He has reasons to want one. Europe is in a war the United States will almost certainly enter, and his domestic programmes are unfinished.1
And a hard constraint:
If he revealed his desire for a third term, though, he would stir up opposition even within his own party. Many had already accused him of dictatorial tendencies.1
That is the shape of the problem, and it is more common than presidential: the thing you want is available, and wanting it visibly is what makes it unavailable. Any move toward it becomes evidence for the charge already circulating against you.
So Roosevelt decided to get what he wanted through a form of passive aggression.1
One: deny it, repeatedly.
In the months leading up to the Democratic Convention, he repeatedly stated his lack of interest in a third term.1
Two: actively promote rivals.
He also actively encouraged others in the party to seek the nomination to replace him.1
That is the move that separates this from ordinary coyness. He was not passive about the field — he worked on it.
Three: never quite close the door.
At the same time, he carefully crafted his language so that he never completely closed the door on running himself, and he pushed enough candidates into the nomination race that no single one of them could come to the convention as the favorite.1
Two operations in one sentence, and together they are the whole machine. The language leaves a gap; the crowded field guarantees that when the convention arrives, the gap is the only thing in it that looks like a solution.
Four: be absent.
Then, as the convention opened, Roosevelt withdrew from the scene, making his large presence known by his absence: without him the proceedings were incomparably dull.1
Reports came back to him that people on the floor were beginning to clamor for him to appear.1
Letting that desire reach its peak, the president then had his friend Senator Alben Barkley insert into his own convention speech a message from Roosevelt:1
"The president has never had, and has not today, any desire or purpose to continue in the office of president, to be a candidate for that office, or to be nominated by the convention for that office."1
Read the sentence on its own and it is an unambiguous refusal. Read it in a hall that has spent two dull days wanting him and cannot settle on anyone else, and it is a release — it removes the last obstacle, which was the appearance that he was seeking it.
After a moment of silence, the convention floor began to ring with the delegates' cry: "WE WANT ROOSEVELT!" The appeal went on for an hour.1
The next day the delegates were to vote, and chants of "ROOSEVELT!" again filled the hall. The president's name was entered for the nomination, and he won by a landslide on the first ballot.1
He did not break the third-term convention. The party did, over his stated objection.
Greene draws it out immediately:
it is never wise to seem too eager for power, wealth, or fame. Your ambition may carry you to the top, but you will not be liked and will find your unpopularity a problem.1
Better to disguise your maneuvers for power: you do not want it but have found it forced upon you.1
And the compressed version:
Being passive and making others come to you is a brilliant form of aggression.1
The move looks like modesty and is doing something more specific, which is worth separating out because it is the transferable part.
A stated ambition invites the question should he have it? — and that question has opponents, precedents and a whole apparatus of resistance already attached to it. Roosevelt's third-term ambition would have been debated on its merits, against a norm and against an existing accusation of dictatorial tendencies.
The denial changes the question the room is answering. It becomes what do we do? — which is a problem the delegates now own, in a hall with no favourite, two dull days behind them, and one obvious answer.
And it relocates the burden of breaking the norm. Somebody had to violate the two-term convention; the manoeuvre ensured it was the convention that did it, in public, by acclamation, which is a considerably more defensible position than a president who asked.
The cost of the technique is the one Greene never mentions: it requires a genuine risk of not getting the thing. If the field had produced a favourite, or the floor had settled, the denial would have stood and Roosevelt would have been out. The move is not free — it is a bet that the demand is real.
You want something where being seen to want it is what would prevent you from getting it.
Start by checking whether that is actually your situation, because it is the only one this works in. If wanting the thing openly costs you nothing, ask for it — this technique is slower, riskier, and pointless where a direct request would succeed.
Then say no clearly, and mean it enough that the sentence would survive being quoted back. Roosevelt's message through Barkley is a flat refusal with no visible hedge. A denial that everyone can see is a manoeuvre does the opposite of its job, because it invites the reading that you are managing them.
Then do the work on the alternatives rather than on yourself. This is the part almost everyone skips and it is what made the outcome inevitable: he pushed enough candidates in that none could emerge. The absence of an obvious alternative is what converts a refusal into a problem, and problems get solved by whoever is available.
Then leave the door technically open and never point at it. The careful language is what allows the room to act; announcing that you have left it open closes it.
Then remove yourself and let the vacancy do the work. His absence made the convention dull, and dullness is what made the floor start looking for him.
