Gentlemen, I love kata applications as much as any of you, but I certainly won't ignore someone the stature of Ota Sensei when he claims certain moves in his lineage aren't meant to have an application to them.
My lineage in goju-ryu karate shares the same opinion as Ota. Sometimes a movement is strictly to practice balance or strength or <gasp> even for artistic merit. Nagamine, the founder of matsubayashi shorin-ryu, was an Okinawan folk dance enthusiast. His line, of which Ota is a part of, certainly believes in kata as a transmission of fighting techniques and concepts but from their viewpoint, it's probably overdoing it to say EVERYTHING (as designed by the pattern's creator) has a martial purpose to it.
Of course that's just what our lines think. Your system or ryu-ha may say differently and that's fine. I just wanted to express the opinion from this side of the aisle.
OK, this post seems a good point to try to crystallize the discussion around some more general issues about the interpretation of kata. The question of whether there are non-combat moves in various kata has been bothering me for a long time, and I've been trying to see the question in terms of the kind of system that kata correspond to, which is part of a more general family of systems in which units of
form are combined by certain rules, applying to such units, to yield
interpretations (which are of various kinds, depending what kind of system we're talking about). Two other systems of the same kind are genetics and human/artificial languages. And in all such systems, there are elements of form which don't correspond to parts of the interpretation. Here's what I mean:
(i)
Genetics
The formal units are certain large molecules; the rules of combination are determined by chemical valence and the physics of the molecular bond; the interpretation of these combinations of macromolecules represent sequences proteins (= tissue), and, at a larger scale, whole organisms.
(ii)
Natural and artificial languages
The formal units are words in NLs (such as human languages), or logical symbol types in ALs (such as logical constants and variables); the rules of combination are defined by the syntax of the language; the interpretation of these combinations of terminal symbols corresponds to truth conditions in human languages and logic, and computational operations in computer languages.
(iii)
Kata
The formal units are specific movements (labeled 'down block', 'double knife-hand block', 'middle punch' etc.), the rules of combination are... well, that's part of what's at issue here.... and the interpretation of these combinations corrends to the fighting moves a defender uses in responding to a violent attack. Basic to this way of putting is the well-known fact that Itosu repackaged karate for school use in a way that deflected attention from the extremely brutal effect of some of the movements Okinawan karate consisted of, using misleading labels such as `pivot', 'punch', 'stance' and 'block' for combat elements that might be more accurately described, respectively, as 'throw', 'neck twist', 'joint break' and 'damaging strike'.
Think of chromosomes, languages, and the kata of a given karate style (including TSD/TKD, i.e., Korean karate) as made up of certain terminal elements, where only certain combinations of those elements are allowed by the system. In each case we have a set of strings of elements—of certain large molecules, terminal symbols and movement types respectively—which are allowed, whereas others are disallowed, i.e., not part of the organic possibilities, forms of the language, or kata set. Each such string corresponds to a particular interpretation in the
semantics of the system (tissue, meaning, combat action). Now the question is, do we know of any elements in the first two that are semantically empty—that play no role in the interpretation?
The answer is, yes, definitely. We know that there are sequences of macromolecules on many chromosomes which appear to do no work: they may contain subsidiary information for the genetic 'readers', the ribosomes, which translate RNA/DNA into protein sequences, but we do not know just what it is they're there for, and it looks as though the tissues the chromosomes in question would encode would be same without these sequences. We know that there are expressions in artificial languages like prop logic which contain parts that are irrelevant to the final interpretation: if
p,q are propositions, then
pV
pV
q has exactly the same truth conditions as
pV
q, so that one of the iterations of
p here does no work. And in natural languages, the word
there in
There's a lion in the closet.
does no semantic work, because the conditions under which this sentence is true are exactly the same as those under which
A lion is in the closet.
are true.
There adds nothing. It seems to be true that in mapping from units of form, combined by syntactic rules, to a semantic result, there is plenty of room for semantically empty forms. Notice that the first of these sentences contains exactly one more word than is present in the second, yet the meanings are the same (there is no state of affairs in which the first is true in which the second is false, or vice versa). It follows that
there contributes nothing to the meaning of the sentence in which it appears; syntacticians call such semantically empty forms 'dummy elements'. They take up space according to the formal rules of combination of the language, the syntax, but contribute nothing to the semantics.
Kata are similarly a formal system—basic elements combined by rules to yield strings or sequences which denote something in a specific domain (in this case, combat actions). So it would not be especially strange if we encountered dummy elements in kata as well, like the
there in the above sentence, or the
that in
I believe (that) Robin is a spy. But it's an empirical question, not something that you can decide in advance.
The syntactic rules which govern English requires the appearance of an overt subject for each declarative sentence (as vs. languages such as Spanish, Italian, Greek or Mandarin). In the same way, the syntactic rules which govern certain classes of TKD forms require an H-shaped performance space, with symmetrical movements on the right and left of the 'crossbar' and the movements along each crossbar being mirror images of each other. Things like the embusen rule, the requirement that the performer of the kata ends up facing the same direction as the one in which s/he did when the kata began, and various other formal conventions, are all part of the rules for stringing movements into kata; they come with the territory, so to speak, in much the same way that English sentences require an overt subject.
