moridinamael

• moridinamael 24 May 2019 3:41 UTC

  2 points

  in reply to: Hazard's comment on: What is your per­sonal ex­pe­rience with “hav­ing a mean­ingful life”?

  To point one, if I feel an ex­cite­ment and ea­ger­ness about the thing, and if I ex­pect I would feel sad if the thing were sud­denly taken away, then I can be pretty sure that it’s im­por­tant to me. But — and this re­lates to point two — it’s hard to care about the same thing for weeks or months or years at a time with the same in­ten­sity. Some pro­jects of mine have os­cillated be­tween pro­vid­ing deep mean­ing and be­ing a ma­jor drag, de­pend­ing on con­tin­gent fac­tors. This might man­i­fest as a sense of ugh aris­ing around cer­tain facets of the ac­tivity. Usu­ally the ugh goes away even­tu­ally. Some­times it doesn’t, and you ei­ther ac­cept that the un­pleas­ant­ness is part and par­cel with the fun, or you de­cide it’s not worth it.

• moridinamael 23 May 2019 13:35 UTC

  19 points

  on: What is your per­sonal ex­pe­rience with “hav­ing a mean­ingful life”?

  As far as I can tell, mean­ing is a feel­ing, some­thing like a pas­sive sense that you’re on the right track. The feel­ing is gen­er­ated when you are work­ing on some­thing that you per­son­ally en­joy and care about, and when you are so­cial­iz­ing suffi­ciently of­ten with peo­ple you en­joy and care about. “Friends and hob­bies are the mean­ing of life” is how I might phrase it.

  Note that the ac­tivity that you spend your time on could be col­lect­ing all the stars in Mario64, as long as you ac­tu­ally care about com­plet­ing the task. How­ever, you tend to find it harder to care about things that don’t in­volve win­ning sta­tus or helping peo­ple, es­pe­cially as you get older.

  I think some peo­ple get them­selves into psy­cholog­i­cal trou­ble by de­cid­ing that all of the things that they en­joy aren’t “im­por­tant” and in­ter­act­ing with peo­ple they care about is a “dis­trac­tion”. They paint them­selves into a cor­ner where the only thing they al­low them­selves to con­sider do­ing is some­thing for which they feel no emo­tional at­trac­tion. They feel like they should en­joy it be­cause they’ve de­cided it’s im­por­tant, but they don’t, and then they feel guilty about that. The solu­tion to this is to rec­og­nize the kind of an­i­mal you are and try to feed the needs that you have rather than the ones you wish you had.

• moridinamael 14 May 2019 17:03 UTC

  2 points

  in reply to: Qiaochu_Yuan's comment on: The Village and The Mission

  I’m in­ter­ested as well. as some­one try­ing to grow the Den­ver ra­tio­nal­ity com­mu­nity, I want to be aware of failure modes.

Com­plex Be­hav­ior from Sim­ple (Sub)Agents

• moridinamael 23 Apr 2019 15:11 UTC

  4 points

  on: AI Align­ment Prob­lem: “Hu­man Values” don’t Ac­tu­ally Exist

  The idea of AI al­ign­ment is based on the idea that there is a finite, sta­ble set of data about a per­son, which could be used to pre­dict one’s choices, and which is ac­tu­ally morally good. The rea­son­ing be­hind this ba­sis is be­cause if it is not true, then learn­ing is im­pos­si­ble, use­less, or will not con­verge.

  Is it true that these as­sump­tions are re­quired for AI al­ign­ment?

  I don’t think it would be im­pos­si­ble to build an AI that is suffi­ciently al­igned to know that, at pretty much any given mo­ment, I don’t want to be spon­ta­neously in­jured, or be ac­cused of do­ing some­thing that will re­li­ably cause all my peers to hate me, or for a loved one to die. There’s quite a broad list of “easy” spe­cific “al­ign­ment ques­tions”, that vir­tu­ally 100% of hu­mans will agree on in vir­tu­ally 100% of cir­cum­stances. We could do worse than just build­ing the par­tially-al­igned AI who just makes sure we avoid fates worse than death, in­di­vi­d­u­ally and col­lec­tively.

