|| Human Nature Review ISSN 1476-1084 | Table of Contents | What's New | Search | Feedback | Daily News | Submit A Manuscript ||
PDF of this article Download Adobe Acrobat Reader Email the reviewer Reviewer's web site Author's web site Search for papers by Miele, F. Publisher's web site Send a response to this article Search the web for related items Contact the Editors
Human Nature Review 2003 Volume 3: 289-292 ( 9 May )
URL of this document http://human-nature.com/nibbs/03/jensen.html
Racial differences and Jensen on trial
A review of Intelligence, Race and Genetics - Conversations with Arthur R. Jensen
by Frank Miele
243 pp, Westview (2002)
Are there significant and genetic intelligence differences between human subpopulations? Or perhaps this question should be asked in a way that some (Westerners) hear this question anyway: Are there intelligence differences between Blacks and Whites? Perhaps almost equally important is the question of what should we think about scientists who claim that there are. I have to admit that without the following quotation from E. O. Wilson on the cover I would not have bought this book: “The work of an honest, courageous man interviewing an honest courageous man.”
But before reviewing the book I feel I must reveal some reasons why I haven’t, probably not even in my thoughts, given these subjects the fair trial that they perhaps deserve. It’s not that I would have been afraid of the consequences if the alleged differences had turned out to exist and then become widely known by the public. Racism never needed scientific proof, and many statements are accepted as fact without evidence, as long as they support the racist’s worldview. I admit that this might be just moral posing, but I think there have been more important scientific matters at hand.
First of all, we cannot know what the intelligence differences were between, for example, ancient reindeer keepers, pygmies or Papuan hunter gatherers. Nor can we say anything about the roles of genes or heritability in such hypothetical differences. The obvious genetic variation that exists inside populations cannot tell us if differences between populations are also genetic.
Secondly, intelligence tests, upon which these claims are based, try to measure mostly analytical skills instead of social, creative or practical skills. The fact is that there is no generally accepted definition of intelligence. And even if there were, there probably would not be a generally accepted and culturally unbiased test that could measure it.
In a sense, intelligence tests are like artificial intelligence: the “highest” brain functions, such as the ability to do algebra, are among the easiest things for machines to replicate - and tests to measure - whereas the deeply intuitive ways we make sense of each other are hardest to produce artificially. Also, which genes “determine” the test scores, and how, and why they do it (depending on the environment), can perhaps be discovered in theory but in practice the answers lie far in the future.
Thirdly, even though different environments made subpopulations look different, it doesn’t mean that they made people’s minds and brains different too. The evolution of intelligence, unlike the evolution of skin colour, is not due to exposure to UV-radiation. Also, for intelligence to evolve it makes no difference if one hunts seals, kangaroos or giraffes, or how one stays warm in the north and cool in the south. The most important selective pressure for human intelligence was probably other intelligent human beings: relatives, friends, competitors, mate candidates, etc. And these things haven’t changed much in human history.
Finally, it is known that in many cases differences within populations are greater than differences between populations. And besides, human populations are in a genetic continuum with no clear boundaries and with plenty of overlaps: “Despite perceptions to the contrary, geneticists have shown that humanity is not divided into distinct racial types” (Cosmides et al 2003).
We can imagine that these and some other minor points could have been taken into account, and that we still were to find an average genetic intelligence difference. So what? Do average differences matter? The answer depends crucially on the society we live in. Think of a situation where there are two applicants for a job, one white and one black. If the employer makes her choice using her “knowledge” that blacks are on average less intelligent than whites, the information would definitely be harmful. But in a more meritocratic society where an individuals status on the job market is based on her merits, the information about average differences becomes superfluous. Only a prejudiced and racist society has use for racial IQ-differences. Sadly, even if conscious attitudes toward different races are “politically correct”, racial stereotypes can still influence reactions that are subconscious (Wilson 2002, p. 188-194). It is possible that there will always be some major negative consequences if scientists indeed prove and publish findings on these differences, no matter how prepared or educated the public becomes.
The situation is somewhat similar, but not in the least as heated, with regard to differences between sexes. Let’s imagine I apply for a job consisting of throwing objects as far as possible, and along with me there is one woman applying. Now, the employer could use the fact that men are on average better throwers than women. But if my competitor happened to be Trine Hattestad or some other female Olympic javelin champion, the firm would definitely loose if they hired me. Indirectly, I too would lose, since the best forces in society would not be in proper use. (Naturally, it works both ways. A hospital hiring nurses could well discriminate male applicants because men on average are less capable of empathising than women.)
On the other hand, like individual differences, sex differences are a totally separate question from racial or population level differences. Selection pressures affecting each sex have been distinct through out our mammalian history.
With these attitudes and this amount of preparedness I thought I should give the issue as fair a trial as I could. I don’t know if reading Frank Miele interviewing Arthur Jensen was the best way to do it, but from Miele’s earlier interviews in Skeptic Magazine, I was convinced that he was more than qualified for an unbiased and thorough job. And besides, this kind of email interviewing, consisting of short and sober comments from both sides, made the book an easy read.
Arthur Jensen rose into public awareness in the 1960s, when he made three statements that even gave rise to a new concept, Jensenism. According to Miele it’s basic tenets are: “(1) the failure of compensatory education, (2) the evidence for a genetic basis to IQ, and (3) the likelihood of some genetic component to the Black-White IQ difference.” There hardly can be a matter as arousing as that! According to Jensen himself this is a fair and accurate definition of Jensenism, but only “so long as no one views it as some kind of dogma but simply conclusions I have reached for the time being based on my studies of these matters.”
Even though the book’s aim (or Jensen’s for that matter) supposedly is not to clean Jensen’s reputation as a scientifically motivated researcher, for me, up to a certain point, it had such an effect. Jensen acknowledges most of the four complaints or reservations I expressed earlier: for example, definition of race cannot be clear, and more importantly, environment (but not the shared home environment) affects intelligence too, but less than is generally thought. This latter point has been nicely proven by twin research. Jensen also readily admits that “at least half of the population variation in IQ is within families” so that “full siblings within the same family, on average, differ in IQ by about as much as the average difference between Blacks and Whites of the same social class.”
And, believe it or not, in his own words Jensen is not at all interested in races or politics, even though he does have some pretty clear ideas about politics: “Policy decisions… emerge from weighing scientific knowledge along with all the other, and at times conflicting, factors outside the province of science - ideals and goals; economic feasibility; traditional social, cultural, and religious values; and the prevailing consensus of public opinion at given time. Policy is often a matter of compromise.”
Anyway, he feels that he is not capable of thinking through what others see as political in his work. Whether this is extreme sincerity or naivety or anything like that is of course not the point. (But still, I can’t help thinking that some of the personalities that don’t feel the need to beware about speaking of intelligence differences are, for the same personality reasons, strongly attracted to these political minefields.) According to Jensen whether he is “right or wrong in any particular instance isn’t the really important thing. What is important is that scientific research on these matters should be encouraged and allowed to advance unfettered.”
Mostly I agree with this, but with a few reservations. First of all, since anyone can have a different idea of what intelligence is, scientists shouldn’t be talking about it at all. In a way, that is exactly what has been happening: at least Jensen’s tools and main subjects are mostly IQ (intelligence quotient, developed in 1904 to help teachers in Parisian schools) and g factor. Miele explains g factor, which is also the name of Jensen’s recent book, by a different factor, a hypothetical a factor (a standing for athletic ability):
We might start with a hunch that individuals who excel in one sport (say, the 40-yard dash) are more likely to perform better than average in other athletic events as well. They don’t have to be the best or even better than average in every athletic event. But those who do better in one event, we might predict, should be more likely to do better in most other events. Or, to put it the other way, those who do below average in some events should be more likely to do poorly in others as well. If so, we have evidence for a general factor of athletic ability… The theory of general mental ability (the g factor) is like our hunch about general athletic ability. It says that on average, those who do well on one mental test also tend to do well on other tests. The statistical methods we use to test the g factor (or our a-factor hunch) are correlation and factor analysis.
But after hearing from Jensen that g factor says nothing about the ability to recognise faces, for instance, or other supposedly universal and/or monomorphic human traits, my interest in the concept was lost. What is the use of a g factor if it doesn’t deal with human nature; human emotions or any kind of (probably modular?) psychological adaptations? Maybe it is time we had some scientists investigating the possible differences in these matters too. But as Plomin and Spinath (2002) said: “…if the construct of modularity were to be extended to consider individual differences, it would predict that genetic correlations among cognitive processes should be low, which would conflict with the high genetic correlations found among psychometric tests of cognitive abilities.”
There are at least three solutions to this paradox, the simplest being invoking a single and independent trait (neural processing speed, for example) that is responsible for g.
Anyway, it would be interesting to know if there really were genetic emotional differences, especially between populations that have been separated the longest possible time (for example, !Kung in Africa and Aboriginals in Australia). We know that the level of technology was practically the same all over the world as recently as 10 000 years ago, but still, it is possible that there has been enough time for natural and sexual selection to create variation - in musical talents or in seeing colours, for example, if not in emotional reactions. And if there are such differences, we could try to find out whether they are due to selection or just pure luck (because the diverging founder groups represented only a tiny fraction of the “original” population, thus eventually creating population level variation). I think that the IQ-score debate between many different cultural explanations (which are all briefly represented in the book) and different genetic explanations may prevent us from investigating these more interesting matters.
On the whole, though, the book made a difference. It didn’t convince me about the validity of the IQ-tests (actually, it didn’t even try), but it forced me to think these matters openly. Still, I don’t find them important, but surely important enough not to be left in the hands of racists.
Intelligence, Race and Genetics has lots of information value too: it explains nicely, for example the always difficult terms such as heritability and heredity. On some occasions Miele, the inquisitor sounds too provocative and even blunt, but I guess he and Jensen had agreed upon that beforehand. In the end, it will be the future generations of scientists and their empirical findings that judge the scientific and moral value of each scientist’s pursuits. My wild hunch, and this might be just wishful thinking, is that future generations will see Jensen’s work as a waste of his mathematical and statistical intelligence, equivalent of hiring me instead of Hattestad for the throwing job.
Cosmides, L., Tooby, J. & Kurzban, R. 2003. Perceptions of race. Trends in Cognitive Sciences. Vol. 7 No 4.
Plomin, R. & Spinath, F. 2002. Genetics and general cognitive ability (g). Trends in Cognitive Sciences. Vol. 6 No. 4.
Wilson, T. 2002. Strangers to Ourselves - Discovering the adaptive unconscious. Belknap Press of Harvard University Press.
Buy Intelligence, Race and Genetics - Conversations with Arthur R. Jensen from:
Amazon.co.uk Amazon.ca Amazon.de Amazon.co.jp Amazon.fr
© Osmo Tammisalo.
Osmo Tammisalo, M.Sc., is Chairman of the Darwin Society of Finland.
Tammisalo, O. (2003). Racial differences and Jensen on trial. Human Nature Review. 3: 289-292.