-In my ontology of how-the-world-works, this is _not_ a boring hypothesis. In my ontology, this is a _shockingly weird_ hypothesis, where I can read the English words, but I actually have a lot of trouble parsing the English words into a model in my head, because the antecedent, "If you identify as a gender, and you're attracted to that gender, then ...", already takes a massive prior probability penalty, because that category is multiply disjunctive over the natural space of biological similarities: you're grouping together lesbians _and_ gay men _and_ heterosexual males with a female gender identity _and_ heterosexual females with a male gender identity, and trying to make claim about what members of this group are like.
+because I suspect my actual crux comes down to a [Science _vs._ Bayescraft](https://www.lesswrong.com/posts/viPPjojmChxLGPE2v/the-dilemma-science-or-bayes) thing, where I'm self-conscious about my answer [sounding weirdly overconfident on non-empirical grounds](https://slimemoldtimemold.com/2022/01/11/reality-is-very-weird-and-you-need-to-be-prepared-for-that/) to someone who doesn't already share my parsimony intuitions—but, frankly, I also expect my parsiony intuitions to actually get the right answer in the real world, and modesty/[Outside View](https://www.lesswrong.com/posts/FsfnDfADftGDYeG4c/outside-view-as-conversation-halter)/[Caution on Bias Arguments](https://slatestarcodex.com/2019/07/17/caution-on-bias-arguments/) to get the wrong answer.
+
+If we have a big causal graph with C at the root (E₂ ← E₁ ← C → E₃ ...) with real-valued variables, and someone proposes a theory about what happens to the E_i when C is between 2 and 3 or between 5 and 6 or above 12, that's very unparsimonious: why would such a discontinuous hodge-pause of values for the cause, have consistent effects?
+
+In my worldview, "gender" (as the thing trans women and cis women have in common) looks like a hodge-podge as far as biology is concerned. (It can be real socially to the extent that people believe it's real and act accordingly, which creates the relevant conditional indpendence structure in their social behavior—but sexuality looks more "biological" than "social".)
+
+In my ontology of how-the-world-works, this is _not_ a boring hypothesis. In my ontology, this is a shockingly weird hypothesis, where I can read the English words, but I have a lot of trouble parsing the English words into a model in my head, because the antecedent, "If you identify as a gender, and you're attracted to that gender, then ...", already takes a massive prior probability penalty, because that category is multiply disjunctive over the natural space of biological similarities: you're grouping together lesbians _and_ gay men _and_ heterosexual males with a female gender identity _and_ heterosexual females with a male gender identity, and trying to make claim about what members of this group are like.