Appendix on Biological Studies
from "On the Position on Homosexuality in the New Draft Programme"
The biological research seeking to uncover possible biological bases underlying sexual orientation (or directed at exploring related questions) is extensive and substantial and cannot be discussed with any degree of thoroughness in the context of this paper. For now we will limit ourselves to making just a few brief observations about the direction studies have taken and the types of problems they have encountered, and will refer readers to some additional references which they might find useful should they wish to study more about all this on their own.
Discussions of the supposed biology of sexual orientation have tended to focus on the results of studies in three distinct fields of biology: neuroanatomy, genetics, and the study of hormones and behavior. In the field of neuroanatomy and related areas much was made in the early '90s when Simon LeVay claimed to have uncovered substantial differences in the morphology (size and shape) of particular sections of the brains of men vs. women on the one hand, and heterosexuals vs. homosexuals on the other. This led to all sorts of speculation about the possible functional influences (on sexual orientation) of these supposed morphological differences. Not surprisingly LeVay's claims have been substantially criticized for conceptual, methodological, experimental and interpretive errors (see Gould 1981 for more on the uninspiring history of such "brain shape and size" studies and Fausto-Sterling and Stein for specific discussions of the problems with LeVay's and similar studies) . Furthermore, attempts by other scientists to replicate and confirm his findings have essentially been unsuccessful.
In the field of heredity and genetics, much has been made in recent years of studies of family trees of homosexuals which--at least at first--strongly suggested that especially male homosexuality "ran in families" and which further suggested that the hypothetical biological "trait" for homosexuality seemed to be getting passed on through the female line in a family, in a pattern typical of what are called "X-linked recessive traits." Very frequently cited studies of the self-reported family histories of male homosexuals having twin brothers who were also homosexual at a rate which (at least at first) seemed strikingly higher than the rate of homosexuality in the general population (though it is always difficult to ascertain what that rate is) also generated tremendous buzz, and the popular press went way overboard trumpeting that homosexuality had now been "proven" to be a genetic and inheritable trait. The twin studies in question (Bailey, Pillard, et al., in particular) strongly suggested a high degree of "concordance" (similarity) between the sexual orientations of identical twin brothers, a slightly lower level of concordance between non-identical twin brothers, and then a progressively lower degree of concordance in the sexual orientation of non-twin biological brothers and finally a lowest level of concordance with adopted (non-biologically related) brothers who had grown up in the same household.
All this certainly gave the appearance of the kind of pattern you would expect to find if homosexuality were in fact a genetically transmitted heritable trait. But there were many problems with how these studies were conducted (and some of these problems have actually been acknowledged by the principal investigators involved). In particular there were some problems inherent to the experimental design and sampling methods (especially in the non-random ways that volunteers for these studies were recruited through ads in gay magazines and other selective factors which suggest that the volunteers, and their cooperating family members, were quite likely not really representative of the broader population of male homosexuals and their family members and dynamics).
Beyond that, there were also conceptual problems with some of the underlying working assumptions from the very start. In particular there was a working assumption that any possible environmental effects or influences could simply be discounted for the purposes of these studies (would have had no effect on the data collected) because all the different kinds of brothers in each family studied grew up together and thus were presumed to have shared the exact same physical and social environment. And there was an underlying assumption that the twins would not have experienced or interacted with the external physical and social environment in any ways that might be significantly different than their non-twin biological brothers or adopted brothers.
A number of biologists (Lewontin among others) have often made the point that this is one of the many problems inherent in studies of twins who were not raised apart. Twins raised together, and in particular identical twins, often grow up experiencing and interacting with the external world as twins and are often viewed and treated by this external world (beyond even their families) in particular ways because they are twins. This is especially true of the experience of many monozygotic (identical) twins. So it would not actually be unreasonable to suggest that the identical twins in these studies might have experienced and interacted with the outside world and all its social influences (including possibly formative ones) in ways which were actually much more similar to each other than to their other brothers. Nevertheless the assumption that twins in a given family would have experienced no greater similarity of environment (and possibly formative effects or experiences from that environment) than their other brothers is just one of the seriously flawed working assumptions which made it easier to jump to the conclusion that the only way to explain the strikingly higher degree of observed concordance in the studied twins' sexual orientations was to ascribe it to their greater genetic similarity.
Of course there is in all this the more overarching and fundamental problem of reductionism involved with trying to locate the source of any complex social behavior at the level of simple genes. This is something which has never been successfully accomplished for any complex social behavior and which, as many scientists have pointed out, is an incorrect approach as a matter of scientific principle (see discussions of this and related questions in Lewontin, for instance, as well as Gould, Hubbard, etc.). Nevertheless, the results of these twin studies were at first striking enough that there was a great rush to attempt to track down the exact location on the X chromosomes of homosexual men of any gene sequence that they seemed to disproportionately have in common with their homosexual brothers which might possibly be implicated in the transmission of sexual orientation. The genetic linkage studies of Dean Hamer, S. Hu, et al., were soon heralded as having found the location of these supposedly shared gene sequences at the now famous (or infamous!) Xq28 location on the X chromosome. "Gay genes found!" trumpeted the press just a few years ago...and to this day a great many people believe that solid "proof" has been found that homosexual orientation is coded in the genes.
