(I mean, the beautiful pure sacred self-identity thing is much harder to talk about _clearly_, but talking about it _un_-clearly is less shameful and requires much less bravery)
-—are immediately provided with "Oh, that means you're not a cis boy; you're a trans girl" as the definitive explanation. But it was a different time, then. Of course I had _heard of_ transsexualism as a thing, in the form of the "woman trapped in a man's body" trope, but it wasn't something I expected to actually encounter in real life.
+—are immediately provided with "Oh, that means you're not a cis boy; you're a trans girl" as the definitive explanation.
+
+But it was a different time, then. Of course I had _heard of_ transsexualism as a thing, in the form of the "woman trapped in a man's body" trope, but it wasn't something I expected to actually encounter in real life.
At the time, I had _no reason to invent the hypothesis_ that I might somehow literally be a woman in some unspecified psychological sense. I knew I was a boy _because_ boys are the ones with penises. That's what the word _means_. I was a boy who had a weird _sex fantasy_ about being a girl. That was just the obvious ordinary straightforward plain-language description of the situation. It _never occured to me_ to couch it in the language of "dysphoria", or actually possessing some innate "gender". The beautiful pure sacred self-identity thing was about identifying _with_ women, not identifying _as_ a woman—roughly analogous to how a cat lover might be said to "identify with" cats, without claiming to somehow _be_ a cat, because _that would be crazy_.
From the standpoint of my secret erotic fantasy, "normal, masculine man wearing a female body like a suit of clothing" is actually a _great_ outcome—the _ideal_ outcome. Let me explain.
-The main plot of my secret erotic fantasy accomodates many frame stories, but I tend to prefer those that invoke the [literary genre of science](https://www.lesswrong.com/posts/4Bwr6s9dofvqPWakn/science-as-attire), and posit technology indistinguishable from magic rather than magic _simpliciter_.
+The main plot of my secret erotic fantasy accomodates many frame stories, but I tend to prefer those that invoke the [literary genre of science](https://www.lesswrong.com/posts/4Bwr6s9dofvqPWakn/science-as-attire), and posit "technology" rather than "spells" or "potions" as the agent of change, even if it's all ultimately magic (where ["magic" is anything you don't understand](https://www.lesswrong.com/posts/kpRSCH7ALLcb6ucWM/say-not-complexity)).
So imagine having something like [the transporter in _Star Trek_](https://memory-alpha.fandom.com/wiki/Transporter), but you re-materialize with the body of someone else, rather than your original body—a little booth I could walk in, dissolve in a tingly glowy special effect for a few seconds, and walk out looking like (say) [Nana Visitor (circa 1998)](https://memory-alpha.fandom.com/wiki/Kay_Eaton?file=Kay_Eaton.jpg). (In the folklore of [female-transformation erotica](/2016/Oct/exactly-what-it-says-on-the-tin/), this machine is often called the ["morphic adaptation unit"](https://www.cyoc.net/interactives/chapter_115321.html).)
Do I have _any idea_ what it would even _mean_ to fix all that, without spending multiple lifetimes studying neuroscience?
-[TODO (working): rewrite this whole section to be more focused on _just_ explaining the math language needed to explain how the transformation mapping would work, using face and height as "easy" examples]
+I think I have just enough language to _start_ to talk about what it would mean. Since sex isn't an atomic attribute, but rather a high-level statistical regularity such that almost everyone can be cleanly classified as "female" or "male" _in terms of_ lower-level traits (genitals, hormone levels, _&c._), then, abstractly, we're trying to take points from male distribution and map them onto the female distribution in a way that preserves as much structure (personal identity) as possible. My female analogue doesn't have a penis (because then she wouldn't be female), but she is going to speak American English like me and be [85% Ashkenazi like me](/images/ancestry_report.png), because language and autosomal genes don't have anything to do with sex.
