Bewitching Science by Val Dusek
[Note by Brian Siano: This article originally appeared in the November/December 1987 issue of _Science for the People_, published by the Science Resource Center, 897 Main Street, Cambridge, MA 02139. I don't even know if the magazine is still being published; it's a shame, because it regularly dealt with such topics as toxic wastes, nuclear power, eugenics, biotechnology, and the like. I have a few back issues, one of which has a dandy article in the psychological experiments of Dr. Ewen Cameron conducted in Canada. I wish a local newsstand still carried it. (BTW, its editorial advisory board includes Stephen Jay Gould and Richard Lewontin.) [Since this article appeared, Bouchard _has_ published his data, although I don't have the specific references for where he published. This article addresses his public statements prior to publishing, and raises important questions as to Bouchard's impartiality over his thesis. [Val Dusek, at the time of publication, taught philosophy at the University of New Hampshire.]
For seven years, popular magazines have regaled us with tales of Oskar and Jack, a pair of twins, one raised in Nazi Germany, the other raised as a Jew in Trinidad, who both think it funny to sneeze in elevators and always flush a toilet before using it. We have also been told about Bridget and Dorothy, British twins who each wore seven rings. These anecdotes issue from scientists undertaking a massive study of identical twins. The study, conducted by Thomas Bouchard and others at the University of Minnesota, is said to show that I.Q., personality traits, and virtually every other mental attribute or behavior is heritable, or capable of being inherited. During the past year, lengthy articles have appeared in U.S. News and World Report (a cover story), Discover, and Science. Shorter pieces have appeared in Time, U.S. News and World Report, the New York Times, and other magazines and newspapers. (1) The Minnesota Twins Study's "latest bombshell" (as U.S. News calls it) purports to show that traits such as shyness, political conservatism, dedication to hard work, orderliness, and intimacy are to a great extent heritable, and that extraversion, conformity, creativity, optimism, and cautiousness are more determined by heredity than by environment. Despite all the media coverage, the scientific data and methods of analysis upon which these conclusions are based have not yet been published in a refereed scientific journal. A December 1986 article in the New York Times and one in the January 12, 1987 issue of Time referred to results "submitted" and "being reviewed" by professional journals. However, in the August 7, 1987 issue of Science, no reference is made to any article having been submitted; it is said only that "the group recently has submitted a paper." This seems like a minor anomaly until one realizes that for the last seven years, Bouchard has been releasing announcements to the media regarding the Minnesota Twins Study and its results. The news section of Science has several times enthusiastically quoted Bouchard. Also since 1980, articles have appeared in Science 80, Newsweek, The New York Times, the New York Times Sunday Magazine, People, the New Orleans States-Item, the Washington Post, and elsewhere. (2) In these articles, traits from political conservatism to toilet flushing have been claimed to be heritable. Bouchard has declared his results "devastating" to feminists. (3) Opponents have been termed "ideological." (4) U.S. News stated, "Unable to hold back the swelling tide of evidence for the importance of genes, supporters of the nurture side try to fight back with words." (5) This public trumpeting of 'science' without data is perhaps the most extreme recent example of popular media releases without scientific publication. Given the popular interest in Bouchard's alleged results and their purported policy impact for child rearing, social welfare programs, the criminal justice system, and the schools, this situation raises questions of ethics and responsibility. These questions relate to not only Bouchard and the Minnesota Twin Study group in releasing these so-far unverifiable claims. They also relate to the journalists who uncritically convey the study's claims and the members of the relevant subdisciplines within the scientific community who have not raised critical notice concerning the twins study and its media coverage.
