Twenty years ago, an experimental psychologist with a background in animal research and a political scientist famous for a scathing critique of the welfare state collaborated on a book about human intelligence and its role in structuring society. Early positive reviews were quickly eclipsed by an outpouring of scathing denunciations and abuse, most of it aimed at the authors rather than their arguments. And so began the public life of Richard J. Herrnstein and Charles Murray’s The Bell Curve: Intelligence And Class Structure In American Life. In the public imagination, The Bell Curve remains controversial, but on its 20th anniversary it is long overdue for a retrospective, particularly as the major scientific developments in cognitive studies over the last two decades have upheld its most salient points. (Here, incidentally, is Murray discussing the book on a Harvard panel convened earlier this year.)
Amusingly, the most consequential claims of The Bell Curve are also its least controversial: there is, in fact, such a thing as quantifiable human intelligence (Murray recently recommended this summary); it is an extremely good predictor of future life outcomes and, despite the vociferous wishes of a large majority of sociologists, it is highly heritable. Murray and Herrnstein use regression analysis, a method of analyzing and isolating variables to determine their relationships to a given outcome, to compare how well IQ predicts social ills like poverty, welfare dependency, out-of-wedlock births, low educational attainment and criminality. At every level, they find that low IQ is a significant – often the significant – causal factor, even taking into account family background.
But if low intelligence is a necessary condition for these social ills, Murray and Herrnstein are quick to point out that it is not sufficient. There has always been a steep difference between the intelligence of the most and least gifted in society, but never before has that difference been so convincingly associated with, for example, delinquency or single-parent families. Beginning an argument that would later become Murray’s Coming Apart, the two examine what they term a phenomenon of “cognitive stratification,” which holds that, with universal high school education and standardized testing, universities have become increasingly good at enrolling the nation’s best and brightest. These men and women, future rulers of the country in one way or another, all come from an increasingly narrow life experience: a small handful of elite universities, sequestered from the American mainstream, followed by internships and careers of the mind (lawyer, doctor, accountant, journalist, academic), and, eventually, marriage to like-minded people of comparable backgrounds. The result is a division of society along intellectual lines, destined to get increasingly more severe as the least cognitively demanding jobs are shipped overseas or automated, and as the payoffs for high intelligence in relatively new fields like financial analysis grow exponentially.
Amusingly, the book’s controversial reputation revolves entirely around one small section, the 13th chapter, coming nearly 300 pages into an 800-page book – this despite the fact that, properly examined, every single chapter of this book represents a radical upheaval of received wisdom. It is in this infamous 13th chapter that Murray and Herrnstein examine ethnic differences in IQ scores. As a primer: IQ tests are standardized such that the mean score in a given population is 100. Roughly half of the US population is concentrated between an IQ of 90 and 110; an additional 20 percent are found at scores of both 75-90 and 110-125, and the final 10% of the population are divided evenly between those with sub-75 IQs (5%) and those with IQs exceeding 125 (5%). The troubling finding is that there are persistent and substantial differences in IQ scores between ethnic groups. The average difference, for example, between black and white scores on over 150 studies is 1.08 standard deviations, which amounts to roughly 16 IQ points. In other words, “a person with the black mean [is] at the 11th percentile of the white distribution, and a person with the white mean [is] at the 91st percentile of the black distribution.” This difference has been consistently observed, not only at the time of The Bell Curve‘s publication but in the intervening years, and while it is far from the only observable ethnic difference in IQ (Ashkenazi Jews and East-Asians have the highest average means, for example), it has been the most explosive, for obvious reasons.
The locus of debate for some time has been on whether or not these results are either accurate or reliable. It is frequently contended that IQ tests exhibit a persistent cultural bias that favors, for example, whites in North America. But the public debate over IQ has been, for some time, many years behind the actual science, and the question of whether or not cultural knowledge is biasing IQ scores has been settled for some time. It turns out, in fact, that the largest observable differences in IQ scores across ethnic groups come on questions that are more abstract (and therefore less vulnerable to cultural bias). Popular academics like Noam Chomsky and the late Stephen Jay Gould have also attacked The Bell Curve (and the scientific study of intelligence more generally), arguing that intelligence cannot be reduced to any one variable, the so-called g or general intelligence factor. Gould’s book The Mismeasure Of Man is invariably cited by outsiders, for example, as the definitive refutation of The Bell Curve. And yet both Gould and Chomsky’s criticisms were laughed at by researchers in cognitive science, 52 of whom published a report in the Wall Street Journal entitled “Mainstream Science On Intelligence,” intended to apprise the public of the middle-of-the-road beliefs of intelligence researchers. Though it was not officially a response to The Bell Curve, and does not in fact mention Murray and Herrnstein’s book, it upholds their central positions.
I will echo the sentiments of Thomas Sowell, who, in an otherwise critical review, still managed to say this: “This is one of the most sober, responsible, thorough and thoughtful books to be published in years.” And indeed it is. Great care went into its research and writing; nothing is presented polemically or dogmatically. It is a sad commentary on the state of our cultural sophistication that the book was so vociferously denounced, and by many who were proud to declare that they had not even bothered to read it. But nor should its thorough and thoughtful composition be taken as a blanket endorsement of the book’s many claims. The ideas of The Bell Curve are too incendiary, bear too directly on important questions of justice and good governance, to be casually endorsed or dismissed. That proviso being registered, this is a staggeringly good book, at once provocative and compassionate, and a model for good social science writing.