Rationalist Hobbies
This is a place to talk about hobbies that teach rationality lessons. I’ll start with a few:
Programming—Lets you practice math and logic, and gain an intuitive understanding of computation. Also teaches that sometimes you can’t argue with reality, but you can fix it. (Also suggested recently by ciphergoth.)
Cryptography—That is, designing/breaking cryptosystems and security protocols. Lets you practice math, logic, and probability theory. Also teaches that almost all human ideas are wrong. After doing this for a while, whenever you encounter a new idea (including your own), you’ll instinctively think “If I can’t find anything wrong with this, it’s probably because I’m not smart or knowledgeable enough or haven’t tried hard enough.”
Science fiction—Reduces status quo bias and gives interesting insights. Also teaches that the way a society is organized depends a lot on the set of technologies it has access to, so if you don’t like how your society works, one lever you have is to change that set.
Video games—Teaches that conventional “success” in life is not much less arbitrary than “winning” in a video game. They’re both fine for a diversion, but there are more interesting goals to pursue.
A couple others I’ve seen suggested in recent comments:
Chess—According to JGWeissman, it “teaches you to carefully consider the consequences of your available actions and choose the action with the best consequences”. (I’ve only played a few games of Chinese Chess, and for me, the lesson was that I don’t like competition, and I should look for things to do that nobody else is doing.)
Poker—Lets you practice statistics and overcoming emotional biases.
Hitting yourself on the head with a rock—Works on important physical skills, similar benefits to martial arts. Helps you overcome limitations. Reinforces actual reality of macro-scale objects, even when quantum mechanics is true. Teaches you about the difference between good and bad ideas.
Yeah, a lot of this seems more like rationalizing hobbies than rational hobbies.
I’m cut to the quick. But that does sound useful for anyone who finds himself doubting whether or not “it all adds up to normality,” if less extreme measures do not suffice. And I honestly do recommend getting some hard but friendly punches under proper supervision; it’s much more fun than it sounds.
“Reinforces actual reality of macro-scale objects”—brilliant :)
Human interaction—teaches human interaction.
Many people consider this activity and skill to be valuable in themselves, but rationalists have a particular reason for pursuing this. This is because it is easier to notice that someone else is confused than to notice it about oneself. And the analysis of other people’s confusion involves skills which transfer directly to the analysis of your own confusion.
The most important skill learned here is the identification of hidden, unshared assumptions. If you and your interlocutor each think that the other is being illogical, then the most likely explanation is not a lapse in logic, but rather an assumption that one of you is making and the other is not. Teasing out that assumption and shining a bright light on it is a necessary rationalist skill, even if you do all of your important thinking alone.
martial arts which include live sparring: As well as encouraging analogies to rationality study, they teach the application of planned strategies under conditions which are sub-optimal for quiet contemplation. As the illustrious sage Tyson has said, “everybody has a plan, until they get punched in the mouth.”
I’m not sure if there are any significant epistemic benefits to be gained, aside, perhaps, from practice using non-reductionist models (ie, chi) without accepting their premises.
From what I understand, the way that people get good at chess is by endlessly honing their ability to recognize a wide variety of visual patterns that happen to recur in the game (and especially at the higher levels, to some extent by just memorizing trees of opening moves), rather than by understanding deep principles or learning deep skills that generalize to other domains. As with anything complex that many smart people have spent a long time thinking about, it has a lot of interesting little details in its culture, and as with anything that takes sustained and concentrated effort and punishes mistakes, it probably teaches qualities like persistence and self-control, but on the whole I would definitely recommend against trying to become good at it if you’re not having a lot of fun.
Strongly agreed. This is largely why the next sentence in my quoted comment was “Becoming a grand master involves increasing your ability to do this in the domain of chess when you should be generalizing the skill to other applications.”
From what I’ve read by Josh Waitzkin, you need the massive memory file of positions, but you also need a system of connecting them together.
That’s true, but only of intensely competitive play. You can reliably beat beginners, amateurs, and even many devoted hobbyists by reading about three books worth of visual patterns and then honing your epistemic/instrumental rationality to a fever pitch.
