An important aspect of the newly emerging folk psychology, if that would be the right term, in America is the notion of a racial ladder: one which extends from ‘Asians’ at the top to ‘whites’ in the middle and ‘blacks’ at the bottom. Sometimes the argument is based on studies of IQ scores, other times the argument is based on anecdotal experiences about American colleges and universities, or about supposed Asian special aptitude for mathematics and natural sciences. These ideas are fairly widespread, and not limited purely to bar talk: even such supposedly hip ‘leftist’ magazines as Slate have contributed to it, and they are fairly prevalent at American universities too, including with the admissions departments.(1)(2) As in the Slate article, often these views are defended from accusations of racism by appealing to the fact that the folk psychology does not place whites at the top. Needless to say, this is not much of an argument, but it is commonly believed to be justified. Because of the nefarious nature of this notion, it is time to confront it with a basic understanding of statistics, which will quickly dispel most ideas of this kind.
First there is the IQ argument. Based on the research of such ‘scientific’ racialists as the Canadian J.P. Rushton, there have been suggestions that the IQ level of Asians is higher than that of whites, and of whites higher than that of blacks – and not just in the United States, but even worldwide. As Saletan’s article notes based on this research:
Among white Americans, the average IQ, as of a decade or so ago, was 103. Among Asian-Americans, it was 106. Among Jewish Americans, it was 113. Among Latino Americans, it was 89. Among African-Americans, it was 85. Around the world, studies find the same general pattern: whites 100, East Asians 106, sub-Sarahan Africans 70. One IQ table shows 113 in Hong Kong, 110 in Japan, and 100 in Britain. White populations in Australia, Canada, Europe, New Zealand, South Africa, and the United States score closer to one another than to the worldwide black average. It’s been that way for at least a century.
This study also proposes that the same racial ladder holds in terms of brain size, with again Asians having the largest brains on average, then whites, and then blacks. Saletan of course warns that one can’t deduce the intelligence of any given person from an average, and there is little suggestion that Saletan himself, or other people who reason this way, are consciously intending to make political divisions based on race. But it is inevitable that such claims, if unchallenged and widely accepted, are going to lead to selection on the basis of race anyway. This is a mere matter of rationality in the most orthodox economic sense: even if you can’t tell from a given black individual whether they are smarter than a white one, while hiring it would still be rational to pick the white candidate because the expected intelligence value would be higher. The claim that since averages don’t tell anything about individuals, the implications are not politically racist or likely to lead to it, is quite clearly intellectually dishonest. Universities too would be wise to credit Asian applicants higher in their internal selection, all else being equal, on the basis of such claims; if they were true.
Fortunately, there is not much reason to believe that they are. There is a number of important reasons why IQ scores of this sort, even if we completely accept the test results and assume that everything is done aboveboard, cannot be used to make such racial distinctions. First there is the Flynn effect: studies by a certain James Flynn have shown that over time, the average IQ score has gone up strongly in Western countries. Because IQ scores are intended to be a measure based on an average for any given population of 100, such increases across the board are incompatible with IQ being a useful measure of intelligence, or even of mental retardation, which it was originally designed to be a measure of. After all, Flynn’s studies pointed out that given the rate of increase of average IQ by 0.3 a year, tracing this increase backwards “the Flynn effect puts the average I.Q.s of the schoolchildren of 1900 at around 70, which is to suggest, bizarrely, that a century ago the United States was populated largely by people who today would be considered mentally retarded.”(5) Needless to say, that just isn’t very plausible. As a result, IQ scores are recalibrated to create a new average of 100 every few decades. This means its value for comparing relative score results improves whenever calibration occurs, but it is inherently incompatible with any ideas of racial heredity of IQ in a meaningful way. After all, if in a few generations the score can go up from ‘mentally retarded’ to ‘a standard deviation above average’, this is an increase much too fast to be explainable in terms of evolutionary adaptations.
