Born That Way

by William Wright

  Having aligned themselves scientifically with the new thinking of a gene-environment mix, they then devote considerable space to proving that the entire gene-environment interaction is too complex, too contingent, too fluctuating to ever nail down in any useful way. And they hammer away at research projects that have done precisely that, which is to say most of the sociobiological and behavioral genetics research of the past twenty years. They suggest, for instance, that kinship studies are meaningless, in that siblings and other relatives may resemble each other for environmental, as well as genetic, reasons. In this helpful way they deftly put their finger on the dynamic that kinship studies have, for twenty years, been successfully unraveling.

  When they turn their sights on studies that compare identical with fraternal twins, both raised together, they quite remarkably fall back on the well-refuted argument that MZs are more alike than DZs because they are brought up to emphasize their similarities.1 As one piece of evidence they cite a study that indicates MZs do homework together more frequently than DZs. While it would not be immediately apparent to most people that this collaboration would bring the twins’ I.Q.s closer together, the most interesting revelation is that the authors have gone to the trouble of finding a study that suggests variation in MZ and DZ environments, while totally ignoring the numerous studies that prove that such variations, to the extent that they exist at all, have zero effect on twin development. The studies, which were reviewed at the end of chapter 9, were planned and executed well before publication of Not in Our Genes, primarily in response to these very criticisms—which had often been made, primarily by these very men—and were aimed specifically at the points they had repeatedly raised. The authors could have criticized the studies and picked holes in the design and methodology, but to pretend this counterresearch didn’t exist was scientifically irresponsible and dishonest.

  When the three authors arrive at the subject of separated twins, they omit any reference to the Minnesota Twin Study, which had been in progress for five years at the time of the publication of Not in Our Genes and which obviated most of the book’s arguments. The authors chose to attack instead the earlier, less airtight reared-apart twin studies, targeting in particular the high concordance in I.Q. scores of separated MZs. Indignantly, they explain this nagging statistic as a result of recruitment bias, contacts between supposedly separated twins, and family connections between the allegedly diverse households.

  Thomas Bouchard had been well aware of all of these criticisms of the earlier twin studies and designed his own study to meet them. Indeed, the elimination of such flaws had for him been a powerful motive in undertaking another separated-twin study. In the eyes of most observers, he was succeeding. He also seems to have been succeeding in the eyes of the Not in Our Genes authors because they chose to ignore the largest and most extensive separated-twin study that had ever been undertaken, while laboriously picking apart the vulnerable previous studies.

  One of the most telling examples of the way Not in Our Genes ignores data concerns I.Q. findings. After attacking the earlier studies on the usual (disproven) grounds, they make no mention of the Minnesota study, which avoided the very pitfalls the authors cite yet had resoundingly replicated the earlier MZA studies’ high I.Q. concordance figures.

  When trying to fathom how eminent scientists could manipulate their evidence in this unconscionable manner, it is perhaps useful to remember that for some, political ideology takes strong precedence over scientific facts. In a conversation I had with James Watson, the codiscoverer of DNA’s structure, he told me that physicist Salvadore Luria had once said to him that “politics are more important than science.” The casual statement of priorities suggests how in those of like mind this might justify scientific skulduggery. To some, noble ends justify ignoble means.

  While science is supposed to examine evidence and draw conclusions, the politically oriented start with their conclusions of how the world should be, then bend or edit the evidence to fit that conclusion. This was standard operating procedure in Russia under Stalin and in Germany under Hitler, and it still thrives in American universities in the guise of political correctness. In researching this book, I was quite surprised to find the media that reports on scientific developments seem unaware that this goes on. They are surely motivated by a desire to be evenhanded, or perhaps they just want to stay out of the fray. For whatever reason, they dutifully report the destructive obfuscations of ideologues like Lewontin; but in so doing, they appear to me to be allowing themselves to be used in ways that have nothing to do with science. As a journalist, I feel that once a source has shown a willingness to lie for higher political goals, that source has disqualified him- or herself from future input.

