When I finished, I asked for questions, bracing myself for what I thought would be a barrage of hostile criticism, perhaps a few hoots of derision, and a generally unpleasant thirty minutes. Instead, Lewontin stood up and graciously announced to the class that I had addressed essentially all the concerns that he and Balaban had about human behavioral genetics studies. As far as he was concerned, Lewontin said, our study was scientifically sound. I had been forgiven. I wasn’t sure if I should kiss Lewontin’s ring, but he ducked out of the lecture hall without speaking directly to me.
No doubt Lewontin was hurrying off to find other behavioral genetics studies to attack, ones more vulnerable than Hamer’s. When he got to his Boston hotel room that evening, Hamer learned that his Maryland computers had indeed found what they had been looking for: the linkage of homosexuality to a downstream region of Xq27 with a 3.0 LOD score (a logarithim of the odds). This meant there was only one chance in a thousand the result was not statistically significant.
WITH OTHER EPISODES LIKE Hamer’s, many behavioral geneticists were concluding that the critics were not interested in serious scientific debate. Thomas Bouchard tries to see a more positive aspect to the relentless criticism than mere validation of their work’s importance. He sees it as guaranteeing that he and other behavioral geneticists will strive for the highest, most painstaking, standards in their research. When setting up studies, they are all too aware of the denigrators who lie in wait to pounce on the most trivial potential flaw. The critics, of course, claim that keeping the science pure is their only motive in carping, but even a casual examination of their complaints shows they are intent on discrediting this science in any way they can.
Most behavioral geneticists know that even their most airtight data will be attacked. If their data withstands the first onslaught, they brace for the character assassinations that usually come next. The geneticists proceed in the hope that more reasonable, less politically motivated scientists will see through the critical smoke screen to the research beyond—as indeed happened when influential scientists from other fields, such as Science magazine’s former editor, Daniel Koshland, examined the studies and found them valid. They are also encouraged by the honesty of fellow psychologists, such as Jerome Kagan and Sir Michael Rutter, who were forced to accept the unwanted implications of their own research.
It has always been a fact of scientific life that the ease of acceptance of new information depends in large measure on its compatibility to the day’s political mood. Little hard scrutiny was given the theory-serving experiments of Watson and Skinner; when their claims were eventually subjected to empirical tests, they were found wanting. It is also illuminating to contrast the nitpicking examination of behavioral genetics studies to the free ride given Sigmund Freud. Although his extravagant theories were never tested, they still revolutionized psychiatry and overhauled the way humans viewed themselves. Freud achieved all this with vague claims of “clinical observation,” details of which he refused to divulge, even to his staunchest supporters in the early years. Considering the very different treatment given behavioral genetics by the rest of science, Sandra Scarr, in a philosophical mood, said, with admirable understatement, “Standards of evidence are raised when one is defying the zeitgeist.”
The attacks on behavioral genetics research, rash and uninformed as they often are, flew completely out of control in one instance and provided a chilling glimpse of the ferocity of the antigenes forces. In Jensen’s incendiary paper on I.Q. differences, he had relied to a degree on the separated-twin research of eminent British psychologist Sir Cyril Burt, who died in 1971 at the age of eighty-nine. Considered by many to be the father of modern British psychology, Burt had reached a level of influence and power rarely achieved by a scientist. Among other posts, he sat on the London County Council and held sway over England’s educational policies in the turbulent postwar years. As might be expected, his research on the heritability of I.Q. informed the policies he advocated.
It is ironic that Arthur Jensen, who agreed with Burt about I.Q.’s heritability, was one of two who first cast doubt on Burt’s research. The other whistle-blower, more predictably, was Leon Kamin. One of the main complaints of Jensen and Kamin was that Burt’s concordance figure for I.Q. in separated twins remained a constant .77 through a number of different studies involving numbers of twin sets. The chances of the number repeatedly coming out precisely the same were nil. Both men were also bothered by Burt’s having apparently estimated the I.Q.s of some twin parents, yet treating these estimates as hard scientific data. These concerns about Burt’s science led to a bizarre drama that was perhaps the blood-lust apex of the nature-nurture wars.
