A crow and his wife have a nest in a banyan tree. A cobra moves into the hollow trunk and eats the chicks as they hatch.
The crow loves the tree and will not move. So he goes to his friend the jackal, and they make a plan.
The crow-hen flies to a pond where the women of the king's court are bathing, with their pearls and necklaces and a golden chain laid out on the shore. She takes the chain in her beak and flies back toward the banyan with the eunuchs chasing her. When she reaches the tree she drops the chain into the hollow.
The king's men climb up after it, find the cobra's swelling hood, kill it with their clubs, take the chain, and go back to the pond.
"And the crow and his wife lived happily ever after."1
That is the marginal fable Greene prints beside Cleopatra, from the Panchatantra by way of R. G. H. Siu's The Craft of Power. It is not the technique the chapter teaches.
Run the chapter's own definition against it. A cat's-paw is "someone who does the dirty, dangerous work for you" — usually "a person from outside your immediate circle, who will therefore be unlikely to realize how he or she is being used."2
The king's men are certainly outside the crow's circle and certainly unaware. But look at what they are told, believe, or are made to want: nothing at all.
They are not persuaded that the cobra is dangerous. They do not learn the snake exists until they are three feet from it. They are not baited into hostility, given false information, flattered, or positioned relative to an enemy. Their beliefs are entirely accurate and entirely unmodified.
The only thing that changed is where a valuable object is.
Compare every other instrument in the law. Cleopatra's Romans are given reasons. Weil's intermediary is given a scheme. Mao's Chiang is given an offer. Jamil is given a secret. In each case the operator does something to the instrument's mind.
The crow does something to the map.
The jackal is a friend and an adviser and contributes one thing: the recognition that the king's retrieval behaviour is a reliable, repeatable, indifferent force which can be pointed.
That is not cunning about people. It is closer to engineering. The court will pursue a stolen royal chain. It will pursue it wherever it goes. It will remove obstacles in its path, without malice, as a matter of routine.
So the plan does not ask who can I get to kill the cobra? It asks: what already-moving thing passes through that hollow if I put the right object in it?
And it is worth noting that the fable makes the plan a collaboration. The crow could not have thought of it — he is the aggrieved party, and the aggrieved party thinks about the enemy. It took someone outside the grievance to think about the road.
The distinction matters because it changes what the technique requires and what can go wrong with it.
A cat's-paw is a person acting on modified beliefs. Its failure modes are the ones Greene lists: the instrument works out what happened, the screen is lifted, the manipulator becomes visible. All of these require somebody to have been given something that can later be examined.
The crow's method has none of that surface. There is no false belief anywhere to be discovered. The eunuchs will remember chasing a bird and killing a snake, and both of those things happened, and no reconstruction of the day produces a story in which anyone was used.
What is hidden is not a fact but a purpose, and purposes leave no residue.
The nearest thing in the chapter to this is Greene's own instruction to "disguise your goal, shrouding it in mystery," which he attaches to the Chuko Liang boats.3 But even there an enemy is reacting to a manufactured appearance. Here nothing appears at all.
The fable's opening constraint is the part that makes it useful, and it is easy to read past.
"The crow did not want to move, since he loved the tree dearly."1
So the obvious solution — leave — is refused, not unavailable. And the second obvious solution, fight the cobra, is unavailable: a crow cannot kill a snake in a hollow trunk.
That leaves an operator who cannot exit and cannot win, which is the position most people are actually in when they reach for something like this. And what the fable demonstrates is that in that position the entire remaining strategy space is about routing rather than force.
Notice too that the crow's contribution is not costless or safe. The crow-hen has to steal a royal chain in full view of the court, in daylight, and be chased the whole way home. If she is caught she is a thief in a story where the punishment is obvious.
The technique is often described as though the operator does nothing. She does the most dangerous thing in the fable. What she does not do is fight the cobra, and the risk she takes is of a completely different kind from the risk she was facing — theft rather than combat, at a moment of her choosing, against pursuers who want an object rather than her life.
The move is not risk-avoidance. It is risk substitution, and the substituted risk is one she is competent at.
There is something in your way that you cannot remove and cannot get away from.
A person, a policy, a piece of the org chart. You have tried the direct route and it does not work, and you have tried the appeal-to-reason route and that does not work either, and the reason both fail is that the obstacle is stronger than you inside the frame you are both standing in.
So stop looking at the obstacle and start looking at what else moves through that space.
Every organisation has forces that run on their own schedule and their own logic and have nothing to do with your problem — an audit cycle, a compliance review, a budget consolidation, a customer escalation path, a security process. They are indifferent. They do not care about your grievance and they will not take your side.
The question is not how do I get one of these pointed at my problem. That is recruitment, and it leaves fingerprints, and it usually fails because the people who run those processes can tell when they are being used.
The question is: is my problem already sitting on one of these paths, and does anyone know?
Watch what you feel as you work through that. There is a temptation, right at the end, to help it along — to add a detail, to make sure the right person is copied, to give it a nudge. That nudge is the whole difference. Without it you have described a situation accurately. With it, you have put a chain down a hole.
And be honest about the second thing you feel, which is that this method is available to you precisely because you cannot win directly. The crow's plan is not what a hawk would do.
The fable ends happily ever after and stops, which is what fables do.
Two things are left out of it, and both are the kind of thing this build has learned to look for.
The tree is now known. A hollow in a specific banyan is the place where the king's men found their chain and killed a snake. The crow's home has appeared, once, on the court's map. Fables do not have sequels, but organisations do.
