Picture the thing you're trying to build — a project, a negotiation, a piece of work — as a block of marble you have to carve into something specific. It's larger than you. The material resists. That's not a problem to be solved; it's simply the accurate size of the task, and a realistic person begins there.
Now imagine a specific disease that alters how you perceive size and proportion. Under its influence, the same block looks smaller. Softer. The people you need to persuade stop looking resistant and unpredictable — they start looking like they already agree with you, like they need you more than you need them.1
Greene calls this disease grandiosity, and its defining symptom is that you cannot feel it happening. It doesn't announce itself as delusion. It arrives disguised as confidence, and by the time the marble has shrunk in your perception, you've usually already made the decision that will prove it hasn't shrunk at all.
The mechanism starts with something entirely ordinary: a small measure of real success. You did something well, and people noticed. That's accurate information, and reacting to it isn't the disease.
The disease begins with what happens next. The success confirms your existing self-opinion — but the confirmation process doesn't stop at "I did this one thing well." It generalizes, quietly and immediately, into "I am the kind of person who does things well," and from there into something closer to "I have a golden touch."2 Two specific subtractions make this generalization possible. First, luck and timing get dropped from the accounting — the market conditions, the lucky break, the moment the audience happened to want exactly what you were offering. Second, other people's contributions get dropped too — the collaborator, the mentor, the predecessor whose groundwork you built on.3 What's left, once both subtractions are complete, is a story in which the outcome was entirely, purely a function of your own superior qualities.
Once that story takes hold, Greene's central image does real explanatory work. The task ahead doesn't just feel more achievable — it's perceptually altered. The people you need to reach stop looking resistant and start looking predictable, because you already know, with a confidence that feels earned, exactly how they'll respond: they'll love it.4
The emphasis shifts, subtly and almost undetectably, from what the task actually requires of you to what you feel entitled to receive from it. And if it fails despite this, the failure cannot be attributed to a bad read of the situation, because the situation was never accurately read in the first place — it has to be attributed to something external instead. Other people. Malice. Envy. Anything except a miscalibrated sense of your own reach.5
Here is the sentence that makes this a genuinely difficult trap rather than a simple error waiting to be corrected: descending from an inflated self-opinion requires admitting the earlier, elevated one was wrong — and Greene's flat assessment is that the human animal will almost never admit that.6
This is what separates the success delusion from an ordinary mistake. An ordinary mistake gets corrected once new information arrives. This one resists correction specifically because correction requires a admission the ego is structured to avoid, which means the distortion doesn't fade naturally with time or evidence — it tends to deepen, since each subsequent failure is a fresh threat to the story rather than fresh information about the story's accuracy, and threats get deflected rather than absorbed.
Greene supplies specific, checkable markers, which matters because the whole difficulty of this condition is that self-diagnosis is nearly impossible while it's running. Overbearing certainty about the positive outcome of your plans, delivered with a confidence that outpaces the actual evidence available. Excessive touchiness when criticized — a defensiveness disproportionate to the substance of what's being questioned. A creeping disdain for any form of authority, since authority above you implicitly contradicts the self-opinion currently forming.7
None of these, taken singly, is damning — everyone feels defensive occasionally, and healthy skepticism toward authority is not itself a symptom. What Greene is describing is a cluster, appearing together and intensifying together, specifically in the period immediately following a real success. The timing is the diagnostic. The same touchiness before any success occurred would simply be touchiness. Appearing right after confirmation arrives, it's something more specific.
Greene is careful to distinguish this from the narcissism covered earlier in the book, and the distinction is worth holding onto because the two conditions call for different responses.
Narcissism, in Greene's usage, is about everything revolving around you — other people converted into instruments, into an audience, into extensions of your own needs. Grandiosity is a related but distinct thing: not merely self-centeredness, but a specific sense of being enlarged — the word's root literally means "big" or "great" — superior, worthy not just of attention but of something closer to adoration. It's less "everything must serve me" and more "I have become, in some genuine sense, more than merely human."9 A narcissist needs an audience. A grandiose person, at the extreme, has started to believe the applause is simply accurate reporting.
Greene's counter-model, stated at the section's opening, is worth holding onto because it's more demanding than it first sounds. Getting what you want is rarely easy. Success depends on real effort plus some genuine luck. Your strategy will probably need to be abandoned and rebuilt as circumstances shift, because circumstances always shift. And the people you're trying to reach will not respond the way you imagined — they have their own needs, histories, and psychology, entirely independent of your plans for them.8
That's not pessimism. It's simply an accurate starting size for the block of marble. The realistic attitude doesn't require you to think less of yourself — it requires you to keep the actual dimensions of the task in view even after a success has made the smaller, distorted version tempting to believe instead.
You've just had a real success, or you're watching someone close to you go through one, and you want to catch the distortion before it compounds into a decision you can't walk back.
