At a palace reception in Paris in 1856, everyone looked at the new arrival: an eighteen-year-old Italian aristocrat called the Countess de Castiglione.
She was stunningly beautiful, and more than that — she carried herself like a Greek statue come to life.1
Napoleon III, a notorious womanizer, could not help noticing. But that was all, for the moment. He preferred more hot-blooded women.
Then he kept seeing her over the following months, and became intrigued despite himself.1
What followed was a campaign conducted almost entirely in absences.
At court events they exchanged glances and occasional remarks. She always left before he could engage her in conversation. She wore stunning dresses. Long after the evening was over, her image would return to his mind.1
And the thing that drove him mad:
he apparently didn't excite her—she seemed only modestly interested in him.1
He began to court her assiduously. After weeks of assault, she finally succumbed.1
Note whose word that is. Greene calls the emperor's pursuit an assault and the countess's yielding a succumbing — and the entire arrangement runs the other way.
Even as his mistress she remained cold. He still had to pursue. He was never sure of her feelings. At parties she drew men's attention like a magnet, making him furiously jealous.
Eventually he tired of her and moved to another woman. Even so, while it lasted, he could think of no one else.1
In Paris at the time was Victor-Emmanuel, king of Piedmont — the countess's home.
Italy was divided into small states. With France's support it would soon become a unified nation, and Victor-Emmanuel harboured the secret desire to be its first king.1
In her conversations with Napoleon, the countess would occasionally talk of him: praising his character, describing his love of France and his strength as a leader.1
The emperor could only agree. Victor-Emmanuel would make the perfect king of Italy.
Soon Napoleon was broaching this idea with his advisers, then actively promoting Victor-Emmanuel for the throne as if it were his own idea—and eventually he made this happen.1
And the reveal:
Little did he know: his affair with the countess had been set up by Victor-Emmanuel and his clever adviser, the Count di Cavour. They had planted her in Paris to seduce Napoleon and slowly insinuate the idea of Victor-Emmanuel's promotion.1
Greene's description of the preparation is what lifts this above an anecdote about a mistress:
The countess's seduction of the emperor had been planned like an elaborate military campaign, right down to the dresses she would wear, the words she would say, the glances she would throw.1
And his reading of the mechanism:
Her discreet way of roping him in was a classic flanking attack, a seductive manoeuvre sur les derrière. The countess's cold beauty and fascinating manner drew the emperor on until he had advanced so far that he was convinced it was he who was the aggressor.1
That conviction is the front. Everything that mattered happened while he was occupied holding it.
Holding his attention to the front, the countess worked to the side, subtly conjuring the idea of crowning Victor-Emmanuel.1
And the counterfactual, which is the useful part:
Had she pursued the emperor directly or suggested the crowning of the king in so many words, not only would she have failed, but she would have pushed the emperor in the opposite direction.1
Drawn forward frontally by his weakness for a beautiful woman, he was vulnerable to gentle persuasion on his flank.1
Greene's transposition is stated as a general model for persuasion:
Maneuvers like this one should be the model for your attempts at persuasion. Never reveal your intentions or goals; instead use charm, pleasant conversation, humor, flattery—whatever works—to hold people's attention to the front. Their focus elsewhere, their flank is exposed, and now when you drop hints or suggest subtle changes in direction, the gates are open and the walls are down. They are disarmed and maneuverable.1
Two structural features are worth separating out, because they are what actually made it work.
The idea arrived as an aside. She talked about Victor-Emmanuel occasionally, in praise, in passing. An occasional remark cannot be argued with, because there is no proposition on the table to argue with.
And it was allowed to be adopted. The measure of success was that Napoleon promoted the throne as if it were his own idea. A conclusion someone reaches themselves is defended by them; one they accept from you is held on your credit and abandoned when you are inconvenient.
You need someone in a position of power to arrive at a conclusion, and you have no standing to propose it.
Stop trying to be persuasive, because being persuasive requires making a proposal, and a proposal from someone without standing gets evaluated on the standing.
The first move is to become interesting rather than useful. Occupy their attention with something that is genuinely about them — their expertise, their taste, a subject they enjoy being good at. What you are constructing is a relationship whose apparent purpose is not the thing you want.
The second move is restraint about access. The countess left before he could engage her. Availability closes a loop; slight unavailability keeps attention allocated forward, and attention allocated forward is attention not examining you.
The third is the aside. Never the proposal — the passing remark, the observation about a third party, the mild praise in a context where nothing is being asked. An idea delivered as an argument invites a counter-argument. The same idea delivered as an aside gets filed and later retrieved as a thought the person had.
And the fourth is giving it away completely. The moment your idea is being advanced by someone else as their own, do not correct the record. The authorship is the price, and it is the whole price — you can have credit or you can have the outcome.
