On the eve of a river battle in third-century China, a strategist is accused of secretly working for the enemy. His commander sets him a test: produce a hundred thousand arrows for the army within three days, or be executed.1
It is not a test. It is a sentence with a formality attached. No workshop in the world makes a hundred thousand arrows in seventy-two hours.
Chuko Liang does not try. He takes a dozen boats, has bundles of straw lashed along their sides, and waits for late afternoon, when mist always covers the river. Then he floats the boats toward the enemy camp.
The enemy, seeing shapes in the fog and "fearing a trap from the wily Chuko Liang," will not send boats out to engage. They shoot instead — from the safety of the bank, in volume, harder as the shapes drift closer. After several hours the men lying flat on the decks turn the boats downstream, and Chuko Liang collects his hundred thousand arrows.1
Take the position seriously before admiring the escape.
The commander has framed the demand so that failure is death and the demand cannot be met. Chuko Liang is not being asked for arrows; he is being disposed of, with the appearance of a fair chance.
And notice what he does not do. He does not protest the accusation, appeal, or explain that manufacture is impossible. He does not attempt the task and fall short honourably.
He accepts the terms exactly as written — a hundred thousand arrows, three days — and then finds a supplier who is not on his side.
The order specified a quantity and a deadline. It did not specify manufacture. Everything that follows is contained in that gap, and the gap was always there; what was rare was reading the sentence as a set of conditions rather than as a task.
The apparatus is almost nothing: a dozen boats, straw, and a weather pattern.
The straw is the only engineered component, and it is doing something precise. It is not armour and it is not disguise — it is a medium that retains what is fired at it. Arrows that hit wood bounce or splinter. Arrows that hit thick bound straw stay put and can be pulled out later.
So the boats are collection devices dressed as targets, which is a much stranger object than a decoy. A decoy absorbs an attack you want spent. These absorb an attack you want delivered.
The mist supplies the condition that makes shooting the right response, and it is not manufactured either — Greene notes it "always" covers the river in the late afternoon.1 Chuko Liang did not create a single element of the situation. He arranged existing elements so that the enemy's ordinary behaviour produced his supply.
This is the part worth being clear about, because the case is usually told as though someone was outwitted.
The enemy commander is looking at unidentifiable shapes emerging from fog, at dusk, moving toward his camp, associated with an opponent famous for elaborate deceptions. His options are to send boats out into low visibility against an unknown force, or to stand on his own bank and put fire onto the approaching shapes.
The second is correct. It is what any competent commander does. It costs nothing but ammunition, it risks no men, and it addresses the threat whatever the threat turns out to be.
Which is the finding: nobody made a mistake. The enemy's caution was appropriate, his response was proportionate, and the more professional he was the more arrows he fired. Greene's own note is that they "redoubled the rain of arrows" as the boats came closer — the escalation is the behaviour of someone taking the situation seriously.
The manipulation is not of anyone's judgement. It is of what their correct judgement produces.
Strip the case to its inputs and ask what could not have been substituted.
The boats could have been anything that floats. The straw could have been any dense fibrous mass. The timing — late afternoon — is determined by the mist and nothing else.
And the mist is not really the asset either. The asset is what the mist does: it makes the boats legible as a threat and illegible as an object. Visible enough to demand a response, unclear enough that the cheap response is fire rather than contact.
That is a very narrow condition, and it is the whole operation. Full darkness and nobody shoots. Clear weather and they see straw and boarders lying flat, and send boats. The window is the band in which an opponent can tell that something is there and cannot tell what.
Now add the second ingredient, which is not environmental. Greene's sentence is "fearing a trap from the wily Chuko Liang." The enemy's caution is not generic military prudence; it is specific to this opponent's reputation. A commander with no reputation for elaborate schemes floats boats out of the mist and someone rows over to look.
So the operation runs on two supplies, and only one of them is free. The mist recurs daily. The reputation took a career to build, is spent slightly every time it is used this way, and cannot be replaced within the three days the commander allowed.