Run the subtraction exercise immediately after any success, before the feeling has time to settle into a fixed narrative. Name the specific luck and timing factors involved, in writing if possible. Name the specific people whose prior work, advice, or collaboration contributed. This isn't false modesty — it's an accurate accounting exercise, and doing it while the memory is fresh is considerably easier than doing it retroactively once the golden-touch story has already calcified.
Watch your own reaction to the next piece of criticism that arrives after a win. If it triggers something disproportionate — real anger, a martyred sense of injustice, an urge to discredit the critic rather than engage the substance — treat that reaction itself as data rather than acting on it. Greene's touchiness marker is most visible in exactly this window.
Before committing to a bigger bet on the strength of recent success, deliberately seek out someone who will argue against it. The realistic attitude requires genuine external friction, and a distorted self-opinion actively avoids exactly that friction because friction is uncomfortable in precisely the way confirmation isn't.
If you're watching this happen in someone else, don't attempt direct confrontation. Greene's own broader advice elsewhere in this material applies here too: telling someone directly that their self-opinion has become inflated rarely produces reflection — it produces a threat response, because you're asking for the exact admission the mechanism is built to prevent. The more effective move is usually distance, or quietly declining to participate in ventures built on the distortion, rather than trying to talk someone down from it.
The mechanism is psychologically plausible and consistent with well-documented findings on overconfidence and self-serving attribution bias in decision science, though Greene cites no specific research and the framing is his own synthesis rather than a direct summary of any cited study. [PLAUSIBLE — needs corroboration]; [POPULAR SOURCE] throughout.
A genuine tension worth naming: Greene's own book elsewhere celebrates confident self-belief as a genuine asset — the expansive attitude, the value of feeling destined for something important, the usefulness of grandiose energy when properly channeled. This chapter treats confidence born of success as close to a disease. The distinction Greene seems to intend — confidence tied to real, demonstrated achievement versus confidence inflated by subtracting luck and others' contributions — is workable, but he never states it as cleanly as the tension actually requires, and a reader moving between chapters could reasonably come away confused about which kind of confidence is being recommended and which is being warned against.
Open question: if the marble-shrinking effect is perceptual rather than merely attitudinal — if the task genuinely looks smaller rather than the person simply talking themselves into overconfidence — what would a reliable external check actually look like? Greene recommends seeking feedback, but the same distorted perception that shrinks the task could plausibly also distort how seriously that feedback gets taken once it arrives, which would make the proposed remedy vulnerable to exactly the disease it's meant to treat.
Eisner and the Disney Arc is this exact mechanism run at full biographical length, and it supplies the piece this more abstract page can't: a demonstration of how long the distortion can persist against overwhelming, repeated, concrete counter-evidence once it's taken hold. Eisner absorbed a genuine business catastrophe (Euro Disney) and continued the pattern for another decade rather than correcting course, which suggests the resistance to "coming back down" isn't a temporary lag before self-correction — it can function as a stable, self-sustaining state that ordinary failure alone doesn't reliably interrupt.
This is the direct parent mechanism, and the pairing clarifies something about why success specifically is so dangerous compared to ordinary daily self-regard. The self-opinion page establishes that people generally hold a self-image somewhat elevated above strict reality, and that this is functional rather than pathological in its baseline form — a small positive bias that supports normal risk-taking and resilience.
The insight this page adds: success functions as a kind of amplifier that the self-opinion mechanism has no natural governor against. Ordinary daily life supplies a steady mix of confirming and disconfirming evidence that keeps the baseline bias roughly stable over time. A genuine, publicly recognized success removes the disconfirming half of that mixture all at once, and the self-opinion — having no internal mechanism to discount a single data point appropriately — treats the win as though it were fully representative of ongoing reality rather than as the specific, non-repeating event it actually was.
The decision-science literature on overconfidence documents the same distortion Greene describes, with better instrumentation: people systematically overestimate the accuracy of their own judgments, and the overestimation reliably increases after a string of correct predictions or successful outcomes, independent of whether the underlying skill has actually improved.
The insight worth drawing from combining the two: Greene's marble metaphor supplies something the decision-science literature usually states more abstractly — a description of what the distortion feels like from inside, rather than only how it measures in aggregate data. The overconfidence literature can tell you the effect exists and roughly how large it tends to be; Greene's framing supplies the subjective texture (the shrinking task, the predictable audience, the shift from effort to entitlement) that makes the effect recognizable in yourself while it's actually happening, which the statistical literature, by its nature, cannot easily provide.
Sharpest implication: the success delusion isn't triggered by failure, and it isn't corrected by ordinary vigilance either — it's triggered specifically by things going right, at precisely the moment self-scrutiny feels least necessary and confirmation feels most deserved. The most dangerous point in any project isn't when it's struggling. It's the first moment it clearly starts working, because that's the exact point where the accurate accounting of luck and contribution becomes both most needed and least psychologically comfortable to perform.
Generative questions: