Learning How To Practice
Practicing is a skill. Learning to learn can really pay off. But it’s hard work, and rarely rewarded.
Background: I trained as a musician for the first thirty years of my life. Music is a fascinating field where there are many orders of magnitude of skill, and you can easily be tremendously better at it than someone else, and yet have many people who are tremendously better than you. (I’m in that middle area, a “talented amateur.”) I only know of a few other areas like this — chess, for example; many sports, due to their competitive nature; some arts; and some mental endeavors like programming may or may not be one, I haven’t decided.
To get better at music, you need the trinity of aptitude, experience, and focus. It’s pretty indisputable that there is some sort of inherent aptitude — some of which may be formed early in infancy, when we’re learning sound and language. Focus is a topic for another day. Experience is training your body and mind to do what you think you can imagine; and the intersection of focus and experience is practice.
Today’s subject is practice how to teach your body and mind to do the thing. I’m going to contrast this with performance, which is when you seek to achieve the main goal of what you’re practicing.
Performance is when you’re graded; it’s where you score real points, earn promotions, face real opposition. Practicing gets you to performance. It’s important to understand that practicing is NOT “performance, but a lot”.
Step one: What Needs Achieving?
It’s tempting to practice “everything”. “I’m going to practice piano”, you say to yourself. But in practice (sorry), it’s important to know what you’re trying to achieve. “I’d like to get Beethoven’s Moonlight Sonata to the point where I can perform it” is slightly better; “I’d like to get measures 72–144 (i.e. the fast part) up to a tempo of 96 bpm with no missed notes” is a much more concrete, measurable goal.
(Side note: this itself is a step along the goal of getting to “performable”: I don’t mean to imply that getting all the notes right is sufficient for a ‘good’ performance, but it’s generally necessary.)
Athletes practice this way. Boxers don’t train for “boxing”, they might train for “punching for a two minute round, solid, with no missed shots”. But most things in our life don’t break down into nice metrics like measures and tempos and perfect placement of fingers. So it’s up to you to find structure in your task, because that’s how you’ll figure out how to practice it.
Step Two: Figure Out How to Make Steps on the Way to Achievement
Got your goal? Good. Now you have to figure out how to trick your body and mind into learning. Generally, we learn best when we stay in the Zone Of Proximal Development (aka ZPD), that region where things are hard but achievable. If it’s too easy, we’re not progressing; if it’s too hard, we’re not learning.
This is an exhausting state to be in. So it’s a state we enter and come out of — pushing ourselves a little, then pulling back a little to “lock in” the learning.
Back to our example. Since the goal is to get the fingers and mind working in concert, we might start with things that are predictable and help forge that connection. Scales are often used for this, though I find that scales aren’t super helpful for me. I prefer something like Hanon’s Exercises, which are repetitive patterns up and down every key signature, and at this point are pretty much rote for me. They get that connection forged, and get your fingers used to the feel of the keys again, the sense of working together, etc. (They also force you to work the two hands in unison, which is another necessary bit.)
There are many stops along the way; I’m going to break them out into their own maxims.
Maxim One: If you Practice it Wrong, you’re Learning it Wrong
A next goal might be to learn the actual notes. Trying to simply play through at tempo is going to learn it wrong: you’re not sufficiently skilled to play it yet, or you would do the thing. With music, once again, there’s a simple dodge: reduce the speed. Playing a piece of music at half or one-third tempo is boring, but it does let you see what’s coming next. (I don’t mean the slow “picking it out” stage, which may precede this depending on your skill level with the instrument; I mean actually _playing_ it at a tempo which is vastly slower than the target tempo.)
Another trick is to play it in slanted tempo: play it with dotted rhythm, slow-fast doubles, like swing. Then play it with reverse-dotted rhythm, fast-slow. Congrats! You’ve just played every pair of notes “fast”. Do it enough, and you can play it all fast.
Another trick is to play a subset: just the right hand, just the left hand. Just the melody line, but using the fingers you’ll use when you put it together. Just the first two voices.
The goal here is to simplify enough that you get it all right. If you get it wrong, you’ve just played it wrong, and you’re teaching yourself the wrong version.
Note! This is very challenging! It’s easy to get fatigued and give up, and say, dang it, I just wanna play. That’s okay. But it’s not practice, and will set you back if you want to improve at the thing.
