Erick Mead

I agree - (though in timing O Sensei had a different, perhaps merely nuanced view, and let's just say that interval (maai) means much more than just relative points in time, IMO).

However, I did not say correct connection was sufficient, but it is absolutely necessary. Connection is the gate into the opponent's structure. By correct physical connection I mean that where, for instance, compression is the active mode of a particular connection -- the outer surface of the connection is extended to take up the slack toward the opponent's center (through the limb or wherever point of connection is obtained). A tension connection must take up the slack from the opponent's center. In a wrist grab, the correct compressive connection slides the whole skin along the forearm structure until it is taut below the point of the grab. In a tension connection, the connection draws the whole skin along the forearm structure until it is taut above the point of the grab.

Good connection need not be fixed like a grab, and can slide along in a dynamic connection just as well -- though obviously requiring substantially more art to perform. The sliding engagement with a strike - literally and figuratively shears like a pair of scissor blades. It can be compressive: i.e. -- the blades (limbs) converging along their length from the initial point of contact, like kiri-otoshi. It can be tensile, like suri-age, where the blades (limbs) diverge along their length from the initial point of contact.

Canonical tenchinage uses one arm in compressive connection and the other in tension, and torques the whole body physically off its supports while interfering with structural reflexes that keep stability (see below about that). Aiki-age also uses both arms, but connects in both compressive and tensile modes simultaneously at each single point of connection. Through proper rotation of the connection that will establish one end of a torsion in each arm reaching though the body to the fixed support on the ground.

Aiki age occurs because the rotations twist along the particular spiral pathway where the golgi tendon organs and muscle spindles provoke the extensor spinal reflex (as does sankyo) putting uke up on his toes, or if in seiza rising off his seat. This reflex is the inverse to the other spiral pathway, which provokes the flexor spinal reflex (as in kotegaeshi or nikkyo), which causes the knees to involuntarily buckle. Even a simple rising hand metsubushi can be a connection without any physical contact at all, and done correctly prompts the visual flinch reflex which has a similar flexor buckling response that is still faster than the voluntary motor cortex can intervene to compensate.

Tenchinage interferes with stability once connected by applying stress (or moving -- they are equivalent) along a path that would requires uke's stabilizing action along that spiral line. But tenchinage is already potentiating the opposite reflex action along that line. IOW, when uke's voluntary effort to resist the instability occurs normally, it simply tightens the line that takes the sub-potential reflex action that tenchinage is potentiating -- and either triggers it over the edge into fully blown reflex action -- opposite to the necessary stabilizing action, and <boom> he falls down -- OR his own stability effort and reflexes get locked into mutually negating conflicting signals at the cusp and he can do little to respond at all.

FWIW I see this where kokyu expressed as in tekubi furi (the oscillations of structural tremor as in a strike termination) help keep his system fracked at that point. You can train to DO it as it happens, but it is hard to DO it when you decide to -- it is a self-reflex you can learn to trigger but not really to voluntarily use as such. YMMV

The structurally provoked reflexes are an order of magnitude faster than untrained voluntary motor action, and about three times faster than highly trained motor actions (e.g. - punching). The visual flinch flexor reflex is about six times faster than ordinary voluntary action, and about twice as fast as trained action. Or, so the literature says - but it sure works out to something like that.