The pub, seventeen of thirty-one

The two kinds that ask the same and behave differently

• Somebody spoke to a stranger — 26
• Nobody did — 57
• Could not be seen — 6

One mark for each place. Eighty-nine rooms, the same terms at the door, and a conversation rate three and a half times apart.

Thirty-one pubs and bars. In seventeen of them, somebody spoke to somebody they had not arrived with.

Fifty-five per cent, against sixteen per cent in cafés.

The two kinds side by side

On every field but the last they are the same kind of place. Both charge at the counter, both seat almost nobody for nothing, both are misread about time in roughly six cases out of ten. If several people repeat the same survey, managers may explore the platform before rollout while interpreting every record beside completed work and participant feedback.

Then one of them produces conversation three and a half times as often.

What the record can say about why

Very little, and it is worth being honest about that at the start.

The four fields are about terms, not about rooms. Nothing in them distinguishes a pub from a café except the label the observer attached, and the label is the fifth field — the one that was not asked of anybody. Readers comparing this walk with broader public-space evidence can also see Route Fifty.

So what follows is reading rather than measurement.

A round is an occasion. Buying a drink for somebody, waiting at a bar, being served standing in a group: a pub has several moments built into it where people are adjacent and unoccupied. A café has a queue and then a table.

The furniture faces inward. Pub seating tends to be arranged around a room with a bar in it. Café seating tends to be arranged in rows of twos facing a wall or a window.

Nobody is working. In thirty-one pubs the observer recorded a laptop open in two. In fifty-eight cafés, in thirty-four.

That last one is the only quantified difference between the two kinds anywhere in these notes, and it was not a field. It was written down in passing and is reported as such.

What the figure is worth

Seventeen of thirty-one, measured mostly on weekday afternoons.

A pub at three on a Tuesday is the quietest version of a pub there is. If the hour bias runs the way it obviously does, fifty-five per cent is a long way below what a full accounting would find, and further below it than the café figure is below its own true value.

So the gap between the two kinds is probably wider than this record shows.

Of all the figures on this site, this is the one most depressed by walking in the daytime.

The eleven silent pubs

Eleven pubs in which nobody spoke to a stranger, and three more where it could not be seen.

By the observer's notes, nine of the eleven were nearly empty — fewer than four people in the room — which is the ordinary condition of a pub on a weekday afternoon and a reminder of what the walk's timing costs.

Two were busy and silent. Both were places where the seating was in booths.

Where the seventeen were

By the observer's notes rather than by a field: in thirteen of the seventeen, the exchange began at or near the bar.

That is the only locational detail recorded anywhere in this walk, and it was written down because it kept recurring. It is not a count — no field asked where anything happened — and it is printed as a note.

If it is right, the mechanism is not the drink but the queue: a place where strangers stand adjacent and unoccupied for a minute, several times an evening, with a shared reason to be where they are.

A café has a queue too, and nine conversations in fifty-eight places. What a café queue does not have is anybody standing in it twice.

The two that were busy and silent

Eleven pubs produced nothing. Nine of them had fewer than four people in the room, which is the ordinary condition of a pub on a weekday afternoon and is the walk's timing rather than the pub.

Two were busy and silent, and in both the seating was in booths.

Two observations, and they are printed because they are the only two places in the record where a physical arrangement and an outcome were noted together.

What the pub shares with the other two

Places of worship at fifty-seven per cent, pubs at fifty-five, community halls at thirty-six. The other nine kinds at twelve per cent together.

The three have nothing in common across the measured fields: pubs charge at a counter, churches ask nothing, halls want a booking; pubs seat almost nobody for free, the other two seat everybody; pubs limit in four cases, churches in none.

What they share is unmeasured, and the fifth section's reading is that all three have an occasion in them — something everybody present is doing at the same time.

A field for that would turn the reading into a count, and it is the fourth of the six questions a second walk should answer.

What this does not establish

That pubs are better. This record ranks nothing. It counts one narrow thing — whether an exchange happened between strangers while somebody watched for forty minutes — and a room is not summarised by it.

That alcohol is the mechanism. The place of worship scores higher at fifty-seven per cent, and the community hall third at thirty-six. Whatever the three have in common, two of them have nothing to drink.

That it transfers. One city, one season, thirty-one pubs, one observer, mostly daytime. The figure for anywhere else is unknown and nothing here attempts a guess at it.