Remembering Stephen Jay Gould: Bully and Boob

Natural History magazine, in which the late Stephen Jay Gould went on at such length for so many years, offers a perspective on Gould's career by the distinguished physical anthropologist Ian Tattersall that's outwardly celebratory, but is actually a pretty funny account of Gould's penchant for projection of all his own intellectual inadequacy, ethical shortcomings, and ethnic hostility on to the morally and technically superior scientists that he fulminated against. (Of course, you have to read it closely and all the way to the end to notice this.)

Remembering Stephen Jay Gould 

By Ian Tattersall 

... Principally in the 1970s—when memories of the struggle for civil rights in the United States during the previous decade were still extremely raw

Or, in other words, when the battles were already won

—Gould devoted a long series of his columns to the subject of racism, as it presented itself in a whole host of different guises. In his very first year of writing for Natural History, he ruminated on the “race problem” both as a taxonomic issue, and in its more political expression in relation to intelligence. He even made the matter personal, with a lucid and deeply thoughtful demolition in Natural History of the purportedly scientific bases for discrimination against Jewish immigrants to America furnished by such savants as H. H. Goddard and Karl Pearson.

A historical farrago, of course, but one that millions of people still believe: Jews didn't score well on IQ tests. A simple historical reality check, such as noting the vast rise of Jewish students at Harvard in the first two decades of the 20th Century and the subsequent imposition of admissions quotas, would suggest that Gould got the story backwards (as indeed he did).

Gould also began his long-lasting and more specific campaign against genetic determinism, via a broadside against the conclusions of Arthur Jensen, the psychologist who had argued that education could not do much to level the allegedly different performances of various ethnic groups on IQ tests. And he began a vigorous and still somewhat controversial exploration of the historical roots of “scientific racism” in the work of nineteenth-century embryologists such as Ernst Haeckel and Louis Bolk. 

But Gould’s most widely noticed contribution to the race issue began in 1978, with his attack in Science on the conclusions of the early-nineteenth century physician and craniologist Samuel George Morton, whom he characterized rather snarkily as a “self-styled objective empiricist.”

It's important to keep in mind this in mind about Gould's jihad against Morton: it's not as if Gould was courageously attacking some giant figure in the history of science with numerous still active defenders. Barely anybody outside of physical anthropologists had heard of Morton when Gould started his crusade against him. The only reason Morton was famous in the 1970s was because Gould was so angry about him. Gould wasn't picking a fight with the current state of the art skull expert, he was picking a fight with a guy who had been dead for 127 years, who had been dead for 8 years when The Origin of Species was published. It was absurd, but it made Gould a lot of money.

In three voluminous works published in Philadelphia between 1839 and 1849—on Native American and ancient Egyptian skulls, and on his own collection of more than 600 skulls of all races—the widely admired Morton had presented the results of the most extensive study ever undertaken of human skulls. The main thrust of this study had been to investigate the then intensely debated question of whether the various races of humankind had a single origin or had been separately created. Morton opted for polygeny, or multiple origins, a conclusion hardly guaranteed to endear him to Gould. Along the way, Morton presented measurements that showed, in keeping with prevailing European and Euro-American beliefs on racial superiority, that Caucasians had larger brains than American “Indians,” who in turn had bigger brains than “Negroes” did. 

After closely examining Morton’s data, Gould characterized the Philadelphia savant’s conclusions as “a patchwork of assumption and finagling, controlled, probably unconsciously, by his conventional a priori ranking (his folks on top, slaves on the bottom).” He excoriated Morton for a catalog of sins that included inconsistencies of criteria, omissions of both procedural and convenient kinds, slips and errors, and miscalculations. And although in the end he found “no indication of fraud or conscious manipulation,” he did see “Morton’s saga” as an “egregious example of a common problem in scientific work.” As scientists we are all, Gould asserted, unconscious victims of our preconceptions, and the “only palliations I know are vigilance and scrutiny.” 

That blanket condemnation of past and current scientific practice was a theme Gould shortly returned to, with a vengeance, in his 1981 volume The Mismeasure of Man. Probably no book Gould ever wrote commanded wider attention than did this energetic critique of the statistical methods that had been used to substantiate one of his great bêtes noires, biological determinism.

Gould was in over his head when it came to the kind of statistics required to contribute to psychometrics.

This was the belief, as Gould put it, that “the social and economic differences between human groups—primarily races, classes, and sexes—arise from inherited, inborn distinctions and that society, in this sense, is an accurate reflection of biology.” 

In Mismeasure, Gould restated his case against Morton at length, adding to the mix a robust rebuttal of methods of psychological testing that aimed at quantifying “intelligence” as a unitary attribute.

Think about how colleges use SAT scores in admission. Do they look at a unitary SAT score: Most do. Do they look separately at Math and Verbal scores separately: Usually. Do they look at the newer Writing score? Some do, some don't. Do they look at SAT Subject Test scores? Some do, most don't. Do they look at subtest scores on the SAT? Maybe a few do, but I've never heard of it.

In other words, we have the usual Lumper and Splitter situation. I.Q. works the same way. The military won't let you enlist if you are in the bottom 30% on the IQ-like AFQT test used in The Bell Curve. But, if they do let you enlist, they are fairly interested in your subtest scores on the AFQT's superset ASVAB in helping determine what kind of job you'll get.

This isn't really that complicated, but Gould was too knuckleheaded to get it.

One of his prime targets was inevitably Arthur Jensen, the psychologist he had already excoriated in the pages of Natural History for Jensen’s famous conclusion that the Head Start program, designed to improve low-income children’s school performance by providing them with pre-school educational, social, and nutritional enrichment, was doomed to fail because the hereditary component of their performance—notably that of African American children—was hugely dominant over the environmental one.

As you can see by looking around you, everything has changed since the publication of Jensen's Harvard Education Review article in the 1960s. Since then, The Gap has completely disappeared, proving Jensen wrong and Gould right.

A predictable furor followed the publication of Mismeasure, paving the way for continuing controversy during the 1980s and 1990s on the question of the roles of nature versus nurture in the determination of intelligence. 

This issue of nature versus nurture, a choice between polar opposites, was of course designed for polemic, and attempts to find a more nuanced middle ground have usually been drowned out by the extremes.

Who exactly are these extremes on the purported Nature Only side that had their hands on the Megaphone to do the drowning? Jensen was on the side of Nature and Nurture, and he sure didn't have the Megaphone. Gould did, and he was the extremist on the Nurture Only side.

So it was in Gould’s case. An unrepentant political liberal, he was firmly on the side of nurture. As a result of his uncompromising characterizations of his opponents’ viewpoints, Gould found himself frequently accused by Jensen and others of misrepresenting their positions and of erecting straw men to attack. 

Yet even after Mismeasure first appeared, the climax of the debate was yet to come. In 1994, Richard Herrnstein and Charles Murray published their notorious volume, The Bell Curve: Intelligence and Class Structure in American Life. At positively Gouldian length

but with unGouldian heapings of data, correctly analyzed

, Herrnstein and Murray gave a new boost to the argument that intelligence is largely inherited, proclaiming that innate intelligence was a better predictor of such things as income, job performance, chances of unwanted pregnancy, and involvement in crime than are factors such as education level or parental socioeconomic status. They also asserted that, in America, a highly intelligent, “cognitive elite” was becoming separated from the less intelligent underperforming classes

But as we can see 19 years later, it's totally clear that Herrnstein and Murray were wrong because society is so much more egalitarian today. Oh, wait ...

, and in consequence they recommended policies such as the elimination of what they saw as welfare incentives for poor women to have children.
To Gould such claims were like the proverbial red rag to a bull. He rapidly published a long review essay in The New Yorker attacking the four assertions on which he claimed Herrnstein and Murray’s argument depended. In order to be true, Gould said, Herrnstein and Murray’s claims required that that what they were measuring as intelligence must be: (1) representable as a single number;

See "Splitter-Lumper" above

(2) must allow linear rank ordering of people;

Which is what Gould's Harvard does in evaluating applicants, what the U.S. Army does, what everybody does. Obviously, rank ordering people has its faults, but the point of The Bell Curve is that tests do have some degree of predictive power, which is what science is about. Gould, however, was far less interested in predicting the future than in controlling the past to control the future.

