Coaches should be ambiguous

[hyperpape]

Remember how Mr. Miyagi taught The Karate Kid how to fight? Wax on/Wax off. Paint the fence. Don’t forget to breathe. A coach is the coach because he knows what the student needs to do to advance. A big problem for coaches is that the most precocious students also (naturally) think they know what they need to learn

More where that came from. This one is sure to provoke some debate...

[LocoRon]

Yes, because movie logic is infallible.

There are a lot of other great examples of ambiguity proving so effective. The main one coming to mind is Master Roshi's amazing training method: Wear a 50 pound turtle shell while delivering milk, and of course fighting off the island's native wildlife (eg, sharks and dinosaurs).

But really, I think the most effective learning/teaching strategy isn't ambiguity... it's the montage.

[Phoenix]

The problem with ambiguous training is neurological. Wax on and wax off ten thousand times and you'll be really, really good at waxing. And the next time someone tries to punch you in the face, you'll come out of it with a nice black eye.

Why? The brain simply hasn't made the association. The purpose behind the movement you've practiced 10,000 times is to wax on/off. This is something I've actually encountered in the martial arts as a sanctioned training method; repeat the same movement a million times without knowing its purpose, then never be told why you did it. Of course when we needed it, it never came out.

So the idea here is:
Practice a movement repetitively ---> the technique will be there when you need it

How reality works:
Practice a movement repetitively ---> associate the movement with its function ---> the technique will be there - albeit sloppy

Or better yet:
Learn a technique and its function ---> practice the movement consciously and in the right context ---> never get punched in the face again

I feel the same applies to everything. In order to learn a new technique/sequence, it's important to know its function. Otherwise you gloss over some Go material, read something like "When attached, extend!", then go on to do these things mechanically and lose all your games, and wonder if what you really need is to pay Guo Juan for lessons to get past 9k.

When I teach a sequence or elaborate on a concept, I always give the student concrete examples of how this thick group helped make territory all the way in the opposite corner, and exactly why the order of moves is just so. This way they don't get creamed when someone goes off-joseki (well sometimes, but don't we all?), and they can use their skills to play similar sequences in similar situations.

Best of all, their fundamental understanding and Go strength grow exponentially faster than with the Miyagi method.

[Javaness2]

I don't think that you can become stronger at Go by being lazy, so what's wrong with a bit of ambiguity in the right places?

[Polama]

There's an interesting history to the idea of hiding a martial arts move in an unrelated action or sequence.

The concern was that it's actually too easy to pass on your techniques, at which point you've got nothing over the student. They might leave and start a competing dojo. They might best you and prove their superiority (you've got the experience, but they are in their physical prime...). They might use their fighting ability in troublesome ways and dishonor you. You want them to improve and learn your techniques, you just want some time to make sure you can trust them (or earn your pay).

So you teach them some convoluted or seemingly pointless action and have them drill that into their muscle memory. Most of it is garbage, but some motion within it is useful. Maybe you take a sequence of moves, and teach each motion as part of a separate drill. Then, when you're finally ready to teach your secret techniques, the parts are all trained up. They already know the motion to block a punch, they've just got to train that action into an instinct for the right triggers.

[Phoenix]

So you teach them some convoluted or seemingly pointless action and have them drill that into their muscle memory. Most of it is garbage, but some motion within it is useful.

I've practiced a few martial arts, and by the time I got to do Kung Fu, where this method was used wayyyy too much, I knew too much for it to really be subtle or obscure.

I remember we did one of these series of movements hundreds of times. In one class, we started sparring and my opponent stepped up and punched me. I grabbed his wrist, did the movements, and he ended up on his back on the floor. This is approximately how the conversation went right after.

"What the hell was that?!?"
"It's the stuff we've been practicing."
"So it's to start a throw?"
"Actually, it's to break your elbow, crush your windpipe and then break your spine on the ground. You're welcome."

There were times when other sequences would come up and everyone would ape the movements. Afterward I would talk to the instructor and ask him if the technique was designed for such or such purpose. He would scowl at me and tell me not to say anything to the others.

I knew for a fact that they were wasting their time, though. Without the right triggers, as you say, the technique will simply not be there when you need it. Even if you connect the dots later, it won't be anything but gross motor patterns. In a fight, where a small mistake can be the end of you, I much prefer knowing what I'm doing and why, so I can have the right response at the right time, and be able to adapt to the surrounding factors.

Just my personal philosophy of staying alive.

[Amelia]

Phoenix, that's very interesting!

On your philosophy of learning, I agree absolutely. Only when you understand what you're doing do you truly learn efficiently.

"Actually, it's to break your elbow, crush your windpipe and then break your spine on the ground. You're welcome."

However, may I say, I'm glad not everyone who practices martial arts gets this right away

[ez4u]

I don't think that you can become stronger at Go by being lazy, so what's wrong with a bit of ambiguity in the right places?

Well, put that way, what's right with ambiguity in the wrong places? As mentioned above, the basis for resolving the ambiguity of the right versus the wrong places is... what works in a kids movie? Let's just skip all the hard work. Show me that radioactive spider that will instantly give me super powers!

[tapir]

"Actually, it's to break your elbow, crush your windpipe and then break your spine on the ground. You're welcome."
...
Just my personal philosophy of staying alive.

How is prison?

You must be serving serious prison time if your philosophy involves murdering everyone who tries to hit you.