But the reality is that none of these studies seem to be holding up particularly well. Recognizing the limitations of their previous studies (especially the sampling methods) the scientists involved in the twin studies have tried to replicate their results with larger pools of more random subjects (anonymous registries of twins in Australia) with much more ambiguous results. The correlations which seemed so striking in the smaller and less randomized studies appear much less significant as the pool of test subjects is expanded, and more random sampling techniques are employed. As for the famed Xq28 reports, it seems that more recent attempts to replicate these results have failed (Rice and Anderson 1999).
In the field of hormones and behavior: This can be considered (at least at first) a particularly intriguing area because there have been more than 40 years of meticulous research involving innumerable animal studies by many different researchers employing a wide variety of species (including some higher mammals) and often generating results which others in these cases have been able to successfully replicate. Although there have been innumerable variations in experimental design, species used, and in the specific focus of the different investigations, it can be said that, in a basic and cumulative sense, what many of these studies boiled down to was the observation that, by manipulating the relative ratios of steroid hormones (such as estrogens and androgens) that embryos are exposed to prior to birth (in utero) or shortly after birth, and by tinkering with the specific timing of hormonal manipulations in relation to developmental stages of those organisms (which varies between species), one can reliably effect some permanent changes in the "sexual" behaviors that these animals will exhibit later on, when they mature sexually. And this even includes observations that you can hormonally manipulate male animals in utero to produce more female-typical mating behaviors later on, and hormonally manipulate female animals in utero to produce more male-typical mating behaviors later on. Beyond these observations it has been suggested that the mechanism involved in such transformations would likely be a hormonally mediated change in the organization of the brain (such things as the quantity and distribution of hormone receptors and estrogen-converting enzymes on the surface of the brain) early in its development, so that in effect it responds in "atypical" ways (and often with what are--seemingly at least--"sex-reversed" behaviors) to the later presence and levels of circulating steroid hormones when the animal comes to sexual maturation.
There is much that is interesting about all this, and there is no doubt still much to learn about the interactions of hormones and brain development and organization among other things. But much of the direct extrapolation that people have tried to make from these kinds of studies to the question of sexual orientation in human beings is quite problematic, to say the least! Making these kinds of interpretive leaps is not justified for a multiplicity of reasons, among which: once again basic conceptual and methodological problems and problems of experimental design and subjective interpretation of results (see Fausto-Sterling for a discussion of some of this); plus there is no evidence to suggest that any particular animal model is particularly useful or relevant for investigating or interpreting any social behaviors in human beings, who are so much more complex and culturally dependent than any other species, and whose behaviors are infinitely more flexible and malleable throughout the course of their lives.
It has also been shown that observations of animal behavior can be heavily clouded by subjective prejudices that experimenters bring to their observations (for example, what they unconsciously think they "see" animals doing, and how they characterize and interpret it as, for instance, "male-typical" or "female-typical"). There is also no particularly good basis to assume that human male and/or female homosexuals have "sex-reversed" sexual attractions and orientations that are "typical of the opposite sex." It should also be noted that there seems to be general consensus at this point that male and female homosexuals have completely sex-typical levels of circulating steroid hormones as adults, and so far at least no study has purported to show evidence that homosexuals were exposed to unusual ratios of steroid hormones prior to birth.
And again, and perhaps most importantly, it is simply not particularly enlightening to observe and interpret social behaviors of social species (even rodents!) in a socially unnatural (artificially limited or socially distorted) environment, which in effect tends to lead investigators to make generalized interpretations of the bases of behaviors without taking into account the possible effects of social interactions (or lack thereof) on the development of those very behaviors. This has been highlighted as a problem with many of the animal studies, but it is even more problematical in relation to much more complex and highly socialized humans! Interestingly, some scientific investigators of the effects of developmental hormones on later behaviors in various animals have begun to experiment with variations in the social environments in which the animals are raised (and/or tested) and have apparently found that these variations in social environment can also have significant effects on the "sexual" behaviors of the test subjects (see, for instance, Wallen 1996).
Again, we cannot here fully dissect and discuss these various studies, and in any case there remains much more to study and better understand. For those who may want to do some further readings on these questions, a good place to start might be biologist Anne Fausto-Sterling's 2000 book, Sexing the Body, which has extensive discussions and some interesting and important methodological critiques of many of the relevant biological studies and who also offers an extensive list of bibliographical references (including many primary sources) which could be helpful to anyone wishing to explore these matters further. Philosopher Stephen Stein's 1999 book The Mismeasure of Desire is also a valuable resource and guide to other references for those wanting to get a further sense of some of the conceptual and methodological difficulties which have plagued biological and also social research aimed at uncovering the bases of the formation of sexual orientations.
Our party is continuing to follow and critically evaluate research developments in these different areas, and we look forward to contributing, along with others, to a broader and more accurate understanding of these important issues.