A con man is working the president of a large bank in Omaha. The mark is rich, so the play is big: a $1,250,000 investment, of which Weil will supposedly put up $900,000 and the banker $350,000.
The banker hesitates. Weil has played golf with him, visited his home, gone to the theatre with him and his wife. He still will not commit.
Then, at dinner one evening, Weil is wearing perfume — Coty's April Violets.
The banker's wife thought it very lovely. "Where did you get it?" "It is a rare blend," I told her, "especially made for me by a French perfumer. Do you like it?" "I love it," she replied.1
What he does next:
The following day I went through my effects and found two empty bottles. Both had come from France, but were empty. I went to a downtown department store and purchased ten ounces of Coty's "April Violets." I poured this into the two French bottles, carefully sealed them, wrapped them in tissue paper. That evening I dropped by the banker's home and presented the two bottles to his wife. "They were especially put up for me in Cologne," I told her.1
The next day the banker comes to his hotel.
*His wife was enraptured by the perfume. She considered it the most wonderful, the most exotic fragrance she had ever used. I did not tell the banker he could get all he wanted right in Omaha… "She said," the banker added, "that I was fortunate to be associated with a man like you." From then on his attitude was changed, for he had complete faith in his wife's judgment … He parted with $350,000. This, incidentally was my biggest score.*1
The single most important fact and it is easy to skate past: the wife's experience was not a deception.
She smelled April Violets, liked it very much, received a large quantity of it, and enjoyed it. Coty's April Violets is a real perfume and it is presumably as pleasant in a French bottle as in a drugstore one.
Everything false is in the provenance — a rare blend, especially made for me, especially put up for me in Cologne — and provenance is not a property anyone can smell.
So the fraud sits entirely in a layer the recipient has no access to and no reason to test. She was not fooled about the perfume. She was fooled about what the perfume implied about the man who gave it to her, and that inference is the product.
Weil states his own method in the opening line:
As time went on I came to look for the little weaknesses… It's the little things that count.1
But the operational insight is structural rather than observational, and it is the reason this is Law 33's sharpest marginal.
The banker was resistant. His wife was not a party to the transaction at all.
Weil had spent weeks on golf, home visits and the theatre — direct approaches to a man whose job is evaluating investments and who was, correctly, unconvinced.
He stopped working on the decision-maker and worked on the person whose judgement the decision-maker trusted more than his own: "he had complete faith in his wife's judgment."
That is Law 33's find the weak link — see Find the Weak Link — executed on a domestic rather than a court structure, and it is more instructive than Richelieu's version because the weak link here has no idea any transaction exists. She is not being deceived about an investment. She is being given perfume.
Run the causal chain and notice how little the money is connected to the money.
The banker did not revise his view of the mining stock. Nothing about the investment changed. No new financial information was supplied.
What changed:
The endorsement is the product, and it was manufactured for about ten ounces of drugstore perfume and two empty bottles.
Which is the same arbitrage the chapter's verdict identifies — see Validation as the Universal Weakness — but with an extra step that makes it far more efficient. Weil did not validate the banker. He validated someone whose good opinion the banker could not discount, because discounting it would mean doubting his wife.
A recommendation from inside a marriage is not evaluated the way a claim from a counterparty is. It arrives through a channel with no scepticism installed on it.
The passage is first-person and it is Weil's own, dated 1875–1976.1
Two things follow and both should be held.
It is a confession, freely given, in a genre — the con man's memoir — that exists to entertain. Every detail that makes the story good is a detail the teller had an interest in.
And it is Weil's second appearance in three consecutive laws. He is Law 32's Authority — quoted approvingly on the value of lying, without comment — supplies this marginal at Law 33, and reappears in Law 34's body as an exemplar of aristocratic bearing.2
⚠ A working confidence man is the most-quoted living practitioner in this stretch of the book, and in one of the three appearances he is elevated to the position normally occupied by Machiavelli, Gracián and Clausewitz. Recorded as a fact about Greene's sourcing, not about Weil.
⚠ Heading OCR-corrupted in the source to "THE THINGS ON".
Someone you are trying to persuade has said no, repeatedly, on the merits — and they are right to, or at least not obviously wrong.
The Weil observation, stripped of the fraud, is a genuine structural point: you have been working on the wrong node.
Ask who this person's judgement defers to. Not who outranks them — whose opinion they treat as better than their own. A spouse, a former mentor, one specific colleague, a predecessor. Most people have exactly one or two, and they are usually not on the org chart.
