Meta-ontology and Meta-fiction

META-ONTOLOGY AND META-FICTION by Denis E. B. Pollard Peter van inwagen's attempt to explain the nature of fiction makes use of Quine's program in meta-ontology.1 This program comprises four basic theses: (i) that being is the same as existence, (ii) that being is univocal, (iii) that this univocal sense is best captured, for the purposes of formalization, by die existential quantifier, and (iv) that deciding what to believe about what there is comes to a decision about what theories we should accept. As the aim of formalization is explicitness, then theories should be rendered into canonicalese and their commitments identified through those sentences which are prefaced with the existential quantifier. For Inwagen, as for his mentor Quine, acceptance of theory expressed in this canonical form entails acceptance of the ontology expressed in those quantified sentences. Bringing these considerations to bear on fiction, Inwagen asks, "What are theories about fiction?" and replies that they are "theories that treat stories as having an internal structure" (p. 72). The upshot of this is that such theories contain sentences which, when formally rendered, revead commitments to fictional objects. It is further claimed that such sentences can be vehicles of objective truth. Thus Inwagen finds he cannot wholeheartedly dismiss Sarah Gamp as a nonentity. In this respect, his view contrasts with two contemporary accounts, the "reductivist" one which denies Sarah Gamp's existence and her possession of any properties whatever, and the more recent "Meinongian" view which would agree that Sarah Gamp does not exist, but would nonetheless strenuously maintain that she is a nurse, fat, old, and decidedly the worse for drink. Inwagen's Sarah is neither ofthese: pace both these positions, she exists all right, but with properties such as being a theoretical entity of criticism, being a satiric villainess, not being a woman, and not being fond of gin. Inwagen's attempt to steer what seems to be a middle course has yielded, if anything, a more counterintuitive result than either ofthe other views. What is die explanation for this? 244 Denis E. B. Pollard245 To begin with, it is important to note a distinction, which Inwagen himself accepts, between two kinds of sentences. Let us take the following elementary examples: (1) "Sarah Gamp is a nurse," and (2) "Sarah Gamp is a character in a nineteenth-century novel." With specimens like (1), Inwagen sides with the reductivist: such sentences asefalse, and are not about anything. But it is with examples like (2) that ontology becomes a real issue, since sentences of this kind belong to literary theory. Neither authors nor critics who make use of sentences of the first kind are committed to the existence of fictional entities. So far, so good. But what is it that compels such commitment in the second case? Inwagen is sceptical about reductivist analyses which seek to paraphrase such sentences in order to remove any "misleading" implications. Such paraphrases he would regard as at best "messy," and more seriously, unlikely to preserve intuitive meaning without appearing to say that there are fictional characters. Unfortunately , he brushes this aside as a "narrow technical issue" at the very point where rather more argument is needed. However, on the issue of preservation of intuitive meaning, it might well be argued that his contention is somewhat question-begging, suggesting a rather literal reading of (2) above as a straightforward predication superficially akin to (3) "McEnroe is a seeded player in this year's Wimbledon tournament." So by a strange alchemy, the expression "Sarah Gamp" in (2) is accorded a referential status denied to it in (1). It is at this point that the notion of"theoretical entity" is introduced to supply the need for a referent dictated by the Quinean program. Inwagen's Sarah Gamp, in this theoretical disguise, can now literally satisfy predicates like "... is a character in a nineteenth-century novel," but not predicates like "... is a nurse." The parallel being exploited here is that with scientific theory. An ideal gas can satisfy Boyle's Law, but cannot literally be ignited, put into containers, or have an offensive smell. But there is more than a suspicion of equivocation over the term...