I see this every year in my role as kendo club coach at Keio University, but when it come to university kendo clubs, the members gather together from all over the country. Even though we can raise suburi as an example of something that each of the students has been taught, the fact is that it varies wildly according to their home region. I felt that in order to be able to instruct all of them, it would be necessary to start by teaching them exactly what correct suburi was. At Keio University, there are not all that many students who achieved superb results in competitions during their high school years; in fact, there are many that are close to beginners [this is a somewhat relative description!]. I thought that by having them master correct basics as best they could during their 4 years here their skill level would rise. Taking up a shinai and striking each other is something that cannot be done without an opponent, but suburi is something that you can do by yourself. I feel that important factors in kendo are kamae and suburi. These days we have machines so we can easily do strength training, but in the old days they didn't have such things. It was most likely suburi itself that was strength training back then.

At every Japan Budo Forum, a research topic is presented. In Heisei 10, Keio University was used as the venue, and we performed the role of hosts. When we then wondered what we should present research on, we hit upon the idea of presenting something about suburi. I looked into the position of the downswing, while Yoshida sensei led those who got to work on the angle of the upswing.

I first turned my eye to the instructional methods of 'swing back until the shinai touches your buttocks' and 'swing up such that the angle of the shinai reaches 45 degrees'.

Talking about the '45 degrees' method, we were somewhat concerned as to whether one could discern what angle the shinai was at, since it was in a position behind the body where it could not be seen. We wondered whether it was really necessary to stipulate the angle to which the shinai should be swung back to, and whether it wasn't possible just to teach a naturally flowing backswing movement without specifying an angle. We also had our doubts as to which stopping position was best for the downswing position. When talking conceptually, suburi is often taught in terms of stopping the right fist at the height of the right shoulder. However, we were aware that when we did suburi we often stopped the kensen at the height of the opponent's head. If you stop with your right hand at the height of your right shoulder, the kensen comes to a halt at quite a high position. I carried out my research into the downswing of the shinai from a stance of trying to find out just which of these suburi methods was most effective.

As far as I am concerned, the suburi method in which you swing back until the shinai touches the buttocks, so called 'jogeburi' suburi, is a method used to instruct children. It's difficult for children to understand if you just tell them to 'swing straight'. By having them line up the shinai with the coccyx, you can teach them a straight shinai path.

In fact, in the suburi described in the All Japan Kendo Federation's 'Points for the Instruction of Young Children' there are only two types, 'jogeburi' and 'sayufuri'. Swinging straight down at an imaginary opponent is distinguished with the name 'shomen strike'. In other words, it is seen not as suburi but a movement that comes at the striking stage. Despite the fact that suburi where one stops the shinai at shomen is actually carried out, it does not exist as a term. In this research, we used the name 'air shomen suburi'.

Most likely 'shomen strike' refers not to suburi but to the movement that comes at the stage when bogu is worn and actual striking is carried out. In Nakano Yasoji's 'An Illustrated kendo Dictionary' there is something called 'advancing and retreating men strike suburi', which is probably safe to think of as shomen suburi.

Kendo suburi and basic techniques came to be standardised after the war, from around the latter half of Showa 30s to the Showa 40s. The central figure of that time was Nakano Yasoji sensei. Since 'An Illustrated Kendo Dictionary' was put out by Nakano sensei, it naturally became the basis of postwar instructional method.

If we go even further back from there, we arrive at Takano Sasaburo sensei's 'Kendo'. This 'Kendo' is the start point. As such, the makeup of many instructional texts that followed was the same as that of 'Kendo', and there are many texts that use almost the same terms.

At present, research into kendo is carried out from every possible viewpoint, but surprisingly, there has been almost no research or data on suburi.

For this investigation, we obtained the cooperation of five subjects with kendo experience, ranging from 3rd dan to 7th dan, and five subjects without kendo experience. Yoshida sensei also used the same subjects in his research.

Yes, Ueda sensei looked at an electromyogram (EMG), and I carried out analysis of movements using a high speed camera. For these observations, we first divided ways of swinging up the shinai into four types (see illustration). We designated swinging back as far as the buttocks as suburi 1, and from there suburi 2, suburi 3, and finally we designated swinging up to 45 degrees as suburi 4. We selected these 4 classifications based upon photographs of suburi that we had seen in various instructional texts and kendo magazines. Actually, while no matter which book you read the explanatory text is pretty much the same; the photographs of suburi are almost completely inconsistent. We looked at these photographs, and while there was a suburi method in which one touches the buttocks, there was also talk of swinging up such that the kensen would thrust into a wall behind you, with the shinai horizontal. Then again there was some where the shinai was swung so that it pointed diagonally up. We divided these into four types.

4 Types of Suburi Tested:

1. Swinging upward at a 270 degree angle
2. Swinging upward at a 225 degree angle
3. Swinging upward at a 180 degree angle
4. Swinging upward at a 135 degree angle

When Yoshida sensei said to me, 'It seems that there's also a method in which you swing back as if thrusting into a wall behind you,' I argued that that was ridiculous. This is because if you swing back as if to thrust into a wall, your elbows will end up going behind your head. I felt that this type of upswing was probably inefficient.

