The case against 'The Bell Curve.' (books that links IQ to race)
Years ago, hoping to persuade this publication to hire me, I quit a decent job in Chicago and moved to Washington. Unemployed and low on money, I lived in a seedy neighborhood behind the Navy Yard in Southeast D.C. Because the editor of this magazine unaccountably took his time in acknowledging my merit as an applicant, to blow off steam I played basketball on the local court several hours each day. I was the only white player in the game, accepted at first as a charity case. After a few weeks on the blacktop, however, I was startled to discover other players wanting me on their team. After two months of daily basketball, I found myself able to hold my own in one-on-one matches against the hot players from nearby Eastern High School. I was squaring my shoulders for accurate jump shots, ducking under other players for layups--the sorts of coordinated, classy-looking moves I had never been able to do before and have not been able to do since.
It would hardly be a wild guess that practice had improved my game, and that lack of practice has since eroded it. Charles Murray and the late Richard Herrnstein would say, however, I had suddenly acquired basketball genes. Then just as suddenly, I lost them!
Page after page of obstruent data and marching columns of Pearson correlations in the new book The Bell Curve by Murray and Herrnstein, which holds that success in life is mainly determined by inherited IQ and that statistically significant differences in inherited intellect exist among the races, imply that the issues at play in the IQ dispute are so sophisticated only readers of high intelligence can grasp them. This isn't so. Most common-sense aspects of the IQ debate are more significant than the statistical motes and jots--and being much better understood, are a sounder basis for social policy. The complex statistical claims of The Bell Curve have received extensive notice in initial reactions to the work. In the end the book's common-sense faults are more telling. Blacktop basketball offers an entry point for understanding why.
The reverse of the notion that blacks are born with less intelligence than whites is that blacks are born with more athletic potential. Well-meaning people who believe that whites are smarter than blacks often quickly add, "But look at how gifted blacks are physically," citing the undeniable black dominance of basketball. Yet if blacks have superior innate athletic ability, why are hockey, tennis, and many other lucrative sports largely dominated by whites? As the writer Farai Chideya will show in a forthcoming book, of the approximately 71,000 Americans who earn livings from sports (broadly defined to include golfers, skaters, and so on), only 10 percent are black.
A likely explanation for black success in basketball is not some mystically powerful jumping gene--natural selection may have favored strength and size in people, but what are the odds it ever favored jumping?--but that many blacks practice the sport intensely. For good or ill, thousands of black kids spend several hours per day through their youth playing basketball. By the time age 18 is reached, it shows: In general, blacks are really good at basketball. Meanwhile, hockey and tennis are usually practiced in youth by whites, who in turn dominate these sports.
In all the complex arguments about inheritability and environment in IQ, the mundane, common-sense question of practice time is often overlooked. Other things being equal, what you practice is what you're good at. As Charles Darwin once wrote to his cousin Francis Galton, founder of the eugenics movement: "I have always maintained that, excepting fools, men [do] not differ much in intellect, only in zeal and hard work."
As a longtime basketball-league participant and a mediocre small-college football player, I have spent a notable portion of my life being knocked down, run past, and otherwise outperformed by black athletes. None ever struck me as possessing any mystical genetic athletic ability, though it may be that as a group they hold some small edge over whites. What often does strike me as a black basketball player in a pickup game hits his shot and I miss mine is the thought: "He's taken that shot maybe five million times in his life, and I've taken it maybe five thousand." It's safe to say that if there had been no color barrier to college basketball in the 1940s and 1950s, blacks would not have dominated in those years, because at that time few blacks practiced basketball as much as the best white players of the period. By coincidence, the week before The Bell Curve was published, the "Science Times" section of The New York Times ran a prominent article on new research showing that the most accomplished violinists and other artistic performers spend significantly more time practicing than the less accomplished--though presumably they enjoy the advantage of genetic gifts. There seemed to me a pellucid connection between this research and the Herrnstein-Murray thesis.