A view arriving through that channel is not assessed the way your view is, because the channel was established for reasons unrelated to you and carries no scepticism.
That is a real and legitimate insight: if somebody you respect is genuinely enthusiastic about your work, their saying so does more than any amount of your saying so, and arranging for good work to reach the right people is not manipulation.
Where it becomes Weil is the manufactured provenance. He did not give the wife perfume she liked and let it stand; he attached a false story about where it came from, because the story was the part being sold.
The test is clean: would this still work if they knew exactly where it came from? Good work reaching a respected third party survives that. Ten ounces of drugstore perfume in a French bottle does not.
Strongest evidence. A first-person account with specific figures ($1,250,000 / $900,000 / $350,000), a named product, a named city, a named target role, and a stated method — and the operator identifies it as his largest score, which is a claim he had no reason to understate.
Tension — it is the operator's own account of his own cleverness. Every element that makes the anecdote satisfying is one Weil benefited from including.
Tension — the wife received exactly what she was promised in the only respect she could check. The deception is entirely in provenance, which reframes what was actually done to her.
🚩 [POPULAR SOURCE] · 🚩 SINGLE SOURCE · 🚩 MOTIVATED REASONING — a confidence man's memoir is the least reliable available genre for a confidence man's methods, and this one is quoted at length without qualification. 🚩 SECONDARY WITHOUT PRIMARY — no date beyond Weil's lifespan, no publication cited.
Open questions. Did any of this happen? The Omaha banker is unnamed, the year is unstated, and the only source is the man it flatters. The method is coherent and instructive regardless — which is worth being explicit about: this page's value does not depend on the anecdote being true, and its evidential status should not be inflated by its usefulness.
Within Law 33, this marginal executes the chapter's own method better than the chapter's cases do, and by a different route.
Richelieu, Duveen and Lustig all work on the target's weakness. Weil works on a third party's delight and lets the target's own trust do the transmission — which is cheaper, faster, and leaves no relationship between operator and decision-maker to be examined afterwards.
The banker was never manipulated. He was correctly reporting his wife's genuine enthusiasm, and that is a much harder thing to defend against than flattery aimed at you.
Against Law 32, the placement of Weil is the more interesting finding. Greene quotes him as the Authority on playing to fantasies — "I found it far more interesting and profitable to romance than to tell the truth" — and here supplies the worked example of that Authority's own practice. The two laws are running one source across two roles, sage and case, which no other figure in the build occupies.
Marketing — provenance as the product. Signal Mimicry: The Arms Race Between Scam and Legitimate Branding tracks how markers of quality get counterfeited faster than they can be established, and how the counterfeit and the genuine article converge on identical surfaces.
Weil's bottles are the pure case: the contents are authentic and the label is the lie.
The insight neither produces alone: the mimicry literature generally assumes a counterfeit good wearing a genuine signal. Here it is inverted — a genuine good wearing a counterfeit origin story — and the inversion matters because it is undetectable by any test applied to the product. The wife could have had the perfume analysed and learned nothing. Provenance is the one attribute that cannot be verified from the object, which is precisely why so much value is attached to it and why it is the cheapest thing to fake. That generalises well past perfume: credentials, sourcing, authorship, and where I got this claims of every kind are all in the same category.
Behavioural economics — the channel with no scepticism on it. Celebrity Endorsement and Second-Order Logic examines why a recommendation's force depends less on the recommender's expertise than on the audience's model of the recommender's incentives — an endorsement is discounted in proportion to how much the endorser gains.
Weil's wife had no stake in the mining stock and did not know it existed.
What the pairing produces is the reason this operation is so cheap. Every claim Weil made to the banker directly was discounted by the banker's model of Weil's interest — correctly, which is why weeks of golf produced nothing. The wife's enthusiasm passed through a channel the banker had no reason to discount at all, because she had no interest, no knowledge of the transaction, and decades of established credibility. So the technique's real object is not a person but an unmonitored channel into a person — and the general defence follows: when a view you were resisting suddenly arrives from a trusted source, ask whether that source was reached deliberately. The people whose judgement you defer to are the least-defended surface you have, precisely because you established that deference for good reasons long ago.
Sharpest implication. He stopped working on the decision-maker and worked on the person whose judgement the decision-maker trusted more than his own — and she never knew a transaction existed. Nothing about the investment changed; what changed was that an endorsement arrived through a channel with no scepticism installed on it, because the banker's model of his wife's incentives was correct and she had none. Cost: ten ounces of drugstore perfume and two empty bottles. The least-defended surface you have is the set of people whose judgement you defer to.
Generative questions.