Before starting the investigation, we had our subjects actually practice the four types of suburi. We made the same subjects perform all four types. Next, we measured which of these swinging methods produced the greatest kensen speed at the point of swinging down, measured at phase 8 (see phase diagram).

Speaking from the results, suburi 3 was found to give the fastest kensen speed. The slowest was suburi 4. If I were to try to explain, it would seem that because with suburi 4 you are in a situation where the shinai has merely been thrust upwards, it is not possible to smoothly switch to a downswing action. In the end, all that you can do is swing using just the power of your hands.

Additionally, from doing this research we became aware of how difficult it is for humans to calibrate angles at a position not visible to themselves.

Speed Comparison Graph

During the investigation, when we had the subjects perform the four types of suburi, we instructed them to 'swing back to this position,' but the point to which they swung was rather varied. At first they swung to roughly the designated point, but there was a trend for their upswing to become steadily larger as they continued to swing the shinai. The result was that suburi 3 was the fastest, but their actual backswing was somewhere between suburi 2 and suburi 3. I think you'll see what I mean if you look at the graph of 'variation in kensen speed'. This graph is of subject M. In his case, suburi 2 is fastest. Next is suburi 1 and suburi 3.

Based upon these results, we tend to feel that while there is no problem as long as the angle is at a position that can be seen, trying to specify angles outside of the field of vision is not a good idea. Expressing it in terms of getting somebody to look from the side and find the rough angle is probably better.

We designated two kensen stopping positions. One was the position reached when stopping the right hand at the height of the right shoulder, and the other was the position reached when stopping the kensen at the position of the head of an imaginary opponent (see illustration). Since opponents come in all shapes and sizes, we went for roughly one's own head height. What we were looking for as a result was a movement close to that of actual combat. The fact is that suburi that has an effect on the actions of actual fighting is good, so the suburi that showed a balanced distribution of power use when viewed on kindenzu would be the best.

We investigated which of suburi 1 and suburi 2 used the muscles most effectively. We took measurements with an EMG, constructed a graph and looked at the results (see 'muscular electrical output of suburi 1 against suburi 2'). For this, we did not make any stipulations with regard to the backswing.

Looking at the results, with suburi 1, where the right hand stops at the height of the right shoulder, we see a situation in which tension remains in the right arm for a long period. Tension in the right deltoid was seen especially frequently. It continued to be tense even after finishing cutting men. On the other hand, with suburi 2, where the kensen was stopped at head height, the tension soon disappeared when cutting down.

In actual fighting, two step strikes and continuous strikes are essential. In order to move quickly to the next movement, it is essential to first shift to a state of relaxation from the state of tension that exists after making a strike. With suburi 1, the muscles are constantly tense, and the shift to the next action is not smooth. We can state that suburi 2 enables muscle usage more in line with the movements of actual fighting.

As an aside, we only specified the kensen stopping height for suburi 2, but when we started the experiment, we saw that most of the subjects brought their fists down to the solar plexus.

In this experiment, we stretched a piece of rubber at each of the stopping positions of right shoulder height and head height, and measured the tip speed when the kensen touched this rubber. In the case of suburi 1, we saw that the maximum tip speed occurred not when it touched the rubber, but prior to that point.

We would like people to think of Yoshida sensei's research on the backswing, and my own on the downswing, not separately but together. That is, that suburi performed with the upswing and downswing that resulted from this research is effective. In conclusion, effective suburi would be something like the following:

First, with regards to the upswing, 'from kamae, raise both elbows as far as they will go'. The position of the kensen is not specified.

However, there is one important condition here, which is not to loosen the grip of the left hand. If you loosen it, you end up with a different movement.

That's right; you must raise the arms whilst gripping the shinai properly. This is an important condition. If you think about it, it's doubtful whether suburi 1 from the upswing investigation can be performed with correct te no uchi in the first place.

That's right. As long as you don't loosen the grip, it's pretty difficult to get the shinai to reach the buttocks.

As for the downswing, you should 'swing down so that the kensen reaches your own head height'. It's best to just specify the kensen height, and not the height of the arms. Since everybody has a different physique, the position of their head will also vary.

One other thing that I can explain with confidence from the results of this research is that you must use the shoulders, then the elbows, then the wrists, in that order, such that it becomes a shoulder joint-centric movement. Best of all is to make full use of the snap of the wrists, rather like a whip. Something often seen with suburi where the right hand stops at shoulder height is people who go directly from the shoulder to the wrist. It ends up being a movement like a pole, not a whip.

If you do suburi in this pole-like state, the extra effort required leads to extra fatigue, and repeated many times, there is even the possibility of damaging the elbows. If the elbows extend flexibly, the burden on them should be less.

There is a teaching often used when striking men, 'after you've struck men, extend both arms'. You'll soon see if you try to do this that if you try to extend the right arm when the body is facing forwards then the left arm bends. To extend both arms, you have no choice but to take a hanmi stance. It's difficult to describe this as correct suburi.
Yoshida: If those teaching just say, 'Stretch out your arms, stretch them!' then it's difficult for those being taught as that's just what they will try to do. Something like 'you should end up with both arms extended in front of you' is probably a better expression.