Now that’s a bit more compelling

I’ve always given the side-eye to proponents of Brain Sex theory (“neurosexism”), the idea that brains can not only be sexed but that those sex states are binary, on the grounds that their methods for measuring brain sex aren’t particularly good: MRIs. I haven’t found a brain scan study that had made me rethink this opinion. They’re still extremely weak for making any argument, even if I appreciate that most Brain Sex proponents tend to be affirmative of trans lives vis-a-vis “born this way” rhetoric.

It’s still a theory ripe for exploitation by sexists, and I still don’t find interpretation from brain scans convincing. There’s too much space between the observation and the interpretation thereof, space filled with magical thinking and after-the-fact justification.

What I have found–and for some reason, didn’t think to look for–is something a fair bit more compelling as evidence for the “gender variance as biological event” hypothesis. It’s a twin study Franksteined together from other twin studies.

Often the difficulty of trans twin studies is a lack of data. So the researchers just sewed together hundreds of twin studies, many of which reported at least one twin in at least one twin pair who was transgender and thus transitioned. Most of these studies weren’t investigating gender variance so it didn’t factor into the analysis, though they were obviously mentioned in the observations.

The findings have interesting implications: (Content Notice–like most cis researchers, the interpretation of results can be belittling or riddled with problems discussed in trans feminist discourse. Emphasis mine.)

It is clear from the findings that there is a greater likelihood of concordance for transsexuality among males than among females and a greater chance of concordance among monozygotic twins than dizygotic twins. These findings are consistent with those reported previously (M. Diamond, 2011b; M. Diamond & Hawk, 2004) and comparable to those found in familial studies and reports of cooccurrence of gender identity disorder within families as reported by others. In a report that appeared while this paper was in press, Heylens et al. (2012) reported that of 23 MZ female and male twins they reviewed, nine were concordant for gender identity disorder (GID; 39%), while none of their 21 DZ twins were concordant. The investigators concluded, “These findings suggest a role for genetic factors in the development of GID.” Their results and those reported here are quite similar. The results of these studies combined, at least for this issue, appear to show that Internet survey data, when coupled with clinically reported material, can provide representative data. Both studies reinforce the significant role of genetics in transsexuality.

Gomez-Gil et al. (2010) in their sample of 995 consecutive transsexual probands (677 male-to-female and 318 female-to-male) report 12 pairs of transsexual nontwin siblings (nine pairs of MtF siblings, two pairs of MtF and FtM siblings, and one pair of FtM siblings). These investigators claim that their data indicate that the probability that a sibling of a transsexual will also be transsexual was 4.48 times higher for siblings of MtFs than for siblings of FtM transsexual probands, and 3.88 times higher for the brothers than for the sisters of transsexual probands. Moreover, the prevalence of transsexualism in siblings of transsexuals (1/211 siblings) was much higher than the range expected according to the prevalence data of transsexualism in Spain (the country of their study). Their study strongly suggests that siblings of transsexuals have a higher chance of being transsexual than the general population and that the potential is higher for brothers than for sisters of transsexuals, and for siblings of MtF than for FtM transsexuals. An excellent review paper by Veale, Clarke, and Lomax (2009) offers a host of references of papers dealing with transsexual familiarity and, while concentrating on the role of genetics and prenatal hormones, also touches on the actual and possible aspects of the rearing environment. They conclude there appears to be a significant role for biology in transsexualism but conservatively caution that attention is given to rearing practices.

Other reports have also shown familial transsexualism. Stoller and Baker (1973) reported on a family in which two brothers transitioned to live as androphilic sisters, and Stoller and Moseley (1974) reported on three siblings with GID. Green (2000), in addition to his report of monozygotic twins that were concordant for transsexualism, simultaneously reported on nine other siblings or parent-child pairs concordant for gender identity disorder. I personally know of three sets of father and son pairs that are concordant for transsexuality. There are no known studies that report any environmental influences significantly able to induce such findings. Results of testing for GID among the large twin data set of Australia found, “The model that best described the data included a significant additive genetic component accounting for 62% of the variance and a non-shared environmental component accounting for the remaining 38% of the variance … Overall, the results support the hypothesis that there is a strong heritable component to GID” (Coolidge, These, & Young, 2002, p251). Bailey, Dunne, and Martin using that same data set concluded, “[C]hildhood gender nonconformity was significantly heritable for both men and women” (Bailey, Dunne, & Martin, 2000, p524).

