+> Some people I usually respect for their willingness to publicly die on a hill of facts, now seem to be talking as if pronouns are facts, or as if who uses what bathroom is necessarily a factual statement about chromosomes. Come on, you know the distinction better than that!
+>
+> _Even if_ somebody went around saying, "I demand you call me 'she' and furthermore I claim to have two X chromosomes!", which none of my trans colleagues have ever said to me by the way, it still isn't a question-of-empirical-fact whether she should be called "she". It's an act.
+>
+> In saying this, I am not taking a stand for or against any Twitter policies. I am making a stand on a hill of meaning in defense of validity, about the distinction between what is and isn't a stand on a hill of facts in defense of truth.
+>
+> I will never stand against those who stand against lies. But changing your name, asking people to address you by a different pronoun, and getting sex reassignment surgery, Is. Not. Lying. You are _ontologically_ confused if you think those acts are false assertions.
+
+Some of the replies tried explain the problem—and [Yudkowsky kept refusing to understand](https://twitter.com/ESYudkowsky/status/1067291243728650243)—
+
+> Using language in a way _you_ dislike, openly and explicitly and with public focus on the language and its meaning, is not lying. The proposition you claim false (chromosomes?) is not what the speech is meant to convey—and this is known to everyone involved, it is not a secret.
+
+—[repeatedly](https://twitter.com/ESYudkowsky/status/1067198993485058048):
+
+> You're mistaken about what the word means to you, I demonstrate thus: https://en.wikipedia.org/wiki/XX_male_syndrome
+>
+> But even ignoring that, you're not standing in defense of truth if you insist on a word, brought explicitly into question, being used with some particular meaning.
+
+Dear reader, this is the moment where I _flipped the fuck out_. Let me explain.
+
+This "hill of meaning in defense of validity" proclamation was just such a striking contrast to the Eliezer Yudkowsky I remembered—the Eliezer Yudkowsky whom I had variously described as having "taught me everything I know" and "rewritten my personality over the internet"—who didn't hesitate to criticize uses of language that he thought were failing to carve reality at the joints, even going so far as to [call them "wrong"](https://www.lesswrong.com/posts/FaJaCgqBKphrDzDSj/37-ways-that-words-can-be-wrong):
+
+> [S]aying "There's no way my choice of X can be 'wrong'" is nearly always an error in practice, whatever the theory. You can always be wrong. Even when it's theoretically impossible to be wrong, you can still be wrong. There is never a Get-Out-Of-Jail-Free card for anything you do. That's life.
+
+[Similarly](https://www.lesswrong.com/posts/d5NyJ2Lf6N22AD9PB/where-to-draw-the-boundary):
+
+> Once upon a time it was thought that the word "fish" included dolphins. Now you could play the oh-so-clever arguer, and say, "The list: {Salmon, guppies, sharks, dolphins, trout} is just a list—you can't say that a list is _wrong_. I can prove in set theory that this list exists. So my definition of _fish_, which is simply this extensional list, cannot possibly be 'wrong' as you claim."
+>
+> Or you could stop playing nitwit games and admit that dolphins don't belong on the fish list.
+>
+> You come up with a list of things that feel similar, and take a guess at why this is so. But when you finally discover what they really have in common, it may turn out that your guess was wrong. It may even turn out that your list was wrong.
+>
+> You cannot hide behind a comforting shield of correct-by-definition. Both extensional definitions and intensional definitions can be wrong, can fail to carve reality at the joints.
+
+One could argue the "Words can be wrong when your definition draws a boundary around things that don't really belong together" moral doesn't apply to Yudkowsky's new Tweets, which only mentioned pronouns and bathroom policies, not the [extensions](https://www.lesswrong.com/posts/HsznWM9A7NiuGsp28/extensions-and-intensions) of common nouns.
+
+But this seems pretty unsatifying in the context of his claim to ["not [be] taking a stand for or against any Twitter policies"](https://twitter.com/ESYudkowsky/status/1067185907843756032). One of the Tweets that had recently led to radical feminist Meghan Murphy getting [kicked off the platform](https://quillette.com/2018/11/28/twitters-trans-activist-decree/) read simply, ["Men aren't women tho."](https://archive.is/ppV86) This doesn't seem like a policy claim; rather, Murphy was using common language to express the fact-claim that members of the natural category of adult human males, are not, in fact, members of the natural category of adult human females.
+
+If the extension of common words like 'woman' and 'man' is an issue of epistemic importance that rationalists should care about, then presumably so is Twitter's anti-misgendering policy—and if it _isn't_ (because you're not standing in defense of truth if you insist on a word, brought explicitly into question, being used with some particular meaning) then I wasn't sure what's _left_ of the "Human's Guide to Words" sequence if the [37-part grand moral](https://www.lesswrong.com/posts/FaJaCgqBKphrDzDSj/37-ways-that-words-can-be-wrong) needs to be retracted.
+
+I think I _am_ standing in defense of truth if have an _argument_ for _why_ my preferred word usage does a better job at "carving reality at the joints", and the one bringing my usage explicitly into question doesn't have such an argument. As such, I didn't see the _practical_ difference between "you're not standing in defense of truth if you insist on a word, brought explicitly into question, being used with some particular meaning", and "I can define a word any way I want." About which, again, a previous Eliezer Yudkowsky had written:
+
+> ["It is a common misconception that you can define a word any way you like. [...] If you believe that you can 'define a word any way you like', without realizing that your brain goes on categorizing without your conscious oversight, then you won't take the effort to choose your definitions wisely."](https://www.lesswrong.com/posts/3nxs2WYDGzJbzcLMp/words-as-hidden-inferences)
+>
+> ["So that's another reason you can't 'define a word any way you like': You can't directly program concepts into someone else's brain."](https://www.lesswrong.com/posts/HsznWM9A7NiuGsp28/extensions-and-intensions)
+>
+> ["When you take into account the way the human mind actually, pragmatically works, the notion 'I can define a word any way I like' soon becomes 'I can believe anything I want about a fixed set of objects' or 'I can move any object I want in or out of a fixed membership test'."](https://www.lesswrong.com/posts/HsznWM9A7NiuGsp28/extensions-and-intensions)
+>
+> ["There's an idea, which you may have noticed I hate, that 'you can define a word any way you like'."](https://www.lesswrong.com/posts/i2dfY65JciebF3CAo/empty-labels)
+>
+> ["And of course you cannot solve a scientific challenge by appealing to dictionaries, nor master a complex skill of inquiry by saying 'I can define a word any way I like'."](https://www.lesswrong.com/posts/y5MxoeacRKKM3KQth/fallacies-of-compression)
+>
+> ["Categories are not static things in the context of a human brain; as soon as you actually think of them, they exert force on your mind. One more reason not to believe you can define a word any way you like."](https://www.lesswrong.com/posts/veN86cBhoe7mBxXLk/categorizing-has-consequences)
+>
+> ["And people are lazy. They'd rather argue 'by definition', especially since they think 'you can define a word any way you like'."](https://www.lesswrong.com/posts/yuKaWPRTxZoov4z8K/sneaking-in-connotations)
+>
+> ["And this suggests another—yes, yet another—reason to be suspicious of the claim that 'you can define a word any way you like'. When you consider the superexponential size of Conceptspace, it becomes clear that singling out one particular concept for consideration is an act of no small audacity—not just for us, but for any mind of bounded computing power."](https://www.lesswrong.com/posts/82eMd5KLiJ5Z6rTrr/superexponential-conceptspace-and-simple-words)
+>
+> ["I say all this, because the idea that 'You can X any way you like' is a huge obstacle to learning how to X wisely. 'It's a free country; I have a right to my own opinion' obstructs the art of finding truth. 'I can define a word any way I like' obstructs the art of carving reality at its joints. And even the sensible-sounding 'The labels we attach to words are arbitrary' obstructs awareness of compactness."](https://www.lesswrong.com/posts/soQX8yXLbKy7cFvy8/entropy-and-short-codes)
+>
+> ["One may even consider the act of defining a word as a promise to \[the\] effect [...] \[that the definition\] will somehow help you make inferences / shorten your messages."](https://www.lesswrong.com/posts/yLcuygFfMfrfK8KjF/mutual-information-and-density-in-thingspace)
+
+One could argue that this is unfairly interpreting Yudkowsky's Tweets as having a broader scope than was intended—that Yudkowsky _only_ meant to slap down the specific false claim that using 'he' for someone with a Y chromosome is lying, without intending any broader implications about trans issues or the philosophy of language. It wouldn't be realistic or fair to expect every public figure to host a truly exhaustive debate on all related issues every time a fallacy they encounter in the wild annoys them enough for them to Tweet about that specific fallacy.
+
+However, I don't think this "narrow" reading is the most natural one. Yudkowsky had previously written of what he called [the fourth virtue of evenness](http://yudkowsky.net/rational/virtues/): "If you are selective about which arguments you inspect for flaws, or how hard you inspect for flaws, then every flaw you learn how to detect makes you that much stupider." He had likewise written [of reversed stupidity](https://www.lesswrong.com/posts/qNZM3EGoE5ZeMdCRt/reversed-stupidity-is-not-intelligence) (bolding mine):
+
+> **To argue against an idea honestly, you should argue against the best arguments of the strongest advocates**. Arguing against weaker advocates proves _nothing_, because even the strongest idea will attract weak advocates.
+
+Relatedly, Scott Alexander had written about how ["weak men are superweapons"](https://slatestarcodex.com/2014/05/12/weak-men-are-superweapons/): speakers often selectively draw attention to the worst arguments in favor of a position, in an attempt to socially discredit people who have better arguments for the position (which the speaker ignores). In the same way, by _just_ slapping down a weak man from the "anti-trans" political coalition without saying anything else in a similarly prominent location, Yudkowsky was liable to mislead his readers (who trusted him to argue against ideas honestly) into thinking that there were no better arguments from the "anti-trans" side.
+
+To be sure, it imposes a cost on speakers to not be able to Tweet about one specific annoying fallacy and then move on with their lives without the need for [endless disclaimers](http://www.overcomingbias.com/2008/06/against-disclai.html) about related but stronger arguments that they're _not_ addressing. But the fact that [Yudkowsky disclaimed that](https://twitter.com/ESYudkowsky/status/1067185907843756032) he wasn't taking a stand for or against Twitter's anti-misgendering policy demonstrates that he _didn't_ have an aversion to spending a few extra words to prevent the most common misunderstandings.
+
+Given that, it's hard to read the Tweets Yudkowsky published as anything other than an attempt to intimidate and delegitimize people who want to use language to reason about sex rather than gender identity. [For example](https://twitter.com/ESYudkowsky/status/1067490362225156096), deeper in the thread, Yudkowsky wrote:
+
+> The more technology advances, the further we can move people towards where they say they want to be in sexspace. Having said this we've said all the facts. Who competes in sports segregated around an Aristotelian binary is a policy question (that I personally find very humorous).
+
+Sure, _in the limit of arbitrarily advanced technology_, everyone could be exactly where they wanted to be in sexpsace. Having said this, we have _not_ said all the facts relevant to decisionmaking in our world, where _we do not have arbitrarily advanced technology_. As Yudkowsky [acknowledges in the previous Tweet](https://twitter.com/ESYudkowsky/status/1067488844122021888), "Hormone therapy changes some things and leaves others constant." The existence of HRT does not itself take us into the Glorious Transhumanist Future where everyone is the sex they say they are.
+
+The _reason_ for having sex-segregated sports leagues is because the sport-relevant multivariate trait distributions of female bodies and male bodies are quite different: men are taller, and stronger, and faster, and have bigger lungs. If you just had one integrated league, females wouldn't be competitive (in the vast majority of sports, with a few exceptions [like ultra-distance swimming](https://www.swimmingworldmagazine.com/news/why-women-have-beaten-men-in-marathon-swimming/) that happen to sample an unusually female-favorable corner of sportspace).
+
+Given the empirical reality of the different trait distributions, "Who are the best athletes _among females_" is a natural question for people to be interested in, and want separate sports leagues to determine. Including male people in female sports leagues undermines the point of having a separate female league, and [_empirically_, hormone replacement therapy after puberty](https://link.springer.com/article/10.1007/s40279-020-01389-3) [doesn't substantially change the picture here](https://bjsm.bmj.com/content/55/15/865).
+
+(Similarly, when conducting [automobile races](https://en.wikipedia.org/wiki/Auto_racing), you want there to be rules enforcing that all competitors have the same type of car for some common-sense-reasonable operationalization of "the same type", because a race between a sports car and a [moped](https://en.wikipedia.org/wiki/Moped) would be mostly measuring who has the sports car, rather than who's the better racer.)
+
+In light of these empirical observations, Yudkowsky's suggestion that an ignorant comittment to an "Aristotelian binary" is the main reason someone might care about the integrity of women's sports, is revealed as an absurd strawman. This just isn't something any scientifically-literate person would write if they had actually thought about the issue _at all_, as contrasted to having _first_ decided (consciously or not) to bolster one's reputation among progressives by dunking on transphobes on Twitter, and wielding one's philosophy knowledge in the service of that political goal. The relevant empirical facts are _not subtle_, even if most people don't have the fancy vocabulary to talk about them in terms of "multivariate trait distributions."
+
+I'm picking on the "sports segregated around an Aristotelian binary" remark because sports is a case where the relevant effect sizes are _so_ large as to make the point [hard for all but the most ardent gender-identity partisans to deny](/2017/Jun/questions-such-as-wtf-is-wrong-with-you-people/), but the point is very general: biological sex actually exists and is sometimes decision-relevant. People who want to be able to talk about sex and make policy decisions on the basis of sex are not making an ontology error, because the ontology in which sex "actually" "exists" continues to make very good predictions in our current tech regime.
