Friday, October 30, 2009

Another blind alley: the ITF “sine wave” theory

I have indicated before that it is dangerous to be obsessed with “power generation” in your martial art; there is so much more, particularly if your approach is oriented towards civilian defence. You don’t want to get hit, for starters. To do that you need not only evasion but “blocking” (deflection). You need a good foundation in terms of grounding and movement – in particular movement that serves as a platform for your evasion and deflection. Of course, you also need skills relating to grappling – whether stand-up or on the ground.

Even when you want to focus on hitting, it is important to remember that what people call “power generation” (ie. the ability to impart force) relies principally on the efficient transfer of momentum (as I’ve discussed in my article “Hitting harder: physics made easy”). The equation for momentum is simple:

p = m x v

If you want to hit harder, you have to move your mass faster. There are glosses on that (as I’ve detailed in my articles “Kime: soul of the karate punch” and "Telegraphing vs. staged activation") but the above formula, in essence sums it up.

So when I see people basing their art around “power generation” I start to become suspicious. I become doubly suspicious when their means of “power generation” focuses on one small variable in the scheme of making you hit “harder” – a variable that then proceeds to overwhelm the entire art.

I’ve mentioned before that I’m not a fan of “koshi” – the pre-loading of hips on every single movement in a kata/form. Ostensibly this is done so as to maximise “power”. What it does, is it ignores the multitude of other ways in which your body, moving as an integrated whole in a dynamic context, imparts greater force. As Marc “Animal” MacYoung says, there is much more to "power generation" than hip use. Sure, the hip is important, but it isn’t everything.

I’ve mentioned before how the Australian Institute of Sport has determined that force applied by a punch is divided as follows:
    15% comes from your arms
    60% comes from your hips
    25% comes from the momentum of your whole body as you lunge or step forward.
Then there are the twin issues of (a) taking extra time to do the pre-load and (b) "telegraphing" your intention...

So now we come to another “blind alley”; the “sine wave” theory of ITF taekwondo.

As an abstract theory it seems to have some merit: General Choi noted that the body in movement (say in walking or running) follows a natural rise and fall motion much like a sine wave. Given that the momentum of your body as you lunge or step into your opponent accounts for 25 percent of your applied force, it seems to make sense to make the most of that by utilising natural body movement. On some movements you can use the “push off” from the ground to impart more force, while on others you can use the “fall” of gravity to add to force.

But what is good in abstract theory falls down when you look at the context in which it is sought to be applied.

The video below is a classic example: the practitioner of the form doggedly tries to apply the “rise and fall” in each movement.

An ITF practitioner performs a pattern with the trademark "bounce" of the "sine wave" theory

What you end up with is, quite frankly, a farcical mess. It is a triumph of dogma over common sense. I keep wondering when someone will shout out “the Emperor has no clothes!” If one had to show this video to someone with no martial arts background they might very easily fall over laughing. Yet all over the world thousands upon thousands of ITF practitioners labour diligently each day trying to force this “sine wave” motion into their patterns.

So what’s wrong with it? For a start it suffers from the same problem as “koshi” – it disconnects movements in the pattern/kata into little distinct parcels, each one separately focusing on “power generation” while ignoring other combat needs, including that of contextual movement and flow.

In this regard, I often wonder facetiously whether “sine wave” and “koshi” could be combined to produce the reductio ad absurdum of focusing on “power” at the expense of everything else. You can just imagine someone bobbing up and down while shaking his or her hips – a gyrating, pulsating train-wreck.

But there is an even more profound problem with “sine wave” theory and that goes to the range in which it is sought to be applied.

Yes the body does go “up and down” while running or walking. But this “rise and fall” starts to be discernible once you have travelled over 2 metres (6 feet). When you are fighting in the “melee range” the distance you travel is in the order of one metre (ie. 3 feet) or less. In the context of, say, lunging toward your opponent with a punch there simply isn’t any relevant “rise” or “fall”. Rather, you want to minimise any “rise” or “fall” tendency because you want to move in a straight line – the shortest distance between 2 points. There simply isn’t time for you to indulge in a circular path to your target. In attack you must move with the greatest economy.

