A is for Art

An A-Z of Arts Policy has to begin somewhere...

”Art”, as W.B. Gallie explained, like “Democracy”, “Freedom”, or “Equality” (in the social, not mathematical sense), is an “essentially contested concept.” We might think that we recognize art when we see it, but we’ve never succeeded in coming up with a definition of art that everyone could agree to, such that when we came across any object or performance there would be consensus on whether it was, or was not, art. It doesn’t mean we cannot try to solve the question of what art “really” is, but it does mean that it is a philosophical problem, not a problem of classification that has simply eluded us up to now but will one day be sorted out.

This does not mean the discussion is fruitless: when we disagree about what we ought to mean when we refer to an artifact as “art” (or refer to a political arrangement as being consistent with “freedom”, or “democracy”), we learn, come to understand that it really is a hard question, and, if we are open to it, see some of the pitfalls in the definition we have been used to working with. In the end I might not be fully persuaded by George Dickie’s “institutional” definition of art, but I am glad to have read it.

---

I taught arts policy for more than twenty years, retiring from the trade last spring. I was trained as an economist, specializing in tax policy and labor economics, and only came to the economics of the arts later in my career. But you can’t be just an economist to do cultural policy: for all its insights, economics is a very narrow - too narrow - lens through which to think about arts policy. And so the arts policy prof needs to pick up other social sciences, some theory of public management and policy formation, art history, and the philosophy of art. It is impossible to do all this without being something of a dilletante; I will cop to the charge.

So I come to the ongoing debate over how to define art as a consumer, rather than a producer. I don’t have the chops to contribute a new perspective. But I do think it is an important question for arts policy: if you are going to have public support for the arts, it is necessary to be able to say what you mean by art. “We fund this sort of thing, but not that sort of thing.” And when asked why, the response has to rely on some idea of why “art” is deserving of public funds, and ought not to be left to whatever markets will produce, but things that are “not art” can make their own way.

---

Collingwood published his Principles in 1938, and he does not rate a mention in the Stanford Encyclopedia entry on the definition of art (although, nothing written in the first half of the twentieth century is cited - was it that barren of ideas?). But for clarifying my own thinking, and as something which makes a kind of sense to me, and which retains its relevance (and provided a very nice prompt for classroom discussion), his position is worth a defense. And so here is a capsule summary:

Collingwood distinguishes two kinds of quasi-art - “magic” and “amusement” - from art-proper.

What does he mean by “magic”? These are performances that adopt what we would think of as genres of art - dancing, singing, painting - but where the purpose of the performance is not artistic expression but something else: in contemporary language, we would say art which has an instrumental purpose. Consider Liverpool fans singing “You’ll Never Walk Alone” before a match. It’s music, sure. But we would not think to evaluate it in the same terms with which we would evaluate a choral recital, or to suggest the fans might try another piece to sing instead. It is ritual, and it doesn’t matter whether the voices are professional-choir quality, nor what Rodgers and Hammerstein were trying to express in this composition. The singing of “The Star-Spangled Banner” before US sporting events is a similar thing - the words of the song don’t matter (I’m not sure how many Americans could pass a quiz on the War of 1812), it has nothing to do with sports; what matters is that this is a thing we do, as a collective. “You’ll Never Walk Alone” is meant to inspire solidarity amongst fans, and to inspire the Reds on the field. “The Star-Spangled Banner” is meant to promote a shared patriotism and respect for the armed forces and its veterans. (This is why some people believe there ought to be a law about how we go about this, which is not something people propose for performances of other songs).

The change of spirit which divides Renaissance and modern art from that of the Middle Ages consists in the fact that medieval art was frankly and definitely magical, while Renaissance and modern art was not.

Collingwood mentions the early poetry of Kipling, designed to inspire pride in Empire, hymns sung in a church service, and, with sadness, the arts of rural areas, then being lost to urbanization and industrialization.

[T]here is the native art of the poor: in particular, that rustic or peasant art which goes by the patronizing name of folk-art. This folk-art, consisting of songs and dances and stories and dramas which in this country [England] (with its tradition of a patronizing contempt for the poor) were allowed to perish almost completely before ‘educated’ persons had become aware of their existence, was largely magical in its origin and motive. It was the magical art of an agricultural people.

Not all “magic” art is bad.

Writing in 1938, he notes that political propaganda as quasi-art is on the rise…

… Orwell sounded a similar warning about writers having their true art corrupted through trying to advance political party goals. Collingwood could not have foreseen how much in our day this sort of “magic” art is considered by so many, even in the arts policy world, as the most valuable art - art that “raises awareness”, that “brings us together”: in how many rural towns (which I will get to in a later post) is the first idea when funding comes available to make an uplifting mural?