-I think I have just enough language to _start_ to talk about what it would mean. If sex isn't an atomic attribute, but rather a high-level statistical regularity such that people can be cleanly classified as "female" or "male" _in terms of_ lower-level traits (like hormone levels, genital shape, _&c._), then, abstractly, we're trying to take points from male distribution and map them onto the female distribution in a way that preserves as much structure as possible. My female analogue doesn't have a penis (because then she wouldn't be female), but she is going to speak American English like me and be [85% Ashkenazi like me](/images/ancestry_report.png), because language and autosomal genes don't have anything to do with sex.
+The hard part has to do with traits that are meaningfully sexually dimorphic, but not as a discrete dichotomy—where the sex-specific universal designs differ in ways that are _subtler_ than the presence or absence of entire reproductive organs. (Yes, I know about [homology](https://en.wikipedia.org/wiki/Homology_(biology))—and you know what I meant.) We are _not_ satisfied if the magical transformation technology swaps out my penis and testicles for a functioning female reproductive system without changing the rest of my body, because we want the end result to be indistinguishable from having been drawn from the female distribution (at least, indistinguishable _modulo_ having my memories of life as a male before the magical transformation), and a man-who-somehow-magically-has-a-vagina doesn't qualify.
-The subtle part as to do with traits that are meaningfully sexually dimorphic but not as a discrete dichotomy: we're _not_ satisfied with the transformation if the magical transformation technology swaps out my penis and testicles for a functioning female reproductive system but doesn't change the rest of my body.
+The "obvious" way to to do the mapping is to keep the same percentile rank within each trait, but take it with respect to the target sex's distribution. I'm 5′11″ tall, which [puts me at](https://dqydj.com/height-percentile-calculator-for-men-and-women/) the 73rd percentile for American men, about 6/10ths of a standard deviation above the mean. So _presumably_ we want to say that my female analogue is at the 73rd percentile for American women, about 5′5½″.
-[...]
+You might think this is "unfair": some women—about 7 per 1000—are 5′11″, and we don't want to say they're somehow _less female_ on that account, so why can't I keep my height? But if we refuse to adjust for every trait for which the female and male distributions overlap (on the grounds that _some_ women have the same trait value as my male self), we don't end up with a result from the female distribution.
-I'm 5′11″ tall, which [puts me at](https://dqydj.com/height-percentile-calculator-for-men-and-women/) the 73rd percentile for men, about 6/10ths of a standard deviation above the mean. So _presumably_ we want to say that my female analogue is at the 73rd percentile for women, about 5′5½″.
+The typical point in a high-dimensional distribution is _not_ typical along each dimension individually. [In 100 flips of a biased coin](http://zackmdavis.net/blog/2019/05/the-typical-set/) that lands Heads 0.6 of the time, the _single_ most likely sequence is 100 Heads, but there's only one of those and you're _vanishingly_ unlikely to actually see it. The sequences you'll actually observe will have close to 60 Heads. Each such sequence is individually less probable than the all-Heads sequence, but there are vastly more of them. Similarly, [most of the probability-mass of a high-dimensional multivariate normal distribution is concentrated in a thin "shell" some distance away from the mode](https://www.johndcook.com/blog/2011/09/01/multivariate-normal-shell/), for the same reason. (The _same_ reason: the binomial distribution converges to the normal in the limit of large _n_.)
-[...]
+Statistical sex differences are like flipping two different collections of coins with different biases, where the coins represent various traits. Learning the outcome of any individual flip, doesn't tell you which which set the coin came from, but [if we look at the aggregation of many flips, we can get _godlike_ confidence](https://www.lesswrong.com/posts/cu7YY7WdgJBs3DpmJ/the-univariate-fallacy-1) as to which collection we're looking at.
+
+A single-variable measurement like height is like a single coin: unless the coin is _very_ biased, one flip can't tell you much about the bias. But there are lots of things about people for which it's not that they can't be measured, but that the measurements require _more than one number_—which correspondingly offer more information about the distribution generating them.
