained constant at about 1.1 SD’s. Using a wider array of tests, no narrowing of the black-white IQ gap has been found (Murray, 2009).
Skuy et al. (2002) looked at 1st-year psychology students and found that an intervention actually increased the black and non-black IQ in Ravens. The IQ score for blacks went from 83 to 97, and for non-blacks it was from 103 to 107. The problem with this study is that it taught people how to solve questions on Raven’s Progressive Matrices; this is called studying for the test – which shows people how to solve specific questions that you’d find on an IQ test (which defeats the purpose of an IQ test). Regardless of this critique, no follow-up was done to see if these IQ scores were permanent gains. Luckily, the Head Start program study did give a follow-up.
Currie and Thomas (1995) looked at black and white kids who were part of the Head Start program. According to the study:
“When selection is controlled in this way, Head Start has positive and persistent effects on the test scores and schooling attainment of white children, relative to participation in either other preschools or no preschool. In contrast, while the test scores of African-American children also increase with participation in Head Start, these gains are quickly lost, and there appear to be no positive effects on schooling attainment.”
For some reason, Head Start helped whites but not blacks. This may suggest that helping raise the IQ of white children is possible, but for Negro children it’s only temporary. Two other programs, The Perry Preschool Program and The Milwaukee Project, have attempted to help the intelligence of blacks, but they too have shown temporary gains (this is talked about extensively in Splitz 1986).
The Perry Preschool Program had black children with IQs from 50 to 85, and they were given special classes for five years and tracked until the age of 19. In the Milwaukee Project, children 6 months of age or younger of low-IQ black women spent large parts of each day for 5 years in an Infant Stimulation Center. Mothers were given remedial education and extensive home visits were given (see Garber 1988; Jensen 1985a, 1985b, 1989). In each of these cases, the experimental children made large gains on IQ and academic performance.
While the kids did make large IQ gains and had an improvement in academic performance, these gains disappeared after the program came to an end (Splitz 1986). Neisser et al. (1996) says that in such studies
“long run gains have proved more elusive.”
These findings of temporary gains are further confirmed by a meta-analysis finding that intervention does raise IQ, but there’s a fade out in IQ gains over time (Protzko 2015). Haier (2014) cautions in trusting the argument that intervention studies, without a follow-up and separate analysis, increase IQ permanently rather than temporarily
Nutrition is strongly needed for a child’s development, which will affect their intelligence as they age throughout in life. Northstone et al. (2012) found that a diet in high fats, sugars and processed foods early on in childhood may result in lower IQ scores which will affect g. A diet rich in healthy foods packed with vitamins and nutrients would work in reverse, though. Since whites are in high SES homes at a higher rate than blacks, it could mean that whites are able to afford healthy food which will assist their IQ.
Black children eat as well as whites. According to a 1985-1986 survey by the U.S. Department of Agriculture described in Levin (1997), black preschool children consume 56.9 grams of protein daily as compared to 52.4 grams for white preschool children. Children in families 75% below the poverty line, a disproportionately black cohort, consume as much vitamins A, 8-6, B-12, and E, thiamin, riboflavin, niacin, phosphorus, and magnesium as children in families 300% above the poverty line.
Since malnutrition retards bone ossification, the greater speed of ossification of cartilage in black children (Jensen 1973) indicates adequate nourishment. Up to 18-24 months, black children outpace white infants in muscular and neurological development (Bayley 1965). This offset favoring blacks is inconsistent with great black malnutrition. Thus, malnutrition is not responsible for the black-white IQ gap.
Stereotype threat refers to situations in which people might feel at risk to confirming to their groups stereotype, and then this causes them to do bad on test. In theory, this should hold true for race and gender.
Brown and Day (2006) found evidence that stereotype threat harms students on Raven’s Progressive Matrices. The American Psychological Association (APA) bolstered the argument by claiming that stereotype threat “undercuts conventional assumptions that genetics or cultural differences lead some students” to do worse on IQ tests. Shelvin, Rivadeneyra, and Zimmerman (2014) also found similar results as Brown and Day.
The original purveyors of the stereotype threat argument were Steele and Aronson (1995) on their paper on stereotype threat and its effect on blacks. Since then, it’s been cited more than 1,000 times in other academic papers. How did Steele and Aronson come to their infamous conclusion that’s still cited to this day as an attack on IQ tests? They simply framed the test as a challenge instead of just that, a test. The belief went that if blacks were told it was a challenge, then they’d perform worse on it. After that, simply remove the threat and race differences went away. In reality, there was many flaws in Steele and Aronson’s work, such as adjusting for ACT scores instead of using mean tests scores, and that a response made later to other researchers by Steele and Aronson said that
“…without this adjustment, they would be shown to perform still worse than White.”Jussim (2015)
Besides Arson and Steele’s fudging, surely stereotype threat should be something that really does affect minorities…right? No. In reality, stereotype threat is under replication bias, meaning that sometimes it’s replicated and other times it’s not.
Wei (2009) looked at 64 papers on race and gender stereotype threat, and stereotype threat was replicated only 58.4% of the time. This shows replication bias, and it was only produced almost 60% of the time, and 30% of the time it wasn’t. To quote an article looking at the replication of stereotype threat:
“In conclusion, a replicability analysis with the R-Index shows that stereotype-threat is an elusive phenomenon. Even large replication studies with hundreds of participants were unable to provide evidence for an effect that appeared to be a robust effect in the original article. The R-Index of the meta-analysis by Flore and Wicherts corroborates concerns that the importance of stereotype-threat as an explanation for gender differences in math performance has been exaggerated. Similarly, Ganley, Mingle, Ryan, Ryan, and Vasilyeva (2013) found no evidence for stereotype threat effects in studies with 931 students and suggested that “these results raise the possibility that stereotype threat may not be the cause of gender differences in mathematics performance prior to college.”