Persecuted Galileos? Doctrines of hereditary I.Q., race differences in I.Q., sex differences in math ability, the sociobiology of aggression and sex roles, and other aspects of biological determinism have flooded the media. In these media presentations, hereditarians play a double game. On the one hand, they claim to be 'pure scientists,' above the political battle. On the other hand, they are not shy in hyping their doctrines to the popular press, and have never, to my knowledge, criticized a favorable presentation of sociobiological doctrine, no matter how vulgar and distorted it may be. The biological determinists often present themselves as persecuted Galileos of science. But they do not hesitate to make policy pronouncements on such topics as the inferiority of black intelligence, the inability of women to pursue careers in science and the law, the ineffectiveness of attempts to educate the disadvantaged, or the 'naturalness' of female depression, rape, capitalism, and war. (6) However, biological determinists tend to claim that their own views are purely scientific, while their opponents' views are purely ideological. Bouchard, his co-workers, and supporters follow this pattern. According to Science, "Bouchard wants to keep his study free from politics." But in the same article, Bouchard is also quoted as saying that his German twins are "devastating to the feminist contention that children's personalities are shaped differently according to the sex of those who rear them, since Oskar was raised by women and Jack by men." (3) Thus, in a sample of one pair of twins, Bouchard is willing to draw conclusions concerning child rearing and sexual politics. Many biological determinists portray themselves as liberals who were brought by the 'harsh facts' of biology to hold conservative doctrines. Sociobiologist E.O. Wilson, psychologist of inherited criminality Sanford Mednick, and others have made this claim. Bouchard is no exception. Despite his discipleship to the scientific racist Arthur Jensen at Berkeley, Bouchard claims to have been engaged in "political activism in the radical sixties." (7) Bouchard also presents himself as having stumbled "almost casually" in 1979 into an interest in twins through reading about a pair of reunited twins. (8) In fact, Bouchard had already published research and review articles years before on the heritability of I.Q. From this work and that of his mentor, Jensen, Bouchard must have realized the centrality of studies of twins reared apart for the I.Q. debate. This importance greatly increased after Cyril Burt's data, a major basis for Jensen's claims concerning black/white I.Q. differences, was discredited as fraudulent. (910) Inferences From Coincidences Despite the claims concerning hard evidence, large samples, and the appeal to the biological sciences, what we find in statements by Bouchard and in material released to the media from the Minnesota Twin Study are anecdotes and amazing stories. What is striking about the anecdotal material is its similarity to the sort of evidence often offered as proof for astrology or extrasensory perception (ESP). Striking coincidences are reported as supposed grounds for belief in the phenomenon itself. In literature about astrology and ESP, cases where forecasts came true or where thought of a friend was immediately followed by a phone call from that friend are offered as evidence. The cases where forecasts failed or where a thought of someone is not followed by a phone call from that person are forgotten or left unmentioned. Bouchard's coincidence anecdotes are of a similar nature: we are told about the similarities (seven rings on fingers, sneezing in elevators) but not about the differences. But some of the similarities are physical ones that are to be expected in identical twins. Other behavioral similarities are not all that amazing. Two twins living east and west of the Mississippi turn out later to live on opposite sides of the river in Louisiana. Even if "the mighty Mississippi divided" the twins, the fact that they both wear cowboy hats and like hunting is not that unusual for two working-class men in the same region of Louisiana.11 Oskar and Jack, the Nazi and Jew -- superficially the most spectacular case of twins reared apart -- both had less isolation from each other and less different environments than the media stories reveal. They were raised by their own relatives in two German households. One of these households emigrated to Trinidad. Bouchard himself admits that their household environments were more similar than their Nazi-vs.-Jew image suggests. In fact, the two men met briefly during the 1950s in Germany, and their wives kept up correspondence since that meeting. (11) Bouchard notes that one function of the media publicity about spectacular coincidences is to recruit more pairs of twins. But such pairs may wish to exaggerate similarities of behavior or wear identical dress to receive publicity and scientific approval for themselves. This sort of recruitment bias has occurred in some earlier twin studies. Some of the coincidences recalled can have nothing to do with the twins' genetics, such as twins being adopted by families which had adoptive brothers with the same name or the twins themselves being given the same name by their adoptive families. (12) Even the language of twin study reportage is similar to that concerning the occult. One of Bouchard's co-workers says that they were still "bewitched by the seven rings." (13) Discover magazine's front cover introduces us to "The Eerie World of Reunited Twins." While admitting that "Genes do not cause