You kneel in seiza, facing a partner. He grabs both your wrists and squeezes. Hard. So hard your wrists ache the next day. Your job is to raise your hands and tip him over.1
Every instinct says find the easy way. Wiggle your hands until you feel the gap in his grip. There is always a gap. Slide along it, and he topples with almost no effort. That feels like the answer. It is the wrong answer.1
The drill asks the opposite. Go straight into his strongest point. Push on the locked door, not the hinges. And the strange thing is, when you do it right, the locked door swings open.1
Kokyū dōsa is the named partner exercise for building internal power. The principle is kokyū chikara, "breath power." The drill is how you train it.1
The setup never changes. Two people kneel in seiza. One grips the other's wrists as hard as he can. The seated person inhales while raising open hands, then exhales and cuts down, tipping the partner over in any direction.1 [PARAPHRASED]
Here is the catch. The exercise is in nearly every aikidō school, and almost nobody does it right.1 When wrists are pinned with full force, raising your hands with ordinary muscle is close to impossible. So students cheat without meaning to. They circle the grip, find the line where the partner's hold is weakest, and tip him over along that groove. It works. It looks clean.1
But that is not kokyū dōsa. That is a seated version of a jūjutsu throw called tenchi nage. It trains a different skill entirely, and it never builds internal power, because internal power never gets used.1 [PARAPHRASED]
The corrective instruction is blunt: go against all of his power. Seek his strongest point, not his weakest.1 [POPULAR SOURCE]
Why would going at the strong point ever beat going at the weak one? On the surface it is backwards. The weak line is the path of least resistance. Physics seems to favor it.1
The answer is that the weak-line route bypasses the very thing the drill exists to develop. When you slide along the groove, you win with leverage and angle. You never have to raise your hands against real load. So the muscles and the breath-timing that produce internal power are never asked to do anything. You get good at finding grooves. You stay bad at kokyū chikara.1 [PARAPHRASED]
Going at the strong point forces the issue. There is no groove to hide in. The only way through is to apply power on the inhale as well as the exhale, to raise the hands under full resistance by drawing power in rather than pressing it out.1
There is a second trap, and it is in your head. When your partner seizes your wrists, your mind gets seized too. You look at your wrists and your attention sticks there. Once the mind is trapped at the point of contact, it cannot move, and neither can your hands.1 [PARAPHRASED]
The fix is to refuse the trap. Do not even think about your wrists. Look off into the distance. Put your attention on the whole body. Inhale and imagine you are growing so large your head touches the ceiling. The wrists are still gripped. But your mind is not gripped. And the hands rise.1 [POPULAR SOURCE]
That is the master move hiding inside the drill. The grip on your wrists is real. The grip on your mind is the one you can drop. Drop the second and the first stops mattering.
This page is the partner drill. The principle it trains lives on Kokyū Chikara, Internal Power — read that for the breath-power mechanism and the balloon-expansion model. Kokyū dōsa is where the principle gets hands laid on it.
The "do not think about your wrists" instruction is a working instance of Mushin, the No-Mind State — attention that refuses to stick to the point of contact. It also shares wiring with Physical Techniques for Mental Principles, where a bodily act is the only honest way to teach a mental one.
The drill is also Lovret's demystification engine running live. The same mind that builds the Unbendable Arm Debunk is at work here: there is no magic, only the right muscles and the right breath. See Ki, Life-Force Demystified for the wider deflation of mystical claims this drill quietly performs.
Picture two pairs of students running the same exercise side by side. Both are seated. Both have their wrists gripped at full force. From across the room they look identical.1
The first pair plays it smart. The seated student feels around the grip, finds the line where his partner's hold has no leverage, and tips him along that line with a flick of effort. The partner goes over easily. Both nod. It looked effortless because it was. This is the tenchi-nage groove — a real skill, a sitting jūjutsu throw, but the wrong skill for this drill. No internal power was demanded, so none was built.1 [PARAPHRASED]
The second pair does the hard thing. The seated student ignores the grooves. He drives straight into the partner's strongest hold. At first it does feel impossible — that is the expected reaction, and the standard cause is that his mind is trapped at his wrists alongside them.1 He stops staring at his wrists, looks off into the distance, inhales as if his head is rising to the ceiling, and cuts upward against the full grip. The hands rise. The partner tips. Tomorrow his wrists will be bruised, and that bruising is the proof he went at the strong point.1 [POPULAR SOURCE]
Lovret names an older, sharper variant: koden aiki happō nage. Here the partner stands while you stay seated. You extend your arms up, he seizes your wrists, and you throw him in any of eight directions. The line that gives the whole logic away: the harder he grabs, the harder he hits the floor.1 That sentence is the inversion in full. In the groove method, more grip strength is more obstacle. In the real method, more grip strength is more fuel. The opponent's commitment to his strongest point becomes the thing that throws him.
You are on your knees on the mat, seiza, partner an arm's length away. He reaches out and clamps both your wrists. The grip bites. You feel the squeeze go white at the knuckles, and your eyes drop straight to your own wrists — there they are, pinned, and for a second the whole world is two gripped wrists and the word impossible.
That is the moment. You catch your eyes going down and you lift them. You look past your partner, past the wall, off into the far distance, at nothing. Your wrists are still crushed. You stop caring where they are.