In sports, this sort of mock practice can be seen in things like slowly practicing a golf swing, or pre-visualizing what you’re going to do. (Some of this is practice for body; some for mind; some for both together. Those are each separate steps.) A while ago I thought up this idea for an exoskeleton that you wear, which limits your motions to the “right” ones for a golf swing or a baseball pitch, literally forcing you to do The Right Thing; it sounds like this idea is finally, 15 years later, becoming reality.
Without such a thing, you have to be the exoskeleton for yourself. Do it right, every time. If you’re not doing it right, slow it down. Start again. Clear your mind of the wrong.
Maxim Two: Pareto-optimality is Insufficient: Focus on the Hard Parts
Pareto-optimality is sometimes misquoted as “good enough, everywhere” or “a 90% solution is better than no solution”. (It really means “the state where you can’t make one thing better without making something worse”.) But we’re not in that sort of zero-sum game here. We want “better, everywhere it matters, and no worse in the rest of the places.”
So the next thing to think about is where you need practice. “We don’t practice the easy parts,” a teacher once said to me, when I complained that practice was hard. This has to be filtered through the ZPD and exhaustion, of course, but in music, people will forget the 999 notes you got right and remember the chord you blew. Another music teacher: “In baseball, if you bat .500, you’d be the best ball player of all time! In music, if you miss half the notes, you wouldn’t make it into the audition in the first place, and even if you get 99% of them, you’re still not going to get through it.” Music requires a certain kind of perfection as a starting point.
(Side note: there are some types of music where this is adamantly not true; some types of jazz, notably, recast certain mistakes as ‘creative exploration’. Nevertheless, there are many categories of mistake which are equally disruptive as a wrong harmony in a Beethoven sonata, and being good at improvisational jazz is about making the right sort of mistakes, and ameliorating the impact of the wrong sorts.)
Programming is a bit like this; any disharmonic mistake is a bug. We ship software with bugs in it all the time, of course, and every one of them has a chance to cause serious problems.
Maxim Three: Make Things Locally Hard, but Easy to Judge
This falls out of the above. To get better, you have to break out of a local maximum of skill by forging into the wilderness and challenging yourself. But it’s hard to see if you’re making progress if all you can see is trees.
In music, a traditional way to judge yourself is with a metronome. It keeps a constant time, which is better than nothing — on your own, the tendency is to slow down for the hard parts and speed up for the easy parts. (Unless you have perfect timing; I have a drummer friend who has such. “Hey, Russell, 72bpm? Thanks.” It took him many years to cultivate, including a stint Trained By A Master at some French school where they literally just drummed all day.)
The metronome gives you constant feedback, once per beat, about where you are with respect to your goal. You can adjust it easily, up or down. As you practice, you adjust the tempo to adjust the difficulty and keep yourself in the ZPD. You can use it as a unit test against your progress.
Well, the combination of the metronome and your ear; you still have to recognize your performance as sub- or up-to-par.
In professional sports, this job falls to the trainers. They watch your swing; they time your lap; they give feedback on whether you’ve lost form after 1m 52s of solid punching. In chess, one way to approach this is to put yourself in situations from past masters, then force yourself to continue the game from there — and then compare your solutions to theirs, giving you feedback on what others have thought about that situation.
Maxim Four: Quantity Has a Kind of Quality
Our bodies and minds are made up out of neurons that require repitition to adjust. (There are exceptions, where a very strong signal causes us to learn something immediately — think food poisoning — but those generally aren’t things you practice.)
So do it a lot. A lot. This is where the myth of “10,000 hours of practice” comes from. Highly-skilled people are finding ways to practice all the time. Professional musicians may devote 4–8 hours EACH DAY to keeping at the top of their game. Writers talk about writing thousands of words a day to keep in practice. We, the mere mortals, might not have to go to such extents, but we do need to practice more than might be obvious.
This is super challenging in a white-collar setting. Most “professional” jobs contain exactly zero opportunities for practice: you’re expected to perform, all day every day. So it is up to us to carve out practice time, from work time when work earns a benefit from it. More clever organizations are realizing that this is necessary, but most are still in this weird Capitalist loop of “hire practiced people, force them to perform 100% of the time, then fire them when ‘somehow’ their skills atrophy/don’t progress”. It’s kinda bullshit.
This “repeat the hell out of it” is what I’ve been taught to call woodshedding, the idea being that you take your instrument and go out behind the woodshed where no one else will be bothered by you playing the same four measures over and over again at slow tempo for two hours. (Not to be confused with bike-shedding, where the uninformed have opinions that derail the central conversation.)