But that isn’t all: the actual IQ results comparing the racial categories are incorrect in the first place, because very often the population compared is not at all equal. Again, it is important to remember that an IQ of 100 is always the average for any given population, so the test population chosen matters enormously. As Richard Dawkins has pointed out, one can safely assume he would have a quite above 100 IQ score when compared to the general population of Britain; but when compared to a population of Oxford University lecturers, his score might not be quite as impressive.(6) The fact that these scores are expressed in absolute numbers, 100, 120, and so on, falsely give the impression of some sort of absolute standard rather than the relative measure for a population of mentally handicapped it was intended to be. Dawkins quite rightly points to the meaning of the old joke about a politician complaining that education is so bad these days that half the population now has an IQ of under 100: it is a mere necessity of statistics. Now the same James Flynn, a social scientist from New Zealand, compared the studies about relative racial IQ performance that another ‘scientific racialist’, Richard Lynn, had done. As Malcolm Gladwell describes it:
Here was a question tailor-made for James Flynn’s accounting skills. He looked first at Lynn’s data, and realized that the comparison was skewed. Lynn was comparing American I.Q. estimates based on a representative sample of schoolchildren with Japanese estimates based on an upper-income, heavily urban sample. Recalculated, the Japanese average came in not at 106.6 but at 99.2. Then Flynn turned his attention to the Chinese-American estimates. They turned out to be based on a 1975 study in San Francisco’s Chinatown using something called the Lorge-Thorndike Intelligence Test. But the Lorge-Thorndike test was normed in the nineteen-fifties. For children in the nineteen-seventies, it would have been a piece of cake. When the Chinese-American scores were reassessed using up-to-date intelligence metrics, Flynn found, they came in at 97 verbal and 100 nonverbal. Chinese-Americans had slightly lower I.Q.s than white Americans.
A third issue with IQ based on heredity is the problem of genes. Because racial categories are based on superficial physical phenomena, people often forget that for any given black person, that individual is likely to have a very significant amount of ‘white’ genes. Nowadays, such matters can be tested, and the results argue completely against the racial ladder idea:
Most tellingly, blood-typing tests have been used to assess the degree to which black individuals have European genes. The blood group assays show no association between degree of European heritage and I.Q. Similarly, the blood groups most closely associated with high intellectual performance among blacks are no more European in origin than other blood groups.
Aside from this, the relation between IQ testing and the assumption of a “general intelligence”, often called ‘g’ or the ‘g factor’, is dubious altogether. As Stephen Jay Gould pointed out in his classic The Mismeasure of Man, there is no particular reason to assume that such a “general intelligence” actually exists, and the statistical ‘g correlation’ may well be an artifact of the manner in which the results from fairly different testing subjects in the common IQ tests are combined.(9) The IQ tests are also widely suspected of being biased in the kind of knowledge and skills that are asked, based often on assumptions of general knowledge that prevail among the majority of Western populations, which might well not apply to the ‘racial’ minorities in such populations. As Nisbett’s article points out:
Important recent psychological research helps to pinpoint just what factors shape differences in I.Q. scores. Joseph Fagan of Case Western Reserve University and Cynthia Holland of Cuyahoga Community College tested blacks and whites on their knowledge of, and their ability to learn and reason with, words and concepts. The whites had substantially more knowledge of the various words and concepts, but when participants were tested on their ability to learn new words, either from dictionary definitions or by learning their meaning in context, the blacks did just as well as the whites.
Given this fact, the purported ability of IQ to predict future performance in society, supposedly based on racial ‘general intelligence’, may well be based on the statistical artifact that something which tests the degree of ‘integration’ with the racial majority of society is also likely to predict the degree of success in terms of such integration in that society. Similar objections as the above two pertain to the argument from brain size; absolute brain size is a very poor predictor of intelligence when a comparison is made between different species. Brain size as a proportion of total mass is a better comparison.(11) Here too Asians are said to score better than whites and whites better than blacks, despite the general weight and height distribution being rather the reverse; but the problem with this is that men also tend to have larger brains on average than women do, even relative to mass, and yet they both score equally on average at cognitive ability tests. Also, the manner of discounting weight differences when measuring the ‘encephalization quotient’ is disputed, and results vary wildly.(12)
Finally, there are good scientific and historical reasons to doubt the meaning of racial categories in the first place. We know that historically the choice in which physical characteristics make one part of which race has been highly variable. In American history, such peoples as Finns, Irish and Italians have in the past been considered not to belong to the white race, even though now they are considered just as ‘obviously’ to belong to it.(13) This casts enormous doubt on the meaning of those superficial morphological characteristics usually considered to be ‘evident’ indicators of the racial label. When one compares different groups of humans as a “geographically circumscribed, sharply differentiated population”, the common evolutionary biological definition of a subspecies, some 25-30% of all genetic differentiation between the individuals within these different groups has to be explained by group differences rather than individual genetic differentiation. When this test was applied to humans (based on 16 different populations from all inhabited continents), however, it found that 84.4% of all genetic differences are explained by individual differentiation and only the remaining 15.6% by group differences. Compared to other wide-spread mammal species, this is particularly low.(14) As a result, one cannot scientifically defend the existence of identifiable races as subspecies of humans.(15) Add to this the fact that the ‘holy trinity’ of races, white, black and Asian, are largely American in origin and not at all the racial classification commonly used in folk psychology in places such as Brazil or Haiti, and the ground for racial classifications of intelligence becomes shaky indeed.
