Self-Improvement or Shiny Distraction: Why Less Wrong is anti-Instrumental Rationality


Less Wrong is ex­plic­itly in­tended is to help peo­ple be­come more ra­tio­nal. Eliezer has posted that ra­tio­nal­ity means epistemic ra­tio­nal­ity (hav­ing & up­dat­ing a cor­rect model of the world), and in­stru­men­tal ra­tio­nal­ity (the art of achiev­ing your goals effec­tively). Both are fun­da­men­tally tied to the real world and our perfor­mance in it—they are about abil­ity in prac­tice, not the­o­ret­i­cal knowl­edge (ex­cept inas­much as that knowl­edge helps abil­ity in prac­tice). Un­for­tu­nately, I think Less Wrong is a failure at in­still­ing abil­ities-in-prac­tice, and de­signed in a way that de­tracts from peo­ple’s real-world perfor­mance.

It will take some time, and it may be un­pleas­ant to hear, but I’m go­ing to try to ex­plain what LW is, why that’s bad, and sketch what a tool to ac­tu­ally help peo­ple be­come more ra­tio­nal would look like.

(This post was mo­ti­vated by Anna Salomon’s Hu­mans are not au­to­mat­i­cally strate­gic and the re­sponse, more de­tailed back­ground in foot­note [1].)

Up­date /​ Clar­ifi­ca­tion in re­sponse to some com­ments: This post is based on the as­sump­tion that a) the cre­ators of Less Wrong wish Less Wrong to re­sult in peo­ple be­com­ing bet­ter at achiev­ing their goals (in­stru­men­tal ra­tio­nal­ity, aka “effi­cient pro­duc­tivity”), and b) Some (per­haps many) read­ers read it to­wards that goal. It is this I think is self-de­cep­tion. I do not dis­pute that LW can be used in a pos­i­tive way (read dur­ing fun time in­stead of the NYT or funny pic­tures on Digg), or that it has pos­i­tive effects (ex­pos­ing peo­ple to im­por­tant ideas they might not see el­se­where). I merely dis­pute that read­ing fun things on the in­ter­net can help peo­ple be­come more in­stru­men­tally ra­tio­nal. Ad­di­tion­ally, I think in­stru­men­tal ra­tio­nal­ity is re­ally im­por­tant and could be a huge benefit to peo­ple’s lives (in fact, is by defi­ni­tion!), and so a com­mu­nity value that “de­liber­ate prac­tice to­wards self-im­prove­ment” is more valuable and more im­por­tant than “read­ing en­ter­tain­ing ideas on the in­ter­net” would be of im­mense value to LW as a com­mu­nity—while greatly de­creas­ing the im­por­tance of LW as a web­site.

Why Less Wrong is not an effec­tive route to in­creas­ing ra­tio­nal­ity.


Work: time spent act­ing in an in­stru­men­tally ra­tio­nal man­ner, ie forc­ing your at­ten­tion to­wards the tasks you have con­sciously de­ter­mined will be the most effec­tive at achiev­ing your con­sciously cho­sen goals, rather than al­low­ing your mind to drift to what is shiny and fun.

By defi­ni­tion, Work is what (in­stru­men­tal) ra­tio­nal­ists wish to do more of. A corol­lary is that Work is also what is re­quired in or­der to in­crease one’s ca­pac­ity to Work. This must be true by the defi­ni­tion of in­stru­men­tal ra­tio­nal­ity—if it’s the most effi­cient way to achieve one’s goals, and if one’s goal is to in­crease one’s in­stru­men­tal ra­tio­nal­ity, do­ing so is most effi­ciently done by be­ing in­stru­men­tally ra­tio­nal about it. [2]

That was al­most cir­cu­lar, so to add meat, you’ll no­tice in the defi­ni­tion an em­bed­ded as­sump­tion that the “hard” part of Work is di­rect­ing at­ten­tion—forc­ing your­self to do what you know you ought to in­stead of what is fun & easy. (And to a lesser de­gree, de­ter­min­ing your goals and the most effec­tive tasks to achieve them). This as­sump­tion may not hold true for ev­ery­one, but with the amount of dis­cus­sion of “Akra­sia” on LW, the gen­eral drift of writ­ing by smart peo­ple about pro­duc­tivity (Paul Gra­ham: Ad­dic­tion, Dis­trac­tion, Mer­lin Mann: Time & At­ten­tion), and the com­mon themes in the nu­mer­ous pro­duc­tivity/​self-help books I’ve read, I think it’s fair to say that iden­ti­fy­ing the goals and tasks that mat­ter and get­ting your­self to do them is what most hu­mans fun­da­men­tally strug­gle with when it comes to in­stru­men­tal ra­tio­nal­ity.

