A study at M.I.T., for example, showed that the average black student there had math SAT scores in the top 10 percent nationwide— and in the bottom 10 percent at M.I.T. Nearly one-fourth of these extraordinarily high-ranking black students failed to graduate from M.I.T.26 More generally, students with a given mathematics level succeeded in getting science, technology, engineering and mathematics degrees more often at academic institutions where the other students were at comparable academic levels.27
It has been much the same story in law schools. At the Georgetown University Law School, for example, the median test score of black students on the Law School Aptitude Test was at the 75th percentile— hardly “unqualified”— but that score was lower than the score of any white student admitted to this elite law school at the same time.28 Studies at a number of law schools indicate that black students admitted with lower qualifications than other students not only do less well academically while in law school but fail the bar examination more often than either the white students at their law school or black students with the same academic qualifications as themselves who attend law schools where the other students have academic qualifications similar to their own.29
In short, ignoring test scores and other academic qualifications when admitting minority students turns minority students with all the qualifications for success into artificially induced failures, by mismatching them with the institutions that admit them under lower standards.
ABSTRACT QUESTIONS
A common finding among groups with low mental test scores, in various countries around the world, has been an especial lack of interest and proficiency in answering abstract questions. A study in England, for example, showed that rural working class boys trailed their urban peers more on abstract questions than on other kinds of questions.30 In the Hebrides Islands off Scotland, where the average IQ of the Gaelic-speaking children was 85— the same as that among blacks in the United States— the Gaelic-speaking youngsters did well on informational items but trailed their English-speaking peers most on items involving such abstractions as time, logic, and other non-verbal factors.31 In Jamaica, where IQs averaged below normal, the lowest performance was on the least verbal test.32 A 1932 study of white children living in isolated mountain communities in the United States showed that they not only had low IQ scores over all, but were especially deficient on questions involving abstract comprehension.33
Indian children being tested in South Africa were likewise reported as showing a “lack of interest in non-verbal materials.”34 Lower class youngsters in Venezuela were described as “non-starters” on one of the well-known abstract tests used there.35 Inhabitants of the Hebrides likewise gave evidence of not being fully oriented toward such questions.36 Black American soldiers tested during the First World War tended to “lapse into inattention and almost into sleep” during abstract tests, according to observers.37
That black-white mental test score differences in America are likewise greatest on abstract questions38 is hardly surprising, in view of this common pattern among groups that score low in various countries around the world, regardless of the race of the particular group. But the fact that low-scoring groups tend to do their worst on abstract questions is also contrary to the claim made by some critics of mental tests that group differences in scores on these tests are due primarily to the words used in these tests or to the culturally loaded subjects in the questions. However, an interest in abstractions is itself something characteristic of particular cultures and not of others. When H.H. Goddard said of the immigrants he tested at Ellis Island that they “cannot deal with abstractions,”39 he overlooked the possibility that they had no real interest in abstractions.
Even if those who take mental tests try to do their best on abstract questions, as on other questions, a lifetime of disinterest in such things can mean that their best is not very good, even if that is not due to a lack of innate potential. If Asian American youngsters were to do their best playing basketball against black American youngsters on a given day, their best might not be nearly as good as the best of youngsters who had spent far more time on this activity before. Similarly if black youngsters try their best on a test measuring mental skills that they have not spent as much time developing as Asian youngsters have.
Neither genes nor a biased test is necessary to explain such results. If there were some group which assiduously pursued intellectual development and yet ended up with low IQs, the case for genetic determinism might be overwhelming. But there seems to be no such group anywhere.
If one chooses to call tests that require the mastery of abstractions culturally biased, because some cultures put more emphasis on abstractions than others do, that raises fundamental questions about what the tests are for. In a world where the ability to master abstractions is essential in mathematics, science and other endeavors, the measurement of that ability is not an arbitrary bias. A culture-free test might be appropriate in a culture-free society— but there are no such societies.
Nor is the importance of particular kinds of abilities constant over time, even in the same endeavors. Criteria that might have been suited to selecting individuals to be shepherds or farmers in centuries past may not be adequate for selecting individuals for a different range of occupations today— or even to selecting individuals to be shepherds or farmers today, in an age of scientific agriculture and scientific animal husbandry.
TEST SCORE DIFFERENCES
Whether or not whatever factors make for high or low mental test scores make these tests a good measure of innate mental potential, what matters from a practical standpoint is whether those factors are important in education, in the economy and in life. Disregarding test scores, in order to get a higher demographic “representation” of black students in colleges and universities, for example, has systematically mismatched these students with the particular institutions in which they have been enrolled.
