That would make sense, but in that I case I wonder which characteristics constitute the "masculinity" or "feminity", and what is their relative weight in the set. Again, to use myself as an example, I don't like drinking beer, and I don't like watching football or hockey -- and I wonder how many masculinity points did I lose by that: 1% or 20%? On the other hand, how many masculinity points do I get for being good at math, or being a computer programmer, or enjoying rationality?
I think that while many people would in general agree that there is a "masculinity" and "feminity" scale, if we tried to talk about details (like: how many points of masculinity scale you lose for not being interested in football or beer), we would find out that we actually don't agree on what the scale consists of, because there would be wildly different definitions given e.g. by different social classes. I imagine that for a working class, beer and football are pretty important masculine characteristics, but much less important for the middle class. (Or by country: I guess Russian men drink vodka, and French men don't lose masculinity points for preferring wine.) So the archetypal example of a "man" could turn out to be mostly an archetypal example of a "working-class man (in my country)".
So where is all this extra variation coming from in the first place?
Testosterone level seems like a reasonable hypothesis.
But if we accept that the concepts of "masculinity" and "feminity" depend on culture (subculture, social group, etc.) then simply coming from a different subculture may cause one to identify differently with expectations of the culture they live in. To give a silly example, if I would move to a hypothetical country where all men are expected to jump on one leg all day long, while women can walk normally, there is a chance I would get labeled as a 'feminine walker', and I might even choose such label for myself. If there would be too many such labels, maybe I would conclude that I am more feminine in general.
But these are, like, two completely different topics: (1) feeling comfortable in a body shaped by given hormones, and (2) feeling comfortable with cultural gender norms. Those are not the same thing. For example, enjoying neither beer nor football is not the same as wanting one's penis removed. I can think that a masculine archetype (some exaggerated version of a working-class man) is silly in many aspects, but still feel quite comfortable in my body.
I think that while many people would in general agree that there is a "masculinity" and "feminity" scale, if we tried to talk about details (like: how many points of masculinity scale you lose for not being interested in football or beer), we would find out that we actually don't agree on what the scale consists of, because there would be wildly different definitions given e.g. by different social classes.
While I expect that different people would wildly disagree about absolute positions of different characteristics on that scale, I ...
As I understand it, there is a phenomenon among transgender people where no matter what they do they can't help but ask themselves the question, "Am I really an [insert self-reported gender category here]?" In the past, a few people have called for a LessWrong-style dissolution of this question. This is how I approach the problem.
There are two caveats which I must address in the beginning.
The first caveat has to do with hypotheses about the etiology of the transgender condition. There are many possible causes of gender identity self-reports, but I don't think it's too controversial to propose that at least some of the transgender self-reports might result from the same mechanism as cisgender self-reports. Again, the idea is that there is some 'self-reporting algorithm', that takes some input that we don't yet know about, and outputs a gender category, and that both cisgender people and transgender people have this. It's not hard to come up with just-so stories about why having such an algorithm and caring about its output might have been adaptive. This is, however, an assumption. In theory, the self-reports from transgender people could have a cause separate from the self-reports of cisgender people, but it's not what I expect.
The second caveat has to do with essentialism. In the past calls for an article like this one, I saw people point out that we reason about gender as if it is an essence, and that any dissolution would have to avoid this mistake. But there's a difference between describing an algorithm that produces a category which feels like an essence, and providing an essentialist explanation. My dissolution will talk about essences because the human mind reasons with them, but my dissolution itself will not be essentialist in nature.
Humans universally make inferences about their typicality with respect to their self-reported gender. Check Google Scholar for 'self-perceived gender typicality' for further reading. So when I refer to a transman, by my model, I mean, "A human whose self-reporting algorithm returns the gender category 'male', but whose self-perceived gender typicality checker returns 'Highly atypical!'"
And the word 'human' at the beginning of that sentence is important. I do not mean "A human that is secretly, essentially a girl," or "A human that is secretly, essentially a boy,"; I just mean a human. I postulate that there are not boy typicality checkers and girl typicality checkers; there are typicality checkers that take an arbitrary gender category as input and return a measure of that human's self-perceived typicality with regard to the category.
So when a transwoman looks in the mirror and feels atypical because of a typicality inference from the width of her hips, I believe that this is not a fundamentally transgender experience, not different in kind, but only in degree, from a ciswoman who listens to herself speak and feels atypical because of a typicality inference from the pitch of her voice.
Fortunately, society's treatment of transgender people has come around to something like this in recent decades; our therapy proceeds by helping transgender people become more typical instances of their self-report algorithm's output.
Many of the typical traits are quite tangible: behavior, personality, appearance. It is easier to make tangible things more typical, because they're right there for you to hold; you aren't confused about them. But I often hear reports of transgender people left with a nagging doubt, a lingering question of "Am I really an X?, which feels far more slippery and about which they confess themselves quite confused.
To get at this question, I sometimes see transgender people try to simulate the subjective experience of a typical instance of the self-report algorithm's output. They ask questions like, "Does it feel the same to be me as it does to be a 'real X'?" And I think this is the heart of the confusion.
For when they simulate the subjective experience of a 'real X', there is a striking dissimilarity between themselves and the simulation, because a 'real X' lacks a pervasive sense of distress originating from self-perceived atypicality.
But what I just described in the previous sentence is itself a typicality inference, which means that this simulation itself causes distress from atypicality, which is used to justify future inferences of self-perceived atypicality!
I expected this to take more than one go-around.
Let's review something Eliezer wrote in Fake Causality:
If you didn't have an explicit awareness that you have a general human algorithm that checks the arbitrary self-report against the perceived typicality, but rather you believed that this was some kind of special, transgender-specific self-doubt, then your typicality checker would never be able to mark its own distress signal as 'Typical!', and it would oscillate between judging the subjective experience as atypical, outputting a distress signal in response, judging its own distress signal as atypical, sending a distress signal about that, etc.
And this double-counting is not anything like hair length or voice pitch, or even more slippery stuff like 'being empathetic'; it's very slippery, and no matter how many other ways you would have made yourself more typical, even though those changes would have soothed you, there would have been this separate and additional lingering doubt, a doubt that can only be annihilated by understanding the deep reasons that the tangible interventions worked, and how your mind runs skew to reality.
And that's it. For me at least, this adds up to normality. There is no unbridgeable gap between the point at which you are a non-X and the point at which you become an X. Now you can just go back to making yourself as typical as you want to be, or anything else that you want to be.