  On the other hand, I agree com­pletely that cou­pling the con­cepts of “AI al­ign­ment” and “op­ti­miza­tion” seems pretty fraught. I’ve won­dered if the “op­ti­mal” en­vi­ron­ment for the hu­man an­i­mal might be a re-cre­ation of the Pleis­tocene, ex­cept with, y’know, im­mor­tal­ity, and care­fully man­aged, ex­cit­ing-but-not-har­row­ing lev­els of re­source scarcity.

• moridinamael 18 Apr 2019 17:34 UTC

  2 points

  in reply to: Said Achmiz's comment on: Epi­sode 1 of “Tsuyoku Nar­i­tai!” (the ‘be­com­ing stronger’ pod­cast/​YT se­ries).

  You may already know this, but al­most all YouTube videos will have an au­to­mat­i­cally gen­er­ated tran­script. Click ”...” to the bot­tom right of the video panel and click “Open tran­script” on the pul­l­down. YouTube’s au­to­matic speech tran­scrip­tion is very good.

• moridinamael 18 Apr 2019 14:29 UTC

  2 points

  on: Epi­sode 1 of “Tsuyoku Nar­i­tai!” (the ‘be­com­ing stronger’ pod­cast/​YT se­ries).

  This ex­ceeded my ex­pec­ta­tions. You kept it short and to the point, and the de­scrip­tion of the tech­nique was very clear. I look for­ward to more epi­sodes.

• moridinamael 26 Mar 2019 16:40 UTC

  5 points

  on: Subagents, akra­sia, and co­her­ence in humans

  Have you—or any­one, re­ally—put much thought into the im­pli­ca­tions of these ideas to AI al­ign­ment?

  If it’s true that mod­el­ing hu­mans at the level of con­sti­tu­tive sub­agents ren­ders a more ac­cu­rate de­scrip­tion of hu­man be­hav­ior, then any true solu­tion to the al­ign­ment prob­lem will need to re­spect this in­ter­nal in­co­her­ence in hu­mans.

  This is po­ten­tially a very pos­i­tive de­vel­op­ment, I think, be­cause it sug­gests that a hu­man can be mod­eled as a col­lec­tion of rel­a­tively sim­ple sub­agent util­ity func­tions, which in­ter­act and com­pete in com­plex but pre­dictable ways. This sounds closer to a gears-level por­trayal of what is hap­pen­ing in­side a hu­man, in con­trast to de­scrip­tions of hu­mans as hav­ing a sin­gle con­voluted and im­pos­si­ble-to-pin-down util­ity func­tion.

  I don’t know if you’re at all fa­mil­iar with Mark Lipp­man’s Fold­ing ma­te­rial and his on­tol­ogy for men­tal phe­nomenol­ogy. My at­tempt to sum­ma­rize his frame­work of men­tal phe­nom­ena is as fol­lows: there are be­lief-like ob­jects (ex­pec­ta­tions, tacit or ex­plicit, com­plex or sim­ple), goal-like ob­jects (de­sir­able states or set­tings or con­texts), af­for­dances (con­text-ac­ti­vated rep­re­sen­ta­tions of the cur­rent po­ten­tial ac­tion space) and in­ten­tion-like ob­jects (plans co­or­di­nat­ing im­me­di­ate felt in­ten­tions, via af­for­dances, to­ward goal-states). All cog­ni­tion is “gen­er­ated” by the ac­tions and in­ter­ac­tions of these fun­da­men­tal units, which I in­fer must be some­thing like neu­rolog­i­cally fun­da­men­tal. Fish and maybe even worms prob­a­bly have some­thing like be­liefs, goals, af­for­dances and in­ten­tions. Ours are just big­ger, more lay­ered, more nested and more in­ter­con­nected.

  The rea­son I bring this up is that Fold­ing was a bit of a kick in the head to my view on sub­agents. In­stead of see­ing sub­agents as be­ing fun­da­men­tal, I now see sub­agents as ex­pres­sions of la­tent goal-like and be­lief-like ob­jects, and the brain is im­ple­ment­ing some kind of pas­sive pro­gram that pur­sues goals and avoids ex­pec­ta­tions of suffer­ing, even if you’re not aware you have these goals or these ex­pec­ta­tions. In other words, the sense of there be­ing a sub­agent is your brain run­ning a back­ground pro­gram that ac­ti­vates and acts upon the im­pli­ca­tions of these more fun­da­men­tal yet hid­den goals/​be­liefs.