+
+[TODO (somewhere around-ish this section): chromosomes at the root of the causal graph: https://www.lesswrong.com/posts/hzuSDMx7pd2uxFc5w/causal-diagrams-and-causal-models ]
+
+Take faces. People are [verifiably very good at recognizing sex from (hair covered, males clean-shaven) photographs of people's faces](/papers/bruce_et_al-sex_discrimination_how_do_we_tell.pdf) (96% accuracy, which is the equivalent of _d_ ≈ 3.5), but we don't have direct introspective access into what _specific_ features our brains are using to do it; we just look, and _somehow_ know. The differences are real, but it's not a matter of any single, simple measurement you could perform with a ruler (like the distance between someone's eyes). Rather, it's a high-dimensional _pattern_ in many measurements you could take with a ruler, no one of which is definitive. [Covering up the nose makes people slower and slightly worse at sexing faces, but people don't do better than chance at guessing sex from photos of noses alone](/papers/roberts-bruce-feature_saliency_in_judging_the_sex_and_familiarity_of_faces.pdf).
+Notably, for _images_ of faces, we actually _do_ have transformation technology! (Not "magical", because we know how it works.) AI techniques like [generative adversarial networks](https://arxiv.org/abs/1812.04948) and [autoencoders](https://towardsdatascience.com/generating-images-with-autoencoders-77fd3a8dd368) can learn the structure of the distribution of facial photographs, and use that knowledge to [synthesize faces from scratch](https://thispersondoesnotexist.com/)—or [do things like](https://arxiv.org/abs/1907.10786) sex transformation, as demonstrated by the likes of [FaceApp](https://www.faceapp.com/), the _uniquely best piece of software in the world_.
-People are [verifiably very good at recognizing sex from (hair covered, males clean-shaven) photographs of people's faces](/papers/bruce_et_al-sex_discrimination_how_do_we_tell.pdf) (96% accuracy, which is the equivalent of _d_ ≈ 3.5), but we don't have direct introspective access into what _specific_ features our brains are using to do it; we just look, and _somehow_ know. The differences are real, but it's not a matter of any single measurement: [covering up the nose makes people slower and slightly worse at sexing faces, but people don't do better than chance at guessing sex from photos of noses alone](/papers/roberts-bruce-feature_saliency_in_judging_the_sex_and_familiarity_of_faces.pdf).
+If you let each pixel vary independently, the space of possible 1024x1024 images is 1,048,576-dimensional, but the vast hypermajority of those images aren't photorealistic human faces. Letting each pixel vary independently is the wrong way to think about it: changing the lighting or pose changes can change a lot of pixels in what we would regard as images of "the same" face. Instead, our machine-learning algorithms learn a [compressed](https://www.lesswrong.com/posts/ex63DPisEjomutkCw/msg-len) representation of what makes the tiny subspace (relative to images-in-general) of _faces in particular_ similar to each other, and that [latent space](https://towardsdatascience.com/understanding-latent-space-in-machine-learning-de5a7c687d8d) is a lot smaller—say, 512 dimensions.
-[TODO: Mathematically,
-Joel et al. and response—maybe in next paragraph
-Beyond the Binary: https://www.pnas.org/content/112/50/15468
-http://cogprints.org/10046/1/Delgiudice_etal_critique_joel_2015.pdf
+[TODO: separating hyperplane / Face editing with Generative Adversarial Networks: https://www.youtube.com/watch?v=dCKbRCUyop8 ]
-http://zackmdavis.net/blog/2019/05/the-typical-set/
-> once you draw a boundary around a group, the mind starts trying to harvest similarities from the group. And unfortunately the human pattern-detectors seem to operate in such overdrive that we see patterns whether they're there or not; a weakly negative correlation can be mistaken for a strong positive one with a bit of selective memory.
