Make a list of the places in your city where you can sit indoors, in warmth, for two hours, without buying anything or explaining yourself. It is a short list. In most European cities it is libraries, some museums on some days, a handful of municipal buildings, and very little else.

This is a relatively recent scarcity. The decline of the pub, the church and the community hall has not been matched by anything, and the retail environments that replaced them are conditional: you may stay as long as you are consuming. That condition sounds mild until you are fifteen, or unemployed, or new to the country, or simply cold.

Civic buildings are what is left. That gives them a job that has little to do with their nominal function, and it should change how they are designed.

Entry is the whole problem

The design difficulty in a civic building is concentrated almost entirely in the first ten metres.

A library user who already uses libraries will find the door. The person the building most needs to reach is the one who is not sure they are allowed in — who suspects there is a membership, a form, a person who will ask what they want. Every threshold detail either confirms or dispels that suspicion, and most confirm it.

The usual offenders: a raised entrance reached by steps, a lobby with a desk facing the door, a security barrier, a single door in a large facade, glazing that reflects rather than reveals. None is a decision anyone would defend as exclusionary. Together they say this is not for you clearly enough that people who take the hint never test it.

At a district library in Oslo we put the entrance at the lowest corner of a sloping site, level with the pavement and directly opposite a tram stop, and made it possible to see nearly the entire collection from outside the door. You can tell what the building is before committing to entering it. That sounds like a small thing. The visitor numbers tripled against a projection of forty percent growth, and the building now hosts language classes, a repair café and municipal advice sessions, none of which were briefed.

Legibility beats signage

A civic building should be understandable without being explained. If you need a sign to know where to go, the plan has failed and the sign is a patch.

The most reliable device is a single organising move that the whole building submits to. At Halden that move is the section: one continuous room stepping up the slope in seven half-levels, connected by a wide stair that doubles as seating. Quiet increases with height. The noisiest public functions sit nearest the tram; the reading rooms are at the top. Nobody has to be told this — you can feel it as you climb, and the building sorts its own users.

This is not a novel idea. It is what makes a market hall or a station concourse comprehensible to someone who has never been there. It has become unusual in civic buildings largely because the brief arrives as a schedule of accommodation, and a schedule of accommodation resolves naturally into rooms off a corridor unless somebody actively resists it.

Permission to stay

A building that is free to enter but uncomfortable to remain in has solved half the problem.

The practical requirements are unglamorous. Somewhere to sit that is not a chair at a table, because a chair at a table implies you should be doing something. A range of positions — alone, alongside strangers, in a group — so that people can choose their degree of exposure. Power sockets, which have quietly become an access issue. A toilet you do not have to ask for a key to use. Somewhere to leave a coat.

And crucially, no ambiguity about how long is acceptable. Fixed opening hours, clearly posted, with nothing in the building's behaviour suggesting a shorter welcome — no music timed to encourage turnover, no seating designed to become uncomfortable, no staff sightlines that feel like supervision.

The stair-seating at Halden gets used for hours by people who never touch a book. There was an early conversation with the client about whether that was a problem. It is not a problem. It is the building working.

The objection worth taking seriously

The obvious criticism is that this is a case for civic buildings as social infrastructure, made by an architect, when the actual determinants are funding and staffing.

That is largely correct. A beautifully planned library with reduced hours and no staff is a worse civic space than an ugly one that is open and adequately staffed, and the last fifteen years across most of Europe have been a story about the second variable, not the first. No amount of section-drawing compensates for a building that is shut on Mondays.

What architecture can do is make the case easier to defend. A library that is measurably the busiest public space in its district is politically harder to cut than one nobody visits. Usage is the strongest argument available to the people fighting for these budgets, and usage is partly a design outcome. That is a modest claim, and it is the honest limit of what the discipline contributes here.

What civic buildings owe

Not grandeur. The monumental civic tradition — the flight of steps, the portico, the double-height lobby with nothing in it — was designed to impress upon citizens the seriousness of the institution, and it works exactly as intended, which is the problem.

What is owed is easier: a door at pavement level, a plan you can read from the entrance, somewhere to sit without buying anything, and no moment at which anyone is asked to justify being there. Those are cheap requirements. They are omitted routinely, mostly because briefs do not mention them and nobody notices their absence except the people who quietly do not come in.