  None of this is at all in con­tra­dic­tion to any­thing in your Se­quence. It’s more like a slightly differ­ent fram­ing, where a “Pro­tec­tor Subagent” is re­duced to an ex­pres­sion of a be­lief-like ob­ject via a self-pro­tec­tive back­ground pro­cess. It all adds up to the same thing, pretty much, but it might be more gears-level. Or maybe not.

• moridinamael 4 Mar 2019 22:05 UTC

  5 points

  in reply to: algekalipso's comment on: Subagents, in­tro­spec­tive aware­ness, and blending

  Could you elab­o­rate on how you’re us­ing the word “sym­met­ri­cal” here?

• moridinamael 29 Jan 2019 19:39 UTC

  2 points

  in reply to: ryan_b's comment on: “AlphaS­tar: Mas­ter­ing the Real-Time Strat­egy Game StarCraft II”, Deep­Mind [won 10 of 11 games against hu­man pros]

  The best I can do af­ter think­ing about it for a bit is com­pute ev­ery pos­si­ble com­bi­na­tion of units un­der 200 sup­ply, mul­ti­ply that by the pos­si­ble po­si­tions of those units in space, mul­ti­ply that by the pos­si­ble com­bi­na­tions of build­ings on the map and their po­ten­tial lo­ca­tions in space, mul­ti­ply that by the pos­si­ble com­bi­na­tions of up­grades, mul­ti­ply that by the amount of re­sources in all available min­eral/​vespene sources … I can already spot a few over­sim­plifi­ca­tions in what I just wrote, and I can think of even more things that need to be ac­counted for. The shields/​hit­points/​en­ergy of ev­ery unit. Com­bi­na­to­ri­ally gi­gan­tic.

  Just the num­ber of po­ten­tial po­si­tions of a sin­gle unit on the map is already huge.

  But AlphaS­tar doesn’t re­ally ex­plore much of this space. It finds out pretty quickly that there’s re­ally no rea­son to ex­plore the parts of the space the in­clude build­ing ran­dom build­ings in weird map lo­ca­tions. It ex­plores and op­ti­mizes around the parts of the state space that look rea­son­ably close to hu­man play, be­cause that was its start­ing point, and it’s not go­ing to find su­pe­rior strate­gies ran­domly, not with­out a lot of op­ti­miza­tion in iso­la­tion.

  That’s one thing I would love to see, ac­tu­ally. A ver­sion of the code trained purely on self-play, with­out a ba­sis in hu­man re­plays. Does it ever dis­cover proxy plays or other es­o­teric cheese with­out a start­ing point pro­vided in the hu­man re­plays?

• moridinamael 25 Jan 2019 16:12 UTC

  38 points

  in reply to: james_t's comment on: “AlphaS­tar: Mas­ter­ing the Real-Time Strat­egy Game StarCraft II”, Deep­Mind [won 10 of 11 games against hu­man pros]

  Be­fore now, it wasn’t im­me­di­ately ob­vi­ous that SC2 is a game that can be played su­per­hu­manly well with­out any­thing that looks like long-term plan­ning or coun­ter­fac­tual rea­son­ing. The way hu­mans play it re­lies on a com­bi­na­tion of past ex­pe­rience, nar­row skills, and “what-if” men­tal simu­la­tion of the op­po­nent. Build­ing a su­per­hu­man SC2 agent out of noth­ing more than LSTM units in­di­cates that you can com­pletely do away with plan­ning, even when the ac­tion space is very large, even when the state space is VERY large, even when the pos­si­bil­ities are com­bi­na­to­ri­ally enor­mous. Yes, hu­mans can get good at SC2 with much less than 200 years of time played (al­though those hu­mans are usu­ally study­ing the re­plays of other mas­ters to boot­strap their un­der­stand­ing) but I think it’s worth­while to fo­cus on the in­verse of this ob­ser­va­tion: that a so­phis­ti­cated prob­lem do­main which looks like it ought to re­quire plan­ning and model-based coun­ter­fac­tual rea­son­ing ac­tu­ally re­quires no such thing. What other prob­lem do­mains seem like they ought to re­quire plan­ning and coun­ter­fac­tual rea­son­ing, but can prob­a­bly be con­quered with noth­ing more ad­vanced than a deep LSTM net­work?