  FOR THE BEHAVIORAL GENETICISTS exposed to such politics-driven criticism, an added frustration is their opponents’ failure to conduct their own research to substantiate their nihilistic arguments. Counter-research is, of course, the accepted method of resolving scientific disputes. If you feel a colleague’s conclusions are wrong, you design and execute a study that will expose the error. The critics of behavioral genetics, on the other hand, confine themselves to the far less arduous course of attacking their opponents’ research, finding weaknesses, possible flaws, and so on. As Arlen Price put it, “Kamin and the others sit back in their armchairs and take shots.”

  Except for Leon Kamin, most of the critics do not bother to inform themselves thoroughly about the research they are attacking. This sometimes results in their citing nonexistent flaws. A frequent criticism of separated-twin studies was that the twins had actually had contact during the so-called years of separation. While this was true, to a degree, with the earlier studies, it is definitely not true of the Minnesota study. Yet the critics continue to hint darkly of twin contact that undermined claims of “separated.”

  Jonathan Beckwith, a molecular biologist at Harvard and a political activist, is one of the more responsible and fair-minded of the behavioral genetics critics. Yet when I spoke with him, he mentioned this as a problem of the Minnesota study, citing a specific example, Jack and Oskar, the Hitler Youth and his kibbutzim brother. It was bad luck for Beckwith that he cited a pair of twins who were separated in the first six months of their life and did not see one another for twenty-one years, during which time they both lived in dramatically different environments.

  “They were not really separated,” Beckwith said authoritatively. “They had contact.”

  I pointed out that the first contact occurred when the twins were twenty-one, a fact ascertained by both twins, who had no reason to lie about it, and substantiated as well by Jack’s ex-wife, hostile to her former husband, who was present at the reunion. Beckwith shifted his ground: “Well, they were both brought up in German households,” he said. “Their upbringings were similar.”

  Aside from the odd supposition that two boys brought up in German households, any German households, would turn out with Jack and Oskar’s catalog of shared quirks and traits, the fact is Oskar’s German household was his Christian grandmother’s house in a village on the German-Czech border, while Jack’s “German household” was a series of Jewish households—one in Trinidad, one in Venezuela, and two in Israel. If this would satisfy a scientific mind as explaining the Jack-Oskar similarities, it would suggest an even more staggering list of identical tics and idiosyncrasies that would inevitably be shared by any two boys growing up in different homes in, say, Berlin or Munich.

  In a paper criticizing the Minnesota Twin Study, Beckwith and coauthors elaborate on the failings of Jack and Oskar as separated identical twins. With a smug tone of this-is-the-real-story, they point out that, after their meeting, their wives corresponded. When I interviewed Crystal Yufe, she was quite certain there had never been anything more than a few Christmas cards between households. Even if they had corresponded, it is remarkable that a scientist of Beckwith’s standing would suggest letters between wives could result in a precise duplication of their husbands’ oddities, perso
nalities, and abilities. Beckwith goes on to suggest that the physical similarity in the two men could have resulted in similar treatment while growing up (in Germany and Trinidad). This ingenious, if slightly far-fetched, possibility has been previously tested empirically, as was mentioned at the end of chapter 9, and was found to have no effect.

  In the paper, Beckwith and the others make the charge that separated-twin studies are based on the assumption that each twin’s environment is markedly different. There is no such assumption. The studies assume that in measuring a group of reared-apart twins, their environments are on average more disparate than those of two twins growing up in the same house. The Beckwith paper also questions the assumption, which he falsely terms essential to twin studies, that twins growing up in the same household have the same environment (his doubt on this point puts him in the forefront of behavioral genetics thinking). On the preceding page, however, he made the opposite point by positing a reason that Jack and Oskar are so similar: They were both raised in “households with a traditional German character.” Not the same household, but German, and this, he suggests, made the homes similar enough to bring about the twins’ long list of parallels.

