1 Title: The Categories Were Made for Man in Order to Make Predictions
4 Tags: epistemology, Scott Alexander
7 > I said, "The truth is whatever you can get away with."
9 > "No, that's journalism. The truth is whatever you can't escape."
11 > —_Distress_ by Greg Egan
13 In ["The Categories Were Made for Man, Not Man for the Categories"](http://slatestarcodex.com/2014/11/21/the-categories-were-made-for-man-not-man-for-the-categories/), the immortal Scott Alexander eloquently argues that proposed definitions of concepts aren't true or false in themselves, but rather can only be evaluated by their usefulness. Our finite minds being unable to cope with the unimaginable complexity of the raw physical universe, we group sufficiently similar things into the same category so that we can make similar predictions about them—but this requires not only a metric of "similarity", but also a notion of which predictions one cares about enough to notice, both of which are relative to some agent's perspective, rather than being inherent in the world itself.
15 And so, Alexander explains, the ancient Hebrews weren't _wrong_ to classify whales as a type of _dag_ (typically translated as _fish_), even though modern biologists classify whales as mammals and not fish, because the ancient Hebrews were more interested in distinguishing which animals live in the water rather than which animals are phylogenetically related. Similarly, borders between countries are agreed upon for a variety of pragmatic reasons, and can be quite convoluted—while there may often be some "obvious" geographic or cultural Schelling points anchoring these decisions, there's not going to be any instrinsic, eternal fact of the matter as to where one country starts and another begins.
17 All of this is entirely correct—and thus, an excellent [motte](http://slatestarcodex.com/2014/11/03/all-in-all-another-brick-in-the-motte/) for the less honest half of _Slate Star Codex_ readers to appeal to when they want to disrupt and obfuscate discussions about empirical reality by insisting on [...]
21 Alexander goes on to attempt to use the categories-are-relative insight to rebut skeptics of transgenderedness: referring to trans people as their desired gender is a category boundary
31 We can at least discuss _in detail_
42 You can call a tail a leg, but you can't stop people from _noticing_ that of a dog's five legs, one of them is different from the others, so different that people habitually distinguish between the walking-legs and the wagging-leg.
51 * First, the basic point is obviously correct.
52 * Things become muddier when we get to the section about national borders:
53 it's true that countries may agree that their borders work in this
54 noncontiguous way, and that matters for some purposes, but these legal
55 fictions don't always capture what people actually mean when they talk about
57 * During my Facebook meltdown, an acquaintance of mine gave some really
58 solid examples that I should ask him if I can borrow (with or without
60 * An attempted clarification (Scott does address this, but I want to
61 emphasize/rephrase it?): categories are value-laden because they're grouping
62 things together on the basis of the predictions that are decision-relevant
63 to what you care about. But this should be distinguished from the categories
64 themselves being _arbitrarily_ value-laden.
65 * Scott Alexander understands this, but the bottom 50% of _Slate Star
66 Codex_ readers do not.
67 * An important subtlety: in the case of countries and gender but _not_
68 whales/fish, we're not just classifying things that exist, but also making a
69 political decision of how we're going to organize ourselves. It's not that
70 trans women and nonbinaries already exist in fixed proportions, and we're
71 trying to decide how to parse them: that we have socially-recognized
72 transition as a thing creates an affordance for people to make the choice to
73 transition (linky ["Lesser Known..."](/2017/Dec/lesser-known-demand-curves/))
74 * The argumentation in section IV is _uncharacteristically_ weak for Scott:
75 basically just "We've established that categories are neither true nor
76 false, so if you care about transgender people, then you should use their
78 * To do better: let's _discuss in detail_ the detailed consequences of
79 different ways of drawing gender categories, and analyze the conflicts
80 that different people have.
81 * The case for using identified gender rather than biological sex is
82 strongest for binary trans people who actually pass.
83 * The tack where you show a picture of Buck Angel and say, "You're not
84 really going to call this person a woman, are you?" makes a good point
85 * It's less strong for ...
86 * People who don't pass
87 * Passing is a continuum rather than a binary and is also
88 observer-dependent, which is inconvenient from the perspective of
89 categorization, which tends to stick to bright-lines and Schelling
91 * Re observer dependence: quote Serano (do I rely/pick on Serano too
92 much?) or someone about how it's actually _harder_ to pass in urban
93 areas because people have a higher prior
94 * Ref-to-rebut Zinnia Jones on "passing is subjective, therefore it's
97 * Normies don't have nonbinary gender in their ontology; at least
98 acknowledge that you're making a political demand when you want
100 * What are the decision criteria for nonbinary, anyway? People can
102 * Rundown of social consequences of different criteria—
103 * When you have people who are _identifiably_ distinct from natal-sex
104 people _and_ not drawn from the same psychological distribution, it
105 becomes socially profitable for people to notice and adjust their
106 expectations; you can't stop them from doing this
107 * separate post "Stereotypes, Models, and Cognition"
108 * People are making probabilistic inferences all the time whether
109 they realize it or not
110 * Being drawn from a different psychological distribution but _not_
111 identifiably (AGPs who pass really well) doesn't hurt the dynamics
113 * AGPs aren't drawn from the same psychological distribution as cis
114 women. (Briefly explain the typology, but refer to external sources
115 for justification. For a more hard-facts empirical justification of
116 "not drawn from the same dist'n", cite data on sexual orientation (and
117 [constrast](https://twitter.com/SteveStuWill/status/905572666332987392))
119 * Making it not-OK for people to _talk_ about the categories that they
120 internally use to make sense of the world is bad
121 * experiences in LWish spaces with lots of trans women: if you
122 doctrinairely call everyone women, my brain rebels and wants to
123 say, "That's not what I meant and _you fucking know it_." And
124 honestly? (And I think they do, in fact, fucking know it.)
125 * The inability to have women's clothing swaps is a _real loss_
126 * Negotiation-structure: we've been using this word to refer to this
127 thing for the past 200,000 years since the invention of language;
128 if you want us to stop, you need to offer us something we value
129 (and you have nothing to trade with); threatening to kill yourself
130 is easily (if callously) countered with "We don't negotiate with
132 * For crime/medical statistics, you need natal sex or third-category.
134 * The case of Emperor Norton looks cute at first glance, but
135 ostensibly-benevolent gaslighting is still problematic (we call it
137 * I've been crazy (link "Memoirs"), and I'm glad my friends patiently told
138 me why I was wrong rather than saying "That's nice dear" (maybe quote
140 * Tell a story about what this could have been like for Norton beginning to
141 doubt the reality of his reign.
142 * Link to Maria Catt's "Baby Jessica" essay (maybe write her fan mail and
143 ask her to put it back up again)
146 /2017/Feb/if-other-fantasies-were-treated-like-crossdreaming/
148 Similarly, [discussion of borders]
150 [point out that legal fictions aren't always taken seriously by people who are trying to talk about the world, use "Europe" examples from acquaintance; Seeing Like a State]
152 [point out that Alexander agrees that some categories suck]