Pennington et al. (2019) found that stereotype threat didn’t harm student performance, further leading to the argument that this isn’t a problem in IQ test.
Home Environments and Child Abuse
In the case of IQ and home environments, it maybe that low IQ parents make bad environments for their kids by treating them worse. Herrnstein and Murray found that the evidence suggest that child maltreatment is mostly from low income families (most low income families tend to be blacks and other minorities).
In their book The Bell Curve, they found that low IQ parents were more authoritarian to their children than high IQ parents, and that high IQ parents responded better to their kids than low IQ parents.
“480 infants of women registered for prenatal care at an urban hospital for indigent persons and their children found that less educated mothers, even with this disadvantaged population, were more likely to neglect their children.
A quantitative study of 113 two-parent families in the Netherlands found that parents with a high level of reasoning complexity (a measure of cognitive ability) responded to their children more flexibly and sensitively, while those with low levels of reasoning complexity were more authoritarian and rigid, independently of education and occupation.“
Since low IQ parents are more likely to neglect their kids, it explains why child abuse is higher in black families than in white families, considering their 1 SD gap in IQ.
Using twins, Currie and Tekin found that when one twin is abused and the other is not, the abused twin is, on average, more criminal, They also become less intelligent (Koenen et al. 2003). This holds true even after controlling for birth order, maternal education, paternal criminality, religion and family structure, according to Currie and Tekin.
A problem with this is that parents might be abusing their kids because of their behavior. So if a child is violent and lacks self-control, their parent might use physical force on them. This might be why blacks have a higher child abuse rate considering black kids act out more and have less self-control than whites:
“Teachers reported better social adjustment and less hostility-aggression from Mongoloid children than from Caucasoid children, who in turn were better adjusted and less hostile than Negroid children“Rushton (1999)
“negroes are impulsive, indulge themselves, settle for next to nothing if they can get it right away, do not work or wait for bigger things in the future.”
I am aware of no study controlling for race differences in behavior. To the degree that child abuse, specifically, is responsible for the black-white IQ gap is maybe 3-4 points as that’s how much child abuse lowers IQ by (Paolucci, Genius, and Violata 2001 ; Klika and Herrenkohl 2013; Koenen et al. 2003). A problem with this, however, is that race differences in IQ are seen at a young age, so that can be a confounding variable. Child abuse on IQ also seems to have an effect on low IQ children rather than high IQ children (Klika and Herrenkohl 2013), with the former being predominately black.
The impact of education on IQ, especially since most studies show an increase in IQ with education, should blow the gene theory – that’ll be talked about later – out of the water. If education does affect IQ, it doesn’t seem to be on g.
In general, it’s commonly argued that blacks get inferior education when compared to whites due to lower school funding, but school funding is actually somewhat progressive with blacks getting more school funding per student when compared to their white counterparts.
Past studies have avoided per-pupil spending and others used have used small samples that aren’t representative of the country, but Richwine did the opposite of this. A good critique of this is that black schools may not be allocating their funds correctly, but I’m aware of no study that finds to be the case.
Since blacks get more school funding, school funding doesn’t seem to correlate with academic achievement since blacks have shown that not to be the case – in respects to school funding. It could be that school funding doesn’t matter, but just getting an education increases IQ. This does seem to be the case.
Ritchie and Tucker-Drob (2018) found that education increases IQ by about 1 to 5 IQ points. So if somehow we just made blacks stay in school, it should probably raise their IQ past their current 85 IQ level. While education may increase certain cognitive abilities, it doesn’t seem to be on g, the primary area where the races differ in.
Huston (2018) notes that Richard Haier, the editor-in-chief of Intelligence, and Huston say that
“’IQ points are useful metrics,’ he says, ‘but they’re not really a measure of intelligence directly.’ Any effects of education on IQ may be related to improvements in particular skills, as opposed to a broader elevation of general cognitive ability.“
The fact that race differences in IQ are seen in a young age before blacks are put into school may show that education doesn’t increase the general intelligence of blacks, and that education doesn’t increase g overall.
It should be no surprise that lead exposure affects intelligence in a negative way (Stewart et al. 2007). Since Hood (2005) found that lead exposure is higher among poor communities, and poorer communities tend to black, it makes sense as to why blacks have 27% higher lead exposure in their tibia,
“likely because of sustained higher ongoing lead exposure over the decades”
Theppeang et al. (2008)
Nevin (2012) showcases that blacks, regardless of SES, were more likely to have higher exposure to lead than whites. The answer on if lead plays a role in lower IQ is an obvious yes, but the question is: “By how much does it lower IQ by?” Luckily, we have data on blood level by race with a national representative sample. CDC (2005) data goes all the way back to 1991:
The difference between blacks and white varies by age group, but the most important is the 1-5 year old age range. The difference is about 2 micro grams. A difference in 2 micro grams give a difference of about 2 IQ points, and Nevin cites a study saying that IQ declined by 7.4 IQ points as lifetime average blood lead concentration increased from 1 to 10 ug per deciliter.