The allegations against Burt’s work were picked up and augmented by two English psychologists, Ann and Alan Clarke of Hull University, who were former students of Burt’s. The Clarkes raised the possibility of fraud. Others joined in and eventually the rumbles of a major scientific scandal were overheard by an investigative reporter at London’s Sunday Times. The newsman examined Burt’s research and found other curiosities, such as a dearth of data about Burt’s twins and his claim that World War II bombings had destroyed many of his records. When the reporter was unable to find evidence of the existence of two research assistants Burt claimed to have had, he decided to go ahead with a major expose. It appeared in the Sunday Times in October 1975. England’s most distinguished psychologist was denounced as a mountebank who had lied about his work and falsified data.
It would be hard to overestimate the furor that swept the scientific community. Most found it inconceivable that such an eminent scientist, a man upon whom every honor had been thrust, was in reality “a confidence trickster,” as the Clarkes termed him. The many distinguished scholars who came forward to defend Burt could offer little more than character references; none could provide satisfying answers to the many questions about Burt’s decades-old research, and, even more suspect, none could verify the existence of the two phantom assistants. The attackers finally prevailed, and the defenders backed away. The chorus of denigration climaxed with Burt’s biographer, Professor Leslie Hearnshaw of the University of Liverpool. In his 1979 book Cyril Burt, Psychologist, he sadly agreed that Burt had fudged his data. This was the clincher that led the British Psychology Society formally to accept the guilty verdict about one of its most illustrious members.
For the next twenty years Cyril Burt served as the poster boy for scientific fraud; any references to his work were contemptuous dismissals designed to prevent gullible young psychologists from taking his research seriously. But hints began to emerge that perhaps Burt had been misjudged. Finally, two books appeared, one in 1989 (The Burt Affair by Robert B. Joynson) and the other in 1991 (Science, Ideology and the Media by Ronald Fletcher), that set out to show that the charges against Burt were unfounded or exaggerated, that at worst he was guilty of careless methods, or more precisely, methods considered careless by the standards of the later period in which he was judged. Neither book pretended to answer all questions about Burt’s research, but both were convincing that the evidence was insufficient for the knockout charge of willful fraud.
Both authors have little trouble disposing of one of the most incriminating pieces of evidence against Burt, the identical correlation numbers for I.Q. in different studies involving different numbers of twin sets. Investigation indicated that Burt had not retested in the particular category and had simply used the figure from his earlier tests. This is dubious practice and should have been announced by Burt; but without further evidence, it cannot be termed fraudulent. Both writers, with sound logic, point out that anyone deliberately inventing data would take the obvious precaution of altering the numbers slightly to throw off suspicion. Since this ruse is known to anyone who has ever estimated tax deductions or filled holes in an expense account, it would surely have occurred to one of England’s finest minds, bent on a higher order of chicanery.
The books offer no help in unraveling the mystery of the perhap
s nonexistent research assistants, which lent an Agatha Christie air to the scandal. The best each author could do was track down and interview associates of Burt who had vague recollections of the women but could not swear to their names. In the two books, this was insufficient to lift the charge that he invented the assistants. (Since publication of the exonerating books, one of the women was found, a Miss Conroy, who had emigrated to Australia.)
Both books succeed in their main purpose of establishing that Burt’s reputation was destroyed on too little evidence, and both make fascinating reading as scientific detective stories. Their chronological presentation of the steps leading to Burt’s disgrace offer a terrifying view of a burgeoning lynch-party mentality. Numbers of journalists and distinguished psychologists were like schoolchildren circling a victim, each pushing their taunts a bit further than the one before, all waiting to see who would muster the courage to strike the first real blow. London’s Sunday Times proved the bravest.
