In the finale of the first series of Sherlock, the high-functioning-sociopath-cum-detective is being forced to decipher a series of puzzles in exchange for the lives of four hostages. Given just ten seconds to prove why a painting - supposedly a lost masterpiece by Vermeer - is a fake, Holmes scrutinises the depicted night sky before turning from the frame with a snort of triumph.
“The Van Buren supernova!” he declares. “Exploding star. Only appeared in 1858.”
“So how,” asks Watson, the penny dropping with an audible clang, “could it have been painted in the 1640s?”
This bit of razzle-dazzle deduction may be fictional (the Van Buren supernova doesn’t actually exist) but the technique that Sherlock used to date the painting is most definitely real. This sort of artistic forensics is known as ‘astronomical chronology’ and it has just helped scientists from Texas State University date a series of paintings by the Impressionist master Claude Monet.
Using astronomical and geophysical clues from Monet's paintings, astrophysicist Donald Olson and his team were able to piece together the date of a series of works, painted by the Frenchman during a trip to Normandy in 1883. In fact, their calculations were so detailed that they were able to narrow down the timestamp of one painting to the exact minute.
The painting that proved fruitful was Étretat: Sunset (as seen above), as it contains both a distinctive rock formation and a sunset - clues for both the geography and the chronology of the piece. Once Olson and his team had found area of coast depicted (a landmark to the locals) they took a series of angular ‘declination’ measurements to calculate the path that the sun must have taken to appear as it does in the painting.
They then fed this information into a piece of software capable of winding back the clock, astronomically speaking, by reversing the motion of the stars and working out how the night sky would have appeared on any given day in the past.
However, this only gave them a rough date around 3-7 February. To finesse this further they looked at letters written by Monet during his trip to the area, and after picking out some further clues (the painter was particularly careful in his observations of the tides) they arrived at a final time and date: 5 February, 1883, 16:53pm. Sherlock Holmes eat your heart out.
Unfortunately, despite the romanticism of these findings – to be published in the February issue of Sky & Telescope – it has to be said that they do have an air of triviality to them. Like the date of Creation worked out by 17th century clergyman Bishop Ussher (Sunday, 23 October, 4004 BC, apparently) there comes a point where too much precision can look ridiculous.
This sort of criticism is well established in scholarly circles. Writing in a 2003 paper titled “The Use and Abuse of Astronomy in Establishing Absolute Chronologies", historian John Steele admits that while astronomical chronology can be “a powerful tool” it can also be misleading, “easily [producing] precise and impressive looking results based on invalid assumptions – results so precise and impressive they may not be questioned by scholars in other fields.”
But although the artistic sleuthing by Olson and his team might unfortunately have more in common than we'd like to believe with the fictional deductions of Sherlock Holmes, the idea that we can know exactly when Monet sat down to put paintbrush to canvas is undeniably enticing. And unlike Holmes and his Vermeer, the end result of this bit of 'astronomical chronology' is to make the painting seem more real, not less.