From 50376699800cb462c55924b685bf56190427511d Mon Sep 17 00:00:00 2001 From: "M. Taylor Saotome-Westlake" Date: Thu, 9 Jun 2022 11:07:17 -0700 Subject: [PATCH] drafting "Consilient Cultural Worldbuilding": It's Just Bayes --- ...nd-the-incoherence-of-nondiscrimination.md | 53 +++++++++++++------ 1 file changed, 36 insertions(+), 17 deletions(-) diff --git a/content/drafts/consilient-cultural-worldbuilding-and-the-incoherence-of-nondiscrimination.md b/content/drafts/consilient-cultural-worldbuilding-and-the-incoherence-of-nondiscrimination.md index e28134a..4c5196a 100644 --- a/content/drafts/consilient-cultural-worldbuilding-and-the-incoherence-of-nondiscrimination.md +++ b/content/drafts/consilient-cultural-worldbuilding-and-the-incoherence-of-nondiscrimination.md @@ -4,7 +4,36 @@ Category: commentary Tags: Eliezer Yudkowsky, worldbuilding Status: draft -Realistic worldbuilding is a difficult art: unable to model what someone else would do except by imagining oneself in that position, authors tend to embarrass themselves writing [alleged aliens or AIs that _just happen_ act like humans](https://www.lesswrong.com/posts/Zkzzjg3h7hW5Z36hK/humans-in-funny-suits), or allegedly foreign cultures that _just happen_ to share all of the idiosyncratic taboos of the author's own culture. The manifestations of this can be very subtle, even to authors who know about the trap. +Realistic worldbuilding is a difficult art: unable to model what someone else would do except by the ["empathic inference"](https://www.lesswrong.com/posts/9fpWoXpNv83BAHJdc/the-comedy-of-behaviorism) of imagining oneself in that position, authors tend to embarrass themselves writing [alleged aliens or AIs that _just happen_ act like humans](https://www.lesswrong.com/posts/Zkzzjg3h7hW5Z36hK/humans-in-funny-suits), or allegedly foreign cultures that _just happen_ to share all of the idiosyncratic taboos of the author's own culture. The manifestations of this can be very subtle, even to authors who know about the trap. + +[TODO: minimal Planecrash summary] + +[TODO: Keltham's proposal] + +... and that's where my suspension of disbelief in Keltham as + +The entire idea of "The government can't discriminate by race or sex" as a _principle_ (as contrasted to most laws happening to not refer to race or sex because they happen to not be relevant to that specific law) is a specific form of Earth-craziness that only makes sense as a reaction to other Earth-craziness. It's not something you would ever spontaneously invent or think was a good idea if you _actually_ came a 140 IQ Society that thoroughly educated everyone in probability theory as normative reasoning. Let me explain the principles first, and then dissect the example of military conscription (possibly _the worst_ possible example the authors could have gone with). + +The issue is that probability theory doesn't have any built-in concept of "protected classes." On pain of paradox, Bayesians _must_ condition on all available information. If groups differ in decision-relevant traits, _of course_ you should treat members of those groups differently! What we call "discrimination" in America on Earth is actually just Bayesian reasoning; P(H|E) = P(E|H)P(H)/P(E) doesn't _stop being true_ when H happens to be "I should hire this candidate" and E happens to be "The candidate is a halfling". Furthermore, there's no reason for the law to behave differently in this respect than a private individual: is Governance supposed to be _less_ Bayesian _because it's Governance_?! + +(As an aside, it's actually kind of _hilarious_ how far Yudkowsky's "rationalist" movement has succeeded at winning status and mindshare in a Society whose [_de facto_ state religion](https://slatestarcodex.com/2019/07/08/gay-rites-are-civil-rites/) is [founded on eliminating "discrimination."](https://richardhanania.substack.com/p/woke-institutions-is-just-civil-rights) Did—did anyone besides me "get the joke"? I would have expected _Yudkowsky_ to get the joke, but I guess not??) + +Of course, as Keltham correctly points out, if you have more specific information about an individual that [screens off](https://www.lesswrong.com/posts/5yFRd3cjLpm3Nd6Di/argument-screens-off-authority) information from their demographic category, then you should use the more specific information: once you measure someone's height, the fact that men are taller than women with an effect size of 1.7 standard deviations is no longer relevant to the question of that person's height. In very many situations, if there's a cost associated with acquiring more specific individuating information that renders information from demographic group-membership irrelevant, you should pay that cost in order to get the more specific information and therefore make better decisions. + +But crucially, getting individuating information is an [instrumental rather than a terminal value](https://www.lesswrong.com/posts/n5ucT5ZbPdhfGNLtP/terminal-values-and-instrumental-values); you should do it _when and because_ it improves your decisions, not because of some alleged principle that you're _not allowed to notice_ someone's race or sex. If there's a _cost_ associated with taking individual measurements, and the cost exceeds the amount you would save by making better decisions, then you shouldn't take the measurements. If your measurements have _error_, then your estimate of the true value of the trait being measured [regresses to the group mean](https://en.wikipedia.org/wiki/Regression_toward_the_mean) to some quantitative exent; again, this just falls out of _ordinary_ Bayesian reasoning, which continues to work even