There is a skill level beyond which you cannot win without sinking thousands of hours into memorizing trees of opening moves, but you can comfortably enjoy chess while learning new skills for thousands of hours before you ever reach that point, and even once you reach that point you can always play the occasional friendly game with someone who shares your distaste for otherwise pointless memorization.
That said, there are very few reasons to play any game repeatedly if you’re not having a lot of fun with it. If you’re going to bother to force yourself to do something, force yourself to learn a skill or read a textbook or earn money or help people, not to play a game. The whole comparative advantage of games is that they can teach us things without more than a trivial cost in terms of morale & willpower.
This sounds like a list of ‘any nerd hobby’ (with the honorable exception of 1 out of the 6, poker).
I tried to justify some of my hobbies to see if I could come up with anything that couldn’t be called a “rationalist hobby” to determine if it’s a useful designation or not.
Knitting—Trains your attention to fine detail. After you knit a pair of socks, whenever you wear knitted clothes you’ll instinctively think, “I could make this. There are no great unknowable secrets in manufacturing, only time and labor.”
Music Radio DJing—You learn how to speak fluently and without pause, and put together an entertaining set of music, which are both useful for signalling in social situations.
Reddit—Its up/downvoting system teaches how to quickly decide whether or not something is interesting to you, and the ability to submit content to be judged by the crowds can train your ability to write short copy that will appeal to large audiences.
I feel like I’m stretching, but also like science fiction, video games, chess and poker are probably also stretches.
Building up a really huge record collection—you learn that this is not such a great idea when you have to move house. It’s a subclass of this one.
I think DJ’ing and Reddit teach skills but not necessarily the skill of rationality.
Knitting, especially when you design your own patterns, does teach you the ability to make things. It sometimes disturbs me how possible it is to go through an adult life without making anything, and how long I go without making anything. (I’m not just talking about handicrafts; I’m talking about the process of designing and producing an entire new thing, that you can call “yours.” Writing a program is “making something.”)
Which quickly generalizes to any sort of physical making—woodworking, metalworking, etc.
Knitting is more like programming than woodworking and metalworking are, at least if you are designing your own knitting patterns. Knitting is a digitizable activity. To design a knitting pattern is to create an executable digital script, whose success is subject to digitizable constraints.
While these skills do help in some things, fluent speaking and making snap decisions are not really rationality skills in an epistemic sense. With luck, most any hobby can help you in an instrumental sense.
Science fiction and video games are stretches, but I can see poker and probably chess. For what it’s worth, I like and do the former two but not the latter.
Well, for example, anime is not included, because when I asked Eliezer, he apparently didn’t think it offered any rationality lessons.
Why do you make an exception for poker? Is there anything else you think should be on the list?
I except poker because, while there is nontrivial discussion of it here and especially on Hacker News, I don’t think poker is strongly associated with nerds. My gut impression is that poker as a hobby is still predominantly non-nerds. Hence, it’s the only entry that was not a nerd hobby.
(I would agree that anime is nothing special for rationality. This includes the many science fiction anime.)
I always figured the good poker players were all nerds, and the folk view of poker as this manly dominance contest of nerves etc just provided an endless stream of prey for them. I used to play Magic: The Gathering and read MTG blogs etc, and the pros drifted into poker because there was more money in it.
Poker, like other games of hidden information such as MTG, very strongly rewards rationality because it presses upon an important bias. Few people understand that the right play is the one that maximises your chances of winning given your information, not the one that will win given all the information.
Most people just can’t accept this. When they make the right bet and lose it messes them up—they start thinking of what they should’ve done differently, and their play diverges from the correct line. They can’t accept that they played correctly. The same thing happens when they make an incorrect play that happens to win.
Of course, if you abstract “Chess” and “Poker” to “board games” (which can then be abstracted together with video games to “games”) then this distinction vanishes!
Some of these hobbies are not like the others. I would classify hobbies based on whether or not rationalism is an essential prerequisite for engaging in the hobby. Programming and poker make sense to me but the rationales for the rest seem to be thinner, ascribing lessons that could be ascribed to almost any activity.
