A book isn't selling. The instinct is to blame the writing, or the concept. A company is underperforming. The instinct is to blame the product. Both instincts point at a visible thing — a "what" — because a single named cause is easier to hold in mind than a web of relationships. It's also usually wrong, or at least incomplete, because most real problems aren't single components failing. They're structures failing.1
Greene's reframe: stop asking what's broken and ask how the parts relate to the whole. A poorly organized book usually traces back to poorly organized thinking, which the prose merely reflects rather than causes. An underperforming company usually traces back to how information moves between people, not to the product itself — no amount of product tweaking fixes a communication structure that was never working.2 The shift is from nouns (a broken thing) to verbs (a broken relationship).
Greene names the underlying habit directly: minds naturally separate things out, thinking in terms of nouns rather than verbs.3 A "what" answer is always a noun — the writing, the product, the person. A "how" answer is a verb, or at least a relationship: how the parts move, connect, and depend on each other. The bias toward nouns isn't a personal failing. It's the same shorthand instinct Alter Your Perspective names as the Conventional Mind's default mode, applied specifically to diagnosis.
Worth stating plainly, since Greene doesn't: not every problem is structural. Sometimes a single broken component really is the cause, and hunting for a relational explanation where none exists just wastes time. The strategy isn't "structure is always the answer" — it's "check structure before assuming a single visible cause is the whole answer," because the visible cause gets checked automatically and the structural one usually doesn't.
Greene's own example: the relationship between electricity and magnetism, and the relativity of their effects on each other, is what field theory is actually about — a revolution in physics that ran from Michael Faraday's original insight through to Einstein's later elaborations. Neither man was staring at electricity as a fixed "what." Both were staring at how electricity and magnetism behaved in relation to each other.4
Greene's own paired example, alongside the underperforming book: a struggling company. The default move is scrutinizing the product. His redirect: look instead at how people in the organization actually communicate — how quickly and fluidly information passes between them. If people aren't on the same page internally, no amount of product-level change fixes what's actually broken.5 The book and the company are the same diagnostic move applied to two different kinds of "thing" — a manuscript's structure and an organization's communication structure are both relationships mistaken, by default, for objects.
Greene makes a claim worth flagging on its own: the electricity-magnetism relational insight was a revolution in physics, but the same reframe — relationship over component — is, in his words, "a revolution that is waiting to happen on a more mundane level, in our everyday thinking."6 The strategy isn't scaled for laboratories. It's pitched, deliberately, at ordinary diagnostic situations most people never think to apply it to.
Something in your work has stopped functioning well, and you've already named the broken component. Pause before fixing it. Ask instead how it connects to everything around it — the relationship might be the actual site of the failure, with the component just where the symptom happens to show up.
You're diagnosing a team or organizational problem. Resist naming a single culprit — a bad hire, a bad process. Look at how information and decisions actually move between people first.
The Faraday-to-Einstein field-theory lineage is accurate history, though compressed — Greene skips the intervening physicists (Maxwell chief among them) who did most of the actual mathematical bridging between Faraday's intuitive picture and Einstein's formal theory.
Open question: Greene gives no method for identifying which relationships matter when a system has many. "Look at structure" is a reorientation, not yet a diagnostic — the page doesn't specify how to narrow a genuinely complex web of relationships down to the ones actually worth examining first.
This is the narrowest and most purely diagnostic of the four shorthands, and it converges cleanly with its parent page, Alter Your Perspective — it's the "cause versus structure" axis of that page's four-part taxonomy, without independent tension elsewhere in the corpus worth preserving here.
Mechanical Intelligence — the Wright brothers' insight was, at bottom, a "how" answer to a "what" problem: instead of asking which parts of a flying machine to improve, they asked how pilot and machine should relate to each other, choosing the bicycle's active-correction relationship over the ship's passive-stability one. The insight the pairing produces: the what/how distinction isn't confined to diagnosing failure — it applies just as much to original design, where the choice of relationship to build around precedes and determines every component choice that follows.
The Social Force — this Laws of Human Nature page treats the group as a physiological field that exists between and around individuals, not reducible to any one person in it. That's a structural, relational claim about groups exactly parallel to Greene's claim about companies here: the group's actual behavior lives in the relationships between members, not in any named individual — which means diagnosing a group problem by pointing at one person is the social equivalent of blaming the writing instead of the structure.
Sharpest implication. Every organization has a standing incentive to name a single scapegoat component, because relationships are harder to fix and harder to blame than things. That incentive, not analytical difficulty, may be the real reason structural diagnosis stays rare even when it's more accurate.
Generative questions.