Drawing Less Wrong: Technical Skill

The abil­ity to ob­serve is prob­a­bly at least 2/​3rds of what sep­a­rates non-artists from am­a­teur artists. But those 2/​3rds are near-use­less with­out the abil­ity to move your pen­cil the way your eyes want to it to go. And once you’ve tran­si­tioned into an am­a­teur artist, around 9,000 hours of hon­ing your tech­ni­cal skill is what sep­a­rates you from a pro­fes­sional.

“Tech­ni­cal Skill” is a broad term—kind of a catch all for all term for var­i­ous mo­tor skills you’ll need to de­velop, back­ground knowl­edge about how par­tic­u­lar types of lines and shapes are per­ceived by most hu­mans, and how to com­bine those skills and knowl­edge to pro­duce par­tic­u­lar effects with your draw­ing.

I can’t even be­gin to cover all of it, and most of it isn’t re­ally ap­pro­pri­ate for Less Wrong. But I will talk about some key mo­tor skills that tie in with the next ar­ti­cle, and a sig­nifi­cant bias that plays a role in them.

This ar­ti­cle was challeng­ing to write—dis­till­ing a kines­thetic pro­cess into writ­ten words is difficult. This ar­ti­cle will not be a sub­sti­tute for hav­ing a teacher and a model, nor will it tell you ex­actly what ex­er­cises to do. But it will try to lay down some con­cepts that I’ll fur­ther ex­pound on later.

Hold­ing the Pencil

For many of you this may seem ba­sic, but at least one reader com­mented that they went for years with­out un­der­stand­ing this, and be­cause it seemed ba­sic, no­body ever no­ticed it and cor­rected them.

Hold­ing a pen­cil should look ap­prox­i­mately like this:

But its a bit more com­pli­cated that that. Many skil­led artists hold the pen­cil in differ­ent ways. The biggest things to keep in mind are:

  1. Don’t grip the pen­cil too tightly. You’ll hurt your­self, and it won’t help.

  2. If you hold the pen­cil closer to the tip, you will have more con­trol over it, which is use­ful for fine de­tails.

  3. If you hold the pen­cil to­wards the back of the pen­cil, you’ll have greater range of mo­tion, and al­lows you to quickly draw larger lines in a sin­gle stroke. It also will be looser, which can feel hard to con­trol but can also pro­duce cer­tain line qual­ities you may want.

(I per­son­ally tend to hold my pen­cil similar to the image above, but closer to the mid­dle of the pen­cil)

A few ex­am­ples of a pen­cil grip in mo­tion:

This man’s grip is similar to mine, al­though the tech­nique he de­scribes isn’t some­thing I think you should be wor­ry­ing about just yet. (I’ll be talk­ing about Dar­rel Tank’s web­site later on—I think he has good tu­to­ri­als on tech­ni­cal skills, but does not pri­ori­tize them based on their low-hang­ing-fruit-ness.)

This car­toon­ist switches grips a few times, demon­strat­ing how they can be use­ful at differ­ent stages of draw­ing. This video is par­tic­u­larly in­ter­est­ing be­cause his “loose” grip is ac­tu­ally closer to the front, which I haven’t seen of­ten.

Slow Draw­ing and the Sunk Cost Fallacy

I’ve spo­ken a few times about “slow, small, pre­cise lines,” and im­plied that they are a ter­rible idea. They of­ten are. You’ll be draw­ing slowly dur­ing some ini­tial ex­er­cises that de­velop ob­ser­va­tional skills. But as soon as pos­si­ble, you’ll want to start de­vel­op­ing a form of hand-eye co­or­di­na­tion that in­volves mov­ing quickly us­ing long lines. Un­til you achieve that, the small, metic­u­lous lines will prob­a­bly have a choppy qual­ity, and cer­tain com­po­si­tions will be harder to cap­ture.

