You had the better case. Everyone in the room could see it. And then there was a conversation afterwards about how to actually implement it, and by the end of that conversation you had roughly half of what you'd won.
Nothing was overturned. You just discovered that winning and keeping are two separate events, and only the first one had an audience.
Greene's version of this is the coldest sentence in Law 15, and it is not about aggression at all: "Negotiation is the insidious viper that will eat away at your victory."1
Greene builds the strategic half of the law on Carl von Clausewitz, analysing Napoleon's campaigns:
"We do claim that direct annihilation of the enemy's forces must always be the dominant consideration... Once a major victory is achieved there must be no talk of rest, of breathing space... but only of the pursuit, going for the enemy again, seizing his capital, attacking his reserves and anything else that might give his country aid and comfort."2
Two separate instructions are packed in there and only one of them is the famous one.
The first is about priority — annihilation of forces as the dominant consideration, above territory, above position, above everything else that looks like winning.
The second is about time, and it is the harder one: no rest, no breathing space, immediately after a major victory. Which is precisely the moment when everyone involved most wants both, and has the best possible argument for taking them.
Greene supplies the reasoning in one sentence and it is entirely administrative: "The reason for this is that after war come negotiation and the division of territory. If you have only won a partial victory, you will inevitably lose in negotiation what you have gained by war."3
Notice what that is not. It is not a claim about enemies hating you, or about vipers growing venom, or about pity being fatal — the emotional argument that carries the rest of the chapter. It is a claim about what a negotiating table does to an incomplete result.
And it is falsifiable in a way the emotional argument is not. It predicts a specific thing: that gains held by force convert into concessions when a process begins that requires both parties to have standing.
Work out why the erosion happens, because Greene asserts it and does not explain it.
A negotiation requires a counterparty with something to offer. A totally defeated enemy has nothing — no army, no capital, no reserves — so there is nothing to trade and no table. The victor simply carves.
A partially defeated enemy has residue. And residue is not just capability, it is standing: the fact of still existing is what qualifies them to be talked to. So the first thing a partial victory produces is not a weakened enemy but a legitimate participant.
Then the asymmetry does the work. The victor is negotiating over marginal gains — the last increment of an advantage already largely secured. The defeated party is negotiating over survival. Each side's willingness to spend on the outcome is proportional to what the outcome means to them, and it means vastly more to the loser.
So the partial victor arrives at the table holding more and caring less, against someone holding less and caring more. That is a losing position in any bargaining, regardless of the military facts, and it is why battlefield gains reliably convert into concessions.
Greene's instruction — "give your enemies nothing to negotiate, no hope, no room to maneuver"4 — is aimed exactly at this. It is not about hurting them more. It is about not manufacturing a counterparty.
Worth separating, because the chapter fuses them: Clausewitz's argument survives with the emotional argument removed entirely.
Suppose enemies never hate you. Suppose the half-crushed viper simply goes home and gets on with its life, bearing no grudge at all. Clausewitz's claim is untouched — a partial result still converts at the table, because the conversion mechanism is about standing and asymmetric stakes, not about resentment.
That matters for what transfers. Everything in this law that requires you to become a person who cannot feel pity belongs to the other argument. This one asks only that you understand what a partial result becomes when the fighting stops, and it applies without modification to a negotiation about a budget.
Two places, and the chapter visits neither.
It assumes the negotiation is optional. Clausewitz is writing for a party that can choose to keep going. Most people cannot: the process starts whether or not you are finished, because someone above you convenes it. In that case the instruction do not stop halfway is not advice, it is a description of a condition you do not control — which is precisely Chiang Kai-shek's position in this same law, cut short by a Japanese invasion.
It assumes annihilation is available and priced. "Direct annihilation of the enemy's forces must always be the dominant consideration" — always, dominant, no cost term anywhere in the sentence. Clausewitz was analysing Napoleon, who eventually demonstrated the cost of exactly this doctrine in Russia. Greene quotes the doctrine and not the demonstration.
You have won something and the follow-up meeting is on Thursday.
Notice the feeling first, because it is the whole trap. You feel finished. The hard part is behind you, the argument landed, and Thursday is administrative — a formality about implementation details.
Thursday is where it goes.
So do the unglamorous thing this page is actually about. Before that meeting, write down exactly what you won, in specifics, as though someone were going to hold you to it. Not the principle — the mechanism, the number, the date, the person responsible.
Then look at your list and ask which items still require the other party's cooperation to be real. Those are the ones that will convert, because those are the ones where they still have standing. The items that are already done — announced, budgeted, in someone's contract — are not on the table, because there is nothing left to discuss about them.
That is the whole operational content, and it does not require crushing anybody: the difference between a total and a partial victory is not how much you took, it is how much of what you took still needs their agreement. Anything that does is not yours yet, and the meeting where you find that out is always framed as a formality.
