Skip Navigation
Search

Measuring beauty

October 15, 2004 By Philip Ball This article courtesy of Nature News.

Philip Ball asks whether a method of persuading people to tell the truth in subjective surveys might give us a more robust way to judge the quality of art.

Please log in to rate this page.

View Comments

Can art be assessed objectively? John Ruskin thought so in the nineteenth century, but today such views are often seen as elitist and arbitrary. The reviews of critics are typically dismissed as just 'one person's opinion', and postmodernists insist that there is no privileged position from which to make aesthetic judgements.

But what would it mean if there were indeed some way to reach a reliable view on the intrinsic value of a work of art, as a new method of data-gathering proposed by Drazen Prelec of MIT in Science this week1 (see " Truth serum could make polling more accurate" seems to imply?

In his paper, Prelec shows how his technique for introducing an incentive-based 'truth serum' to surveys that require subjective judgements might be applied to groups of experts and lay people asked to evaluate the originality of an exhibition by an artist. Both groups are best rewarded by offering an honest opinion.

Our aesthetic judgements are corrupted by other people's judgements.
Drazen Prelec
Mathematical psychologist, MIT
At face value, this is just a way of conducting a better survey - it means you can be more confident in a finding that, say, 80% of lay people dislike an exhibit. That doesn't in itself make the art 'bad' - just unpopular. But couldn't a technique for improving the accuracy of subjective assessments encourage the belief that the judgement has more validity? Might it bolster attempts to measure, and thus to define, 'quality'?

Corrupted views

The problem with asking people what they think about art, or any other subjective question, is that whether through lack of confidence or the wish for acceptance, we tend to fall in line with what others say. "Our aesthetic judgements are corrupted by other people's judgements," Prelec says. This can suppress unusual views, or it can also encourage them, out of a reactive impulse to 'be different'. Neither response is authentic.

But if, using a method like Prelec's, we are able to ferret out people's genuine subjective truths, can they help us to find our way to a trustworthy 'interpersonal' truth: to pronounce on what is and isn't beautiful, say?

People have very little confidence in their own evaluations. That's a dangerous situation for society to be in.
Brian Goodwin
Biologist, Schumacher College in Devon, UK
Biologist Brian Goodwin of Schumacher College in Devon, UK, believes that science needs to find a reliable way of doing just that, rather than focusing solely on quantities. He advocates a technique called consensus methodology, which claims to find a robust and consistent consensus on subjective questions.

"This procedure preserves an essential aspect of scientific methodology: reliable knowledge depends upon consensus between individuals who practise an agreed procedure of investigation", Goodwin says.

Cultural theorist Charles Jencks is wary of this kind of thing: he fears it is a prescription for mediocrity, whereas beauty and innovation don't usually come from a centrist consensus. Goodwin argues that consensus methodology avoids that danger, however. "Every individual has the right to disagree with the consensus, and they can change it by convincing others that their evaluation is more consistent with the evidence available."

Goodwin thinks that such procedures can boost the trust we place in our opinions. "People have very little confidence in their own evaluations", he says, "and that's a dangerous situation for society to be in."

Tide of opinion

The common view seems to be that if art appreciation were based on opinion polls, innate public conservatism would crush innovation in the same way that it condemned Impressionism in the 1870s or Stravinsky's ballet The Rite of Spring in 1913.

But new art surely does emerge from a collective view of quality, based on the opinions of both lay and 'expert' audiences. There is no reason to think that important innovations arouse universal condemnation; if they did, they would not survive.

El Greco had royal patrons; J. M. W. Turner painted successfully in a virtually abstract style in the early nineteenth century. The Impressionists always had discerning and vocal advocates (it's just that most of them weren't French). And the near-riot in Paris at the premiere of The Rite of Spring was in response to Diaghilev's challenging and non-classical choreography, not Stravinsky's music - it was drowned out by the catcalls anyway, and most reviews barely mentioned it.

The most 'iconoclastic' of today's art tends to attract the richest sponsors and the most critical acclaim, despite the public ridicule that often greets it. The idea of brave new art breaking down a wall of unrelenting opposition is largely a popular fantasy.

Applying 'science' to subjective matters has a justifiably bad press - it conjures up images of chemistry Nobel laureate and colour theorist Wilhelm Ostwald denouncing Titian for using a blue that was "two tones too high".

But Prelec is not advocating an art criticism based on some number spewed from a computer, nor from the equivalent of a public show of hands. Rather, he is suggesting a way to get a more accurate view of what both the public and the 'experts' really think.

We'd be foolish to interpret what comes out of such surveys as an indication of what one ought to think. But there is nothing like a reality check for keeping us honest. As Prelec points out, "A futurist, or an art critic, can comfortably spend a lifetime making judgments without the reality checks that confront a doctor, scientist, or business investor."

It's not a question of determining whether such judgements are 'right', but of finding a way to ensure they are made honestly rather than being the product of spineless conformity or contrarian posturing. That might even be a handy attribute to bring to the peer review process.

References

  1. Prelec D., Science, 306. 462 - 466 (2004).

Comments

User Tools [+] Expand

User Tools [-] Collapse

Pinterest button

Favorites

Please log in to add this page to your favorites list.



Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.