The method is repeatable by anybody who watched. The eunuchs saw a bird lead them to a treasure in a hollow. That is a legible pattern, and legible patterns invite the second, more careful visit.
Neither of these is a flaw in the fable, which is doing something else. They are what the fable would owe if it were the strategic instruction Greene's placement implies it is.
The attribution is worth stating plainly because it is unusually layered: a fourth-century Panchatantra tale, retold in R. G. H. Siu's The Craft of Power, 1979, quoted as a marginal in a 1998 book about power.1
Siu is in Greene's own bibliography, and the vault holds him as a separate source. So this is one of the places where Greene's marginal column exposes his sources rather than decorating them: the material is coming from a named twentieth-century synthesis of exactly the same kind Greene is producing, and he prints it beside his own text without comment.
⚠ The marginal's heading is OCR-corrupted in the source file, reading "IIII (ROW COBRA AND" — one of several typographic corruptions the build has logged in marginal headings.
Attribution is unusually good, in the usual place. Named collection, named century, named modern retelling, named year.1 The body text beside it — Cleopatra's motives, her knowledge of what Ptolemy would do — carries no citation at all. The marginal is better sourced than the chapter it decorates, for the twentieth-odd time in this build.
The fable does not illustrate the law it is printed beside. Greene's law is about instruments who act on modified beliefs. This is about an object relocated into the path of an indifferent process. He does not remark on the difference, and there is no evidence he registered it.
🚩 The moral is not stated. Unlike the other complete fables in this cluster, this one carries no closing line. What it teaches has to be inferred, and the inference here is the page's own reading rather than the text's claim — flagged rather than smuggled.
Open question. The method requires a powerful indifferent process whose behaviour is predictable. Is that a general feature of organised societies, or a property of specific systems — and does the technique simply become unavailable where institutions are erratic?
The nearest case in Greene's own chapter is Chuko Liang's arrows, and setting them together sharpens both.
Chuko Liang manufactures a situation — boats, straw, mist — so that an enemy's correct response produces what he needs. The crow relocates an object so that a court's routine produces what she needs. Both operators avoid recruiting anybody and both leave no false statement behind.
The difference is what the operator has to supply. Chuko Liang needs a reputation: the enemy shoots because of who he is, and without the prior belief the boats are just boats. The crow needs no reputation at all. She needs one fact about the world — that the court retrieves its property — and one dangerous flight.
Which makes the fable the more portable of the two. Greene's version of this technique is gated on an asset that takes a career to build; the fable's version is gated on nothing but observation, and the book prints the second in the margin without noticing it is the more useful one.
Eastern spirituality — The Asura Boon-Loophole
The demon has done years of brutal discipline and earned a wish. He knows immortality is off the table, so he writes himself a contract with no gaps: not by a man, not by a beast, not in the day, not in the night, not inside the house, not outside, not with a weapon, not with bare hands. He closes every door he can think of and walks away feeling safe. Then a thing that is neither man nor beast kills him at dusk on a threshold with its claws.
That is the same operation as the crow's, run from the opposite side, and the pairing produces the general form of both.
The demon's protection is a specification, and specifications describe a bounded set of ways in. The god's answer does not break the contract; it finds the region the contract does not cover, and every clause remains true at the moment of death.
The crow's cobra is protected the same way — by being a snake in a hollow trunk, which is impregnable against birds. The crow does not break that protection either. She changes what is coming, and against clubs and a climbing man the hollow is not a fortress but a container.
So the shared principle, which neither the fable nor the chapter states: defences are written against an enumerated threat model, and the enumeration is always shorter than the world. Neither the demon nor the cobra is defeated by superior force in the category they defended. Both are removed by something that arrives from a direction their defence did not have a term for.
And it names what the operator is really doing. Not out-thinking an opponent, and not overpowering one — searching the complement of somebody's threat model, which is a much larger space than the space of attacks they have anticipated, and which grows every time they add a clause.
History — Gekokujo — The Principle of Displacement
Gekokujo — the low overcoming the high. A vassal is sent to reinforce his lord, turns the army around, surrounds him in a temple, and burns him alive. Oda Nobunaga died that way in 1582, killed by an officer who was supposed to be marching to his aid. For two centuries this was not an occasional crime in Japan but something closer to a structural feature.
Both the fable and gekokujo are inversions of a power gradient. What separates them is who supplies the violence, and the difference is the whole insight.
Akechi Mitsuhide had to bring an army. That is what makes gekokujo possible and what makes it catastrophic: the inferior must accumulate enough force to act directly, which means the act is attributable, which means the successor's entire subsequent career is spent defending a position taken by visible treason. Akechi held it for thirteen days.
The crow supplies no force at all. She has none. And because she supplies none, there is no act to attribute — the cobra is killed by people with no grievance, no relationship to her, and no memory of having done anything unusual.
Which yields a scoping rule about power inversion generally. Direct displacement converts a weak party into a target the moment it succeeds, because the force it had to gather is now the evidence against it. Routed displacement leaves the weak party weak — and that weakness, which was the original problem, becomes the protection, because nobody investigates a crow. The price is that you can only remove what happens to sit on somebody else's path, and you never get to choose when.
Sharpest implication. Every other instrument in this law is a mind that has been worked on, and every one of them is therefore a record: a false belief that can be examined, a scheme that can be traced, a screen that can be lifted. The crow works on the position of an object, and the technique has no surface at all — no lie, no persuasion, no participant who could later realise anything. Which suggests the most durable form of indirect action is not the one that fools people best, but the one that modifies the environment rather than anybody's beliefs about it.
Generative questions