Figur­ing out goals is fairly per­sonal, of­ten sub­jec­tive, and can be difficult. I definitely think the deep philo­soph­i­cal el­e­ments of Less Wrong and it’s con­tri­bu­tions to epistemic ra­tio­nal­ity [3] are use­ful to this, but (like psychedelics) the benefit comes from small oc­ca­sional doses of the good stuff. Goals should be re-ex­am­ined reg­u­larly, but oc­ca­sion­ally (roughly yearly, and at ma­jor life forks). An an­nual re­treat with a mix of close friends and dis­tant-but-re­spected ac­quain­tances (Burn­ing Man, per­haps) will do the trick—read­ing a reg­u­larly up­dated blog is way overkill.

And figur­ing out tasks, once you turn your at­ten­tion to it, is pretty easy. Once you have ex­plicit goals, just con­sciously and con­tin­u­ously ex­am­in­ing whether your ac­tions have been effec­tive at achiev­ing those goals will get you way above the av­er­age smart hu­man at cor­rectly choos­ing the most effec­tive tasks. The big deal here for many (most?) of us, is the con­scious di­rec­tion of our at­ten­tion.

What is the en­emy of con­sciously di­rected at­ten­tion? It is shiny dis­trac­tion. And what is Less Wrong? It is a blog, a suc­ces­sion of short fun posts with com­ments, most likely read when peo­ple wish to dis­tract or en­ter­tain them­selves, and tuned for pro­duc­ing shiny ideas which suc­cess­fully dis­tract and en­ter­tain peo­ple. As Mer­lin Mann says: “Join­ing a Face­book group about cre­ative pro­duc­tivity is like buy­ing a chair about jog­ging”. Well, read­ing a blog to over­come akra­sia IS join­ing a Face­book group about cre­ative pro­duc­tivity. It’s the op­po­site of this clas­sic piece of ad­vice.

Now, I freely ad­mit that this ar­gu­ment is rel­a­tively brief and min­i­mally sup­ported com­pared to what a re­ally good, solid ar­gu­ment about ex­actly how to be­come more ra­tio­nal would be. This laz­i­ness is de­liber­ate, con­scious, and a di­rect ex­pres­sion of my be­liefs about the prob­lem with LW. I be­lieve that most peo­ple, par­tic­u­larly smart ones, do way too much think­ing & talk­ing and way too lit­tle ac­tion (me in­cluded), be­cause that is what’s easy for them [4].

What I see as a bet­ter route is to gather those who will quickly agree, do things differ­ently, (hope­fully) win and (definitely) learn. Note that this gen­eral tech­nique has a dou­ble ad­van­tage: the small group gets to en­joy im­me­di­ate re­sults, and when the time comes to change minds, they have the pow­er­ful ev­i­dence of their ex­pe­rience. It also re­duces the prob­lem that the stated goal of many par­ti­ci­pants (“get more ra­tio­nal”) may not be their ac­tual goal (“en­joy the com­pany of ra­tio­nal­ists in a way which is shiny fun, not Work”), since the call to ac­tion will tend to se­lect for those who ac­tu­ally de­sire self-im­prove­ment. My hope is that this post and the de­scrip­tion be­low of what ac­tual per­sonal growth looks like in­spire one or more small groups to form.