The distinction, as I see it, is that both programming and poker require rationalist discipline in depth that must be internalized to be effective. I can play video games or read/watch science fiction and benefit from the entertainment value without any investment in rationalism. By contrast, the very act of programming requires a considerable amount of logic and careful reasoning to produce anything but the most trivial result. Without a significant investment in rationalist thinking, you can’t participate in a constructive way, which to my mind defines a “rationalist hobby”.
Improv theater—Practice in examining and altering your motivations. Sharper-than-usual distinction between terminal goals (humor, drama) and immediate goals (those of your character).
Game design—Similar benefits to chess. You’re in a situation of perfect information and clearly-defined rules, and yet you still bump very quickly against the limits of your own cognition. You’re forced to acknowledge and try to work around your limitations, because there are no excuses. Unlike in chess, there are no unsolvable problems; at worst, there are problems with no elegant solution.
Writing pages on the tvtropes wiki—Heavy overlap with the goals of lesswrong; there are even tvtropes pages for fallacies and biases. Learning to spot patterns in literature has the same challenge as learning to spot them in thought: the most omnipresent ones are also the least visible.
Anything to do with language—Translation, conlangs, writing in English with arbitrary restrictions...they all help the same way that Rationalist Taboo does.
It also illustrates the gap between what rules directly allow/prohibit, and what they actually incentivize doing. It requires that bit of economic thinking—what incentive systems are your rules setting up? What feedback systems does this result in?
Edit: Actually, it requires careful clear thinking about human biases in quite a few other things too! In order to even do what I wrote above, you have to avoid obvious biases in playtesting, if you want to figure out what a good player will do. But then you also have to account for the fact that most players will probably not be seriously focused on winning, and you will have to consider what their biases will lead them to do instead, so you can make sure the game is fun for them as well.
Yes, and game design also provides some very good examples where people not thinking broadly enough can lead to serious problems. One good example of this is Dungeons and Dragons 3.0, where the play testing occurred with characters all playing as certain archetypes (the fighters hit, the wizards throw fireballs, the clerics healed in combat). As a result, no one noticed that minimal amounts of tactical sense could make the wizards and clerics massively outshine the other classes.
And, of course, Fun Theory!
Wei_Dai:
From what I’ve seen of it, I disagree. With a very few exceptions, even the highest-ranking SF in terms of popularity and critical acclaim is usually badly written, and instead of exploring truly imaginable and inventive developments, it projects the prejudices and illusions of its own time embodied by the author.
I’d be very curious to hear about some counterexamples from other commenters, though.
There has already been a few threads on SF recommendations. This one for example.
Read Arkady and Boris Strugatsky, Stanislav Lem, Robert Sheckley, that’s just the first names that came to my mind.
As for a parent post, society is more dependent on it’s economical system more than on anything else. And technology advancements, as history teaches us, don’t always define social structure. Ancient Greece and Roma are the great examples of enormous scientific and technological advancements living side by site with an underdeveloped dead-end social structure. Yep, there is a good Sci-Fi novel about this, read “Hard to Be a God”.
Rainbow’s End.
Hobbies that you do not actually enjoy but you chose because you thought they’d improve your rationality—You will become miserable and rapidly quit, thus learning not to get distracted by potential side benefits (improving rationality) if it makes you fail to address the reason you were looking for a hobby in the first place (to have fun).
Just read through the comments. Ouch. As Nesov said in the absolute denial macro discussion, “this is turning into a Let’s Spout Nonsense!!! thread”, but at least that one didn’t make me want to unsee. Cheers to Manfred.
I’ve “deleted” the post, given your criticism, and that (judging from the number of upvotes on Manfred’s comment at this point) a large number of readers apparently consider the post and/or followup comments to be ridiculous. I’m not really sure why though. It doesn’t seem obviously crazy to provide information about the rationality benefits of some hobbies that would-be rationalist can take into account, along with other costs and benefits, while choosing hobbies.
Perhaps the title, “Rationalist Hobbies”, gives the impression that I and others are recommending hobbies that every rationalist should have? If so, maybe I should have just renamed the post to “Rationality Benefits of Various Hobbies”?
