Fictional characters and literary practices

British Journal of Aesthetics

Volumn 43, Issue 2, 2003, Pages 138-157

  Thomasson, Amie L ^a  

^a Department of Biology   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 57049083210     PISSN: 00070904     EISSN: None     Source Type: Journal    
DOI: 10.1093/bjaesthetics/43.2.138     Document Type: Article

References (36)

1. Terence Parsons, Nonexistent Objects (New Haven: Yale U.P., 1980)
2. Nicholas Wolterstorff, Works and Worlds of Art (Oxford: Clarendon Press, 1980).
3. Edward Zalta, Abstract Objects (Dordrecht: Reidel, 1983).
4. Robert Howell notes the creation problem for Meinongian and possibilist accounts, in 'Fictional Objects: How They Are and How They Aren't', Poetics, vol. 8 (1979), pp. 133 and 139.
5. A. L. Thomasson, Fiction and Metaphysics (Cambridge: Cambridge U.P., 1999).
6. See Parsons, Nonexistent Objects, p. 188.
7. See Wolterstorff, Works and Worlds of Art, pp. 144-149.
8. Cf. Thomasson, Fiction and Metaphysics, pp. 111-114.
9. K. Donnellan, 'Speaking of Nothing', Philosophical Review, vol. 83 (1974), no. 1, pp. 3-31.
10. This is a different analysis of negative existentials involving fictional names from the one in Fiction and Metaphysics, pp. 112-113. There 1 suggest that they are true provided we understand them as engaging in restricted quantification (for example, over people, not fictional characters)-as saying, for example, 'There is no such person as I Holmes.' I have since come to think that a better way of handling these is to follow Donnellan as above.
11. This view hears some resemblance to Charles Crittenden's view in Unreality: The Metaphysics of Fictional Objects (Ithaca, NY: Cornell U.P., 1991) that a fictional character is a mere 'grammatical object' which 'is just what corresponds to accepted linguistic practice' (p. 97). Nonetheless, there are also important differences. Crittenden treats critical practices as constitutive even of the particular properties of individual fictional characters, and treats the characters spoken of as having 'no sort of reality whatever' or no 'metaphysical status'. I argue for a thesis at one level of remove: that literary practices are constitutive of general ontological features of the nature of fictional characters as a kind, from which it does not follow that any character has a given property just in case the corresponding property-attributing expression is (considered to he) appropriately applied. Nor do I accept that, because their ontological nature is determined by our practices, fictional characters have no reality or metaphysical status-although T think these practices determine what their metaphysical status is.
12. Gregory Currie, in The Nature of Fiction (Cambridge: Cambridge U.P., 1990), p. 127 seems to embrace the view that we are simply wrong in thinking that fictional characters can appear in multiple works of literature.
13. Cf. Bcrtrand Russell, Tile Problems of Philosophy (Oxford: Oxford U.P., 1959), pp. 25-26.
14. Michael Devitt and Kim Sterelny, Language and Reality, 2nd edn (Cambridge, MA: MIT Press, 1999), p. 91, similarly note that even in grounding a natural kind term such as 'human', 'there must be something about the grounding situation that makes it the case that it is a grounding of a natural kind term and not talk about, say, an artifact. It seems that something about the mental state of the grounder must determine which putative nature of the sample is the one relevant to the grounding.'.
15. In fact this is a general form of argument applying (at a minimum) to any kinds of entities where, to establish reference to that kind (rather than to an underlying natural kind), an ontological concept is necessary. Thus, it seems also to apply to other kinds of artefacts and works of art, whether or not they are abstract (see my 'The Ontology of Art', in Peter Kivy fed.], The Blackiwell Guide to Aesthetics [Oxford: Blackwell, forthcoming] for further discussion regarding kinds of works of art).
16. The idea that fiction is a public, rule-governed practice (much as games like baseball are) is developed and defended in Peter Lamarque and Stein I laugom Olsen, Truth, Fiction, and Literature (Oxford: Clarendon Press, 1994), pp. 29-52.
17. Of course, as with any case of reference-grounding, these need not be just those people who initially introduce the term-the term may be 'grounded just as effectively by subsequent groundings' (Devitt and Sterelny, Language and Reality, p. 89), and if the use gradually changes over time, so may the reference.
18. It follows from this that, where there are greatly differing critical practices, different corresponding ontologies of fiction may be presupposed by the practices and appropriately drawn out by philosophers. This suggests one sense in which character identity may indeed be 'interest-relative' rather than absolute (as Peter Lamarque argues in 'Objects of Interpretation', Metaphilosophy, vol. 31 [2000J, nos. 1-2, pp. 120-121