You have been given a target you cannot hit by the obvious method, and the framing makes clear that missing it is the point.
The first thing to do is unusual and it is the whole technique: read the requirement again as a set of conditions rather than as a job. Not how do I do this — what exactly has been specified, and what has been left open? A number. A date. A form of evidence. Almost never a method, because the person setting it assumed the method was obvious and did not bother to close it.
Now look outward instead of inward. The instinct is to search your own resources for capacity you don't have. The move is to search the environment for something already producing the thing you need, in volume, for its own reasons.
Notice how uncomfortable that is. It means the solution will not look like effort, and when it lands, nobody will be impressed in the way they would have been by three days of visible labour. You will hear well, that's not really what we meant, and it will be true, and there is no answer to it.
Then the part that decides whether you can run it at all. What is the enemy's cheapest correct response, and what does it emit? Not what mistake do you want them to make — what will a careful, competent, well-advised opponent do, and is there any version of the situation in which the careful response hands you something?
And the honest accounting at the end, which the case makes easy to miss. Chuko Liang spent a reputation to get the arrows. The enemy shot because of who he was, not because of what they saw. If you have nothing like that — if the room has no prior belief about you at all — then the mist is just weather and the boats come back empty.
This is the second time in the build that Chuko Liang's reputation does the work instead of his forces.
The other is the empty-city ploy: a general with two and a half thousand men opens all four gates of his city, has the streets swept by twenty men in commoners' clothes, and sits on the wall playing a zither while a hundred and fifty thousand troops approach. The enemy commander stops, looks at the open gates and the calm man, and turns his army around.
Different event, different decade, same asset. In both, an opponent is shown something that admits two readings, and in both he chooses the reading that assumes a scheme — because of who is standing there.
That makes the pair unusually informative about what a reputation for cleverness actually is. It is not credibility, and it is not fear. It is a standing instruction to over-interpret: a bias in the opponent's inference engine that converts ambiguous evidence into evidence of design. And it is exploitable in exactly two directions — make him act when acting is expensive, or make him refuse to act when refusing is fatal.
The vault holds the empty-city episode from a scholarly source rather than this one; see The K'ung-ch'eng Chi — The Empty-City Ploy. Greene tells the two stories nineteen laws apart and never connects them.
The case appears in a chapter about keeping your hands clean, and Greene's framing is: "Chuko Liang would never do work that others could do for him—he was always thinking up tricks like this one."2
That is a strange lesson to take. The enemy is not doing work for him in any sense they would recognise; they are defending their camp. And the operation has nothing to do with the chapter's actual subject, which is arranging for blame to attach elsewhere.
What Greene does draw out is worth keeping, because it names the real requirement: "the ability to think far ahead, to imagine ways in which other people can be baited into doing the job for you," and then the general rule — "disguise your goal, shrouding it in mystery, like the strange enemy boats appearing dimly in the mist. When your rivals cannot be sure what you are after, they will react in ways that often work against them."2
Disguise the goal, not the act. The boats were fully visible. What was hidden was what they were for, and that is a lighter and more durable form of concealment than hiding anything.
🚩 SECONDARY WITHOUT PRIMARY · 🚩 SINGLE SOURCE. The episode is one of the most famous in the Romance of the Three Kingdoms, a fourteenth-century novel, and its historicity is contested. Greene presents it as an event, names no source, and does not distinguish the novel from the chronicles. The vault's scholarly material on Chuko Liang notes that his reputation as the era's surpassing strategist is itself substantially literary.
Tension with the vault's own portrait. Ts'ao Ts'ao — The Villain Who Was the Real Strategist argues, from Sawyer, that Chuko Liang "came to be seen as the era's surpassing strategist without any justification," while the man cast as the villain unified China. This page's central asset — the reputation that makes the enemy shoot rather than investigate — may be exactly the reputation that scholarship says was inflated. The inflation, if it is one, is not a flaw in the mechanism; it is the mechanism.
