Florence hires the best builder alive to repair the dome of Santa Maria del Fiore. Then, for reasons that have nothing to do with the work, it hires a second man to share the job.
The first man says nothing. He does not complain, does not threaten to walk, does not explain to anyone why this is a bad idea.
At the critical moment of construction, he gets sick.
Not sick enough to die. Sick enough to stop. And he makes a point, on his way out, of reminding the city officials that they had hired a second architect who ought to be perfectly capable of continuing.1
Greene sets it up precisely. The commission was "important and prestigious." Then the city hired Lorenzo Ghiberti to work alongside him.1
Brunelleschi's private assessment, which he kept private: Ghiberti "had gotten the job through his connections, and that he would do none of the work and get half the credit."1
Note that this is Law 7's problem arriving from the other end. Brunelleschi is about to be the person whose labour someone else's name gets attached to, and he can see it coming with total clarity.
Greene's phrase for his response is "the great artist brooded in secret."1
The brooding is the interesting part, because of everything it isn't.
He did not explain to the officials that Ghiberti was unqualified — the natural move, which fails for the reason Law 8 keeps giving: it makes the complainant the problem, and asks a committee to admit a bad appointment.
He did not resign, which surrenders the commission. He did not sabotage the work, which is detectable and would have been blamed on him.
He did nothing visible, and waited for the moment when his absence would cost most.
"At a critical moment of the construction, then, Brunelleschi suddenly developed a mysterious illness."1
Two words are doing all the work: critical and mysterious.
Critical is the timing. Any competent builder can stop working. Stopping at the point where stopping is catastrophic is a different act, and it requires knowing the project well enough to identify the moment.
Mysterious is the deniability. An illness cannot be argued with, cannot be proven false, and does not require him to have made a demand. Nobody had to lose an argument, because no argument occurred.
Here is the move that makes this worth a page rather than a story about a sulk.
Brunelleschi did not merely withdraw. He "pointed out to city officials that they had hired Ghiberti, who should have been able to continue the work on his own."1
He publicly backed his rival.
And then, when it became clear Ghiberti was useless and the officials came begging, Greene says: "He ignored them, insisting that Ghiberti should finish the project."1
He kept backing him. Twice. While the officials pleaded.
That is the whole engine. Brunelleschi never made the case against Ghiberti — he arranged for the city to make it, out loud, to itself, at its own expense, and then he declined to accept their conclusion until they had fully committed to it. By the time they fired Ghiberti, the decision was entirely theirs and there was nothing for anyone to resent him for.
"Finally they realized the problem: They fired Ghiberti."1
"By some miracle, Brunelleschi recovered within days."1
Greene delivers it dry, and the dryness is right, because the transparency is not a flaw in the plan. Everyone involved could see what had happened. Nobody could do anything about it, and more importantly nobody needed to — the officials had got what they wanted, which was a finished dome by the only man who could build one.
Then Greene's own summary, and it is unusually generous: "He did not have to throw a tantrum or make a fool of himself; he simply practiced the art of 'making others come to you.'"1
This case sits directly on top of a contradiction in the chapter and is the best evidence for resolving it.
Eight lines earlier Greene insists the target "does not feel the strings that pull him."2 Then, immediately before Brunelleschi, he says the opposite: "it is sometimes better to let them know you are forcing their hand. You give up deception for overt manipulation… The person who makes others come to him appears powerful, and demands respect."3
Brunelleschi is the overt case, and it worked. Napoleon is the concealed case, and that worked too.
Greene supplies no rule. The two cases supply one: concealment is for targets who can walk away, and overt hand-forcing is for targets who cannot. Florence had one dome and one man capable of finishing it. Seeing the strings changed nothing, because there was no alternative at the end of them.
Then the closing generalisation, which is the law's real payoff: "If on one occasion you make it a point of dignity that others must come to you and you succeed, they will continue to do so even after you stop trying."1
The manoeuvre is run once. What it buys is a standing.
You've been paired with someone who will contribute nothing and share the credit.
First, work out whether you're actually irreplaceable. This is the gate, and getting it wrong is fatal rather than merely unsuccessful. If the work can be reassigned, withdrawal is resignation.
Second, find the moment where your absence costs most. Not the moment you're most annoyed. Brunelleschi's timing was a technical judgement about the build, made by someone who understood it better than his employers.
Third, back your rival publicly and mean it. Every word against Ghiberti would have made Brunelleschi the difficulty; every word for him accelerated the demonstration.