I agree with your statement. The reasons for my criticism: a) it seems hard to determine objectively how much a given hobby helps your rationality, b) almost any hobby can be presented as “helping rationality” given enough ingenuity, c) some people will want to present their own hobbies this way. I’m not sure we can get much value from such discussions until we have some solution for at least one of these points.
If an idea isn’t obviously crazy, then the fact that it has problems seems to call for a discussion into the nature of the problems and possible solutions, instead of ridicule. In this case, b) and c) both seem surmountable. For b, many people are apparently able to tell when it is a “stretch” that some hobby offers rationality benefits, and appropriately discount such claims. For c, perhaps a warning to avoid rationalizing would be sufficient.
Overall you’re right. I apologize.
But your original post listed video games as a rationalist hobby. That certainly felt like a “stretch” to me, and to many other commenters too, I think. (Crypto, on the other hand, sounds like a very good recommendation.)
The title said “rationalist hobbies” (which I now think was a big mistake) but the post explained that these are hobbies that teach rationality lessons. I rarely play video games anymore, but I still think the lesson that hobby taught was a very valuable one that I may not have gotten from any other source. Most of us here probably do not need that specific lesson at this point, but I thought I’d include it to help explain how some of us got to where we are today.
That was unnecessary, given the post’s current score of +7. If all those people who upvoted Manfred really disapproved of the post (as opposed to lauding a display of wit), they could have voted it down to −28.
I don’t have any problem with the post itself, its title, or most comments; and in fact I don’t think your proposed title change would have “helped” one bit with folks like Manfred or cousin_it, since the idea of exploiting hobbies for their rationality benefits seems to be precisely what they find ridiculous.
Tabletop Roleplaying Games: Two main benefits:
Problem Solving and Lateral Thinking: With a good GM, you should be able to be challenged in ways that you can beat, but aren’t obvious. Particularly in GURPS, it doesn’t devolve into “I hit the monster until it dies”. Ever thought about how you’d negotiate with terrorists? Escape imprisonment? Survive a zombie apocalypse as a vlogger newscrew? When you play in person, you’re forced to think really fast. (“I can attack this which is doing the most damage, but it’d take a while and there’s still lackeys”, “Do I start healing this guy now, or just finish the fight?”, “What can I possibly say to make this seem innocuous?”). Also teaches the importance of planning ahead (“What can I say to the guard so that X will agree with me, even though we can’t talk?”).
You also have rules to brush up against, and designing a good character is difficult. Do you try and remain a generalist? Specialize at something really useful? What’s your role in the party?
Social: You have to play and deal with other people. You play the game in-person and it is prone to making a lot of in-jokes. Brings friends together and whatnot. It also has a lot of the same benefits as improv with regards to trying out new character traits. You’re not supposed to be pretending to be yourself, and you can try out personality tweaks in a way that’s not reflective of yourself. I played outgoing characters before I became less awkward as a person.
I’d say especially running tabletop role-playing games.
You have to balance different players’ tastes—which means learning to measure their tastes. You have to predict the actions of the people you know and plan for different choices—and you will be surprised by things they decide to do. You have to be organized, and especially you have to figure out how organized you have to be.
Agreed.
Well, they do slightly different things but I agree with your point. I think that players need to optimize more (aiming for a smaller slice of search space, with fewer possible actions), but that there’s lots of useful rationality training you get from running the game that you don’t get from playing.
Running a game gives you more experience working with different epistemic states. Like, nothing exists in the game-universe until you say it does so you have in your head what you know and are thinking of having happen, and what the players know. Good GMs are able to keep the two straight and change what they thought was happening, but wasn’t yet revealed, behind the scenes in able to adapt. For example, if you had another ambush in an encounter but the party’s already beat up you can just not have the ambush happen.
GMs also need to better predict other people’s actions, agreed. And design challenges that aren’t so easy as to be boring, but not so hard as to total party kill.