when some of the hypotheses are about groups of people. + +If this still seems counterintuitive, it may help to consider that from the standpoint of Just Doing Bayesian Decision Theory, the distinction between "information from demographic group membership" and "information from individual measurements" isn't fundamental. + +"Sex" and "race" aren't + +https://www.lesswrong.com/posts/vhp2sW6iBhNJwqcwP/blood-is-thicker-than-water + +* "sex" and "wisdom" are both abstract categories that it would suck if the category mispredicted you; there's no principled difference +* The specific military example + +---- + +[TODO— make intro much shorter; minimal amount of words/info to set up the context for my complaint In _Planecrash_, a collaborative roleplaying fiction principally by Iarwain (a pen name of Eliezer Yudkowsky) and Lintamande, our protagonist, Keltham, hails from [dath ilan](https://www.lesswrong.com/tag/dath-ilan), a more smarter, more rational, and better-coordinated alternate version of Earth. Keltham has somehow survived his apparent death and woken up in the fantasy world of [Golarion](https://pathfinderwiki.com/wiki/Golarion), and sets about uplifting the natives using knowledge from his superior civilization. @@ -21,8 +50,13 @@ It makes sense that Keltham wants to smash the patriarchy in Osiron, but I'm sur Of course, here in America on Earth, there are historical reasons that _our_ culture has come to uphold equality under the law as a _principle_, rather than most laws just happening not to treat different groups differently. It's a sensible precaution if you don't trust your government or your culture: if a law that distinguishes demographic groups could be used to oppress one of those groups, don't allow _any_ such laws, even if they come with a purportedly benevolent rationale attached. But Keltham isn't _from_ America. Everything we've heard about his world says that they educate everyone thoroughly in probability theory as normative reasoning, and that citizens end up trusting the existing government on the basis that they would know about and could overthrow a corrupt government. In _that_ context, equality under the law is ... much less obvious of a desideratum? +] + + + +A principle that the law can only refer to lower-dimensional concepts (like "Wisdom") but isn't allowed to [refer to clusters](/2021/Mar/link-see-color/) in [thick subspaces of configuration space](https://www.lesswrong.com/posts/esRZaPXSHgWzyB2NL/where-to-draw-the-boundaries) (like "is a halfling") is a principle that _decreases the expressive power of the law_, restricting the ontology that the law is allowed to reason about: effectively, saying that the government has to be _less Bayesian_ because it's the government. + -The issue is that probability theory doesn't have any built-in concept of "protected classes." A principle that the law can only refer to lower-dimensional concepts (like "Wisdom") but isn't allowed to [refer to clusters](/2021/Mar/link-see-color/) in [thick subspaces of configuration space](https://www.lesswrong.com/posts/esRZaPXSHgWzyB2NL/where-to-draw-the-boundaries) (like "is a halfling") is a principle that _decreases the expressive power of the law_, restricting the ontology that the law is allowed to reason about: effectively, saying that the government has to be _less Bayesian_ because it's the government. In America, we're used to objecting, "But it's unfair to treat someone as representative of their race or sex, because some people are atypical for their group; you need to look at their individual traits, like Intelligence or Charisma". But really, individual "traits" are _also_ an abstraction that sums over individual differences: someone might be more charming to certain people or in certain contexts in complicated ways that a single Charisma score can't express. In that light, it's not obvious why the objection against using demographic categories as predictors is more compelling than, "But it's unfair to treat someone as representative of their Intelligence or Charisma, because some people are atypical for their trait score, you need to look at individual sub-traits" ... and so on recursively? Are all forms of abstraction-for-statistical-prediction inherently oppressive? @@ -59,18 +93,3 @@ neglect of probability (https://www.lesswrong.com/posts/q7Me34xvSG3Wm97As/but-th > They end up wanting 42 major things and 314 minor things (on the current count of what's known and thought to be distinct in the way of adaptation) https://www.glowfic.com/replies/1801140#reply-1801140 - ----- - -OUTLINE - -* realistic worldbuilding is hard, people write aliens or foreigners just like themselves; it's easy to fall into this trap even if you ostensibly know better -* Planecrash summary -* Keltham's proposal -* This entire idea of "The government can't discriminate by race or sex" is specific Earth-craziness that only makes sense in relation to other Earth-craziness; it's not something you would spontaneously invent if you actually grew up in an IQ 140 Society that taught everyone probability theory as normative reasoning: in this essay, I will first explain the principle, and then dissect Keltham's example of military service which is the _uniquely worst possible example_ -* Probabily theory has no protected classes -* If you happen to have more fine-grained information, fine, but you don't always -* Expense of taking additional measurements -* Measurement error -* "sex" and "wisdom" are both abstract categories that it would suck if the category mispredicted you; there's no principled difference -* The specific military example -- 2.17.1