Another missed connection concerns a 1990 flap at the University of California at Berkeley. There, a tenured anthropologist, Vincent Sarich, began to say that black success in basketball proved the inherited basis of talent, which in turn supported the view that whites could inherit superior mental faculties. Sarich's argument is revealingly faulty: He would tell classes that "There is no white Michael Jordan. ..nor has there ever been one." Actually there was a white Michael Jordan--the late Pete Maravich. Maravich scored much more than Jordan in college and had the same league-leading scoring average in the NBA, 31 points per game. Maravich had the same ability as Jordan to throw the no-look pass, to dunk in ways that appeared to defy certain laws of physics, and so on. Jordan became a sports legend because his college and pro teams were champions; this happened because Jordan was a highly disciplined defensive performer and an astute judge of the court situation. Maravich, in contrast, became something of a standing joke, even to sportswriters eager for white stars, because his teams always lost. Maravich was a hopelessly selfish performer, inert on defense and he never passed up a shot. The comparison between Jordan and Maravich both defies the stereotype of the white player as disciplined and the black player as the gunner, and undermines the notion of black genetic dominance generally.
So if white kids as a group spend more time practicing schoolwork, should we then be surprised that they score better on school-related tests? Herrnstein and Murray acknowledge that 150 hours of extra study will raise the typical student's SAT score by 40 points--a common-sense confirmation that scholastic practice makes for scholastic success. True, the score-boosting effects of extra study on SAT tests reach a plateau beyond which further practice adds little. Yet seeing that behavior (study time) alters brain-test outcome, and then concluding as The Bell Curve does that brain performance is mainly genetic, is an inverted form of the logic that Stalin's favorite scientist, Trofim Lysenko, employed to contend that genetic characteristics are acquired during a person's life. That many white kids may spend more hours studying than many black kids may well be an argument that some minority parents are negligent in compelling their children to hit the books. But this is an argument about environment, not inheritance.
It is not racist for Herrnstein and Murray to study whether there are differences in inherited IQ. Some commentators have attempted to reject The Bell Curve out of hand on grounds of racism, and thereby avoid dealing with its discomfiting contentions. Yet obviously people talk about the mental abilities of various groups, usually in whispers; better to talk about this in the open. For this reason, in my affiliation with The Atlantic Monthly, I favored that magazine's publication of some of Herrnstein's earlier work. I agreed with the decision of The New Republic to put an excerpt from The Bell Curve on its cover. And I am glad Herrnstein and Murray (the principal author) wrote The Bell Curve, which is not a racist work, though it is fantastically wrong-headed. Bringing the arguments about race, inheritance, and IQ out into the open in Murray's straightforward writing style is a useful service--especially because the more you know about this line of thought, the less persuasive it becomes.
Now, other objections to The Bell Curve, concentrating on those not already raised by other commentators:
* The Hollywood corollary. Perhaps black overrepresentation in basketball is essentially a fluke telling nothing about the general relationship between practice and achievement. Yet consider that blacks are also overrepresented in several performing arts, notably singing and comedy. Is this because they have superior singing and joking genes? It's hard to imagine why natural selection would have favored DNA for human song. On the other hand, African Americans as a group have spent generations learning various forms of performance. Most African culture is oral; and until recent decades, owing both to discrimination and poverty, when American blacks wanted entertainment they had to entertain each other. That is, they practiced song and comedy, and they got good at it.
* Is everybody too dumb to know who's smart? In The Bell Curve there are numerous assertions that society has handicapped itself by failing to favor the smart. For instance, the book asserts that the American economy loses as much as $80 billion per year because a 1971 Supreme Court decision bars most forms of workplace IQ testing. High-IQ workers are more productive, Herrnstein and Murray say; promoting them would increase productivity. But if high-IQ employees are more productive, that should be self-evident to employers regardless of tests. Are employers so dumb they don't promote the productive workers? On a common-sense basis, society has long been attuned to what can be accomplished by the smart, and almost always rewards this already.
* The Hiram College contradiction. Early in The Bell Curve comes a section describing how in the fifties the freshman class at Harvard was not composed exclusively of the brightest of the bright; many were slow-witted kids entering on Dad-um's alumni connections. This was actually to the good, Murray writes, because it meant that many bright kids who otherwise would have been consolidated at Harvard instead had no choice but to attend Hiram or Kenyon or some other school, distributing IQ throughout society. These days, The Bell Curve says, owing to accurate SAT testing (which is now quite accurate, but only so far as it goes), Harvard gets the brightest of the bright, withdrawing the "cognitive elite" into a small, isolated world. This, the book says, is bad.