Then the two guards this page will not skip.
Check whether you'd be content for them to learn how it worked. Napoleon never knew he had been placed. If the discovery would constitute a betrayal, you have crossed from advocacy into something else, and the technique cannot tell you which side you're on.
And notice who bore the cost here. Cavour and Victor-Emmanuel planned this campaign; the eighteen-year-old placed in the emperor's bed executed it. In the version of this you run, be clear about who is doing the exposed part and whether they chose it.
Virginia Oldoini, Countess di Castiglione, was a real figure, was Napoleon III's mistress, was Cavour's cousin, and was sent to Paris by him with an explicit political brief — that much is well attested and she herself was not discreet about it. [POPULAR SOURCE]
The causal chain is the part to hold loosely. French support for Italian unification, and Napoleon III's backing of Victor-Emmanuel, had substantial strategic causes — Austrian power in northern Italy, French interests in the peninsula, and the Plombières agreement that Cavour negotiated with the emperor directly. Greene's account makes a mistress's occasional remarks the operative cause of a European settlement. Her influence was real and was one input among many, and the chapter needs it to be decisive.
And the account is written entirely from the male principals' side. The countess appears here as an instrument — placed, planned down to her dresses and glances, deployed. She was eighteen, she was sent by her own relatives to sleep with a head of state, and she spent the rest of her life on the far side of it, increasingly isolated. Greene's framing has her as the strategist executing a flanking maneuver; the arrangement she was inside had her as the maneuver. Both readings are available from the same facts, and the chapter takes only the flattering one.
The technique's success condition is also its unfalsifiable part. Napoleon promoting the idea as his own is offered as proof the insinuation worked. It is equally consistent with an emperor who found the proposal independently sensible and was pleased to be encouraged. The chapter has no way to separate an idea successfully planted from an idea that was going to be adopted anyway — which is precisely the ambiguity that makes practitioners of this technique so confident of their own influence.
Against Metternich and the Close Embrace, built six pages earlier in this cluster, the resemblance is exact and Greene never remarks on it. Both are campaigns of years conducted inside an intimacy, aimed at a Napoleon, using a woman placed in the emperor's private life as a channel — Caroline Murat and Marie Louise in one case, the countess in the other. Both worked. Both are filed under different strategies.
The difference worth naming: Metternich was reading Napoleon I in order to find a weakness. Cavour was using an already-known weakness in Napoleon III to deliver a specific payload. Intelligence-gathering versus insertion — the same posture in service of two different operations, and the chapter's vocabulary of flanking covers both without distinguishing them.
Against Strategy 13: The Intelligence Strategy, this is what a close embrace is for once the reading is done.
And against Genghis Khan and the Shah of Khwarizm, the mechanism at the far end is the same one: the shah's own son delivered the conclusion the Mongols wanted him to hold, and Napoleon delivered his own to his advisers. The most durable belief is one the target manufactured.
Behavioral Mechanics → The Aside Beats the Argument. An idea introduced as a proposition activates evaluation; the same idea introduced as an incidental remark is processed without a defence being raised, and remains available later without a source attached to it.
The insight the pairing produces: the strength of a belief and the memory of where it came from decay at different rates, and that gap is what the technique exploits. People retain conclusions long after they have lost the provenance, which means an idea encountered casually enough, often enough, becomes indistinguishable from one they worked out. This also names the only real defence, and it is not scepticism — scepticism operates on arguments and nothing here is presented as an argument. The defence is provenance discipline: periodically asking of a position you hold, where did I first hear this, and who benefits if I believe it? That is an unusual and effortful habit, which is why the aside remains the most reliable persuasion technique available.
Business → Seeding a Decision You Can't Propose. Anyone without formal authority — a junior analyst, an outside adviser, a supplier — faces the countess's constraint: the proposal will be judged by the proposer. The standard workaround is to route the idea through someone senior enough to be taken seriously.
The insight neither domain gives alone: routing is not merely a credibility fix, it changes what the idea is by giving it a defender. An idea you propose remains yours and remains attackable by anyone who wants to attack you; an idea a senior person adopts becomes theirs and is now defended by their standing. This is why "getting credit" and "getting the outcome" are so often mutually exclusive in organisations, and why people who consistently shape decisions frequently have no visible record of having done so. It also explains the failure mode of the ambitious: insisting on attribution converts a durable adopted idea back into a contestable proposal, usually at the exact moment it was about to become policy.
Napoleon III advanced the crowning of Victor-Emmanuel to his own advisers as his own idea, and it was the whole point. Which means the measure of a successful influence operation is that it leaves no trace of itself in the mind of the person influenced — and the corollary is that the ideas you are most certain you reached yourself are exactly the ones whose provenance you have never checked.