fires," one popular book entitled Twins: Nature's Amazing Mystery moves easily between enthusiastic reports of Bouchard's coincidences and discussions of telepathic communication between twins and synchronous events such as fires in the lives of distant twins. (15) The anecdotes that Bouchard relates would seem more at home on the pages of the National Enquirer than in those of Science. It is ironic that Bouchard, in his reviews of the critics of the twin studies, dismisses their work as ad hoc and unscientific. (14) In reviewing Howard Gardner's criticism of I.Q. tests, Bouchard says. "This book is primarily an opinion piece, a collection of anecdotes... Gardner's scheme is not, however, a theory in the rigorous (or even the non-rigorous) scientific sense." (14) This remark is particularly ironic since all that Bouchard has so far released are anecdotes of strange coincidences that "struck" him. Given that the largest study of identical twins reared apart fraudulent by even Burt's students and admirers, and that earlier studies of twins are replete with tester and surveyor bias,15 it would seem especially desirable that Bouchard and the Minnesota group open to public scientific scrutiny their data and experimental design. However, all we have in the popular reports are assertions of the heritable nature of various traits and anecdotes concerning a few of the twin pairs. The only paper in a refereed journal which makes use of the Minnesota Twin Study data is a study of homosexuality in twins reared apart. (16) This study relies on the huge data base of six pairs of twins -- four pairs of females and two pairs of males. Both members of one of the male pairs are gay. Only one member of the other male pair is gay. Of the four pairs of female twins, only one member each is lesbian or bisexual and one member each is heterosexual. From these results, Bouchard and McGue conclude that male homosexuality has a strong heritable component, while lesbianism does not. That such a grand conclusion can be drawn from this sample of two gay male twins is even more mind-boggling than some of the coincidences that Bouchard relates. The Science review of earlier I.Q. correlation studies 11 and the study on the heritability of homosexuality are the only articles in peer-reviewed journals closely relevant to or based upon the twin study material. A central feature of science is its public and critical nature. Scientific data, unlike the lore and traditions of some religious cults or such esoteric practices as alchemy, are made publicly available in journals whose contents are reviewed, evaluated, and published by members of the scientific community. Peer review is meant to subject articles to critical scrutiny prior to being accepted as worthy of publication. Despite the fact that peer review does not always function to ideal effectiveness, it is better than outright cronyism or nepotism. Once the scientific article has been deemed worthy of publication by a group of fellow scientists, the publicly available account of data and methods is available to the entire scientific community for further examination and criticism. Methods of data collection, sources of sample populations, statistical techniques. and the logic by which conclusions are drawn can be carefully analyzed and criticized by other scientists. The failure of Bouchard and his colleagues in the Minnesota Twin Study to participate in the peer review process is an extreme example of circumventing the scientific process and using the media for public relations. But scientists in competitive fields such as high-energy physics, genetic engineering and medicine have also announced their discoveries to the press before they are published in the organs of the scientific community. Editors of the New England Journal of Medicine and Physical Review Letters have complained about this practice, (17) and have tried to discipline scientists who publish in the popular press before their work is refereed by other scientists through refusal of publication in their journals. For seven years, Bouchard and the Minnesota group have been announcing their 'conclusions' concerning the heritable nature of personality traits. They have been relating anecdotes of coincidences to convince the general public that subtle characteristics such as "beringedness" (wearing rings) have heritable predispositions and that complex behaviors, such as double toilet-flushing, sneezing in elevators, and naming one's dog Toy, are relatively independent of upbringing and environment. Most recently, the Minnesota group has released a list of group's representatives have also expounded on such topics as the heritable nature of Chuck Yeager's bravery (although Yeager is not known to be a subject of their survey). (18) It is possible that Bouchard's survey is exhaustive and his logic impeccable. But as long as the Minnesota Twin Study does not publish its data and the methodological basis for its conclusions in a peer review journal, we cannot tell. To investigate the background, upbringing, and circumstances of recruitment for the twins involved in Bouchard's research, a book-length study would first have to be released. The Discover article promises such a book by 1989, but by the time critical evaluations are published by scientists, a decade of media coverage will have made its impression. The media anecdotes about "eerie" and "freakish" coincidences that "struck" Bouchard must remain on a par with tales about astrology and ESP. And Bouchard's data and methods must remain in that limbo in which Cyril Burt's imaginary assistants and unverifiable data existed.