-https://www.lesswrong.com/posts/veN86cBhoe7mBxXLk/categorizing-has-consequences
-[a higher-dimensional statistical regularity in the _conjunction_ of many variables](https://www.lesswrong.com/posts/cu7YY7WdgJBs3DpmJ/the-univariate-fallacy-1)
-96.8% classification from MRI https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6374327/
-]
-[the wrists: http://unremediatedgender.space/papers/yune_et_al-beyond_human_perception_sexual_dimorphism_in_hand_and_wrist_radiographs.pdf]
-[talk about mapping from one distribution to another: e.g. height]
+[...]
The same moral applies to sex differences in psychology. I'm a pretty weird guy, in more ways than one. I am not prototypically masculine. Most men are not like me. If I'm allowed to cherry-pick what measurements to take, I can name ways in which I'm more female-typical than male-typical. (For example, I'm _sure_ I'm above the female mean in [Big Five Neuroticism](https://en.wikipedia.org/wiki/Big_Five_personality_traits).) But "weird" represents a much larger space of possibilities than "normal", much as [_nonapples_ are a less cohesive category than _apples_](https://www.lesswrong.com/posts/2mLZiWxWKZyaRgcn7/selling-nonapples).
If you _sum over_ all of my traits, everything that makes me, _me_—it's going to be a point in the _male_ region of the existing, unremediated, genderspace. In principle, you could define a procedure that maps that point to the female region of configuration space in some appropriately structure-preserving way, to compute my female analogue who is as authentically _me_ as possible while also being authentically female, down to the proportion of gray matter in her posterior lateral orbitofrontal cortex and—the love of a woman for a man.
-(Note that we can already basically do this for _images_ of female and male faces, using the [latent spaces found by generative adversarial networks](https://arxiv.org/abs/1907.10786) and [autoencoders](https://towardsdatascience.com/generating-images-with-autoencoders-77fd3a8dd368), as demonstrated by the likes of [FaceApp](https://www.faceapp.com/), the _uniquely best piece of software in the world_. Doing it for _actual whole people in the real world_ and not just flat images is a task for future superintelligences, not present-day GANs, but some of same basic principles should apply.)
-
-[TODO: mention https://thispersondoesnotexist.com/ ; we can synthesize images from scratch]
+[...]
-Okay. Having supplied just enough language to _start_ to talk about what it would mean to actually become female—is that what I _want_? I mean, I would definitely be extremely eager to _try_ it ...
+Okay. Having supplied just enough language to _start_ to talk about what it would mean to actually become female—is that what I _want_? I mean, if it's reversible, I would definitely be extremely eager to _try_ it ...
-I had said we're assuming away engineering difficulties in order to make the thought experiment more informative about pure preferences, but let's add back one constraint to _force_ the thought experiment to be informative about preferences, and not allow the wishy-washy evasion of "I'm eager to _try_ it."
+I had said we're assuming away engineering difficulties in order to make the thought experiment more informative about pure preferences, but let's add one constraint to _force_ the thought experiment to be informative about preferences, and not allow the wishy-washy evasion of "I'm eager to _try_ it."
What if I can't just "try" it? What if the machine can only be used once? Come up with whatever frame story you want for this: maybe the machine costs my life savings just to rent for two minutes, or maybe the transformation process is ever-so-slightly imperfect, such that you can't re-transform someone who's already been transformed once, like a photocopy being a perfectly acceptable substitute for an original document, but photocopies-of-photocopies rapidly losing quality.
-In that case, if I have to choose—I _don't_ think I want to be Actually Female. I _like_ who I am on the inside, and don't need to change it. So, I guess ... if I'm being honest ... I guess I _want_ to be a normal man wearing a female body like a suit of clothing.
+In that case, if I have to choose ... I _don't_ think I want to be Actually Female? I _like_ who I am on the inside, and don't need to change it. So, I guess ... if I'm being honest ... I guess I _want_ to be a normal man wearing a female body like a suit of clothing.