Shedding is also where you start to learn your instrument, by which I mean yourself. It can be a kind of meditative practice in which the actual thing you’re practicing goes away, and you learn what is between you and it. It lets you see what you bring to the situation, and seek to improve on it.
Shedding is HARD. That leads us to the next step.
Step Three: Figure Out When to Stop, Rest, and Reflect
Practice is hard work. If it’s not, you’re not practicing; you might be reinforcing, but you’re not improving.
As a result, it’s easy to exhaust yourself practicing, and then you start taking shortcuts. You start doing it wrong. You’ve stopped learning, and in fact, you’re unlearning — you’re undoing the hard work you just put in! It’s time to stop and catch your breath.
This point is different for different endeavours. When trying to practice for cardio, say, for a marathon, it’s crucial to push yourself past what your body tells you is “enough”. The learning process — the practice — includes improving your ability to judge the point between “I can’t do this any more (it’s so hard™)” and “I *CAN* *NOT* do this any more (or I will hurt myself)”.
This stopping-point-decision can be a very tricky for non-physical endeavours; we’re not trained to recognize it, and in fact receive lessons all the time that we’re super human and should be able to do it for infinitely long.
[Far Side]Teacher, can I be excused? My brain is full.[/Far Side]
So go ahead. Rest. Weightlifters know that their bodies need to heal — traditionally, you have “leg days” and “arm days”, and alternate; as you get older, your muscles need more time to heal the microtears, and so this might become a three- or four-day cycle.
But! The downtime is not just for rest. It’s for rumination. After you’re not quite so blasted by the practice, as soon as you can, you need to assess what happened during practice. Did you make progress on something? Hurrah! Put it in the “lock it down” pile for later (see step four). Was something unexpectedly hard? Put it in the “woodshed” pile. Is something confusing? Break it down into smaller pieces. Something has come together? Lock it down.
Step Four: Lock It Down
Once you’ve got things in a good place, it’s time to lock it down: get it inside you so it can’t escape. We’ve all had the experience where we cram for a test, pass the test, then can’t remember whatever we spent four hours learning. That’s because we didn’t lock it down, and it got away.
Locking it down is a different sort of practice. Instead of challenging yourself to learn, you’re challenging yourself to integrate that learning into the larger body of work that constitutes “you”.
Some of this is achieved through Mere Repetition (more woodshedding); some is achieved by integrating the work into what surrounds it. You might play the pieces from a few measures before the hard section, and from the hard section into the next section, to practice on the transitions. You might luxuriate in playing the whole thing, if it’s “good enough” and start thinking about the next layer you’re going to learn (for music, traditionally, this is where ‘interpretation’ comes in).
In sports, this sort of integrative work is seen in more complex routines (e.g. kata or long weapons forms in martial arts), in sparring or scrimmages, etc. These are highly stylized spaces where practice can be performed without the actual rigors and stress of performance — you’re still in “practice” mode — but they provide a way to integrate the words and sentences you’ve been teaching your body into a full-fledged statement.
This is the actual goal of business practices that usually get relegated to the BS pile, like “retrospectives” and “hansei” and “performance reviews”. The ideal goal is not to judge performance, it’s to lock in the learning and achievement (as well as do some of the “ruminate” part of step 3). Sadly, in corporate America, these steps are usually used to punish, shame, or otherwise derail any attempt at learning.
Finishing Line: lather, rinse, repeat
The great thing about learning is that it never ends. Every time you learn something, conquer a bit of possiblity space, you realize that you’ve just increased the surface area of things you could learn, could improve. (With rare exception. Sometimes you’re filling in holes. But those are usually times when you can abstract away an entire nugget into a whole that you can use for other things. “Okay, I understand how to construct a major and minor scale in any key! Now I can ‘play scales’. What can I do with that knowledge?”)
You’ve learned all the hard parts, by identifying them, isolating them, woodshedding them, and assessing your progress. What you’ve got, you’re going to keep, thanks to locking it down. You’re rested and think there isn’t anything left in your original goal. Woo! Celebrate! Perform! Dance in the streets!
Then get back to practice, because there’s always more to do.
Conclusion
And that’s it. Four steps, with many ways to achieve them. Four hard steps which require challenging the ways we are taught to learn, and contradict the way that most of us are rewarded.
No one said it was going to be easy. But it’s better than spinning your wheels, lying to yourself about measure 76 and that tricky run in the left hand that you mumble your way through, knowing that you could get it if you just spent a few hours teaching it to your body and mind.
