The Grammar — The Norm, Plate v

The volume has shown that unwritten rules exist and that they matter — conventions that coordinate, trust that enables, reciprocity that binds — but it has not yet said precisely what a norm is, what makes a regularity of behavior carry the force of an ought, or why a rule that has held for generations can crumble in a season. Cristina Bicchieri supplied the missing anatomy. A social norm, she showed, is not a thing inscribed in people or in institutions but a structure of mutual expectation: we keep a norm on two conditions — that we believe others keep it, and that we believe others expect us to keep it. This conditional structure is the grammar of the norm, and from it follow both the norm's great power over a whole community and its surprising fragility.

At the heart of Bicchieri's account is the claim that conformity to a social norm is conditional, and conditional on two distinct kinds of expectation. The first she called empirical expectations: I conform because I believe that most of the people who matter to me — my reference group — also conform. The second she called normative expectations: I conform because I believe that those same people believe I ought to conform, and may disapprove or punish if I do not. A social norm proper requires both together. I tip the waiter, queue for the bus, deal honestly with a stranger, return a favor — not from unconditional principle, but because I expect others to do likewise and expect them to expect it of me. Take away the first belief, and let me come to think that no one else conforms, and my reason to conform weakens; take away the second, and let me think that no one really expects it or would mind, and it weakens again. The norm hangs on the pair of beliefs held together, and snaps if either is cut.¹

1. Bicchieri's core claim: conformity to a social norm is conditional on two expectations. Empirical — I believe most of my reference group conforms. Normative — I believe they believe I ought to conform, and may sanction if I do not. A social norm proper needs both. I tip, queue, deal honestly, return favors not from unconditional principle but because I expect others to and expect them to expect it. Cut either belief and the reason to conform weakens. The norm hangs on the pair held together.

This conditional structure lets Bicchieri sort the unwritten rules into kinds that look alike but behave very differently. A pure convention, in the sense of the volume's first plate, rests on empirical expectations alone: I drive on the right because the others do, and there is no sense that I ought to, no one to take offense if a deserted road tempts me to wander, because the matter is one of pure coordination. A merely descriptive regularity — what people simply happen to do — likewise carries no ought. A moral rule, at the other extreme, is unconditional: I refrain from cruelty regardless of what others do or expect, so my conformity hangs on no expectation at all, and such a rule is, in her terms, not a social norm but a personal one. The social norm proper occupies the ground between: it requires both expectations, the empirical and the normative, and it governs exactly those matters — fairness, honesty, reciprocity, the done thing — where we conform because the others do and because the others expect it of us. The grammar tells us, in advance, which rules will bend with circumstance and which will hold against it.²

2. The structure sorts look-alike rules. A pure convention (first plate) rests on empirical expectation alone — drive on the right because others do, no ought, no offense on an empty road. A descriptive regularity carries no ought either. A moral rule is unconditional — refrain from cruelty whatever others do — so it is a personal, not social, norm. The social norm proper needs both expectations and governs fairness, honesty, reciprocity, the done thing. The grammar predicts which rules bend and which hold.

⁂

Because a norm consists in expectations, it can be brought into the open and studied, where before it could only be gestured at. Elicit what people believe others do, and what people believe others approve, and one can diagnose whether a norm genuinely exists, how widely it is held, and how firmly — turning the vague sociological notion of the done thing into something measurable. And because a norm depends on its expectations being present to mind, it can be switched on and off. A norm is not always active: the same person may behave one way when a situation makes the relevant norm salient and quite another way when nothing calls it to attention. A cue that focuses attention on what others do and expect — a posted sign, a watching eye, a quiet reminder of the rule — can activate a dormant norm and summon conformity; its absence can leave the norm sleeping while behavior drifts. This is why the same crowd can be orderly in one setting and heedless in another, and why those who would change behavior have learned to work less on people's private convictions than on the expectations that switch the norms on.³

3. Made of expectations, a norm can be studied. Elicit what people think others do and others approve, and one can diagnose whether a norm exists and how firmly — the done thing made measurable. And since a norm needs its expectations present to mind, it can be switched on and off: the same person conforms when a cue makes the norm salient (a sign, a watching eye, a reminder) and drifts when nothing calls it to attention. Hence the same crowd is orderly here, heedless there — and reformers work on expectations, not private conviction.

