The Plans

To analyze a plan, it must be of good quality. You can't pick up a plan from a brochure or an old book! If you compare several plans of the same cathedral, you'll often notice differences in measurements. In the case of Amiens, I found a discrepancy of over three meters in the length of the transept!

In addition to the problem of measurements, there's another problem: the precision with which the buildings are executed. As you will see, some cathedrals suffer from irregularities and asymmetries. The work is splendid, inspired and perfectly executed, but the layouts are sometimes empirical. The eye doesn't see these details, only the plans reveal them. It's important to understand that, for the builder, the monument's stability, overall harmony and quality of detail were paramount. It's important to understand that a master builder couldn't transcribe a layout on the ground and then sit back and build. As work progressed, the cathedral's footprint gradually emerged. This took years.
Construction began with the chevet, the layout with the transept crossing. To find each enclosure, each proportion, we had to base ourselves on built elements. I'm talking about masonry elements, columns that had already been erected. You had to find the axes, the exact centers, an operation that's only easy on a computer or a drawing board. This difficulty is probably at the root of the inaccuracies that litter floor plans.

Fortunately, Gothic techniques offered great freedom of execution. While masonry blocks were "standardized" in the quarry, each stone used for an arch or vault could be adjusted on site. This meant that slight irregularities between spans were of only relative importance. In the worst case, all that was needed to correct the discrepancy was to modify the profile of the arch (see glossary) or the templates. Note that the latter was usually invisible to the eye and had no impact on a structural plan.

In the Middle Ages, intention, the concrete, took precedence over strict numerical precision, which admittedly made no sense at the time.
The question then arises: how can we justify work based on approximate plans and involving buildings that are sometimes irregular? Methodologically, the position would be untenable, but recently the situation has changed. Thanks in no small part to the work of Professor Andrew Tallon, laser scanning techniques have made it possible to digitize a number of French cathedrals. A new analysis based on these documents has confirmed my work in principle. The gamble of placing logic and intention above formal precision was the right one.

Leica Geosystems ScanStation C10 - Credit: Andrew Tallon - CC-BY 3.0

Laser scanning modeling of Chartres Cathedral (Leica-geosystems)

Scanning the abbey church of Saint-Denis - Credit: Andrew Tallon - CC-BY 3.0

It should be noted, however, that even a scan accurate to a few millimetres cannot erase the building's irregularities. The problem remains the same. The compass reveals the idea, but the digital result is lost in a cloud of meaningless data. We need to be able to correlate the information and interpolate it for statistical processing. For the time being, these processing algorithms, the idea itself, do not exist. While the availability of metric data is invaluable, traditional compass analysis remains the rule.

I would like to point out that the plans used on this site are intended solely to illustrate and explain. I have redrawn some of them for greater clarity. I invite anyone wishing to reproduce or verify the work shown as an example to use only plans produced by laser scanning.