Musical performance. A few things I think are rationality-enhancing:
Realizing the difference between what your perception and others’—and between your perception and your recall. A confident, assured performance that has many mistakes you are aware of will often be impressive to an audience who is not expert or does not know the music—presentation goes a long way. But an expert will often perceive mistakes that you don’t. And hearing yourself recorded will very sharply inform you of the difference between what you recall and what you actually did.
It’s also useful for learning to understand others—especially playing in a section. (Orchestral winds usually play 2-4 to a section, with parts partially independent and partially in harmony; the first player is the “principal”, while the 2nd and subsequent players generally attempt to match the tone and playing style of the principal so that when the parts harmonize, it sounds like a single player. Orchestral strings play in larger sections--8 or more—all playing the same part as the principal.) Learning to model and anticipate others’ musical behavior—how they will react to conductors’ gestures, to unfamiliar music—and imitate it. How will they think about the music; what do their physical gestures indicate that their future actions will be? Get it correct and be rewarded by making beautiful sounds and impressing those around you.
Self-improvement through imitating and analyzing successful techniques: when I want to play better, it often helps me to consciously imitate people who are better, and then am able to identify what it is I’m doing that is causing the improvement. If I want a richer wind sound, I’ll try to imitate a college friend’s style and mannerisms; in doing so, I realize, I’ll sit straighter, take deeper breaths, relax more, be less hesitant at the beginnings of notes. If I want to be more agile in a tricky string passage, I’ll try to imitate a different person, and realize I’m not tensing up as much, I’m using less bow motion, and moving my fingers faster.
Instant feedback: you can hear when you have or haven’t mastered a task. You can try again continually until you reach success. (Also, not-so-instant feedback: will strangers will pay to hear you more than once?)
Emotional control under stress: performance can be very emotionally stressful. Learning to control this through practice is valuable. (In my first college recital, I had to stop halfway through because I was in a full panic: I’d forgotten the music and I was shaking too hard to continue. I left the stage to calm down before returning. I’d panic similarly when forced to give speeches in public. Through sheer repetition I’m now mostly unfazed by giving recitals or speeches in front of large crowds.)
No upper bound: I’m not going to exhaust the available material, as new challenges always exist and new ways to further refine existing skills are always possible.
And not a lesson but a benefit: status-raising (as some Overcoming Bias posts tickle the back of my brain). It has pretty much always been a plus in my personal and professional life to mention that I am a classical musician, particularly as many high-achieving people had some musical training. It makes people think both that I come from a higher social class than I do, and that I am more diligent and conscientious than I actually am...
Not to mention the warm-fuzzies you probably get from playing and being surrounded by beautiful sound. I sing in a church choir, which isn’t exactly status raising; I tend to get odd looks from my university-age friends. However, I started out as a tone-deaf eleven-year-old and can now sing solos in front of the entire congregation, which contributes hugely to a self-image of ‘I can accomplish anything that I’m willing to work hard enough on.’
Have you studied theory? Do you compose any music of your own?
I upvote programming. This gets you used to the concept that all problems are fixable with the right algorithm. Being in a problem-solving mindset can benefit you in fields orthogonal to CS, life your social life.
I downvote video games. They are too much of a time-sink. I don’t think that video games teach me anything useful, so I avoid them now.
Chess — I’d like to spend more time playing chess and go. My ability in these games are awful but I think I could learn to be awesome at them. I’m not sure that these skills are transferable to another domain, but maybe the meta-skill is.
Editing Wikipedia, especially articles on controversial topics—gives you good practice on extracting useful information from biased sources, and also gives you a better idea of how much you can trust information on Wikipedia (or any other source whose truth-filtering-mechanism isn’t better than Wikipedia’s). The history of distant countries or religions is a good start, because you’re less likely to start already attached to a particular interpretation but the sources you meet will still be mostly biased.
The knowledge of when to get a new hobby when you’ve squeezed the juice out of your current hobby might be the master skill, though I supposed it can’t count as a hobby in itself.
Remembering that almost all growth follows a sigmoid can be helpful.
Once I understood the reference I realised that was very clever. So when is the optimum point to dettach?
All the rationality benefits here look small to me relative to fun / mental health benefits, non-rationality educational benefits, and time costs.
