Abstract
Non-actualist theories promise straightforward accounts of meaning, truth and reference of fictional discourse but are ostensibly saddled with a Selection Problem, that multiple possible candidates satisfy the role-descriptions associated with names used in fictions and no principled way to distinguish between them; yet if names are referential, there can only be one referent. The problem is often taken to be a serious—even decisive—obstacle for non-actualism, and the aim of this article is to show that the challenge can be met. I suggest that storytellers and authors fix the referents of referential terms they use by arbitrary selecting referents through simple acts of stipulation, and then determine which of the worlds containing the relevant individuals serve as truth-makers for the story by adding properties to the characters. The resulting view of reference is consistent with reasonable, foundational requirements on reference, such as a causal-historical theory (properly understood), and can moreover be used to support plausible theories of truth in fiction and facilitate elegant models of fiction as a type of discourse.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
An alternative to treating sentences asserted in fiction as elliptical is to interpret them as non-elliptical but understand fiction talk as involving a shift in relevant circumstances of evaluation. This would obviate the need for explicit or tacit fiction operators and perhaps make certain concerns about the nature of such operators (e.g. Sainsbury 2014), moot. The choice will not matter here.
Since my aim is to defend a possibilist analysis that give the truth-conditional content of fictional talk, I will not discuss metafictional claims in detail. Examples like ‘Holmes is famous’ are tricky, as they appear to predicate properties Holmes cannot have in worlds in which he doesn’t exist, yet are intuitively true in the actual world. Possibilism is not obviously worse off than its competitors with regard to such examples, however. ‘Holmes is famous’ seems to require paraphrase on any irrealist or Meinongian view (or, on the latter, ambiguity between ascription of the encoded and the exemplified property of fame), and on abstract artifact theories, according to which Holmes is an abstract object, Holmes is not a detective and thus intuitively the wrong target for such ascriptions. Indeed, such ascriptions intuitively require paraphrase, since Holmes isn’t really famous in the way, say, Caesar is. I will leave further discussion aside, however.
The Selection Problem arises for other versions of realism as well, including Meinongian and Platonist theories. See Berto (2012) for a response on behalf of the Meinongian. Berto’s response is not available to non-Meinongian possibilists.
My strategy is thus somewhat reminiscent of Priest (2005), who uses a Meinongian framework in which ‘Holmes’ refers to a non-existent object in the actual world that could have existed in a different world (a move that allows Priest to use a fixed-domain approach to modality). Priest, however, does not attempt to justify the assignment of an individual to ‘Holmes’ in any detail.
There may be ways of fixing reference to certain mere possibilia through causal links (Kaplan 1973). I can perhaps use ‘Tim’ to name the bookshelf that could have been assembled from the IKEA set I purchased but never assembled (and never will). But Tim is a merely possible individual constructed from actually existing materials to which I am properly causally related. Holmes is not.
Divers (2002) describes a technique for constructing definite descriptions that single out particular worlds by relating them to the actual world given some appropriate principle of recombination. The technique is unable to give us Holmes, however.
I am of course not suggesting that Doyle must have started with the name; rather, I claim that when the name—or perhaps the intention to refer—occurs, a selection takes place.
We could of course take the semantic value of ‘Holmes’ itself to be a choice function, along the lines of certain treatments of anaphoric pronouns, and hence insensitive to any particular input from the domain of the function. If we did we would have a version of a descriptivist approach, however.
That restriction does not automatically rule out the possibility that Holmes is identical to some other fictional character. To rule out the possibility that Holmes is Hamlet (in a different world) we need to constrain the selection also by an intention to talk about a new character.
It should be noted that essential relational properties are less problematic. For those sympathetic to essentiality of origin-type properties, Holmes’s parents may automatically be determined upon selecting Holmes. However, that Holmes’s parents get selected by default along with Holmes is not itself a challenge as long as Doyle is free to ascribe any descriptive properties he wanted (including names) to those parents.
Moreover, our model of fiction as a type of discourse allows us to identify the distinction between counterfactual and fictional discourse as a function of restrictions on the context set: Counterfactual discourse is ultimately about the actual world, and counterfactual scenarios take place in worlds accessible from worlds that are candidates for being the actual world (and remove worlds from which these scenarios are not related to the actual world in the right manner from the context set). It is accordingly a rule that the context set for discourse involving counterfactuals contain the actual world.
It follows that the tautology ‘Holmes either has blood type A or he doesn’t’ is true-in-the-fiction since it is true in all the truth-making worlds. The consequence that all tautologies are true in any fiction has been criticized (Proudfoot 2005: p. 11) but the criticism rests on a mistake. There would be a problem if, as Proudfoot appears to assume, true-in-fiction-f distributed over the disjunction—that is, if true-in-fiction (p ∨¬p) entailed true-in-fiction-p ∨ true-in-fiction-¬p—but, as should be clear from the interpretation of in-fiction as a restricted box operator, it doesn’t. Given that the connectives mean the same in fiction as in ordinary speech all tautologies should be true in all fictions.
