A drip painting by Jackson Pollock

The Institute of Clinical Evaluative Sciences in Toronto recently conducted some unusual research. It used its database of 100,000 patients and tried to correlate the prevalence of specific diseases in people with their astrological birth signs. We will skip the details, except to say that the hospital found an association between each sun sign and at least two diseases. The Canadian hospital was not trying to prove that astrology is scientific, but was merely trying to warn about the dangers of interpreting statistical correlations simplistically.

While this experiment was going on, a professor who was born in India and a physics graduate student at Case Western University in the US were trying to do something similar in a completely different way. They were telling physicists and art aficionados, in a communication to the journal Nature, that mathematical patterns found in some paintings were just as insignificant. These paintings were supposedly by Jackson Pollock (1912-56), the controversial American painter.

Taylor claimed fractals were part of all Pollock’s works

R.P. Taylor, a professor of physics at University of Oregon, found repeating patterns called fractals in these works, patterns he claims to have seen in all of Pollock’s “drip paintings”. The paintings under contention would be worth tens of millions of dollars if Taylor was right. But the two Case researchers, Harsh Mathur and Katherine Jones Smith, believe that he is wrong. And they are following up the Nature article with more analysis. “We would not be able to tell you the results now, but they clearly substantiate our earlier findings,” says Mathur.

Jones Smith, a theoretical cosmologist, works on topics like gravitational radiation, issues far removed from the commercial world of celebrity paintings. She stumbled upon Pollock’s world by accident. She had made some doodles, and found that they too exhibited patterns that Taylor claimed were part of all Pollock’s works. She kept this finding to herself but started investigating the topic more and more when she heard that Taylor’s method was going to be used to identify the disputed paintings. She then took the help of another professor in the physics department, Harsh Mathur, and published her work.

A condensed matter physicist, Mathur did his undergraduate degree from the Indian Institute of Technology, Madras. He had been working on topics like quantum liquids and although he too had not worked on fractals earlier he was willing to help Jones Smith. That was in November last year.

Now the duo continues to work on the problem, analysing more Pollock paintings, and generally gathering more evidence to prove their point. They are now ready to publish another paper.

Fractals are patterns that repeat within themselves on magnification, whorls within whorls lying apparently ad infinitum. They are nature’s art, and by exhibiting fractals in his drip paintings Taylor contends that Pollock had mastered nature. Drip painting is a technique Pollock pioneered. It consists of pouring paint on a canvas laid on the floor while moving around it in an artistic frenzy. Time magazine once called Pollock Jack the Dripper. His technique resulted in patterns that people call abstract or representational, great or childish, sometimes all in the same breath. Whatever it be, it became vitally important for art lovers to find a way to distinguish Pollock from non-Pollock, the master from the imitator.

Mathur and Jones Smith’s paper immediately created a controversy within the community of scientists and artists. One of their contentions was that the physical range of Pollock’s paintings, from the size of the canvas to the smallest rendering of pattern in paint, was not large enough for the patterns to be called fractals. The smallest fleck of paint was a thousand times smaller than the entire painting itself, and this was not small enough for the scientists. In fact, you can’t call them fractals. Taylor says that this definition excludes 50 per cent of what we recognise as fractals in nature. On the other hand, says Jones Smith, some patterns that Taylor observed in Pollock are repeated in many other works. “We analysed works of student painters in the campus that look similar to those of Pollock. And they also show the patterns,” says Jones Smith.

Mathematical analysis of works of art can indeed be a good way of distinguishing the genuine from the fake. But science could still be some way from achieving this goal. As the Toronto doctors who did the medical experiment warned, scientists sometimes tend to correlate things too easily. That precisely is the reason for citing medicine in a story about physics and art. After they published the Nature communication, Mathur and Jones Smith analysed more paintings of Pollock. Art lovers and physicists will be waiting for their paper.