Greene's opening claim about negotiation is the whole chapter in one sentence:
People will always try to take from you in negotiation what they could not get from you in battle or direct confrontation.1
And a second, about the vocabulary they will use:
They will even use appeals to fairness and morality as a cover to advance their position.1
The instruction that follows is not about tactics but about category:
Do not be taken in: negotiation is about maneuvering for power or placement, and you must always put yourself in the kind of strong position that makes it impossible for the other side to nibble away at you during your talks.1
The chapter's most useful piece of analysis is its account of why people handle this badly, and it is unusually sympathetic.
On one level, war is a relatively simple affair: you maneuver your army to defeat your enemy. . . . You may have to retreat here and there, but your intention is eventually to advance as far as possible.1
Negotiation is not simple:
On the one hand, you need both to secure your existing interests and to get as much on top of them as you can; on the other hand, you need to bargain in good faith, make concessions, and gain the opposing side's trust.1
To mix these needs is an art, and an almost impossible one, for you can never be sure that the other side is acting in good faith. In this awkward realm between war and peace, it is easy to misread the opponent, leading to a settlement that is not in your long-term interest.1
That is a genuine structural problem rather than a failure of nerve, and it explains why the chapter's answer is to dissolve the dilemma rather than to solve it.
Philip of Macedon's solution, which the chapter builds on:
to see negotiation not as separate from war but rather as an extension of it. Negotiation, like war, involved maneuver, strategy, and deception, and it required you to keep advancing, just as you would on the battlefield.1
And the reframing of trust that follows from it — the sharpest and coldest claim in the book:
Trust is not a matter of ethics, it is another maneuver.1
He would buy them from Athens later on, when he was powerful and had things to offer it in exchange.1
The generalisation:
Earning people's trust and confidence is not a moral issue but a strategic one: sometimes it is necessary, sometimes it isn't. People will break their word if it serves their interests, and they will find any moral or legal excuse to justify their moves, sometimes to themselves as well as to others.1
The full case is Philip of Macedon and Negotiation as Advance.
The chapter's operational core is a fork based on where you actually stand.
If you are weak, use negotiations to buy yourself time, to delay battle until you are ready; be conciliatory not to be nice but to maneuver.1
If you are strong, take as much as you can before and during negotiations—then later you can give back some of what you took, conceding the things you least value to make yourself look generous.1
That second half is the mechanism the whole chapter runs on: you cannot be generous with things you have not taken. Concessions are cheap in exact proportion to how much you hold, which means the generosity everyone praises at the end of a negotiation was manufactured at the beginning.
And Greene's dismissal of the standard worry:
Do not worry about your reputation or about creating distrust. It is amazing how quickly people will forget your broken promises when you are strong and in a position to offer them something in their self-interest.1
The chapter's argument against conciliation is stated as an empirical claim rather than a cynical one.
Out of a desire to avoid such unpleasantness, people will often try to be nice and conciliatory to those around them, in the belief that that will elicit the same response in return. But so often experience proves this logic to be wrong: over time, the people you treat nicely will take you for granted. They will see you as weak and exploitable.1
Being generous does not elicit gratitude but creates either a spoiled child or someone who resents behavior perceived as charity.1
And the corrected model of when people actually reciprocate:
People respond in a nice and conciliatory way only when it is in their interest and when they have to do so. Your goal is to create that imperative by making it painful for them to fight.1
If you ease up the pressure out of a desire to be conciliatory and gain their trust, you only give them an opening to procrastinate, deceive, and take advantage of your niceness.1
Greene's evidence is a run of American wars, and it is the chapter's most contestable passage.
In June 1951 the U.S. military halted an effective offensive against the Chinese army in Korea because the Chinese and North Koreans had signalled a readiness to negotiate.
