Gernot Hassenpflug wrote:
Your point about the "chaotic figure of eight" illustrates that dillemma. How can we control that precisely, how can we train that more strongly and use it's power less randomly?
Control in the sense of dictating what happens in aiki is something I have learned to abandon in favor of control over accepting what happens as given. Masagatsu agatsu. Timing, maai, sente -- go no sen, sen no sen sensen no sen, all of these are useful -- but descriptive, not prescriptive -- they are what happens -- not what is done.
Tenchinage is an example where this gyroscopic issue also can be seen. If I held a spinning bike wheel with its hub axis horizontally in front of me (top edge spinning toward me), and push forward on the right side and pull back on the left side, the wheel's axis tilts toward vertical but neither side comes closer or retreats. This motion of my hands and the sense of expression of kokyu in them is equivalent to tenchinage. The manipulation of partner's center in kokyu tanden ho is similar, just with active intelligence feedback.
Practice refines this exploitation of continuous rotation into the exploitaiton of instantaneous rotation, and then, ultimately, if O-Sensei's and other adept's experience be believed, into virtual rotation. Ain't there yet by a long shot, although I can see it. Still fleshing out the second part.
Gernot Hassenpflug wrote:
In other words, the internal mechamics of movement over the external ones.
I think they are in fact the same. O-Sensei said that there are no secret techniques in aikido -- that it is all there to see. What is hidden then, begins as visible. What begins as external, ends as internal. What begin as continuous ends as instantaneous. What begins as actual ends in virtual. What is big, gets small.
Kokyu tanden ho/kokyu dosa exercise is at the heart of this for me. It is one of the reasons that I would ask skeptics to be more explicit in their skepticism of the operation of kokyu. In the katatedori kokyu ho example I gave, the initiating attack and the ending response share the vertical axis, both planes intersect on the vertical axis which provides the medium for converting from one frame to the other, by entry to the center and rotation about it (irimi tenkan).
By accepting pertubation of my own center I touch more deeply the center that is perturbing it, and by joining it, the resulting action must reflect the product of two wills joined, not one overriding will or two wills in contention. If his will is, say, oriented (x,z) to establiosh a plane of attack, then the first component of my will can be oriented with his in (z), and I am free to impart or alter translation or rotation without resistance by a second component of my will in (y); Alternatively, if I blend with him in (x), the second component of my will is free to act without resistance in (z). Tenkan begins in irimi and irimi ends in tenkan. Tthe attacker's two axis planar motion is converted to a three-axis tumble.
If I am right, there are no shortcuts, just better comprehension by dint of serious study and contemplation of these things.