Less Wrong: Nega­tive Value, Pos­i­tive Potential

Un­for­tu­nately, in this frame­work, Less Wrong is prob­a­bly of nega­tive value to those who re­ally want to be­come more ra­tio­nal. I see it as a low-ROI ac­tivity whose shini­ness is tuned to at­tract the ra­tio­nal­ity com­mu­nity, and thus serves as the perfect dis­trac­tion (ra­tio­nal­ity porn, ra­tio­nal­ity opium). Many (most?) par­ti­ci­pants are al­low­ing LW to grab their at­ten­tion be­cause it is fun and easy, and thus si­mul­ta­neously dis­tract­ing them­selves from Work (re­duc­ing their over­all Work time) while con­vinc­ing them­selves that this dis­trac­tion is helping them to be­come more ra­tio­nal. This re­duces the chance that they will con­sciously Work to­wards ra­tio­nal­ity, since they feel they are already work­ing to­wards that goal with their LW read­ing time. (Ad­ding [4.5] in re­sponse to com­ments).

(Note that from this per­spec­tive, HP&TMoR is a pos­i­tive—peo­ple know read­ing fan­fic is en­ter­tain­ment, and be­ing good enough en­ter­tain­ment to dis­place peo­ple’s less ed­u­ca­tional al­ter­na­tive en­ter­tain­ments while teach­ing a lit­tle ra­tio­nal­ity in­creases the over­all level of ra­tio­nal­ity. The key is that HP&TMoR is read in “fun time”, while I be­lieve most peo­ple see LW time as “work to­wards self-im­prove­ment” time. Ironic, but true for me and the friends I’ve pol­led, at least)

That said, the prop­erty of shini­ness-to-ra­tio­nal­ists has re­sulted in a large com­mu­nity of ra­tio­nal­ists, which makes LW po­ten­tially a great re­source for ac­tual train­ing of peo­ple’s in­di­vi­d­ual ra­tio­nal­ity. And while cat­alyz­ing Work is much harder than get­ting pos­i­tive feed­back, I do find it heart-warm­ing and promis­ing that I have con­sis­tently re­ceived pos­i­tive feed­back from the LW com­mu­nity by point­ing out it’s er­rors. This is a com­mu­nity that wants to self-cor­rect—which is un­for­tu­nately rare and a nec­es­sary though not suffi­cient crite­ria for im­prove­ment.

This is tak­ing too long to write [5], and we haven’t even got­ten to the con­struc­tive part, so I’m go­ing to as­sume that if you are still with me you no longer need as de­tailed ar­gu­ments and I can go faster.

Some Ob­ser­va­tions On What Makes Some­thing Use­ful For Self-Improvement

My ver­sion: Growth ac­tivi­ties are Work, and hence feel like work, not fun—they in­volves over­rid­ing your in­stincts, not fol­low­ing them. Any growth you can get from fol­low­ing your in­stincts, you have prob­a­bly had already. And con­sciously di­rect­ing your at­ten­tion is not some­thing that can be trained by be­ing dis­tracted (willpower is a mus­cle, you ex­er­cise it by us­ing it). Find­ing the best tasks to achieve your goals is not prac­ticed by do­ing what­ever tasks come to mind. And so forth. You may ex­pe­rience flow states once your at­ten­tion is fo­cused where it should be, but un­less you have the in­cred­ible and rare for­tune to have what is shiny match up with what is use­ful, the act of start­ing and main­tain­ing fo­cus and im­prov­ing your abil­ity to do so will be hard work.

The aca­demic ver­sion: The liter­a­ture on skill de­vel­op­ment (“ac­qui­si­tion of ex­per­tise”) says that it in­volves “de­liber­ate prac­tice”. The same is very likely true of ac­quiring ex­per­tise in ra­tio­nal­ity. The 6 tenets of de­liber­ate prac­tice are that it:

  1. Is not in­her­ently en­joy­able.

  2. Is not play or paid prac­tice.

  3. Is rele­vant to the skill be­ing de­vel­oped.

  4. Is not sim­ply watch­ing the skill be­ing performed.

  5. Re­quires effort and at­ten­tion from the learner.

  6. Often in­volves ac­tivi­ties se­lected by a coach or teacher to fa­cil­i­tate learn­ing.

One must stretch quite a bit to fit these to “read­ing Less Wrong”—it’s just too shiny and fun to be use­ful. One can (and must) en­joy the re­sults of prac­tice, but if the prac­tice it­self doesn’t take effort, you are go­ing to plateau fast. (I want to be clear, BTW, that I am not mak­ing a Pu­ri­tan fal­lacy of equat­ing effort and re­ward [6]). Med­i­ta­tion is a great ex­am­ple of an in­stru­men­tal ra­tio­nal­ity prac­tice: it is a bor­ing, difficult iso­la­tion ex­er­cise for di­rect­ing and notic­ing the di­rec­tion of one’s at­ten­tion. It is Work.