When proponents of Brain Sex theory refer primarily or exclusively to brain scans, I think I had reason to be skeptical due to the enormous gap between the intrepretation and the data. But when I encountered this paper and fact-checked some of their citations, I found much more compelling evidence to suggest there is one possible explanation in a genetic basis for gender variance. Abandoning the pseudoscience of trying to sex a brain with an MRI (which is about as clever as trying to study atoms with a magnifying glass), we instead find a correlation that gives us a much clearer idea of where to look next.

Note that the study does not suggest whether there is one specific biological event that causes gender variance, nor does it explore whether that event is strictly genetic, nor does it explore if there are other avenues to develop gender variance. There has been some research to suggest a correlation between a certain fetal development event and the onset of gender variance, which could have a gazillion causes, and at the moment I’m not aware of any investigation into said causes. This was the crux of Cordelia Fine’s analysis in Delusions of Gender, a text challenging neurosexism. I certainly don’t feel there is compelling evidence that brains can be routinely or accurately sexed in a binary fashion and that this is the mechanism by which gender variance is produced. Strange that this twin study isn’t getting the same amount of hoo-rah despite finding a considerably stronger correlation with far less subjective nonsense.

Perhaps my favourite consequence of this data is how it blows large holes in many iterations of so-called “gender critical” theory, which usually claims that trans people have internalized gendered oppression and that gender dysphoria is a manifestation of that oppression.

Were this the case, the probability two monozygotic twins separated at birth would both be trans (“concordant”) should be the same as the general population (~0.6%). If gendercrit had any predictive power, we would see dizygotic twins transitioning at the same rate as monozygotic ones, since the proposed mechanism of action is internalized oppression which both sets of twins would experience, given that they are raised in the same culture. Yet the study demonstrates that monozygotic twins are much more likely to both be transgender if at least one of them is trans, even if they’ve been separated at birth–while dizygotic twins and non-twin siblings also exhibit higher probabilities of gender variance than the general population.

Damning implications indeed for the hypothesis that gender dysphoria is merely a manifestation of internalized oppression.

What I would like to see next is the correlation between being trans and a proposed biological event, which might be a bit harder to measure without upsetting some ethics boards considering the running hypothesis is “something in fetal development.”

Here’s to asking better questions.



  1. AMM says

    I notice you speak of “separated at birth”, but the passage you quote doesn’t say “separated” at all. If they grew up in the same family, and thus the same social class, subculture, etc., they would have a lot in common. If separated some time after birth, they would be placed in similar environments, typically relatives.

    But, frankly, I don’t think the obsession with finding “causes” of gender dysphoria or transgender-ness is healthy. It pathologizes being trans, something which I consider simply an example of normal variation. The problem isn’t being trans, the problem is the way society treats being that way. If society treated stamp collecting as some malignant mental illness, then stamp collectors would probably suffer the same sort of dysphoria about their love of stamps.

    We’ve actually been through all this before, only with non-conforming sexual orientations. At one time, being gay or lesbian was believed to be a mental illness, because so many gay & lesbian people were so miserable about their orientations. Now that being gay/lesbian is legal and sort of socially acceptable (in some places), we’re finding that it isn’t being gay or lesbian that’s the problem, it’s society insisting that what you are is an illness or an abomination.

  2. Siobhan says

    I notice you speak of “separated at birth”, but the passage you quote doesn’t say “separated” at all.

    Not in the passage I quoted, no, but it is discussed elsewhere in the link that the trend is still observed in those twin sets who are separated, which suggests limitations on the strength of environmental influences in being gender variant.

    It pathologizes being trans, something which I consider simply an example of normal variation.

    I disagree that finding the cause(s) is itself pathological. As with many other discoveries, the knowledge can be abused once discovered. If we did manage to find a definitive evidence based model for how and why gender variance occurs, that could very well be exploited in many unethical ways. But this evidence also discredits the growing popularity of a certain breed of armchair psychology that attempts to defend conversion therapy. That others will try to use this knowledge to “cure” transness is less a commentary on the knowledge itself and more a commentary on what some people think of gender variance.

    In other words, don’t blame Einstein for Hiroshima.

    The problem isn’t being trans, the problem is the way society treats being that way.

    But these studies aren’t always framing gender variance as a problem. That was certainly the case for professional dickbags Paul McHugh and Kenneth Zucker, but these researchers don’t make any value judgements about their observation. Nor do they discuss the impact of discrimination on mental health. They simply report on a pattern. “if X then Y 40% of the time”