  (I haven’t seen any­one bother to com­pute an es­ti­mate of the size of the state-space of SC2 rel­a­tive to, for ex­am­ple, Go or Chess, and I’m not sure if there’s even a co­her­ent way to go about it.)

• moridinamael 18 Jan 2019 18:26 UTC

  7 points

  in reply to: Kaj_Sotala's comment on: Book Sum­mary: Con­scious­ness and the Brain

  The free­dom to spec­u­late wildly is what makes this topic so fun.

  My men­tal model would say, you have a par­tic­u­lar pat­tern recog­ni­tion mod­ule that clas­sifies ob­jects as “chair”, along with a weight of how well the cur­rent in­stance matches the cat­e­gory. An ob­ject can be a pro­to­typ­i­cal perfect Pla­tonic chair, or an al­most-chair, or maybe a chair if you flip it over, or not a chair at all.

  When you look at a chair, this pat­tern recog­ni­tion mod­ule im­me­di­ately clas­sifies it, and then brings on­line an­other mod­ule, which makes available all the rele­vant phys­i­cal af­for­dances, lin­guis­tic and log­i­cal im­pli­ca­tions of a chair be­ing pre­sent in your en­vi­ron­ment. Rec­og­niz­ing some­thing as a chair feels iden­ti­cal to rec­og­niz­ing some­thing as a thing-in-which-I-can-sit. Similarly, you don’t have to puz­zle out the im­pli­ca­tions of a tiger walk­ing into the room right now. The fear re­sponse will co­in­cide with the recog­ni­tion of the tiger.

  When you try to in­tro­spect on chair­ness, what you’re do­ing is toss­ing imag­ined sense per­cepts at your­self and ob­serv­ing the re­sponses of the chari­ness de­tect­ing mod­ule. This al­lows you to gen­er­ate an ab­stract rep­re­sen­ta­tion of your own chair­ness clas­sifier. But this ab­stract rep­re­sen­ta­tion is ab­solutely not the same thing as the chair­ness clas­sifier, any more than your ab­stract cog­i­ta­tion about what the “+” op­er­a­tor does is the same thing as the men­tal op­er­a­tion of adding two num­bers to­gether.

  I think a lot of con­fu­sion about the na­ture of hu­man think­ing stems from the in­abil­ity to in­ter­nally dis­t­in­guish be­tween the ab­stracted sym­bol for a men­tal phe­nomenon and the men­tal phe­nomenon it­self. This dove­tails with IFS in an in­ter­est­ing way, in that it can be difficult to dis­t­in­guish be­tween think­ing about a par­tic­u­lar Part in the ab­stract, and ac­tu­ally get­ting into con­tact with that Part in a way that causes it to shift.

  I’m not sure why you say that the un­con­scious mod­ules com­mu­ni­cat­ing with each other would nec­es­sar­ily con­tra­dict the idea of us be­ing con­scious of ex­actly the stuff that’s in the workspace, but I tend to agree that con­sid­er­ing the con­tents of our con­scious­ness and the con­tents of the workspace to be strictly iso­mor­phic seems to be too strong.

  I may be sim­ply mi­s­un­der­stand­ing some­thing. My sense is that when you open the fridge to get a yo­gurt and your brain shouts “HOW DID CYPHER GET INTO THE MATRIX TO MEET SMITH WITHOUT SOMEONE TO HELP HIM PLUG IN?”, this is a kind of thought that arises from check­ing metic­u­lously over your epistemic state for log­i­cal in­con­sis­ten­cies, rather es­o­teric and com­plex log­i­cal in­con­sis­ten­cies, and it seems to come from nowhere. Doesn’t this im­ply that some sub­mod­ules of your brain are think­ing ab­stractly and log­i­cally about The Ma­trix com­pletely out­side of your con­scious aware­ness? If so, then this ei­ther im­plies that the sub­con­scious pro­cess­ing of in­di­vi­d­ual sub­mod­ules can be very com­plex and ab­stract with­out need­ing to share in­for­ma­tion with other sub­mod­ules, or that in­for­ma­tion shar­ing be­tween sub­mod­ules can oc­cur with­out you be­ing con­sciously aware of it.