What Would A Real Ra­tion­al­ity Prac­tice Look Like?

Eliezer has used the phrase “ra­tio­nal­ity dojo”, which I think has many cor­rect im­pli­ca­tions:

  1. It is a group of peo­ple who gather in per­son to train spe­cific skills.

  2. While there are some the­o­ret­i­ci­ans of the art, most peo­ple par­ti­ci­pate by learn­ing it and do­ing it, not the­o­riz­ing about it.

  3. Thus the main fo­cus is on lo­cal prac­tice groups, along with the global co­or­di­na­tion to max­i­mize their effec­tive­ness (mar­ket­ing, brand­ing, in­te­gra­tion of knowl­edge, com­mon in­fras­truc­ture). As a re­sult, it is driven by the needs of the learn­ers.

  4. You have to sweat, but the re­sult is you get stronger.

  5. You im­prove by learn­ing from those bet­ter than you, com­pet­ing with those at your level, and teach­ing those be­low you.

  6. It is run by a pro­fes­sional, or at least some­one get­ting paid for their hobby. The prac­ti­cants re­ceive per­sonal benefit from their prac­tice, in par­tic­u­lar from the value-added of the coach, enough to pay for tal­ented coaches.

In gen­eral, a real ra­tio­nal­ity prac­tice should feel a lot more like go­ing to the gym, and a lot less like hang­ing out with friends at a bar.

To ex­plain the ones that I worry will be non-ob­vi­ous:

1) I don’t know why in-per­son group is im­por­tant, but it seems to be—all the peo­ple who have replied to me so far say­ing they get use­ful ra­tio­nal prac­tice out of the LW com­mu­nity said the growth came through at­tend­ing lo­cal mee­tups (ex­am­ple). We can eas­ily in­vent some evolu­tion­ary psy­chol­ogy story for this, but it doesn’t mat­ter why, at this point it’s enough to just know.

6) There are peo­ple who can do high-qual­ity pro­duc­tive work in their spare time, but in my ex­pe­rience they are very rare. It is very pleas­ant to think that “am­a­teurs can change the world” be­cause then we can fan­ta­size about our­selves do­ing it in our spare time, and it even hap­pens oc­ca­sion­ally, which feeds that fan­tasy, but I don’t find it very cred­ible. I know we are re­ally smart and there are memes in our com­mu­nity that ra­tio­nal­ists are way bet­ter than ev­ery­one else at ev­ery­thing, but frankly I find the idea that peo­ple writ­ing blog posts in their spare time will cre­ate a bet­ter sys­tem than trained pro­fes­sion­als for im­prov­ing one’s abil­ity to achieve one’s goals to be lu­dicrous. I know some per­sonal growth pro­fes­sion­als, and they are smart too, and they have had years of prac­tice and study to de­velop prac­ti­cal ex­pe­rience. Talk is cheap, as is time spent read­ing blogs: if peo­ple ac­tu­ally value be­com­ing more ra­tio­nal, they will pay for it, and if there are good teach­ers, they will be worth be­ing paid. Money is a unit of learn­ing [7].

There are some other im­por­tant as­pects which such a prac­tice would have that LW does not:

  1. The ac­cu­mu­la­tion of knowl­edge. Blogs are in­her­ently re­ward­ing: peo­ple read what is re­cent, so you get quick feed­back on posts and com­ments. How­ever, they are in­her­ently ephemeral for the same rea­son—peo­ple read what is re­cent, and posts are never sub­stan­tially re­vised. The vot­ing sys­tem helps a lit­tle, but it can’t even close to fix the un­der­ly­ing struc­ture. To be effi­cient, much less work should go into ephemeral posts, and much more into ac­cu­mu­lat­ing and re­vis­ing a large, de­tailed, nu­anced body of knowl­edge (this is ex­actly the sort of “”work not fun” ac­tivity that you can get by pay­ing some­one, but are un­likely to get when con­trib­u­tors are vol­un­teers). In the­ory, this could hap­pen on the Wiki, but in prac­tice I have rarely seen Wikis suc­ceed at this (with the ob­vi­ous ex­cept of Le Wik).

