You have not found a way to connect your hobby to rationalism yet. It itches. You are not whole. It is forbidden to post an article entitled Rationalist Hobby on Less Wrong. You lie awake at three in the morning, trying to create puns.
-- Alicorn, Rationality Gothic [2]

(Note from 2017: This sequence was written before Alicorn's classic jab at people shoehorning rationality into random hobbies. I'm fairly embarrassed by it at this point, but don't have time to rewrite this into something I'm not embarrassed by, and several found it a very helpful introduction to drawing.)

((Also, this sequence ends abruptly when my 2011 life got in the way. If you get to the end and go "aarrg! what do I do next!?" the answer is "purchase 'Drawing on the Right Side of the Brain' and finding a figure drawing class))


This post begins a mini-sequence that discusses how to draw, reports on an experiment about teaching people how to draw, and examines how rationality and good drawing practices are related. (As it turns out, a fair amount)

I'm a professional artist. I have a fairly extensive background in traditional drawing, but most of my training is in computer animation. I chose my career because I liked the control offered by the computer - the ability to undo, to manipulate art in procedural ways, and most of all for the flexibility to duplicate things, repurpose them for different projects and combine my love of visual art with my love of game design, animation, and various other mixed media.

But now I work 10+ hours a day at an advertising agency. I spend all day getting paid to stare at a computer screen. Most of my other hobbies also involve staring at a computer screen. And many types of digital work are taxing on the same set of creative muscles, so at the end of the day I didn't have energy to work on the personal projects I wanted.

One option would be to get a different job that didn't tax those creative muscles or involve staring at a screen. I've actually considered getting a "physical" job - after years and thousands of dollars of college to get a nice posh job without physical labor, I actually think it might be better to get PAID to exercise. And instead, use my free time to channel my skills from colleges into personal creative projects that I'm passionate about.

I may do that some day, but I DO like my job, I like the people there, and I continue to learn important skills. So instead of modifying my job, I modified my hobbies. A few months ago I began drawing people - on subways, in coffee shops, in parks, etc. This gave me a new creative outlet, as well as a new social outlet. (Starting a conversation with "Hey, can I draw you?" is a pretty useful technique - not only does it provide an excuse to begin talking, but if you follow up with a good drawing, you've established right off the bat that you're an interesting person with a valuable skill. You've also flattered the other person a bit, and if the conversation enters a lull, it's okay - just draw for a while until you can think of something to say.)

So I've been getting better at traditional drawing, and better at social interaction, and more confident in general. And at a local Less Wrong meet up, it recently it became clear that...

  1. Other people wanted to learn to draw
  2. I wanted to learn to teach
  3. A few people wanted to model. So the "Drawing Less Wrong" meetup was born. I prepared some lesson plans and began holding 4-hour workshops.

What interested me was how much the study of drawing was relevant to rationality. Not only do you have to learn to observe reality (this is surprisingly hard), but you have to pretty much scrap your entire model of how you think drawing works. (Almost everything you will naturally gravitate towards is wrong). Most artists don't notice that they should be applying these lessons to the rest of their life, but I think the skills can generalize if attention is brought to that notion.

In the past, I've been to figure drawing workshops where I saw people go from not being able to draw much at all (one person showed up to class with a *horribly* copied manga drawing that they said had taken them 12 hours), to being able to execute a reasonable gesture drawing1 in about 60 seconds. It took them about 8 hours of dedicated practice. I wanted to try and replicate that.

Soon to follow are a collection of posts discussing the nature of talent, how to draw effectively, and lessons I learned from trying to teach people extremely counterintuitive models of reality.


[1] "Reasonable Gesture Drawing" is a specific phrase that means something to trained artists, which non-artists may misinterpret. It doesn't mean "looks amazing." It does mean that this person improved in important ways in a short time.

[2] I refuse to call it "Rationalism Gothic" because Rationalism means an oddly specific philosophical school that is actually kind of the opposite of what Less Wrong is about.

New Comment
38 comments, sorted by Click to highlight new comments since: Today at 10:46 PM

Several years ago I worked my way through Drawing on the Right Side of the Brain after deciding I wanted to be able to draw, then got busy with other real life pursuits and didn't really follow it up at all.

Earlier this year I had my passion for the visual arts rekindled, and as a result I've been doing a lot of figure sketching, picking up a couple more books on the subject and spending a silly amount of money on various forms of carbon. In the interim I've learned a lot more about the cognitive sciences, neuroscience, the human visual system, principles of design, aesthetics, and the psychology of producing good work. While drawing, I found an awful lot of those concepts and subjects were directly applicable to what I was doing, and also that drawing provided examples and insight into those areas.

One example is the Dunning-Kruger effect: If you train at something for about twenty hours, most non-practitioners will not be able to tell the difference between you and an expert. People who think they can't draw will see some of my stuff and conclude I'm an incredible artist, whereas anyone with a cursory amount of experience will be able to spot the (many) flaws in my work.