Indeed, one of the principles of shotokan karate (from which ITF taekwondo evolved) is to practise “level movement” – ie. to keep the same height while stepping in a low forward stance. “Sine wave” practitioners argue that this is unnatural – and indeed it might appear to be when one is stepping up and down the floor in class. But the error in the criticism lies here: In their stepping exercises, shotokan karateka are not training to cover the length of the floor. Rather each step is a separate attempt to cover the distance of one metre or so. Each step is separate training for movement in the melee. Doing steps one after the other just allows the karateka multiple opportunities to train that single, lunging step. The idea isn’t to get from one side of the dojo/dojang/kwoon as quickly as possible. It is to make each separate step as quick as possible.

Think of a runner at a starting block: he or she crouches down and then blasts off with the starter gun. So far so good – the sine wave theorist would argue that the runner is “pushing up” (I would argue that he or she is pushing forward, but that is another issue). However it is what happens after the initial push off that is interesting. In first metre or 2 the runner takes lunging steps. It is only when the runner has passed that one or 2 metre mark that the runner rises up to a “cruising speed” where his or her gait falls into a “sine wave” pattern. Between the initial push off and the reaching “cruising speed” the runner is lunging – and staying level.

I've found that if you isolate the "lunge" aspect of the initial take off, and put it into a walk, you get the low-slung walk effected by Groucho Marx in his movies. I actually use this “Groucho Marx” walking as a training tool. I get my students to do it occasionally instead of moving up and down the floor in forward stance (which is really a more formal way of doing the same thing). What I have found is that over one or 2 metres the lunge is unbeatable in terms of speed. From a standing start, you simply can’t move faster over a shorter distance. Level, lunging movement is perfect for the melee range. The "sine wave" motion of "post take off" walking or running is, on the other hand, irrelevant to this range.

When there is a rise or fall in the melee, it occurs naturally as the technique you're executing requires. For example, you might drop into shiko dachi (horse stance) to effect a deflection or control or to absorb a push. You might rise out of a low posture to initiate a counter. But you don't "bob up and down" with every punch/strike/kick.

The "Groucho Marx" walk - a low-slung walking exercise to condition and develop lunging speed for melee exchanges

So what has prompted this “sine wave” trend in ITF taekwondo? I think it comes right back to “faux boxing”. When you spend your time bouncing up and down, well out of melee range, I guess it makes sense to incorporate bouncing into your patterns/kata. But if this isn’t an example of “diluted knowledge” then I don’t know what is. Rather than go back to examining techniques suitable for the melee, the sine wave “development” takes workable civilian defence techniques further down a blind alley.

To summarize, the melee affords you very little time. You have to move along the most direct path possible. Every movement must be contextual - ie. there because it has to be. Certainly every kata/pattern/xing has some movements where you rise and others where you drop. But you can't afford to force an artificial "rise and fall" into each kata move. As with "koshi", you can't afford to introduce extra movement just so as to "gain power"; apart from the time required to do the movement, you also "telegraph" your intention.

The sine wave theory of movement is an abstract ideology that ignores these melee dynamics. It loosely fits sports taekwondo fighting/sparring where people “bounce” before engaging. But is starkly at odds with the physics of the actual engagement (which is where civilian defence should be focused).

The precious time wasted, and the intention telegraphed, in artificially forcing an arc where there is no time/space for it is perhaps why you will never see a pragmatic MMA fighter “bobbing” up and down (or, for that matter, wobbling his hips) while fighting “toe to toe” in the melee.

Copyright © 2009 Dejan Djurdjevic

Wednesday, October 28, 2009

The role of traditional stances

It has occurred to me that I have not ever addressed one of the major elements of traditional fighting arts before: stances.

Stances are an integral part of traditional eastern fighting arts. Moreover they are surprisingly consistent in form.

