The science of style and fashion

July 22

Yesterday I recalled art historian Anne Hollander, who brought rigorous critical thought as well as imagination to the historical evolution of clothing and dress, style and fashion.  Hollander did so through the traditional methods of the humanities, which in one of its core functions I suppose might be summed up as … bringing sense to sensibility.  Yet it must be said that Hollander’s disciplined methods of criticism, the humanist tools of her intellectual trade, are largely alien to today’s social sciences and empirical social psychology.  The field that belonged to Anne Hollander seems gradually to have been ceded, in much of its serious intellectual aspirations, to empirical social psychology.  The rise of social sciences comes in its turn with a grant of expert authority in addressing the educated middle class, usually under the headline: #StudyShows.  Consider, then, the following study (published in PLoS One, July 17, 2014) on the “objective features” of fashion:

The Science of Style: In Fashion, Colors Should Match Only Moderately

AbstractFashion is an essential part of human experience and an industry worth over $1.7 trillion. Important choices such as hiring or dating someone are often based on the clothing people wear, and yet we understand almost nothing about the objective features that make an outfit fashionable. In this study, we provide an empirical approach to this key aesthetic domain, examining the link between color coordination and fashionableness. Studies reveal a robust quadratic effect, such that that maximum fashionableness is attained when outfits are neither too coordinated nor too different. In other words, fashionable outfits are those that are moderately matched, not those that are ultra-matched (“matchy-matchy”) or zero-matched (“clashing”). This balance of extremes supports a broader hypothesis regarding aesthetic preferences-the Goldilocks principle-that seeks to balance simplicity and complexity.  Authors: Gray K, Schmitt P, Strohminger N, Kassam KS.

PLoS One is a real journal (if you’re unfamiliar with plos.org), by the way, though an unconventional online one that aims for open access for medical and scientific research across many fields.  I’m an admirer of the project.  I ran across the article because I was looking to see how aesthetics, including human dress, was addressed in social psychology, in order to understand how it approaches, or will approach, the aesthetics of machines that engage in human interaction.  I read this article in order to understand how, as a matter of method, social psychology deals with human signaling behavior, both sending and receiving–because I need to understand better the capacities and limits of empirical psychology in understanding how people might react to different types of humanoid robot designs.

I learned something useful for my work from this article, in other words. True, what I learned was mostly the limits of this methodology applied to aesthetics. I understand the inevitable comment that this article is merely “Department of the Obvious” confirmation.  Phrases like “maximum fashionableness is attained when outfits are neither too coordinated nor too different,” which, you’ll be pleased to know, exhibits a “robust quadratic effect”–well, with apologies to the authors, on first downloading and glancing at the article, I actually thought it was a joke or academic parody. I had to double check to be sure it was real.

But beyond the cringe-worthy language of ‘sciency-ness’ (swiping from Colbert’s ‘truthiness’), I found myself thinking skeptically about the methodological reductivism built into this kind of empirical study.  Of what is this method “revelatory,” for example, and are there crucial features not captured by this method? Are these seemingly plainly “objective features” quite what they seem? Color coordination seems objective, but it is framed in this study in relation to “fashion” and “fashionableness.” There are subtle differences between those two, conceptually; then there is the distinction between “fashion” and “style,” as Hollander observed; and then the distinction between “style” and “glamour,” of which Virginia Postrel has written.  Whatever objectively might be measured by asking a sample of people a uniform list of questions about color coordination preferences seems to me set against something of a moving target with respect to these other qualities.

Still, the article is right to observe that (even from a purely practical, boringly utilitarian point of view), the fashion and clothing industry is a market of something like $1.7 trillion. Little is understood, it says, about the “objective features,” the scientifically, empirically measurable features, that “make an outfit fashionable.” In an economic activity of this size, I quite agree it’s not irrelevant that, even if this study is merely confirmation of the obvious, possibly it’s the beginning of learning genuinely new things that would inform economic and business decisions.  The hope for big data in the study of consumer behavior is not so very different in its aims, after all; it wouldn’t be impossible to see the categories this study attempts to assess someday undergirding a big data fashion study.  All this said, though, the definitive statement on all these questions remains the big speech Meryl Streep makes to Anne Hathaway’s character in that chick flick my daughter and her friends used to giggle over in elementary school, The Devil Wears Prada. “Pile of stuff,” indeed.

Kenneth Anderson teaches law at Washington College of Law, American University; he is also a non-resident senior fellow of the Brookings Institution, member of the Hoover Institution Task Force on National Security and Law, and senior fellow of the Rift Valley Institute.
Comments
Show Comments
Most Read National
Next Story
Eugene Volokh · July 22