  2. It would in­volve more liter­a­ture re­view and poin­t­ers to ex­ist­ing work. The ob­vi­ous high­est-ROI way to start work­ing on im­prov­ing in­stru­men­tal ra­tio­nal­ity is to re­search and sum­ma­rize the best ex­ist­ing work for self-im­prove­ment in the di­rec­tions that LW val­ues, not to rein­vent the wheel. Yes, over time LW should pro­duce origi­nal work and per­haps even­tu­ally the best such work, but the ex­ist­ing work is not so bad that it should just be ig­nored. Far from it! In refer­ence to (1), per­haps this should be done by cre­at­ing a database of re­views and rat­ings by LWers of the top-rated self-im­prove­ment books, per­haps even­tu­ally with rat­ings tak­ing into ac­count the va­ri­ety of skills peo­ple are seek­ing and ways in which they op­ti­mally learn.

  3. It would be prac­ti­cal—most units of in­for­ma­tion (posts, pages, what­ever) would be about ex­er­cises or ideas that one could im­me­di­ately ap­ply in one’s own life. It would look less like most LW posts (ab­stract, the­o­ret­i­cal, fo­cused on chains of logic), and more like Struc­tured Pro­cras­ti­na­tion, the Pmarca Guide To Per­sonal Pro­duc­tivity, books like Eat That Frog!, Get­ting Things Done, and Switch [8]. Most dis­cus­sion would be about top­ics like those in Anna’s post—how to act effec­tively, what things peo­ple have tried, what worked, what didn’t, and why. More learn­ing through em­piri­cism, less through logic and anal­y­sis.

In form­ing such a prac­tice, we could learn from other com­mu­ni­ties which have de­vel­oped a new body of knowl­edge about a set of skills and dis­sem­i­nated it with rapid scal­ing within the last 15 years. Two I know about and have tan­gen­tially par­ti­ci­pated in are

  1. PUA (how to pick up women). In fact, a so­cial skills com­mu­nity based on PUA was sug­gested on LW a few days ago - (glad to see that oth­ers are in­ter­ested in prac­tice and not just talk!)

  2. CrossFit (syn­the­sis of the best tech­niques for time-effi­cient broad-ap­pli­ca­bil­ity fit­ness)

Note that both in­volve most of my sug­gested fea­tures (PUA has some “read­ing not do­ing” is­sues, but it’s far ahead of LW in hav­ing an ex­plicit cul­tural value to the con­trary—for ex­am­ple, al­most ev­ery work­shop fea­tures time spent “in the field”). One fea­ture of PUA in par­tic­u­lar I’d like to point out is the con­cept of the “PUA lair”—a group of peo­ple liv­ing to­gether with the ex­plicit in­ten­tion of in­creas­ing their PUA skills. As the lair link says: “It is highly touted that the most profi­cient and fastest way to im­prove your skills is to hang out with oth­ers who are ahead of you, and those whose goals for im­prove­ment mir­ror your own.” [9]


If LW is to ac­com­plish it’s goal of in­creas­ing par­ti­ci­pant’s in­stru­men­tal ra­tio­nal­ity, it must dra­mat­i­cally change form. One of the biggest, per­haps the biggest el­e­ment of in­stru­men­tal ra­tio­nal­ity is the abil­ity to di­rect one’s at­ten­tion, and a ra­tio­nal­ity blog makes peo­ple worse at this by dis­tract­ing their at­ten­tion in a way ac­cepted by their com­mu­nity and that they will feel is use­ful. From The War Of Art [10]:

Often cou­ples or close friends,even en­tire fam­i­lies, will en­ter into tacit com­pacts whereby each in­di­vi­d­ual pledges (un­con­sciously) to re­main mired in the same slough in which she and all her cronies have be­come so com­fortable. The high­est trea­son a crab can com­mit is to make a leap for the rim of the bucket.