The conditional structure also explains the most dramatic feature of norms: their capacity to hold for generations and then give way almost at once. Since conformity hangs on the belief that others conform and expect conformity, a shift in that belief can cascade. Let enough people come to think that others have stopped keeping the norm, or have stopped caring whether it is kept, and each one's reason to keep it weakens, which confirms the next one's impression, and the norm can unravel with startling speed; the same logic, run forward, lets a new norm sweep in to replace it. Most treacherous of all is the case Bicchieri and others have called pluralistic ignorance, in which a norm persists though most people privately reject it, because each wrongly believes that the others support it. Such a norm is a structure of mutual misperception, and it can stand for a long time and then collapse in an instant, the moment the truth becomes common knowledge — as an unpopular regime, obeyed only because each subject imagines the others to be loyal, can fall the very day they discover that they are not alone. The volume returns to this sudden giving-way in a later plate; here it is enough to see that a rule made of beliefs is only ever as solid as the beliefs of which it is composed.⁴

4. The structure explains why norms hold for generations then give way at once. Conformity hanging on belief, a shift cascades: let enough think others have stopped keeping the norm, or stopped caring, and each one's reason weakens, confirming the next's, and the norm unravels fast — or a new one sweeps in. Worst is pluralistic ignorance: a norm privately rejected by most survives because each wrongly thinks the others back it. Such mutual misperception can stand for years and collapse the instant the truth is shared — as an unpopular regime falls the day its subjects learn they are not alone.

With Bicchieri's grammar the unwritten rule becomes tractable — measurable, switchable, predictable in its rise and fall — and the practical consequences are large, for if a harmful practice is sustained not by anyone's private conviction but by a web of shared expectation, it can be changed by changing the expectation, as reformers have found in working to end practices that whole communities upheld yet few among them privately defended. But the deepest lesson is about what a norm fundamentally is. It is not a law on a page, nor a fact lodged in any single mind, but a shared belief about a shared belief — a thing held, precisely, in common. That is what gives the unwritten rule its strange double character: mighty enough to govern a whole people, frail enough to dissolve the moment the beliefs beneath it shift. It remains now to see the unwritten rules doing their most impressive work — holding together, with no court and no state to compel it, the long-distance commerce on which prosperity has always depended. The next plate, the flagship of the volume, turns to it. The next plate is the reputation.⁵

5. With the grammar the norm becomes tractable — measurable, switchable, predictable — with large consequences: a harmful practice sustained by shared expectation, not private conviction, can be ended by shifting the expectation, as reformers have found. But the deepest lesson is what a norm is: not a law on a page nor a fact in one mind but a shared belief about a shared belief, held in common — hence mighty over a people yet frail when the beliefs shift. Next, the flagship: unwritten rules holding long-distance trade together with no court at all. Next: the reputation.

Cristina Bicchieri gave the unwritten rule its grammar. A social norm, she showed, is sustained by conditional preferences resting on two expectations at once — the empirical belief that others conform, and the normative belief that others expect us to conform and may sanction us if we do not — and it holds only while both hold. This structure distinguishes the social norm from the pure convention (empirical expectation alone), the descriptive regularity (no ought), and the moral rule (unconditional); and it explains why norms can be measured by eliciting expectations, switched on and off by cues that make them salient, and tipped into sudden collapse or cascade when the underlying beliefs shift — most strikingly in pluralistic ignorance, where a norm privately rejected by all survives on each one's mistaken belief in the others' support. The unwritten rule, at bottom, is a shared belief about a shared belief, which is the source at once of its power over a whole community and of its fragility. The next plate is the reputation.