The Nørreport underpass project is often summarized by one number: incidents fell by sixty percent in the year after completion. It is a useful number, but it should be handled with care. Lighting did not repair the city by itself. Cleaning changed. Patrol patterns changed. The entrances were made clearer. People also changed their habits once the route felt less abandoned.
The lighting work was still specific. We removed glare from the walking line and put more information on the walls. We made faces readable earlier. We made the exits present from farther away. We avoided high contrast around service doors because dark interruptions invite imagination to do unpleasant work.
The sociologist on the project insisted that we measure behaviour before we claimed atmosphere. That meant watching how people moved, where they hesitated, and whether they chose the tunnel when an alternative route was available. It also meant interviewing people who did not care about luminaires and should not have to.
The best result was not brightness. It was ordinary use. People walked through without performing alertness. In a public passage, that is a serious architectural outcome.