And accept that you might lose it. The technique is a bet on the demand being real. If it is not, your refusal stands and you have talked yourself out of the thing — which is the honest price, and the reason the move reads as modest even when it is not.
The 1940 Democratic convention, Barkley's reading of Roosevelt's message, the ensuing demonstration on the floor, and the first-ballot nomination are all established. [POPULAR SOURCE]
The "spontaneous" demonstration was substantially organised, and Greene omits it. The floor demonstration following Barkley's message is widely reported to have been amplified by Chicago's Mayor Kelly through the convention hall's public-address system — the "voice from the sewers," delivered by the city's superintendent of sewers from a basement microphone. That is not a detail; it is the difference between reading the room correctly and manufacturing the room's response. Greene's account has the delegates' desire reaching its peak on its own. 🚩 SINGLE SOURCE, and the omission flatters the technique considerably.
And Roosevelt's private intentions in the preceding months are genuinely disputed. Some historians read the whole sequence as a designed draft; others hold that he was ambivalent well into 1940 and that the war's escalation resolved it for him. Greene presents a fully premeditated campaign. The ambivalent version is at least as consistent with the evidence and would make the "strategy" partly a reconstruction — which is this book's most frequent failure mode and worth flagging here specifically because the technique's whole appeal is its apparent deliberateness.
One thing the case does establish cleanly, whatever Roosevelt's inner state: the structural point stands. A crowded field with no favourite, plus a leader who has not ruled himself out, produces a draft, and that is true regardless of whether anyone planned it.
Against Strategy 32: The Passive-Aggression Strategy this is the chapter's gentlest case and its clearest statement of the passive half — nobody is deceived about anything material, and the aggression consists entirely in arranging a situation and then declining to fill it.
Against FDR and Political Jujitsu this is the fifth Roosevelt treatment of the build, and the pairing is the most coherent. That page holds the 1920 landslide loss, polio, and the deliberate silence that goaded opponents into shrillness. Both are the same instrument — an engineered vacancy that other people rush to fill on his terms — and Greene distributes them across Strategies 9 and 32 as unrelated techniques.
Against Roosevelt and the 1936 Ripening the convergence is exact: leaving the field open until Landon was fixed in place, then pivoting late. Three of the five FDR pages in this vault are the same move at different scales — wait, stay unfixed, let the other side commit, then act — which is a stronger finding about the man than any individual chapter makes.
Against The Mediator Position the mechanism is shared and the purpose differs. Metternich held the centre by never committing so that others would court him permanently; Roosevelt held it briefly so that a specific demand would form. Same instrument, one used as a standing position and one as a single manoeuvre.
Psychology → A thing you demanded and a thing you were asked for are held differently. The delegates who chanted for Roosevelt had, in their own account, chosen him against his stated wishes. The vault's Cognitive Dissonance holds why that matters afterwards: a decision you made and had to overcome resistance for is one you then have reason to defend.
The insight the pairing produces: the same outcome is far more durable when the other party had to work for it, which means engineering their effort is not manipulation of the decision but of the commitment that follows it. Roosevelt did not merely avoid the accusation of grasping — he acquired a convention that had actively overruled him and was therefore invested in having been right. This gives a rule with wide application and an uncomfortable edge: the support you are handed is weaker than the support you made someone argue for, and any arrangement where you did all the persuading has left the other party with nothing to defend.
Business → An unfilled vacancy generates its own candidate. The convention became dull, no favourite emerged, and the room went looking. Set that beside Strategy 29: The Fait Accompli Strategy, where a completed act forecloses a discussion that a proposal would have opened.
Together they produce a genuine pair of opposites worth holding: you can foreclose a decision by making it, or by making yourself the only remaining answer to it — and which works depends on whether the decision is yours to take. Roosevelt could not simply announce a third term; he could ensure the question arrived with one available answer. In organisational terms this is why the person who quietly makes themselves the only one who understands a system ends up running it without ever having applied, and why the most effective move in a contested succession is often to remove yourself early and let the shortlist fail — a strategy that is indistinguishable from graciousness and costs nothing if the demand does not materialise, except the thing itself.
Roosevelt did not break the two-term convention — the party broke it, over his flat public refusal, which is a far more defensible position than asking. The technique's real work was done on the field: pushing in enough candidates that none could emerge, so that his refusal arrived in a room with a problem and one obvious answer.