Mandatory Secret Identities

Pre­vi­ously in se­ries: Whin­ing-Based Communities

“But there is a rea­son why many of my stu­dents have achieved great things; and by that I do not mean high rank in the Bayesian Con­spir­acy. I ex­pected much of them, and they came to ex­pect much of them­selves.” —Jeffreyssai

Among the failure modes of mar­tial arts do­jos, I sus­pect, is that a suffi­ciently ded­i­cated mar­tial arts stu­dent, will dream of...

...be­com­ing a teacher and hav­ing their own mar­tial arts dojo some­day.

To see what’s wrong with this, imag­ine go­ing to a class on liter­ary crit­i­cism, fal­ling in love with it, and dream­ing of some­day be­com­ing a fa­mous liter­ary critic just like your pro­fes­sor, but never ac­tu­ally writ­ing any­thing. Writ­ers tend to look down on liter­ary crit­ics’ un­der­stand­ing of the art form it­self, for just this rea­son. (Or­son Scott Card uses the anal­ogy of a wine critic who listens to a wine-taster say­ing “This wine has a great bou­quet”, and goes off to tell their stu­dents “You’ve got to make sure your wine has a great bou­quet”. When the stu­dent asks, “How? Does it have any­thing to do with grapes?” the critic replies dis­dain­fully, “That’s for grape-grow­ers! I teach wine.”)

Similarly, I pro­pose, no stu­dent of ra­tio­nal­ity should study with the pur­pose of be­com­ing a ra­tio­nal­ity in­struc­tor in turn. You do that on Sun­days, or full-time af­ter you re­tire.

And to place a go stone block­ing this failure mode, I pro­pose a re­quire­ment that all ra­tio­nal­ity in­struc­tors must have se­cret iden­tities. They must have a life out­side the Bayesian Con­spir­acy, which would be wor­thy of re­spect even if they were not ra­tio­nal­ity in­struc­tors. And to en­force this, I sug­gest the rule:

Ra­tion­al­ity_Re­spect1(In­struc­tor) = min(Ra­tion­al­ity_Re­spect0(In­struc­tor), Non_Ra­tion­al­ity_Re­spect0(In­struc­tor))

That is, you can’t re­spect some­one as a ra­tio­nal­ity in­struc­tor, more than you would re­spect them if they were not ra­tio­nal­ity in­struc­tors.

Some notes:

• This doesn’t set Ra­tion­al­ity_Re­spect1 equal to Non_Ra­tion­al­ity_Re­spect0. It es­tab­lishes an up­per bound. This doesn’t mean you can find ran­dom awe­some peo­ple and ex­pect them to be able to teach you. Ex­plicit, ab­stract, cross-do­main un­der­stand­ing of ra­tio­nal­ity and the abil­ity to teach it to oth­ers is, un­for­tu­nately, an ad­di­tional dis­ci­pline on top of do­main-spe­cific life suc­cess. New­ton was a Chris­tian etcetera. I’d rather hear what Laplace had to say about ra­tio­nal­ity—Laplace wasn’t as fa­mous as New­ton, but Laplace was a great math­e­mat­i­cian, physi­cist, and as­tronomer in his own right, and he was the one who said “I have no need of that hy­poth­e­sis” (when Napoleon asked why Laplace’s works on ce­les­tial me­chan­ics did not men­tion God). So I would re­spect Laplace as a ra­tio­nal­ity in­struc­tor well above New­ton, by the min() func­tion given above.

• We should be gen­er­ous about what counts as a se­cret iden­tity out­side the Bayesian Con­spir­acy. If it’s some­thing that out­siders do in fact see as im­pres­sive, then it’s “out­side” re­gard­less of how much Bayesian con­tent is in the job. An ex­per­i­men­tal psy­chol­o­gist who writes good pa­pers on heuris­tics and bi­ases, a suc­cess­ful trader who uses Bayesian al­gorithms, a well-sel­l­ing au­thor of a gen­eral-au­di­ences pop­u­lar book on athe­ism—all of these have wor­thy se­cret iden­tities. None of this con­tra­dicts the spirit of be­ing good at some­thing be­sides ra­tio­nal­ity—no, not even the last, be­cause writ­ing books that sell is a fur­ther difficult skill! At the same time, you don’t want to be too lax and start re­spect­ing the in­struc­tor’s abil­ity to put up prob­a­bil­ity-the­ory equa­tions on the black­board—it has to be visi­bly out­side the walls of the dojo and noth­ing that could be sys­tem­atized within the Con­spir­acy as a to­ken re­quire­ment.

• Apart from this, I shall not try to spec­ify what ex­actly is wor­thy of re­spect. A cre­ative mind may have good rea­son to de­part from any crite­rion I care to de­scribe. I’ll just stick with the idea that “Nice ra­tio­nal­ity in­struc­tor” should be bounded above by “Nice se­cret iden­tity”.

But if the Bayesian Con­spir­acy is ever to pop­u­late it­self with in­struc­tors, this crite­rion should not be too strict. A sim­ple test to see whether you live in­side an elite bub­ble is to ask your­self whether the per­centage of PhD-bear­ers in your ap­par­ent world ex­ceeds the 0.25% rate at which they are found in the gen­eral pop­u­la­tion. Be­ing a math pro­fes­sor at a small uni­ver­sity who has pub­lished a few origi­nal proofs, or a suc­cess­ful day trader who re­tired af­ter five years to be­come an or­ganic farmer, or a se­rial en­trepreneur who lived through three failed star­tups be­fore go­ing back to a more or­di­nary job as a se­nior pro­gram­mer—that’s noth­ing to sneeze at. The vast ma­jor­ity of peo­ple go through their whole lives with­out be­ing that in­ter­est­ing. Any of these three would have some tales to tell of real-world use, on Sun­days at the small ra­tio­nal­ity dojo where they were in­struc­tors. What I’m try­ing to say here is: don’t de­mand that ev­ery­one be Robin Han­son in their se­cret iden­tity, that is set­ting the bar too high. Selec­tive re­port­ing makes it seem that fan­tas­ti­cally high-achiev­ing peo­ple have a far higher rel­a­tive fre­quency than their real oc­cur­rence. So if you ask for your ra­tio­nal­ity in­struc­tor to be as in­ter­est­ing as the sort of peo­ple you read about in the news­pa­pers—and a mas­ter ra­tio­nal­ist on top of that—and a good teacher on top of that—then you’re go­ing to have to join one of three fa­mous do­jos in New York, or some­thing. But you don’t want to be too lax and start re­spect­ing things that oth­ers wouldn’t re­spect if they weren’t spe­cially look­ing for rea­sons to praise the in­struc­tor. “Hav­ing a good se­cret iden­tity” should re­quire way more effort than any­thing that could be­come a to­ken re­quire­ment.

Now I put to you: If the in­struc­tors all have real-world anec­dotes to tell of us­ing their knowl­edge, and all of the stu­dents know that the de­sir­able ca­reer path can’t just be to be­come a ra­tio­nal­ity in­struc­tor, doesn’t that sound healthier?

Part of the se­quence The Craft and the Community

Next post: “Be­ware of Other-Op­ti­miz­ing

Pre­vi­ous post: “Whin­ing-Based Com­mu­ni­ties