The Philosophy Resource

Teaching resources for philosophy students and teachers

What is art?

Usually when philosophers ask the question ‘what is art?’ they are looking for a certain kind of answer. Namely they are looking for, at least, a small set of features (perhaps even one) shared by all artworks, which can rightly be called the ‘essence’ of art, or what is ‘essential’ or ‘necessary’ to it. That is, features which art HAS TO have, in order to count as such.

Equally, philosophers on the hunt for an answer to the question of ‘what is art?’ tend to want to respect what is sometimes called the ‘Critical Practice Constraint’. Roughly speaking, the Critical Practice Constraint says that any account or definition of what art is must respect the way the term ‘art’ tends to be applied in contemporary critical and appreciative practice – i.e. the way it is applied by artists themselves, critics, and other relevant experts.

Given the wide variety of objects now counted as artworks, philosophers have been pessimistic about finding any manifest (perceptible) properties shared by all. The objects in question seem to be too different in terms of their perceptible properties. Instead philosophers have looked for non-manifest (non-perceptible) relational properties shared by all artworks. Relational properties are properties which an object has in virtue of standing in a certain (as yet unspecified) relation to something else. So for instance: a manifest property of a person is having red hair. A non-manifest relational property of a person is coming from Aberdeen. You can’t tell whether a person comes from Aberdeen by just looking at them (normally).

Broadly speaking, in the 20th and 21st century, philosophers of art have fallen into three camps when trying to answer the question ‘what is art?’ in terms of non-manifest relational properties:

The essence of an artwork is that it has a particular function

Functional accounts of art say, roughly, that art is defined in terms of standing in a relation to a particular function, or restricted set of functions. A functional relation is often a non-manifest property because often you can’t tell what functions an object has by looking at it, or otherwise perceiving it.

An example of a functional account of art would be any theory that says that it is the function of art to produce aesthetic experience. Philosophers haven’t been very happy with this as a definition because it seems very unclear what counts as aesthetic experience. Additionally, they point out, the account can’t accommodate the wide diversity of artworks that have existed, especially in modern times. In modern times, many modern artworks seem to have been made deliberately to be ‘anti-aesthetic’ – either to provide an ugly experience or a boring one (For example Duchamp’s ‘In Advance of a Broken Arm’, as described in Step One) or even no perceptual experience at all (For example Acconci’s ‘Following piece’, as described in Step One).

Challenge: Can you think of a function common to all artworks?

The essence of an artwork is that it stands in relation to a particular procedure

Procedural accounts of art say, roughly, that art is defined in terms of standing in a relation to a particular procedure. Standing in relation to some procedure is often a non-manifest property because often you can’t tell what procedures an object has undergone by looking at it, or otherwise perceiving it. (For instance, in many cases, you can’t tell whether someone has undergone the procedure of getting married, just by looking at them).

An example of a procedural account of art is George Dickie’s Institutional Theory of Art. Roughly speaking, Dickie argues that an artwork is an artefact which has had a particular status conferred upon it – the status of being ‘a candidate for appreciation’. Moreover, this status must have been conferred by some person(s) acting on behalf of an institution: ‘the artworld’.

An analogy: what makes certain bits of paper money? Roughly, those bits of paper have had a certain status conferred upon them – the status of being legal tender – via a certain sort of procedure – being designed, printed and released for use in a certain way – where this procedure is carried out by people legitimately acting on behalf of a certain institution (the Bank of England, in the case of Sterling).

According to Dickie, artworks owe their status as such to an analogous procedure. Roughly, members of the artworld confer upon artifacts the status of ‘candidate for appreciation’ i.e. put them forward as the sort of thing that should be considered as worthy of appreciation (whether or not it actually turns out to be so worthy).

Dickie’s account looks like it copes with the wide diversity of artworks very well. But one worry about Dickie’s definition is: does the artworld have the formal structure to count as an institution, properly speaking, and so to bestow on its members the power of giving objects a genuine ‘status’ (as opposed to just calling them ‘art’ and acting as if they are art)? Dickie’s institutional definition apparently depends on the coherency of thinking of the artworld as a genuine institution, akin in relevant ways to – for instance – the Bank of England. Institutions like the Bank of England are centrally organized; have clearly defined roles, established in a hierarchical relationship to one another; and have agreed formal procedures for acceding to these roles, and for leaving them. The artworld is not centrally organized, contains far fewer defined hierarchies (many of them separate from one another), and the conditions of membership of many of these roles is very vague. What makes an artist? What makes a choreographer? A curator? A director? Etc.

Question: what makes someone a member of the artworld? Are there identifiable conditions of membership?

Art has no essence

A more radical approach is to deny that artworks as a class have even any relational properties in common. To think this makes you an ‘anti-essentialist’ about art. This sort of position is advanced in a well-known article by Morris Weitz, written in the 1950s (chronologically prior to Dickie’s institutional theory). Weitz was heavily influenced by the later Wittgenstein in his Philosophical Investigations, who was skeptical generally of concepts being definable. Weitz argues that art, like other concepts according to Wittgenstein, is a family resemblance concept: that is to say, some artworks have properties in common with certain other artworks, but there are no properties common or necessary to all artworks. According to Weitz, each time the concept of art is extended to cover new cases (for instance, in the case of the avant-garde), it is extended via a practical decision made on the basis of similarity to some other established artworks but not all of them, rather than more automatically by reference to a fixed set of necessary and sufficient conditions which all artworks must fulfil.

Question: say for the sake of argument that Weitz was right and there is nothing common to all artworks. Would it follow that art did not exist?


Go to…

Step One

Go to…

Step Two

Go to…

Step Three

Go to…

Step Four

Go to…

Step Five