For example, virtually every eastern fighting art has the “forward” stance (sometimes called the "bow and arrow stance"). In Japanese it is called “zenkutsu dachi”. In Chinese it is called “gong bu” (work step). This stance is the “workhorse” of most martial systems.

Typically it is shoulder width between the feet but 2 shoulder widths in length with the front knee bent so that the shin is vertical while the back leg is straight. The hips are usually oriented forwards (hence the term “forward stance”). It even exists in yoga where it takes the form of the “warrior pose” indicating its martial links.

Another ubiquitous stance is “horse stance”, which is typically bow-legged with the feet 1 ½ to 2 shoulder widths apart, toes either pointing straight forward or off at 45 degrees. It is common to most northern (and some southern) shaolin schools of gong fu / wushu where it is known as “mabu” or “horse step”. In karate it takes the form of either the kiba dachi from naihanchi or the shiko dachi from Naha te. The latter is also used in the Japanese art of sumo and various Indian wrestling disciplines.

Then there are the myriad other stances, most of which have their counterparts in different systems. These include the cat stance (“neko ashi dachi”) and the pigeon-toed “battle stance” (called “sanchin dachi” or “zhan bu / chien be”). Both the latter stances are to be found in karate as well as white crane and other Hakka systems of southern China (for example the art of ngo cho kun / wu zu quan or “5 ancestor fist”). Other common stances are the “back stance” (“kokutsu dachi”) which has its counterpart in the san ti posture in the internal art of xingyi.

That stances should take a common form in different martial arts is hardly surprising. The human body can only be put into a limited number of functional postures.

What is surprising is that they should all use stances in the first place, when modern combat sports don’t have any similar “formal” footwork.

So why did different martial arts traditions develop such similar "formal" stances? What purpose do they serve?

I’ve heard many a combat sports practitioner ridiculing traditional stances with the argument “no one fights like that”. One kickboxer told me that if I were to assume a forward stance he would simply kick my front knee and break it. Indeed.

What this particular fighter wrongly presumed is that stances are fixed postures that one might assume and hold (perhaps as an "on guard" posture). This is not the case.

I have previously written an article outlining the differences between taijiquan and yoga. Where both are often lumped together under a “new age umbrella” they really are very different in goal and function. Most relevantly, taijiquan, being a martial art, does not have fixed “postures”. It has stances, but unlike yoga these are not postures to be adopted in a static mode. The same applies to an art like karate. The stances are points of transition; they are snapshots in a continuum of movement. Yes, they are often held statically in training for the sake of conditioning. But this is an exercise only.

In application they are positions of extension (a lunge into the forward stance) or contraction (coiling back into cat stance) or steadying postures (dropping into horse stance to take a load, as per sumo). The traditional martial artist will flow through these postures in a dynamic way, never pausing. A camera taking stills will however be able to pick out individual points where the stance “was held” – however fleetingly.

But why practise stances? Surely a good workout can be had without adopting what seem to be overly formal, classical postures that many argue are akin to dance steps? Why not spar – and go through all the extensions, contractions, steadying etc. without learning formal stances?

The answer lies in:

(a) the need to develop kinaesthetic awareness; and
(b) the importance of muscle-specific exercise.

In the first case (kinaesthetic awareness), by training in stances you give your body a framework of known points in time and space. Your body becomes aware of its positions of stability and strength and its positions of imbalance and weakness in each formal stance. Between all the various stances the full “picture” is covered. Every other “informal” posture can be seen as an amalgam of various stances or more relevantly a point of transition between stances.

Accordingly, by training in traditional stances a martial artist gets to know where and how to step to maximise balance, transfer momentum and impart force. The fixed points of the stances are akin to triangulation in isolating the origin of a radio or other electronic signal; once you have certain variables the body will, through extrapolation, “fill in the blanks”.