+
+Yudkowsky's claim to merely have been standing up for the distinction between facts and policy questions doesn't seem credible. It is, of course, true that pronoun and bathroom conventions are policy decisions rather than a matter of fact, but it's _bizarre_ to condescendingly point this out _as if it were the crux of contemporary trans-rights debates_. Conservatives and gender-critical feminists _know_ that trans-rights advocates aren't falsely claiming that trans women have XX chromosomes. If you _just_ wanted to point out that the organization of sports leagues is a policy question rather than a fact (as if anyone had doubted this), why would you throw in the "Aristotelian binary" strawman and belittle the matter as "humorous"? There are a lot of issues that I don't _personally_ care much about, but I don't see anything funny about the fact that other people _do_ care.
+
+If any concrete negative consequence of gender self-identity categories is going to be waved away with, "Oh, but that's a mere _policy_ decision that can be dealt with on some basis other than gender, and therefore doesn't count as an objection to the new definition of gender words", then it's not clear what the new definition is _for_. The policymaking categories we use to make decisions are _closely related_ to the epistemic categories we use to make predictions, and people need to be able to talk about them.
+
+An illustrative example: like many gender-dysphoric males, I [cosplay](/2016/Dec/joined/) [female](/2017/Oct/a-leaf-in-the-crosswind/) [characters](/2019/Aug/a-love-that-is-out-of-anyones-control/) at fandom conventions sometimes. And, unfortunately, like many gender-dysphoric males, I'm _not very good at it_. I think someone looking at some of my cosplay photos and trying to describe their content in clear language—not trying to be nice to anyone or make a point, but just trying to use language as a map that reflects the territory—would say something like, "This is a photo of a man and he's wearing a dress." The word _man_ in that sentence is expressing _cognitive work_: it's a summary of the [lawful cause-and-effect evidential entanglement](https://www.lesswrong.com/posts/6s3xABaXKPdFwA3FS/what-is-evidence) whereby the photons reflecting off the photograph are correlated with photons reflecting off my body at the time the photo was taken, which are correlated with my externally-observable secondary sex characteristics (facial structure, beard shadow, _&c._), from which evidence an agent using an [efficient naïve-Bayes-like model](https://www.lesswrong.com/posts/gDWvLicHhcMfGmwaK/conditional-independence-and-naive-bayes) can assign me to its "man" category and thereby make probabilistic predictions about some of my traits that aren't directly observable from the photo, and achieve a better [score on those predictions](http://yudkowsky.net/rational/technical/) than if the agent had assigned me to its "adult human female" category, where by "traits" I mean not (just) particularly sex chromosomes ([as Yudkowsky suggested on Twitter](https://twitter.com/ESYudkowsky/status/1067291243728650243)), but the _conjunction_ of dozens or hundreds of measurements that are [_causally downstream_ of sex chromosomes](/2021/Sep/link-blood-is-thicker-than-water/): reproductive organs _and_ muscle mass (sex difference effect size of [Cohen's _d_](https://en.wikipedia.org/wiki/Effect_size#Cohen's_d)≈2.6) _and_ Big Five Agreeableness (_d_≈0.5) _and_ Big Five Neuroticism (_d_≈0.4) _and_ short-term memory (_d_≈0.2, favoring women) _and_ white-to-gray-matter ratios in the brain _and_ probable socialization history _and_ [any number of other things](https://en.wikipedia.org/wiki/Sex_differences_in_human_physiology)—including differences we might not necessarily currently know about, but have prior reasons to suspect exist: no one _knew_ about sex chromosomes before 1905, but given all the other systematic differences between women and men, it would have been a reasonable guess (that turned out to be correct!) to suspect the existence of some sort of molecular mechanism of sex determination.
+
+Forcing a speaker to say "trans woman" instead of "man" in that sentence depending on my verbally self-reported self-identity may not be forcing them to _lie_, exactly. (Because it's understood, "openly and explicitly and with public focus on the language and its meaning", what _trans women_ are; no one is making a false-to-fact claim about them having ovaries, for example.) But it _is_ forcing the speaker to obfuscate the probabilistic inference they were trying to communicate with the original sentence (about modeling the person in the photograph as being sampled from the "man" [cluster in configuration space](https://www.lesswrong.com/posts/WBw8dDkAWohFjWQSk/the-cluster-structure-of-thingspace)), and instead use language that suggests a different cluster-structure ("trans women", two words, are presumably a subcluster within the "women" cluster). Crowing in the public square about how people who object to be forced to "lie" must be ontologically confused is _ignoring the interesting part of the problem_. Gender identity's [claim to be non-disprovable](https://www.lesswrong.com/posts/fAuWLS7RKWD2npBFR/religion-s-claim-to-be-non-disprovable) mostly functions as a way to [avoid the belief's real weak points](https://www.lesswrong.com/posts/dHQkDNMhj692ayx78/avoiding-your-belief-s-real-weak-points).