[lalaithion]

> Or did they followed some method that is today lost, possibly based on the position of a start or a constellation, resulting in an alignment on the Mercator projection?

I would not say that the method is today lost; while not aligned on a geodesic, the cathedrals are aligned on a spiral starting at the North Pole and ending at the South Pole which always travels at a constant bearing. It’s totally possible to recreate a line like this with a compass by walking such that the compass has a constant reading. Doing so over the Mediterranean, multiple mountain ranges, and bits of the Atlantic would be a bit difficult, though.

[mensetmanusman]

I wonder if the answer is along these lines. Eg. a star path relative to the horizon from some reference point…