Yet later, in a section assailing affirmative action (The Bell Curve really despises affirmative action), Murray says that offering special admissions consideration to minority students is awful because it denies some worthy white students entry slots in the top schools. But isn't the effect that a percentage of smart kids end up at Hiram and Kenyon, distributing IQ throughout society? When some smart white kids were denied admission to Harvard because the sons of the landed had a special deal--in other words, when there was a patrician system that favored the affluent--that was great, according to The Bell Curve. Now that some smart white kids are denied admission to Harvard because the sons and daughters of poverty have a special deal--the new system no longer rigged in favor of the affluent--that's offensive, according to the same book.
* Those inherited differences that have been confirmed are small. It's obvious that there exist inheritable physical differences among racial groups. But all such differences are too minor to mean anything, except as sources of the many forms of prejudice. Perhaps there are inherited mental differences among racial groups, but the observed pattern in physical differences suggests any mental differences would also be too minor to matter in practical terms. For instance, African Americans are on average about an inch and a half taller than Caucasian Americans. This distinction is real but just too small to make a difference except in highly competitive situations like, say, entry into the small number of slots in the National Basketball Association: There, a competition between two players of otherwise equal skills might end in the taller being selected. Similarly, suppose there really are on average a few points of difference between whites and blacks in IQ. This is too little to matter in practical terms, except in highly competitive situations like, say, entry into the small number of clerkships to the Supreme Court, where an extra margin of IQ might carry the day.
* If this stuff is really true, it's whites that ought to feel inferior. The same IQ tests that Murray says show blacks one "standard deviation" (in this case, very roughly 15 percent) less smart than whites show white children duller than Asian-American children by almost the same margin. Simple-minded me might say that is mainly because of the phenomenal (probably excessive) study time many Asian-American parents impose on their kids. But if genes are the IQ destiny that The Bell Curve asserts, shouldn't whites be maneuvering to protect themselves against Asians, given that Asians already out-number Caucasians worldwide? Instead, nearly all of the book's prescriptive material focuses on reasons to retaliate politically against blacks: end affirmative action, shift money from compensatory education programs like Head Start to programs for the "gifted" (that is, white students), scale back welfare.
* What's the mechanism? All human beings are physically similar because they share a line of descent and have all been subject to about the same "selection pressure" from evolution. For there to be significant inherent mental differences among racial groups, there would have to have been significantly different selection pressure. Scientists call this the "What's the mechanism?" question. Any researcher claiming to have found a substantial genetic difference among similar creatures is expected to propose a selection mechanism by which the differential arose.
Early in The Bell Curve controversy, The New York Times ran an op-ed article asserting that from an evolutionary standpoint, differential intellect among human groups could not have evolved in fewer than "hundreds of thousands of years." This is weak science: Most recent discoveries tend to support the notion that natural selection can operate relatively quickly in geologic terms. So differential intellect is not precluded. Yet neither Herrnstein and Murray nor any credentialed believer in the brain-gene theory has suggested how, on an evolutionary basis, black and white intelligence DNA could have diverged significantly.
The sole researcher asserting a hypothesis in this category is J. Philipe Rushton, a psychologist at the University of Western Ontario. The Bell Curve makes a point of praising Rushton as "not ... a crackpot." But a crackpot is precisely what Rushton is. He believes that among males of African, European, and Asian descent, intellect and genital size are inversely proportional, and that evolution dictated this outcome in an as-yet-undetermined manner. Sound like something the 16-year-olds at your high school believed? That should not stop Rushton or any researcher from wondering if there might have been different selection pressures on different racial groups. But Rushton's "research" methods, defended by The Bell Curve as academically sound, are preposterous. For instance, Rushton has conducted surveys at shopping malls, asking men of different races how far their ejaculate travels. His theory is the farther the gush, the lower the IQ. Set aside the evolutionary absurdity of this. (Are we to presume that in prehistory low-IQ males were too dumb to find pleasure in full penetration, so their sperm had to evolve rocket-propelled arcs? Give me a break.) Consider only the "research" standard here. Is it possible that one man in a hundred actually knows, with statistical accuracy, the average distance traveled by his ejaculate? Yet The Bell Curve takes Rushton in full seriousness.