You breathe in. Not into your chest — into everything. You imagine the air going to your scalp, your spine, your heels. You are a balloon filling. You picture your head drifting up toward the ceiling, your whole body swelling outward in every direction. The grip has not loosened a thread. It does not matter now, because your attention is not down there with it.
Your open hands begin to rise. Not by yanking. By expanding. You inhale and draw power in as the hands lift against the full weight of his hold, and they keep coming up even though they should not. Then the breath turns over. You exhale and cut down with a feeling of cutting through, and your partner pitches over to whichever side you chose. He hits the mat. You kneel there, hands still open, and you notice you barely strained. Tomorrow the bruises on your wrists will tell you that you went at his strongest point, and went through it.
You are not doing kokyū dōsa. You are doing tenchi nage in disguise. Here is how to tell.
The throw felt easy from the first try, with no impossible phase — a sign you found a groove instead of going at the strong point.1 Your wrists never bruise, because you never loaded them.1 You caught yourself making small hand-position adjustments, hunting for the line of least resistance — that is the jūjutsu circle-around, not the straight drive.1 [PARAPHRASED]
The mental tell is just as clear. Your eyes keep dropping to your wrists. You can feel your attention stuck at the point of contact, and your hands feel welded in place when it does.1 If the harder your partner grabs, the harder it gets for you, you are still in the groove logic — in the real drill, the harder he grabs, the harder he falls.1
Everything here comes from one practitioner source — Lovret, writing in 1987.1 He is a teacher describing a teaching exercise, not a researcher reporting a study. Treat the claims as instruction inside a tradition, not as tested findings. [POPULAR SOURCE]
The internal tension worth naming: Lovret insists there is no magic, only correct muscles and correct breath. Yet the corrective instruction — "go at the strongest point and it opens" — reads almost mystical until you accept his frame that the groove method simply never recruits the right muscles. Whether the difference is purely mechanical, as he claims, or whether attention and expectation do real work he is not crediting, the source does not settle. He asserts the mechanical reading. He does not prove it.
Open question: how much of the effect is muscle recruitment and breath timing, and how much is the attentional shift away from the wrists? Lovret folds both into one mechanism. They might come apart.
Lovret is the sole voice on this page, so the tension is not between authors but between two pages in his own system. The principle page, kokyū chikara, tells you what internal power is: breath power, the balloon feeling, power applied on both halves of the breath. This page tells you how it gets trained into the body — through a specific partner drill with a specific, counterintuitive rule. The two are not rivals. One states the idea; the other is the only honest way to install it. Lovret is consistent across both: no ki, no mystery, just muscles and breath used correctly. The convergence is total. The drill exists precisely to keep the principle from becoming a parlor trick — which is exactly the fate Lovret warns the unbendable-arm demonstration suffers when it is taught wrong.
This drill keeps turning into a lesson about minds and pressure long after the wrists let go.
Behavioral mechanics — confront the strong point; the power of directness. Lovret's core instruction is to attack where your opponent is strongest, not weakest. That cuts directly against the standard tactical reflex, which is to probe for soft spots and exploit them. The structural parallel is to the operator move of going straight at someone's stated position rather than maneuvering around it — naming the objection out loud, pressing on the thing the other person has braced hardest to defend. In persuasion this is the counterintuitive truth that the defended point is often the unstable one: a person who has committed maximum force to holding a position has, like the partner gripping with both hands, locked himself into a single line he cannot adjust. The harder he grips, the harder he falls. The insight neither domain reaches alone: the groove-hunting student and the soft-spot-hunting persuader are making the same error in two registers — both win small, both train themselves to avoid the only contact that builds real capacity. Easy compliance found in a weak spot is the persuasion equivalent of the tenchi-nage groove: it tips the person over without ever developing the operator's actual power, and it does not transfer to a harder case. Going at the strong point is slower, it bruises, and it is the only thing that builds something that holds up under full resistance.
Psychology — the mind not trapped by the point of contact. The second instruction is mental: do not let your mind get seized along with your wrists. The trap is attentional capture. Pain and pressure pull awareness to a single point, the mind sticks there, and stuck attention cannot generate action. This is the precise structure of how acute stress and pain narrow the field of awareness — attention collapses onto the threat, and the collapse itself is what disables a wider, more resourceful response. Lovret's fix is to deliberately widen the aperture: look into the distance, attend to the whole body, refuse to let the gripped point own your mind. The structural parallel runs to any practice that breaks fixation under duress — the deliberate redirection of attention away from a hot spot so the larger system can keep functioning. The insight the pairing produces, which neither side states alone: the physical grip and the mental grip are two different grips, and only one of them is actually holding you. The partner controls your wrists. He cannot control where your attention goes — unless you hand it to him by staring at the contact. The drill is therefore a body-level rehearsal of a purely psychological freedom: that the point where you are being squeezed is not where you have to live. Drop the mental grip and the physical one loses most of its power, not because the force changed, but because you stopped feeding it your whole mind.
Both handshakes share one spine. In each, the obvious move — find the weak spot, attend to the pain — is the move that keeps you small. The drill trains the reverse in the muscles, where it cannot be faked.
The Sharpest Implication. The thing pinning you and the thing stopping you are not the same thing. Your partner has your wrists. What actually freezes your hands is your mind going to your wrists. Loosen the grip you control — the mental one — and the grip you do not control stops mattering. The locked door opens not because you forced the lock but because you stopped pushing where the lock is and started expanding the whole house.
Generative Questions.