Home Articles of Interest Uechi Breathing & Cooperative Drills

Uechi Breathing & Cooperative Drills

by George Mattson

From the Newsletter archives: June, 2005


In my last newsletter I described an experiment that I hoped would validate what I had learned from Kanei Uechi and Ryuko Tomoyose involving Sanchin breathing and Uechi-ryu breathing. Although I felt comfortable with what I was doing and teaching, there was a part of me that questioned the validity and effectiveness of the training, based on what some of my colleagues and associates were saying on the forums.

On the second “thread” in Van Canna’s Self Defense Realities Forum, after eleven pages of posts, all the posters agreed that what I was teaching was OK or had merit and although new and different, warranted further evaluation and testing.

Needless to say, this consensus of opinion on such a volatile subject has been a huge breakthrough for a subject that up until now, saw few Uechi practitioners agreeing on anything. Conceding that there is something called “Sanchin Breathing” which teaches a  repeating breathing method and when used during “stressed” effort, enables the student to draw on and use in a “natural” manner (Uechi breathing) a variation of the Sanchin Breathing. (Which often is very similar to more recognized and recommended breathing methods. WHEW. . .)

Ok, one Uechi target eliminated. Now. . . If we could only find someone to defend the old and venerable “cooperative drills”!!! 🙂

And as though  “Doctor (MD) to be”, Greg Postal read my mind, in the mail today I received the today’s featured article,

Uechi-Ryu Yakusoku Kumite As A Self Defense Training Progression

I hope you enjoy the article as much as I did. If you wish to communicate with Greg and continue the discussion of the “cooperative” drills, please visit Bill Glasheen’s Forum and “get involved”!

George E. Mattson

Click “read more” to read the article on “cooperative” drills.




Why “Cooperative” Drills!!!Uechi-Ryu Yakusoku Kumite As A Self Defense Training Progression

The yakusoku kumite of Uechi-Ryu, Kyu Kumite and Dan Kumite, were developed with the goal of training students in such essential skills as ma’ai, tai sabaki, hyoshi and uke waza(distancing, body movement, timing and reception techniques respectively).

In recent years, these drills have been increasingly criticized by some proponents of “realistic training” as being stilted and worthless for developing an efficacious skill set for self defense. This point is well taken, especially in regard to thoseuechika who continue to practice these sets in precisely the way they initially learned them: as halting, staccato movements with a uniform pacing throughout. This error is compounded by those yudansha who practice both kumite at what they feel to be black-belt level by simply increasing the speed and/or force of the movements while retaining the same training methodology they used as white belts. The problem with this approach is the possibility that these individuals are fooling themselves (and in some cases their students) into believing that by practicing this way they are coming closer to reality and therefore better preparing themselves for “street self- defense.”

This raises the question: if there are so many dojo utilizing these training methods in arguably backward and counterproductive ways, why not eliminate these drills entirely? In fact, many dojo have done just that, most frequently advocating substitution by jiyu kumite as a method of teaching the same skills that Dan Kumite and Kyu Kumite claim to teach. The benefits and drawbacks of jiyu kumite have been much debated elsewhere and I will not discuss them here, except to point out that it is often extraordinarily difficult to address specific training goals utilizing jiyu kumite. Many dojo recognize this and will add short drills designed to teach techniques or inculcate principles which later can be employed in sparring, or in self defense situations. Though they may differ markedly from Kyu Kumite as practiced in that same dojo, these mini-drills are frequently not much different in composition from sequences in the yakusoku kumite. In short, I would argue that the reasons for continuing to practice yakusoku kumite are twofold: first, if properly utilized, they can in fact serve as valuable training tools; second, they are familiar in some form to Uechi students everywhere, and can thus form a common “language” which allows students from different dojo to train together.

Some of the best models of alternative training methodology for yakusoku kumite can be found in the Filipino martial arts. Due to a variety of cultural and economic factors, these martial arts are, on the whole, generations closer to the reality of fighting than are the Chinese/Japanese arts on which Uechi-Ryu is based. In the traditional Filipino arts, one needs only to look back a generation or so to find an instructor who was involved in challenge matches where victory and defeat were demonstrably clear. Leaving aside any moral implications of these practices, the pragmatic result was a culling of those techniques and training that left an escrimador the loser. It is therefore instructive to look at these training methodologies when reality-focused training is a primary goal. Like yakusoku kumite, the attacks and defenses are prearranged and well known to both students. A common example of this is sumbrada sometimes termed a “rolling pattern,” some version of which is utilized in many Filipino styles, notably Cabales’s Serrada Escrima. The Filipino model yields good results because the pattern is utilized as part of a training progression rather than remaining a static exercise. Any successful training progression must help develop the necessary basic tools in a beginner and then allow the more advanced student to bridge the gap between drill and reality. In contrast to the “traditional” advanced way of practicing Dan Kumite and Kyu Kumite – moving faster and more forcefully without changing anything else – the advantage of the Filipino model is the development of considerable fluidity (readily apparent to anyone who has seen skilled escrimadors “playing”) while training within a set pattern. Nevertheless, there are also students of the Filipino arts who are tremendously fast and strong when they “roll,” but who will be hit consistently when faced with an opponent who departs from the pattern. If a student skips too many steps in his/her mad dash on the road to reality in any martial arts practice, the building blocks are weak and will not hold up under the barrage of even a moderately skilled opponent determined to inflict damage. At the same time, if one keeps hammering away at the same basic drills ad infinitum without adapting them to meet more advanced training needs, one merely becomes an accomplished beginner regardless of what rank one holds.