A meta-analysis on 7 studies by Lanphear et al. (2005) found that
“the estimated IQ point decrements associated with an increase in blood lead from 2.4 to 10 µg/dL, 10 to 20 µg/dL, and 20 to 30 µg/dL were 3.9 (95% CI, 2.4–5.3), 1.9 (95% CI, 1.2–2.6), and 1.1 (95% CI, 0.7–1.5), respectively.”
A model from their paper can be seen below.
Using their model and race differences in lead exposure, lead exposure explains less than 1 IQ point. This means that the 1 SD gap in IQ (15 points) goes down to 14 points after lead exposure is taken into account. Past studies have shown higher IQ drops within children who have been affected by lead, but some didn’t control for confounding variables and other things (refer to Kaufman 2001 for more).
Despite all this, lead does have negative exposure effects that should not be ignored. In the case of its effect on intelligence, it doesn’t seem to be the route that should be taken to explain away race differences in IQ.
One hypothesis that could make sense is the fact that Africans were brought to the U.S. by force and were forced to live their lives as slaves. In turn, slavery somehow could’ve caused the black-white IQ gap as slavery would have stunted their IQ. In order to make this hypothesis true, Africans in Africa would’ve needed to be intelligent in the first place – and the evidence suggest that this isn’t the case.
In modern times, Africans are less intelligent than whites. Lynn (2011) did a meta-analysis on studies looking at the average IQ of sub-Saharan Africans, and the tables can be seen below.
The IQ scores above average to 75, which would put Africans in the very dull range. Wicherts, Dolan, and Mass (2009) object to this and claim that the IQ of Africans is actually higher due to Lynn testing people who have been exposed to diseases and other stuff – that of which will lower IQ.
Lynn has responded to Wicherts et al. by arguing that those are real problems in Africa that must be taken into account. Even then, evidence shows that things like malaria may not have an adverse effect on intelligence (Muntendam et al. 1996), and that there may not be a correlation between diseases and IQ (Sailer 2010). If we ignore Lynn’s analysis and only focus on Wicherts et al., we can see that they tested an unrepresentative sample of the African population. Findings from Wicherts et al. can be seen below.
|Nigeria||2/6||118||McCarth||89||Ashem & Janes (1978)|
|Nigeria||13||803||CCF||95||Nenty & Dinero (1981)|
|Sierra Leone||8||202||DAM||91||Ohuche & Ohuche (1973)|
|South Africa||19||228||SPM||97||Crawford-Nutt (1976) |
[Reported in Wicherts]
|South Africa||24||40||WAIS-3||84||Shuttleworth et al. (2004)|
|Zimbabwe||8||52||PMA||84||Wilson et al. (1991)|
[Reported in Wicherts]
The studies cited by Wicherts et al. were done on elite African samples. All the people used were largely unrepresentative of the general African population (refer to Lynn  for more on this).
The strongest case against the slavery hypothesis is the fact there have always been racial differences in intelligence even before slavery happened. Lynn (2009) looked at race differences in brain size, since they correlate with IQ, cold winters theory, and Bakers (1974) criteria for civilization. He found that race differences in intelligence have been present for the past 10,000 years.
Thus, the slavery hypothesis that it handicapped the black IQ can not work. In conjunction with the historical oppression narrative on IQ is that of segregation.
Since blacks and whites were both segregated for a time, it makes sense to think that the type of education given to each race was “unequal,” a position supported by Hammond (1998); of course, some individuals like Sowell contest to this and argue that desegregation didn’t improve black performance and that blacks did better academic wise under segregation (Sowell 2006). In order to test this argument, we can compare IQ scores during and after segregation. If there is a shift in the gap after segregation ended, then the segregation argument will be shown to hold merit.
We have some data on the IQ of blacks prior to the ruling of Brown v. Board Education (1954) and after. If there’s a shift in IQ scores due to integration, then we can see that segregation did temporarily stunt the black IQ.
As can be seen from the chart above, the IQ’s of blacks from ages 3-18 has remained constant since 1916 all the way to 1965. Rivkin and Welch (2006) analyzed studies on the effects of desegregation on blacks, and two study found that desegregation did not have a large effect on the black-white academic achievement gap. The change caused by desegregation was very small. Herrnstein and Murray (1994) also found similar results supporting the fact that the black-white IQ gap didn’t close after desegregation.
If the IQ of blacks from 1916-65 didn’t shift at all, and if desegregation didn’t close the black-white IQ gap or narrow it, then this argument from the environmentalist position can be thrown away.
The last resort in the oppression narrative argument is that racism is responsible for the black-white IQ gap. This argument is a null-hypothesis, and frankly, there’s no way to really disprove it as you can’t control for racism in the first place. Despite this fact, there are a few lines of reasoning that suggest racism doesn’t play a role in the black-white IQ gap.
(i) Self-Esteem: If racism has had long-term effect on blacks, it doesn’t seem that blacks have internalized this due to them having higher self esteem levels than whites (Bachman et al. 2012; AAUW 1991; Parker et al. 1995; Steele 1992). Despite blacks performing worse in school, this doesn’t play a role in their self-esteem. So it seems that blacks haven’t internalized racism that may lead to a lower IQ.
(ii) Blacks Do Worse in IQ Tests All Over The World: Some countries do not have the same history as the U.S., and so you would expect blacks to perform better on IQ tests there. Regardless of where blacks are, their IQ is always lower than that of whites (Lynn 2011).
(iii) The Performance of Asians: If blacks have lower IQs because of racism, then the same should hold true for Asians, but it doesn’t seem to be the case. Asians outscore whites in IQ tests and are generally recognized to be more intelligent (Herrnstein and Murray 1994; Rushton and Jensen 2005; Lynn 2011; Levin 1997).