The flip side is, as I have said, muscle-specific exercise. Lyoto Machida has demonstrated just how important stancework is to fighting. He is one of the few MMA fighters who is able to dart in and out of what I call the “melee range” with astounding speed and precision. His ability to “reset” himself while his opponents are left wrong-footed is not lost on commentators. Looking at his pre-MMA karate sparring style (which relied heavily on traditional stances), I’m not surprised that he can do this. Timing issues aside, very few people have the leg muscle conditioning to “explode” into action the way he can and does.

Nor can weight training and other exercise replicate the effect of stance training. It just doesn’t provide the same conditioning. Sure, weights are an important part of any athlete’s workout. But ask a gymnast whether there is any substitute for handstands or working on the rings… If you want to be a gymnast you have to do those activities and no amount of other weight-bearing exercise will do.

Some will argue: “Surely traditional stances are nothing like fighting stances – even the lunges don’t necessarily look like a forward stance?” Look again. A forward stance, as I’ve frequently argued, is just a boxer’s stance elongated with the back knee straightened. Virtually every traditional stance is just a lower, harder to hold, more “formal” manifestation of a move you will make in sparring.

Just as Roger Bannister was able to break the 4 minute mile by increasing "load" (he ran up hills and with weights in a backpack), so martial artists hold lower and deeper stances so that higher ones become easier.

And just as gymnasts rely on core strength muscles to assist them, so stance-work (particularly learning to move from stance to stance) helps the traditional martial artist work the core muscles needed for stability in combat - particularly in the melee range.

I mention above that stances are more "formal" manifestations of postures you move through in sparring - but this is particularly so if you are using traditional melee range strategies such as tenshin (evasion) or taisabaki (body movement), combined with traditional “blocks” (deflections) and strikes. The latter tactics rely on effective, solid grounding and lithe and decisive footwork.

So stances are indeed the metaphorical as well as physical foundation on which traditional techniques for civilian defence are built.

Copyright © 2009 Dejan Djurdjevic

Thursday, October 22, 2009

The "naihanchi stance"

Readers of my blog will be aware that I am at odds with many karate practitioners in relation to how the kata naihanchi/naifanchi should be performed.

I have previously detailed my dislike of "hip shaking" - ie. pre-loading or telegraphing the hips to gain extra power - in practically every technique in naihanchi/naifanchi.

However I have recently become aware of another point of disagreement I have with many schools over their practice of this kata: the stance.

Many schools today practise naifanchi in what amounts to a relaxed, shoulder-width stance, with feet parallel. This is known as "heiko dachi" in Japanese. Even if it isn't exactly a heiko dachi, it is very near it (it certainly isn't a horse stance which is one and a half to 2 shoulder widths between the feet).

Consider, for example, the video below:

Naihanchi shodan by Onaga Michiko - performed in what is, to all intents and purposes, a normal shoulder width stance

I presume that the basis for this variation is the feeling that more modern incarnations of mainland Japanese karate have lengthened and deepened stances for training and aesthetics at the expense of functionality. This is particularly said of the innovations by Funakoshi Yoshitaka, son of shotokan karate founder Funakoshi Gichin.1

That Funakoshi Yoshitaka did lengthen and deepen karate stances is beyond doubt: an examination of photographs from the early 20th century shows karateka in higher and narrower stances.

Stills of Funakoshi Gichin performing naihanchi or tekki shodan

Not only did Funakoshi Gichin himself use higher stances, but so did his rival Choki Motobu (see the picture at the start of this article). In the case of naihanchi kata both men adopted a stance which was considerably narrower and higher than that seen in modern shotokan, for example.

But was this older stance a heiko dachi (natural stance, shoulder width, knees largely straightened)? I think the answer is definitely "no". As far as I can tell, naihanchi has always been characterised by a distinctive stance: a horse stance with the feet parallel, often called "kiba dachi" - sometimes "naihanchi dachi" - to distinguish it from the "shiko dachi" (a stance used in naha te as well as arts such as sumo, where the feet are angled outward rather than parallel).

Kiba dachi is, and always was in my view, the stance known in China as "mabu" - literally "horse step". Now it's true that Funakoshi and Motobu both had a narrow kiba dachi/mabu - however I see this is as within the usual range for horse stance.2

As far as I can tell, naifanchi was designed to be performed in a horse stance. It appears to be the case historically (see any old photos of naihanchi, for example) and I would argue that it is also the case pedagogically. It is not a heiko dachi kata. When I see it performed with feet shoulder width (heiko dachi) I can't help but think that someone has missed the point.

Kata were, in my view, designed for training and putting the body under load. This might not involve the very deep stance depicted earlier in this article, but it involves a load of some kind nonetheless. Horse stance isn't a relaxed, high stance of the kind one might adopt when waiting in a queue.

In this regard take careful note of Motobu's stance at the start of this article and see how low he is sitting; there is nothing "easy" about this stance. It requires considerable leg strength just as any traditional or formal stance does. This is quite different to heiko dachi, which has virtually no knee bend at all (and accordingly requires and generates no conditioning).

But what of the argument that stances should be "natural" for fighting purposes?

Motobu was himself a very practical man who insisted on functional fighting techniques - yet his naihanchi still featured a stance which can be described as anything but "natural" or "relaxed".

The problem with the argument that stances should be "natural" lies with the definition of of that term. If by "natural" one means "non-injurious" and "biomechanically sound" then I agree - all stances should be "natural". However if "natural" is interpreted as meaning "the way one stands in ordinary day-to-day discourse" then I disagree entirely. If this were true, then we'd be doing kata every day - as we walk to the bus stop or when we stand talking at a barbecue.

The problem with the latter definition is it confuses training with fighting. It also sees stances as fixed postures rather than snapshots in a continuum.

And I want to make an unequivocal statement that there is no danger in people "fighting in stances that are too deep". This argument is often raised by combat sports practitioners to criticise traditional forms, but it holds no water at all: When you fight you will naturally rise to a more relaxed posture - however you will be conditioned to lunge and deepen where necessary (the "snapshots" to which I refer above).

Kata aren't meant to look like fighting. They are training drills for conditioning the body. Accordingly stances don't have to be "real" - they can and should be lower so as to add "load", as discussed earlier.

Moreover, if you can do something in a low stance it is easier in a high stance.

I look to the Chinese arts for their mabu - horse stance. It isn't as low and wide as that seen in modern shotokan, but it isn't quite as narrow as Funakoshi's or Motobu's kiba dachi (it is just a tad wider than their's). In particular I recall my instructor teaching me a form designed by Hong Yi Xiang of Taipei - Da Peng Zhan Chi (see below) and telling me to use "mabu" - a stance slightly narrower than kiba dachi. My experience with Chen Pan-Ling's various arts is consistent with this.

Da peng zhan chi - a form designed by Hong Yi Xiang which features "mabu" or horse stance

There are many forms I've seen that operate along the sideways line in China which also use horse stance (mabu). In China, if it isn't a mabu, then it's either a gong bu or zenkutsu dachi (forwards stance), or cat stance or a stance like xingyi's zhan bu (or a few other odds and ends, like "chi bu" or "chicken step").

So in the end I see the stance in naihanchi in terms of Ockham's razor: what looks (in Motobu's and Funakoshi's case and in every other surviving early photo of naihanchi) to be a horse stance is probably going to be a horse stance. This is particularly so given its ubiquitous nature in Okinawan karate and Chinese quan fa and its importance in the far eastern martial arts. I think it is far more likely to be a horse stance than a relaxed upright stance or some other "strange" stance existing only in this kata and which meets the complex "koshi" interpretations we give the kata today.

With respect to those who would compare naihanchi's stance to sanchin, I don't think there is any evidence to support the assertion that it is a modification of an hourglass stance. The mere fact that it is used by some schools for "shime" testing doesn't qualify it to be "sanchin in another guise".

I figure if it looks like a horse stance, it probably is and always was.


1. See the article "Master Funakoshi's karate" by Graham Noble.

2. See this article from wikipedia on the horse stance.

Copyright © 2009 Dejan Djurdjevic