I had the opportunity this week to listen to a wonderful talk by Charles Falco, a physicist from the University of Arizona who has done fascinating work in recent years with the painter David Hockney on the use of optics in early renaissance painting. As he mentioned at the beginning of his beautifully constructed and entertaining talk, his ambition used to be to have his name in the index of a physics textbook; instead, he now encounters PhD theses on the "Hockney-Falco thesis."
As recounted in an article and his FAQ, Falco and Hockney showed via careful analysis that a number of famous renaissance painters (e.g., van Eyck) made use of lenses and mirrors in their paintings. They do this not (only) by showing that the paintings are unreasonably accurate, but also (in some cases) that they are inaccurate in unexpected ways.
This work is interesting for several reasons: first, it was generally assumed that optics were far less advanced at that time, and second it provides interesting insights into how van Eyck and others worked. It is also a fascinating example of how even the most studied objects can have surprises to reveal.
Falco observed that many art historians responded to this analysis by claiming (often stridently) that it was "wrong, irrelevant, or obvious--and sometimes all three." That a more nuanced response is possible was evidenced by Barbara Stafford, an art historian at UChicago, who observed after the talk that precisely because optics distort in various ways, their use represents an aesthetic choice, not (just) a labor-saving device.