The Concept of Art

Is there a universal way to define the word “art”?

If by “universal” you mean a way to define art that captures everything that is a work of art and excludes everything that is not a work of art, as opposed to a definition of art that is universally accepted, the answer is yes.

At least, there are ways of trying to define art universally, although none has been conspicuously successful or satisfying to all. However, that doesn’t mean we can’t learn something by making the attempt and by reflecting on attempts made. There was quite a lot of that in the last half of the twentieth century, though in the last couple of decades the focus has shifted to artistic value – why art matters – and issues involving aesthetic properties. Beauty, especially, has gotten a good deal of (in my view) needed attention.

There are other ways to form a conception of something than by defining it, especially if you think of a definition as a statement of individually necessary and jointly sufficient conditions. In the last century, however, George Dickie seemed to have found one. He said, roughly speaking, that something counted as an artwork just in case it was offered as a candidate for appreciation by someone institutionally authorized to do so. The institution in question was usually thought of as the “art world,” a rather loose but not utterly incoherent arrangement of practices, beliefs, values, authorities and other things we associate with an ongoing and purposive social activity.

Another definition was formulated by Arthur Danto, who thought an artwork must at the very least be about something, even if only itself. Being about something implies the expression of an idea or attitude towards something, and an artwork does this in a manner that is uniquely tied to its form. An artwork expresses an attitude by taking a form.

If these definitions seem rather broad, it’s partly because their creators were trying to understand how something like this could be a work of art.

This is Marcel Duchamp’s Bottle Dryer, an unaltered manufactured item no different from thousands of others. Its creator designated it as a work of art in 1917.

It was also necessary to explain how something like this could be a work of art.

This work by Dan Flavin consists of an arrangement of ordinary commercially produced fluorescent lights, used by the artist just as they came out of the box and not enhanced or otherwise modified in any way.

A possible limit-case of this kind of thing might be Robert Barry’s All of the Things I Know (1969).

Here we have a work of art that never has been and never will be experienced by anyone – not even its creator. At least, that’s the case if we take seriously Barry’s claim that the work of art itself really is the totality of that portion of his knowledge that he was not reflecting on for the duration of the instant of time that occurred at exactly 1:36pm on June 15, 1969.

As I said, a definition that states necessary and sufficient conditions isn’t the only way to arrive at a concept of art. There’s another broad kind of concept which is variously understood as a descriptive, cluster, or family resemblance concept. The kind of definition Dickie composed looks something like this:

(A ≡ X)

Here, an artwork would be an instance of a closed, or exclusive, single determinant or set of conjunctive determinants. “Art if and only if X.”

(A ≡ df X)

Here, an artwork would instantiate a disjunctive set of determinants, but this set is also closed. “Art if and only if d or f.”

(A ⊃ d + e + f + g + [?] X)

Here, an artwork would instantiate an open set of disjunctive determinants. “Art if d or e or f or g [or….].”

In this kind of family resemblance concept, there is a range of properties that can instantiate the concept, but new properties can be introduced into the set of possible properties and properties that had been present can stop being instantiated. In the practice of art, according to some, the criterion for inclusion is sufficient resemblance to what’s already included.

If the art practice is like this, then a work can appear that resembles others only in ways that had up to then been regarded as as irrelevant, and still be admitted to the practice. This approach can accommodate a theory of how Duchamp’s kitchen appliance, Flavin’s office equipment, and Barry’s untapped knowledge can count as art.

The key is to think of the history of art as a process that retains its identity by changing, much like a river. A river is flowing water: when you step into one, you step into different water than you would have if you’d stepped in earlier or later. But it’s the same river, because a river is flowing water.

When we think about change, we typically define what changes against an element that doesn’t change. Although a river is flowing water, the river-bed doesn’t change. If the practice of art were like that, it would be a process of change, but an orderly one. But of course the practice of art isn’t like that. It’s much less predictable.

What if the fixed element does change, only more slowly than what we’ve designated as the changing element? The river-bed over which the water flows does erode, only very slowly. The unchanging/ changing distinction would then depend on the area of our frame of reference, and nothing in the process would have to be regarded as permanently or necessarily fixed. Identity or sameness would be change that we cannot perceive or choose to ignore.

On the other hand, identity will be preserved: without at least a relatively enduring constant, there can be no practice. Admissions decisions must be made about every applicant, and the degree of scrutiny will be high for unusual candidates, at least if the practice is healthy. A case for the newcomer has to be made, and the bona fide members of the practice must be persuaded. So although the art practice maintains its identity by changing as practitioners come and go, the changes aren’t arbitrary.

The Painter’s Studio (1855) by Gustav Courbet.
Nathan Louis Finckelstein, Factory panorama with Andy [Warhol], 1966.

Leave a Reply

Your email address will not be published.