Dave Miller, Ph.D.
Brad Harrub, Ph.D.
BAILEY AND PILLARD—
THE FAMOUS “TWINS” STUDY
THE FAMOUS “TWINS” STUDY
One of the most frequently cited studies used in promoting the genetics of sexual orientation is a 1952 study by Kallmann. In this famous work, he reported a concordance rate (or genetic association) of 100% for sexual orientation among monozygotic (identical) twins (1952, 115:283). This result, if true, would prove nearly insurmountable for those people who doubt the biological causation of homosexuality. However, Kallmann subsequently conjectured that this perfect concordance was an artifact, possibly due to the fact that his sample was drawn largely from mentally ill and institutionalized men (see Rainer, et al., 1960, 22:259). But Kallmann’s research opened the door to twin studies in regard to sexual orientation.
Michael Bailey and Richard Pillard, researchers at Northwestern University and the Boston University School of Medicine, carried out a similar experiment, examining 56 pairs of identical twins, 54 pairs of fraternal twins, 142 non-twin brothers of twins, and 57 pairs of adoptive brothers (1991, 48:1089-1096). Bailey and Pillard were looking to see if homosexuality was passed on through familial lines, or if one could point to environmental factors as the cause. Their hypothesis: if homosexuality is an inherited trait, then more twin brothers would be expected to have the same orientation than non-twin or non-biological brothers.
Their Reported Findings
- 52% of identical (monozygotic) twins of homosexual men were homosexual
- 22% of fraternal (dizygotic) twins were likewise homosexual
- 11% of adoptive brothers of homosexual men were homosexual
- 9.2% of non-twin biological siblings reported homosexual orientations (Bailey and Pillard, 1991, “A Genetic Study of Male Sexual Orientation”)
- 48% of identical twins of homosexual women were likewise homosexual
- 16% of fraternal (dizygotic) twins were likewise homosexual
- 6% of adoptive sisters of homosexual women were likewise homosexual (Bailey and Benishay, 1993, “Familial Aggregation of Female Sexual Orientation”)
Problems with Bailey and Pillard’s Study
While the authors acknowledged some of the flaws with their research, they still were quoted in Science News as saying: “Our research shows that male sexual orientation is substantially genetic” (as quoted in Bower, 1992, 141:6). However, the most glaring observation is that clearly not 100% of the identical twins “inherited” homosexuality. If there was, in fact, a “gay gene,” then all of the identical twins should have reported a homosexual orientation. And yet, in nearly half of the twins studied, one brother was not homosexual. In a technical-comment letter in Science, Neil Risch and colleagues pointed out: “The biological brothers and adoptive brothers showed approximately the same rates. This latter observation suggests that there is no genetic component, but rather an environmental component shared in families” (1993, 262:2063). In fact, more adoptive brothers shared homosexuality than non-twin biological brothers. If there was a genetic factor, this result would be counter to the expected trend. Byne and Parsons noted:
However, the concordance rate for homosexuality in nontwin biologic brothers was only 9.2—significantly lower than that required by simple genetic hypothesis, which, on the basis of shared genetic material, would predict similar concordance rates for DZ [dizygotic] twins and nontwin biologic brothers. Furthermore, the fact that the concordance rates were similar for nontwin biologic brothers (9.2%) and genetically unrelated adoptive brothers (11.0%) is at odds with a simple genetic hypothesis, which would predict a higher concordance rate for biological siblings (1993, 50:229).
A more recently published twin study failed to find similar concordance rates. King and McDonald studied 46 homosexual men and women who were twins. The concordance rates that they reported were 10%, or 25% with monozygotic twins—depending on whether or not the bisexuals were included along with the homosexuals. The rates for dizygotic twins were 8% or 12%, again, depending on whether bisexuals were included (King and McDonald, 1992). Byne and Parsons commented: “These rates are significantly lower than those reported by Bailey and Pillard; in comparison of the MZ [monozygotic] concordance rate, including bisexuals (25%), with the comparable figure from Bailey and Pillard (52%)” (p. 230). They went on to observe: “Furthermore, if the concordance rate is similar for MZ and DZ twins, the importance of genetic factors would be considerably less than that suggested by Bailey and Pillard” (p. 230, emp. added).
Another factor that may have had a drastic affect on the results of this study (and other similar studies) centers on methodology. Bailey and Pillard did not study a random sample of homosexuals. Instead, the subjects were recruited through advertisements placed in homosexual publications. This method can be deemed questionable because it is highly dependent on the readership of those publications and on the motives of those who respond. Thus, it may lead to skewed results—for example, inflated rates of concordance in identical twins owing to preferential participation (see Baron, 1993). Hubbard and Wald observed:
The fact that fraternal twins of gay men were roughly twice as likely to be gay as other biological brothers shows that environmental factors are involved, since fraternal twins are no more similar biologically than are other biological brothers. If being a fraternal twin exerts an environmental influence, it does not seem surprising that this should be even truer for identical twins, who the world thinks of as “the same” and treats accordingly, and who often share those feelings of sameness (1997, p. 97).