Hopefully, all this evidence has shown that environmental explanations can not explain the totality of the black-white IQ gap. The most infamous position that has been argued is that of a genetic influence on IQ. This will manifest itself via brain size differences, gene allele differences, and other things which are largely genetic.
The Genetics of Intelligence
An argument frequently assessed by egalitarians and those alike is that race is not tied to intelligence. For example, just because someone may have black skin it doesn’t mean that it’s their skin color that causes low IQ; race, your genetic ancestry, is not tied to IQ directly. Meaning that just because you’re of European ancestry, it doesn’t mean that that’s what determines your intelligence. This is a point many hereditarians concede to, but an argument can be made against this position.
If the races evolved differently to have differently morphological features and gene allele differences, then race (since you can classify people based off of morphological features and gene alleles) is tied to IQ. This will become more clear later when the race differences in morphological features and genes that correlate with IQ are discussed.
In order to first understand the genetics of intelligence, one must first grasp what a gene is and what the gene-environment expression/ correlation is.
What is A Gene?
To put it simply, a gene is a unit of heredity. Much like blueprints, genes are instructions for characteristics of individuals like eye color, height, intelligence, temperament and so on. Genes are small sections of DNA that contains the instructions for a specific molecule like protein. The purpose of genes is to simply store information, and each gene contains the information required to build a specific protein needed for an organism.
In cells, long strands of DNA are compacted into form units called chromosomes. Humans inherit two sets of chromosomes from their mother and their father. Genes are inherited through chromosomes.
When a gene comes in different forms, it’s referred to as an allele. Alleles for a particular gene often come in pairs on each chromosome. When genes express themselves, they express themselves phenotypically (the physical traits and characteristics of an individual/ organism that’s a result of their genetic makeup). So an individual’s phenotype is a result of the combination of gene alleles they have.
Think of someone’s height. For a gene that determines height, there may be different alleles. One allele may result in being short, and another may result in being over 6’ foot. Someone’s full height would be a result of how many alleles they have and how they interact with the environment. It’s important to note that the result of a phenotype are a result of the genes and the environment. The characteristics associated with a particular allele can sometimes be dominant or recessive. A gene is said to be dominant when the dominant alleles show their effect phenotypically even when there is only one copy in the genome, and a gene is said to be recessive when the effects of an allele only show its effects when when there are two copies in the genome.
A gene can express itself differently in different environments. This is referred to as the gene-environment expression. Despite the interaction of genes and the environment, it is wrong to say that environmental manipulation can make a gene express itself in any way due to the range of the gene’s expression. So if blacks are found to have a higher frequency of genes associated with lower intelligence, it’s wrong to say that changing the environment would make said gene(s) express itself in a way where it makes it control for higher intelligence. Along with the gene-environment expression comes the gene-environment correlation.
The gene-environment correlation means that genes shape the environment (DeFries, Loehlin, and Plomin, 1977). The gene-environment correlation can be divided into three types.
(i) A correlation is said to be active when a gene affects its environment via its phenotypic expression. Someone genetically predisposed to knowledge will seek environments that encourage the consumption of education.
(ii) A correlation is passive when the gene that causes it is not a correlate. High IQ parents making environments that encourage education (such as making their kids room filled with books on mathematics, science, etc.) is a passive/ gene environment.
(iii) The final one is reactive when the phenotypic expression of a gene causes others to modify their behavior towards the gene’s bearer. For example, a kid who may be predisposed to playing music will encourage his parents to buy him a guitar or a piano. This particular reaction will cause a feedback loop. The reaction caused by the phenotype changes the phenotype so that it elicits a further reaction. The feedback may be positive, amplifying the phenotype provoking its reaction, which encourages the kids musical curiosity. Or the reaction may be negative, which will diminish the phenotype. So someone predisposed to depression will shape their environment so it encourages depression, causing a feedback loop.
This is important to understand this for things like environmental influences if we treat them as phenotypic variables. If genes shape the environment and they cause a reactive correlation, then blacks are making their environments worse due to certain genetic pre-dispositions. To put this into perspective, think of the black-white crime gap in the United States (Rubinstein 2016).
If we have two neighborhoods, A and B, that are completely isolated in respects to it having humans living there, and then fill A with blacks and B with whites and assume that blacks are pre-disposed to violence and whites aren’t, then each neighborhood will turn on vastly different. Neighborhood A will be filled with violence and other problems, and neighborhood B will be kept tidy and have less problems. So most of the environmental influences discussed could be a result of the gene-environment correlation and the reactive correlation.
What’s worth mentioning are the criticisms that’t you can’t separate genes and the environment (e.g. Burt and Simons 2015). This is not necessarily true; it is true that nature vs. nurture is a false dichotomy as a gene needs an environment to express itself in, but it is wrong to say that you can’t separate genetic influences and the environments. In reality, you can separate genetic influences from the environment, as shown in Jensen (1973), Wright et al. (2015; 2017) and Sesardic (2005).
Now that this is out of the way, a discussion on the genetics of intelligence can be discussed.
The Heritability of IQ
It’s important to note what heritability actually is. Heritability is the variance of a trait that is attributable to genetics. As the Rushton and Jensen paper cited notes:
“Heritability refers to the genetic contribution to the individual differences (variance) in a particular group, not to the phenotype of a single individual. Heritability is not a constant that holds for all groups or in all environments. A heritability of 1.00 means all the observed differences in that group are due to genetic differences and not at all to their differences in the environment. A heritability of zero (0.00) means the converse. A heritability of 0.50 means the observed variation is equally the result of genetic and of environmental differences. The heritability of height in modern industrial populations, for example, is about 90%, which means that most of the differences in height among the individuals are due to their genetic differences.”