Open question the chapter avoids. The operation depended on an opponent's prior belief about Chuko Liang. Where does the first such operation come from, for someone with no reputation yet? Greene's instruction assumes an asset his own case shows to be a precondition.
Greene tells the Chuko Liang material twice — the empty city at Law 5, the arrows here — and the two tellings serve incompatible laws.
At Law 5, So Much Depends on Reputation, the empty city proves that a reputation is a defensive asset: it protects you when you have nothing else. Here, at Law 26, the arrows prove that a reputation is a productive asset: it can be converted into materiel.
Both are true and the second is the more interesting claim, because it implies a reputation is spendable rather than merely protective — and spending is a thing that draws down a balance. Neither chapter says what the balance is or how it refills. The empty city works because the enemy expects a trap; the arrows work for the identical reason; and each successful use is also a demonstration, to that enemy and everyone he talks to, of exactly what the trick looks like.
Greene's build-once problem shows here too: the same figure carries two laws separated by twenty-one chapters, with the shared asset unremarked in both.
Behavioral mechanics — Animal Decoy Misdirection
The scene on that page: a guard hears a twig snap in the dark and goes rigid, and for one second the night hangs on what he decides the sound was. Then a fox darts across the path. His shoulders drop, he mutters, he writes it off — and never knows a man was crouched ten feet away, or that the fox was released at that exact moment to hand his ears an answer he could live with.
The principle is stated there as giving the telltale sound an alibi: you do not remove the evidence, you supply an interpretation of it that the target can accept and act on.
Chuko Liang's boats are that mechanism with the sign flipped, and the flip is the insight.
The fox supplies an interpretation that produces inaction — nothing to see, stand down. The mist supplies one that produces action — a trap, don't approach, fire from here. Both are alibis for ambiguous evidence; the operator chooses which behaviour the alibi licenses.
What the pair generates is a general form neither states. The unit of manipulation is not the target's belief but the response their belief licenses, and those can be selected independently. You are not choosing what they will think; you are choosing which of several true-enough readings will be nearest to hand, and then harvesting whatever that reading makes them do. The fox harvests silence. The mist harvests a hundred thousand arrows.
And it exposes the constraint both cases share. The supplied interpretation must be one the target would have reached anyway under a slightly different prompt — the fox is a plausible cause of a snap, a trap is a plausible reading of shapes in fog. An alibi that has to be argued for is not an alibi, which is why neither operation involves saying anything to anyone.
Psychology — Messenger Effect & Source Credibility
The experimental finding is that identical content lands differently depending on who is understood to be behind it. Hovland and Weiss put the same article under two attributions and measured a large difference in persuasiveness with no difference in the message. Source changes reception, independent of content.
The boats in the mist are the same effect operating on a situation rather than a text, and the extension is where the value is.
The enemy commander receives an ambiguous stimulus: shapes, fog, movement toward his camp. What he does with it is determined not by the stimulus but by the attribution — this is Chuko Liang. Greene's sentence records the mechanism precisely: "fearing a trap from the wily Chuko Liang." Remove the name and the identical shapes produce a scouting party.
Here is what the pair yields. The messenger effect is usually described as making people believe you more. This case shows a second and more exploitable consequence: it makes them resolve ambiguity in your favour, which is not the same thing and is far more useful. Belief requires you to make a claim. Ambiguity-resolution operates on situations where you have said nothing at all — and it works hardest exactly when the evidence is thinnest, because that is when the prior does the most work.
Which supplies the resource model the chapter lacks. A reputation for cleverness is not stored credibility to be drawn on when you assert something. It is a standing distortion in how strangers read the neutral facts around you, and its practical value is highest in situations where you have almost nothing to show — a dozen boats, some straw, and weather.
Sharpest implication. No one in this case is deceived and no one errs. The enemy sees something genuinely ambiguous, reasons correctly, and takes the response that is cheapest and safest — and that response is the supply chain. Which locates the technique somewhere other than deception entirely: the operator's work is to find a situation in which an opponent's best available move emits something he wants, and the only thing he manufactures is the situation.
Generative questions