Over on the Online Photographer Ctein has posted a variant of the immortal "what is art?" question in relation to some photographs of Michael Paul Smith. What was he thinking? Ctein, has nobody told you to let sleeping MFA students lie? The last thing any reasonable person wants is a plague of black-turtle-necks getting the idea that we are interested in their opinions. If we wanted their opinions, we would read their blogs.

Against my better judgement and with apologies to you, reader, I actually have a few things to say about this subject. Like most problems of this sort, the difficulty is not finding the evasive answer, but rather defending against the muddled question. The question 'what is art?' asks for a definition and thus assumes a definition is useful or even possible. Take a look at the comments following Ctein's post and you will find an army of proposed definitions. Some are nonsensical: "Art is an expression of the infinite." Some from the dictionary, "the expression or application of human creative skill and imagination." Some a little more solipsistic, "Art is that which I happen to admire."

We learn early on that when we don't know what something means, we should go to the dictionary. The dictionary helps provide the meaning of the unknown term by providing other terms which we presumably know. It's a funny sort of arrangement when you think about it because we somehow learned what the words in the definition meant without a dictionary or we looked them up in a dictionary which provided yet more terms, which we may or may not know. The entire structure of definitions depending on each other seems circular without ever grounding itself in meaning. Jacques Derrida (I feel a collective swoon from the black-turtle-necks at the mention of his name) referred to this as postponing or deferring the meaning. This became one of the foundations of his idea of différance. It is a strong critique of the idea that we understand meaning through definitions of words. Definitions are good for giving you a sense of a word based on what you already know, but aren't necessarily the way we understand language and meaning. We seem to be able to use words correctly even when we can't define them.

So how do you get out of the dilemma of defining art?

It's simple, actually: you don't need to define art, you simply need to recognize that art is a category. Of course this begs the question, what conditions must be be met for inclusion in the category of art. But here, unlike a quest for a definition, we are on philosophical terra firma with lots of theory to help us.

Lets go back to Aristotle. Aristotle had the idea that everything belonged to a class and although the definition of the class could be elusive, with enough work a smart enough person could identify the defining characteristics of any class. These categories were in nature and preceded us. Horses belonged to one category, camels to another and the categories were hierarchical; horses and camels are both animals, for instance. One simply had to come up with the necessary and sufficient conditions for the category and everything would fall into place.

This idea was definitively challenged by Ludwig Wittgenstein, who blamed this kind of thinking for many philosophical conundrums.

The idea that in order to get clear about the meaning of a general term one had to find the common element in all its applications has shackled philosophical investigation.

Wittgenstein looked at games and noticed that although we all know what a game is and have no problem using the word game, we are never really able to define it in such a way that all things considered games would be included and all non-games excluded. For instance, games are not all amusing; chess masters are rarely amused. Some involve skill, others luck, some require a ball, others cards, etc. We never identify necessary or sufficient conditions for the category game. Wittgenstein offered a solution in the idea of family resemblance. He noticed that the way we actually use language and categories depends on an overlapping system of common features, which aren't necessarily common to all members of a category. It's much the same way that family members look alike. Some have the same chin or the same widow's peak, his eyes, her cheeks, but nobody shares all the same features. Nevertheless, you still see the family resemblance. Once we understand this, we can stop searching for the elusive definition and start considering common features.

Cognitive research has gone a step further with Prototype Theory. In the 70s Eleanor Rosch noticed that membership in a category is not really equal among its members. Some members are more prototypical of the category than others. Consider furniture. Clearly a sofa and dining room table are furniture, but what about a piano? A stereo? If not a stereo how about a old cabinet Victrola? We all know what furniture is, but when you think about it, there are some things that are more "furnitury" than others. Rosch found ways of testing this empirically with techniques like measuring subject response times to questions of categorical inclusion. She found that we form categories around exemplary members and use something akin to family resemblance to decide if other things belong to the same category. We are quite sure about things that closely resemble the prototype, and become more uncertain as potential members move further from the prototype. Categories for the most part have fuzzy boundaries. This has had a profound effect on the way we understand language. When you think about it, there are very few categories which don't exhibit prototype effects. And it's not just difficult concepts like beauty or goodness; concrete categories exhibit prototype behavior. A car is certainly a vehicle, but what about a skateboard? Ice skates? Although skates are more of a vehicle than, say, goldfish, when I say vehicle, ice skates are probably not the first thing to come to mind. As you move away from the prototype, category boundaries seem to fade. Even categories in science are slippery. We all learned a little taxonomy or at least touched on it school—it's the ultimate categorical scheme with the enormous ambition to categorize all living things. I was taught that it was pretty cut and dry and never suspected it was open for debate. In a very amusing essay, What if anything is a Zebra, Stephen Jay Gould discusses what happens when competing taxonomies conflict. In one type of taxonomy based on the branching order of the family tree, called cladistics, the coelacanth ends up more closely related to elephants than fish. But clearly everything about it, from a morphological point of view, suggests it's a fish.

At this point, many biologists rebel, and rightly I think. The cladogram of trout, lungfish, and elephant is undoubtedly true as an expression of branching order in time. But must classifications be based only on cladistic information? A coelacanth looks like a fish, tastes like a fish, acts like a fish, and therefore—in some legitimate sense beyond hidebound tradition—is a fish.

If professional taxonomists can't tell a fish from an elephant, certainly categorizing and defining art is a lost cause.

Rather than picking up Ctein's gauntlet, a better approach is to accept that art is a category that exhibits prototype behavior. Not all things will fit into the category as well as others. We learn a handful of exemplars from culture—the Mona Lisa, Venus de Milo, a Beethoven Symphony—and extrapolate from there deciding individually how closely an item resembles the prototypes we know. People from different backgrounds will naturally pick up different prototypes, and our prototypes will change over time as we grow and expose ourselves to different traditions. The MFA student learns early on, for instance, that to accept Marcel Duchamp's Fountain as a prototype of art is a necessary and sufficient condition of donning the black turtle neck.