To aid growth at ra­tio­nal­ity, Less Wrong would have to be­come a skill prac­tice com­mu­nity, more like mar­tial arts, PUA, and phys­i­cal fit­ness, with an ex­plicit fo­cus of helping peo­ple grow in their abil­ity to set and achieve goals, com­bin­ing lo­cal chap­ters with global co­or­di­na­tion, in­fras­truc­ture, and knowl­edge ac­cu­mu­la­tion. Most dis­cus­sion should be among peo­ple work­ing on a spe­cific skill at a similar level about what is or isn’t work­ing for them as they at­tempt to progress, rather than ob­scure the­o­ries about the in­ner work­ings of the hu­man mind.

Such a prac­tice and com­mu­nity would look very differ­ent, but I be­lieve it would have a far bet­ter chance to ac­tu­ally make peo­ple more ra­tio­nal [11]. There would be dan­ger of cultism and the re­li­gious fer­vor/​”one true way” that self-help move­ments some­times have (Land­mark), and I won­der if it’s a profound dis­taste for any­thing re­motely smelling of cult that has led Eliezer & SIAI away from this path. But the op­po­site of cult is not growth, it is to con­tinue be­ing an opi­ate for ra­tio­nal­ists, a pleas­ant way of mak­ing the time pass that feels like work to­wards growth and thus feeds peo­ple’s de­sire for guiltless dis­trac­tion.

To be growth, we must do work, peo­ple must get paid, we must gather in per­son, fo­cus on ac­tion not words, put forth great effort over time to in­crease our ca­pac­ity, use peak ex­pe­riences to knock peo­ple loose from in­grained pat­terns, and copy these and much more from the skill prac­tice com­mu­ni­ties of the world. Devel­oped by non-ra­tio­nal­ists, sure, but the ones that last are the ones that work [12] - let’s learn from their em­bed­ded knowl­edge.


That was 5 hours of my semi-Work time, so I re­ally hope it wasn’t wasted, and that some of you not only listen but take ac­tion. I don’t have much free time for new pro­jects, but if peo­ple want to start a lo­cal ra­tio­nal­ity dojo in Moun­tain View/​Sun­ny­vale, I’m in. And there is already talk, among some re­view­ers of this draft, of putting to­gether an in­tro­duc­tory work­shop. Time will tell—and the next step is up to you.


[1] Anna Salomon posted Hu­mans are not au­to­mat­i­cally strate­gic, a re­ply to the very prac­ti­cal A “Failure to Eval­u­ate Re­turn-on-Time” Fal­lacy. Anna’s post laid out a nice rough map at what an in­stru­men­tally ra­tio­nal pro­cess for goal achieve­ment would look like (con­sciously choos­ing goals, met­rics, re­search­ing solu­tions, ex­per­i­ment­ing with im­ple­ment­ing them, bal­anc­ing ex­plo­ra­tion & ex­ploita­tion—the ba­sic recipe for suc­cess at any­thing), said she was keen to train this, and asked:

So, to sec­ond Lion­hearted’s ques­tions: does this anal­y­sis seem right? Have some of you trained your­selves to be sub­stan­tially more strate­gic, or goal-achiev­ing, than you started out? How did you do it? Do you agree with (a)-(h) above? Do you have some good heuris­tics to add? Do you have some good ideas for how to train your­self in such heuris­tics?

After read­ing the com­ments, I made a com­ment which be­gan:

I’m dis­ap­pointed at how few of these com­ments, par­tic­u­larly the highly-voted ones, are about pro­posed solu­tions, or at least pro­posed ar­eas for re­search. My gen­eral con­cern about the LW com­mu­nity is that it seems much more in­ter­ested in the fun of de­bat­ing and an­a­lyz­ing bi­ases, rather than the bor­ing repet­i­tive trial-and-er­ror of cor­rect­ing them.

Anna’s post was up­voted into the top 10 all-time on LW in a cou­ple days, and my com­ment quickly be­came the top on the post by a large mar­gin, so both her agenda and my con­cern seem to be widely shared. While I rarely take the time to write LW posts (as you would ex­pect from some­one who be­lieves LW is not very use­ful), this feed­back gave me hope that there might be enough un­tapped de­sire for some­thing more effec­tive that a post might help cat­alyze enough change to be worth­while.

