Friday, February 10, 2012

A Russian answer to Escher’s day/night print

In the last issue of ISCC News (# 454), Hugh Fairman reported on Henry Hemmendinger’s search for an M.C. Escher print that seemed to transform from a day-lit scene to a night scene when the illumination spectrum changed. Was this a deliberate trick from clever use of more than three colorants? As is clear from Hugh’s article, it was not, but before I learned that fact, I extrapolated Henry’s search far afield, to the heartland of Russia.

In the summer of 2008 I was in Tambov, teaching Russian college students how to use English to advance their science careers (see Issue # 435, p. 6). For three weeks I stayed in the same dorm room, and every morning and evening had the same view from the same chair of a picture on the wall. By day the picture was a sun-lit landscape with some water and green shrubbery in the foreground. By night (in tungsten light) the picture appeared instead to be moon-lit, partly because the sky around the sun/moon orb was darker, but mostly because the green of the shrubbery appeared relatively lighter. The tungsten light was evidently rendering the blue sky darkly, but was raising the lightness of the shrubbery as if to mimic the Purkinje shift (shift to rod dominance in low light levels---hence greater lightness of green). Of course, the Purkinje shift here was illusory and not real, because the light was still bright enough to render colors: my cones still ruled the night.

I thought this might be an example of the art object Henry sought that conveyed two scenes in two lights due to colorant manipulation. I considered trying to purchase the picture, but even if I got a fair deal on it the trip home would not be easy. As I sat in that same chair one evening, I thought I’d have a closer look at the picture before I made my purchase offer. So I rose from the chair. Instantly the green shrubbery darkened.

Oops! This wasn’t related to metamerism at all. The shrubbery was brighter when I sat in the chair because I was receiving a specular reflection from the tungsten light. In daylight, I didn’t get a specular reflection, so that is why the shrubbery looked darker by day.

I had to marvel at this picture, which had different gloss in different areas. The shrubbery had the greatest gloss. Do reproductions in Russian dorm rooms have such texture and gloss differentiation, or was I looking at an original painting? Later I learned that, in printing, ink over-loading (hence gloss) is more likely in the green (and purple) than in other colors because more than one colorant maximizes its load. But meanwhile, I had reached the end of my skills as an art connoisseur, and the end of my time as well---I had to return home the next day.

Was this day/night trick deliberately set up? Perhaps not. Was it related to metamerism? Definitely not, unless you count the much-disparaged term, “geometric metamerism.” Alas, I couldn’t discuss the matter with Henry Hemmendinger, who by that time had gone where colors are more real and permanent.

Michael H. Brill
Datacolor