What then can explain the folk psychology in America about the different racial performances at universities, including the perception that Asians outperform the rest? Some statistics can clear this up. Since we have shown that racial categories are made up by humans, and have no systematic biological basis, we can see if an assumption of overall equal average intelligence can sustain such results (arguendo, we will assume also that there is such a thing as one general intelligence). Given the large quantities of population involved and the assumption of equal average intelligence, we can assume that intelligence among the different three racial groups is distributed according to normal distribution. Now essential in this context is to note that the racial classification has for historical reasons arisen in such a manner that the number people bearing the label ‘Asian’ is much larger than that of the other two races. Depending on how the label is applied, fully half the global population can be classified as ‘Asian’ in this manner. Given this fact and normal distribution, we would then expect the result to be that the smartest people on the planet are racially ‘Asian’, and the dumbest people on the planet are also racially ‘Asian’. This simply follows from a hypothetical comparison of the three bell curves: ‘Asians’ being in absolute amounts more numerous, their bell curve would be wider on both ends as well as higher in terms of absolute population at each point in the curve. Now this result has to be tempered in the case of American universities by a number of factors, such as the ‘Asian’ population of the United States (which is relatively small), the degree to which each race is capable of equally accessing a university given equal intelligence levels, and so forth. Also important would be the relative degree of emigration for the purposes of study in the United States on the part of Asians.
In California, for example, there have been indications that since the abolition of ‘affirmative action’ by referendum in 1996, the number of Asians & Pacific Islanders (designated by census) at Californian high level universities has strongly increased. For example, the University of California at Berkeley has a 42% Asian undergrad population, on a general Asian population in California of slightly over 12%.(16) It seems safe then to assume that Asians are equal to whites in most (if not all) top universities in their ability to enter on the basis of intelligence, whereas blacks likely still are not, to say the least. Now if we assume that foreign applicants to American universities are generally drawn from the upper and upper middle classes from their areas of origin, and that the percentual amount of upper class people in Asia and in Africa are roughly similar, one would expect a numerically larger number of Asian foreign students. Combine this with the relatively larger Asian populations in states such as California and New York with high level universities, and one gets substantial ‘expected’ minorities at top level American universities. This in turn combined with the aforementioned normal distribution gives some indication of how statistics could explain this folk psychology idea, without there being any need for actually assuming any hereditary capabilities of intelligence on the part of actually existing ‘races’. Given paucity of data, the still small overall Asian population of the US, the limited amount of foreign students at American universities, differences in tuition fees and so forth, this statistical exercise by no means intends to prove the entirety of the perceived gap. But it shows to demonstrate that even if we assume such generalizations of folk psychology to be true, we do not and should not draw racial conclusions from such statistically dubious data.
1), 3) William Saletan, “Created Equal”. Slate (Nov. 18, 2007).
2), 16) Tim Krueger, “Asian: The New White”. The Cornell Daily Sun (Sept. 20, 2007).
4) See: Rushton & Jensen, “Thirty Years of Research on Race Differences in Cognitive Ability”. Psychology, Public Policy and Law, Vol. 11:2 (2005), p. 235-294.
5), 7) Malcolm Gladwell, “None of the Above”. The New Yorker (Dec. 17, 2007).
6) Richard Dawkins, The Ancestor’s Tale (London 2005), p. 87n.
8), 10) Richard Nisbett, “All Brains Are the Same Color”. New York Times (Dec. 9, 2007).
9) See: Stephen Jay Gould, The Mismeasure of Man (New York, NY 1996).
11) Dawkins, p. 81.
12) http://pubpages.unh.edu/~jel/brainIQ.html; Sarah Blaffer Hrdy and Richard Lewontin, “Women Versus the Biologists: An Exchange”. The New York Review of Books Volume 41, Number 13 (July 14, 1994).
13) See e.g. Noel Ignatiev, How the Irish Became White (New York, NY 1995); J. Sakai, Settlers (Chicago, IL 1983).
14) Alan Templeton, “The Genetic and Evolutionary Significance of Human Races”, in: Jefferson Fish (ed.), Race and Intelligence: Separating Science from Myth (Mahwah, NJ 2002), p. 36.
15) This means that ‘scientific racialist’ Arthur Jensen, a well-known proponent of the hereditary racial intelligence thesis, was quite wrong when he claimed human races equate to subspecies. See: Arthur Jensen, The ‘g’ Factor: The Science of Mental Ability (Westport, CT 1998).