I think we’re referring to the same ship.
This is a bigger topic than what we’re dealing with here, but I’ll give you my take to see if it’s compatible, or if this is another deeper diconnect that underpins our disagreements in this thread.
It’s true that a person is her body, and in some sense a person, like a river, only exists in a single moment, and changes in the next as calories are burned, cells are born and die, and neuronal connections form, strengthen, weaken, or die. Every moment, the set of contingencies around an individual changes. That’s true whether that individual is receiving hormone therapy or just typing on a philosophy forum. There is a very pedantic sense in which the continuity of personhood is illusory.
In practice, people ignore that, and it’s a pragmatic choice. We treat people as though they are the same person from day to day, from year to year; we punish sober people for crimes they commit while drunk; we keep people in prison for decades. We think of ourselves as continuations of the people we were before puberty, even though our thoughts, motivations, and physical and mental abilities are mostly different. We do that because it’s useful, it’s predictive, it’s pragmatic.
So too should we treat a person who undergoes a hypothetical complete sex change as still the same person they were. Not because they are identical in every detail, but because the change is comparable to other changes in which we recognize continuous personhood: normal human aging; significant bodily injury; drug use; etc.
There’s a different question, as I allude to in invoking the Ship of Thesesus, of at what point e.g. a man becomes a woman in the process of the complete sex change. I agree with your point that the hypothetical relies on the premise that at the beginning we have a man, and at the end we have a woman, and I think that’s correct. I don’t know if most trans advocates would agree with me, but I don’t know that we can have a consistent ontology of gender without that; if an earlier version of themselves self-identified as their birth sex and sought to conform with societal expectations of that sex, it is difficult to say on what grounds we would say they were not (other than an assumption of e.g. a female soul in a male body that is inherently female no matter what it believes about itself).
But that doesn’t touch personhood. I don’t disagree that it would entail behavior and thought changes, and that’s my understanding of what trans people report about the effects of hormone therapy. But personhood is generally accepted to survive such changes.
I think this is significantly more of a constraint than you seem to think. People aren’t going to be convinced of things that are obviously empirically false, so getting everyone to agree about something entails getting a bunch of people to independently verify the world and align their descriptions of it. When we run tests in a laboratory setting, having many sensors that agree on the outcome is a good thing, and that agreement is taken as a strong indication about what the reality is.
And to preempt your rebuttal that the present case is a case of people being “convinced of [something that is] obviously empirically false”, again I will point out that that would be question begging. When a transwoman says “I am a woman”, she isn’t making a claim about chromosomes, and she would happily acknowledge that such a claim would be obviously empirically false.
Let me provide a few defeaters for two distinct claims:
Claim: there is a widely accept sense of “woman”/“man” that does not depend on biology.
If you could demonstrate that people don’t think of Princess Toadstool as a woman (and, perhaps more importantly, didn’t before she got voice-overs from human women). If people didn’t see her and other fictional characters, who lack biology and are only sex-defined by their superficial social sexual signaling, as actually being the social sex which is being signaled, that would be strong evidence that my claim is false.
Claim: calling a trans person by their chosen (i.e. sincerely and accurately signaled) sex accurately describes the world.
If you could show me that people’s expectations are significantly misaligned by calling a trans woman a “woman” or a trans man a “man”, e.g. in a contrived setting in which someone describes a non-present trans person using their chosen sex and is then asked about how they expect that person to look, act, think, etc., and see how well those expectations align as compared with control cases of their birth sex or third options like “transsexual”. Similarly, measuring their surprise using something like pupil dilation when that person walks in following various descriptions (this test may need to be normed against other similarly uncommon traits and their various descriptions).
This is useful for clarifying what we’re even disagreeing about. I note than one aspect of our disagreement is that I am more focused on descriptive claims about how we use language, and how language is understood, and you are more interested in normative claims about how we should use or understand language. For example, I learned recently of the idea of corpus linguistics, the study of language through statistical analysis of a given corpus, and I think that would be very probative to my claims, but not say very much about your claims (except to the extent that the actual use of language in a speaker population goes to the utility because being understood is a big part of the utility of language).