Notes
1. "The Eerie World of Reunited Twins," Discover, September 1987; "How Genes Shape Personality," U.S. News and World Report, April 13, 1987; "The Genetics of Personality," Science, vol. 237, 1987; "Exploring the Traits of Twins," Time, Jan 12, 1987; "Genes: Little Things that Mean a Lot," U.S. News and World Report, Dec. 15, 1986; "Major Personality Study Finds that Traits Are Mostly Inherited," The New York Times, Dec. 1, 1986.
2. "Twins Reunited," Science 80, Nov. 1980; "Identical Twins Reared Apart," Science, vol. 207, 1980; "Twins, Nazi and Jew," Newsweek, Dec. 3, 1979; "Twins Reared Apart, a Living Lab," New York Times Sunday Magazine, Dec. 9, 1979; "Two Ohio Strangers Find They're Twins at 39 -- and a boon to Psychologists," People, May 7, 1979; "The Twins," States-Item, Feb. 25-29, 1980; "Me, Myself, and Us: Twins," Science Digest, Nov./Dec. 1980.
3. op. cit. Science, vol, 207, 1980.
4. Bouchard's coworker David Lykken quoted describing Leon Kamin as one of the "psychologists who object to genetic research on ideological grounds" and "do not understand its true implications." U.S. News, Dec. 15, 1986.
5. op. cit. U.S. News, April 13, 1987.
6. For a sample of the history and criticism of biological determinist doctrines, see: Biology as a Social Weapon, the Ann Arbor Science for the People Editorial Collective; The Mismeasure of Man by Stephen Jay Gould; Not In Our Genes, by R. C. Lewontin, Steven C. Rose, and Leon Kamin; Biology As Destiny, Science for the People Sociobiology Study Group.
7. op. cit Science 80, November 1980.
8. op. cit Discover, Sept. 1987.
9. L.S. Hearnshaw, Cyril Burt, Psychologist, 1979. Hearnshaw, eulogist at Burt's funeral and "official" biographer of Burt, was only reluctantly led to his conclusions that Burt invented both data and research assistants. Leon Kamin (The Science and Politics of I.Q., 1974) had already raised doubts about Burt's data. Interestingly, the only person who by 1980 still seemed to have doubts that Burt's data were fraudulent was Science journalist Constance Holden, author of three articles in praise of Bouchard (Science 80, Nov 1980, and Science, vol. 207, 1980, to cite two).
10. A look at Bouchard's previous publications in psychology does not increase one's trust in the so-far-unpublished twin data. Bouchard and McGue's "Familial Studies of Intelligence: A Review," (Science, vol 212, 1981) reviews previous studies of correlations of I.Q. among relatives, omitting Burt's discredited studies. This article is obviously meant to show that despite the loss of Burt's supposed data, there is a large body of work on which hereditarians can base their assertions. The survey has many faults. One is that Bouchard and McGue do not mention or bother to deal with the faults already found in the early studies that they resurrect (dating back to the 1920s, and largely from the 1930s and 40s). Many of these studies were biased in their methodology and reported as "separated from birth" twins who actually lived next door to one another, attended the same school, played together, and had frequent social interaction. These studies are also vitiated by neglecting to correct for the age bias in I.Q. tests. Despite the fact that I.Q. is supposedly corrected for age, the I.Q. tests used in these studies show I.Q. rising with age. Thus, part of the weaker correlation between nontwin siblings than between twins arises from the fact that twins are exactly the same age, while other siblings may differ in age. Finally, Bouchard and McGue simply pooled the samples from very different tests and from tests which gave extraordinarily divergent results. For instance, one test of siblings gave an I.Q. correlation of 10 percent, while another test gave a correlation of 90 percent. Bouchard and McGue simply averaged the two to give a correlation of 50 percent. Given the radically opposite results of the two surveys, it is likely that they were performed with radically different biases and methodologies. They could not have been randomly sampling two subpopulations of the same homogeneous population of data -- a basic requirement for drawing valid statistical inference.