Although the actual charges of fraud against Burt came mostly from the English scientists and journalists, it was Kamin who had submitted Burt’s research to a thorough review and declared it worthless. While this is a devastating judgment on a colleague’s life work, scientists, especially Kamin, do this to each other with nerve-wracking frequency. When such a charge was leveled at the august Burt, however, it opened the door to the crescendo of escalating accusations that took place in England. Although Kamin had only raised the possibility of fraud, one can only imagine the satisfaction with which he watched from across the Atlantic the pillorying he had set into motion.
One of the authors defending Burt, Ronald Fletcher, subjects Kamin’s critique of Burt to analysis and finds many inaccuracies and omissions, all on the side of Burt’s purported fraudulence. Fletcher documents this skewed critiquing so thoroughly, in fact, that it raises the suspicion that Kamin felt confident no one would trouble to check him out as he was checking out Burt. However deliberate the distortions in Kamin’s analysis may have been, it was widely accepted and carried considerable weight. It provided what many wanted to hear: I.Q. still had not been proven to be heritable. It also provided the tasty subtext that those who said it was were scoundrels and con men (which is the central theme of Stephen Jay Gould’s book on I.Q. testing, The Mismeasure of Man).
Burt based his claim of I.Q. heritability on a reared-apart twin correlation of .77. When the attacks on Burt began, his correlation figure had already been corroborated by three earlier studies and has been replicated by two subsequent ones—Bouchard’s in Minnesota and the Swedish study of identical twins. Although Burt’s figures clearly contained irregularities and puzzles, his bottom-line conclusion has been confirmed, not only by the two Burt-scandal authors, but also by a wealth of unrelated research. This wholesale substantiation only adds to the irony of so many people expending so much energy in a wild-dog effort to destroy Burt. It is even more ironic that for twenty years they were successful.
GIVEN BURT’S GRISLY FATE, it is surprising that anyone, Swedish or American, would venture into the shark-infested waters of I.Q. heritability. In fact, when Bouchard set up his twin study, the mangled reputation of Cyril Burt hung heavily in his mind, with so much of the doubt based on the absence of documentation, not its falsification. As a result, Bouchard took stringent measures to ensure that every aspect of the Minnesota study was extremely well documented—recruitment of twins, correspondence, interview tapes, medical test results—all the accumulated data of each twin’s involvement with the study was meticulously recorded and stored. Perhaps the most poignant example of Bouchard’s obsessive documentation was his videotaping of his interviews with each twin set.
“I had no specific scientific purpose in mind,” he later said, “but I didn’t want them coming along years later saying I never had any twins, that I made the whole thing up.” As “they” had with Cyril Burt, he might have added. Since Bouchard is not a man of exaggerated fears or paranoia, his belief in the need for hard evidence that the Minnesota Twin Study had actually taken place provides a chilling picture of the ruthless opposition he felt himself to be facing.
Some two dozen scientists had been involved in Bouchard’s study at varying times, as well as hundreds of twins. The project was well known at the University of Minnesota and around the country, his laboratory had been visited by many journalists and scientists, among them Leon Kamin himself, and his twin study had been written up in numerous publications from Sunday supplements to the nation’s most prestigious scientific journals. Still, Bouchard felt that without actual videotape of each twin set, he was vulnerable to charges of fraud. Anyone familiar with the Cyril Burt case could not scoff at Bouchard’s fears.
In fact, a complaint already made against the Minnesota study had implications of fraudulence. The critics spoke often about the unavailability of specific data about Bouchard’s twins, although knowing that releasing such data is forbidden by federal law. On agreeing to take part in the study, the twins signed an informed-consent agreement in which the Minnesota scientists promised not to divulge information that could be traced to specific twin volunteers. Bouchard points out that under these rules, Juels-Neilsen and other twin researchers could not have published as they did; too much of their information could have been traced to particular twins. Even though the critics have been repeatedly informed of this restriction, they complained noisily about Bouchard’s refusal to divulge specific twin data. This evokes the specter that Bouchard and his sixteen coconspirators were falsifying data.