19. How to Create a Fictional Character. The Creation of Art 2003, Paisley Livingston and Berys Gaut eds Cambridge: Cambridge U.P.
20. and 'How to Create a Fictional Character', in Paisley Livingston and Berys Gaut (eds), The Creation of Art [Cambridge: Cambridge U.P., 2003].
21. and as Lamarque and Olsen argue in Truth, Fiction and Literature, p. 132). It may also provide a way to explain why, as Robert Howell puts it ('Ontology and the Nature of the Literary Work', Journal of Aesthetics and Art Criticism, vol. 60 [2002], no. 1, pp. 67-79) there may be no single ontological kind for literature (or other kinds of work of art). For where different interests form the basis for different critical practices, or where there are different conceptions of literature (for example, among those in our culture, versus those with oral traditions that do not care about fixed texts) there will be correspondingly different ontologies for the objects discussed in those cultures.
22. Or, following Currie's Gricean analysis of fiction-writing (The Nature of Fiction, p. 30) pretending to refer to a person, and intending readers to make-believe that there is such a person by recognizing the author's intention that they so make-believe.
23. Cf. Kendall Walton, Mimesis as Make-Believe (Cambridge, MA: 1 Iarvard U.P., 1990), pp. 399-400.
24. S. Schiffer, 'Language-created Language-independent Entities', Philosophical Topics, vol. 24 (1996), no. 1, p. 156.
25. John Searle, Expression and Meaning: Studies in the Theory of Speech Acts (Cambridge: Cambridge U.P., 1979), pp. 71-2.
26. and Saul Kripke, The John Locke Lectures (unpublished, 1973), Lecture 3.
27. Or, on Walton's theory (Mimesis as Make-Believe, p. 410), the first may be analysed straightforwardly, while the second brings us into a make-believe context, invoking the ad hoc game of make-believe that 'to author a fiction about people and things of certain kinds is fictionally to create such'. What Walton calls an 'unofficial game' of make-believe here, I think is in fact a language game (not involving make-believe) that is used in such a way that the conditions sufficient for the truth of the former are guaranteed to be sufficient for the truth of the latter.
28. This does not, however, mean that fictional characters are 'minimal' in the sense of having some sort of second-class or deflated ontological status, as claims that certain entities are minimal or pleonastic are sometimes used to imply. See my 'Ontological Minimalism', American Philosophical Quarterly, vol. 38 (2001), no. 4, pp. 319-331, for critical discussion of claims of this sort.
29. Quine, From a Logical Point of View (Cambridge, MA: Harvard U.P., 1953), p. 13. Quine's original claim is motivated primarily by examining Russell's treatment of negative existentials, bringing us to the conclusion that the use of names does not commit one to the existence of something named (since the truth of a sentence like 'Pegasus doesn't exist' does not require us to posit Pegasus)-a point that I am not contesting. But from this it does not follow that, for all grammatical categories, 'to be is to be in the range of reference of a pronoun' (ibid.)-so that accepting, for example, that some dogs are white does not commit us to the existence of the properties of doghood or whiteness. Nor does it follow that, if we have a statement that appears to commit us to entities of a certain kind K, if it has the same meaning as a paraphrase that does not involve quantifying over Ks, we need not accept that there are Ks. If the two really have the same meaning, then the apparently less committing paraphrase can be transformed back into the original committing sentence, and as long as these connections of meaning are preserved, we have not thereby avoided commitment to any entities.
30. Schiffer, 'A Paradox of Meaning', Noûs, vol. 28 (1994), no. 3, pp. 304-305.
31. P. van Inwagen, Material Beings (Ithaca, NY: Cornell U.P., 1990), pp. 101-102 and 108-114.
32. Thomasson, 'Ontological Minimalism', pp. 327-330.
33. K. Fine, 'Vagueness, Truth and Logic'.
34. reprinted ill Rosanna Keefe and Peter Smith (eds), Vagueness: A Reader (Cambridge, MA: MIT Press, 1997), p. 120.
35. Lamarque ('Objects of Interpretation', p. 120) similarly notes that character identity may be indeterminate in certain cases.
36. An earlier version of this paper was presented at the workshop 'Do Ficta follow Fiction?' in Vercelli, Italy, June 2002. My thanks go out to all those present in the discussion, and to my commentator, Marco Nani, for helpful comments and questions, which certainly improved the final product. Portions of this paper also developed out of my 'Reply to Critics' in the American Society for Aesthetics session in October 2001, and I wish to express my thanks to Robert I lowell and Peter Lamarque for their detailed and incisive commentaries that spurred me to rethink many of the issues discussed here.