  A third pos­si­bil­ity would be that you were ac­tu­ally con­sciously think­ing about The Ma­trix in a kind of inat­ten­tive, dis­tracted way, and it only seems like the thought came out of nowhere. This would be far from the most shock­ing ex­am­ple of the brain sim­ply ly­ing to you about its op­er­a­tions.

• moridinamael 17 Jan 2019 20:21 UTC

  7 points

  in reply to: Jan_Kulveit's comment on: Book Sum­mary: Con­scious­ness and the Brain

  The most ob­vi­ous ex­am­ple of this kind of thing is the “flash of in­sight” that we all ex­pe­rience from time to time, where a com­plex, multi-part solu­tion to a prob­lem in­trudes on our aware­ness as if from nowhere. This seems to be a clear case of the un­con­scious work­ing on this prob­lem in the back­ground, iden­ti­fy­ing its solu­tion as a valid one still in the back­ground, and in­ject­ing the fully-formed idea into aware­ness with high salience.

  It’s a bit like the phe­nomenon of be­ing able to pick out your own name from a bab­ble of crowded con­ver­sa­tion, ex­cept ap­plied to the un­con­scious ac­tivity of the mind. This, how­ever, im­plies that much com­plex in­ter-agent com­mu­ni­ca­tion and ab­stract prob­lem solv­ing is hap­pen­ing sub­con­sciously. And this seems to con­tra­dict the view that only very sim­ple con­cep­tual pack­ages are passed through to the Global Workspace, and that we must nec­es­sar­ily be con­scious of our own ab­stract prob­lem solv­ing.

  My own per­cep­tions dur­ing med­i­ta­tion (and dur­ing nor­mal life) would sug­gest that the sub­con­scious/​un­con­scious is do­ing very com­plex and ab­stract “think­ing” with­out my be­ing aware of its work­ings, and in­ter­mit­tently in­ject­ing bits and pieces of its ru­mi­na­tions into aware­ness based on some­thing like an ex­pec­ta­tion that the gestalt self might want to act on that in­for­ma­tion.

  This seems con­trary to the view that “what we are aware/​con­scious of” is iso­mor­phic to “the Global Workspace”. It seems that sub­con­scious mod­ules are chat­ter­ing away amongst them­selves al­most con­stantly, us­ing chan­nels that are ei­ther in­ac­cessible to con­scious­ness or severely muted.

• moridinamael 7 Jan 2019 19:41 UTC

  10 points

  on: Se­quence in­tro­duc­tion: non-agent and mul­ti­a­gent mod­els of mind

  I re­ally look for­ward to this Se­quence.

• moridinamael 4 Jan 2019 16:13 UTC

  6 points

  on: What is a rea­son­able out­side view for the fate of so­cial move­ments?

  I would sus­pect the failure of most so­cial move­ments is overde­ter­mined. So­cial move­ments by de­fault are de­signed to change the sta­tus quo, and the sta­tus quo tends to be sta­ble and in­trin­si­cally re­sis­tant to change. So­cial move­ments are of­ten ide­olog­i­cally origi­nated and may be aimed at achiev­ing some­thing prac­ti­cally im­pos­si­ble.

  Another phras­ing might be that most so­cial move­ments fail be­cause a sober anal­y­sis would have shown that there was never any re­al­is­tic pos­si­bil­ity for most so­cial move­ments to suc­ceed, even if they had more re­sources, smarter peo­ple and bet­ter plan­ning.

• moridinamael 1 Jan 2019 22:13 UTC

  11 points

  on: How can I most effec­tively im­prove my writ­ing abil­ities?