---

The second kind of quasi-art, and this leads to more controversy these days, is what he calls “amusement”, or what we might call “entertainment.” Like “magic”, it also has an instrumental purpose. But while the emotions stirred by magic art are meant to last, to go beyond the moment of its performance, “amusement” is all about the moment, stirring emotions in the audience - laughter, suspense, horror, romance, awe at a technical achievement - but leaving no trace. The creator is not interested in expressing anything as an artist - the attention is all upon achieving the desired effect: a comedy that makes you laugh rather than cringe, a thriller that thrills, instead of just making you smirk. But there is nothing else there.

It is not about genre, but about intent: there are some films, songs, visual arts, that are nothing more than amusements, and some where the artist is actually using the medium of her art to express something to the viewer. The discussion generated a few years ago by Martin Scorcese making just this point about movies shows how the issue is a live one.

Collingwood was writing at the time of an explosion of entertainment through Hollywood movies and the radio, and he was not alone (see various issues of F.R. Leavis’s journal Scrutiny, for example) in worrying about amusements swamping both “magical” art and art-proper. I’ll come back to this…

---

Well, what is “art” then?

“If I could tell you what it meant, there would be no point in dancing it.” Isadora Duncan.

For Collingwood, art is the expression of emotion by an artist through some medium. It is not a description of an emotion - “I was just so angry!” - but an expression of it, and one that does not lend itself to simple, ordinary language.

The expression of emotion, simply as expression, is not addressed to any particular audience. It is addressed primarily to the speaker himself, and secondarily to anyone who can understand. … The speaker’s attitude towards his audience is quite unlike that of a person desiring to arouse in his audience a certain emotion. If that is what he wishes to do, he must know the audience he is addressing. He must know what type of stimulus will produce the desired reaction of a particular sort; and he must adapt his language to his audience in the sense of making sure that it contains stimuli appropriate to their peculiarities. If what he wishes to do is express his emotions intelligibly, he has to express them in such a way as to be intelligible to himself; his audience is then in the position of persons who overhear him doing this.

A person who sets out to make an entertaining amusement is concerned with what he thinks his audience will respond to - what they will think is funny, or exciting - and tries to craft something that will achieve this. An artist has a different motive, an attempt to express an emotion in a way intelligible to herself.

---

Difficulties:

If the difference between art and amusement comes down to the aims of the person who created it, there will be cases where reasonable people can disagree as to which side of the line a particular work falls, since we aren’t always able to know for certain what the intent was. After all, great artists - Shakespeare, Mozart, Hardy - hoped their audiences would be entertained by their works … but they were also looking for something more.

Furthermore, there can be good entertainments - great pop songs, funny comedies, well-crafted mysteries - and bad art, where the artist’s emotions are not particularly deeply examined, or where the artist lacks the technical skills to achieve in a score or on canvas what they are wanting to convey, or both. Collingwood is deeply worried about what he saw (correctly, I think) as a coming dominance of amusements, cheaply and widely available, and the decline (or, more accurately, corruption) of both “magic” art and art-proper. I, most people, enjoy what Collingwood calls amusements: favorite songs and movies, even silly ones.

But what I take from him, in 2025, is this: while we might enjoy our entertainments, there is a distinction between them and art. You can think Taylor Swift is a talented entertainer - she is - but there’s a conceptual difference between “We Are Never Ever Getting Back Together” and Strauss’s “Four Last Songs”, in the same way there is a conceptual difference between “Deadpool & Wolverine” and “Persona”. The people who created these different works had completely different intent.

---

I don’t much write about AI, but here is a parting thought. AI can create amusements. It can be programmed to create “rainy Sunday afternoon light jazz” for people seeking the emotional state of “listening to rainy Sunday afternoon light jazz”, or can be programmed to create a print of a landscape whose colors really go nicely with the wallpaper and couch. It can create a locked-room murder mystery for people who like to read that sort of thing. If asked “what would be funny to put on a museum wall?” it might answer “A banana would be lol”.

But by Collingwood’s definition, AI simply cannot create art. The emotions the artist wishes to convey cannot be programmed because they are beyond the capabilities of even the most experienced writer of code.

Will AI put some entertainment jobs at risk? Yes, it already has. Will AI replace artists? Only if we - listeners, viewers - give up on the idea of art.