A building is a thermal argument

A building is a thermal argument with three terms: which way it faces, how much mass it has, and where the openings are. Everything after that is commentary. The cladding is commentary. The render is commentary. The award is commentary on the commentary.

This sounds reductive until you try to find a building where it is false. The ones that come to mind are the ones with enormous plant.

Orientation is free and is set in the first week. Mass is cheap and is set in the first month. Openings are expensive and are argued about for a year. We spend our attention in exactly the reverse order of the leverage available, which is the single most reliable finding across every project this journal has looked at.

Take the three in order. Rotating a block on a loose site costs a drawing. On a tight urban site it costs nothing at all, because the rotation is already fixed by the street and the only decision left is which rooms face which way — also a drawing. The cost of getting orientation right is the cost of thinking about it on the correct afternoon.

Mass costs a little more. A heavier floor build-up or an exposed soffit in place of a suspended ceiling is a real decision with a real price, and it conflicts with acoustics and with services routing. But it is a decision about a build-up, taken once, early, and it is substantially cheaper than the plant it displaces.

Openings are where the money is, and they are where the argument happens. Every window is a negotiation between daylight, view, heat loss, solar gain, fire, security, cleaning and planning. There is no right answer, which is exactly why it absorbs a year: it is the term that rewards discussion, so it gets all of it.

The cover study is a worked example of the inversion. Its fins are 520mm deep at 820mm centres across a reading room 11400mm wide, and that geometry was fixed in month two by a cooling load calculation, before anyone had drawn the elevation it now defines. It is the cheapest decision in the project and the one you photograph.

The study tower stands 9.5m and the upper floor cantilevers 2.0m west past the room below it. That was a planning move. It also shades, because a horizontal thing over a window always does, and it cost nothing because it was already there. It is not in the energy model, which assumed a flat facade.

Draw the shade and the free lunch turns into the standard argument against horizontal devices on a west face. At midday the overhang covers the top two metres of the glass. By six, when the west sun is doing its actual damage, it covers eleven centimetres. A projection fails exactly as the sun it is shading from gets low, which is why a west elevation wants vertical fins and a south elevation wants a brise-soleil — and why the cover study has its fins on the wrong wall to help here.

If the leverage is so lopsided, why does nobody reorder the work? Because the three terms differ in how visible they are to a client. Orientation and mass are invisible at the stage they are decided; there is nothing to present. Openings produce images. A fee structure that pays for stages, and a client who approves what they can see, will pull attention toward the term that photographs — and that term is the one with the least leverage per hour spent on it.

The practices that escape it do one thing in common, and it is not virtue. They produce a number in the first fortnight — a peak load, a glazing ratio, a free-running hours count — and they put it in front of the client before there is an image to look at. Once a number is on the table it has to be argued with, and arguing with it is the design.

That is the whole method, and it is the reason this journal publishes the quantities. Not because numbers are more honest than drawings, but because a number asked for in week two changes a building and the same number produced in month fourteen is a report.