Rembrandt's command of light and darkness makes characters in his paintings appear real enough to touch, from the musketeers setting off on a mission in ''Night Watch" to Christ's crumpled figure in ''Descent from the Cross." Now, two Harvard scientists say a secret of the Dutch master's success may be that the world actually looked flat to him, making him unusually aware of shadows and other details that helped him judge distances.
Margaret S. Livingstone and Bevil R. Conway, neurobiologists at Harvard Medical School, say Rembrandt's many self-portraits reveal that his eyes are focused in slightly different directions, depriving him of the ''stereo" effect that makes vision three-dimensional. As a result, they argue, Rembrandt would have struggled with depth perception -- though he may never have known he had a vision defect.
Rembrandt's flat world view may have helped him more precisely capture reality on a flat canvas, where painters create the illusion of three-dimensions through techniques such as shading. In fact, Livingstone and Conway say that visual artists are far more likely to be ''stereoblind" than the general public, suggesting that limited depth perception may actually be an advantage over normal sight.
''Art teachers often instruct students to close one eye in order to flatten what they see," the researchers write in today's New England Journal of Medicine, explaining their theory about Rembrandt. ''Stereoblindness might not be a handicap -- and might even be an asset -- for some artists."
But the neurobiologists' theory is reopening old wounds for art scholars who say scientists too often jump to simplistic conclusions about creativity -- as if Claude Monet's cataracts can fully explain the beauty of his late paintings. They say the theory that Rembrandt had a vision defect is speculative and not very useful.
''This is just one cog in a gigantic wheel attempting to scientifically explain art away and. . . bring the mysteries of creativity down to scientific fact," said Alan Chong, curator of the Isabella Stewart Gardner Museum in Boston. ''I hate to be really cynical, but it is total nonsense."
Chong acknowledged that the Rembrandt self-portrait at the Gardner Museum is ''a little cross-eyed," but that doesn't mean Rembrandt had a vision defect. He said that many portrait painters in the 17th century and before painted their eyes slightly out of alignment, but that may reflect their training rather than the way they actually looked.
Even if Rembrandt was stereoblind, Chong added, scientists should avoid ''looking for a single explanation for the look of a painting or a style."
Livingstone and Conway say they understand the wariness, but insist they want to enrich the experience of art rather than cheapen it. ''Artists tend to be really anxious that scientists are out there trying to develop some recipe for a good picture," said Conway, himself a painter and photographer. ''The better analogy is, if we understand what's going on in our minds and brains, then the whole experience can be richer."
Livingstone has long taken an interest in applying the science of perception to art and made waves four years ago with her theory that Mona Lisa's smile is an optical illusion. She argued that when viewers look at Mona Lisa's eyes, their peripheral vision perceives a smile because of her high cheekbones, but when they look directly at the mouth, they see that her lips form nearly a straight line. Some artists said Livingstone had used scientific terms for something they long understood.
Livingstone became interested in Rembrandt's eyesight a couple years ago when she saw four of his nearly 90 self-portraits in a single room at the Louvre Museum in Paris. In each image, the eyes appeared to be looking in slightly different directions, an indicator of stereoblindness, a minor visual defect suffered by about 10 percent of the US population.
Measurements of Rembrandt's pupil positions in self-portraits confirmed what Livingstone suspected: 23 of 24 oil paintings showed his right eye looking directly at the viewer while the left eye is looking slightly to the side. In 12 self-portraits done as etchings, the left eye is the one looking at the viewer directly, but the researchers say that may reflect the fact that etchings are printed from an original metal plate, so the artists' rendering is reversed.
Until now, Rembrandt scholars have said little about possible defects in the 17th century painter's vision, instead crediting him with an unusually fine eye for detail both in facial expressions and scenes. Among the artist's nearly 3,000 paintings, etchings, and drawings are some of the most valuable and famous paintings in the world, including ''Storm on the Sea of Galilee," stolen from the Gardner Museum in 1990.
Livingstone and Conway contend that visual artists are several times more likely to be stereoblind than the general public based on photographs of prominent artists at the National Gallery of Art in Washington.
The pair say that 20th century artists such as Man Ray and Frank Stella may have been stereoblind, based on the positions of their pupils. The researchers' broader study of artists and stereoblindness is awaiting publication.
''Some people are good at things for different fundamental reasons," explained Livingstone. ''This is one tiny biological reason why some people might be better at painting."
Scott Allen can be reached at email@example.com