A similar point is made by Priest (2005, p. 141).
Kripke’s specific target is predicativist theories. I am extrapolating.
Challenges to Breckenridge and Magidor may for instance include what to do when the relevant domain is empty, and perhaps how to define the validity of a universal generalization from Φ(α) to ∀xΦx when Φ(α) is simply a normal singular proposition where α refers to an actual, real individual.
It may, admittedly, be possible to raise a (familiar) general worry for modal realists insofar as the truth-maker for ‘p is necessary’, for instance, is the epistemically inaccessible set of every world.
Note that if we require that Doyle couldn’t have been wrong about Holmes’s non-existence to refer to him, then we cannot extend our account to empty or mythical names, such as ‘Vulcan’, introduced to denote a hypothetical planet between Mercury and the sun. Of course, Vulcan may count as fictional at present, but we need something to say about Le Verrier’s original context, and that context seems relevantly different from Doyle’s insofar as Le Verrier was mistaken about Vulcan’s ontological status. If my account does not extend to empty or mythical names, this is no obvious weakness; as opposed to Doyle, Le Verrier intuitively did fail to select an appropriate referent. It is in any case far from clear that empty names ought to receive the same treatment as names for fictional characters (see e.g. Salmon 1998).
References
Benacerraf P (1965) What numbers could not be. Philos Rev 74(1):47–73
Benacerraf P (1973) Mathematical truth. J Philos 70(19):661–679
Berto F (2012) The Selection Problem. Rev Int Philos 66(4):519–537
Breckenridge W, Magidor O (2012) Arbitrary reference. Philos Stud 158(3):377–400
Caplan B (2016). The extraordinary impossibility of Sherlock Holmes. Res Philos 93(2):335–355
Currie G (1990) The nature of fiction. CUP, Cambridge
Divers J (2002) Possible worlds. Routledge, London
Evans G (1979) Reference and contingency. Monist 62:161–189
Everett A (2005) Against fictional realism. J Philos 102(12):624–649
Hanley R (2004) As good as it gets: Lewis on truth in fiction. Australas J Philos 82(1):112–128
Hawthorne J, Manley D (2012) The reference book. OUP, Oxford
Horwich P (1998) Meaning. OUP, Oxford
Jeshion R (2010) Singular thought: acquaintance, semantic instrumentalism, and cognitivism. In: Jeshion R (ed) New essays on singular thought. OUP, Oxford, pp 105–140
Kaplan D (1968). Quantifying in. Synthese 19(1/2):178–214
Kaplan D (1973) Bob and Carol and Ted and Alice. In: Hintikka J et al (eds) Approaches to natural language. Reidel, Dordrecht, pp 490–518
Kripke S (1980) Naming and necessity. HUP, Cambridge
Kripke S (2011) Vacuous names and fictional entities. In: Philosophical troubles: collected papers, vol 1. OUP, Oxford, pp 52–75
Lewis D (1978) Truth in Fiction. Am Philos quart 15:37–46
Lewis D (1986) On the plurality of worlds. Blackwell, Oxford
Liebesman D (2014) Necessarily, Sherlock Holmes is not a person. Anal Philos 55(3):306–318
Priest G (2005) Towards non-being. The logic and metaphysics of intentionality. Clarendon Press, Oxford
Proudfoot D (2006) Possible worlds semantics and fiction. J Philos Log 35:9–40
Russell B (1910). Knowledge by acquaintance and knowledge by description. Proceedings of the Aristotelian Society, NS 11, 108–28
Sainsbury M (2010) Fiction and fictionalism. Routledge, London
Sainsbury M (2014). Fictional worlds and fiction operators. In M. García-Carpintero, G. Martí (eds) Empty representations: reference and non-existence, Oxford University Press, Oxford, pp 277–289
Salmon N (1998) Nonexistence Noûs 32(3):277–319
Stone J (2010) Harry Potter and the spectre of imprecision. Analysis 70(4):638–644
Thomasson A (1999) Fiction and metaphysics. CUP, Cambridge
Wright B (2014) Many, but almost Holmes?. In M. García-Carpintero, G. Martí (eds) Empty representations: reference and non-existence, Oxford University Press, Oxford, pp 290–306
Acknowledgements
The research behind this article was made possible in part through a grant from the SASPRO/FP7-Marie Curie Actions-COFUND scheme [Grant Number 0086/01/03/-b] as well as VEGA Grant 2/0049/16: Fictionalism in philosophy and science. I am grateful in particular to the Department of Philosophy, Slovak Institute of Sciences, for hosting me during the grant period, and to two anonymous referees, Zsófia Zvolenszky, Martin Vacek, Robin Neiman and audiences at the Proper Names in Fiction workshop organized by the Institute of Philosophy, University of Warsaw, for helpful comments on earlier drafts of this article.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Haraldsen, F. How to Pick Out a Dragon: Fiction and the Selection Problem. Topoi 39, 401–412 (2020). https://doi.org/10.1007/s11245-017-9517-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11245-017-9517-1