In and era when researchers Have located the genes for conditions such as Huntington's disease, cystic fibrosis, and Duchenne muscular dystrophy, some scientists believe that "homosexual" genes will soon be found. It comes as no surprise, then, that lately the popular press has been humming with stories about whether homosexuality is genetically determined.
We are uneasy about the current unbridled enthusiasm for studies relating genes with human behavior. Scientists' arguments for a biological basis for human differences have previously been used for insidious ends; the arguments by German scientists before World War II for the genetic inferiority of Jews is just one example.
Moreover, much of the older scientific analysis of the origins of human behavior, particularly using biological methods, has been debunked. In the nineteenth century, for instance, "phrenologists" claimed that they could predict aspects of an individual's personality, such as sexuality, intelligence, and criminal tendencies, merely by examining the skull's structure. Despite its popularity, this "science," which often included explicitly racist implications, was not based on any reliable evidence. More recently, after studies in prisons in the 1960s, geneticists jumped to the conclusion that males with an extra Y chromosome were more likely to be criminals than other men. Followup studies in the general population showed that this claim was unwarranted.
But that is the past. It's always possible that the field of human behavioral genetics has shaken off its tawdry history. Today, when sophisticated techniques can be used to analyze human DNA, maybe the renewed interest in connecting biology and behavior portends the development of a more scientific era. On the other hand, maybe not. A look at recent studies seeking a genetic basis for homosexuality suggests that many of the problems of the past have recurred. We may be in for a new molecular phrenology rather than true scientific progress and insight into behavior.
Even for the living, scientists must decide whether, when trying to classify a person as gay, to rely on the frequency of homosexual behavior, the age at which it began, or, as some psychiatrists have suggested, the presence of homoerotic fantasies. Compounding this difficulty is the reluctance of many gay people to admit their sexual preference. In this situation one can easily create inaccurate or inconsistent study groups.
A related problem cropped up in a study by Michael Bailey, a psychologist at Northwestern University, and Richard Pillard, a psychiatrist at Boston University School of Medicine. In research published in 1991 in Archives of General Psychiatry, they reported that among identical twins reared together, each individual had a high probability (52 percent) of being homosexual if the other twin was. For nonidentical twins, adopted brothers, and non-twin siblings, the correlations were much lower (22, 11, and 9 percent, respectively).
The researchers solicited participation for their study through advertisements in gay newspapers. But this could well have produced a biased sample. It is not unreasonable to think that identical twins who are both openly homosexual would be more likely to volunteer for such a study because of the shared knowledge that they are both comfortable with their sexuality. Furthermore, if one gay twin "comes out" it might be easier for the other to do the same.
While the authors interpreted their findings as evidence for a genetic basis for homosexuality, we think that the data in fact provide strong evidence for the influence of the environment. On average, both non-identical twins and non-twin siblings share 50 percent of their genes. If homosexuality were a genetic trait, the pairs in these groups should besexual a similar percentage of the time. They certainly should be homosexual more often than adopted siblings. But Bailey and Pillard's data do not fit those predictions.
Their results could instead stem from the uniform conditions under which some identical twins are raised, frequently dressed identically by parents and viewed similarly by outsiders. The environment could also figure in the fact that almost half the identical twins did not share sexual preference.
The problem of distinguishing between cause and effect is perhaps most clearly illustrated by considering genetics and criminality. There is a correlation between genes and the likelihood of incarceration in the United States: a majority of prisoners have inherited dark skin color. But genetics clearly has nothing to do with the fact that the rate of black male incarceration has quadrupled over the last four decades, as Troy Duster, a University of California at Berkiologist, has pointed out. Rather, racism and the tenuous economic power of so much of the African-American population likely play causal roles.
Studies of human behavioral genetics could benefit society. But until geneticists pay more than lip service to the problems in their studies and the complex interactions of genes and the environment, history may simply repeat itself.
It is time to hold behavioral-genetics research to higher standards than in the past, and to recognize that the work is conducted in a society colored by prejudice, stigma, and discrimination.
PAUL BILLINCS, formerly chief of the Division of Genetic Medicine at Califomia Pacific Medical Center in San Francisco, is now head of general internal medicine at the Palo Alto Veterans Administration Hospiral. JONATHAN BECKWITH is American Cancer Society research professor in the Department o f Microbiology and Molecular Genetics at Harvard Medical School.