Social & Critical Psychology · Simplified

How Many, and At What

The older account of crowds and its successor do not mainly disagree about how much happens. They disagree about whether it has a shape.

What the numbers here are. They come from a simple teaching model whose rules are printed further down, not from any real crowd. Use them to compare what the two accounts predict. Neither account is refuted here, both have serious critics, and no event, police operation or town on this page is real.

The two accounts

What each one says to look at

Both are serious positions with serious critics. They are set out here because they tell you to measure different things.
Account	Where it puts the cause	What it predicts about the shape of the action	What it therefore measures

Step 1 of 3

First, a judgement

Footage of an afternoon shows that almost everything the crowd did was directed at one thing, and that a shuttered shop, a bus shelter and a line of parked cars a few metres away were left alone. Which account does that fit better?

Step 2 of 3

Six sections of a crowd

A fictional market town, a fictional afternoon, and six groups of people who came for different reasons. Choose how the day went, then decide whether people could be identified.

Key terms

Common category: Whether the people present come to see themselves as one group with one thing happening to them. It is something that happens during an afternoon rather than something a crowd arrives with.
How concentrated the action was: Whether what happened was aimed at one thing or spread across whatever was nearby. Zero is an even spread across the four targets, one is everything on a single target.
The disinhibition index: Identifiability together with how many sections joined in. It is what an account of lost restraint tracks, and it is printed here so that you can watch two afternoons score the same on it.

How the afternoon went

Nothing else about the afternoon changes when you press this.

One bar per target, each labelled with its share. A flat set of bars is action spread across whatever was nearby; one tall bar is action with a shape to it.


Section of the crowd	Readiness before the afternoon	Took part?

Step 3 of 3

Two afternoons, one index

The same number of sections took part and people were equally anonymous, so the index an account of lost restraint tracks is identical. The afternoons are not.
Afternoon	Sections taking part	Disinhibition index	How concentrated the action was	Biggest single target

What this shows

The disagreement is about shape, not amount

Key idea: An account built on lost restraint tracks how many people joined in and whether anyone could be identified, and on that measure the two afternoons in the last table are the same afternoon. They are not. In one, what happened was spread across whatever was to hand; in the other it was almost entirely on one thing and three targets a few metres away were left alone. The second account predicts that difference, because on its view a crowd that has come to see itself as one group with one thing happening to it has a shared understanding of what this is about, and a shared understanding has limits in it as well as permissions. That also explains the direction anonymity moves things. Where there is a shared understanding, being unidentifiable makes the action more concentrated rather than less, which is the opposite of what a story about losing your inhibitions would predict, and it is the point at which the two accounts part company on evidence rather than on emphasis.

The figures come from a teaching model with invented numbers, and the concentration measure is a device written for this page rather than a published statistic. Neither account is refuted here. The older one is not a straw man: it was built to explain real observations, it continues to be defended in modified forms, and the anonymity result that goes against it is one line of evidence among several. The successor account has critics of its own, chiefly that identifying a shared understanding after the fact is easier than predicting it beforehand, and that both accounts can be fitted to footage once the footage exists. And a model in which six sections of a crowd have fixed readiness scores is a considerable simplification of an afternoon in which people talk to each other, change their minds, arrive and leave.

The longer version lets you set each condition independently rather than choosing a scenario, adds further presets, and works through what each account would have to observe in order to be wrong. It is at Crowds: Deindividuation and the Elaborated Social Identity Model in the main collection.