As long as the concepts discussed above remain paramount, any prearranged drill (e.g. Dan Kumite, or the various Yakusoku Kumite of other Uechi organizations) can become the foundation for a successful training progression. The following example using Kyu Kumite is representative of the program I use with my students. The steps below are arranged in roughly the order in which I introduce them, but can also be mixed and matched to work on the development and/or sharpening of specific skills. In each step, changes introduced in previous iterations can be retained while incorporating new aspects into the drill, or students can return to the more traditional version with only the newest variable added. Initially, of course, students start off simply learning the basic sequences and performing them in typical “white belt” Kyu Kumite fashion: attacks aimed at non vital targets (e.g. sternum) for safety; one move per count; slow unfocused attacks. As students demonstrate basic competence, speed and focus of attacks should be increased, though attacks are still delivered singly with a count for each attack. Students should be encouraged even at this level to practice in a fairly realistic fashion within the confines of the basic sequences and strikes should land if the defense (i.e., ma’ai, tai sabaki and uke waza) is not sound. After students have achieved this basic level of competence in the drill, any of the variations enumerated below can begin to be added:

  • Uke (attacker) should now direct attacks at more vital targets such as face, solar plexus, and groin. Returning to non-vital targets for a brief time as each new segment of the training progression is added may be advisable for safety.


  • Same as A, but now uke launches attacks at variable times after the count. Attention should be paid to ensure tori isn’t anticipating the attack by moving or blocking early. Uke should be discouraged from making a game of this segment by “faking” the attack.

  • An entire sequence (e.g. the first half of #1) is performed in one direction after each count; uke performs all attacks, and then becomes tori on the next count.

  • Same as C, but uke should attempt to attack fluidly with no breaks in the attack other than those forced by tori (e.g. when tori has trapped uke’s side kick in #4). At this point, it will be necessary for tori to adapt traditional uke waza and tai sabaki toward “what works.” Also by this stage uke’s punches should be launched from sanchin kamae with no chambering, and should return to sanchin immediately after full extension.

  • Tori adds a takedown or restraining technique at the end of each sequence.

  • No count. It may be advisable to reintroduce a count as new aspects are introduced to the drill to allow controlled learning.

  • No count and should be performed with one partner as uke all the way through (i.e. #1 through #5), with the goal of “chaining” together all the attacks in one direction. Some sequences may need to be altered slightly to accomplish this.

  • Tori may try to interrupt uke and counterattackbefore what is traditionally the last attack in that sequence.

  • As in H, but after tori successfully interrupts the attack, s/he immediately begins attacking with another one of the sequences (thus becoming uke).

  • Sequences performed singly but uke may attack with any sequence s/he chooses.

  • Tori starts in sanchin kamae, uke starts the attack from a neutral stance with hands at his/her sides.

  • Both uke and tori start in a neutral stance with hands at sides.

  • Uke may “break the pattern” for one move at a time, adding one attack drawn from elsewhere in the drill but should then complete the sequence that s/he initially began.

  • Uke can begin with one sequence and end with another.

  • Uke may attack with a random mix of 2-3 techniques – all of which should be from Kyu Kumite.

  • Uke may change the attacks in the kumite to more “realistic” ones. Examples could include hook punches instead of straight punches in #1, varying the angle of attack for the club attack in #4, substituting a low round kick in #5, etc. This will necessitate tori changing some of the uke waza, some of which will not work for specific attack variations.

  • Two or more attackers (as in kanshiwa bunkai), each of whom attacks with a different sequence. Of primary importance is that tori remain aware of all of the attackers. To that end, s/he may need to alter the tai sabaki or ma’ai, and should ideally try to use the uke against one another.


Although the various aspects of the drills above can be introduced in a different order depending on the needs of particular students, the direction of the training progression should proceed in a way that takes into consideration both safety and skill development. To this end it is useful to balance the concepts of risk and predictability as aspects of the progression are introduced. As students progress in their training, either risk can be increased or predictability can be reduced, but care should be taken not to change both parameters at once.

The concept of maintaining correct structure when in the role of tori is crucial to this entire process. One should not make the mistake of interpreting this as an injunction to perform these drills as if they were kata, but rather to ensure that the techniques within the drill are being performed correctly regardless of which thematic aspect of the drill is being emphasized at the moment. Although the execution of the technique may not appear smooth or pretty, especially when uke is throwing multiple attacks in quick succession, the true test is one of efficacy. For example, assuming that the attacks are being performed honestly – i.e. with the intent of landing the strike – it should be rapidly apparent whether or not the uke waza has been performed correctly.

I have attempted to outline above what I consider to be the essential elements of any yakusoku drill which has self defense skill optimization as its goal. Given the limitations of the print medium, some of this may not be as clear as I would like. Please feel free to contact me with questions or comments.




Related Articles

Leave a Comment