* Are the natives doltish as well as restless? Herrnstein and Murray note that not only do African-Americans score somewhat below white Americans on IQ tests but tribal black Africans score significantly low even in pure-logic tests designed to correct for language differences. (Mazes and so on.) Though many claims of cultural bias are exaggerated by the we're-all-victims lobby, here IQ tests would be expected to be of little reliability, considering the gulf between cultures as different as those of the United States and tribal Africa. Yet The Bell Curve takes the low scores of tribal Africans in earnest, implying this proves the existence of an entire continent of morons.
Just what mechanism of selection pressure would have caused this wide disparity? The authors do not say, gliding past this and all other complications of genetic science. More, they make a tee-hee implication, citing IQ scores among South African "coloureds," that American blacks are smarter than African blacks because of interbreeding with whites. Objection One: If black-white interbreeding in North America were substantial enough to transfer the presumed white intellect to the black gene pool, would not the same process have transferred the presumed black athletic gifts to the white gene pool, leading to an NBA dominated by guys named Blaine and Todd? Objection Two: Though geologic time would probably not be required for differential intellect to arise if a selection mechanism could be shown, a couple of centuries seems insufficient. That is, unions between whites and blacks since colonial times would be unlikely to account for African-Americans doing substantially better on IQ tests than tribal black Africans. The exposure of African-Americans to an educational system teaching (as it should) book-based culture would, on the other hand, explain it pretty neatly.
* Geneticists don't claim genes explain IQ. "The people who say intelligence is genetic are the ones with no training in genetics," says Evan Balaban, a former professor of evolutionary biology at Harvard and now a fellow at The Neurosciences Institute, a research organization. Murray is a social scientist; Herrnstein was a psychologist. Balaban continues: "Any serious biologist would be horrified by the idea of using the little we know about genes as the basis for social policy. Current genetic research cannot even explain how basic body parts form." Nearly all contemporary discoveries about human genetics concern only markers or genes associated with protein coding, vastly less complicated than a developmental trait like intellect.
Troy Duster, a sociologist at the University of California at Berkeley who has studied the history of claims of inherited intellectual inferiority, notes that, "Since the turn of the century the people making political assertions about population genetics always reason backwards from the phenotype [observed trait] to the presence of a gene. This is the reverse of the way molecular biology reasons. Since molecular biologists have discovered genes for diseases like cystic fibrosis and Tay Sachs, the public has begun to believe biologists already understand the human genome. People like Herrnstein and Murray use the halo effect of that belief to give their views a sheen of modem genetics, when in truth their assertions run counter to what can be supported by modem genetics." I called prominent molecular biologists at Harvard, MIT, Penn State, Stanford, and the University of Washington, and all asserted that the notion of a traceable gene line for intelligence has no grounding in present research.
Telling in this regard is The Bell Curve's misunderstanding of Mendelian genetics. The authors treat inheritance from parents as if it could be charted in straight lines: Smart parents A beget smart kids B, etc. This is a common blunder. Trait-inheritance charts more often look like zig-zags, as phenotypes bounce around among offspring and may skip entire generations. Two red-haired parents may have two brunette children, each of whom in turn have one red- and one black-haired child, and so on. Herrnstein and Murray allude in a few sentences to the common outcome that the children of very bright parents may be only somewhat above average in intellect, but otherwise depict IQ as reliably passed through the generations in straight-line fashion. If IQ does pass down generations in straight lines, then the cause must be mainly the environment families create, since genetic traits don't express so predictably.
* Nonsense dysgenics. A substantial doom section of The Bell Curve is devoted to "dysgenics," the reverse of eugenics--the fear that high fertility rates among those of low mental prowess will swamp society with dumbness. At least since Malthus, this has been a belief of the privileged classes whose concerns Murray and Herrnstein hold foremost. It was the central fear of Darwin's cousin Galton, and was a reputable paranoia among the educated in the United States as recently as the years when the Nazi use of eugenics became known. Even Norman Thomas, the most important American socialist of this century, in the thirties denounced the high rate of fertility among "those of a definitely inferior stock."