[2] There are many other other ar­gu­ments as to why im­prov­ing one’s abil­ity to do work is un­likely to be fun and easy, of course. With a large space of pos­si­ble ac­tivi­ties, and only a loose con­nec­tion be­tween “fun” and “helps you grow” (via evolu­tion­ary biol­ogy), it seems a pri­ori un­likely that fun ac­tivi­ties will over­lap with growth­ful ones. And we know that a gen­eral recipe for get­ting bet­ter at X is to do X, so if one wants to get bet­ter at di­rect­ing one’s at­ten­tion to the most im­por­tant tasks and goals, it seems very likely that one must prac­tice di­rect­ing one’s at­ten­tion. Fur­ther­more, there is ev­i­dence that, speci­fi­cally, willpower is a mus­cle. So the case for grow­ing one’s in­stru­men­tal ra­tio­nal­ity through be­ing dis­tracted by an en­ter­tain­ing ra­tio­nal­ity blog is...awfully weak.

[3] What are the most im­por­tant prob­lems in the world? Who is work­ing most effec­tively to fix them and how can you help? Un­der­stand­ing ex­is­ten­tial risks is cer­tainly not easy, and im­por­tant to set­ting that por­tion of your goals that has to do with helping the world—which is a minor part of most peo­ple’s goals, which are about their own lives and self-in­ter­est.

[4] I also be­lieve the least effec­tive form of de­bate is try­ing to get peo­ple to change their minds. There­fore, an ex­ten­sive study and doc­u­men­ta­tion to cre­ate a re­ally good, solid ar­gu­ment try­ing to change the minds of LWers who don’t quickly agree with my ar­gu­ment sketch would be a very low-re­turn ac­tivity com­pared to get­ting to­gether those who already agree and do­ing an ex­per­i­ment. And in­stru­men­tal ra­tio­nal­ity is about max­i­miz­ing the re­turn on your ac­tivi­ties, given your goals, so I try to avoid low-re­turn ac­tivi­ties.

[4.5] A num­ber of com­menters state that they con­sciously read LW dur­ing fun time, or read it to learn about bi­ases and ex­is­ten­tial risk, not to be­come more ra­tio­nal, in which case it is likely of pos­i­tive value. If you have suc­cess­fully walled off your work from shiny dis­trac­tions, then you are ad­vanced in the ways of at­ten­tion and may be able to use this par­tic­u­lar drug with­out nega­tive effects, and I con­grat­u­late you. If you are read­ing it to learn about top­ics of in­ter­est to ra­tio­nal­ists and be­lieve that you will stop there and not let it af­fect your pro­duc­tivity, just be warned that many an opi­ate ad­dic­tion has be­gun with a le­gi­t­i­mate use of painkil­lers.

Or to go back to Mer­lin’s metaphor: If you buy a couch to sit on and watch TV, there’s noth­ing wrong with that. You might even see a sports pro­gram on TV that mo­ti­vates you to go jog­ging. Just don’t buy the couch in or­der to fur­ther your goal of phys­i­cal fit­ness. Or claim that couch-buy­ers are a com­mu­nity of peo­ple com­mit­ted to be­com­ing more fit, be­cause they some­times watch sports shows and some­times get out­side. Couch-buy­ers are a com­mu­nity of peo­ple who sit around—even if they watch sports pro­grams. Real run­ners buy jog­ging shoes, sweat head­bands, GPS route track­ers, pe­dome­ters, stop­watches...

[5] 1.5 hrs so far. Time track­ing is an im­por­tant part of at­ten­tion man­age­ment—if you don’t know how your time is spent, it’s prob­a­bly be­ing spent badly.

[6] Speci­fi­cally, I am not say­ing that growth is never fun, or that growth is pro­por­tional to effort, only that there are a very limited num­ber of fun ways to grow (tak­ing psychedelics at Burn­ing Man with peo­ple you like and re­spect) and you’ve prob­a­bly done them all already. If you haven’t, sure, of course you should do them, and yes, of course dis­cov­er­ing & cat­a­loging such things is use­ful, but there re­ally aren’t very many so if you want to con­tinue to grow you need to stop fool­ing your­self that read­ing a blog will do it and get ready to make some effort.

[7] Referenc­ing Eliezer’s great Money: The Unit of Car­ing, of course. I find it ironic that he un­der­stand ba­sic eco­nomics in­tel­lec­tu­ally so well as to make one of the most elo­quent ar­gu­ments for donat­ing money in­stead of time that I’ve ever seen, yet seems to be try­ing to cre­ate a ra­tio­nal­ity im­prove­ment move­ment with­out, as far as I can tell, in­volv­ing any spe­cial­ists in the art of hu­man change or growth. That is, us­ing the method that grownups use. What you do when you want some­thing to ac­tu­ally get done. You use money to em­ploy full-time spe­cial­ists.

[8] I haven’t ac­tu­ally read this one yet, but their other book, Made To Stick, was an out­stand­ing study of memetic en­g­ineer­ing so I think it very likely that their book on habit for­ma­tion is good too.

[9] In­deed. I hap­pen to have a back­ground of liv­ing in and found­ing in­ten­tional com­mu­ni­ties (Tor­tuga!), and in fact cur­rently rent rooms to LWers Divia and Nick Tar­leton, so I can at­test to the value of one’s so­cial en­vi­ron­ment and per­sonal growth goals be­ing syn­chro­nized. Ben­ton House is likely an ex­am­ple as well. Groups of ra­tio­nal­ists liv­ing to­gether will au­to­mat­i­cally prac­tice, and have that prac­tice re­in­forced by their pri­mate de­sire for sta­tus within the group, this is al­most surely the fastest way to progress, al­though not re­quired or suited to ev­ery­one.

[10] The next para­graph ex­plains why I do my best not to spend much time here:

The awak­en­ing artist must be ruth­less, not only with her­self but with oth­ers. Once you make your break, you can’t turn around for your buddy who catches his trouser leg on the barbed wire. The best thing you can do for that friend (and he’d tell you this him­self, if he re­ally is your friend) is to get over the wall and keep mo­tat­ing.

Although I sup­pose I am vi­o­lat­ing the ad­vice by turn­ing around and giv­ing a long speech about why ev­ery­one else should make a break too :). My the­ory is that by say­ing it right once, I can re­frain from wast­ing any more time say­ing it again in the fu­ture, should this at­tempt not work. But that may just be ra­tio­nal­iz­ing. On the other hand, do­ing things “well or not at all” is ra­tio­nal in situ­a­tions where the re­turn curve is steep. Given my low eval­u­a­tion of LW’s use­ful­ness, I ob­vi­ously think the early part of the re­turn curve is ba­si­cally flat zero. We will see if it is hubris to think the right post can re­ally make a differ­ence, and that I can make that post. Cer­tainly plenty of op­por­tu­nity for bias in both those state­ments.

[11] Note that helping peo­ple be­come per­son­ally more effec­tive is a much eas­ier meme to spread than helping peo­ple bet­ter un­der­stand how to con­tribute to pub­lic goods (ie how to bet­ter un­der­stand effi­cient char­ity and ex­is­ten­tial risk). They have ev­ery in­cen­tive to do the former and lit­tle in­cen­tive to do the lat­ter. So train­ing peo­ple in gen­eral goal achieve­ment (in­stru­men­tal ra­tio­nal­ity) is likely to have far broader ap­peal and reach far more peo­ple than train­ing them in the as­pects of epistemic ra­tio­nal­ity that SIAI is most in­ter­ested in. This large com­mu­nity who have grown through the in­di­vi­d­u­ally benefi­cial part of the philos­o­phy is then a great tar­get mar­ket for the so­cietally benefi­cial part of the philos­o­phy. (A clas­sic one-two punch used by spiritual groups, of course: provide value then teach val­ues. It works. If ra­tio­nal­ists do what works...) I’ve been mean­ing to make a post on the im­por­tance of per­sonal benefit to spread­ing memes for awhile, this para­graph will have to do for now...

[12] And the ones with good memetic en­g­ineer­ing, in­clud­ing use of the Dark Arts. Many difficult de­ci­sions will need to be made about what tech­niques are and aren’t Dark Arts and which are worth us­ing any­way. The fact re­mains that just like a sports MVP is al­most cer­tainly both more skil­led and more lucky than his peers, a suc­cess­ful self-help move­ment is al­most cer­tainly both more effec­tive at helping peo­ple and bet­ter memet­i­cally en­g­ineered than its peers. So copy—but filter.