  There is no particular reason why Beckwith, who is engaged in important molecular biological research unrelated to human behavior, should be well versed in the details of behavioral genetics research—unless, of course, he chooses to attack it. But attack it he and his coauthors do because of the dire consequences they see for liberalism in the implications of the Minnesota twin studies. While they may feel they help their cause with their half-baked attacks, tripping up by any means scientists with sinister motives, they are doing little for their own reputations for scientific thoroughness and accuracy.

  Bouchard has frequently addressed the criticism that twin similarity results from placement in socioeconomically similar homes. Aside from the obvious rebuttal that fraternal twins raised in the same home are not as similar as reared-apart twins, Bouchard makes an additional point. “If similar home environments made people similar,” he said, “close correlations would be found between twins and their adoptive parents (living in the same home). But a number of studies have shown that such correlations are not found.”

  Beckwith’s errors about the twin studies are typical of the criticisms, often from men and women who are meticulous in their own science, and from opponents like Beckwith, who, in spite of their hostility to behavioral genetics, would not consciously fabricate discrediting information. When, however, they shoot from the hip this way, their improvised and inaccurate swipes are given credibility by scientific reputations derived from unrelated research. Their attacks are picked up by a press that struggles to tell both sides of controversies, and the cry of “fatal flaws” is heard throughout the land.

  Bouchard is even more bothered that the critics segue quickly from pointing at weaknesses that might skew study results to claiming that these weaknesses have skewed results. And they do this, he points out, without offering any evidence that this is so. As mentioned above, they also do it in the face of abundant research data that have tested the premises and found them to have no effect. Yet the critics persist in the charge that these potential weaknesses undermine twin studies.

  THE ACCUSATION OF FLAWS in research is a common weapon in scientific debate but has particularly plagued twin and adoption studies. The complexity of the study designs—their unwieldy numbers, varying subject ages, varying environments, varying lengths of separation—have made them relatively easy targets for critics bent on inflicting harm. When, in spite of the lack of counterresearch, the critics refer darkly to “serious flaws,” or the most lethal, “fatal flaws,” these tidy phrases, so easy to put into the air, have a way of sticking, as does most destructive gossip. Sideline observers of the fray, even those with no stake in the matter, pick up the hurtful phrases and repeat them, unaware they are baseless. When in doubt about scientific matters, it is better to come down on the side of caution. As an added bonus, a pungent phrase like “fatal flaws” also lends a quick dose of self-promotion in that it places the speaker’s standards above those being debunked. In this way, much injustice is perpetrated against meticulous and scrupulous researchers, and much significant data brushed aside.

  Bouchard and other behavioral geneticists accept this as the inevitable price of working in a controversial field. They also see it as inevitable when studying a highly complex organism like the human, who does not fit into experiments as handily as mice and fruit flies. Some, like Minnesota’s Matthew McGue, view this carping in a positive light. “Behavioral genetics gets people excited because it is important. Most things don’t warrant any controversy because they just don’t matter, but our field is considered significant enough to argue about.”

  Robert Plomin is another behavioral geneticist who accepts the criticism as coming with the territory but wishes it were not so. He wistfully cites an illustration of his notion of the proper way to wage scientific battle. The eminent British psychologist Sir Michael Rutter was convinced that autism was environmentally caused and was distressed by the growing tide of scientific opinion that held it to be genetic. To prove his hunch correct and halt the tide, Rutter set up his own experiment. The results, to his surprise, proved the opposite of his expectations: Autism showed a significant genetic basis. Quite publicly, Rutter made it clear he had changed his mind completely on the subject. Rutter’s acquiescence to the evidence was, in Plomin’s view, the only honorable response for a scientist, but one sadly lacking among the radical environmentalists.

  Others in the behavioral genetics camp object to what they see as the unfairly high standards to which they are held as opposed to other branches of the social sciences. Sandra Scarr said, “Yes, we are complaining, not because of the final verdicts, but because the process is so manifestly unfair. Our work is subjected to the scrutiny of an electron microscope while the rest of psychology is examined through the wrong end of a telescope.” Scarr continued: “Leon Kamin approves of the close scrutiny because he sees dire consequences in the possible errors of genetic claims. I am sympathetic with his politics in this case, but my civil-libertarian bias impels me to claim the right to make as many mistakes as anyone else in psychology.”