Is that weird? Is that wrong?
If the magical transformation technology were real, I would want a mirror. (And in the real world, I would probably crossdress a _lot_ more often, if I could pass to myself in the mirror.)
-What's going on here? The sexologist James Cantor [speculates](https://youtu.be/q3Ub65CwiRI?t=281): mirror neurons.
+What's going on here? The sexologist [James Cantor speculates](https://youtu.be/q3Ub65CwiRI?t=281): mirror neurons.
https://www.lesswrong.com/posts/NMoLJuDJEms7Ku9XS/guessing-the-teacher-s-password
The rules don't change when the entity X happens to be "my female analogue" and the category Y happens to be "me". The ordinary concept of "personal identity" tracks how the high-level features of individual human organisms are stable over time. You're going to want to model me-on-Monday and me-on-Thursday as "the same" person even if my Thursday-self woke up on the wrong side of bed and has three whole days of new memories. When interacting with my Thursday-self, you're going to be using your existing mental model of me, plus a diff for "He's grumpy" and "Haven't seen him in three days"—but that's a _very small_ diff, compared to the diff between me and some other specific person you know, or the diff between me and a generic human who you don't know.
+[TODO: address Yudkowsky on "the change is too sharp"]
+
In everyday life, we're almost never in doubt as to which entities we want to consider "the same" person, but we can concoct science-fictional thought experiments that force [the Sorites problem](https://plato.stanford.edu/entries/sorites-paradox/) to come up. What if you could "merge" two people—construct a human with a personality "in between" yours and mine, that had both of our memories? (You know, like [Tuvix](https://memory-alpha.fandom.com/wiki/Tuvix_(episode)).) Would that person be me, or you, or both, or neither? (Derek Parfit has [a book](https://en.wikipedia.org/wiki/Reasons_and_Persons#Personal_identity) with lots of these.)
[TODO: change scenario to interpolate between people, _at what point_ does it become]
-The thing about Sorites problems is that they're _incredibly boring_. The map is not the territory. The distribution of sand-configurations we face in everyday life is such that we usually have an answer as to whether the sand "is a heap" or "is not a heap", but in the edge-cases where we're not sure, arguing about whether to use the word "heap" _doesn't change the configuration of sand_. You might think that if [the category is blurry](https://www.lesswrong.com/posts/dLJv2CoRCgeC2mPgj/the-fallacy-of-gray), you therefore have some freedom to [draw its boundaries](https://www.lesswrong.com/posts/d5NyJ2Lf6N22AD9PB/where-to-draw-the-boundary) the way you prefer—but [the cognitive function of the category is for making probabilistic inferences on the basis of category-membership](https://www.lesswrong.com/posts/esRZaPXSHgWzyB2NL/where-to-draw-the-boundaries), and those probabilistic inferences can be quantitatively better or worse. Preferences over concept definitions that aren't about maximizing predictive accuracy are therefore preferences _for deception_, because "making probability distributions less accurate in order to achieve some other goal" is the only thing _dece
+The thing about Sorites problems is that they're _incredibly boring_. The map is not the territory. The distribution of sand-configurations we face in everyday life is such that we usually have an answer as to whether the sand "is a heap" or "is not a heap", but in the edge-cases where we're not sure, arguing about whether to use the word "heap" _doesn't change the configuration of sand_. You might think that if [the category is blurry](https://www.lesswrong.com/posts/dLJv2CoRCgeC2mPgj/the-fallacy-of-gray), you therefore have some freedom to [draw its boundaries](https://www.lesswrong.com/posts/d5NyJ2Lf6N22AD9PB/where-to-draw-the-boundary) the way you prefer—but [the cognitive function of the category is for making probabilistic inferences on the basis of category-membership](https://www.lesswrong.com/posts/esRZaPXSHgWzyB2NL/where-to-draw-the-boundaries), and those probabilistic inferences can be quantitatively better or worse. Preferences over concept definitions that aren't about maximizing predictive accuracy are therefore preferences _for deception_, because "making probability distributions less accurate in order to achieve some other goal" is the only thing _deception_ can mean.