In summarizing their findings, Byne and Parsons stated: “Critical review shows the evidence favoring a biologic theory to be lacking” (50:228). Commenting on Bailey and Pillard’s report, researchers Billings and Beckwith wrote:
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 (1993, p. 60).
When evaluated scientifically, twin studies fail to provide any valid support for the longed-for “gay gene.”
DEAN HAMER—THE GAY GENE
ON THE X CHROMOSOME
ON THE X CHROMOSOME
Two years after Simon LeVay’s report, a group led by Dean H. Hamer of the National Cancer Institute allegedly linked male homosexuality to a gene on the X chromosome. His team investigated 114 families of homosexual men. Hamer and his colleagues collected family history information from 76 gay male individuals and 40 gay brother pairs as they searched for incidences of homosexuality among relatives of gay men.
In many families, gay men had gay relatives through maternal lines. Thus, they concluded that a gene for homosexuality might be found on the X chromosome, which is passed from the mother alone. They then used DNA linkage analysis in an effort to find a correlation between inheritance and homosexual orientation.
Their Reported Findings
Because many of the families with a prevalence of homosexual relatives had a common set of DNA markers on the X chromosome, Hamer’s group assumed a genetic etiology. Of the 40 pairs of homosexual brothers he analyzed, Hamer found that 33 exhibited a matching DNA region called q28—a gene located at the tip of the long arm of the X chromosome. In summarizing their findings, Hamer and colleagues noted: “Our experiments suggest that a locus (or loci) related to sexual orientation lies within approximately 4 million base pairs of DNA on the tip of the long arm of the X chromosome” (1993, 261:326, parenthetical item in orig.). This discovery prompted Hamer and his colleagues to speculate:
The linkage to markers on Xq28, the subtelomeric region of the long arm of the sex chromosome, had a multipoint lod score of 4.0, indicating a statistical confidence level of more than 99 percent that at least one subtype of male sexual orientation is genetically influenced (261:321, emp. added).
It is important to note that Hamer did not claim to have found a “gay gene,” or even the set of genes, that might contribute to a propensity for homosexuality. According to Chicago Tribune staff writer, John Crewdson, what Hamer claimed to have found was “statistical evidence that such genes exist” (1995).
Problems with Hamer’s Study
One of the most significant problems with Hamer’s approach is that he and his colleagues did not feel that it was necessary to check whether any of the heterosexual men in these families shared the marker in question! Would it not be useful to know whether or not this “gay gene” is found in heterosexuals? Even if only a few of them possess the gene, it calls into question what the gene or the self-identification signifies. Additionally, Hamer never explained why the other seven pairs of brothers did not display the same genetic marker. If this is “the gene” for homosexuality, then one must assume all homosexual individuals would possess that particular marker—and yet that was not the case in Hamer’s study.
In a letter to Science, Anne Fausto-Sterling and Evan Balaban pointed out some of the additional problems with Hamer’s study. They noted:
Despite our praise for aspects of Hamer, et al.’s work, we feel it is also important to recognize some of its weaknesses. The most obvious of these is the lack of an adequate control group. Their study demonstrates cosegregation of a trait (which Hamer, et al. have labeled “homosexuality”) with X chromosome markers and the trait’s concordance in homosexual brothers. This cosegregation is potentially meaningful if the mother is heterozygous for the trait. In this case, segregating chromosomes without the markers should show up in nonhomosexual brothers, but Hamer, et al. present no data to that effect (1993, 261:1257, emp. added).
Fausto-Sterling and Balaban continued:
This sensitivity to assumptions about background levels makes Hamer, et al.’s data less robust than the summary in their abstract indicates.... Finally we wish to emphasize a point with which we are sure Hamer, et al. would agree: correlation does not necessarily indicate causation (261:1257).
In other words, Hamer’s methodology leaves something to be desired. One also should keep in mind that Hamer’s sampling was not random, and, as a result, his data may not reflect the real population.
George Rice and his colleagues from Canada looked intently at the gene Xq28. They then observed: “Allele and halotype sharing for these markers was not increased over expectation. These results do not support an X-linked gene underlying male homosexuality” (1999, 284:665, emp. added). Rice, et al., included 182 families in their study. They noted:
It is unclear why our results are so discrepant from Hamer’s original study. Because our study was larger than that of Hamer et al., we certainly had adequate power to detect a genetic effect as large as was reported in that study. Nonetheless, our data do not support the presence of a gene of large effect influencing sexual orientation at position Xq28 (284:667).
That is a tactful way of saying that any claims of having found a “gay gene” were overblown, if not outright false, and that Hamer’s results are dubious at best. Commenting on the study of Rice and his colleagues, Ingrid Wickelgren remarked: “...the Ontario team found that gay brothers were no more likely to share the Xq28 markers than would be expected by chance.... Ebers interprets all these results to mean that the X linkage is all but dead” (1999, 284:571, emp. added).