Much of the “en­ergy” of your draw­ing1, and the qual­ity of the com­po­si­tion, will be es­tab­lished within the first one to two min­utes. This isn’t a hard and fast rule, but holds true most of the time. Yes, you can erase, and re­work things. With pure ob­ser­va­tion, in­finite time and brute-force-re­work­ing, you can craft a draw­ing that perfectly cap­tures re­al­ity. But ev­ery time you erase and fix a draw­ing, two things hap­pen:

One is that the pa­per smudges, tears slightly and oth­er­wise de­grades. This might be okay if you’re work­ing on a com­puter tablet, but so long as you’re prac­tic­ing with a real pen­cil, it’s an is­sue. After eras­ing 5-10 times, your draw­ing will have no­tice­ably de­graded. It’s not game over, but you’ll have to work harder to over­come it.2

The other, more im­por­tant con­cern, is that the more you’ve drawn, the more you’re at­tached to the ex­ist­ing sec­tions of the draw­ing. Say you’ve drawn an arm bent awk­wardly. You can erase it and fix it. But the arm doesn’t ex­ist in iso­la­tion. It con­nects to the shoulder, which con­nects to the torso and neck. Fix the arm, and you have fix the all those other things.

You prob­a­bly won’t want to fix them all, be­cause it will feel too sad for you to have to erase large sec­tions of your draw­ing. And even if you DO fix them all, the re­sult won’t be a fluid, grace­ful image that cap­tures the mo­tion and in­ter­con­nected mus­cles of your sub­ject—it’ll be a hodge podge of Franken­stei­nian body­parts, awk­wardly sewn to­gether.

(There’s also an­chor­ing in­volved: once a line ex­ists, you’ll have trou­ble eval­u­at­ing new lines on their own mer­its, in­stead of how they com­pare to the ex­ist­ing ones.)

Sev­eral times over the past year, I’ve worked on a draw­ing of a per­son for 5-10 min­utes. By the 1 minute mark, I know some­thing’s off about the draw­ing. By the 2 minute mark, I’ve started eras­ing and re­work­ing things. I have a nag­ging sense that I’ve done this be­fore, and that the next 8 min­utes will in­volve lots of eras­ing, and a draw­ing that still isn’t very good.

10 min­utes, and lots of eras­ings later, I have a dis­ap­point­ing draw­ing.

The nag­ging sen­sa­tion that this isn’t go­ing to work well… that’s what the sunk cost fal­lacy feels like. You’ve put in a few min­utes of work (some­times much longer—you can go down this path for hours). Start­ing over would feel like defeat, like your pre­vi­ous work was a waste.

The cor­rect ac­tion is to start over any­way. It’s true when you’ve only been work­ing for a minute and have just no­ticed the feel­ing. It’s still true 10 min­utes later. It’s still true if you’ve spent 4 hours painstak­ingly eras­ing and dou­blecheck­ing against re­al­ity, adding lots of nu­anced shad­ing. It’s an oddly near-uni­ver­sal tru­ism, not just in draw­ing but in many pro­jects, that the thing you just spent 4 hours work­ing on, which would take an­other 4 hours to finish, could be done in 10-30 min­utes if you started over.3

There are rea­sons to spend 4 hours on a draw­ing. Those rea­sons will not be rele­vant to you in the near fu­ture. All the most im­por­tant el­e­ments of a draw­ing should take you no more than a few min­utes. After that, you’re get­ting dis­tracted by de­tails, which might make the draw­ing more in­ter­est­ing, but won’t ac­tu­ally fix the ex­ist­ing prob­lems with it.

Most im­por­tantly, it won’t help you learn to avoid those prob­lems in the first place.

Fast, Con­fi­dent Lines

So, you need to cap­ture the most im­por­tant el­e­ments of a draw­ing, quickly:

  • You need to cap­ture how differ­ent body parts con­nect to­gether, as a seam­less whole.

  • You need to es­tab­lish a good com­po­si­tion, so that the de­tails you work in later aren’t just re­in­forc­ing a bad de­sign.

  • I haven’t elab­o­rated on it yet, but you’re go­ing to want an en­er­getic, in­ter­est­ing draw­ing, which is sim­ply hard to com­pose with­out work­ing quickly.

To do all this, you need to de­velop a par­tic­u­lar kind of hand-eye co­or­di­na­tion, which is prob­a­bly differ­ent than what you’re used to. You need to be able to draw large sec­tions of the body, us­ing a sin­gle line. That line can change di­rec­tion. But it needs to be done in one fluid mo­tion.

This artist demon­strates what I mean by “con­fi­dent lines.” She blocks out large chunks of body with long curves, with­out sec­ond guess­ing. She doesn’t bother draw­ing the arms or feet, but she does end those lines AFTER they’ve curved in a new di­rec­tion, so if/​when she con­tinues them she’s set her­self to con­tinue them grace­fully.