The strongest thing about this passage is that it is the only genuinely structural argument in Law 15 — it does not depend on any claim about human psychology, and it can be checked against any negotiation you have been in.
Tension: the chapter runs two arguments as one. The emotional case (a wounded enemy's hatred grows) and this structural case (a partial result erodes at the table) point the same way in Greene's examples and would diverge elsewhere — against an enemy with no capacity for resentment but full capacity to bargain, or the reverse. Greene never separates them, so the law cannot handle either case.
Tension: Napoleon. Clausewitz derived the doctrine from Napoleon's campaigns; Napoleon's career ends as the standing counter-example to unlimited pursuit. Greene takes the doctrine and leaves the outcome. This build has now logged Napoleon appearing as both exemplar and cautionary tale within single chapters (Law 8) — here he is the source of a doctrine whose refutation he also supplies.
🚩 SECONDARY WITHOUT PRIMARY — the Clausewitz passage is quoted without work, book, chapter or translator, and On War is a large and much-disputed text whose surrounding argument includes extensive qualification about the relationship between war and policy. The excerpt as given reads as unconditional; whether it is unconditional in context is not checkable from this source. [POPULAR SOURCE] · [PLAUSIBLE — needs corroboration]
Open questions. Does the conversion effect have a threshold — is there a degree of victory past which the loser's standing collapses and the table disappears, and is it the same threshold Chiang missed at 87%? And in settings where you cannot decline the negotiation, is there a defensive version of this: a way to make gains non-negotiable by making them non-dependent, rather than by making the enemy non-existent?
Within Law 15 this passage is quietly at odds with the chapter's own closing qualification. Greene ends by saying the law is "not, of course, a question of murder, it is a question of banishment."5 But Clausewitz's mechanism does not care whether the enemy is dead or exiled — it cares whether they retain standing to be negotiated with. A banished enemy with residual claims is exactly the counterparty the argument warns about, and the qualification therefore does not qualify this half of the law at all. It only softens the other half.
The sharper convergence is with the law's parent page and its Siu counter-position. Siu's Law of Reversal says approaching total dominance summons countervailing power. Clausewitz says stopping short surrenders your gains. Between them they describe a corridor with a wall on each side, and neither names the width. Stop short and the table eats your victory; go all the way and you manufacture the coalition. The chapter's cases sit on both sides — Chiang short, Versailles long — and Greene reads only the first as a lesson.
Psychology — Loss Aversion and Asymmetric Valuation
Loss aversion's core finding is that losses loom larger than equivalent gains — the same amount of value hurts more going out than it pleases coming in, by a consistent factor.
Put that under the negotiation trap and it supplies the mechanism Greene asserts without explaining.
At the post-victory table, the two parties are not in symmetric positions with respect to the same quantity. The victor is negotiating over gains; the loser is negotiating over losses. Even holding the objective stakes identical, the loser is operating in the steeper part of the value function — every increment they concede costs them more than the identical increment is worth to the party receiving it.
The insight neither generates alone: this predicts that partial victories erode even between parties of equal skill and equal information, with no coercion and no bad faith. Greene's account implicitly blames the victor's softness or the loser's cunning. The pair says neither is required. The asymmetry is in the frame each side occupies, and it is created by the victory itself — which is a genuinely uncomfortable result, because it means the erosion is not a failure of negotiating technique that better preparation could fix. The only intervention that works is upstream, in whether a counterparty exists at all.
Business — The Monopoly Trap
That page reports Greene's own Reversal at Law 11: the weakness of making others depend on you is that you become dependent on them, and the monopolistic escalation has three failure modes — of which coalition is the strongest, because it is a manufacturing process. The monopolist gives scattered, mutually hostile parties the one thing they have in common.
Set that against Clausewitz and the two halves of a single problem appear.
Clausewitz says: an incomplete victory leaves a counterparty, and counterparties eat gains. The monopoly page says: a complete victory creates counterparties who did not previously exist, by supplying a shared enemy to parties who were previously fighting each other.
Together they specify the failure at both ends of the same dial, and identify what the dial is actually measuring — not force, but how many other parties can see themselves in the loser's position. Stopping short leaves one party with standing. Going all the way converts every observer into a party with standing, because the demonstration tells them what happens to people in that seat. Which reframes Greene's whole law: the question is never how completely should I win, it is who is watching, and does my completeness give them something in common that they did not have this morning. Empress Wu's sealed court had no such audience. Versailles had a continent of one.
Sharpest implication. A victory is only as durable as the portion of it that no longer requires the other party's agreement. Everything still requiring their cooperation is not a gain you are holding; it is an item on an agenda, and agendas are where gains are spent.
Generative questions.