EC 2.9 (11-11-94)

EC 2.9 (11-11-94)

by Douglas Davis, Ph.D.

Denn die einen sind im Dunkeln
Und die andern sind im Licht.
Und man siehet die im Lichte
Die im Dunkeln sieht man nicht.
Bertoldt Brecht, Die Dreigroschenoper

For the first time my editors have suggested a theme for a column, prompted by the attention given a new book and its authors in the slick magazines and talk shows Webster's is destined to replace: Race and Intelligence. The request that I speak to this issue was motivated, I understand, both by my union membership in the sub-field of psychology that makes and interprets aptitude tests and by my interest in an instance of what might be called anthropological psychology, the role of cultural factors in most aspects of mental life. As a result I've been reading The Bell Curve by R.J. Herrnstein and Charles Murray and perusing some of the reviews and commentary generated by the book and by the second author's interviews.[1] I've also returned to some of the writing generated by the last iteration of the same argument, Arther Jensen's 1969 monograph in The Harvard Educational Review, R. J. Herrnstein's article in the Atlantic Monthly two years later, and Steven J. Gould's impassioned critique of the uses of mental measurement a decade later.[2] I'll probably be caught up with these matters for a couple of columns, and then we'll see whether they have turned out to be about Eros as well as Psyche.

The Bell Curve

Here's Herrnstein and Murray's argument:

1. Something quantifiable, consistent across many standardized tests, and stable throughout most of life is measured by intelligence tests, whether of the traditional individually-administered IQ type or the established group tests of scholastic ability (SAT, GRE). Such tests correlate substantially with each other, and it is plausible to think of them as primarily meaures of a very general aptitude for problem-solving and for performance in most of the tasks required for success in academic settings. These relationships have been demonstrated for at least 70 years, but they have become much stronger in the past 40, as more Americans have continued their education beyond high school and colleges have become more stratified in the aptitude levels of those attending. Some colleges and universities now have students whose average ability is below that of the nation as a whole, while the most selective draw their students from among the top few percent on the basis of both measured aptitude and pre-college success.

2. These measures of intelligence are strongly correlated not only with success in school but with success in American life, as measured by income, power, and prestige; and these correlations have become much stronger in the last 30 years, as economic pressures have created incentives for identifying, training, and rewarding exceptional ability. Graduates of top colleges and professional schools command much higher salaries and exert much more influence over society than do the products of average institutions. Those at the bottom of the measured ability distribution are vastly over-represented in the welfare roles and as targets of the criminal justice system.

3. Performance on aptitude tests, and all the correlated life outcomes, is strongly influenced by inheritance. Indeed, intellectual success is much better predicted for society as a whole from knowledge of close blood relatives' aptitude than from knowledge of social class or other broad details of one's environment. Children of mothers from the top fifth of the aptitude distribution are likely to become educated and to prosper, and children of those from the bottom fifth are likely to drop out, and to require expensive social support and control. In the most extreme case, identical twins' similarity in ability is far greater than that of genetically unrelated family members, even if the twins have been reared apart. The best twin studies[3] suggest that more than half of the variability in IQ is inherited in the middle class samples from which the twins have primarily come.

4. African-Americans score significantly lower than White Americans on standardized tests of intelligence. Indeed the difference found in many studies, fifteen IQ points or one full standard deviation, places only 16% of African-Americans above the White average.

5. Since the heritability of IQ is so high, and since attempts to remove aptitude differences by welfare and public intervention programs have not in general succeeded it is reasonable to assume that the observed aptitude difference between the two races is in large measure due to genetics.

Much of The Bell Curve's argument is made in terms of the fate during the past several decades of the very brightest Americans -- people like Herrnstein and Murray's Harvard colleagues (and perhaps you, gentle reader) -- and of the very dullest, whose likelihood of finishing high school is low and whose likelihood of being arrested, of drawing welfare, and of bearing illigitimate children is high. As society, moved by the invisible hand of profit-maximization, rewards the brighest, they and their above-average children will prosper. Unless society, moved by the invisible hand of loss-minimization, stops rewarding the dullest, they and their below-average children will persist. Current welfare and Aid to Dependent Children policies have the effect of increasing the fertility (number of births per woman) of those whose children will be most likely to fail. The result has been a USA ever more polarized between caste-like cognitive upper and lower classes.

