Thursday, August 13, 2015

Concretes and Abstracts

Technique-based training is concrete. "He throws a straight punch and you outside block, side step and throw an inside knife-hand strike. Go do a thousand reps." It's easy to teach. He does X, you do Y. Reps. But I can think of zero actual fighters who find this valuable (except as a business model). To deal with chaos you need to train with chaos. And train is the wrong word. You need to play.
Partially because play is the way animals naturally learn, partially because, in a complex system working rote drills hampers more than helps.

Principles-based training involves understanding the principles and applying them in chaos. It's much harder to teach, because knowledge isn't enough, the instructor must have understanding. It's less measurable, less "objective" but infinitely more useful under stress.

Technique repetition may lead to knowledge. Actual experience leads to understanding. Play, if the games are done well, can give you a start on understanding, maybe some insight.

As understanding deepens, you are able to "batch" more and more things. To integrate techniques that seem disparate into single thoughts. As you do so, you process things faster, you become more efficient and decisive.

A technique-based practitioner may go into a fight with a rolodex of forty hand strikes and twenty kicks in his head. He'll try to use this unwieldy mental rolodex and probably get his ass kicked. Memory is simply too slow. Taught in a principles-based way, one level of abstraction up is to understand that striking is just power generation, targeting, and conformation. If you understand it, your rolodex of sixty has become a rolodex of three, with a vast reduction in reaction, action and decision time but an increase in flexibility and adaptability. That's if you understand it. The problem is that if you only know it, you're going into the fight with three mental rolodexes that have to be cross-indexed under pressure. That's bad.

As your understanding deepens, your integrating concepts become simpler and more efficient. In Meditations on Violence I wrote about meta-strategies. Many of the extraordinary fighters I know have complete battle systems that can be expressed in a single sentence. "Destroy the base." "Defang the snake." "Take the center."

Simpler and more efficient, but also, expressed in words, they will seem more abstract. Memorizing techniques is easy. Nice and concrete. Teaching power generation, targeting and conformation is a good size to chunk the information. It gives beginners efficient tools and increases flexibility in hours instead of months. But every so often I want to go really deep, experiment with teaching a workshop on "Structure and Void". I think it would be a really powerful integrating concept, a good framework to teach. But I fear it's too abstract for most people. It would probably only be useful to people with a good depth of understanding already. There are far fewer of those.


Mike said...

Thank you, Rory. I was thinking about exactly this topic this afternoon. The "rolodex" method of "learning" has always rubbed me the wrong way (and not just in SD or MA). It's much more helpful and efficient to learn high-leverage principles. Not that there's no room for details, little jewels that can be unearthed and pointed out by someone with loads of technical knowledge. Just that, in bootstrapping a mental pattern to the next level, principles trump rolodexes. Learning how to think is in many ways better than learning what to think.

Louie said...

I think Bruce Lee feared the man who practiced one kick ten thousand times because about a quarter of the way through, the guy probably got bored out of his skull. The last 7500 was just him playing around, finding ways to make the reps more fun.


Jim said...

Question: can you start purely with principles, or do you have to start first by giving some sort of technique based structure, so that students have a toolbox to work principles within?

In other words... Do you think you can dump all the basic technique stuff like "this is a straight punch; this is an uppercut; front kick" and just introduce ways to generate power (getting the body behind it, aligning the wrist and joints, etc.) and then go to work-play on applying them in motion -- or do you "practice the scales" (music analogy) and use those fundamental sets of techniques under varied situations..

Kim said...

After watching Rory's joint locks video, I've been trying to do more of this with everything I teach. I'm amazed at how much it helped me (with 13 years experience) as well as helped brand new beginners who had never seen a joint lock. Simplify rather than complicate.

Jim - I think you can practice the mechanics of some specific techniques in isolation, but I don't think they're necessary to get students to do things (unless you want them to do those specific techniques). I think that if you teach them a front kick then have them go play, they're going to try to find ways to use the front kick, rather than find what works. If you just help them adjust the mechanics of what they're doing to make it more effective, then it seems to come more naturally.

God's Bastard said...

I wonder how many SD teachers who bang on relentlessly about how "joint locks don't work" ought to more accurately say "I do not know how to teach joint locks in a useful fashion, so I have thrown them out the window".

Brandon said...

I've been thinking that most of the time when people say martial "art" or "system," the word they should really be using is list.

John Thomas said...

I just found the blog, and really appreciate it. Thank you.

Some thoughts here, from an instructor perspective - please ignore at will.

The difference between Concrete and Abstract applies in pretty much every discipline - including engineering and liberal arts, as well as martial arts.

Consider a sketch artist that repeats and repeats to learn how to draw the perfect Mickey Mouse. If they don't abstract those concepts into general principles, they'll never be able to draw another type of cartoon, let alone something as complex as a portrait. An artist who only knows how to draw Mickey Mouse isn't an artist, they're a technician. And an art technician does about as well in an art competition as a martial technician does in a real fight.

So it might be helpful to ask ourselves: Are we martial artists, or martial technicians?

Many Aikido practitioners discuss the concept of "discovering your own Aikido". This is the point at which, instead of repeating the "standard" moves, you've internalized the principles enough to apply them in your own way - including "inventing" new moves on-the-fly. (Although I trained with a number of people at this level, I don't pretend to have even gotten close).

My Tae Kwon Do instructor said that it takes at least ten years to learn how to make a fist. Which sounded odd until, twelve years later, my mind and the muscles in my hand understood the principles in a way my white belt self never could have. There's one middle punch, but the middle punch of a white belt and a master instructor are vastly different, because the latter applies principles the former doesn't even know exist yet.

There are a number of challenges to teaching principles instead of techniques.

One is that principles can't really be described. The Theory of Relativity (in physics) is simple to describe, but it's said that only a handful of people in the world truly understand it. I don't think martial principles are nearly that difficult, but the same general rule applies - being able to describe it does not equal understanding of it.

Another challenge is that most new students can't begin to grasp the principles. So training usually starts with the concrete. Although the true purpose of learning the concretes probably should be as an example to help teach the principles, the step of moving from concretes to principles sometimes gets lost, for exactly the reasons you mentioned - lack of instructor understanding being a large one.

And another challenge is that, even if the instructor does understand them, principles are still hard to teach. That's because they can't actually be taught, they can only be discovered. The concrete, when used properly, can be used to help discover the principles, rather than as an end in itself.

None of those challenges prevent principles-based teaching at all; but they do make it less common than may of us might wish.

Unfortunately, I also think some of the most powerful martial principles are the most difficult to really understand. And I think the more difficult the concepts are to learn (discover), the more likely those arts are to be taught as rote-repetition instead of principle-directed.

From a teaching perspective, this is somewhat self-correcting (students who desire real understanding will seek out and find the instructors teaching the principles - and will become tomorrow's principles-based instructors). But I think it does sometimes make it more difficult to evaluate the contributions of different approaches - especially those that have gained wider recognition (more instructors = more chance some of them use rote-repetition as a stand-in for understanding).

All real understanding comes from learning the principles. But not all instruction is geared toward that. I think that leaves it up to us - as both students and instructors - to focus on the former.

Paul said...

You can have a 'Rolodex' of techniques. Hundreds of them in fact. The trick is to practice diligently and then you just to "let it happen" and not think technique once you see a fight is coming. Your subconscious will pick what is necessary and pick it far faster than your conscious thought.