When the top tier colleges and universities accept black students whose test scores are like those of students in the second tier of academic institutions, then those colleges and universities in the second tier, which now find themselves with a smaller pool of black applicants whose qualifications are suited to their institutions, are thus left to accept black students whose test scores are more like those of students in the third tier— and so on down the line. In short, mismatching at the top tier institutions has a domino effect across the field of academic institutions, leading to far higher rates of academic failure among black students than among other students.
A widely-praised book on the effects of affirmative action in college admissions— The Shape of the River by former college presidents William Bowen and Derek Bok— claimed to have refuted this mismatching hypothesis with data showing that black students “graduated at higher rates, the more selective the school that they attended” (emphasis in the original).40 But what would be relevant to testing the mismatching hypothesis is the difference in test scores between black and white students at the same institutions— and this difference has been less at Harvard (95 points on the combined SAT test scores) than at Duke (184 points) or Rice (271 points).41 Other data likewise indicate that black students graduate at a higher rate in colleges where their test scores are more similar to those of white students at the same institutions.42 As Bowen and Bok themselves say: “There has been a much more pronounced narrowing of the black-white gap in SAT scores among applicants to the most selective colleges.”43
That the high rate of college dropouts found among black students in general is not as great at institutions where the racial mental test score gap is not as great is a confirmation of the mismatching hypothesis that Bowen and Bok claim to have refuted. The fact that access to their raw data has been refused to others44 suggests that the great praise showered on their book in the media may reflect agreement with its message and its vision, rather than a critical examination of its evidence and reasoning.
Although most controversies about racial differences in intel
ligence focus on averages, such as those of IQ scores, what is also relevant is the range of these scores. As already noted, much of what was said in the early twentieth century seemed to indicate a belief that there was some ceiling to intelligence that was lower for some races than for others. This was another way in which Professor Jensen differed from early twentieth century believers in genetic determinism, since he acknowledged that “as far as we know, the full range of human talents is represented in all the major races of man.”45
As for that supposed lower intellectual ceiling, among the “beaten men from beaten races” disdained during the Progressive era were Jews— who were later in the forefront of those whose scientific work made the United States the first nuclear power, and Jews have been wholly disproportionately represented among Nobel laureates worldwide.46 International chess championships have been won by any number of members of another group of “beaten men from beaten races,” the Slavs— and the first human being to go into space was a Slav. The idea of an intellectual ceiling for particular races seems unsustainable, whatever might be said of intellectual averages.
There have been studies of blacks with IQs significantly above the national average, these studies having lower cutoff IQ scores of 120, 130, and 140.47 One of these studies turned up a nine-year-old girl “of apparently pure Negro stock” with an IQ of 143 on the Porteus mazes test, 180 on the Otis test and “approximately 200” on the Binet IQ test.48 If there is an intelligence ceiling for blacks, and it is up near an IQ of 200, then its practical significance would be wholly different from what was proclaimed by genetic determinists of the Progressive era, who depicted some races as being unfit for survival in any role above that of the proverbial “hewers of wood and drawers of water.” No one of course knows whether there is a racial ceiling on anyone’s IQ, much less what that ceiling might be.
Although the most common and most heated controversies about racial differences in IQ have centered on black and white Americans, the singling out of any given racial or ethnic group for comparison with the national average in any country creates an implication of uniqueness that is belied by empirical facts, since the national average itself is simply an amalgam of very different IQ levels among a variety of racial, social, regional and other groups.*
There is nothing unique about the average black American IQ of 85, compared to a national average of 100. At various times and places, other racial or social groups have had very similar IQs. Studies during the era of mass immigration to the United States in the early twentieth century often found immigrant children from various countries with average IQs in the 80s. A 1923 survey of studies of Italian American IQs, for example, found their average IQ to be 85 in one study, 84 in three studies, 83 in another study and 77.5 in still another study. A 1926 survey of American IQ studies found median IQs of 85.6 for Slovaks, 83 for Greeks, 85 for Poles, 78 for Spaniards, and 84 for Portuguese.49
Similar IQs in the 80s have been found among people living in the Hebrides Islands off Scotland and in white mountaineer communities in the United States in the 1930s50— both groups being of Nordic extraction, people who were supposed to be intellectually superior, according to Madison Grant and others. A 1962 study of the children of people from India tested in South Africa found them to have a mean IQ of 86.8, the same as that of African children there.51
Although mental test pioneer Carl Brigham wrote in 1923 that the Army mental tests during the First World War provided an “inventory” of “mental capacity” with “a scientific basis,”52 in 1930 he recanted his earlier view that low mental test scores among various immigrant groups in the United States reflected low innate intelligence. He belatedly pointed out in 1930 that many of the immigrant men tested by the Army during the First World War were raised in homes where the language spoken was not English. Although Brigham said in his 1923 book that he and other testers had “demonstrated the accuracy of the combined scale as a measure of the intelligence of the groups under consideration,”53 he said candidly in his 1930 article that his previous conclusions were— in his own words— “without foundation.”54
For blacks who took those same tests, their very low level of literacy at the time was likewise a factor to be considered, though few commentators took that into account. One sign of that low level of literacy among black soldiers taking the Army mental tests, and how that affected the results, was that black soldiers were more often able to answer some of the more difficult questions that did not require understanding the meaning of written words than they were able to answer much simpler questions that did.*
In addition, one section of one of the Army tests required information such as the color of sapphires, the location of Cornell University, the profession of Alfred Noyes and the city in which the Pierce Arrow automobile was manufactured.55 Why blacks would have had any reason to know any of these things at that time is a mystery— and why such questions could be considered measures of either black or white innate intelligence is an even bigger mystery. But here, as in other very different contexts, statistical data that seemed to fit prevailing preconceptions among intellectuals have been accepted and proclaimed, with little or no critical examination.
