When we look at plans of buildings, we are accustomed to two important conventions that indicate the building's orientation: the plan usually has a north-arrow; and north is usually "up," at the top of the page. However, these are modern conventions, and so it's easy to be misled by looking at old drawings and inferring orientation at a glance. I was reminded of this issue recently while re-reading Dean Hawkes' excellent book Architecture and Climate: An Environmental History of British Architecture, 1600-2000.
Here's an early engraving of the Chiswick House (London, 1725-29):
You'd be forgiven if you assumed the front of the house faces south. (The assumption is not only understandable for reasons of graphic convention, but also because we are now trained, to some extent, to assume 'wisdom' in the environmental strategies of pre-modern architects --- of course they wanted a south-facing porch.)
Hawkes mentions (almost parenthetically), the Chiswick House faces
south-east. I've rotated it to show how it might be better-portrayed for modern readers:
And Hawkes also noted that Chiswick's diagonal orientation was derived from Palladio's Villa Rotonda, although the "main entrance" at the Villa Rotonda faces northwest. Is that a surprise? (It surprised me ... and I've been there!) Here is Palladio's canonical drawing:
The unconscious association of the graphic orthogonality with the cardinal points is amazingly powerful. You project the drawing onto the world. Now, here is a more 'accurate' representation of the plan:
To be clear, this is not to criticize the illustrators (Herisset and Palladio) for orienting their drawings in this manner, or the architects (Lord Burlington and Palladio) for the orientation of the buildings.
Another famous example, I think, of a historic building with a misunderstood orientation is Hardwick Hall:
Hardwick Hall is sometimes cited as a forerunner of passive solar design. This is the east facade however, not the south.
Any others? Please comment below!