Thanks. Not to dive too far into the weeds ... but I'm wondering if there's something I misunderstand about Popper:

> If your theory is capable of "explaining" every possible piece of evidence, it's not a scientific theory.

Do you mean, if your theory is not falsifiable? If your falsifiable theory is not (yet) falsified, if you magically know it never will be falsified, isn't it still a theory?

(If it's just a matter of wording on an Internet comment, then sorry to bother you - I don't mean to play gotcha.)

I've misused the word "possible": I meant to say that if the theory is equally-capable of "explaining" two mutually-incompatible states of affairs, it does not actually have the explanatory power that all scientific theories must have. Which, yes, I think "falsifiable" is the correct word for that. (Although, you'll note that my explanation of "falsifiable" is not precisely correct. I think I've convinced myself that there is no logically-precise definition of "falsifiable", and that it inherently leans on some intuitive notion of "possibly-falsifying evidence" which we have to take for granted, like philosophical induction.)

However, that was itself me being lazy, because Blanchard's typology does make concrete predictions about what sorts of people exist. The assertion that people can be classified into "attracted to women" and "not attracted to women" is refuted by the Kinsey Reports (and many better studies besides); the assertion that being trans is a sex thing is refuted by the existence of certain ace trans people; the assertion that transition is about imitation is refuted by the existence of certain non-binary people…

Of course, Ray Blanchard himself bolts on new categories whenever he can't fudge the analysis to fit a pre-existing conclusion – last I checked, he had around 10 categories – but most adherents treat the original two categories as dogma. (The lengths to which people will go for a gender binary…)

Really, I should say that if you're cis and relying on thought experiments, Blanchard's typology is unfalsifiable: you will simply not be able to observe any evidence that refutes it. I myself (not then having much contact with out trans people (if we discount Reddit)) had to do a lot of introspection (could I rule out self-deception? wishful thinking? how?) before I concluded that either my internal subjective experience refuted the typology (and its variants), or I was too insane to be capable of reasoning. This tendency towards solipsism did not do me many favours (and was ultimately quite harmful, for reasons I will not discuss here), but it did at least prepare me for the many other academics who talk out of their arses about things they have no legitimate claim to expertise in, save for being among the first academics to study it. (Or, in Ray Blanchard's case, among the first academics to study it after the Nazi purge.)

I am not advocating for Blanchard in any way, in case I'm giving that impression, just talking about Popper in general.

Thought experiments, unless something provable like mathematics, are worse than useless, they are deceiving. There is an evil demon that deceives us, that we cannot overcome.

The only source of knowledge is to go out and ask, 'what do you mean?' Thanks for answering!

You're not giving that impression, but your criticism was warranted. (I don't understand falsification as well as I thought I did.)

I wouldn't quite say that thought experiments are always deceiving (Albert Einstein found them quite useful, for example), but they are very parodiable. https://existentialcomics.com/comic/183

> "Did he just… not see any of us?" "Rationalists are terrible at seeing stuff. He probably just tried to deduce who was here from reason alone."

> criticism

Really, purely curiosity. At first, one thing I was wondering was if some new theory might have caught hold in some corner of academia, for example. Then I wondered if I misunderstood falsification myself.

> Einstein found them quite useful, for example

He is why I made the exception for math. His 'experiment' used math (exclusively?) - and wasn't relied upon until confirmed empirically.