Instead they drew out the talks as long as they could while they recovered their forces and strengthened their defenses. When the negotiation failed and the war was resumed, the American forces found that their battlefield advantage was lost.1
He reports the same pattern in Vietnam and to some extent in the Gulf War of 1991 — and identifies the motive as partly a desire to reduce casualties, partly a wish to appear conciliatory.1
What they did not realize was that the enemy's incentive to negotiate in good faith was lost in the process. In this case, trying to be conciliatory and save lives led to much longer wars, more bloodshed, real tragedy.1
The counterfactuals Greene asserts are strong ones: continued advance in Korea would have compelled negotiation on American terms; continued bombing would have forced the North Vietnamese to negotiate rather than procrastinate; a march to Baghdad in 1991 would have removed Saddam Hussein as a condition of peace, "preventing a future war and saving countless lives."1
The lesson he draws:
by continuing to advance, by keeping up unrelenting pressure, you force your enemies to respond and ultimately to negotiate. If you advance a little further every day, attempts to delay negotiation only make their position weaker. You are demonstrating your resolve and determination, not through symbolic gestures but by administering real pain.1
And the clarification that keeps it from being simple greed:
You do not continue to advance in order to grab land or possessions but to put yourself in the strongest possible position and win the war.1
The chapter's best practical section is about negotiating from genuine weakness.
Sometimes in life you will find yourself holding the weak hand, the hand without any real leverage. At those times it is even more important to keep advancing.1
By demonstrating strength and resolve and maintaining the pressure, you cover up your weaknesses and gain footholds that will let you manufacture leverage for yourself.1
De Gaulle in London in 1940 is the case: no constituency, no army, no recognition, and an all-or-nothing posture that produced all three. See De Gaulle and Manufactured Leverage.
Greene's rule from it:
if you are weak and ask for little, little is what you will get. But if you act strong, making firm, even outrageous demands, you will create the opposite impression: people will think that your confidence must be based on something real.1
You will earn respect, which in turn will translate into leverage. Once you are able to establish yourself in a stronger position, you can take this further by refusing to compromise, making it clear that you are willing to walk away from the table—an effective form of coercion.1
With a note on what to do if it's tested:
The other side may call your bluff, but you make sure there's a price to pay for this—bad publicity, for instance. And if in the end you do compromise a little, it will still be a lot less than the compromises they would have forced on you if they could.1
The chapter closes on Harold Nicolson's typology of negotiators — warriors who use talks to gain time and position, shopkeepers who work to establish trust and reach a mutually satisfying settlement.
the problem arises when shopkeepers assume they are dealing with another shopkeeper only to find they are facing a warrior.1
The asymmetry is what forces the recommendation:
The difficulty is that skillful warriors will make themselves masters of disguise: at first they will seem sincere and friendly, then will reveal their warrior nature when it is too late.1
In resolving a conflict with an enemy you do not know well, it is always best to protect yourself by playing the warrior yourself: negotiate while advancing. There will always be time to back off and fix things if you go too far. But if you fall prey to a warrior, you will be unable to recoup anything.1
In a world in which there are more and more warriors, you must be willing to wield the sword as well, even if you are a shopkeeper at heart.1
That gets its own page: Warriors and Shopkeepers.
Greene's Reversal is short, real, and points directly at Strategy 22.
in negotiation as in war, you must not let yourself get carried away: there is a danger in advancing too far, taking too much, to the point where you create an embittered enemy who will work for revenge.1
The evidence is Versailles — the Allies imposing terms on Germany so harsh that they "arguably laid the foundations for World War II."1
And the counter-model is Metternich, who a century earlier made it his goal to prevent the other side from feeling wronged.1
Your purpose in any settlement you negotiate is never to satisfy greed or to punish the other side but to secure your own interests. In the long run, a punitive settlement will only win you insecurity.1
You are going into a negotiation and your instinct is to open reasonably, build trust, and find a fair settlement.
Test that instinct against one question first: do you know which kind of negotiator you're facing? If not, the chapter's asymmetry argument applies — a shopkeeper meeting a warrior loses everything, a warrior meeting a shopkeeper can back off later. Playing it firm is the recoverable error.
Then establish your position before the talks rather than in them. This is the chapter's real mechanism and the part people skip. The concessions you make in the room are drawn from what you hold when you walk in, which means the negotiation was substantially decided by whatever you did in the preceding weeks — the alternative you developed, the deadline you set, the third party you brought in, the ground you took.
Then take more than you need, deliberately, so that you have things to give. Not to keep them — to spend them. The person who arrives with exactly what they want has nothing to concede and will end up conceding something real.
Then advance during the talks rather than pausing them. Greene's specific claim is that pausing removes the other side's reason to settle: if talking is cheaper than the alternative, they will talk indefinitely. Keep the cost of delay running.
If you are genuinely weak, ask for more rather than less, and understand why it works: confidence is read as evidence of something behind it. Then make refusal expensive — not by threatening, but by ensuring that calling your bluff has a visible cost to them.