DURATION OF MENTAL TEST RESULTS
During the Progressive era, one of the strongest arguments advanced for eugenics was that the tendency of people with lower IQs to have larger families would, over time, lead to a decline in the national IQ. But the later research of Professor James R. Flynn showed that, in more than a dozen countries around the world, the average performance on IQ tests rose substantially— by one standard deviation or more— in a generation or two.56 Only the fact that IQ tests are repeatedly renormed, in order to keep the average IQ at its definitional level of 100, as the average number of questions answered correctly increased, had concealed this rise— and only the fact that Professor Flynn went back to the original raw scores revealed the facts which the renorming had concealed.
Much has been made of the fact that the average IQ among blacks has remained at about 85 over the generations, suggesting that the tests are measuring an unchanging genetic potential. But the apparent permanence of the performance of black Americans on IQ tests is an artifact of the renorming of those tests. The average number of questions answered correctly on IQ tests by blacks in 2002 would have given them an average IQ of 104 by the norms used in 1947–1948, which is to say, slightly higher than the average performance of Americans in general during the earlier period.57 In short, the performances of blacks on IQ tests have risen significantly over time, just as the performances of other people in the United States and in other countries have, even though the renorming of those tests concealed these changes. While the persistence of a gap between blacks and whites in America on IQ tests leads some to conclude that genetic differences are the reason, the large changes in IQ test performance by both black and white Americans, as well as by the populations of other whole nations around the world, undermine the notion that IQ tests measure an unchanging genetic potential.
The fervor and persistence of the racial IQ debate cannot be assumed to be a measure of its practical implications,* as distinguished from its ideological importance for competing social visions. As already noted, even the leading advocate of genetic theories of IQ differences, Professor Arthur R. Jensen, has seen scholastic achievement as amenable to different teaching methods and has treated IQ differences as an over-estimate of differences in intelligence between children from lower socioeconomic classes and others. Since concrete capabilities matter much more in the real world than do abstract potentialities, educational outcomes are the practical issue, however much this practical issue has been overshadowed by ideological issues.
The leading scholar in the opposing, environmentalist school of thought, Professor James R. Flynn, expressed the narrowness of the practical issues in 2008:
The race and IQ debate has raged for almost forty years. I have been entangled in it for thirty years. It has been a constant and unwelcome companio
n, rather like living with an uncongenial spouse from an arranged marriage. It has occupied the time of legions of scholars and laid waste acres of trees. Will we ever see the end of it? At least the debate is entering a new and more sophisticated stage. Given the relatively high values for black IQ in infancy and age 4, the focus should now be on whatever causes the decline of black IQ (compared to white) with age. If that can be settled, the main event will be over.58
Professor Flynn has argued that the culture in which most black Americans grow up has had a negative effect on their intellectual development. He pointed out that the offspring of black and white American soldiers, who fathered children with German women during the American occupation of Germany after the Second World War, had no such IQ differences as that among black and white children in the United States. Professor Flynn concluded that the reason for results being different in Germany was that the offspring of black soldiers in Germany “grew up in a nation with no black subculture.”59
There is other evidence that the black subculture has a negative effect on intellectual achievement. An empirical study published by the National Bureau of Economic Research found that “a higher percentage of Black schoolmates has a strong adverse effect on achievement of Blacks and, moreover, that the effects are highly concentrated in the upper half of the ability distribution.”60 In other words, brighter black students do not perform as well in settings where there are many other black students around them, contrary to the theory that what is needed in educational institutions is some larger “critical mass” of black students, in order to make them feel socially comfortable and thus able to do their best work academically. Yet the unsubstantiated “critical mass” theory has flourished from academic journals to Supreme Court briefs.61
Intellectuals and Race Page 9