No Erasing

You need to work quickly, and fluidly. Stop­ping to erase will in­ter­rupt your flow. So you need to learn to draw with­out eras­ing. There are two ways of do­ing that:

1) Don’t make mis­takes ever.

This ac­tu­ally is not as un­rea­son­able as it sounds. In the first 30 sec­onds, iden­tify the most im­por­tant lines of the draw­ing, and draw them. It’s what the artist in the pre­vi­ous video did. Ob­vi­ously, this is… es­sen­tially im­pos­si­ble for a be­gin­ner. You’re go­ing to make mis­takes. But what you CAN do is draw boldly, con­fi­dently, let the mis­takes hap­pen, and then rather than try­ing to fix them, move on to the next draw­ing af­ter 30 sec­onds. Over time you’ll get bet­ter.

I haven’t watched new stu­dents try to learn with JUST this philos­o­phy, and I have no idea how long it’d take to de­velop from scratch. But if you DO have some pre­vi­ous draw­ing ex­pe­rience, I think this may be a good ap­proach, at least to try out. If your goal is to pro­duce some­thing like the woman in the video, draw­ing 5 draw­ings in 30 sec­onds with sim­ple, bold lines will prob­a­bly pro­duce at least one draw­ing that’s bet­ter than the one you’d do in two and a half min­utes.

2) Be okay with your draw­ing be­ing messy.

This is what I ac­tu­ally recom­mend for be­gin­ners.

A big hur­dle young artists have, when they’re tran­si­tion­ing onto the path of a “pro­fes­sional”, is they feel that “messy is bad.” They’re draw­ing like this:

[2017 Note: this used to be a link to a perfectly bad high school Dragonball Z fan-art, which was re­ally use­ful for high­light­ing the sort of er­ror modes I was point­ing at, but a) it was sort of mean to use it, b) in 2017 the link is ap­par­ently dead, c) I was un­able to find an­other ex­am­ple that illus­trated the ex­act qual­ities I want to point to]

[Fol­lowup: Some­one vol­un­teered this old high school art, which doesn’t quite hit on the same set of is­sues but works well enough to illus­trate the ba­sic con­cept]

When they should be draw­ing like this:

They’re look­ing at the former, and see­ing it as a fairly clean draw­ing that just needs to be fixed a lit­tle.

A col­lege pro­fes­sor looks at Ex­am­ple A, cringes, and thinks “man this per­son is go­ing to need to sys­tem­at­i­cally bro­ken down over the course of two semesters un­til they’re ready to be­gin learn­ing, and it’s go­ing to be painful for the both of us.” They look at Ex­am­ple B and think “This per­son knows ex­actly what to do already, they just need to do it for an­other 10,000 hours.”

The prob­lem with Ex­am­ple A is that the artist is copy­ing su­perfi­cial el­e­ments of a par­tic­u­lar style, with­out un­der­stand­ing the un­der­ly­ing prin­ci­ples that make good a good figure draw­ing. Ex­am­ple B has lots of over­lap­ping lines, and vague messy shapes. But the figures there com­mu­ni­cate a good un­der­stand­ing of anatomy, a grasp of weight, de­cent com­po­si­tion.

As an as­piring artist, don’t ask if your draw­ing is bet­ter than A. Ask if it’s at least as good as Ex­am­ple B. If you want to draw truly good Manga art, you must first learn things OTHER than the su­perfi­cial char­ac­ter­is­tics of Manga. And while it may look like a mess at first, as you learn to draw that way, you’ll un­der­stand that there’s ac­tu­ally a lot of in­for­ma­tion there that Ex­am­ple A has missed.

(No offense to those of you out there cur­rently draw­ing Ex­am­ple A. I’ve been there. It’s a rite of pas­sage. In par­tic­u­lar, no offense to the blog I took Ex­am­ple A from. The blog­ger iden­ti­fies it as one of their old, mid­dle school works and gives other ex­am­ples that show a lot of im­prove­ment. I tried to draw my own ver­sion of Ex­am­ple A, but it’s ac­tu­ally re­ally hard for me to draw that par­tic­u­lar way now, and I can’t find any older ex­am­ples).