That's why defining your personal identity to get the answer you want is cheating. If the answer you wanted was actually _true_, you could just say so without needing to _want_ it.
When [Phineas Gage's](/2017/Dec/interlude-xi/) friends [said he was "no longer Gage"](https://en.wikipedia.org/wiki/Phineas_Gage) after the railroad accident, what they were trying to say was that interacting with post-accident Gage was _more relevantly similar_ to interacting with a stranger than it was to interacting with pre-accident Gage, even if Gage-the-physical-organism was contiguous along the whole strech of space time.
-Same principle when Yudkowsky wrote, "If I fell asleep and woke up as a true woman [...] I don't think I'd call her 'me'"—maybe he was wrong, but if so, that cashes out as being wrong _about_ some precise statement about the actual properties of similarity metrics on persons (which could be computed in principle, even if _we_ don't know how), _not_ whether we prefer the "call her me" or "don't call her me" conclusion and want to _define the interpretation of those phrases in order to make the claim come out "true."_
+Same principle when Yudkowsky wrote, "If I fell asleep and woke up as a true woman [...] I don't think I'd call her 'me'"—maybe he was wrong, but if so, that cashes out as being wrong _about_ the actual properties of similarity metrics on persons (which could be precisely computed in principle, even if _we_ don't know how to make it precise), _not_ whether we prefer the "call her me" or "don't call her me" conclusion and want to _retroactively redefine the meaning of the words in order to make the claim come out "true."_
[TODO: being at peace with what's real ...]
Imagine my surprise to discover that, in the current year, my weird sexual obsession is suddenly at the center of [one of the _defining political issues of our time_](https://en.wikipedia.org/wiki/Transgender_rights).
-All this time—the dozen years I spent reading everything I could about sex and gender and transgender and feminism and evopsych and doing various things with my social presentation (sometimes things I regretted and reverted after a lot of pain, like the initials) to try to seem not-masculine—I had been _assuming_ that my gender problems were not of the same kind as people who were _actually_ transgender, because the standard narrative said that that was about people whose ["internal sense of their own gender does not match their assigned sex at birth"](https://www.vox.com/identities/21332685/trans-rights-pronouns-bathrooms-sports), and I had never interpreted my beautiful pure sacred self-identity thing as an "internal sense of my own gender."
+All this time—the dozen years I spent reading everything I could about sex and gender and transgender and feminism and evopsych and doing various things with my social presentation (sometimes things I regretted and reverted after a lot of pain, like the initials) to try to seem not-masculine—I had been _assuming_ that my gender problems were not of the same kind as people who were _actually_ transgender, because the standard narrative said that that was about people whose ["internal sense of their own gender does not match their assigned sex at birth"](https://www.vox.com/identities/21332685/trans-rights-pronouns-bathrooms-sports), whereas my thing was obviously an outgrowth of my weird sex fantasy—I had never interpreted the beautiful pure sacred self-identity thing as an "internal sense of my own gender".
+
+_Why would I?_ In the English of my youth, "gender" (as a single word, rather than part of the phrase "gender role") was understood as a euphemism for _sex_ for people who were squeamish about the hypothetical ambiguity betweeen _sex_-as-in-biological-sex and _sex_-as-in-intercourse. In that language, my "gender"—my sex—is male. Not because I'm necessarily happy about it (and I [used to](/2017/Jan/the-erotic-target-location-gift/) be pointedly insistent that I wasn't), but as an observable biological fact that, whatever my pure beautiful sacred self-identity feelings, _I am not delusional about_.
+
/2018/Jan/dont-negotiate-with-terrorist-memeplexes/
projects / Ultimately_Untrue_Thought.git / blobdiff