Given this basic framework—that kata are governed by certain rules of combination which put basic elements (kihon movements) together to form sequences which have certain combat meanings (typically, each such combination has five or six or so 'meaningful' subcombinations, each of which represents a complete attack-initiation/defense-completion scenario)—one of the big problems with interpreting kata is to decide when a given movement or movement sequence is just part of the formal requirements, vs. being 'meaningful' (in terms of combat content). Take, say, taikyoku shodan. You can think of the first four moves (downblock+lungepunch—(180Āŗ turn)—> downblock + lungepunch) as simply a two sequence combat scenario to the left followed by the same movement to the right. A plausible
bunkai for the downblock+lungepunch sequence would be:
(A)
(i) Attacker, face to face with defender, grabs defender's forearm, shirt, etc. with his right hand; defender closes right fist over attacker's right hand, pivots 90Āŗ away from attacker pulling attacker's right arm straight, pinning attacker's arm by thrusting left forearm into attacker's right arm above the elbow and driving bodyweight into the pin to hyperextend the elbow joint, forcing attacker's upper body down.
(ii) Defender pulls left forearm out of the pin and delivers left-arm elbow spear-thrust strike to attacker's face, immediately followed by downward striking knifehand or hammerfist strike to attacker's lowered larynx, exposed by preceding elbow strike.
(iii) Defender applies muchimi-shift of striking left hand to gripping left hand, immobilizing the injured attacker's head by gripping his hair or ear, and steps forward to deliver finishing righthand punch to attacker's lowered head (strike to jaw or, pulling his head backward, again striking and damaging exposed throat).
The same bunkai are supplied in moves 3 and 4 of the kata, for a grab by the left-hand, i.e., the sequences 1/2 and 3/4 are just mirror images.
But it is entirely possible and realistic to see moves 3 and 4 as
continuations of the the scenario depicted in (1), as follows:
(B)
(i) Attacker, face to face with defender, grabs defender's forearm, shirt, etc. with his right hand; defender closes right fist over attacker's right hand, pivots 90Āŗ away from attacker pulling attacker's right arm straight, pinning attacker's arm by thrusting left forearm into attacker's right arm above the elbow and driving bodyweight into the pin to hyperextend the elbow joint, forcing attacker's upper body down.
(ii) Defender pulls left forearm out of the pin and delivers left-arm elbow spear-thrust strike to attacker's face, immediately followed by downward striking knifehand or hammerfist strike to attacker's lowered larynx, exposed by preceding elbow strike.
(iii) Defender applies muchimi-shift of striking left hand to gripping left hand, immobilizing the injured attacker's head by gripping his hair or ear, and steps forward to deliver finishing righthand punch to attacker's lowered head (strike to jaw or, pulling his head backward, again striking and damaging exposed throat).
(iv) Defender applies muchimi again, gripping the attacker's ear with his right hand, and attacker's right shoulder or arm with his left hand, and pivoting 180Āŗ to throw the attacker to the defender's right,
(v) then stepping in to deliver a third strike to the attacker's head with the left fist.
In other words, in Taikyoku Shodan, moves 3 and 4 can simply be the mirror of moves 1 and 2, as in (A),
or the four moves 1–4 can constitute a single longer fighting sequence as per (B).
I see this as a kind of 'parsing problem': in working out bunkai for maximally effective oyo, how should you group the separate movements recorded in the kata so that the result gives you the best applications? The problem is that kata, like natural languages, are ambiguous: a single string can have several different structures (e.g.,
I saw the student with the telescope). A similar problem arises in genetics, where it turns out that a single string of chromosomal molecules can be read by the ribosomal 'translator' in one of two or more different ways, yielding very different results depending on where the ribosome starts the reading operation that maps the chromosome into protein tissue.
Even in Okinawa, where the bunkai analysis traditions are unquestionably the best-preserved, it seems possible that different lineages could have interpreted the parsing of a given kata in two different ways, along the lines of (A) vs. (b). In the interpretation (A), the 180Āŗ pivot is a purely formal element, a part of the syntax of the kata but semantically empty (devoid, that is, of fighting content, only present as part of the 'display', as a transition to repetition of moves 1/2 on the opposite side); in this sense, it's like the empty pronoun
there in my example above from English. In the interpretation (B), though, the pivot has very definite combat content; it's a crucial part of the throw which, in (B), is a continuation of the counterattack following (iii). Two different interpretations, each of which might have gone with one of two different karate lineages. Neither is right or wrong, but the founders of the '(A)-lineage', as we can call it, and of the '(B)-lineage', simply had different takes on the matter. This wouldn't be surprising. We know that Motobu vehemently disagreed with some of Funakoshi's bunkai for Niahanchi, and, without taking sides, it seems possible that they came to different respective conclusions without ever having gotten formal directions from Itosu, who I can well imagine might have been less than explicit about the bunkai that he taught (Motobu certainly thought that Itosu had concealed the 'true' bunkai from Funakoshi, but then, he seems to have had a strong personal dislike for GF, so... ?)
My point is just that unless you've gone out of your way, as logicians and computer scientists have, to eliminate ambiguity from your formal system, it's entirely possible to have ambiguous parses of some given sequence in your system, with people deriving different views of the way in which the syntactic rules have combined the basic elements to yield complete structures, and therefore holding different views of how those structures should be interpreted. And even the most distinguished practitioners might take different perspectives on the right parsing, in the case of such ambiguity....