[Splatted]

"Actually, it's to break your elbow, crush your windpipe and then break your spine on the ground. You're welcome."
...
Just my personal philosophy of staying alive.

How is prison?

You must be serving serious prison time if your philosophy involves murdering everyone who tries to hit you.

On the subject of selective quoting.

You must...murder...everyone

[Phoenix]

"Actually, it's to break your elbow, crush your windpipe and then break your spine on the ground. You're welcome."
...
Just my personal philosophy of staying alive.

How is prison?

You must be serving serious prison time if your philosophy involves murdering everyone who tries to hit you.

Very funny.

I've been taught movements in the martial arts with the purpose of seriously harming others. I'm really not sure why this was important to some of my instructors. My personal approach to self-defense is the same as I've been taught: fight only when you're in danger and only to the extent necessary.

I've been in dangerous situations where my skills might be needed and every single time I've managed to defuse the situation without violence. This includes an attempted mugging.

I just like to be clear.

[EdLee]

every single time I've managed to defuse the situation without violence. This includes an attempted mugging.

Phoenix, congrats -- that's excellent. May I ask for more details of the attempted mugging and how did you defuse it?

[Polama]

In college I studied math/CS, but my friends were all humanities types. Which meant I ended up helping a few people get through the required Caclulus I class. Invariably, the core issue for them was that the class would start out by jumping back and forth between the formal definition of derivitives (the limit as h goes to 0 of (f(a+h)-f(a))/h) and the simple algorithms you can use to solve derivitives when you don't care about formality (multiply by the exponent and reduce the exponent by one, e.g.) In the end, all they needed for the tests were the simple algorithms, but they were lost in these quasi-methods somewhere between the two. In about an hour I could start them over and just focus on the simple algorithms and what they meant in terms of a graph and get them solving equations they had zero understanding of before. It was, in a sense, the two extremes discussed here: ambiguity (not unrelated information in this case, but more information and jumbled information) and simply providing just the information useful to them in an approachable manner. Once they had a basic understanding I could rattle through the rules for them (how to derive a logarithm, a fraction, a power to x ...) and they could use them and pass the class.

On the flip side, I doubt any of them could still take the derivitive of a simple equation. I hadn't been tutored, so I had to fight through making the interesting connections of Calculus: why the simple algorithms and the limit formula said the same thing, what the intuitive relationships were between derivitive, slope and velocity, etc. I put a lot of thought into how Calculus related to the real world and the equivalance of equations and graphs. And because of that, I could distill the information down into something somebody else could use, and it's still with me today.

Which is a round about way of saying you can go overboard in either direction. Adding ambiguity for ambiguities sake isn't helpful. But you can go to the other extreme of spoonfeeding everything: this move is good in this situation because of this, this one is bad because of that, and the student can remember your instructions and improve. But there's a lot to be said for not showing them which moves are bad, and letting them learn how to figure that out for themselves. Give a man joseki and he won't let all his stones in the corner die. Teach a man to discover joseki, and he won't let all his stones in the center die either.

[Bill Spight]

Give a man joseki and he won't let all his stones in the corner die. Teach a man to discover joseki, and he won't let all his stones in the center die either.

There is much to that. As I have mentioned, the first joseki book I got was about joseki mistakes. To understand joseki you have to understand why non-joseki are not in general good.

But discovering joseki is beyond the power of the individual. That is why go history is cluttered with obsolete joseki. Patterns of play that were accepted as reasonable are refuted, or come to be seen as not so reasonable. (Actually, I am surprised how many ancient joseki have survived. The Mini-Chinese goes back centuries. )

[Phoenix]

May I ask for more details of the attempted mugging and how did you defuse it?

That one was a matter of psychology. Assaulting someone doesn't come naturally, especially when unprovoked. There's a whole primal system that kicks in which forces us to avoid fighting when we can. Otherwise with our short tempers and irrational minds, we'd be extinct as a species.

Anyway, a mugger has to work himself up to the mugging. 'Pump up' psychologically, get the adrenalin flowing, etc. I was walking across a bridge when the man came out from cover at the end. He had a hand in his pocket and looked dead serious, and there was no one anywhere close. I just knew instinctively what to do. I had all the training and knew I was in no real danger. So I showed it.

I relaxed, held my head high, looked him in the eye, smiled confidently, and kept walking straight. It was sort of funny, really, because at that moment he took a step towards me, then froze, glanced around nervously, did a little back-and-forth dance of indecision, then just retreated back and let me pass untouched.

Put yourself in the other person's shoes and things aren't so scary.

I put a lot of thought into how Calculus related to the real world and the equivalance of equations and graphs. And because of that, I could distill the information down into something somebody else could use, and it's still with me today.

Which is a round about way of saying you can go overboard in either direction. Adding ambiguity for ambiguities sake isn't helpful. But you can go to the other extreme of spoonfeeding everything: this move is good in this situation...

[/quote]

This is the kind of thing I do as well. Ambiguity isn't such a bad thing IMO as long as it serves the purpose of learning, instead of simply obscuring information. In Go we talk about initiative, influence, thickness, strength of groups, etc., as a way to cope with the thousands of variations we simply can't read in a whole- or even part-board setting. Concepts are one way to be ambiguous while being helpful. They're a form of metaphor, and the mind is very metaphorical.

Going through and distilling the learning process into something that is instinctive and easily teachable is part of what makes a great teacher great. We need more people like you in the Go world, and in other disciplines as well.