If a cis kid was made to transition…

[Epistemic status: somewhat speculative, due to lack of data.]

One argument that’s sometimes made in favor of the existence of an innate gender identity is the case of David Reimer, a boy who was brought up as female due to damage to his penis after a circumcision. Later in life, he ended up gender dysphoric and transitioned back to living as male, but eventually committed suicide.

The Reimer case is likely not strong evidence, though. It’s only n=1 (obviously), and John Money made Reimer do things that could very easily be seen as abusive. Thus, it’s not a great case to rely on.

Instead, a better case to consider might be boys with cloacal exstrophy (a serious condition that among other things leads to an underdeveloped penis) who were raised as female. In the main study of such boys that I’m aware of, they found high levels of gender dissatisfaction, with the majority returning to live as boys, and 2 to 4 of the 5 living as girls wishing to be boys.

On first glance, this supports the notion of innate gender identity. However, one thing that’s already worth noting is that all of these kids were very masculine from young; thus, it also supports some sort of link between this gender identity and masculine behavior, a link that isn’t very compatible with narratives of repressing or hiding gender nonconformity.

In addition, I think if we read the study more carefully, we see some issues with the idea that it supports notions of innate gender identity. The kid who had the least gender issues, and was the most satisfied with being a girl, subject #1, appeared to be in many ways similar to the other kids. So why the different outcomes? Here’s my suspicion:

If you read the study carefully, only two of the subjects living as male, 9 and 10, spontaneously declared their “gender identity”. These, along with two other subjects who were not living as male, were the only ones who formed the idea of “I am male” independently, indicating that whatever is motivating the gender issues of most of the subjects, it’s more complicated than an internally-generated feeling of being male (as gender identity is usually but not always defined). This leaves open the possibility that for all the subjects, including these two, the gender issues came from a more-complex interaction between their behavioral masculinity and society.

Subjects 11-13 adopted a male identity after their parents told them of their medical status at ages 5-7. Since there are quite a few people who have gender issues in their childhood but get over them when older, this makes them imperfect examples. For instance, we could hypothesize that telling masculine girls that they are in some sense male will often lead to a desire to transition. For these subjects, I’d wonder how many of them would’ve desisted if they had not been told.

Subject 14 assumed a male identity after being told at age 18. This is probably old enough that the effect above cannot explain it; thus, I’d categorize this subject in a way similar to subjects 9 and 10.

Subjects 7 and 8 are really similar to subjects 9 and 10, except that their families aren’t supportive, limiting their ability to start living as male. Both for these, and for subjects 9 and 10, there’s another issue that’s worth considering; due to being natal males, they need to take estrogen medically, rather than having the body produce it naturally at puberty. Refering the desistance study again, it is worth noting that some masculine girls feel uncomfortable at puberty but eventually find that they like being girls:

The second factor the desisting girls associated with their decrease in gender discomfort was the feminization of their bodies, primarily the growth of their breasts. At first they reported that this was unpleasant. They felt embarrassed and uncomfortable, and felt it interfered with their freedom to move. However, before long their feelings shifted in a positive direction and they desired even more physical feminization.

♀ Desister #11
Before puberty, I disliked the thought of getting breasts. I did not want them to grow. But when they actually started to grow, I was glad they did. I really loved looking like a girl, so I was glad my body became more feminine.

One thing I would wonder is if the need to take the estrogen exogenously leads to more gender issues, as the effects of it are seen as more foreign and avoidable than if this is what the body naturally produces. As such, while subjects 7-10 are probably the most-unambiguously gender-dissatisfied of the bunch, the situation isn’t completely unambiguous.

Subjects 1 to 5 have their own set of ambiguities, though. The only info we had on how they did in adulthood was based on parent report, which raises the question of how accurate it is. However, the parent’s reports that the subjects are generally content is compatible with more-reliable observations that masculine girls with gender issues generally get over them when they grow up. (On the other hand, the kids in the cloacal exstrophy study were attracted to girls, while the masculine girls who tend to get over their gender issues tend to be attracted to boys.) For most of them, their gender issues were also somewhat limited in scope at the initial assessment, further supporting the possibility that they did fine in adulthood.

Subject 6 is really unclear, though. They appeared to be doing ok – not perfect, but ok – at the initial assessment, but after being told of their medical status, would not discuss the topic with any. However, they did comply with estrogen treatment. Due to lack of better info, I’d classify them with subjects 11-13, as having ambiguous gender issues.

Group	Count	Subjects
Ambiguously no gender issues	5	1-5
Ambiguous gender issues	4	6, 11-13
Unambiguous gender issues	5	7-10, 14

I’d be inclined to drop subjects 6 and 11-13 for being told at a young age, making them very difficult to compare to e.g. masculine natal females. This yields about half with no gender issues, and about half with clear gender issues, and I can’t help but point out that this is a similar to the rate of people who tend to identity as cis-by-default in surveys (this survey found 54% identifying as cis-by-default, 46% identifying as affirmatively cis).

[Epistemic status for the followup: questionable math, mainly for sanity-checking. The following math can probably be adjusted to “prove” anything by fiddling with the assumptions.]

How well does this fit with a model where masculinity interacting with society is the driving factor for these sorts of gender issues? I usually estimate there to be a D~2 gender difference in psychology, which implies that 15% of people are more like the opposite sex than like their natal sex. This is wayyy to much if just used directly, as this would suggest that 15%/2 = 7.5% of natal females become trans men.

However, doing this estimate directly would also be somewhat ridiculous, as the vast majority of the 15% would still be more feminine than the average boy, and because the gender issues appear to be much stronger among those attracted to girls than those attracted to boys.

Thus, the real question we need to know is how many lesbians are more masculine than the average man. The estimate of the difference between lesbians and straight women in masculinity/femininity varies depending on study, but let’s go with the gender diagnosticity difference from this study and assume d~0.5. This means that lesbians are d~1.5 more feminine than men. The estimated rate of lesbianism also varies, but let’s go with a middle ground answer of 2%.

By these estimates, about 7% of lesbians should end up with serious gender issues and transition to end up as trans men, which in total should make up a bit more than 0.1% of the natal female population, or, if we estimate trans rates to be about 0.3%, a bit less than half of the FtM population.

This is twice as high as the rate they truly make up (23%, according to the USTS), but there’s infinitely many places that the numbers and calculations can be tweaked, so I don’t think this problem with fitting it should be taken too seriously.

My main conclusion for this is that probably a lot of men would do fine living as women if they were raised as girls, but also that quite a few probably wouldn’t. This appears to very roughly match the cis-by-default self-identification situation, at least to around an order of magnitude or two (which, admittedly, is a pretty bad match). I don’t think that these results are as compatible with a universal innate immutable gender identity, as much as they might be compatible with more-complex mechanics, involving significant individual variation in how well any given man would do living as female.

EDIT 2019-08-02: This study also appears to find that 50% of boys with cloacal exstrophy who are raised as female end up non-gender-dysphoric, but I haven’t read it very carefully so I don’t know. Might be worth looking into.