UNDER THE MICROSCOPE: WHY COMPARE ART TO SCIENCE?: I just don't understand it

I REMAIN suspicious and uncomfortable with the idea that science and the visual arts have a great deal in common. However, I now feel the need to rethink my position as, for the past six months, the leading science journal Nature has run "Art and Science", a column by distinguished Oxford University art historian, Martin Kemp.

His principle has been to seek what he calls "structural intuitions". By this, he says he means, "those elements in our perception and modelling of the seen world (including what we believe lies behind appearance) that are the common province of anyone who indulges in the critical exercise of our visual faculties in the most searching manner - whether artist or scientist. The 'structural' element relates both to the patterns that can be extracted from nature at various levels of complexity and to the structures that operate in our perception, cognition, and visualisation." I find this very difficult to understand; his example of the recurrence of decorative motifs such as spirals does not help me in relation to science.

Some artists he claims, were able, during the "Scientific Revolution", to play an active role in the dialogue between seeing and knowing. The fiery emissions of Joseph Wright's late 18th-century paintings of volcanoes are given as an example. Wright's painting of Vesuvius erupting is dramatic but owes nothing to geology. Poetic license is needed to claim that this picture did much to convince European scientists that the great mystery, life, might be explained in terms of electrochemical forces. It may be, however, that Jan Vermeer indeed discovered that more compelling images can be achieved through a kind of optical illusion that makes special use the perceptual system inside our brains rather than through the details that reach our eyes.

Kemp regards Andy Goldsworthy unrivalled in skill as a "nature artist". He makes sculptures out of leaves and pebbles, and in others blades of grass are arched into a snake-like form. These forms, according to Kemp, cut across our conventional ideas of genus and species. That is fine and imaginative but when he likens him to a modern D'Arcy Wentworth Thompson who uses mathematics and physics to discuss biological forms in his wonderful and influential book Growth and Form, the analogy seems absurd. When Goldsworthy writes: "the snake has evolved through a need to move close to the ground, sometime below and sometimes above, an expression of the space it occupies", it may sound like science but it is not - it merely uses some of the same words.

Computer generated images are now quite common, particularly since Mandelbrot generated beautiful fractal images. New images based loosely on an evolutionary paradigm are indeed strange and monstrous and show the importance of new technologies for art, but do not in any way illuminate biology. And the work of the artist Herman de Vries is about collections of, for example, varieties of earth which may be unclassifiable and so "alerts us to a multitudinous variety that far outstrips our leaden vocabulary." Natural historians should not be impressed for theirs is the true triumph of classification.

Why the visual arts should be singled out for its relation to science remains puzzling. There can be little doubt that science has influenced artists and that both are influenced by the prevailing culture. But of all the arts, painting is the one with least intellectual content and requires the least effort to appreciate. One has to work so much harder with poetry or music, and to understand science can require enormous effort. Why not sport? Just think of the possibilities of waxing eloquent about the similarities between the Newtonian mechanics of golf, cricket and tennis.

But I do remain very impressed by Kemp's scholarship and look forward to his new series in which he will explore the use of visual images in science. That I may understand.