Reconsidering the Artworld's Role in Art's Permanence — Epoche B2
Two Questions About a Symphony A worry is often raised against George Dickie's institutional theory of art [1] . If a thing is art because the artworld has conferred that status on it, then works with no physical body — a symphony [2] , a novel, a performance piece — seem to hang by a thread. Their art status came from a social act, they have no object to fall back on, and social acts can be withdrawn. The worry is usually met by saying that the artworld keeps conferring: that exhibition, criticism and re-performance sustain the status continuously rather than once. That answer is worthless, because the worry it answers is confused. It runs together two questions which have different subject matters and different answers: what makes a thing art, and what kind of thing a given work is. The institutional theory is an answer to the first and says nothing about the second. This note argues that the separation is not a technicality but a constraint on any theory of art status, and that the ontology of non-physical works has to be settled by something else [3] . The claim is deniable — Amie Thomasson denies it [4] , and her reasons are the strongest thing in the vicinity — and the last two sections take her objection seriously. The theory, stated Dickie's definition, in Art and the Aesthetic (1974), is that a work of art in the classificatory sense is an artefact, a set of the aspects of which has had conferred upon it the status of candidate for appreciation by some person or persons acting on behalf of a certain social institution, the artworld. Two things follow immediately. Art status is relational: nothing in an object settles it. And art status is conferred on something: there is an artefact, and a status is bestowed upon it. The second point is the one that matters here, and it is not an incidental feature of Dickie's wording. Conferral is a relation, and relations need relata. When a committee confers a degree, there is a person who receives it; when a gallery presents a urinal, there is a urinal. The urinal's existence, its persistence and its identity over time are settled entirely without reference to the gallery. This is obvious for physical objects and it is exactly as true for works with no body — but there, because the object is unfamiliar, the obviousness disappears and the confusion becomes tempting. Where the questions come apart Take a specific case. Beethoven's Fifth Symphony is not the manuscript, since the manuscript could burn and the symphony survive. It is not any performance, since it has had thousands and would exist if it had none. It is not the set of all its performances, since it would still be the same work if one more were given tomorrow. Whatever answer we give, notice that the question has not mentioned art status once. A piece of music that no artworld ever recognised — a hymn from a tradition no institution has heard of — raises exactly the same problem of identity, and the answer we give for the symphony must apply to it. Conversely, a work whose art status is beyond dispute can have thoroughly disputed identity conditions: musicologists argue about whether a given version of a Bruckner symphony is the same work as another while nobody argues about whether either is art. So the two questions vary independently, and a theory that answers one has, so far, said nothing about the other. Making the artworld's engagement continuous does not change this. Continuous conferral is still conferral, and it still needs something to be conferred upon. What does settle the second question The serious candidates locate non-physical works among types or kinds rather than particulars. On the view defended by Nicholas Wolterstorff in Works and Worlds of Art (1980) and by Peter Kivy and Julian Dodd since [5] , a musical work is a sound structure [6] : an abstract pattern that exists timelessly and is discovered rather than made. Composing is finding, as a mathematician finds a proof. This explains repeatability at once — a performance is a token of the type — and it explains why the work survives the loss of every score. The cost is that nothing is created. Jerrold Levinson's alternative, in 'What a Musical Work Is' (1980), keeps creation. A work is not the bare sound structure but that structure as indicated by a particular person at a particular time in a particular musical context. Two composers who wrote down the same notes two centuries apart would then have made two works, not one, because the indicating differs, and the later one would be a different piece with different properties: derivative where the first was original, conventional where the first was daring. Both are answers to the ontological question and neither mentions an institution. Levinson's does mention a context, and it is worth being clear about the difference. Levinson's context supplies the work's identity by fixing what the composer did; it is not a body of people who approve. The same is true of Noël Carroll's proposal that art status is secured by an identifying narrative connecting a work to earlier practice [7] : a narrative is something one gives and defends, and it can be a bad one, which is precisely what conferral by an institution cannot be. A case where lineage does the work The Korean solo instrumental genre sanjo makes the ontological question concrete in a way that Western examples do not. A sanjo is a long improvisatory piece for gayageum, geomungo or daegeum with drum accompaniment, moving through a fixed sequence of rhythmic cycles from the slow jinyangjo to the fast jajinmori . The genre took shape in the late nineteenth century and is traced to Kim Chang-jo; it is transmitted in named schools, so that one speaks not of a piece but of a lineage, such as the gayageum sanjo of Kim Juk-pa. The interesting feature is how identity is fixed. Two lineages may share the modal material and the whole sequence of rhythmic cycles and still count as different pieces, while within one lineage two