Medieval Architecture

French philosopher Jacques Maritain wrote the following about architecture in the Middle Ages:

[A. Vailant] points out that since in those days building plans could not be made on paper as nowadays, and since the only drawing-material at their disposal was rare and costly vellum, which they used sparingly and washed so as to use again, "the projected work was represented in its essential elements chiefly by means of a reduced model. One worried about details only at the moment when they were about to take shape, when one knew the scale exactly -- and then by using familiar rules and elements. It was on the job that the solution of all the building problems was considered and discovered, and the difficulties overcome.

"When one thinks of the enormous quantity of paper required by us in order to plan and prepare the erection of our modern buildings, and of the calculations indispensable to the working out of our slightest projects, one is amazed at the depth of intellectual power, the range of memory, and the positive talent of the master builders and foremen of those times, who were able to build these grandiose and magnificent monuments, inventing daily and ceaselessly perfecting."

Jacques Maritain, Art & Scholasticism ISBN  0828898545

JustInTime design is not a new idea.

In some cases, they also had two hundred years or more to build a cathedral. And a lot of them fell down. Big, expensive, failures of JustInTime design are part of medieval history just as much as the successes. In some cases, they're the same building (see Ely cathedral - and you should).

Seconded. Medieval structural engineering should not be our goal to emulate. And do see Ely cathedral; it's my favorite in England. -- RobertEikel

Were the failures due to the master builders' JustInTime design? If they'd had square miles of paper, would they have had better success rates? If you're not going to finish the cathedral for another eighty years, do you need to work out where your successor's successor's successor is going to place the Ladychapel's columns?

The failures were due to ambition without structural engineering. We can put up amazing structures today because lots of people have built tools to work out the numbers. Very, very few of them fall over (although quite a lot seem to leak). Incidentally, one of the interesting things about most buildings until, say, the neo-classicists turned up, is how malleable they were. The cathedrals are full of bits that got added and removed as requirements changed.

This brings up a fairly interesting point: For the most part, in today's large, monumental, architecture, the design is fixed, because the design is the product. It's hard to imagine, for example, the Bilbao Guggenheim adding a new wing to their fancy FrankGehry?-designed building. And, hey, as long as you're adding a new wing, why not get away from all that stainless steel and make this wing with a wood-and-glass facade!

MedievalArchitecture was the opposite of this: A design needed to be useful at all times, so if it's a hundred years after the initial construction, and you need to add a new wing, then you add it, no problem. Making the building useful seemed more important than making the piece an expression of the original architect's singular, majestic idea. You could make the point that this seems to be a good example of Modernism vs PostModernism, and an interesting refutation of BigDesignUpFront. (Also, ChristopherAlexander probably has something interesting to say about the whole topic.)

See HowBuildingsLearn: a fascinating book about how buildings can adapt to new uses.

I'd challenge the assertion that little drawing was done before these buildings were put up. At RoslynChapel? (a remarkable building for many reasons) on the very oldest parts of the building may be seen drawings for the more recent parts, scratched into the stonework.

As mentioned above, what the mediaeval builders suffered from the lack of was not "big design upfront", which in fact they did have, but structural calculation. Designing a building by scaling up a model without any clear, quantitative understanding of what this means for the forces in the structure is bound to fail above a certain size.

There's some evidence that classical greek building proceeded in a similar fashion. Of course, although it looks very heavy, the structural forces in a "classical" building are much lower than in a light, gingerbread looking gothic building.