"Intelligence" and "Race" in America

When I was a lad in Introductory Psychology at the University of Minnesota in the early sixties, the class of 2500 students met in Northrup Auditorium, the University's formal concert hall and home of the Minneapolis Symphony. The class was graded on the curve, based on expertly validated multiple-choice exams. And beautiful bell curves they were too. I have preserved none of my course notes and but few relics in memory of these sessions -- spotty images of the charismatic professors moving about the huge stage and gesturing to a slide on the overhead projector, perhaps a joke or two. One thing I do seem to recall came from the lecture on IQ and assessment. Dave LaBerge (or was it J. J. Jenkins?) announced the topic and (perhaps pausing slightly to allow Bics to approach notebooks) offered us the to-be-memorized definition of the key term: "Intelligence is what the intelligence test measures." It's funny, no? Maybe you had to be there in those heady Skinnerian behaviorist times. The definition, Herrnstein and Murray would want me to tell you, is misleading in its implication that this measurement is a tightly circular process, that that's all intelligence is. But the point's an important one, once we have noted a major conclusion of The Bell Curve and of the aptitude-testing literature cited therein -- whatever the intelligence test measures seems also to be an important part of what the course exam will measure, since grades correlate significantly with measured intelligence.

Intelligence is what the intelligence test measures,, and it seems to predict a great deal about how folks do in the US of A. "Race," by the same dust-bowl-empiricist logic is, IMHO, "what White and Black Americans differ in." Never mind that Americans are seldom good genetic examples of the breeding populations their great-great-great grand-mommies came from in Norway or Senegal, that they are usually race-mixed, and that you can be "Black" in the US with a majority of "White" blood. In the USA you don't know where your DNA was living 200 years ago, but you do know if you're White or Colored -- and your mommy knew, and your teachers all know, and the waitress at Denny's knows, and the people who draw the an-encephalic camels on the billboard outside your row-house know, and the cops who stop you and your friends when you drive up to a party in the Main Line know. "Race" is with the Black American from the moment of conception, and it takes cognitive and emotional shape from the moment of birth. It reduces the likelihood that a Black child will acquire the disposition to attend to formal instructions on which problem-solving is based, even as it reduces the likelihood there will be books in the home, adults with the time and inclination to read to the child, privacy from TV and siblings in which to read (or think), support for the long-term assumption that attention, problem-solving, and reading are satisfying and important. Under such circumstances the Black child can easily be a standard deviation down on the derivative skills on which aptitude-test performance -- and all the similar performance on which education is largely premised -- depend. And these effects of race will be different from those on an Asian child of similar apparent socio-economic status and greater apparent language handicap who is being raised in the American iteration of a very different set of cultural behaviors where attention, problem-solving, and reading are concerned. And these effects of race are entirely, IMO, the result of environmental (cultural) factors. And assignment to the universe of environmental factors that comprise the Black Experience in America is almost entirely predictable from "race," from whether one is raised Black, or White.

And that's the gist of my argument with Herrnstein and Murray. Aptitude tests are in fact good measures of a range of problem-solving skills important for school and life success, and these skills as inferred either from the tests or the success are in shorter supply in the ghetto than the 'burbs, and color is a good predictor of which of these settings you live in -- and therefore "race" correlates with "intelligence."

Next week: The Bell Curve II.


[1]Herrnstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life. New York: The Free Press. R.J. Herrnstein had the ill fortune to die while the book was in press.

[2]Jensen, A.R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 39, 1-123.

Herrnstein, R.J. (1971). IQ. Atlantic Monthly (September), 43-64.

Gould, S.J. (1981). The mismeasure of man. New York: Norton.

[3]Bouchard, T. J., Jr., Lykken, D. T., McGue, M., Segal, N. L., and Tellegen, A. (1990). Sources of human psychological differences. Science, 250, 223-228.

Douglas Davis, Ph.D. <>

About the author ...

Copyright (C) Douglas Davis 1994. All rights reserved.