Vincent van Gogh’s “The Starry Night” may have been based on one specific night sky: July 22, 1889, just before dawn over France. If that’s true, it means the famous painting was based on real observations, not just van Gogh’s imagination, astronomer Elijah Jeong-hyun Kim and colleagues report October 22 in Scientific Reports. About a month earlier, van Gogh had entered a hospital in Saint-Rémy-de-Provence, France.
A short time later, van Gogh experienced an illness during which he didn’t paint for about six weeks, so most historians think “The Starry Night” must have been painted before then. That letter is taken as strong evidence, says astronomer Donald Olson of Texas State University in San Marcos, who has written about how astronomy shows up in art history. Historians favor the June date because “van Gogh himself said so,” he says.
That assumption has led to some confusion, though. In the 1980s, two astronomers tried to match the painting’s stars to real skies in Saint-Rémy in mid-June. One proposed that the sky was from June 19, 1889, and the collection of swirly stars is the constellation Aries.
But the other concluded it couldn’t have been painted on June 19, based on cloudy weather reports in Aix-en-Provence, France, about 70 kilometers away. That study claimed the painting was “a celestial composite, constructed to make an image of the morning sky.” Kim, who founded a telescope-building company called SLABB INC in Seoul, South Korea, didn’t know about the June date when he became interested in “The Starry Night.” “Unlike previous astronomical studies, we could start freely,” he says. Given his job, Kim had an intuition for which stars would be visible when.
He was skeptical of the Aries theory. The stars in that constellation are faint enough that they wouldn’t be all that visible in the predawn sky, “hardly enough to inspire a painter,” he says. Instead, Kim saw the stars as the Hyades star cluster, which forms a sideways V shape in the constellation Taurus.
To test that theory, he and his colleagues applied a statistical technique called Procrustes shape analysis to put a number to how well the stars in the painting matched any particular constellation. Researchers compared the V-shaped collection of stars in “The Starry Night” (first image) with those in the Hyades star cluster (middle image). The resulting graph (last image) maps the real stars (blue) and the painting’s stars (orange) against each other.
Extract — continue reading at the source.
