Musical Ontology, Musical Reasons

Abstract

IntroductionSince the publication in 1968 of Nelson Goodman's Languages of Art, many philosophers with an interest in musical ontology have been concerned to discover, or to elaborate, identity conditions for musical works.1 They were spurred to this in the first place by a desire to improve upon Goodman's own specification of such conditions, which convinced no one;2 and they have kept at it in the meantime out of dissatisfaction with one another's solutions. I cannot pretend to try to settle the matter here.3 But I believe that I can show that the range of potentially plausible solutions is circumscribed in a way that has not been recognized. Specifically, I think I can show that any account that fails to acknowledge, or lacks the room to acknowledge, the role of musical reasons in understanding the issues that Goodman's problem raises cannot hope to offer a satisfactory solution to it - where, included among such accounts, is one that has recently been said to represent the intuitive 'default position'.4The fundamental question has been taken to be: what are the identity conditions of a musical work? Or: what conditions must something satisfy if it is to count as a token of the type that is to be identified with a given musical work? In practice, this has meant that philosophers have attempted to unearth the conditions that a sound-event (typically a performance) must satisfy if it is to count as an instance (typically a performance) of such-and-such a work, where these conditions establish the identity ofthat work. The results of such enquiry have the following shape: a set of properties, Σ, is posited as normative with respect to a given type, so that - here I am following Julian Dodd's way of putting the matter - 'for something to be a properly formed token' of that type, 'it must possess each member' of the relevant set of properties; '[t]o be any sort of token' of that type, even if only a 'malformed' one, 'a sound-event must possess enough of Σ's members'.5 Philosophical disagreement is, then, disagreement about what is to be included in S. Some philosophers have included only temporally ordered patterns of relations between pitched sound (a position sometimes termed 'sonicism')6; others have also included the timbrai properties of those pitched sounds ('timbrai sonicism')7; others have added the means by which such timbrai properties are produced ('instrumentalism'); while still others have added considerations concerning musico-historical context and the relations holding between that context and a given sound-event ('contextualism').8 And there are further possibilities, too.My eventual conclusion is that, of these possibilities, only some version of contextualism can conceivably be right. Before moving on to the argument for that, however, it will be helpful to raise a preliminary issue. What, we might ask, is it for this or that sound-event to count, not as a token of this or that musical work, but as music at all!1. On Being Music1.1One reason for asking this question might seem to be that the category 'music' clearly comprises more than just (the totality of) musical works. So, for example, and as many have claimed, it may be plausible to think that the concept 'musical work' is not much more than a couple of hundred years old, and so that music composed prior to about 1800 probably shouldn't be thought of in its terms.9 Such music is, however, still music. It is also uncontroversial to say that there are sound-events of a type being produced for the first time now that are music but are not works or performances of works: free improvisation, for instance, belongs to this category.It is sometimes suggested that the upshot of these points, and of points like them, is that, in focussing on the ontology of musical works, specifically, philosophers have misrepresented or misunderstood the wider category - music - to which such works belong, and so have, in some sense, taken themselves off up a blind alley. …