Be­gin Light, Em­pha­size with Darks

One im­por­tant part of tech­ni­cal skill is be­ing able to draw lines in the lo­ca­tion you want them. Another im­por­tant part is be­ing able to ad­just the light­ness or dark­ness of those lines (as well as thick and thin-ness)

Your draw­ings are go­ing to be messy. But you want a par­tic­u­lar kind of mess. If you look at the right-most figure in Ex­am­ple B, you’ll see that all the lines are the same thick­ness. This is okay—the artist has enough skill that they’re all ap­prox­i­mately cor­rect, and the ones that are off have been re­pur­posed—in­stead of be­ing pure mess, they end up rep­re­sent­ing the vol­ume of the figure.

The cluster of scrib­bles in the face sug­gest its round­ness, and hav­ing a bunch of them de­val­ues the im­por­tance of each in­di­vi­d­ual line, so that even if none of them end *perfect*, your brain doesn’t re­ally care—it sees that they’re all sort of fuzzy and ac­cepts the av­er­age po­si­tion in a sort of “Wis­dom of Crowds” way.

It’s okay that all the lines are the same thick­ness, be­cause none of them are *com­pletely* off. There’s no gi­ant leg that ac­ci­den­tally stuck out way too far and ru­ined the image. If it had, it’d be re­ally hard to re­pair the draw­ing. Espe­cially since you’re try­ing to work quickly, with­out eras­ing.

You’re go­ing to be mak­ing sig­nifi­cant mis­takes, and you won’t want to start over ev­ery sin­gle time.

The solu­tion is to do your early work lightly, and then, once you’ve iden­ti­fied the parts you like, use dark lines to em­pha­size those ar­eas. This tu­to­rial demon­strates how to draw like this. No­tice that within 30 sec­onds, he’s es­tab­lished a frame­work, with­out wor­ry­ing about mak­ing any “clean” shapes. Over the course of 2 min­utes, he builds on that frame­work, filling in the mass of his sub­ject mat­ter, and even­tu­ally adding much darker lines to em­pha­size the fi­nal shape.

To do this, you need to be able to ad­just the “value” of your lines (how light or dark they are). This takes some prac­tice. A good ex­er­cise is to cre­ate a se­quence of value-swatches like this:

Be­gin with the swatches on the far sides—make the dark­est dark and light­est light you can pos­si­bly do. Then try and fill in the rest, grad­u­ally dark­en­ing.

Be­gin your draw­ing with some­thing close to the sec­ond-light­est swatch. For now, try not to get much darker—it’s easy to ac­ci­den­tally get too dark too quickly, and then are your lines are uniformly black and you can’t em­pha­size the parts you want.

So… Re­cap:

Th­ese are only some of the skills you’ll need to ac­quire, but they’re the most im­por­tant in the im­me­di­ate fu­ture. So in sum­mary:

  • Hold the pen­cil with three fingers, not too tightly.

  • Don’t be afraid to start over.

  • Work quickly, with­out stop­ping to erase.

  • Draw strong, con­fi­dent lines.

  • Be okay with your draw­ing be­ing messy—let ex­tra lines help define the form.

  • Start with light lines, make your mis­takes, then em­pha­size the good parts with dark lines.

[0] There’s some­thing akin to an­chor­ing bias here as well—once part of the draw­ing ex­ists, even if you’re try­ing to com­pletely ig­nore it, it’ll be warp­ing your per­cep­tion of what’s ac­tu­ally go­ing on.

[1] I promise I’ll ex­plain what I mean by “en­ergy” soon.

[2] Each draw­ing tool has sep­a­rate rules that need mas­ter­ing. This in­cludes pen­cils, char­coal, foun­tain pens… and com­puter tablets. I’ll be speci­fi­cally talk­ing about the pen­cil here. In­for­ma­tion here WILL still gen­er­al­ize to tablets, but I’ll warn you that you’ll ex­pe­rience some awk­ward­ness tran­si­tion­ing to or from them.

[3] Why start­ing over saves time is a com­plex ques­tion. Part of it has to do with you already hav­ing stud­ied the prob­lem. Part of it is that a fresh can­vas frees you from bias to­wards your old solu­tions. Part of it is that your ex­ist­ing work is sub­op­ti­mal, and you’d need to spend ex­tra time fix­ing it.

[Fi­nal 2017 note: We are now at the abrupt end­ing I warned you about. Sorry!]