And run the Reversal before you finish. What does this settlement make the other party want to do next? If the answer is get even, you have optimised the wrong variable, and the bill arrives in the next round.
The Philip and Metternich cases are well documented and the Nicolson typology is real. [POPULAR SOURCE]
The military counterfactuals are the weakest passage in the chapter and they are stated with the most confidence. "Had the United States continued its bombing campaigns in Vietnam, it could have forced the North Vietnamese to negotiate instead of procrastinate" is a live and heavily contested claim, not an established one; the North's demonstrated capacity to absorb bombing is precisely what the historical debate turns on. "Had it continued its march all the way to Baghdad in 1991, it could have forced Saddam Hussein out of office as a condition of peace, preventing a future war and saving countless lives" is remarkable in a book published in 2006 — three years into an occupation that had already tested the proposition. Greene presents all three as illustrations of a principle; each is an argument that would need making.
And the chapter has a self-contradiction it never resolves. Its central instruction is to keep advancing and never ease pressure. Its Reversal is that advancing too far creates an enemy who works for revenge, and it praises Metternich for ensuring the other side did not feel wronged. Those cannot both be operationalised without a stopping rule, and the chapter supplies none — that rule is the culminating point of victory, and it lives in the next chapter.
One thing worth crediting. The claim that generosity offered from weakness produces contempt rather than reciprocity is not merely cynical — it is consistent with a good deal of observed negotiating behaviour, and the chapter's formulation of the corrected model (people reciprocate when it is in their interest and when they have to) is more careful than its reputation would suggest.
Against Strategy 22: The Exit Strategy, immediately following, the two chapters are a matched pair that Greene never presents as one argument. This chapter says advance without pause; the next says the whole art is knowing when to stop. The Reversal here is the thesis there — and a reader taking the chapters in order gets an instruction and its correction with nothing joining them.
Against Metternich and the Close Embrace, Metternich is the chapter's counter-example on restraint and its exemplar of manipulation four chapters apart. The reconciliation available: he took everything he wanted and left the appearance of dignity intact, which is the settlement design this chapter recommends and its Reversal requires.
Against Strategy 18: The Turning Strategy, there is a genuine tension. That chapter argues that Caesar's strategic generosity — clemency, gifts, taking an enemy's side — was what defeated the militarily superior Pompey. This one argues that generosity produces spoiled children and resentment. Both are Greene, three chapters apart. The distinction available and unstated: Caesar was generous from strength, after winning; the conciliation this chapter attacks is offered from weakness, in order to obtain something. Generosity is a settlement instrument, not a negotiating one.
Business → BATNA as the Thing That Was Built Beforehand. Negotiation theory's central concept — your best alternative to a negotiated agreement — matches this chapter exactly: your position at the table is a function of what you have outside it, and skill in the room is a small term.
The insight the pairing produces: this is why negotiation training so consistently underperforms, and why the people who are good at it are rarely the ones who studied it. Tactics operate on the margin of a position that was fixed in advance; a strong alternative makes an unskilled negotiator effective and a weak one makes a skilled negotiator articulate about losing. Which means the actionable version of Greene's "advance before and during negotiations" is not aggression but scheduling — the work that determines the outcome has to happen in the weeks before anyone books a meeting, and by the time the meeting is in the calendar most of the available leverage has already been won or forfeited.
Psychology → Reciprocity Has a Precondition Nobody States. Reciprocity is among the most robust findings in social psychology, which is exactly what makes Greene's contrary claim interesting rather than merely contrarian.
The insight neither field gives alone: reciprocity operates between parties who perceive themselves as roughly equal in standing, and concession from perceived weakness is not processed as generosity at all — it's processed as information about what can be extracted. The reciprocity literature's experiments are overwhelmingly conducted between peers, which is why the norm looks universal. Greene's negotiators are in explicitly asymmetric relationships, and there the same act reads as capacity rather than as gift. This resolves the apparent contradiction and yields a usable rule: a concession only generates reciprocal obligation if the recipient believes you could have withheld it — so the sequence matters more than the substance, and generosity offered before you have demonstrated the ability to refuse buys nothing.
You cannot be generous with things you haven't taken. Which means every settlement praised for its magnanimity was manufactured earlier, by someone who took more than they needed specifically so they would have something cheap to give back — and the person who arrives at the table holding exactly what they want will end up conceding something they actually needed.