Since the racial groups in the U.S. share an equal environment, we can assume that the heritability of IQ is the same for all races in the U.S., respectively (some commentators argue that within-group heritability can not be extrapolated to between-group heritability, Levin (1997) responds to this point elegantly and so does Jensen ).
The heritability of IQ can be seen in the table below.
The Heritability of IQ
|0.86 (MZ), 0.55 (DZ)||Hunt (2010)|
|0.86||Panizzon et al. (2012)|
|0.8||Plomin and Deary (2015)|
|0.78 – 0.8.||Pederson et al. (1992)|
These high estimates may seem dubious as some organizations have claimed that the heritability of IQ is .5, not the high estimate of .8 (U.S. National Library of Medicine). It is true that the heritability of IQ is .5, but it increases with age due to something called the Wilson Effect. As you get older, your genes start playing a large role and the environmental effects start canceling out, as can be seen in the chart below from McGue et al. (1993).
The Wilson Effect is also further supported by the fact that the same genes that explain intelligence during childhood are also there during adulthood (Trzaskowski, Visscher, and Plomin, 2014).
Intelligence is also very stable throughout ones life. In a longitudinal study of English children, there was a correlation of 0.81 between intelligence at the age of 11 and on scores on national tests of educational achievement 5 years later (Deary et al. 2007). A single test of general intelligence taken at the age of 11 correlates highly with the results of an IQ test taken at 90 (Deary et al. 2013).
Something that should be mentioned quickly is the Burt study, which found a heritability estimate of 0.77. When it comes to Burt, Kamin (Princeton 2018) and Gould (1996) assert that Burt fabricated his figures on the heritability of intelligence. The main evidence of fraud has been the high concordance of Burt’s figures over time and the elusiveness of his co-authors. Joynson (1989), Fletcher (1991) and Rushton (2001) indicate that Burt’s figures were stable because of his re-use of the same data. While not the best science, it is certainly not fraud. Some witnesses also do remember Burt’s co-authors.
Turkheimer et al. (2003) finds the the heritability of IQ differs by social class – with the heritability of IQ being lower in lower SES positions (this theory is based off of the Scarr-Rowe hypothesis, essentially saying that genes play a smaller role in lower SES positions). Based off of this study, it’s natural to assume that since blacks are more likely to be in lower SES positions than their white counterparts, this means that genes should play a smaller role for them than whites in higher SES positions, an argument made by Gladwell (2007) when he said that
“it’s hopelessly naive to assume that the same rules apply to suburban, middle-class whites as apply to, say, urban, inner-city black families.”
The problem with this assertion is that Turkheimer et al. didn’t report their results based on race, rather it was for their entire sample. Therefore, the hypothesis that the heritability of IQ is near zero for blacks isn’t supported by Turkheimer et al.’s own paper.
Replication attempts of Turkheimer et al. have failed (refer to to Kirkegaard 2016 for more), but this could be attributed to different methods used (Nagoshi and Johnson 2005; Asbury, Wachs, and Plomin 2005; Bates et al. 2015; McGue et al. 2005). Of course, though, a meta-analysis by Tucker-Drob and Bates (2015) found no reliable overall effect for the Scarr-Rowe hypothesis. Other studies, like Fuerst (2014), found the heritability of IQ to be the same regardless of race. Figlio et al. (2017) didn’t find support for the Scarr-Rowe hypothesis, rather the contrary.
While we shouldn’t disregard the Turkheimer study, we also shouldn’t take it at face value. More research needed, of course.
Within and Between Group Heritability
It has been argued before that you can’t extrapolate within group heritability (WGH) for heritability between groups (BGH). Meaning that twin studies on the heritability of IQ, if done on whites only, just shows what the heritability of IQ is for whites rather than for blacks or Asians. Thus, you can’t extrapolate within group heritability to between group heritability, Indeed, this is a position that has been argued against the hereditarian position on race and intelligence. For example,
“the existence of signiﬁcant heritability for IQ within the populations that have been studied does not imply that average IQ differences between races are in whole or in any part due to genetic differences . . . Various writers – the most prominent being Arthur Jensen . . . – have taken the heritability of IQ to show that these differences must have a genetic base. No such conclusions follow” (Papineau 1982).
Richardson (1984) wrote:
“On the basis of evidence supporting a high heritability value within a subpopulation, Jensen infers that heritability will be (correspondingly?) high in the population as a whole, and that variation between groups has a (correspondingly?) high genetic basis . . . But there is no intrinsic connection between the magnitude of the heritability within groups and the magnitude of between-group differences” (Richardson 1984).
This line of argument is seen as “persuasive” and “compelling” (Longino 1990). This criticism was first said by Lewontin to Jensen, with an example using two identical seeds and different soils. The example goes that if you take two seeds from the same homogeneous sample and then planted them in two different soils (one rich in nutrients and the other poor), you will get a result of phenotypic differences within each of the two groups of plants that will be 100% heritable, but the differences between the two plants will be entirely environmental due to the different environments (essentially saying the heritability of x between the two plants, or in this case race, is 0).
There are three ways to argue within group heritability to between group heritability, as proposed by Sesardic (2005): (H1) High WGH means high BTW; (H2) High WGH, by itself and nothing more, entails high BGH; (H3) High WGH, together with other clusters of data, establishes a non-zero BGH.