11. Cassil, Kay. Twins: Nature's Amazing Mystery. 1982.
12. Ibid. p. 180. Also, op. cit Discover, Sept. 1987.
13. Op. cit. Cassil, p. 134-5, 158-164, 189.
14. Bouchard, Review of Howard Gardner's "The Intelligence Controversy."American Journal of Psychology, Vol. 95, 1987.
15. Kamin, Leon. The Science and Politics of I.Q. 1974.
16. Eckert, Bouchard, Bohlen, and Heston, "Homosexuality in Monozygotic Twins Reared Apart," British Journal of Psychiatry, vol. 148, 1986.
17. "Gene Cloning by Press Conference," New England Journal of Medicine, March 27, 1980: New York Times article, Nov. 18, 1974.
18. U.S. News, Dec. 15, 1986, and Time, Jan. 12, 1987. David Lykken is the source of the claim about Chuck Yeager.
Sidebar: "Financing Racist Research"
The first New York Times report about the Minnesota Twin study quoted Bouchard as saying, "I'm going to beg, borrow, and steal" to pursue the twin study. In fact, Bouchard has solicited money from the Pioneer Fund, a foundation with racist and radical right-wing connections. the University of Minnesota has received grants from the fund for Bouchard's twin study. Butthe Pioneer Fund is best known for its support of research purpoting the inferiority of blacks. Once headed by directors such as the Chairman of the House Committee on UnAmerican Activities, Representative Francis E. Walter, and Mississippi Senator James O. Eastland, the fund has long subsidized research and publication of the works of scientific racists, including William Shockley and Arthur Jensen, Jensen served on the scientific advisory board of the German Neo-Nazi journal Newe Anthropologie. (SeeBarry Mehler's article "The New Eugenics" in the May/June 1983 issue of _Science for the People_.) The Pioneer Fund financed the work of Roger Pearson, quthor of _Eugenics and Race_. Pearson also helped organize the 1978 World Anti-Communist League meeting in Washington, D.C. The League has united old European Nazis with leaders of Third World death squads. Bouchard, in his grant application to the Pioneer Fund, noted that the National Science Foundation has repeatedly refused funding for his study and has made numerous criticisms of his method. Bouchard has claimed that the NSF and the National Institutes of Health are packed with left liberals who deny him funds on ideological grounds.
[Additional Commentary by Brian Siano: In fairness to the Pioneer Fund, they also provided some funding for _The Atomic Cafe_, a savagely funny documentary about the ridiculous claims on the harmlessness of nuclear war circulated in the 1950s. It's hard to call this film 'right wing.' [As for the comments on the World Anti-Communist League, they're certainly true; the WACL was even condemned by the John Birch society as being too fanatical. A good resource on this organization (which has numbered Roberto D'Aubuisson, Rev. Sun Myung Moon, Ferdinand Marcos and John Singlaub as its members) is _Inside The League_, by Scott Anderson and Jon Lee Anderson. [Roger Pearson deserves a study by himself. Many of his books (some still sold my the American Nazi Party) argue classic racist themes, mainly against the dilution of the white race's genetic stock through intermarriage with blacks and Jews. _Inside the League_ provides a quick thumbnail study of Pearson's views and activities.]