Another is the notion that drawing (and to an extent all representational art) is tricking your perceptions into recognising something that doesn't exist, and as a result is parochial to human beings. Things people like to look at tend to be things that engage specialised recognition patterns, (faces, food, idyllic landscapes, human bodies, tools, animals, etc.) Much to Clippy's distaste, it's very easy to recognise specific types of visual subject in a relatively simple set of geometric shapes.

There's definitely a lot of scope for a sequence like this to have bearing on LessWrong-conducive subjects.

In the interim I've learned a lot more about the cognitive sciences, neuroscience, the human visual system, principles of design, aesthetics, and the psychology of producing good work.

I don't actually know too much about some of those things, and in particular how they work together. I'd be interested if you shared relevant insights for each of my follow up posts, or did one of your own if you think you have enough content.

I'm really not an expert in any of them, but I've had enough exposure to recognise a few unifying trends. I've also been thinking about it a lot over the past few months, so my enthusiasm probably outstrips how well-informed I am. Still, I'll contribute where I can.

Fair enough. Look forward to the discussion.

What amazed me was how much the study of drawing is a good exercise in rationality. Not only do you have to learn to observe reality (this is surprisingly hard), but you have to pretty much scrap your entire model of how you think drawing works. (Almost everything you will naturally gravitate towards is wrong).

My prior confidence (relative to what the follow-up posts have to say) of the first sentence above is low. Yes, you have to learn how to observe reality, but you'll do so as an artist, not as a more generalized agent. Drawing may call your attention to previously-unnoticed details, but these details may be irrelevant for all purposes other than drawing them. To that extent, learning to draw will not add epistemic benefit.

And the second listed benefit above (you will learn the truth about how drawing works) is not a benefit either, except in the narrow sense that it involves updating your map. This is helpful, but it's not a rationality technique (i.e., something that aids in map-updating generally).

I look forward to reading the follow-ups, because if I am wrong then it will give me an opportunity to become less so. And the existence of repeating meet-ups devoted to drawing is evidence against my claim.

My prior confidence (relative to what the follow-up posts have to say) of the first sentence above is low. Yes, you have to learn how to observe reality, but you'll do so as an artist, not as a more generalized agent. Drawing may call your attention to previously-unnoticed details, but these details may be irrelevant for all purposes other than drawing them. To that extent, learning to draw will not add epistemic benefit.

This statement is completely true, and I should have qualified it better (I'm going to rewrite it slightly). Future posts are going to elaborate on how drawing CAN be an exercise in rationality, if you apply the lessons in a particular way.

I'm thinking it's a deconstruction of how a (visual) algorithm feels from the inside.

Not the direction I was going, and I'm not quite sure whether you're joking, but that's an interesting way of framing it.

Of relevance is Scott McCloud's Understanding Comics, which offers many interesting links between visual art styles/techniques and their human interpretations. It is not a drawing tutorial, but it is definitely of interest to those who draw, or who otherwise want to imply specific ideas through imagery (i.e. GUI designers). It's also a lot of fun to read. :-)

HIGHLY recommend this book. I don't know that I actually consider it the "second or third best book that I've read," but when I'm asked the question "What books have inspired you?" it pops into my head immediately.

What's the very best book you've read?

Harry Potter and the Methods of Rationality. (Even among Less Wrongians I seem to be on the far end of the bell curve of "How mind blowingly amazing that story was." In addition to opening me up to a big wide world of rational thinking, and inspiring me in a way that no other work of art has, all the creative elements of the story exactly match my preferences for humor and narrative.)

I once considered doing a mini-sequence on marksmanship along similar lines to your drawing sequence, but decided it was too tangential to the Less Wrong project to be of much value. Was I wrong?

"Hey, can I shoot you?" is somewhat less appropriate as a conversational gambit, compared to "Can I draw you".

What interested me was how much the study of drawing was relevant to rationality. Not only do you have to learn to observe reality (this is surprisingly hard), but you have to pretty much scrap your entire model of how you think drawing works. (Almost everything you will naturally gravitate towards is wrong). Most artists don't notice that they should be applying these lessons to the rest of their life, but I think the skills can generalize if attention is brought to that notion.

It seems to me that everything about drawing that makes it a good training ground for rationality is even more true of, say, long range precision rifle shooting. Also, evaluating your progress is easier because the metrics are more objective (targets don't lie) and your biases are more apparent (you can go so far as quantifying your biases by fractions of a centimeter and even giving them directionality in 3-dimensional space).

"Hey, can I shoot you?" is somewhat less appropriate as a conversational gambit, compared to "Can I draw you".

Your response misses the point. Obviously, I didn't have marksmanship in mind for helping you approach people on the subway. I believe the primary goal of this sequence was to explore how to improve general rationality through drawing, not merely better social skills through drawing. It is my contention that various shooting sports teach skills that are just as or more readily transferable to other domains that require rationality. Of course, I am willing to be corrected by Raemon as to his/her intention in writing this sequence.

Your response misses the point

I think your response misses the joke.