  Scarr is underestimating Kamin’s opposition. He is not simply combating errors; he combats any evidence of a genetic role in human behavior. An examination of the indiscriminate ferocity of his attacks raises the question of whether he launches them because he believes there is no genetic role or because he feels such a belief is bad for business—that is, liberal politics. If pinned down, he acknowledges a genetic component to human behavior. Yes, he is saying, there may be a genetic component to behavior, but no, they haven’t proved it yet. His rhetoric implies something quite different: that genes are in no way involved in behavior and all efforts to establish a connection are bogus.

  Other beleaguered behavioral geneticists complain about the practice, when research is published, of publishing in the same issue vigorous rebuttals from one or more opponents. This too, they say, is a major deviation from normal scientific practice and appears an effort at preventing dangerous hereditarian positions to float unchallenged. When other claims of scientific advances are published, they are granted their moment in the sun—at least until the publication’s next issue, when letter writers can do their discrediting worst.

  The antigenes critics seek to deny behavioral geneticists even this degree of a hearing and do not always wait for papers to appear before launching their negative campaigns. Dean Hamer told me that when he was setting up his major study on genetic links to homosexuality, the gene police at Harvard got wind of it through a pamphlet Hamer issued to enlist volunteers. He received an immediate letter from biologist Ruth Hubbard, whose gene phobia was displayed in her 1970s efforts to organize Harvard office workers to demonstrate against recombinant DNA research. Hubbard told Hamer that his planned study was not the way to go about finding linkage (Hamer’s scientific c
redentials are every bit as good as Hubbard’s). A Harvard colleague, she told him, was in fact using the pamphlet in a course as an example of how not to conduct linkage studies.

  Hamer next got a letter from the colleague, Evan Balaban, asking Hamer for methodological details about his study. With less candor than Hubbard, Balaban said that his aim was to survey “conceptual advances” in behavioral genetics research. Still smarting from being “blind-sided by some Harvard professor I never met,” Hamer received yet another letter from Richard Lewontin, who was co-teaching the stamp-out-genes course with Balaban and was, he said, writing at the suggestion of Ruth Hubbard.

  Like Hubbard, Lewontin proclaimed his opposition to the methodology of Hamer’s study, stating that behavior was “very, very far from genes.” He cited language as an example of a behavior that is not influenced by genes, pointing to the world’s diversity of languages, which “we know for sure are not influenced by genes.” (This would surely come as jarring news to Noam Chomsky, Stephen Pinker, Myrna Gopnick, and other experts on language theory.) Lewontin also cited the failed effort to locate specific genes for various behaviors (alcoholism, for one) as reason to abandon this line of investigation.

  Hamer was stunned that prominent scientists were ganging up on a study that had not yet begun. More than attacking it, they were trying to stop it. When I met with Hamer in Washington, he expressed amusement at the disingenuousness of Balaban’s feigned interest in the gay-gene quest,2 but was less amused by Lewontin’s condescending lecture on the distance between genes and behavior. In responding to Lewontin’s point about diverse languages, Hamer said, “That’s like saying that the urge to eat is not genetic because some people like hot dogs and other people prefer tacos.”

  When he realized that the Cambridge group’s intentions were less friendly than Balaban had indicated, Hamer volunteered to give a guest lecture on his DNA study to the Harvard class, hoping to call their bluff about interest in his methods. To his surprise, they accepted his offer. At the lecture, Hamer outlined in careful detail what he saw as the criteria for a sound behavioral genetics study, stressing the statistical validity of teasing out genetic influence on complex behaviors by comparing the DNA of siblings who share the behavior. In his fascinating 1994 book on the entire gene-hunt adventure, The Science of Desire, Hamer describes the scene at his lecture’s end: