Tuesday, September 30, 2008

Naha te and its Chinese cousins

Further to my articles "Karate and the Chinese martial arts Parts 1 and 2"...

Many questions have been raised in on the internet about white crane sanchin and its relationship to goju-ryu and uechi-ryu. For those who haven't done so I invite you to visit the website of my friend Martin Watts, a long-time practitioner of Yong Chun white crane. His website (http://www.fujianbaihe.com) has many videos with links to his master's site in China. The videos are very interesting and I think show that uechi-ryu is arguably the closest Okinawan form to Yong Chun baihe (white crane). In my many discussions with Martin on this topic I know he certainly thinks so. I don't think they are related in a linear sense, but they are certainly "cousin" arts. In goju, the nearest we come to white crane is our form "tensho" (there is at least one Yong Chun form with similar movements).

I can see why many would argue a close connection between naha te (both uechi ryu and goju ryu) and southern mantis. However here is my reasoning as to why this relationship might more distant than, say, with white crane.

There is a group of external arts that can be broadly grouped together in a technical sense. I shall for these purposes call them the "Fujian external group". They include white crane, ngo cho kun (5 ancestor fist) and wing chun. It is commonly held that karate (in all its manifestations) is descended (principally) from this group of Chinese arts (with a heavy influence from the native Okinawan "te"/"ti").

As a student of xingyiquan as well as karate, I can see that there is a great deal of credibility in the theory (espoused by many) that the Fujian group is an "external" offshoot of early xingyi. By this I mean that the Fujian group are descended from the "internal" art of xingyiquan but have morphed into artforms that have predominantly "external" biodynamics (notwithstanding their obvious soft" elements).

There is a fundamental difference between external and internal biodynamics as I detail in my article “Understanding internal arts” - neither is better or worse, and nor is is true that one is necessarily "softer" or "harder" than the other. The labels "external" and "internal" are, for me, just labels to differentiate these biodynamics.

A simple example of the differences in biodynamics is xingyi's predominant use of the "advancing foot" in its forms ("yori ashi" in karate - front leg moves first, legs sliding, no passing; and "suri ashi" - legs sliding, rear leg passing to take the lead), where external arts predominantly use normal stepping ("ayumi ashi"). There are exceptions in the Fujian group of course - the ryuei ryu kata anan uses a lot of suri ashi and many versions of seisan in Okinawa use yori ashi. Ngo cho kun or wu zu quan (5 ancestor fist) occasionally uses suri and yori ashi in their forms. However, in the end the dominant form of sabaki remains "normal stepping". In xingyi the reverse is true.

An impressive demonstration of southern mantis by Andy Chung

Now we come to the "Hakka" schools of southern mantis and its close cousin bak mei (white eyebrow). These clearly show a far greater internal influence. Take a look at the dominant use of yori ashi for starters. Bear in mind, this is just one technical example - I have compared xingyi notes with my brother in law who is also a senior bak mei practitioner and there are many points of technical similarity. The Hakka schools are, in short, almost "internal" in their classification. Traditionally only xingyiquan, baguazhang and taijiquan (as well as liu he ba fa or "water boxing") are entitled to the "internal label", however there is certainly scope for arguing that the Hakka schools should fall under this umbrella as well, even though they contain some external (Shaolin) elements.

Now there were undoubtedly examples of the Hakka school in Fuzhou during Uechi Kanbun and Miyagi Chojun's visit. What sort of influence did they leave on those respective master's arts?

Well many researchers (Akio Kinjo, for example) believe that Miyagi's kata shisochin is a mantis form.2 It is true that the hand thrusts are very reminiscent of mantis. On the other hand, it's biodynamics are still inherently from the Fujian external group and - most importantly - bear all the hallmarks of an Okinawan artform.

I agree with Mario McKenna when he argues that Uechi and Miyagi (like their predecessors in China, Higashionna and Nakaima etc.) taught their own eclectic forms of martial art, placing elements of their Chinese study onto an Okinawan base (which, in turn, reflects a much earlier Fujian method rather than what they learned on the mainland).1 As an example, see my article “Seisan - the universal kata”. It seems to me that Uechi borrowed or preserved many elements of Okinawan vesions of seisan in his kata.

So while I agree that Uechi and Miyagi may have borrowed elements of mantis/bak mei, they remained true to their external nature. Their arts remain, essentially, part of the Fujian external group, rather than the "Hakka" (semi-internal) group by classification.

On the other hand, what strikes me about Yong Chun is just how similar its biodynamics are to uechi ryu (and to a lesser extent, goju ryu). It's san shr tai bau (13 treasures) is reminiscent of uechi's seisan (though not goju's). It is certainly not a predecessor to uechi or goju, but to my mind it reflects a mainland descendant of the Fujian external group circa late 1800s (pre-boxer revolution), much like karate.

But it is the emphasis on sanchin/sanzhan that is, I feel, the best barometer of relation. While many schools in China have sanchin/sanzhan, Yong Chun's version shows the most similarity to that in uechi ryu which leads me to believe that Yong Chun is the closest extant relative to uechi ryu in China today (though by no means a direct ancestor). Ngo cho kun would, by comparison, be goju's nearest relative. It is important to note that Yong Chun and other white crane schools (see Eric Lim's ancestral crane, for example) use a different sanchin stance to goju/uechi - their front foot points forward and the back foot points out to the side (curiously this stance survives in some shorin - and even goju - schools as "seisan dachi" – again see my article “Seisan - the universal kata”). By contrast, ngo cho kun have a more "pigeon toed" (or at least "straight line") sanchin/saam chien like goju, uechi, ryuei-ryu etc.

A demonstration of goju, uechi and white crane sanchin/sanzhan

A shorin mate of mine in the US has long argued the mantis/naha te connection, however my guess is that if mantis is related to karate, the relation is far more distant (about as distant as white eyebrow and xingyi). I think that modern similarities are more likely due to cross influence and parallel development than historical derivation.


1. See Mario McKenna’s article “So what did you think you were doing”.
2. See Joe Swift's article "The Kempo of Kume Village" at Meibukan Magazine No. 6.

Copyright © 2008 Dejan Djurdjevic

Martial arts and practicality


Readers of my blog will be familiar with my view that "it is the man, not the art". However it has recently been said to me on an internet forum that this is nothing but a myth. Rather, it is contended, some arts are plainly better than others when it comes to applicability for civilian defence (something I shall label "practicality" for the purposes of this article). Principally this view has been expounded to me by "modern" combat sports practitioners who deride traditional martial arts (with their forms/kata) as "impractical". At the centre of their argument is the observation that traditional arts are typically not practised or tested in "live" or "resistant" conditions. Why? "Because they don't work", is the inevitable retort.

So are some fighting systems (in particular traditional ones) less "practical" than others (in particular the modern combat sports)? Is "the man not the art" just a screen of political correctness erected to hide something "everyone already knows"?

In order to examine these questions I think it is important to understand what is meant by "practicality". In a martial context this term is actually short hand for 2 different things, namely:

1. practical training methods;

2. practical techniques.

For example, it is generally accepted that most aikidoka don't have a "practical" training method in the sense that they do not practise or test their skills in a "live" or "resistant" environment - but does this mean that aikido techniques are necessarily impractical? Muay Thai training might have a more "live" training environment, but does this mean that it has more practical techniques for an ordinary person in a civilian defence scenario?

Practical training methods

It is true that most traditional martial artists are not inclined to practise in a live environment. But is this really enough to consign all traditional arts to the "impractical" bin? I certainly don't think so. Why not? Learning the finer points of an advanced traditional martial art (eg. aikido/wing chun/taiji/karate) takes a lot of "impractical" isolation training just like a golf swing or tennis serve (more on this in a minute).

So you can't just grab out selected traditional techniques and say "I like your karate front kick - I'll just throw that in to my sparring". If you did the latter you'd probably end up with techniques such as the "pushing kick"... not the same thing and not nearly as good as good (see my articles "Visible force vs. applied force", "Hitting harder: physics made easy" and "Understanding the internal arts" for an explanation of what I mean). I don't know many "non-traditional" martial artists or combat sports practitioners who can do anything like a good front snap kick - most are not even aware that they are lacking what I consider to be a basic, essential fighting skill.1

Leaving aside specific technique acquisition, what about the lack of "live" training? The argument goes, "one does not swim best by flailing one’s arms on dry land. One learns to swim by getting into the water and then later moving into the deep end." Very true. On the other hand this doesn't mean that the best training is to jump into the pool and start swimming with the doggy paddle assuming this to be an optimum stroke. It isn't - irrespective of the fact that it is "natural", "no-nonsense", "straightforward", "easy to learn" and "useful immediately". And it doesn't matter how long you perfect it - in the water or out.

It is certainly the case that if you take a student with no experience and train him/her for one year mostly with gloves/bags and sparring, and simultaneously train the student's clone mostly in kata, the latter will be less prepared for combat than the former. But he or she needn't stay that way: the student will learn a formal version of "freestyle", "backstroke", "breastroke" and "butterfly". Now all he or she needs to do is start applying it.

Of course the swimming analogy is deeply flawed: unlike combat, swimming does not require reacting to another party. Leaving this aside, the main difference is that unlike swimming, combat has almost infinite angles of movement and variables. This means that "doggy paddle" will be much more tempting and that "breastroke" will be harder to apply. It doesn't mean it is impossible. And it doesn't mean that the "extreme training" that has produced doggy paddle will be producing the optimum method. The "extreme training" if anything, discourages experimentation and development of better technique. It encourages conservative reliance on what you already know.

No one disputes the value of "live" training. But all too often I hear people devaluing the importance of technical development (or the assumption that honing the right cross to within an inch of its life is the same thing). You need a mix of both live training and technical progression. Just as you need to get out into the "real world", you shoudn't rely on your grade 7 texts for ever.

As an aside, I think that BJJ is a good example of technique (not surpising since it is firmly based on judo and jujutsu, depending on how you want to label the activity that spread to Brazil). It is much richer and more complex than what people in ring sports adopt as a stand up strategy.

Grappling is however different from striking arts in that it can only be practised with a partner - hence it has never had single person kata etc. even in the traditional realm. But the technical complexity (stemming from traditional martial arts) is there - same as "stand up" arts like karate. That this complexity (eg. the art of deflection, front kicks) are not being absorbed by ring fighters is, in my view, a function of the fact that they require a lot of dedicated study and are therefore largely misunderstood, not that they are inapplicable.

Unlike grappling, stand up dynamics are also a lot more variable, making it harder to apply skilled techniques cleanly. The fact that application also means a missing tooth (rather than a twisted arm) goes without saying.

Practical techniques

The reference to "practical techniques" is itself short hand for a number things, namely techniques that are:

1. capable of being used in combat; and
2. easily learned; and
3. in themselves sufficient for civilian defence.

But is this a truly useful definition for determining whether a martial discipline is worth studying? Grappling meets the first 2 requirements - but not the third (it is generally accepted by all except die-hard grapplers that you need good stand-up skill for self-defence). Yet few have been game to say that grappling is "impractical" - particularly after the first Ultimate Fighting Championships where the Gracies showed the effectiveness of their Brazillian jujitsu. Rather most combat sports practitioners stress the importance of acquiring at least a grappling skill subset.

By extension, if I choose to augment my skills in another highly useful sideline to punching/kicking (eg. trapping skills from Wing Chun), I cannot see why I am being "impractical" even if I feel that these skills are not, in themselves, sufficient to constitute a "complete" civilian defence system.

Most importantly, I cannot see why I should preclude myself from studying an art that is hard to master, yet once mastered is very good in combat (ie. an art with techniques that fail the second requirement but pass the first and third). Rather I feel that senior students should practise such an art in tandem with other (more easily acquired) skills.

I know that my views in respect of the latter are not shared by many combat sport practitioners who have argued to me that it is best to rely on a much smaller subset of skills that is immediately practical to some extent, and forget about other "hard to learn skills". But this doesn't make much sense for someone who has trained for as long as I have. Why wouldn't I expand my range? Is the need for immediate practicability really that much of an issue to me? I am not able to punch and kick as fast and as hard as when I was young. I see little point in trying to refine my right cross by tiny increments in the hope of maintaining a small edge over the big new bruiser in the gym who can naturally hit twice as hard as I can...

On the flip side, I know of martial artists and even footballers who have "cross-trained" in such activities as ballet hoping to get some small unexpected edge. I get the feeling that most combat sports practitioners wouldn't even consider something as obtuse as that. But they can and should look to acquiring traditional skill sets, just as traditional martial artists should look to other traditional martial arts and combat sports.

"But", some will argue, "I just can't see some techniques working - aikido for example. You just can't do that stuff in MMA." As an example they will offer a video like the one below:

An example of "non-live" aikido training

It is quite clear to me however that many of the defences in the video (at least the standing ones) are solid - move in early and intercept the attack, then drop the opponent. I have no doubt that they could be practised realistically and then applied in a "real world" environment. They might require some adjustment in terms of attack interception etc. - but the principles are sound.

Presumably what some people find objectionable is the training method of aikido. Yes, the video shows too much compliance for realistic training. However the art of aikido is based on solid principles related to other forms of jujutsu and this must be distinguished from the "intensity" issue.

Here's some aikido done a bit more realistically: http://www.youtube.com/watch?v=0aJv48aVUko

I have no doubt the techniques are capable of being used in reality - maybe not as cleanly as demonstrated, but more or less (at least the majority). I've seen many of the techniques applied (usually in a more limited way) in ring fights (including the kick defences at the end).

That aikido is hard to apply is true. The reason is 2 fold:

First you need to learn fairly refined movement. To learn movements of this refinement requires many years of "non-live" isolation practise. After all, you can't learn a good golf swing unless you do it without pressure. And a golf swing has nothing on an aikido throw, believe me.

To use a better analogy, you couldn't expect to learn a proper baseline or net shot in tennis if you never spent time isolating the movement. If you just go straight into playing full tennis games because "you don't have time for isolated practise - you're too busy returning balls" you'll be wiped off by even a slightly coached player.

Second, once you've learned the technique you need to introduce it gradually into a live environment. While most aikidoka don't do this, it doesn't invalidate the technique, nor make it impossible. Free fighting has many more variables than a game of tennis, which means it is harder environment to apply technique of any kind. Refined technique can be applied in a free fighting environment provided you actually go to the trouble of training to do so. The fact is that neither MMA nor aikido practitioners ever have: Most people who have the inclination and patience to learn aikido aren't minded to train in more live enviroment. It is a matter of natural selection. Are people drawn to ring fighting ever going to start aikido? Are people who do aikido the types who might have joined and stayed in an MMA gym?

Criticise the lack of intensity all you like - you'll probably be right. But as for the techniques - aikido is a highly skilled discipline with a similar technical base to BJJ or judo - trashing it or its experienced exponents by simplistically focussing on the training method and ignoring the technique shows some other "issues" in my opinion. It is also deeply insulting of the many years of sweat and dedication aikidoka put into the mat. The techniques work - an armbar is an armbar, a shoulder lock is a shoulder lock. Take a look at the video below: when aikido type locks or projections are inadvertently applied in MMA fights they work; why shouldn't they - they are based on sound biomechanical principles.

An example of aikido type techniques applied in MMA

What people are presumably saying when they criticise traditional techniques is that they disagree with self defence "formulae" - ie. "if he attacks you with X, step forward and do Y" etc. Such formulae hardly ever work - even when the attacks are more "realistic" - and they need to be distinguished from the techniques which comprise them.


Is there an optimum way of learning, practising and combining martial skills for civilian defence? There are probably many. It is even likely that what is "ideal" will vary greatly from person to person. Consider that in tennis Steffi Graf was reknowned for her ability to run around a ball just to use her powerful forehand (where others would have simply used a backhand). In combat sports Bill "Superfoot" Wallace was reknowned for his skill with his right roundhouse/reverse roundhouse kick combination which he often used in preference to simpler kicks and strikes. The fact that in both of these cases an athlete made something quite "unorothodox" work for them suggests to me that it is unlikely "one size will fit all".

My personal approach has been to recognise the need for a sequentially relativistic syllabus - one that ensures a continual path of skill development mixed with the necessary elements of realism in training. Such a syllabus will vary in its implementation depending on why people train in the martial arts. In the case of a person wishing/needing to use his or her skills in the ring or in a military/security/law enforcement context the mix will require less "long term" development and more "live training". In the case of the average person and his or her needs for civilian defence, exercise, fitness, health and fun, the mix will favour more "grooving in isolation". Regardless, I favour a gradual shift from simpler, more practical techniques to progressively more complex/advanced/efficient ones - in other words, a transition from "external" to "internal". For more on these subjects read my articles "My quest for the martial holy grail", "Hitting harder: physics made easy" and "Understanding the internal arts".

So is any particular art "less effective" than another? Sure - depending on your goals, experience and other individual differences one art will suit you more than another. Are some "arts" absolute nonsense? Indeed. Some newly made-up systems are not even based on sound biomechanical principles. But the established traditional systems don't suffer from this malaise in my opinion. They have had long periods of evolution to iron out the biomechanical kinks. While some arts will statistically confer more "practical" benefits sooner (or perhaps overall) I don't know if this provides any useful data when it comes to evaluating an opponent. Why? You have no idea of your opponent's individual differences - his or her level of skill, experience and ability to apply techniques which are commonly regarded as "less practical" - think Bill Wallace and his high roundhouse kicks or Steffi Graf and her "run around" forehand shots. In my experience these individual differences trump "style" or "art" any day.


1. I have recently encountered the argument that "traditional ball of foot front kicks aren't used because they carry a higher risk of toe injury - heel kicks are better". Except that this misses the point I make in the article "Visible force vs. applied force": you can't do a snap kick with the heel (at least not very easily or effectively) - making it hard to impart hydrostatic shock. And in almost 3 decades of hard and fast sparring I've broken a toe only 3 times... But I have what I consider to be a very hefty front snap kick! In my opinion people don't avoid front snap kicks because of "toe" issues; they avoid them because they don't know (a) just how effective they are; and/or (b) how to do them correctly. This kind of argument is usually countered with "front snap kicks are not used because don't work". There is little I can do to answer this except say: "you obviously haven't been on the receiving end of one".

Copyright © 2008 Dejan Djurdjevic

Wednesday, September 17, 2008

Is mawashi uke goju's rising block?

Like most other karateka, practitioners of goju ryu faithfully practise the standard age uke (rising block) during basics training. They will apply it in ippon kumite (one-step sparring), "find" it in kata bunkai (application analysis) and desperately try to apply it in sparring.

But is it really a goju technique?

What karateka call "age uke" is really a basic shorin technique. The only kata in which it is found are the 2 gekisai forms, developed and introduced by Miyagi in the early 1940s as basic kata for school children. Prior to that one wonders whether it was even practised in goju dojos...

This question has lately led me on a journey to discover whether goju ryu has its "own" rising block. What did goju/naha te practitioners use for defences to head height attacks before age uke was incorporated into the syllabus? As summarised in the video below, I feel that the answer is to be found in goju's famous "mawashi uke" or roundhouse block.

I discuss mawashi uke and its relationship to age uke

There are 2 forearm deflections found in goju ryu kata that might be candidates for ancestry of the modern age uke; both are found in the internal arts of China of which xingyi in particular is thought to be a distant relative of goju ryu.

The first is a deflection commonly known in karate as "haiwan nagashi uke" (top of forearm sweeping block). This is an internal arts deflection that relies on redirecting an attack both over the head and simultaneously to the side. It occurs in xingyiquan, baguazhang and taijiquan. In the photograph to the right you will see my shifu Chen Yun-Ching, his brother Chen Yun-Chow and James Sumarac demonstrating a variant from xingyi.

In goju ryu the haiwan nagashi uke occurs in only one kata - seiyunchin. This is significant because it means that the haiwan nagashi uke cannot really qualify as "goju's answer to age uke". One would expect it to occur far more frequently if it were goju's standard rising deflection. [Note that by contrast, haiwan nagashi uke occurs often in shorin ryu kata, including naifunchin/naihanchi (from which some argue seiyunchin might be derived - see my article "The origins of goju-ryu kata: Part 4".)]

Importantly, haiwan nagashi uke, as manifested in seiyunchin kata, is performed turned fully side-on - not at all like the age uke (or even the xingyi version demonstrated by my shifu above). In this respect it is not dissimilar to another version of this block (shown to the right) used in Hong Yi Xiang's Taipei-based Tang Shou Dao.

The second possible ancestor of age uke is a deflection colloquially referred to as the "steeple block". Unlike haiwan nagashi uke, this deflection is applied "front on" as one would an age uke. In fact the only real feature that distinguishes it from the standard age uke is that the elbow is not raised. Why? Raising the elbow means that the shoulder girdle will also rise at the end of the movement. This not only takes more time, it also creates an inherent weakness; for a movement to transfer momentum efficiently it must effect a staged activation of larger to smaller body parts - not the reverse (see my article "Hitting harder: physics made easy").

Raising the elbow also creates a huge opening for further attacks: the higher you raise your elbow, the more you expose the delicate underarm area and the further you have to return your arm in order to cover that area.

It is for these reasons that the internal arts eschew the standard age uke in place of either the haiwan nagashi uke discussed previously (where the above issues are mitigated by a small or large body turn) or by using the steeple block which does not raise the elbow. In respect of the latter most instructors will tell you that while the steeple block is more efficient and leaves less of an opening, it is correspondingly harder for beginners to apply because the margin for error is so small.

I demonstrate the steeple block

As you will note from the above video, the steeple block occurs most obviously in the goju kata seisan and kururnufa. Accordingly I think of it as "truly goju" because it is found in both "cluster H" and "cluster M" - the 2 principal technical (and arguably historical) groupings of goju kata (see my article "The origins of goju-ryu kata: Part 1").

It has long been my belief that it also occurs in sanseiru and shisochin as the first part of what I call "sokumen awase uke". Importantly, in the case of sanseiru and kururunfa the block is followed immediately by an "ura te osae uke" (a depressing block with the back of the hand. This, combined with a forward movement of the body, not allows increases the effectiveness of the deflection, it feeds directly into a friction hold/control or grab while the other hand effects a strike. Finally, it mitigates against the fact that the steeple block usually has "little room for error"; the back of the wrist can still curl around and "save" you if you haven't quite got the deflection right.

But herein lies the problem: for the steeple block to be "goju's age uke" it must not only be present in a few kata; it must be ubiquitous. Steeple block does not fit that model. Or does it?

I was wondering the other day (for the umpteenth time) about the true meaning of goju's "mawashi uke" or roundhouse block. I have lost count of the number of times that I've been told it is "the most advanced block of goju" containing "the secret" of that system. I have myself made much of this over the years. And in truth mawashi uke does have many potent/impressive applications. Only one aspect has ever troubled me and that is this:

There is a movement half-way through the mawashi sequence where one arm inscribes a vertical circle in what some call "the window-wiper" movement. Try as I might I've never been able to make sense of this. I've tried to think of it as a standard hiki/kake uke (grabbing or hooking block) but this is not correct; if you apply it as such the whole dynamic of the mawashi uke changes. This is because hiki/kake uke is preformed at an angle of 45 degrees to your body not 90 degrees like the basic mawashi uke. Even a brief attempt to perform mawashi uke with a hiki/kake emphasis will show you that it is completely at odds with the general movement/principle of a "roundhouse" block.

I demonstrate the traditional mawashi uke

So what is the function of the "window wiper" movement?

As far as I'm concerned, only one answer makes perfect sense of the dillemma. If you cut short the "wiper" movement by making it slightly less circular - in other words if you "cut upwards" instead of making a circular action with your principal deflecting arm - you effect the standard steeple block (as it manifests in, say, kururunfa). As I demonstrate in the video at the start of this article, the mawashi uke requires very little modification in order to achieve this. One might not even realise that any modification had been made - natural variations in individual body movement will often be greater.

I demonstrate the mawashi uke with the steeple variant

Accordingly I would argue that my opening question has been answered: Mawashi uke is goju's age uke. I say this for 2 reasons: First, it contains the steeple bock - a cousin of shorin-ryu's basic age uke. Second, mawashi uke qualifies as an essential goju movement since it is in virtually every kata of that system.

If there is any indigenous "age uke" in goju, mawashi uke must be (or contain) it!

Copyright © 2008 Dejan Djurdjevic

Friday, September 12, 2008


“Grounding” (sometimes called “rooting”) is an essential skill in traditional martial arts and is often associated with the sanchin/sanzhan stance in many schools of karate and particularly external southern Chinese arts. It also features strongly in the internal arts of xingyi and its offshoot yi quan (see my articles “Sanchin in the Chinese martial arts”, “The naming of sanchin” and “Seisan - the universal kata” where I suggest a link between sanchin and xingyi’s “san ti” posture which utilises a stance sometimes called “zhan bu” (battle stance)).

In our school we practise a kind of “standing pushing” exercise in sanchin intended to develop and test grounding. Both partners stand in sanchin at bent elbow range – one hand on the hip, one hand on the shoulder. There are no “rules” other than “no leaning” and “no sudden or pulsating thrusts”. The drill is illustrated in the video below:

Sanchin pushing

So far I've been able to resist such “standing pushing” by far stronger/heavier men in the gym (provided they don't lean in with a significantly larger body mass).

As a matter of interest, I filmed this after I'd lost 18 kg in hospital - yet I was still able to push over larger people without much loss of "grounding" ability. It goes to show how grounding relies on correct technique for momentum transfer, not power (see my article “Hitting harder: physics made easy”).

In this context it is also easy to see how “sanchin pushing” might be assumed to have a “paranormal” element (more on that in a moment). I can assure you it does not.

So how does it work? It relies on efficient posture and lowering your centre of gravity. It is not, of itself, a "combat technique", nor is it just a cheap "parlour trick". Rather it is evidence of good, basic grounding.

I'm not going to start a lengthy physics dissection here (as I did more generally in my article “Hitting harder: physics made easy”). Suffice it to say that good grounding gives you a foundation for generating or receiving any applied force; weak grounding does not. Good grounding involves the correct (optimum) posture to create or receive a push/force etc. That's why leaning isn’t allowed – we are testing each other's stability, not who has the greater mass.

The pushing exercise shown above does not need to be done in sanchin. It can be done in any stance once you acquire sufficient skill. I've found that grounding translates to every other stance. I have the same results in shiko dachi (horse stance) and even natural stance. However grounding is easier to learn in sanchin. Why? Sanchin lets you use the “shime” or pelvic thrust (see my article “Whole lotta shakin’: addendum”) to add a forward and upward moment.

The "shime" is obvious with beginners and becomes more subtle as you gain experience until it is barely discernible. This is just one variable and works in sanchin particularly.

Good grounding is a prerequisite for applying "softer"/"internal-type" techniques (ie. those that focus on the efficient transfer of momentum rather than on power – see my articles “Internal vs. external martial arts”, “Hitting harder: physics made easy” and “Understanding the internal arts”).

Sanchin pushing is one of the "internal" things I do which I could, if I choose, pass off as some kind of “chi”/“ki” because it has people scratching their heads. The biomechanics are not obvious. The fact that I have bent my knees to a particular point and no further, aligned my back and hips in a particular way, performed a subtle "shime" as noted above goes unnoticed to the untrained eye.

It is important to note that in the video Tim is not being "compliant" in letting me push him (except in so far as he isn't "cheating" in this exercise by leaning in). In fact, most of my students try to beat me in this exercise and I encourage it (I even get "challenges" after class - we all know it isn't "combat" but people are, by nature, competitive).

Sometimes the more senior students do beat me (not often!) (in my experience skill in sanchin pushing/grounding correlates with years of training). This is a big difference to many purported “internal” schools of the kind I refer to in my article “Understanding the internal arts” where students let themselves be “thrown” many metres by the mere wave of a hand. As discussed in that article, I do not, for one minute, think this is anything but fake. The students are complicit in their teacher's faking (whether consciously or subconsciously). As it happens, no one I know is in such debt, or holds me in such "awe", that they will fall down when I merely wave my hand!

In this regard, you'll note that Tim doesn't fly away with great speed, only to fall, flailing in agony, on the floor. In this case I was more stable (and able to withstand his "standing push"). Any more grandiose claim than that would be false in my experience and opinion.

And in case there is any doubt, Tim was trying to push hard – except that I’d got the upper hand and he’d lost any base from which to push. The "no leaning rule" also means he couldn't go down to a rugby posture and “salvage” his situation.

In this case I wasn't pushing hard - I had lots to spare; my stance was doing the work. He was attempting to push from a weaker base, so I could use minimal arm force to keep him at bay. With most beginners (even those who are big and heavy) I often don't need to use my arms at all...

So how does it feel? If your opponent is better than you it should (and does) feel like you're pushing a brick wall. If you are better than your opponent, it feels like he's pushing himself backwards and you feel “like a brick wall”...

If you're very evenly matched with your partner then it is touch and go as to who gets the upper hand – but once someone does, it goes back to the "wall" feeling and you know then that “it's all over, red rover”.

Copyright © 2008 Dejan Djurdjevic

Wednesday, September 10, 2008

Whole lotta shakin': an addendum

A colleague of mine on gojuryu.net recently said the following in response to my article: “Whole lotta shakin’: pre-loading the hips”:

“When stepping, there is an inherent motion to the hips. If this is utilized to load a technique, then there is no telegraphing or slowdown.

I think examples where there is a block THEN hip load THEN strike will never work against a properly motivated attacker. However, why can't all three of those things be the same - ie, block and punch with the same hip motion at the same time?”

I think this is an excellent point. It occurred to me many martial arts movements are specifically designed this way. Consider the humble sanchin dachi (3 battles stance) as it occurs in goju ryu and other karate kata. That stance involves a pelvic tension which occurs as you step up into the stance just before you effect a technique (whether it is a block or a punch or both). That pelvic tension produces a 45 degree movement in the hips, creating forward and “upward” momentum.

As you will note from the video below, this momentum can, in itself, be used as a weapon and produces surprising force when applied directly in a clinch.

However the real purpose of this movement is shown at the end of the video: the hip movement adds momentum to your technique in much the same way as the standard “horizontal” plane hip rotation can for a standard reverse punch. As discussed in my article “Hitting harder: physics made easy”, momentum transfer (and hence the application of force to your target) is maximised when you use “staged activation” of body parts, specifically joints, moving from the larger joints (in this case the hip) through to the next largest joint (in this case the shoulder), to the next largest (elbow) and (depending on your technique) the wrist.

The movement is natural and hence does not involve any unnecessary “telegraphing”. Consider by way of contrast the following video of Tomari te rohai where the demonstrator is loading his hips after he has already entered sanchin for a horizontal hip load; something which is neither part of the movement, nor (in my view) necessary when you consider the natural expression and application of sanchin dachi…

See also: "Whole lotta shakin': contextual hip use" and "The importance of flow" where I discuss a particular technique where a naturally occuring hip load has not been used, while an artificial hip load has been added.

Copyright © 2008 Dejan Djurdjevic

Hitting harder: physics made easy


Most martial artists share the goal of “hitting harder”. This is usually expressed in colloquial terms as hitting with “more force” or “more power”. But even a basic knowledge of physics will tell you that “force” and “power” are not the same thing. Which is it that makes you hit “harder” – force, power or both? And is it more helpful to talk in terms of something else such as momentum?

Understanding force

Many people think of “force” in very nebulous terms (perhaps explaining such misappropriations as “The Force” in Star Wars). So what is “force” as understood in physics?

Force is something that enables you to cause an object with mass to accelerate.

In other words, it is Newton’s famous formula: f = m x a.

Using this formula people commonly argue that in order to hit “harder” they need to –

1. maximise their mass; and
2. maximise the acceleration of their attack.

For the most part this is true: if you have a big mass and accelerate that mass well, you will maximise your chance of applying a greater force to a target. But it is important to note that this argument refers to force you are applying to your own body. It does not refer to the force you are applying to the target.

Put another way, in this argument the equation f = m x a assumes that m = your body mass, not the mass of your target (as it should for the purposes of calculating the force on that target!).

So instead of looking at the force applied to your body, let’s examine the force applied by your body to a target.

For example, consider a stationary object weighing 1 kg. For simplicity’s sake, let’s say you hit/push it for one second causing it to reach a velocity of 1 m/s. We can use the following formula to determine the accleration of the object:
v = u + at

where a = acceleration, v = the final velocity (1 m/s), u = the initial velocity (0 m/s) and t = time.

Accordingly we can determine the acceleration of the object as 1 m/s2.

Now in order to determine how far the object will move in that one second, we can use the formula:


where s = distance.

From this formula we can determine that the object will move 0.5 m. Put another way, if we cause a stationary 1 kg mass to move 0.5 m in one second we’ve caused it to accelerate at 1 m/s2. By reference to the formula f = m x a you’ve applied a net force of 1 N or kg/m2 to effect this result.

Clearly the more force you apply, the farther you will move it in the same time (by accelerating it to a higher velocity). So imagine you have managed to force the 1 kg object 1 m in one second (ie. you’ve managed to accelerate the object from 0 m/s to 2 m/s). In that case you’ve applied a force of 2 N.

So far so good. At this point you might be forgiven for thinking: “If I can knock my opponent across the room with my punch, surely I must be hitting very hard?”

Well yes and no. Let me explain by asking a question: Does “hitting harder” really mean moving something farther? If it does, a strong push or shove arguably constitutes the “hardest hit”, because a strong push is likely to move something farther than a punch or kick. I’m sure you’ll agree that this is an extraordinary proposition (for more on this topic see my article "Visible force vs. applied force"). The former is designed to move, the latter to hurt (with or without moving).

It seems that “movement”, and hence the acceleration of your target, might not be the most significant indicator of “how hard you have hit” after all.

Furthermore, you might hit a brick wall with your fist as hard as you can. You’d be unlikely to move the wall at all. Does this mean you haven’t hit it “hard” (you have) or with any force (you have, except that the wall has exerted an equal force on your fist as per Newton’s Third Law)?

Theoretically if you were heavy enough you could push the brick wall over (ie. move it) – but we’re back to “pushing”. Theoretically if you punched fast enough your peak force1 would be sufficiently high to break through the brick wall without “moving” it (eg. a high velocity bullet can pass through objects without displacing them).2

So what distinguishes a strike from a push? It is your deceleration before hitting the wall. The more you decelerate before you reach your target, the slower the velocity at impact. If your velocity at the point of impact is low, you’ll effect more of a push. If your velocity at the point of impact is high, you’ll effect a blow.

By now you should be starting to see the limitations of using f = m x a as a means of determining “how hard you hit”. The equation might tell me how to increase the force applied to my body (which is necessary in order for me to apply force to someone/something else), but it doesn’t tell me how to apply my force.

Rather than talk about the acceleration of your mass or the mass of the target, it seems to be more useful to consider the velocity of your mass at the moment of impact.

This is momentum.

The importance of momentum and its transfer

In physics momentum is defined by the following equation:

Momentum (p) = mass (m) x velocity (v)

Obviously a one tonne (m) car travelling at 60 km/h (v) will do some damage if it hits you. A fist, while not so destructive, obeys the same physics.

It appears that a well-thrown fist reaches its maximum velocity when the arm is about 80% extended. Accordingly if your punch covers say, 60 cm from a fully chambered position to full extension, then your punch reaches its maximum velocity at 48 cm.

This speed is generated by moving a number of body parts toward the target; eg. the whole body (by stepping or lunging), hips, leg, shoulder, upper arm and lower arm. If the full range of movement is used, and the body parts act in a staged way to transfer momentum in a whip-like sequence from larger to smaller body parts, then the fist will be accelerated to the fastest possible speed. A reverse punch is biomechanically better suited to transfer momentum than in this way than a leading arm punch.

So far so good. But doesn’t this come to more or less the same thing as previously argued – in other words to hit “hard” you want a mass travelling at a high velocity? Again, close. But there is more. You not only want your body to have momentum; you want to transfer it effectively.

Now, momentum transferred is called "impulse". The equation for impulse is as follows:

Impulse = force x time

In other words force = impulse / time

With impulse a fixed quantity, force and time are necessarily inversely proportional. In other words, one can deliver a given amount of momentum by:

1. transferring a large force for a short time; or
2. by transferring a smaller force over a longer time.

So, the longer it takes to transfer momentum, the less force is applied. If you want to maximise your force, you must ensure that as much of your momentum is transferred on impact as possible. This is true for both focused punches and follow-through punches used by boxers. The only real difference is that punching with boxing gloves increases the time it takes to transfer momentum which reduces the force compared to a similar punch with bare knuckles. In other words, when punching with gloves the momentum transferred will be the same as bare knuckles but there will be less impact force and instead the target will feel more of a push.

"Power punching" vs. traditional punching

When you throw a cross punch (see my article “Chambering punches”), the punch travels a longer distance than a straight punch – particularly if (as is inevitably the case) the cross follows a curved path to some extent. The further the punch travels from the chambered position to the target, the more time it has to accelerate and the faster its maximum speed will be (at 80% extension). This is why the cross punch carries more momentum than any other punch.

If this is so, why wouldn't one always use a boxer's cross punch? Why would one ever stop one's punches at a pre-determined point rather than use a follow through? The answer to this lies in the inherently conservative approach of traditional strikes - ie. they are part of a civilian defence strategy that puts "not being hit" ahead of "hitting with maximum force". A traditional martial artist focuses a strike so as to stop it at a pre-determined point in space in order to avoid over-committing (eg. if the target was to duck or evade) (for more on this topic see my article "Stopping strikes at a pre-determined point").

However it is not true that the traditional strike is "weak": it achieves the maximum possible force that is possible if one doesn't use a follow-through. It does so by stopping the punch as quickly as possible at an optimum predetermined point just beyond the target. This is the definition of "kime" or focus, a concept central to the study of karate and many other traditional Far Eastern fighting disciplines - see my article "Kime: soul of the karate punch".

As with striking a makiwara or pad, the strike will be stopped earlier than the predetermined focus point by the target if it does connect. But if the strike does not land, the traditional martial artist's positioning, balance and defensive capacity won't be compromised. For more on this topic see my articles "Stopping techniques at a pre-determined point" and "Karate punches vs. boxing punches".

The role of power

That a competitive boxer can apply a staggering amount of force with a gloved cross compared to, say, a suburban karateka performing a bare knuckle reverse punch, does not detract from the fundamental observation I have made above: a gloved punch has more “push” and less “hit” than any bare knuckle variant. The competitive boxer is big enough, and is moving fast enough, to compensate for any “diffusion” of his/her force resulting from the wearing of gloves. He or she might knock you across the room – and hurt you at the same time.

However this in no way invalidates a karate punch (ie. a straight punch) in which the impulse is maximised. Put another way, while the karate punch might not carry the same momentum as a boxer's cross, it can still achieve an effective result - without necessitating the same amount of work as one sees in a boxing contest.

By “work” I mean, of course, the concept in physics defined by the following equation:

Work (w) = force (f) x displacement (d)

In other words, work is done when a force acts upon an object to cause a displacement. Because of the diffusion of force caused by gloves, a boxer's gloved cross punch will cause more displacement than a bare knuckle punch thrown with the same force. This means that the gloved boxer has necessarily done more “work” than he or she might have had to had he or she been fighting without gloves.

Work can also be described as the amount of energy transferred by a force. In this case the gloved cross punch, using the same force, transfers less energy than it would were the boxer not wearing gloves.

As I have said above, the boxer will usually more than compensate for any diffusion caused by gloves. He or she will do so by adding more “power”. However this also means that a great deal of boxer training is concerned with power, given it's central role in the art of fighting with gloves.

In physics the equation for power is as follows:

Power = work / time

To compensate for the diffusion of his/her force because of gloves, the boxer has to work more in a shorter time. This means he/she needs more power and will train accordingly.

Don’t all martial artists need power? Of course they do. Everyone will be subject to the same laws of physics. However, remember the purpose of this article: to determine how one can “hit harder”. Power and displacement are directly proportional. Yet “hitting harder” doesn’t require greater displacement - rather, greater displacement is (at least to some extent) a foreseeable by-product of gloved fighting.

For more on the issue of displacement, see my article "Visible force vs. applied force".

Accordingly, while boxing training and methodology is eminently applicable in self defence, it is a mistake to assume that it is perfectly suited to a civilian defence system - or that other strategies designed with civilian defence specifically in mind are less effective for their purpose just because they don't develop strikes that are as "powerful" as those in combat sports.


If you want to “hit harder” you should look to maximising your momentum by increasing your velocity and/or your mass. But you also need to look closely at how you transfer that momentum.

Transferring momentum effectively is a question of maximising your impulse; ie. applying a large force over a short time. The alternative (a small force for a longer time) produces a push – not a “hard hit”.

The smaller your impulse (ie. the longer you take to apply a force), the more power you will need to compensate.

It is my view that a martial system that focuses on maximising power is what is classified in China as “external” or "hard" (waijia). On the other hand, a system that focuses on impulse is what is classified in China as “internal” or "soft" (neijia) – see my articles “Internal vs. external martial arts” and “Understanding the internal arts”. As I have noted in that article, no art is purely "external" or "internal" – just as no strike can be effected without some impulse and no technique can be effected without power.

A good “civilian defence system” will attempt to find a pragmatic mix of both – or ideally adopt a sequential system of teaching progressively more “impulse-oriented” techniques and less “power-oriented” techniques – something I call “sequential relativism” in martial arts training (see my article “My quest for the martial ‘holy grail’”).


1. It is important to note that when I refer to maximising “force” I am of course referring to the peak force – force is never applied evenly but is applied in a curve. It is the peak force that is in issue when it comes to “hitting hard” – not the average or median force etc.

2. In this article I have not attempted to address factors such as stress and pressure. In the case of the former we are all familiar with the idea that we can increase damage by targeting softer areas. In the case of the latter we also know that we can increase damage by applying force over a smaller surface area – ie. increasing pressure. These are topics unto themselves which I hope to address at another time.

Copyright © 2008 Dejan Djurdjevic

Sunday, September 7, 2008

The origins of goju-ryu kata: Part 4 - seiunchin kata

Continued from Part 1, Part 2 and Part 3 of this article.

In his book “Okinawa kempo” Choki Motobu mentions the kata seisan, seiunchin and naihanchi as kata that were in existence in Okinawa long before Kanryo Higaonna’s trip to China.

He writes1:

“Among those styles or katas which have been used in Ryu Kyu from ancient days are:
Sanchin, Jo-Ju-Shi-Ho, Seisan, Seiunchin, Ippakku-Re-Hachi, Naihanchi (Ichidan, Nidan, Sandan), Passai, Chinto, Chinte, (bamboo-yari spear style), Wanshu, Rohai and Kusanku.
And especially the three styles Nai-Hanchi, Passai (great and small), and Kusanku which are very widely known to many islanders.

As I have mentioned, Ryu Kyu Kempo-Karate originally came from China. Sanchin, Jo-Ju-Shi-Ho, Seisan and Seiunchin have been used there for many centuries.”

Is Miyagi’s seiyuchin the same as the original shorin form? We’ll never know for sure. Today’s seiyunchin appears to contain enough Naha te elements to suggest some modification by Miyagi. Certainly it has a right/left “balanced” design similar to kata such as naihanchi/naifunchin, however it is important to note that in this respect it follows the distinctive “cluster M” design (for details on this design, see Part 1 of this article) rather than the naihanchi/naifunchin template.

It is impossible to determine of course just how much “original” seiyunchin, if any, exists in Miyagi’s creation and how much it is a composite of other Okinawan and Chinese techniques that Miyagi might have come across. Perhaps the only constructive thing one can do is to examine other kata known to have “seiyunchin-like” elements in the hope that this might shed some light as to seiyunchin’s origins.

In this regard, I have always found seiyunchin has similar elements to versions of rohai, in particular the Matsumura and Tomari te versions. Depending on the version, the kata features open hand “sukui to hiki/kake” blocks reminiscent of the opening moves of seiyunchin — consider the pictures to the left of an unknown performer demonstrating the kofukan variant contrasted with shito ryu’s matsumura version on the right.

All versions of rohai also feature a 2-handed high and low block occurring in sagi ashi or "one-legged stance(see the picture on the left) startlingly similar to that in seiyunchin where the same block is performed in shiko (see the picture on the right).

Shito ryu Matsumura rohai

Rohai has movements that are clearly “naihanchi like” - yet there are also movements that are reminiscent of both seiyunchin and seipai. Consider the (somewhat altered) version in suikendo below:

Suikendo rohai

I think this version is quite interesting in that it seems to go over the same ground that Miyagi might have in either modifying or synthesising his own seiyunchin and seipai (in as much as "shorin ryu" moves have been applied in a distinctly Naha te manner...).

However what is most intriguing is that seiyunchin displays elements of Gokenki’s nepai which survives in what is arguably its most “reliable” form in goju’s sister school, tou’on ryu. The most striking similarity between nepai and seiyunchin is an identical opening move (albeit that it is not repeated 3 times and does not feature the sukui uke discussed above).

Another “crane” form that has clear “cluster M-like” elements is the kata hakutsuru of Kume village, as researched by Patrick McCarthy. These include the "closing" move from seiyunchin:

Finally we see seiyunchin elements in Aragaki forms, notably in sochin which (at least in the version shown below) has an identical closing move.

Aragaki sochin kata

Of course with all these similarities we have no way of knowing whether they are forerunners of seiyunchin or whether they have been influenced by seiyunchin. All we can do is make an educated guess: Juhatsu Kyoda did not practise seiyunchin for example, but he did practise nepai. Did Kyoda choose to incorporate the opening move of a Miyagi form (seiyunchin) or did Miyagi choose an element of a Gokenki form (nepai) practised by Kyoda and Mabuni, among others? Or were the techniques congruent/coincidental...?

Next in Part 5: The origins of seipai kata


1. Choki Motobu Okinawan Kempo, Rising Sun Productions (September 3, 1995) ISBN-10: 092012917X ISBN-13: 978-0920129173

Copyright © 2008 Dejan Djurdjevic

Wednesday, September 3, 2008

The origins of goju-ryu kata: Part 3 - shisochin kata

Continued from Part 1 and Part 2 of this article.

Shisochin begins with 3 opening sanchin stances making it superficially similar to cluster H. However it departs from cluster H in almost every other sense: the kata is “symmetrical” and has a high proportion of “soft” techniques. Moreover the opening thrusts are performed as nukite — knife hand thrusts. While it is said that cluster H were originally practiced open hand, it is more likely that, like the uechi-ryu kata, the nukite where executed palm down to a point just below the attacker’s nipple, not with a vertical hand to the solar plexus as per shisochin.

All of this points to shisochin being from a very different source than the cluster H kata (and of the same family/source as saifa, seiyunchin, seipai and kururunfa). But what was that source?

As I stated in Part 2 of this article, shisochin might have existed in Okinawa before Kanryo Higaonna even left for China: there is written record of Seisho Aragaki performing a kata named "chisaukin" at a demonstration in 18671. Seisho Aragaki was, of course, Kanryo Higaonna’s first teacher. Could Kanryo Higaonna have taught chisaukin/shisochin to Miyagi, perhaps as an “extra” or “non-syllabus” kata? Or might Miygai have remembered forms taught to him by his first teacher Ryuko Aragaki (who would likely have known Seisho’s forms)? Either is a possibility: but the odds of today’s shisochin being faithful to the original Aragaki chisaukin seem slim at best. For a start, forms known to have been taught by Aragaki have an architecture and design (including the right side bias) that is far closer to cluster H than cluster M (see Part 1 of this article for a detailed description of, and discussion as to, these features). The same applies to the “feel” of the katas.

While shisochin might not be an exact rendition of Aragaki chisaukin or one of his other katas, it doesn’t mean that Aragaki’s techniques don’t form at least a part of today’s shisochin. On the contrary: an examination of surviving Aragaki kata show elements that are, if not identical, at least highly reminiscent of shisochin. For example, unsu – a kata known to have been taught by Aragaki – opens with 3 “shisochin-like” nukite, albeit it from neko ashi dachi. The video below is the modern shito ryu version. It is hard to know what the original Aragaki form might have looked like, but I note that the nukite feature even in the shotokan version. Generally speaking, the kata seems to vary little from school to school giving me some encouragement that we have a reasonable idea of Aragaki’s karate.

Shito ryu’s unsu kata

You will note that other “shisochin-like” features include:

rising elbow strikes

low counterbalancing pullbacks or shuto blocks/strikes

hiki uke (open hand “pulling” blocks)

open hand haito uke (ridge hand blocks)

Unsu is otherwise clearly a different kata from shisochin, following a different embusen, different techniques and a “cluster H-like” right side bias. However in the absence of a “chisaukin”, kata , might unsu have provided Miyagi with some of the techniques that found their way into his shisochin? Certainly Miyagi was friends with Kenwa Mabuni, so even if Kanryo Higaonna or Ryuko Aragaki never taught unsu to Miyagi, Mabuni could easily have done so...

Another “shisochin-like” Aragaki form is the kata sochin. Again, it opens with 3 strikes in neko ashi dachi (albeit these are closed fist punches). It has some similarity to shisochin, but I would argue that this is less than unsu. I will discuss sochin in the context of seiyunchin in the next Part of this article.

Importantly, one needn’t just look to Aragaki to find “shisochin-like” Okinawan kata: For example the following “Tomari te” version of rohai (a kata I have long suspected of having an influence on Chojun Miyagi’s system) features a start that is identical to the start of shisochin, right down to the 3 sanchin dachi...

Tomari te rohai

This “Tomari te” rohai is practised in Gohakukai, the karate style developed by respected karate teacher and researcher Iken Tokashiki who studied both goju-ryu under sensei Fukuchi and Tomarite under sensei Nakasone. How “original” this version of Tomari te rohai I do not know, but if it is it would be an obvious candidate for at least part of Miyagi’s shisochin (or at least a form that is inspired by the same “source kata” – perhaps the original chisaukin). Alternatively it is also possible that shisochin has influenced this “Tomari te” version of rohai…

In terms of Chinese arts existing at the time, shisochin has a passing similarity to the “Hakka” systems of white crane, bak mei and southern mantis. Researcher Akio Kinjo2 has suggested that shisochin kata has its origins in the cricket/mantis systems of Fujian and that the original characters may well have been which mean "cricket/mantis battle" (pronounced "shisauchin" in the Hokkien/Amoy dialect or "xishuaizhan" in Mandarin). This seems quite possible if one examines southern mantis forms which feature finger thrusts while moving in a stance equivalent to sanchin (it is important to note that southern and northern mantis schools are completely unrelated and look very different — there are no “obvious” northern-style “mantis” postures and hand movements in the southern school).

I think it is possible that Miyagi was exposed to southern mantis during his visits to China. At the very least this might have inspired him in the development of shisochin. Consider the following video examples:

Andrew Chung, a master of various Chinese arts, gives an impressive performance of southern mantis

Another southern mantis form – note the arm bars and their similarity to shisochin

If I were to express my gut feeling about shisochin, it would be that it is Miyagi’s synthesis of techniques from Okinawan forms (if not Aragaki chisaukin, then some other Aragaki kata and/or Tomari te rohai) – possibly with elements inspired by southern mantis that Miyagi saw in China. This might be pure speculation, but it fits with the available evidence...

Next in Part 4: The origins of seiyunchin kata


1. See Mario McKenna’s article “So what did you think you were doing”.
2. See Joe Swift's article "The Kempo of Kume Village" in Meibukan Magazine No. 6.

Copyright © 2008 Dejan Djurdjevic