  I’ve im­proved most dra­mat­i­cally at writ­ing by get­ting very spe­cific feed­back from peo­ple who are clearly bet­ter than me. I con­sider my­self lucky to have had a small hand­ful of teach­ers and pro­fes­sors will­ing to put in the time to cri­tique my writ­ing on the sen­tence- and word-level.

  Re­cently I had a work of fic­tion of mine minutely cri­tiqued by a pro­fes­sional au­thor and ex­pe­rienced a similar sense of “lev­el­ing up”. For ex­am­ple, I’ve thought for years that I un­der­stood what “show don’t tell” means. But my gra­cious ed­i­tor in this case was able to point out mul­ti­ple in­stances in my story where I was “tel­ling” when I could be “show­ing”. Once he pointed these out, I un­der­stood on a deeper level what to pay at­ten­tion to.

  One in­ter­est­ing thing about get­ting feed­back on writ­ing is that some­one who is truly bet­ter than you can usu­ally provide sug­ges­tions that you im­me­di­ately rec­og­nize as cor­rect. You may think your writ­ing is fine, even great, but you’ll rec­og­nize true im­prove­ments as be­ing ob­vi­ously cor­rect and bet­ter, once pointed out. The pro­cess of be­com­ing bet­ter at writ­ing is the ac­cu­mu­la­tion of such in­di­vi­d­ual cor­rec­tions.

• moridinamael 13 Dec 2018 17:06 UTC

  15 points

  on: The Bat and Ball Prob­lem Revisited

  My daugh­ter is just start­ing to learn sub­trac­tion. She was very frus­trated by it, and if I ver­bally asked “What’s seven minus five?” she was about 50% likely to give the right an­swer. I asked her a se­quence of sim­ple sub­trac­tion prob­lems and she con­sis­tently performed at about that level. In the course of our back and forth I switch my phras­ing to the form “You have seven ap­ples and you take away five, how many left?” and she im­me­di­ately started an­swer­ing the ques­tions 100% cor­rectly, very rapidly too. Ex­per­i­men­tally I switched back to the prior form and she started get­ting them wrong again. It was ap­par­ent to me that sim­ply phras­ing the prob­lem in terms of con­crete ob­jects was ac­ti­vat­ing some­thing like vi­su­al­iza­tion which made the prob­lems easy, and just phras­ing it as ab­stract num­bers was failing to ac­ti­vate this switch. So as you say, for more tricky ar­ith­metic prob­lems, it may be the case that what men­tal cir­cuits are “ac­ti­vated au­to­mat­i­cally” de­ter­mine the first an­swer you ar­rive at, and you can ex­ploit that effect with edge cases like this.

• moridinamael 20 Nov 2018 21:19 UTC

  40 points

  on: Preschool: Much Less Than You Wanted To Know

  It seems so ob­vi­ous to me that the benefits of preschool would wear off af­ter a short num­ber of years that I feel like I must be miss­ing some­thing. How could it be oth­er­wise, given the cur­rent sys­tem? This is all com­pletely set­ting aside the de­vel­op­men­tal limi­ta­tions of small chil­dren.

  Let’s take two kids, Jamie and Alex. Pre­tend that there are no de­vel­op­men­tal limi­ta­tions on chil­dren’s brains and that they can be taught to read equally well at ages 3, 4, 5, and 6.

  Alex starts preschool at age 3 and they can read at a 1st grade level by the time they en­ter Kin­der­garten.

  Jamie does not do any preschool and can­not read at all when they en­ter Kin­der­garten.

  By the end of Kin­der­garten, Alex can read slightly bet­ter than 1st grade level, but not a lot bet­ter, since the cur­ricu­lum hasn’t been challeng­ing. It’s ba­si­cally been a re­hash of what they already can do. Jamie can read at the ex­pected grade level by the end of Kin­der­garten.

  By the end of first grade, no ac­com­mo­da­tions have been made for the fact that Alex is a slightly ad­vanced reader. Both kids are given es­sen­tially the same pool of books to read. Alex has not skipped a grade or put in an some se­cret fast-track pro­gram for kids who went to preschool, be­cause this does not ex­ist. So by the end of first grade, they can read about equally well. Maybe Alex reads slightly bet­ter, but since no real pres­sure is be­ing put on this ad­van­tage that would cause it to com­pound rather than diminish, it nat­u­rally diminishes un­til both stu­dents are at the same level.

