The distance at which holding a door becomes an obligation

A rule expressed entirely in metres

Nobody has ever been taught how far away a person must be before you stop holding a door for them, and yet almost everybody applies a threshold, and the thresholds broadly agree. Close behind, you hold it. Some distance away, you do not. Somewhere in between lies a band where neither choice is comfortable.

It is one of the purest examples of a convention with no author. There is no authority, no written form, and no penalty beyond a moment of mild embarrassment. Despite that, it is applied hundreds of times a day in every public building, and the failures are rare enough to be memorable.

What the threshold is actually balancing

The calculation is between two costs. Holding a door for somebody far away means standing still for several seconds, which is a real imposition on you and, oddly, on them. Not holding it for somebody close behind means letting a door swing into a person, which is worse.

The distance where these balance is not fixed. It stretches when the door is heavy, when the person approaching is carrying something, when a child or an older person is involved, and when it is raining. It contracts when you are in a hurry and when the following person is obviously not. That flexibility is what makes the convention robust: it is a judgement about relative cost, not a measurement, which is why it survives situations nobody anticipated.

The courtesy jog exists to correct an error

The most visible artefact of the whole system is the small acceleration people perform when somebody holds a door slightly too far in advance. It is rarely fast enough to arrive meaningfully sooner, and that is not its purpose.

The jog is a signal. It acknowledges that a cost has been imposed and demonstrates willingness to reduce it, which is why a person who does not jog and simply walks at the same pace is felt to have taken something without acknowledging it. The whole exchange is a small repair of a misjudged threshold, performed by the person who did not misjudge it, and it works so reliably that most people have never thought about it.

Thanks is doing structural work

The verbal acknowledgement at the end is not decorative. It confirms the transaction was received and closes it, which matters because the holder has invested a few seconds and has no other way of knowing whether it registered.

This is why a held door that goes unacknowledged is irritating out of all proportion to the effort involved. The cost was trivial; the failure was in the closing of the loop. And it explains why an over-elaborate acknowledgement is also slightly uncomfortable, since it prices a two-second courtesy as though it were a favour, and the convention depends on both parties agreeing that it is not.

Doors that break the convention

Certain doors defeat the whole system, and it is worth noticing which. Automatic doors remove the possibility, which is why people sometimes make an odd half-gesture at them out of pure habit. Revolving doors have their own separate protocol involving compartments and speed. Doors that open outwards into a corridor create a hold that blocks the passage, so the courtesy imposes on a third party.

The interesting case is the double door, where holding the first without the second leaves the follower in a small vestibule facing an unheld door. Most people either hold both or hold neither, and the awkwardness of the halfway version is a good illustration that the convention is about the journey rather than about the individual door.

Why this one varies less than most

Door-holding is more consistent across cultures than many conventions, and the likely reason is that it is grounded in physics. A heavy door swinging back is unpleasant everywhere, the time cost of waiting is the same everywhere, and the distance at which those two costs balance does not depend much on local custom.

The variation that does exist is around gender and age, where expectations differ considerably between places and have shifted over time within many of them. That part is genuinely contested and not something to be settled here. The distance calculation underneath it, though, appears to be close to universal, which is a slightly surprising thing to be able to say about a rule that has never been written down anywhere.

What the whole thing demonstrates is how much coordination people manage without instruction, provided the situation is visible and repeated. Two strangers who will never speak again resolve a small logistical problem in under two seconds, using an estimate neither could state numerically, and both walk away with the sense that nothing happened. Most of the conventions in this section work the same way. Their success is measured by how completely they go unnoticed.