In 1545 Cosimo I de' Medici commissioned frescoes for the main chapel of San Lorenzo in Florence, and chose Jacopo da Pontormo — an ageing painter who intended this to be his masterpiece and his legacy.1
His first decision was not about the painting. It was to close the chapel off with walls, partitions and blinds.
"He wanted no one to witness the creation of his masterpiece, or to steal his ideas. He would outdo Michelangelo himself. When some young men broke into the chapel out of curiosity, Jacopo sealed it off even further."2
Note the direction. The breach did not prompt a reconsideration. It prompted more sealing — which is the same escalation pattern as Shih Huang Ti's passageways, running on one man and a room instead of an emperor and an empire.
He worked on it for eleven years, "rarely leaving it, since he had developed a phobia for human contact and was afraid his ideas would be stolen."3
He died before finishing. None of the frescoes survives.
The only description we have is from Vasari, a friend of Pontormo's, who saw the work shortly after his death — and it is the reason this case is in the chapter rather than being a sad anecdote.
The ceiling was filled with biblical scenes: the Creation, Adam and Eve, Noah's ark, on and on. At the top of the middle wall, Christ in his majesty raising the dead on Judgment Day.4
And then the assessment: "There was a total lack of proportion. Scenes bumped against scenes, figures in one story being juxtaposed with those in another, in maddening numbers. Pontormo had become obsessed with detail but had lost any sense of the overall composition."5
Vasari's closing line is the one that gives the case its force. He broke off his description by writing that if he continued, "I think I would go mad and become entangled in this painting, just as I believe that in the eleven years of time Jacopo spent on it, he entangled himself and anyone else who saw it."6
Greene's reading is the sharpest sentence in Law 18: the frescoes were "visual equivalents of the effects of isolation on the human mind: a loss of proportion, an obsession with detail combined with an inability to see the larger picture, a kind of extravagant ugliness that no longer communicates."7
That claim is stronger than the political version of the same law, and worth separating out.
When a ruler loses proportion in a fortress, the loss is inferred — we see the bad decisions and reason backwards to the state of mind. Nothing is directly observable.
Pontormo's frescoes are the state of mind made visible. Loss of proportion is not a metaphor here: it is a formal property of the picture, measurable in the relationship between figures and the whole. The same word describes the psychological condition and the compositional fault, and in this case they are the same thing.
This is the only case in Law 18 where the interior consequence of isolation is directly inspectable, and it is doing evidentiary work the political cases cannot.
Work out why proportion specifically, rather than skill or invention.
Proportion is a relational judgement. It is not a property of any part — it is the answer to how much of this, relative to that, relative to the whole. And a relational judgement cannot be made from inside a single element; it requires stepping back, which in painting means physical distance, and in judgement means comparison.
Isolation removes exactly the operations that produce proportion. No visitors means nobody entering with a fresh eye, seeing the whole before seeing the parts. Eleven years in one room means never returning to the work after an absence, which is the only way an artist gets to see their own picture as a stranger would.
The other faculty needs nothing isolation takes away. Detail can be refined indefinitely, from close up, alone, and each individual improvement is genuinely an improvement.
So the two do not decay together — one is starved and the other is fed, and the result is precisely what Vasari describes: obsessive local quality with no global structure. The painter is not getting worse at painting. He is getting better at parts and blind to arrangement, and from inside the room the first is visible and the second is not.
Worth being fair to Pontormo, because Greene is not and the chapter's argument does not require it.
His stated fear was that his ideas would be stolen. In sixteenth-century Florence that is not paranoia — it is an accurate reading of a competitive workshop culture in which motifs, compositions and inventions circulated rapidly and attribution was contested.
And the second motive, "he would outdo Michelangelo himself," is the ordinary ambition of a serious painter with a major commission.
Neither motive is irrational. What was fatal was the duration. A period of closed working while an idea is fragile is a normal artistic practice. Eleven years without the corrective is a different thing, and there was no moment at which the transition announced itself — which is the case's real content and the reason it generalises.
Greene reads Pontormo as an example of a mistake. He is better read as an example of a reasonable practice with no natural end, and no signal that it has become the problem.
Greene's counter-example is Shakespeare: "the most famous writer in history because, as a dramatist for the popular stage, he opened himself up to the masses, making his work accessible to people no matter what their education and taste. Artists who hole themselves up in their fortress lose a sense of proportion, their work communicating only to their small circle. Such art remains cornered and powerless."8
This is the weakest part of the passage and should be flagged rather than repeated.
It reads accessibility as a consequence of exposure, when for a working dramatist on the Elizabethan stage it was a commercial constraint. And it uses fame as evidence for the claim, which makes the argument unfalsifiable — any circulating artist who is famous confirms it, and any isolated artist who is famous is not mentioned.
The Pontormo case does not need it. Vasari's testimony is about composition, not about audience size, and the compositional claim stands on its own. Greene's move from isolation damages proportion to therefore make work accessible to the masses is a separate argument with separate evidence, and he supplies none.
You are deep in something and you have stopped showing it to anyone.
The reasons are real. It is not ready. Showing it early invites the wrong feedback and you will end up defending a decision you have not finished making. And there is a version of this where somebody else runs with it first.
All of that is true, and Pontormo's version of it was true too. The question is not whether the reasons are good. It is how long they have been operating.