Such smear innuendoes are a tradition among the antigenes group. In spite of Konrad Lorenz’s Nobel Prize for his animal studies, the opposition constantly alleges that he cooperated with the Nazis when a young scientist working under their regime. In the same McCarthyite spirit, the critics point to Minnesota’s acceptance of funds from the politically suspect Pioneer Foundation as evidence of their evil intentions. Bouchard’s arrest for student radicalism, on the other hand, is never mentioned. The critics are as selective in their use of ad hominem data as they are with scientific data.
When Ruth Hubbard accused Dean Hamer, before he had begun his study, of not only bad science but bad politics, she came closer than most of the environmentalists to admitting that their arguments rested on the belief that certain knowledge is “bad” and should be avoided or ignored. Most scientists take the position that knowledge is neutral, value free; the use to which it is put might be good or bad, beneficial or hurtful to society in general. First learn as much as we can, then let society decide how new information will be used. The opponents of behavioral genetics have consistently feared such a climate of unfettered inquiry.
1In addition to Kenneth Kendler’s exhaustive refutation of this point summarized at the end of chapter 9, Robert Plomin cites five studies that show differences in treatment of MZ and DZs are slight and do not have any effect on personality and behavior. The five studies came out years before Not in Our Genes: P. T. Wilson, 1934; A. Lehtovaara, 1938; R. Zazzo, 1960; R. T. Smith, 1965; J. C. Loehlin and R. C. Nichols, 1976.
2Ironically, this same Evan Balaban won headlines in 1997 with an experiment that demonstrated, perhaps more vividly than had Hamer, the direct link between genes and behavior. Working at the San Diego Neurosciences Institute, Balaban planted brain cells of quail into the brains of embryo chickens. The result was chicks that did not sound off like chickens but crowed and bobbed like quail. Remaining true to the faith of his Harvard fathers, Balaban dismissed the implications for humans of this landmark behavioral transplant, announcing that “most human behavior is learned” (Time, March 17, 1997). Alfred Wallace and Bishop Wilberforce must have smiled down at this reaffirmation of human uniqueness.
SIXTEEN
SCIENTISTS IN DENIAL
NOT ALL OF THE CRITICS of behavioral genetics resort to attacks on the integrity of the scientists involved. Some restrict themselves to scientific argument and, because of this, attract less attention than the character assassins. Prominent am
ong this more temperate group is Jerry Hirsch, a professor of psychology at the University of Illinois. Unlike many psychologists, Hirsch is well grounded in gene theory and is considered one of the founders of behavioral genetics.
When the furor over Arthur Jensen’s paper erupted, Hirsch was as appalled as the environmentalists and turned his knowledge of the new science against Jensen and his methods. Disagreement between members of the same discipline happens constantly, but Hirsch’s outrage at what he saw as Jensen’s malevolent misapplication of behavioral genetics sent him into a frenzy of contradiction, from which he never fully recovered. For a good part of the years that followed, Hirsch devoted considerable energy to castigating what he saw as the excesses and oversimplifications of behavioral genetics.
His fundamental argument is that, with humans, there is too much variation in both environment and genetic makeup for any coherent study of gene-environment effects. In the epilogue to a book he edited, Behavior Genetic Analysis, Hirsch writes, “Since genotypic diversity and genotype-environmental influence are ubiquitous, attempts to study the laws of environmental influence have been grasping at shadows.” Having worked extensively with fruit flies, Hirsch developed a two-tiered view of the problem facing behavioral genetics: the ease of controlling fly environments as compared to human, and the enormous potential for genetic variation and anomalies in any organism. Looking at the messy, unmanageable human from this vantage point, Hirsch threw up his hands at ever sorting out the elements with any scientific accuracy. He also saw little use to heritability figures.
Born That Way Page 26