Fourth, decline the first request. Accepting early would have left Ghiberti's competence unsettled and the pairing intact.
Fifth, don't gloat and don't explain. He recovered and went back to work.
And the honest part: this cost Ghiberti a major commission and his public reputation. Greene records it as elegance. It was also a man deliberately set up to fail in front of a city.
Strongest support. The move is fully specified — the pairing, the private assessment, the timing, the two refusals, the recovery — and the mechanism is legible without needing any claim about anyone's psychology.
Tension — it isn't the law it's illustrating. Nothing is baited. Brunelleschi offers no lure and creates no attraction. He removes himself, which is closer to Law 16's use of absence than to Law 8's honeyed trap. Greene notices the difference enough to introduce it with "it is sometimes better to let them know you are forcing their hand," and then files it under baiting anyway.
Tension — the technique requires exactly the position the reader probably lacks. This is Law 7's problem again: the chapter's best cases belong to people who were already indispensable.
🚩 SINGLE SOURCE · 🚩 SECONDARY WITHOUT PRIMARY — no date, no source, no citation to Vasari or any other authority. Brunelleschi and Ghiberti's rivalry over the Florence cathedral is historically real and well documented; the mysterious-illness account as Greene tells it, with the insistence and the days-later recovery, is given without support. [POPULAR SOURCE]
Open questions.
Greene builds Rubens and the Production Line at Law 7 out of an artist managing a single afternoon of visibility — shooing his specialist painters out so a client could watch him work alone at superhuman speed.
Both men manipulate what an employer observes. But they're solving inverse problems, and the pair maps the whole territory.
Rubens had more output than his reputation could account for, and staged a scene to make the output look like his.
Brunelleschi had less credit than his output warranted, and staged an absence to make the shortfall visible.
Rubens adds a performance. Brunelleschi subtracts one. And the deciding factor is which direction the discrepancy runs: if the work exceeds the credit, remove yourself and let the gap show; if the credit exceeds the work, appear and fill it.
Which produces something neither law states: an artist's reputation problem is always one of these two, and they require opposite behaviour. Greene has both cases, ninety lines apart, and treats them as illustrations of unrelated principles.
→ behavioral-mechanics · Aggression as Loss of Control
The mechanism page carries Law 8's diagnosis and its central prescription: stay back, keep calm, don't mistake aggressive action for effective action.
Brunelleschi is the cleanest small-scale demonstration in the book, and he sharpens the prescription into something checkable.
Greene's version of stay back is mostly a temperament — patience, composure, not being provoked. Brunelleschi's version is a technical calculation: identify the point in a process where your absence is most expensive, and be absent then. That is not restraint. It is a precisely timed act that happens to look like doing nothing.
The insight neither page produces alone: what reads as patience in this book is frequently scheduling. Talleyrand waited because he was waiting for a congress; Togo waited because the Russians had six thousand miles to sail; Brunelleschi waited for a specific week in a construction sequence. In none of these is waiting a disposition, and a reader who takes the lesson as be calmer has taken the least useful half.
→ history · Sen no Rikyū and the Wooden Statue
At Law 1, the tea master places a wooden statue of himself, wearing sandals, inside the palace temple — and is ordered to take his own life for it.
Both men assert standing in front of a superior. One is thanked; one is destroyed. The variable is not tact, and it is not the size of the assertion.
Rikyū's art was prestigious and substitutable. Hideyoshi valued tea ceremony enormously and could obtain it from other masters. His refinement was real and it was not a bottleneck.
Brunelleschi's was a bottleneck. Florence had one unfinished dome and one man who could close it.
The insight the pair produces: the survivability of forcing a superior's hand has nothing to do with how the gesture is received in the moment, and everything to do with whether the superior can obtain the thing elsewhere. Rikyū's statue and Brunelleschi's illness are the same assertion, and the outcomes diverge because of a fact about supply that neither man's conduct could change.
Which is a hard finding for anyone in the arts, because it means the safety of asserting yourself is largely decided before you assert anything — and prestige is a poor proxy for it. Rikyū was the most celebrated practitioner of his art in Japan, and celebrated is not the same as unreplaceable.
Sharpest implication. Brunelleschi never once argued that Ghiberti was incompetent. He publicly maintained the opposite, twice, while a city discovered it independently and at its own expense. The most durable version of winning an argument is arranging for the other side to reach your conclusion and believe they got there alone — and the price is that you must be willing to let a project visibly fail in the meantime, which is the part almost nobody can stomach.
Generative questions.