Here is the check that actually discriminates, and it is not about feedback. When did you last see the whole thing at once? Not the section you are working on — the whole. If you cannot remember, you are in the failure mode, and the symptom is not that the work is bad. The parts are probably excellent. The symptom is that you have been improving details for weeks and could not say what the overall shape is doing.
The corrective is distance rather than opinion, which is why get feedback is the wrong prescription. Leave it alone for long enough to come back to it as a stranger — a week, not an afternoon — or show it to one person and watch what they look at first, since a fresh eye goes to the composition and yours goes to the part you were fixing.
And notice the escalation, because that is the tell that it has stopped being a choice. Pontormo did not seal the chapel once. He sealed it, someone broke in, and he sealed it further. If your response to an interruption is to close the door harder rather than to consider whether the door should be open, the closure has stopped being a working method and become something you are protecting.
The evidentiary strength here is unusual for this book: a named contemporary witness who knew the artist personally, describing a specific work, with a specific formal criticism. Vasari is a partisan and a fabulist in places, but he is a real, checkable source rather than an uncited anecdote.
Tension: the case is over-determined and Greene picks one cause. Pontormo was, on Greene's own account, developing a phobia of human contact. Whether the isolation caused the compositional collapse or a deteriorating mental state caused both is not separable from this evidence, and the chapter assumes the first.
Tension: the Shakespeare argument does not follow from the Pontormo evidence. Vasari's claim is about composition; Greene's conclusion is about accessibility to a mass audience. Two different arguments, one body of evidence.
Tension: the work does not survive. Everything rests on one description by one friendly witness who was also an interested party in the historiography of Florentine art. There is no possibility of a second opinion.
🚩 SINGLE SOURCE · 🚩 SECONDARY WITHOUT PRIMARY — Vasari is named but no work, edition or translator is given, and Greene's paraphrase of the Lives is not distinguished from quotation. The eleven-year figure and the phobia are uncited. [POPULAR SOURCE] · [PLAUSIBLE — needs corroboration]
Open questions. Was there a moment at which the sealed chapel stopped working and started harming — and would Pontormo have been able to detect it, given that the thing he lost was the faculty that would have noticed? And: Vasari says he would "go mad and become entangled" if he continued describing it. Is the account partly a literary construction, and how much of the frescoes' reputation for madness comes from the work rather than from that sentence?
Within Law 18 this case does something none of the political cases can. Shih Huang Ti's loss of proportion has to be inferred from outcomes; Pontormo's is on the wall. Greene's phrase "visual equivalents of the effects of isolation" is doing real work — this is the chapter's only piece of direct evidence for its central psychological claim.
The escalation parallel is exact and unremarked: the emperor's passageways answered being seen, the different bedroom answered being found, the executions answered the leak. Pontormo sealed the chapel, was broken into, and sealed it further. Same shape, different scale, and in both cases the response to a breach is more of the thing that caused the vulnerability.
Against the Reversal, this case is the strongest argument for the Reversal's warning and the chapter does not use it that way. The Reversal concedes that isolation is the one condition that permits thought, and then warns that "the more isolated you are, the harder it is to break out of your isolation when you choose to—it sinks you deep into its quicksand without your noticing." Pontormo is the quicksand clause with a date on it, and Greene places the case four paragraphs earlier without connecting them.
Psychology — The Paradox of Mental Control
That page's finding is the white-bear problem: suppressing a thought requires monitoring for it, and the monitoring sustains the thing being suppressed.
Pontormo's fear of theft has that structure, and it explains the escalation the chapter reports without accounting for.
Protecting an idea from being stolen requires continuously assessing exposure — who could see, what could be inferred, whether the partitions are sufficient. That monitoring is itself a form of sustained attention to the possibility of theft, which keeps the fear supplied with material. Each new precaution generates a new set of gaps to worry about.
What the pair produces: the sealing could not reach a stable state, by construction. There is no configuration at which the monitoring returns safe, because the monitoring is what defines the threat's dimensions. Which is why the broken-in-upon chapel got more sealed rather than prompting a reassessment — the breach was not new information, it was confirmation the monitoring had been correct all along.
And it locates the intervention. You cannot resolve this by making the room more secure, since security is judged by the process that produces the fear. It resolves only by changing what is at stake — publishing early, accepting the theft risk — which is a decision about exposure rather than about protection.
Business — Work for Free to Manufacture Proof
That page's logic is about putting work into the world before it is commercially ready, in order to generate evidence — the output circulates, produces reactions, and the reactions become the asset.
Pontormo is the maximal counter-case, and the pair sharpens what circulation is actually for.
The obvious reading is that early exposure buys reputation. Pontormo's case says it also buys something the artist needs more and cannot get any other way: calibration. Eleven years of unpublished work produced no reputation and, more damagingly, no correction — and the correction is what proportion is made of.
What the pair produces: releasing work early has two entirely separate payoffs, and only one of them is about other people. The external payoff is proof, audience, opportunity. The internal payoff is that you get to see your own work as an object rather than as a process, because the act of releasing it forces you to view it whole, from outside, in the state a stranger would encounter.
That second payoff is available even if nobody looks. Which means the strongest argument for shipping is not about the market at all, and it survives every legitimate objection about the work not being ready — because the thing being corrected is not the work's quality but the maker's ability to see it.
Sharpest implication. Proportion is a relational judgement and requires distance; detail is a local one and does not. So isolation does not degrade a practice evenly — it starves one faculty and feeds the other, and from inside the room the fed one is visible and the starved one is not. The work gets better in every part and worse as a whole, and the maker experiences only the first.
Generative questions.