In June of 1998, University of Chicago psychiatrist Alan Sanders reported at the meeting of the American Psychiatric Association that he, too, had been unable to verify Hamer’s results. Looking for an increase in Xq28 linkage, Sanders’ team studied 54 pairs of gay brothers. As Wickelgren indicated, Sanders’ team had found “only a weak hint—that wasn’t statistically significant—of an Xq28 linkage among 54 gay brother pairs” (284:571). Commenting on the validity of Hamer’s study, Wickelgren quoted George Rice: “Taken together, Rice says, the results ‘suggest that if there is a linkage it’s so weak it’s not important’” (1999, emp. added). Two independent labs failed to reproduce anything even remotely resembling Hamer’s results.
CHANGEABILITY OF HOMOSEXUALS—
EVIDENCE AGAINST GENETICS
EVIDENCE AGAINST GENETICS
An individual born with diabetes has no hope of changing that condition. Likewise, a child born with Down’s syndrome will carry that chromosomal abnormality throughout his or her life. These individuals are a product of the genes they inherited from their parents. Homosexuality appears to be vastly different. Many people have been able to successfully change their sexual orientation. [Truth be told, some individuals experiment with a variety of sexual partners—male/female—often, going back and forth. One might inquire if the bisexuality denotes the existence of a “bisexual gene?”] Ironically, however, the removal of homosexuality as a designation from the Diagnostic and Statistical Manual of Psychiatric Disorders by the American Psychiatric Association has kept many physicians from attempting to provide reparative therapy to homosexuals.
Robert Spitzer conducted a study on 200 self-selected individuals (143 males, 57 females) in an effort to see if participants could change their sexual orientation from homosexual to heterosexual (2003, 32:403-417). He reported some minimal change from homosexual to heterosexual orientation that lasted at least five years (p. 403). Spitzer observed:
The majority of participants gave reports of change from a predominantly or exclusively homosexual orientation before therapy to a predominantly or exclusively heterosexual orientation in the past year (p. 403).
In summarizing his findings, Spitzer declared: “Thus, there is evidence that change in sexual orientation following some form of reparative therapy does occur in some gay men and lesbians.” He thus concluded: “This study provides evidence that some gay men and lesbians are able to also change the core features of sexual orientation” (p. 415).
Six years earlier, the National Association for Research and Therapy of Homosexuality (NARTH) released the results of a two-year study stating:
Before treatment, 68 percent of the respondents perceived themselves as exclusively or almost entirely homosexual, with another 22 percent stating that they were more homosexual than heterosexual. After treatment, only 13 percent perceived themselves as exclusively or almost entirely homosexual, while 33 percent described themselves as either exclusively or almost entirely heterosexual (see Nicolosi, 2000, 86:1071).
The study also reported:
Although 83 percent of respondents indicated that they entered therapy primarily because of homosexuality, 99 percent of those who participated in the survey said they now believe treatment to change homosexuality can be effective and valuable (p. 1071).
These data are consistent with the ongoing research project of Rob Goetze, who has identified 84 articles or books that contain some relevance to the possibility of sexual orientation change (2004). Of the data reported, 31 of the 84 studies showed a quantitative outcome of individuals able to change sexual orientation. These are not studies that merely speculate on the ability to change; they actually have the numbers to back it up! All of these data come on the heels of warnings from the Surgeon General, The American Academy of Pediatrics, and all of the major mental health associations, which have issued position statements warning of possible harm from such therapy, and have asserted that there is no evidence that such therapy can change a person’s sexual orientation. For instance, the 1998 American Psychiatric Association Position Statement on Psychiatric Treatment and Sexual Orientation noted:
...there is no published scientific evidence supporting the efficacy of reparative therapy as a treatment to change one’s sexual orientation.... The potential risks of reparative therapy are great, including depression, anxiety, and self-destructive behavior (see American Psychiatric Association, 1999, p. 1131).
Thus, physicians are caught in a quandary of a double standard. On the one hand, they are told that it is “unethical” for a clinician to provide reparative therapy because homosexuality is not a diagnosable disorder, and thus one should not seek to change. Yet, they contend that not enough studies have been conducted to determine the effectiveness of reparative therapy. The message is loud and clear: “Do not do this because it is unethical to ask a homosexual person to change. However, truth be told, we have not collected enough data to know if a person can safely change his or her sexual orientation.”
In situations where sexual orientation is being measured, studies face serious methodological problems (i.e., follow-up assessment, possible bias, no detailed sexual history, random sampling, etc.). But even given these serious shortcomings from behavioral studies such as these, there are sufficient data to indicate that an individual can change his or her sexual orientation from homosexual to heterosexual—something that would be an impossibility if homosexuality were caused by genetics.
Consider the obvious problem of survival for individuals who allegedly possess a gay gene: individuals who have partners of the same sex are biologically unable to reproduce (without resorting to artificial means). Therefore, if an alleged “gay gene” did exist, the homosexual population eventually would disappear altogether. We now know that it is not scientifically accurate to refer to a “gay gene” as the causative agent in homosexuality. The available evidence clearly establishes that no such gene has been identified. Additionally, evidence exists which documents that homosexuals can change their sexual orientation. Future decisions regarding policies about, and/or treatment of, homosexuals should reflect this knowledge.