That said, creating plausible alternate history involves basically the same skills as predicting the future; I think the genre emphasizes accurate depictions of what would have happened more than SF emphasizes accurate depictions of what will happen. Of course, in neither case can one actually check one’s answers.
Agreed on cryptography: I started reading up on it (specifically, Bruce Schneier’s Applied Cryptography) after seeing a LWer’s post on the matter. That made me realize that crypto bears directly on very deep questions, while at the same time having everyday relevance.
But how come none of the experts is smart enough to recognize that a “public key” is more like a lock than a key?.
Sports—By watching sports and being a fan, you can learn about biases, first recognizing them in others and then in yourself. You can learn to evaluate strategies based on how well they help one win. By taking part in arguments and analysis, especially in more statistically sophisticated (sabermetric) communities, you can learn to reason based on data and evidence and to judge the quality of an argument. You can test yourself by making predictions and observing the results, especially if you gamble or play fantasy sports. (See also here and here.)
I upvote sports, although I’m not a sports fan, because it’s another area where you bump up against reality. Teams have to make the right choices or they lose. It’s another way to gain intuitive experience with probability.
Of all the things you could watch on TV, sports are probably the best for your brain. Scripted TV shows accustom you to narrative logic, not real-world probabilities. TV news does the same thing.
When I read this comment, it strongly reinforced the part of my mind that was wondering how many hobbies existed that couldn’t be construed as rationalist exercises if one wished to do so. However, you may want to discount me slightly because I’m not a fan of sports at all, and am sometimes slightly irritated at seeing sports on TV for reasons that I have not paid much close attention to.
With that said, I propose that we come up with a list of hobbies that shouldn’t be considered “rationalist,” and see if anyone else can come with a convincing enough note regarding them to convince us otherwise.
So, to semi-stick my head out, my hypothesis is that the space of hobbies that could be considered “rationalist” is too large to be very useful to us. Grouping these according to the skills that they purportedly train may help counteract this; perhaps it would be useful to rank skills in each group so that we could quantify opinion on which ones increase certain skills more than others.
Politics and history—lets you see what answers you would give to difficult real-life problems, and you have to keep in mind that the people who dealt with them at the time were really extremely smart and politically able. They might not have been experts on rationality, but they were certainly experts at it. Learn from their failures.
I like history because it supplies lots of counter-examples. You sometimes see people say silly things in discussions that they wouldn’t say if they knew their history.
(For example, on lukeprog’s blog, I saw a commenter say that he didn’t know of anyone shedding blood in the name of the Buddha, as compared to Mohammed or Jesus. Sure, as far as he knew...)
I play video games because:
Do video games enhance cognitive abilities?
Video Games Boost Brain Power, Multitasking Skills
I just spent a few moments puzzling over this because I thought it was saying that by analogy, we should be reaping the benefits of encouraging girls to improve their graphics.
To which my immediate response was “don’t they get plenty of encouragement already?”
Not really...
When I said “encouraging girls to improve their graphics,” I wasn’t talking about video games.
Magic: the Gathering is good.
I think that some explanation of how you thought that playing M:TG was ‘good’ would be more useful as a comment.
Sorry.
Anyway, one thing that distinguishes Magic from other games is that it keeps changing: every few months, there’s a new set released, which changes the relative effectiveness of various strategies. You have to keep experimenting and learning in order to continue to do well; figuring out what the best deck to play at a tournament basically requires Doing Science, and it can—and does—go wrong in all the same ways that real science goes wrong.
Agreed, but I think that a strong point against Magic is the large amount of money that it costs to build a competitive deck. I enjoyed playing it for a few years, but eventually ran out of steam (partially for this reason, partially because I especially enjoy B/W control, which wasn’t so good anymore.)
Yeah, Magic is ridiculously expensive when compared to many other games. :(
Hardly at the same level, but I wonder how much can be gained just by doing something like switching from Chess to something like Chess960. (Hardly at the same level because it’s just a one-time injection of randomness; if people had been playing that all along, there still would not be new chess expansions released every few months.)