The Marriage of Reason and Squalor: drawings by Pier Vittorio Aureli

The Marriage of Reason and Squalor: drawings by Pier Vittorio Aureli — “Pieces of architecture-in-waiting that, despite rejecting a precise programme, are waiting sheet after sheet, to turn into something better defined, which is not necessarily architecture.”


On first principles

From a profile of Friedrich Kiesler in Architectural Forum (1947)

“If Kiesler wants to hold two pieces of wood together, he pretends he’s never heard of nails or screws. He tests the tensile strengths of various metal alloys, experiments with different methods and shapes, and after six months comes up with a very expensive device that holds two pieces of wood together almost as well as a screw”

Reality reduced to a model


Reading through Geoff Manaugh’s interview with Nicholas de Monchaux, author of Spacesuit: Fashioning Apollo, I came across this:

“If you lay, side by side, quotations from USC’s discourse on parametric urbanism now and USC’s discourse on cybernetic urbanism thirty years ago, for better or for worse, you can read them as a complete narrative. It’s impossible to distinguish which is which. Both are born out of a fundamental faith in technology and a fundamental notion that, if you feed enough variables into a problem-solving system — now we call it parametric, then we would have called it cybernetic — that an appropriate and robust solution will emerge. I’m not, myself, so sure that’s the case; in fact, I’m pretty certain that it’s not.”

Models are translations of reality, and all translations are partial, reducing some dimensions in order to allow a degree of fidelity in other dimensions. Something is abstracted away in order to concentrate something else: a white paper model might leave out any suggestion of materials to highlight volume; and acoustic properties may be excluded from an electronic model that simulates solar gain. Breaking a situation down into variables and parameters so it can be modelled is no stranger than any other kind of modelling. Problems arise when reality is reduced to the model.

I agree with de Monchaux that appropriate and robust solutions don’t emerge automatically given enough parameters; and appreciate his reminder that, despite a varnish of novelty, parametric approaches are not new.

(Again, I’d plug Adam Curtis’s All Watched Over by Machines of Loving Grace (2011) on this topic).