There is a single, overriding principle in the current controversy about I.Q., accounting for its tone and quality far more than the actual pros and cons. It is that any criticism, no matter how false or incompetent, of I.Q. tests or of the argument that intelligence has a genetic component will be widely circulated, mostly believed, often quoted, and in due course, almost certainly published. If the incompetence or falsity of such a statement comes to light, those who embraced it will simply turn to the next document purporting to show the I.Q. to be meaningless, immoral, culturally biased, capitalist, racist, etc. For many people, these episodes appear to generate no cumulative skepticism about the criticism. If you possess a thoughtful cast of mind, you should be piqued by this curious phenomenon, for it is a rare specimen of emotion submerging reason, of wish outweighing fact. Later, I will give some examples, drawn from last Tuesday's Crimson (November 20).
First, I should state my own views on the controversy, for they are routinely misstated by my critics. People, I note, vary in mental capacity. The variation is partly genetic, partly environmental. In addition, people obviously achieve different degrees of success (as society defines it, in terms of status, income, admiration, etc.) in their lives. Moreover, success often appears to demand at least a modicum of mental capacity. Consequently, the status differences among people will also be to a degree genetic. Finally, it follows that the more equally society distributes wholesome surroundings, education, and opportunities for social advancement, the more will social status depend on genetic differences, for the environmental differences will then have been minimized. I gloss over the details of the argument, so as to get on to other matters.
Note, however, that my basic argument might be valid in principle even if there were no such thing as an I.Q. test. For, all it really says is that people vary genetically and that genetic variations are sometimes socially important. What the I.Q. does for the argument is put some teeth into it, which is no doubt why it provokes such a furor. It has been shown that I.Q. scores reflect genetic differences between people (though not entirely), that they correlate with the ordinary meaning of intelligence (though far from perfectly), that they index a necessary (though not sufficient) ingredient in success for most occupations in a modern society. You may disbelieve each of those assertions, but if you do, it is only because you have fallen victim to the central principle of the controversy--that any criticism of the I.Q. test gains automatic credibility, independent of its merits.
Last Tuesday's Crimson (November 20) featured an article by a Dr. Clemens E. Benda, who was said to be an international authority on mental retardation and child development. He argued that I.Q. tests do not really measure intelligence because intelligence is not "linear;" that I.Q. scores are not fixed throughout life and are anyway meaningless for adults, and, furthermore, that many "enormous contributions" have been made by people who did poorly on I.Q. tests. He claimed to have found "embarrassing" misinformation in my writings; he said that the genetics of human intelligence is not scientifically established; he dismissed out of hand the biologically and socially significant possibility that husbands and wives are correlated in I.Q., and he was indignant about my conclusions about racial differences in I.Q. As my critics go, this is a fair sampling. It will not strike those who have been following the controversy in the popular press as unusual. Let us, therefore, look a bit more closely and gauge the caliber of the criticism.
Though he was indignant about my racial conclusions, he did not say what they were. His reason may have been that I do not draw any racial conclusions. In everything I have written on the subject, I have been at pains to state that the question of population differences in I.Q. remains open, and so I state here. We cannot exclude by fiat the possibility of inborn I.Q. differences between all sorts of human groups--men and women; Democrats and Republicans; English and Italian; black and white and yellow and red--simply because we hope there are no such differences. Instead, we should bear in mind that if such differences are shown to exist, they are bound to be differences on the average only, for all such broad human groups overlap fully in I.Q.
On the mating habits of people, the facts are clear, and clearly different from Dr. Benda's unusual impressions of the matter. Husbands and wives correlate in I.Q. by about .5, which is roughly the same as the value for brothers and sisters. As a result, there is likely to be a greater tendency for high or low I.Q.s to run in families than there would be if marriage were independent of I.Q.
On the estimates of I.Q.s genetic ingredient--they are firmly established and widely cited. I know of no published technical refutation, nor even any major criticism, by a quantitative geneticist, of I.Q.'s substantial heritability, though there are countless allusions to such refutations or criticisms in articles like Dr. Benda's. The technical literature contains a veritable chorus of agreement on the fact that the I.Q. is heritable--not as heritable, let us say, as physical height, but more heritable than any other broad human psychological trait yet measured. Without a doubt, the confusion about I.Q.'s heritability is the controversy's most regrettable distortion. Geneticists who, in the technical literature, acknowledge I.Q.'s large genetic component refrain from adding their findings or judgments to the public debate, no doubt for fear of having articles like Benda's written about them.
How about the I.Q. itself? Dr. Benda says, "Early I.Q. testing and its modifications by Stanford, Wexler, Terman et all are all based on the studies of Binet and Simon at the end of the last century." Stanford? Could this be Stanford University, where Lewis Terman was when he developed the "Stanford-Binet"? Wexler? Could this be David Wechsler, developer of the Wechsler-Bellevue, the Wechsler Adult Intelligence Scale, and other modern tests? (Note, incidentally, that this Wechsler has not heard that the testing of adults is meaningless.) Reading that sentence, I thought for a moment that the entire piece was a prank, but this is neither the time nor the topic for April Fools'. I quickly realized that the sentence was just another manifestation of the first principle of this controversy. Dr. Benda could have included Charles the Simple in his list of early contributors and it would have been printed--and, what's far worse, widely believed.
It is, to be sure, true that the I.Q. is not utterly fixed throughout one's life. In fact, I.Q. scores before about the age of five are not highly correlated with adult I.Q. With each passing year of childhood, the I.Q. converges more narrowly on the adult score. Precisely the same could be said about your height relative to your peers. A tall five-year-old may be somewhat more likely to grow up tall than a short five-year-old. However, the odds are still higher for a tall twelve-year-old. But who said any different? I know of no serious student of the subject who thinks I.Q. is fixed like a tattoo. The I.Q. score is actually more like the outcomes of a first-rate medical examination: considerably diagnostic and moderately predictive. And a recent test is always better than an old one. Many people's I.Q.s change by ten points or less after the age of eight; only a few people's change by as much as 20 points.
Serious students also know that no single scale can capture all of human intelligence. That is why testers work so hard learning and using multiple-factor analysis and other statistical techniques to parcel out the numerous (up to 120, some experts say) components of intelligence. Nevertheless, the one-dimensional I.Q. continues to be the best single predictor of academic performance and one of the best single predictors of social status. The I.Q. is essentially a combination of many of the dimensions of intelligence, just as your height is essentially a combination of the vertical lengths of your feet, your shins, your knees and so on up to your head.
Dr. Benda said that some people who had low I.Q.s made enormous contributions to society. I do not doubt it. The little Dutch boy who put his finger in the dike may not have been exceptionally bright, though he may well have been exceptionally brave. While perhaps interesting, such unusual cases are beside my point. My theory calls only for a substantial correlation between I.Q. and status, for which the evidence is overwhelming.
Is there, then, really a controversy on I.Q.? There is clearly a good deal of agitation, but is there counterevidence instead of emotion, fact instead of fantasy, reasoned dispute instead of vituperation? Not much, it seems. The proper message to extract from the ersatz controversy on I.Q. is that many people are not yet ready or willing to hear the news about human differences, because they fear it will be bad. It is unlikely to be as bad as they fear, but it does challenge common egalitarian visions of the good society. The real news--rather than what you've been reading in the papers--about human differences is that there are some, that some of them are important, and that any functioning society needs to husband scarce and irreplaceable resources, such as good sense.
Richard J. Herrnstein, professor of Psychology, is author of IQ and the Meritocracy.