Stopping the Watch

The watch stops at the church door.

That sentence opens this book because it is the only rule kept through eight years of walking, and because it sits badly against everything that follows. What follows is a book of counts. One hundred churches. Two hundred and thirty-two recorded visits. A hundred and forty photographs, one of them a frame from a video, each dated to the minute as the photo library records it. Every church placed, every historical claim traced to a published source, and everything else to what the photographs show or what the walks recorded. You could be forgiven for thinking the counting was the point.

The counting came afterwards, and it came out of photographs.

The clearest evidence is an afternoon in Neuilly. The photographs of 26 June 2024 show the church’s rounded end, a Rosary chapel and a carved doorway, and no name. That visit entered this record on 26 October 2026, when a pin on a map resolved to Saint-Pierre de Neuilly, seven weeks after a second visit on 8 September. Nothing happened in Neuilly in October 2026. What happened was that the record caught up with an afternoon already lived.

No book was planned while any of it was happening. The walking came first, and the record is a second thing, assembled later out of what the walking left behind. That is worth saying plainly, because the two are easy to confuse once they are bound together, and because the order matters. A hundred is not an achievement.

The record is built as carefully as it is for a reason, and the reason is not to make the number impressive. A number becomes impressive when its keeper is generous with it, and generosity is exactly what the counting rules at the end of this book prevent. A church enters the record only once its door has been crossed. A visit is a day on which that happens. Several photographs taken in the same church on the same day count as one visit, which gives up visits the record could otherwise claim. Where a photograph and memory disagree, the photograph wins.

The rules are there so that you can check the record, and so that it can be shown to be wrong. A list at the end names ten churches with no recorded visit yet, three of them photographed only from the street; they stay out of the count.

None of that apparatus tells you what the walking is like.

On a Thursday morning, a church in the eleventh arrondissement is not expecting you. The door is heavy. It is colder inside than out, and darker than your eyes are ready for. Somebody is doing something practical with flowers. Whatever you carried in with you is still there, but it has stopped talking for a moment, and that moment is the whole of it. There is nothing to count while it is happening, which is why the watch stops.

The book serves those two things in two different ways. The photographs keep moments from the visits. The texts present the churches and their history. The source notes and the counting rules let you check the information, and show what the record leaves out.

Saint-Étienne-du-Mont has eight recorded visits. It does not know that. The number describes a habit, not the building, and that asymmetry is most of the reason to keep going back.

The record stands at a hundred because the cataloguing stopped there. The walking did not.