Lewontin and others have attacked H1 and H2, while the hereditarian position primarily rests on H3. While resting their arguments on H1 and H2, critics have accused hereditarians, primarily Jensen, of an “elementary error” (Lewontin 1975). Gould (1977) also attributes H1 to Jensen. The fact that Lewontin and others (e.g. Block 1995; Rose et al. 1984) think that the seed example disproves a connection between BWG and WGH shows that their attack is on H1 and not H3.
Lewontin (1973) states that
“The error of confusing the heritability within a population with the causes of differences between populations was clearly made by Arthur Jensen in his famous article in the Harvard Educational Review, when he tried to infer from heritability studies within the American white population the causes of differences between races. This elementary blunder would not be tolerated in a freshman class in statistics or genetics. We may well wonder how it came to be made by a professor!”
The fault in the argument proposed on top for BGH not being the same as WGH is that Jensen never intended to even defend H1 or H2. If he did, then it would make no sense as to why he tried to rule of environmental factors like SES, birth order, etc. Jensen has also claimed that Lewontin attacked a straw man:
The main thrust of Lewontin’s argument, as he sees it, actually attacks only a straw man set up by himself: the notion that heritability of a trait within a population does not prove that genetic factors are involved in the mean difference between two different populations on the same trait. I agree. But nowhere in my Harvard Educational Review discussion of race differences do I propose this line of reasoning, nor have I done so in any other writings (Jensen 1976b).
Jensen has also stated:
“So all we are left with are various lines of evidence, no one of which is deﬁnitive alone, but which, viewed all together, make it a not unreasonable hypothesis that genetic factors are strongly implicated in the average Negro–white intelligence difference” (Jensen 1969).
This statement by Jensen doesn’t fit the criticisms made by others against him, especially when they rest on H1 and H2. Jensen clearly states that high BGH doesn’t automatically mean high WGH, rather that the inability of environmental factors to explain the totality of the black-white IQ gap could mean that genes play a larger role than the environment.
Thus, attacks on H1 and H2 can not be the same for H3. As has been cited in the evidence above, environmental factors have been shown to not explain the totality of the black-white IQ gap. In order for the environment to produce a 15 point IQ gap, as in the case of the black-white IQ gap, blacks would need to have an environment that is 90% – 98% worse than that of white Americans to produce the 1 SD gap we see in IQ between blacks and whites today (Jensen 1972). There’s no reason to think that the Wilson Effect is for a single race, making it plausible that the black-white IQ gap is due to genes and that the heritability of IQ for both groups is 80%.
The (safe) hereditarian hypothesis argues that heritability of IQ in the U.S. is at .8, any person who argues that it’s .8 around the world is wrong due to different environments (readers interested in learning about the heritability of IQ in different countries are referred to Lynn ). Now we will move onto race differences in gene alleles, morphological and other features that lead to differences in IQ.
The Neurology of Intelligence
Brain Size and Neurology
Race differences in brain size, while often used a joke of sorts against the concept of race (Contra 2017), do exist and they correlate with IQ. A table adopted from Lynn can be seen below on the brain size of Europeans and Africans.
Brain Size of Europeans
|World||mf||52||1,401||Morton, after Gould (1981)|
|USA||mf||811||1,370||Ho et al. (1980)|
|World||mf||–||1,369||Smith and Beals (1990)|
|World||m||1,476||1,476||[Reported in Lynn]|
Brain Size of Africans (with B-W Difference)
|Location:||Sex:||N:||Cranial Capacity:||B-W Difference:||Source:|
|World||mf||29||1,360||41||Morton, after Gould (1981)|
|USA||mf||450||1,267||103||Ho et al. (1980)|
|World||mf||–||1,283||86||Smith and Beal |
|World||m||–||1,416||60||[Reported in Lynn]|
Brain size in general has a heritability estimate of 87% (Peper et al. 2007), and race differences in brain size can be seen at a young age, too. Schultz (1922) found that white fetuses had a larger cranial capacity than black fetuses, and Kirkegaard (2018) noted that race differences in brain size could be see in children as young as 3. Similarly, Rushton (1997) found that, at birth, Asians had larger brains than whites, and whites had larger brains than blacks. Fitting the hereditarian position, mixed black-white individuals actually have brain sizes that fell between that of blacks and whites (Pearl, 1934; Bean, 1906).
A common objection to race differences in brain size is to cite Boas (1912), the founder of American Anthropology. Boas looked at European immigrant children and their parents: he found that the brain size of kids matched, not their parents, but non-immigrant kids. It turns on that Boas didn’t even control for brain size differences with the function of age. Sparks and Jantz (2002) re-looked at Boas’s findings and controlled for the age difference between the kids and their parents. The brain size of the parents and their kids didn’t differ, and the statistical findings were negligible. The brain size of the immigrant kids and non-immigrant kids also did differ, contrary to what Boas claimed. Boas and his study are still cited in the most updated anthropology textbook I could find (e.g. Larsen, 2017).
These findings could be attributed to the fact that poverty may decrease brain size, a position akin to Noble et al. (2015). They found that poor children scored worse when compared to rich children, but the gap shrunk – but didn’t reduce completely – when controlling for brain size. These results suggest that nutrition and the prenatal environment cause poor children to have poor cognitive skills than rich children. The researchers didn’t falsify the more likely explanation between brain size and SES. Namely that the parents with larger brains tend to make more money, and then they in turn pass on the genes that control for brain size onto their children which makes them more intelligent. Instead, the authors gave possible hypothesis for their findings with insufficient support. When it comes to brain size and IQ, it’s been empirically verified that brain size correlates with IQ.
Three meta-analysis on over 100 studies have found that people with bigger brains are more likely to score higher on IQ test; they produced correlations ranging from .24 to .40. (McDaniel 2005; Rushton and Ankey 2009; Pietschnig et al. 2015).
McDaniel (2005) found a correlation of .21 – .41 in his meta-analysis for brain size and IQ. Using MRI scans on brain size and IQ on college students after body size was controlled for, Willerman et al. (1991) found an r of .35 between brain size and IQ. Andreason et al. (1993) obtained r’s ranging from .26 to .56 between IQ and the size of specific brain structures, and an r of .38 between IQ and full-scale IQ and grey matter volume. Raz et al. (1993) and Wickett, Vernon, and Lee (1996) got replications ranging from .41, .47 – .49. Overall, the data does suggest a significant correlation ( .21 <r< .56). Thus, bigger brain volume, is associated with higher intelligence.
Andreason et al. (1993) suggest that the remainder of variance is due to
“aspects of brain structures that reflect quality rather than quantity of brain tissue: complexity of circuitry, dendritic expansion, number of synapses [or neurotransmitter] efficiency.”
IQ scores correlate with intracranial, cerebral, temporal lobe, hippocampal, and cerebellar volumes (Andreason et al. 1993). These together encompass almost the entire brain. Using Voxel-based morphometry (VBM), a neuroimaging analysis technique that allows you to estimate the focal differences in brain structure, it allows researchers to see if whether any such areas are clustered together or distributed throughout the brain. Using VBM on brain data has shown us that there is a positive correlation between intelligence and cortical thickness that are located primarily in areas of the frontal and temporal lobes (Hulshoff Pol et al., 2006; Narr et al., 2007; Choi et al., 2008; Karama et al., 2009).
It may also be that the brains of intelligent people function differently from those who are in the lower IQ range, and that there maybe brain differences, too. Rushton (1997) looked at brain neurons (in millions of excess neurons) by race, and found that blacks, on average, have less brain neurons than whites and Asians:
Brain Neurons by Race
Researchers at the Free University Amsterdam and Amsterdam University Medical Center have found that smarter people have thicker neurons than do people of lower intelligence. People with higher IQs had more complex dendrites and faster action potentials which can process more information coming in and can pass more detailed information onto other neurons (Goriounova et al. 2018). Colom et al. (2010) also note neurological variables that correlate with intelligence.
It may also be that there are neurological differences between blacks and whites in processing brain glucose metabolism. Brain glucose metabolism is needed when the brains needs fuel for energy generation (Rao, Oz, Seaquist 2006). More intelligent people’s brain consumes less energy than those who are less intelligent (Fidelman 1992).
While the researchers for these studies didn’t look to see if the races differ in neurological features, there’s no reason to think that they don’t differ. The question shouldn’t be: Why would the race differ in neurological features, rather Why wouldn’t they?
The genes associated with brain development may also not be equal across racial lines. Wu and Zhang (2011) found that the races differ in genes associated with brain development at a higher frequency than they do in genes that control for skin color.
Something correlated with neurology and IQ is reaction time. Reaction time simply consists of the speed of a reaction to a simple stimulus, like clicking a light when it turns on. A simple understanding of this can be seen in games – ones for children and ones found within application stores. Any game that requires you to react to something, for example clicking a tile when it lights up or placing your foot on top of a light when it lights up, is an example of a stimulus that requires a quick reaction. Jensen (1998) and Deary (2000) have found that the correlation between reaction time and intelligence is at 0.2 to 0.3. If the races differ in reaction time, then it can show race differences in brain efficiency which some portions of the black-white IQ gap.
Jensen (1993) looked at 585 Europeans and 235 African 10 year old children in the United States whose IQ’s were assessed by using Raven’s Progressive Matrices. Within the study, Jensen used computer-controlled apparatuses so that no human error can potentially fudge the data. Three different kinds of reaction times were measured: Simple reaction time (SRT), which consists of reactions to a single light; Choice reaction time (CST), which consists of reaction to one of eight lights; and finally Odd-man reaction time (OMRT), which consists of reaction to one of three lights that was farthest from the other two. Each of the three reaction times were measured for four components consisting of the reaction time proper (the decision time), the movement time (time it takes to move the finger to the button), and the standard deviations of the reaction and movement time. A similar study was also carried out by Lynn and Holmshaw (1990). The results of both studies can be seen in the table below.
Correlations between RT and IQ are all positive, and all the data in 16 of the 24 correlations are statistically significant but the correlations are also low. Reaction times in row 1, 5, and 9 are faster in whites than in blacks except for CRT in Jensen. Simple movement time shows no race differences, but Africans are significantly faster than whites in both CMT and in OMMT in the Lynn and Holmshaw data. The SD’s are consistently greater in blacks in the Lynn and Holmshaw data than in Jensen’s data. In Lynn and Holmshaw’s data, the mean difference of the 6 reaction times and SD’s between blacks and whites amount to only 0.67j as compared to a 2.5d difference in IQ. The higher a persons IQ, the lower and less variable their reaction time.
This means that approximately a quarter of the black-white IQ gap may be attributable to race differences in the speed of neurological processing. Reaction time also has a heritability of 50% (Deary 2000).
Intelligence and Genes
Besides neurological features, the races also differ in IQ related genes. At the time when Jensen, Rushton and Herrnstein were alive, race differences in IQ related genes were not known. Now with the advancements made in science, their hypothesis on a genetic influence on IQ can be confirmed.