(3) be primarily heritable;

Or partly heritable. Remember, Gould was the Nurture Only extremist.

and (4) be essentially immutable.

Or, as Herrnstein and Murray said, IQ gaps appear to be "intractable" by plausible government remediation programs. For example, today you see that the most fashionable solution for The Gap is to more or less kidnap tiny black children and hand them over to college graduates to raise during almost all of their waking hours, then drop them off with their kin only for sleeping. Maybe that will work, maybe not, but you can see how far fashion has gone for the Borrowed Generation to be the conventional wisdom of the moment.

None of those assumptions, he declared, was tenable. And soon afterward he returned to the attack with a revised and expanded edition of Mismeasure that took direct aim at Herrnstein and Murray’s long book. 

The funny thing about Gould is that he seemed to have a very high impression of his own IQ, and lots of people shared that view, but there were always doubters, such as Paul Krugman. I doubt if Gould was as smart as Dave Barry.

There can be little doubt that, as articulated in both editions of Mismeasure, Gould’s conclusions found wide acceptance not only among anthropologists but in the broader social arena as well. But doubts have lingered about Gould’s broad-brush approach to the issues involved, and particularly about a penchant he had to neglect any nuance there might have been in his opponents’ positions. Indeed, he was capable of committing in his own writings exactly the kinds of error of which he had accused Samuel Morton—ironically, even in the very case of Morton himself. 

In June 2011, a group of physical anthropologists led by Jason Lewis published a critical analysis of Gould’s attacks on Morton’s craniology. By remeasuring the cranial capacities of about half of Morton’s extensive sample of human skulls, Lewis and colleagues discovered that the data reported by Morton had on the whole been pretty accurate. They could find no basis in the actual specimens themselves for Gould’s suggestion that Morton had (albeit unconsciously) overmeasured European crania, and under-measured African or Native American ones. What’s more, they could find no evidence that, as alleged by Gould, Morton had selectively skewed the results in various other ways. 

The anthropologists did concede that Morton had attributed certain psychological characteristics to particular racial groups. But they pointed out that, while Morton was inevitably a creature of his own times, he had done nothing to disguise his racial prejudices or his polygenist sympathies. And they concluded that, certainly by prevailing standards, Morton’s presentation of his basic data had been pretty unbiased. What is more, while they were able to substantiate Gould’s claim that Morton’s final summary table of his results contained a long list of errors, Lewis and colleagues also found that correcting those errors would actually have served to reinforce Morton’s own declared biases. And they even discovered that Gould had reported erroneous figures of his own. 

It is hard to refute the authors’ conclusion that Gould’s own unconscious preconceptions colored his judgment. Morton, naturally enough, carried all of the cultural baggage of his time, ethnicity, and class. But so, it seems, did Gould. And in a paradoxical way, Gould had proved his own point. Scientists are human beings, and when analyzing evidence they always have to be on guard against the effects of their own personal predilections. 

There is no doubt whatsoever that Gould’s humane and passionate writing in defense of racial equality will be looked upon by future anthropologists and historians as a beacon of rational positivism in an age in which genetic reductionism was showing alarming signs of resurgence—as indeed it still is, as race-stratified genome-wide association studies continue to dominate research on human variation.

In other words, the more science advances, the more Gould is left behind.

As Gould’s longtime friend, the anthropologist Richard Milner, told a correspondent from Discover magazine: “Whatever conclusions he reached, rightly or wrongly, he did with complete conviction and integrity. He was a tireless combatant against racism in any form, and if he was guilty of the kind of unconscious bias in science that he warned against, at least his bias was on the side of the angels.”

The ends justify the means.

This essay is fairly damning, but I wonder: how many readers noticed. I have to imagine that Tattersall (or the editors) calculated this pretty nicely: Nick Wade will notice it's a takedown of Gould, but 98% of the subscribers will just take away "side of the angels."