- Modal Fictionalism1 and the unusual problems it faces have been widely discussed of late. Realizing the scope for fictionalism2 about a variety of domains, many have wondered whether recent insights into the problems for modal fictionalism3 have any further implications. We will argue that one of the main problems facing modal fictionalism4 is merely a special case of a general problem that faces a genus of fictionalisms5 of which modal fictionalism6 is merely a species. We will suggest that one particular diagnosis of this problem becomes more obviously attractive when viewed in this broader context, arguing that not only modal fictionalism7, but other interesting fictionalisms8 (including fictionalism9 about mathematics and fictionalism10 about universals) cannot provide the ultimate grounding for the areas of discourse in question.
- To introduce this class, we will begin by recapping the 'story so far' in the case of modal fictionalism11. The form of modal fictionalism12 presented by "Rosen (Gideon) - Modal Fictionalism" (1990) will be our focus for simplicity. It is the thesis that while claims made using modal operators are often literally true (for instance, the claim 'possibly swans are blue' is perfectly in order), it nevertheless is false that there are any non-actual possible worlds, concrete or otherwise, and so the claim 'there is a possible world where swans are blue' is literally false. However, since talk of possible worlds is at the very least quite useful when discussing modal issues, and perhaps even indispensable, the modal fictionalist wishes to rescue the usual 'Possible Worlds translations' of modal claims, and does so by translating claims involving modal operators into claims about what is true according to a certain fiction about possible worlds. Talk of possible worlds is then useful in talking about modality because there is a procedure for moving back and forth from the fiction about possible worlds to workaday claims involving modal operators. According to Rosen's fictionalist, the connection between 'possibly swans are blue' and 'there is a possible world where swans are blue' is that possibly swans are blue iff, according to the modal fiction, there is a world containing blue swans. In general
Possibly p iff according to the modal fiction, at some world p, and
Necessarily p iff according to the modal fiction, at all worlds p.
- Using the device of a story about possible worlds and the above 'translation procedure', Rosen claims that we may be able to have the benefits of
The modal fiction Rosen chooses to rely on is very nearly the one presented as non-fiction by David Lewis (1986) - with a few minor alterations.
- having ordinary modal talk come out as literally true most of the time, and
- having recourse to the elegance of a possible worlds framework for modal thinking, but without the ontological cost of non-actual concrete possible worlds and possibilia.
- A serious problem has been identified for this approach in recent literature ("Rosen (Gideon) - A Problem for Fictionalism about Possible Worlds", 1993, "Brock (Stuart) - Modal Fictionalism: A Response to Rosen", 1993), which can be illustrated by considering the following claim:
Treat the scope of the implicit quantifier in (1) as quantifying over absolutely everything - (1) is to be read in the widest possible way. The modal fictionalist does not want to be forced to endorse (1), since at best the fictionalist will wish to remain agnostic about whether or not there are many worlds, and at worst the fictionalist will want to reject outright the claim that many worlds exist, understood literally.
- Necessarily, there exist many worlds.
- Should it be necessarily true by the fictionalist's own lights that there are many worlds, the fictionalist will seemingly be led to embrace modal realism after all, since necessarily p entails p. Unfortunately, it is hard for the fictionalist to resist such a conclusion. Consider the fictionalist's translation scheme. (1) will be true iff
and Lewis's fiction does say that the modal universe contains many existing worlds, and the 'at all worlds' is just a redundant quantifier when 'there exist many worlds' is taken to have a very wide scope. 'At all worlds, there exist many worlds', when interpreted in the intended way, is true according to "Lewis (David) - On the Plurality of Worlds", 1986, so it is true according to the modal fiction. And since (2) is true iff (1) is, by the fictionalist's lights, then (1) must be true by the fictionalist's lights. That there are in fact many worlds follows from (1), and so the modal fictionalist is committed to the existence of the many entities that fictionalism13 was introduced to avoid commitment to. Rosen's initial modal fictionalism14 in the form proposed is unsuccessful15. A broader understanding of the problem is afforded, we believe, by recognizing that the problem isn't peculiar to modal fictionalism16. Moreover, we will be offering a strategy for dealing with each manifestation of the problem, one that appears rather compelling from the broader perspective.
- According to the modal fiction, at all worlds, there exist many worlds.
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2019
- Mauve: Text by correspondent(s) or other author(s); © the author(s)