Based off of polls done on intelligence researchers, researchers do seem to acknowledge a moderate effect when it comes to genes and the black-white IQ gap.
Rindermann, Becker, and Boyle (2016) asked intelligence researchers: “What are the sources of U.S. Black-White Differences in IQ?”
|x% due to genes:||Expert Opinion:|
Synderman and Rothman (1987) also asked 661 experts on the source of the black-white IQ gap.
|Insufficient Evidence to Answer||24%|
|Environmental and Genetic||45%|
So from this, we can tell that researchers – not only agree that there are race differences IQ – but that there is a genetic component to it.
Race Differences in IQ Related Genes
Piffer (2015) used data from the 1000 Genomes project and world IQ data from Lynn and Vanhanen (2012). The goal was to see how 9 random SNPs associated with IQ differed between populations. The results can be seen in the table below.
Frequency of Genes Associated with Intelligence by Race
|Gene Variant||Association with IQ||East Asians||White||Black|
Positive (+) = High IQ Related Gene
Negative (-) = Low IQ Related Gene
From this, we can see that Asians and Europeans have a higher frequency of genes associated with higher IQ, while blacks have a higher frequency of genes associated with lower IQ’s. It is true that IQ is a highly polygenic trait (meaning that it’s influenced by many genes), and that these 9 SNPs aren’t all the genes that control for IQ, but if these genes differ by race – then there should be no reason to think that all the other genes that control for high and low IQ won’t differ by racial populations.
According to the hereditarian hypothesis, black admixture will lower IQ while European admixture would actually increase it. Luckily, we have some evidence showing that this maybe the case.
It maybe that American blacks have a higher IQ than pure blooded blacks due to European admixture. Reed (1971) and Chakraborty, Kamboh, Nwamko, and Ferrell (1992) estimate that American blacks have 25% white ancestry. Lynn (2011) estimates that IQ would increase 0.2 points for every 1% of European genes, and that American blacks with 50% European admixture would have an IQ of 90. As the European admixture increases, for example to 75%, it would lead to an IQ of 95.
In Scarr and Weinberg (1983) , they found that mixed black/ white kids had an IQ between those who were black and those who were white. This is well expected and falls inline with the hereditarian hypothesis. Shuey (1966) found that in 16/18 studies were skin color could be used as a proxy for amount of admixture, blacks with lighter skin had a higher IQ than those with darker skin color (although, the correlation was small at only .10).
A counter to the admixture hypothesis is that Scarr et al. (1977), Loehlin, Vanderberg, and Osborne (1973), and Neisser et al. (1996) found no link between skin color and IQ, but these studies don’t actually test this hypothesis (Reed, 1997).
From all this evidence using biological traits and how they differ by race, it’s enough justification to defend the position that race is tied to IQ. No matter how much someone wants to deny it, race is tied to IQ if morphological and genetic features that influence IQ differ by groups.
No single line of evidence can suggest that environmental factors have failed to explain the totality of race differences in intelligence and that genetics are the primary reason as to why the races differ, but when you put all the evidence together the answer becomes clear. The final passage that will be argued is that IQ matters for society, answering the question on why race differences in IQ matter at all.
Why IQ Matters
A big reason as to why IQ matters is because racial inequality in the U.S. can partially be explained by IQ. I ask the read to think of some racial inequalities in the U.S. that differ by race. Some things that may come up is the black-white income gap, graduation gap and incarceration gap. When IQ is controlled for, these disparities shrink.
Herrnstein and Murray found that controlling for IQ shows that whites and Latino’s are equally likely to graduate from college, and it’s higher for blacks:
Controlling for IQ makes the black-white incarceration gap go away almost entirely.
The black-white income gap goes from thousands to a couple of hundreds after IQ is controlled for.
The chances of being in poverty for each race is reduced when IQ is controlled for.
These are just a small set of examples. From this, we can see that racial inequality in the U.S. can be explained by race differences in IQ. Other reasons as to why IQ matters come in things like the demographics of a country and the values it espouses.
IQ and National Values
The IQ of a nation matters since certain IQ’s are related to certain values, and some populations have lower IQ’s. Many of these values are good for society, but a decrease in a nations IQ (for example, letting in low IQ populations) can cause irreversible damages.
Kanazawa (2009) found that even after controlling for economic development, education and history of communism, high IQ societies were more liberal, less religious, and more monogamous.
Average intelligence in societies also increases the highest marginal tax rate; this means that people are more willing to contribute to the private resources for the welfare of people genetically unrelated to them. Each increase in IQ increases the highest marginal tax rate by more than half of a percentage point.
These are some reasons as to why IQ matters. Once we acknowledge that the races differ in IQ, we can see why some races fail and others prosper in life. The egalitarian mindset that racial inequality exists because of racism can be thrown away, and that biological differences seem to be the primary suspect.
Towards A Conclusion
From all of this, we can see that race differences do exist and that they matter. The failure of environmental positions to explain race differences in intelligence have shown that the hereditarian hypothesis is true. Some maybe quick to appeal to individualism, but this line of thinking ignores that fact that groups are made up of individuals, and most people, at least in the context of IQ, don’t differ much from their populations average (Resnick 2017).
IQ measures something that differs by population, and the correlation between IQ and many variables can suggest future policy implications. Race differences in intelligence do exist, and the evidence – especially when it’s all seen together – suggests that genes are responsible for more than half of the race differences in IQ.