Yet during the very century in which, The Bell Curve says, dysgenics has run wild globally, overall scores on IQ tests have consistently risen by decade, among blacks as well as whites. Now, how can it be that overall IQ scores are going up, yet society simultaneously is being swamped by fertile dullards? One possible explanation is that in decrying high fertility rates among low-achieving inner-city women (a problem, to be sure, though first for the women themselves), The Bell Curve conveniently overlooks a parallel social phenomenon: the rise of the American black middle class. Today, for every one African-American whose life pattern fits the dysgenic nightmare, there are roughly two following the eugenics prescription--moving out of the city, having smaller families, advancing financially and scholastically. Black middle class school achievement trails comparable white numbers, but a small trailer effect seems easily explained as a remnant of segregation.
Herrnstein and Murray say little about the black middle class, a significant group which for good or ill is busily embracing suburban American norms. The authors can't deal with this factor because not only would it foul up claims of dysgenics; dealing with it forces you to confront the fact that many studies show children's IQs tend to be higher in smaller families. This is what might be expected, as other things being equal smaller families offer children more attention and have better social and economic circumstances. That's er, ahem, nurture rather than nature, which falls outside the desired conclusion of The Bell Curve.
* Spin disguised as scholarship. The most disquieting aspect of The Bell Curve is its insistence on phrasing as detached data analysis what is in truth an ideological argument about social policy. Ideology regarding social policy is fine, but should be presented as such. The authors of The Bell Curve adapt a weary tone of "we hate these conclusions, yet as scientists we are driven to them by impartial reading of neutral data." The data they offer as impartial has, however, been elaborately scrunched to fit the desired ideological boxes.
The book's main artifice in this regard is to present the work of those researchers who do conclude that IQ is mainly inherited and is the main determinant of life outcomes (there are a few such researchers, with full credentials), then describe their studies as generally accepted or no longer seriously contested by other researchers. This is duplicitous. Most academic researchers now accept the notion that IQ tests have become reasonably fair and reasonably predict performance in school. Beyond that there exists a fantastic range of opinions about what the tests really tell you. Many credentialed academic "psychometricians" (students of IQ) come to conclusions dramatically at odds with what Herrnstein and Murray think about IQ, genes, and mental determinism, Robert Sternberg of Yale probably standing as the leading example. The Bell Curve makes passing reference to the existence of prominent academics who would reject its thesis, but in the main represents to readers that few researchers now contest the notion that IQ rules. This borders on intellectual dishonesty.
* Spin disguised, period. Murray's work on The Bell Curve was underwritten by a grant from the Bradley Foundation, which National Journal in 1993 described as "the nation's biggest underwriter of conservative intellectual activity." Bradley is a respectable foundation about whose financial support no author need apologize. But Bradley backs only one kind of work: that with right-wing political value. For instance, Bradley is currently underwriting William Kristol. The Bell Curve identifies Murray as a "Bradley Fellow" but gives readers no hint of the foundation's ideological requirements. Telling readers this would, needless to say, spoil the book's pretense of objective assessment of research.
Slipping down the slope from the respectable Bradley Foundation, Herrnstein and Murray praise some research supported by the Pioneer Fund, an Aryan crank organization. Until recently, Pioneer's charter said it would award scholarships mainly to students "deemed to be descended from white persons who settled in the original 13 states." Pioneer supports Rushton and backed the "Minnesota Twins" study, which purports to find that identical twins raised apart end up similar right down to personality quirks. The Aryan crank crowd has long been entranced by the Minnesota Twins project, as it appears to show that genes for mentation are entirely deterministic. Many academics consider the protocols used by the Minnesota Twins study invalid.
Lesser examples of disguised ideological agenda are common in The Bell Curve. For example, at one point Murray presents an extended section on problems within the D.C. Police Department, saying their basis lies in "degradation of intellectual requirements" on officer hiring exams. Information in this section is attributed to "journalist Tucker Carlson." No one who lives in Washington doubts its police department has problems, some of which surely stem from poor screening of applicants. But who is the source for the particularly harsh version of this problem presented in The Bell Curve? "Journalist Tucker Carlson" turns out to be an employee of the Heritage Foundation; he is an editor of its house journal Policy Review. Heritage, for those who don't know it, has a rigid hard-right ideological slant. Its Policy Review is a lively and at times insightful publication, but anyone regarding its content as other than pamphleteering would be a fool. The article The Bell Curve draws from lampoons the intelligence of D.C. police officers because some cases have been dismissed owing to illegible arrest records. And just how many high-IQ white doctors have unreadable handwriting? If an article in Policy Review were an impartial source of social science observations, Murray would simply come out and say where his citation originates. Instead he disguises the source, knowing full well its doctrinaire nature.
