Paris, 1643. Ninon de Lenclos is twenty-three and doing well: her lovers come from royalty and the aristocracy, they pay her properly, they entertain her, and — the detail Greene picks out — they treat her "almost as an equal."1
Then her mother dies suddenly and she is left with no family, no dowry, nothing to fall back upon.1
A kind of panic takes her. She enters a convent, turns her back on everything, leaves a year later, moves to Lyons.
When she reappears in Paris in 1648, the suitors come back in larger numbers than before, because she has been missed.
And she has changed the terms.
The system she puts in place divides everyone who wants access to her into two groups, and lets them pick.
Payeurs. The dukes and princes who want to pay may continue to pay. But *"they were no longer in control—she would sleep with them when she wanted, according to her whim. All their money bought them was a possibility."*2 If that meant once a month, so be it.
Martyrs. Those who did not want to be a payeur could instead visit for her friendship, her wit, her lute-playing, and the company of Molière, La Rochefoucauld and Saint-Évremond.3
And the martyrs have their own possibility. Periodically she selects from them a favori — a lover who pays nothing, to whom she abandons herself completely for as long as she wishes. A week, a few months, rarely longer.
The rule that makes the machine run: a payeur can never become a favori. A martyr can, and might wait a lifetime.3
The Interpretation is unusually clear-eyed about the trap she was in, and it is worth stating because it explains why the design is so specific.
The courtesan's problem was ownership. "The man who paid for the courtesan's services in essence owned her, determining when he could possess her and when, later on, he would abandon her."4 As she aged, fewer men chose her, and the options narrowed. So she had to accumulate a fortune young.
Which produced a second trap on top of the first: the greed this required "also lessened her allure, since the illusion of being desired is important to men, who are often alienated if their partner is too interested in their money."4
So the standard arrangement demanded that she pursue money and punished her for appearing to. Two incompatible requirements, and the ordinary courtesan's career is what happens when you cannot satisfy both.
Her system dissolves the contradiction rather than balancing it. The money is collected from a group whose payments buy no control, so taking it does not read as being purchased. And the desire is demonstrated in a channel where no money moves at all.
This is the case Greene explicitly names as the riskless grade of option-control, in contrast to Ivan:
Where the alternatives set up by Ivan the Terrible involved a certain risk—one option would have led to his losing his power—Ninon created a situation in which every option redounded to her favor.5
Run the branches.
Choose payeur. She gets the income that funds the salon, and — this is the part a cruder scheme would miss — the payeur gets something real too. *"The fact that Ninon would only sleep with them when she wanted to gave them a thrill unavailable with every other courtesan: She was yielding out of her own desire."*6 He is buying the one thing money normally makes impossible to buy.
Choose martyr. She gets the salon populated with the best minds in France, at no cost, and "the ultimate in power: She could surround herself with a bevy of admirers, a harem from which to choose her lovers."7 The martyr gets superiority over the payeurs and a live hope.
Choose neither. Nothing is lost. The system is opt-in and its scarcity is what recruits.
There is no branch where she loses, and — the harder design problem — there is no branch where the man feels he got the worse deal, because each group is compensated in the currency the other group cannot access. The payeur has certainty of access; the martyr has the possibility of being wanted. Neither can have both, and each is told the other's position is inferior.
Greene identifies the single component the whole thing depends on, and it is not the money:
The system, though, depended on one critical factor: the possibility, however remote, that a martyr could become a favori. The illusion that riches, glory, or sensual satisfaction may someday fall into your victim's lap is an irresistible carrot… That hope, however slim, will make men accept the most ridiculous situations, because it leaves them the all-important option of a dream.8
And the evidence is a name. The poet Charleval "never enjoyed Ninon's favors, but never stopped coming to visit."3
A man who attended for years, received nothing he had come for, and kept attending.
That is not a satisfied customer and it is not a deceived one. He knew the terms. What kept him was that the door was genuinely open and never opened for him — a real, small, non-zero probability, which is a completely different psychological object from either a certainty or a lie.
Greene names this mechanism once and does not give it a chapter. It runs on its own logic, separate from the illusion of choice: choice-control narrows the menu; this keeps one item on the menu permanently visible and permanently out of reach. See The Option of a Dream.
The reaction is the detail that proves the design rather than merely illustrating it.
As word of this system reached polite French society, Ninon became the object of intense hostility. Her reversal of the position of the courtesan scandalized the queen mother and her court. Much to their horror, however, it did not discourage her male suitors—indeed it only increased their numbers and intensified their desire.9
Two audiences, opposite responses, from the same information.
The court was scandalised because the arrangement inverted a status order — a courtesan choosing was a courtesan not owned. The suitors were drawn for the same reason: an arrangement she controlled was evidence that being chosen by her meant something.
The hostility was free advertising, and it advertised precisely the property that made her valuable. A system that produces outrage in one audience and desire in another, on identical facts, is a system whose value derives from the outrage — and she could not have bought that reaction, only provoked it.
⚠ Greene does not note the risk this carried. Scandalising the queen mother in seventeenth-century France was not a neutral cost; Ninon was in fact imprisoned in a convent by royal order at one point in her life. The chapter presents the hostility as pure upside.
You have something people want access to and you are being ground down by supplying it on their terms.
The instinct is to raise the price or ration the supply. Both keep you inside the same transaction — you are still selling access, just less of it, and the resentment accumulates on both sides.
Ninon's move is to build two doors with different currencies, and the discipline is in the rule connecting them.
One door takes money and buys certainty of a limited thing — a defined, bounded, reliably delivered service. Reliable, unglamorous, and explicitly not the good stuff.
The other door takes no money and buys eligibility for the thing you actually care about doing. The people through that door get your real attention, sometimes, when you choose.
Then the rule that makes it work rather than making it a pricing tier: the paying door cannot lead to the free one. The moment money can buy eligibility, the free door's currency collapses and you are back to selling access.
Two cautions the chapter does not give. Whatever sits behind the second door has to be genuinely available sometimes — Charleval waited a lifetime, but other men did become favori, and a door that never opens for anyone is discovered. And be honest with yourself about what you are running: this is a lottery, its main product is hope, and the people holding tickets are paying in years.
Strongest evidence. Dated (1643, 1648), named (Molière, La Rochefoucauld, Saint-Évremond, Charleval), with a stated mechanism and a named negative case — Charleval is unusually good evidence because he is a non-beneficiary who stayed. The two-audience reaction is checkable in principle against period sources.
Tension — this is the riskless grade and Greene never generalises it. He names the Ivan/Ninon distinction5 and then lists six methods flat. Documented at The Illusion of Choice.
Tension — the chapter presents the scandal as costless. It was not.
Tension — the mechanism's beneficiaries are described in the operator's vocabulary only. "The illusion of choice, married to the possibility of future good fortune, will lure the most stubborn sucker into your glittering web."8 Charleval, who attended for a lifetime and got the company of Molière and La Rochefoucauld out of it, may not have been the sucker in that arrangement.
🚩 [POPULAR SOURCE] · 🚩 SECONDARY WITHOUT PRIMARY — no memoir, no letter, no biography cited, for a figure with a substantial documentary record. 🚩 The terms payeur, martyr and favori are given as hers ("men she called her martyrs") with no source for the usage.
Open questions. Did the payeurs know the martyrs existed, and on what terms — because the whole design depends on each group knowing the other's position and preferring its own? And how many martyrs became favori? The ratio is the difference between a functioning lottery and a story.
This is Ninon's third page in this build, and the three do not describe the same person.
Ninon de Lenclos and the Marquis (Law 3) has her as a coach — advising a young man on a seduction and watching him destroy six weeks of work with one word. Ninon de Lenclos on Timidity in Courtship (Law 28) has her as an author, writing first-person advice about how women receive hesitation. This page has her as an operator, running a system on the men in her own house.
She is the only woman in this stretch of the book, and she appears three times in three different roles, which is more range than any male figure gets. Greene distributes her across laws without noting the recurrence. She is also in the Preface's roll-call of practitioners, and the PRD's recurrence table does not list her — a correction made during this cluster's Pass 6.
Against Ivan in the same chapter, the pairing is Greene's own and it is the most analytically useful thing in the law: the same technique at two grades of risk, with the constrained party in both cases inventing leverage they were not given. Ivan lacked the power to reform; Ninon lacked a dowry. The Reversal's admission that this law "works best for those whose power is fragile" is visible in both cases and stated only at the end.
Behavioural economics — the non-zero probability as the product. The Regret Lottery works on the finding that people respond far more strongly to a live-but-unclaimed possibility than to its expected value — a ticket that would have won produces behaviour change that a straightforward payment does not.
Charleval held a ticket for a lifetime, and the salon was the price.
The insight neither frame gives alone: lottery research treats the near-miss as a distortion in how people price probability. Ninon's system shows it functioning as a payment mechanism — the hope is not a side effect of the arrangement, it is what the martyrs receive in exchange for their attendance, and it is the only thing they receive. Which reframes what she built: not a filter on suitors but a currency she could issue at no cost, in unlimited quantity, backed by a reserve she rarely had to spend. The constraint on issuing it is credibility — some martyrs had to become favori, or the currency devalues — and that is the same constraint a central bank operates under. She was running a monetary policy with one instrument.
Craft — the reader kept one step behind. Reader as Proxy, Writer One Step Ahead describes the writer's management of a reader's anticipation — the piece works when the reader is always close to arriving and never quite there, because arrival ends the engagement.
Ninon is running that on people rather than on a page, and the comparison exposes the honest version of what she did. A writer who keeps a reader one step behind is not cheating them; the withholding is the experience being sold, and both parties know it.
What the pairing produces is the test for whether this arrangement is exploitative, which the chapter's vocabulary cannot ask. Greene calls the martyrs suckers and treats the hope as the con. But the craft frame says a sustained not-yet is a legitimate good when the waiting itself is worth having — and Charleval's waiting came with Molière, La Rochefoucauld, Saint-Évremond and the best conversation in Paris. He was not paid in hope alone. The arrangement is a con only if the antechamber is empty, and hers was the most interesting room in France. That is a real distinction, it separates Ninon from most of the operators in this book, and Greene's framing cannot see it because his vocabulary has only web and sucker.
Sharpest implication. The design's load-bearing part is not the money and not the scarcity — it is an impossible door that is genuinely, occasionally possible. A payeur could never become a favori; a martyr could, and mostly didn't. That non-zero probability is a currency the operator issues at no cost, in unlimited quantity, and the only constraint is that it must sometimes be honoured or it devalues. And the honest test of whether this is a con is not whether people are waiting — it is whether the antechamber is worth being in.
Generative questions.