But yes, I think this could be an interesting. I have a mild phobia of guns (that I don't see a need to override), but I would be interested to read the material. In general, I am a fan of Less Wrong articles that talk about a real, instrumental goals we have had and how we achieved them, as long as the experience can be tied to rationality tools.

Put another way: Technically, ANYTHING you can win at relates to instrumental rationality, insofar as you can always rationally choose not to think about too hard about it (i.e. dancing and sex may require you to deliberately turn off certain parts of your brain). My litmus test for "should this be a Less Wrong article" is: "Did your 'rational approach' to this subject consist primarily of identifying that Less Wrong skills weren't that relevant and then following conventional wisdom, or did you actually have to use tools that you learned here?"

I know others may disagree with me on this (my post's already been downvoted at least once), but anything that connects rationality to actual instrumental success seems like a good idea to me.

Metaphors are cheap. With a little effort you can take any arbitrary hobby, from abstract math to whittling little wooden bear statues, and come up with plausible rationality lessons to be learned from it. Some of them will even generalize to most practitioners of the art. And that's fine as far as it goes; you certainly shouldn't reject any insights that fall into your lap, and on the communicative side of things there's always a need for compelling examples with a personal flavor.

But outside of a few tasks that're tightly and explicitly bound to specific rationality skills (poker comes to mind), I don't think it's a good idea for us to spend a lot of time debating which hobbies are better at inculcating good habits of thought, let alone recommending rationalist (sic) ones to people. There's an information gap and a lot of potential confirmation bias there, and that's usually a recipe for unproductive debate. At most it might be a good idea to write a few survey posts regarding the applications of common hobbies -- and while the OP might pass muster under that light if you turn your head and squint, I think it's probably better viewed as a case study.

At most it might be a good idea to write a few survey posts regarding the applications of common hobbies -- and while the OP might pass muster under that light if you turn your head and squint, I think it's probably better viewed as a case study.

I wanted to respond to this, but I'm not actually sure what you meant. Can you clarify a bit?

long range precision rifle shooting

It wasn't until this point that I realized you were not talking about archery.

Rifle shooting was just an example (the one I'm most familiar with), but you can practice the art of marksmanship with a bow or even a slingshot for that matter.

I would be interested in a sequence like that. Of course, if it only touches rationality tangentially then maybe LessWrong is not the best place for it. But again, I personally would be very interested in it.

Perhaps this is true only for me, but the title led me to believe that this post was literally going to be about drawing Less Wrong (as in metaphorically drawing the community Less Wrong) not about drawing as it applies to Less Wrong-ian concepts and interests.

Anyways, this sequence sounds interesting. I've always - maybe inaccurately - identified myself as a person who could not draw well, and I'm sure I have various wrong assumptions about drawing that will be corrected.

Me too, I was expecting something like this.

one person showed up to class with a horribly copied manga drawing that they said had taken them 12 hours

It took them about 8 hours of dedicated practice.

So, shouldn't that guy have stopped sucking about two thirds of the way through his picture?

Pretty sure the drawing was from before the workshop.

I think the point is that you progress much faster with proper methods than when doing things naturally. Also, you assumed the person was male.

The person was in fact male, but in fact, saying that it was a manga drawing is slight evidence that he was female. And in general, yeah, it's better not to make the assumption.

It took them about 8 hours of dedicated practice.

Either this is missing a zero or everyone is wrong about "practice a lot".

The people who talk about practicing a lot are talking about reaching an expert skill level. Raemon's example was a basic technique executed with "reasonable" competence.

Sometimes it's just the techniques. My drawing of faces became significantly better once I read a "how to draw" book that told me that the eyes are in the middle of the head (many beginner artists draw them towards the top). Likewise, knowledge of basic anatomy is helpful in drawing figures.

That said, I'm very interested in this series. Good luck!

[-][anonymous]12y20

.

I have briefly considered that. (I didn't dismiss it, it just seemed like too big a switch at the time and I had more important things to evaluate). I may look into that again.

I don't doubt that it would be a useful skill, and I'd like to have it, but I seriously doubt that it would be easy.

Is this sequence going to be about drawing specifically, or visual arts in general?

This will focus on drawing. There's obviously plenty of information about art in general that can be relevant to rationality, but drawing is where I can talk more about actual results.

Wow, sounds great. I'm really looking forward to this :)

Rationality can be useful when drawing. It allows you to avoid simple mistakes which you could otherwise make. I think this is especially true when you are for example inking your work, or doing some other other task which is mostly mechanical. However, sometimes following mere feelings can provide very interesting results. I am not a good drawer, nor do I actually know anything about drawing, but I draw a little bit every now and then. I find drawing most enjoyable when I draw quided by intuition, just letting the pen draw curve after curve the way it feels. I have found that when I do this, I achieve results more to my liking than when I actually think about what to draw and how. Maybe this is simply because I don't have much actual knowledge about drawing, I don't know.

Anyway, interesting post, thanks.