* Even the worst-case claimed by the brain-gene believers just doesn't sound so bad. Herrnstein and Murray estimate that intelligence is 60 percent nature, 40 percent nurture. Since genes get the majority number here, to them this clinches the argument for inborn intellectual determinism.
But think about this worst-case-intelligence as 40 percent nurture. "Forty percent variability based on environment would make intelligence an exceptionally pliant trait," Balaban says. It's known, for example, that better nutrition can improve height--but only by a few inches, about a five percent swing based on the potential range of human statures. If IQ swings by 40 percent owing to circumstances and life experiences, then human society has more control over intelligence than virtually anything else in its genetic inheritance. Thus, even The Bell Curve's own contentions would seem solid ground upon which to support further attempts to improve the school and home environments of underprivileged children.
In the end, The Bell Curve should be seen not as racist or violating a taboo, but simply as an attempt to torment data to make it support a right-wing agenda. That's fine so far as it goes: Right-wing ideas have as much claim on society's attention as any other kind, and some of the conclusions Herrnstein and Murray offer are surely correct ones. (They're surely correct, for example, in contending that in most cases small, stable, "legitimate" parents-wedded families are in the best interests of the child.) It is essential, however, that The Bell Curve be seen as a tract advocating a political point of view, not a detached assessment of research. In that regard two final common-sense objections to the book are particularly strong:
* You don't have to be real smart to grasp test-score convergence. For decades black scores on IQ and aptitude tests have been converging upward toward white scores, even as white scores rise. Exceptionally high intelligence is not required to theorize that this is happening because of improved educational opportunity.
The Bell Curve makes a passing mention of black IQ score increases, calls them encouraging, then quickly switches back to doom pronouncements about genetic determinism and the feeble-mindedness of minorities and the poor. Anything more than a passing mention of black IQ test convergence would have kicked the chair out from under the premise of Herrnstein and Murray's tract. If someday black scores stop rising toward white scores, that might be alarming. But this hasn't happened yet, and until it does all the marching data in The Bell Curve and similar works will contain a huge common-sense defect.
* Even if The Bell Curve were right about genes, then it's still wrong about policy.
It turns out that since IQ testing became common, approximately in the 1920s, the scores of American blacks have shifted upward by about two "standard deviations"--that is, about twice as much positive shift as the negative gulf Herrnstein and Murray find between whites and blacks today. But then almost every American group's IQ score has upshifted by about two standard deviations in recent decades. Blacks, whites, yellows, reds, browns: According to IQ testing, we're all getting smarter dramatically fast. The explanation would seem obvious--quality and quantity (especially number of years of schooling) of education has gone up for everybody, so everybody now does better on tests of educational aptitude. Herrnstein and Murray reject this view, saying it must be mainly genes.
Suppose they're right. If rising IQ levels are mainly genetic, then some evolutionary force must be propelling genus Homo in the direction of more DNA for brainpower. Modern society rewards education and mental prowess, so evolution may now be rewarding the same. (Genes do not change during life, but changing circumstances influence which genes are deemed fit and passed to offspring--this is the definition of selection pressure.) Thus if The Bell Curve is correct about intellect being mainly genetic, then some aspect of modern social circumstances and government policy must be encouraging or at least neutral to a fantastic wave of improvements in the human genetic endowment for IQ.
Yet The Bell Curve concludes by calling for drastic changes in social circumstances and government policy--the very forces which, in Herrnstein and Murray's analysis, seem to be causing natural selection to favor IQ as never before. The book ends up mired in such illogic either because its authors do not understand the science of genetics on which they pretend to premise their case, or have produced what should properly be seen as an unusually lengthy promotional brochure for a rather unattractive political package.
|Printer friendly Cite/link Email Feedback|
|Date:||Dec 1, 1994|
|Previous Article:||How Washington insiders ambushed Mickey Mouse.|
|Next Article:||Welfare lines.|
|Race, earnings and intelligence.|
|IQ's evolutionary breakdown: intelligence may have more facets than testers realize.|