  Act­ing as though any­thing else would hap­pen doesn’t make sense to me. It’s not like each year a child spends in school ex­erts some kind of Ed­u­ca­tion Force on their brain which ac­crues gen­er­al­ized scholas­tic abil­ity. Kin­der­gart­ners are taught kinder­garten level math and read­ing skills; kids en­ter­ing kinder­garten who already pos­sess these skills only benefit un­til the other kids catch up.

  So IMO the prob­lem isn’t the preschool “doesn’t do any­thing”. The prob­lem is that the sys­tem as it stands doesn’t ac­tu­ally uti­lize the po­ten­tial ad­van­tage of preschool. We are pretty far away from a sys­tem that would do so; such a sys­tem would need to be one that tai­lors the spe­cific ed­u­ca­tional con­tent to the spe­cific child.

  My four year old can read pretty well and can write well enough that you can puz­zle out what he’s try­ing to com­mu­ni­cate. But there is no ex­pec­ta­tion that he’s go­ing to skip kinder­garten be­cause of this. So in what sense could this ever be a long-term aca­demic ad­van­tage?

• moridinamael 16 Nov 2018 16:49 UTC

  2 points

  in reply to: Dagon's comment on: Sto­icism: Cau­tion­ary Advice

  “Real” sto­icism seems to de­mand to­tal re­lin­quish­ment of all at­tach­ments, to al­most ex­actly the same de­gree that “real” Bud­dhism does. I think this is a patholog­i­cal thing to want.

  Yes, it’s psy­cholog­i­cally benefi­cial to be less up­set about be­ing stuck in traf­fic. When you’re already stuck in traf­fic and can’t do any­thing about it, your choice to not be up­set about it is sim­ply a choice to avoid need­less suffer­ing.

  One might ar­gue that it’s even bet­ter to let your­self be re­ally an­noyed by be­ing stuck in traf­fic, and then per­mit your an­noy­ance to mo­ti­vate you to take ac­tions to avoid be­ing stuck in traf­fic in the fu­ture.

  The sort of per­son who would le­gi­t­i­mately not care if their child died would also be differ­ent from me in a num­ber of other very im­por­tant ways in or­der to be a rea­son­ably con­sis­tent agent. For ex­am­ple, if a stoic claims to be emo­tion­ally in­differ­ent be­tween “child death” and “child flour­ish­ing”, then what ac­tu­ally mo­ti­vates them? Why do any­thing, why make any choice? At least Bud­dhist thought is hon­est about this, and ad­mits that the only truly con­sis­tent solu­tion is a purely monas­tic life of med­i­ta­tion and ag­gres­sive pur­suit of non-ex­is­tence. Sto­icism, as far as I can tell, re­fuses to bite the bul­let of the con­clu­sions of its premises.

• moridinamael 15 Nov 2018 16:49 UTC

  2 points

  on: Sto­icism: Cau­tion­ary Advice

  Two things:

  1) I think a lot of peo­ple think they’re stoic when in ac­tu­al­ity they’ve just never had any­thing bad hap­pen to them. Modern life offers rel­a­tively few op­por­tu­ni­ties to test sto­icism, and by de­fault, ev­ery­one fails such tests with­out truly sig­nifi­cant prepa­ra­tion.

  2) Sto­icism is ac­tu­ally a huge drag.

  With re­gard to what­ever ob­jects give you delight, are use­ful, or are deeply loved, re­mem­ber to tell your­self of what gen­eral na­ture they are, be­gin­ning from the most in­signifi­cant things. If, for ex­am­ple, you are fond of a spe­cific ce­ramic cup, re­mind your­self that it is only ce­ramic cups in gen­eral of which you are fond. Then, if it breaks, you will not be dis­turbed. If you kiss your child, or your wife, say that you only kiss things which are hu­man, and thus you will not be dis­turbed if ei­ther of them dies.

  - Epicte­tus, The Enchiridion

  Who wants to live like this? I want to be dis­turbed if a loved one dies.