Abstract
David Lewis objected to theories that posit necessary connections between distinct entities and to theories that involve a magical grasping of their primitives. In On the Plurality of Worlds, Lewis objected to nondescript ersatzism on these grounds (and thus branded it as ‘magical’). The literature contains several reconstructions of Lewis’ critique of nondescript ersatzism but none of these interpretations adequately address his main argument because they fail to see that Lewis’ critique is based on broader methodological considerations. I argue that a closer look at his methodology reveals the broader objection he presented against nondescript ersatzism. This objection, I further argue, remains a challenge for the ersatzer who posits structure-less entities as possible worlds.
Similar content being viewed by others
Change history
13 July 2019
Please note that this article belongs to the Special Issue on the Legacy of David Lewis but was included in issue 195:5 by mistake. It should be regarded as part of this selection of articles.
Notes
This view is usually called ‘magical ersatzism’. I call it ‘nondescript ersatzism’ to emphasise the fact that it is the theory’s property of being nondescript rather than being magical that is of most importance.
Lewis considers ‘further hypotheses’ about selection only to put them aside as he ‘[does] not think they do any better’ (Lewis 1986b, p. 182).
You might think there is no relation realistically conceived as an entity; the two-place predicate ‘... selects ...’ is a piece of primitive ideology that does not correspond to an entity in the ontology of the theory. So, it makes no sense to say that the relation is not graspable or magically connects the concrete world and elements. However, the issue is not really to do with relations as entities. It is about our concept of selection that is expressed by the predicate ‘selects’ (this is so even in the case of the selection relation being external). If you deny the existence of the relation, you still have the predicate and hence must give an account of how we grasp the concept expressed by that predicate.
David A. Denby says in reply to Lewis that we can grasp the selects relation by positing a further primitive that pairs properties of elements with properties of the world. The world being P selects element E being Q iff some P and some Q stand in the pairing relation R to each other and the world is P and element E is Q (Denby 2006, p. 167). So we do not need to grasp particular properties of elements; rather, the reply goes, we need only grasp the Q-properties in general via quantification over them. However, this proposal is misguided. To label the property \(Q\) as a ‘Q-property’ is to commit the error of describing an entity in terms of the role it fills. It is pointless to tell us this property is called ‘Q’ and that this is what property it is because it plays the Q-property-role. We have no notion of what a property called ‘Q’ is beyond the fact that it plays the Q-property-role. You might reply on behalf of Denby that he hasn’t described or named the Q-properties; instead, he has quantified over them. Even still, to quantify over Q-properties we require in our fundamental ideology the predicate ‘... is a Q-property’ in sentences like: ‘there is an \(x\) such that \(x\) is a Q-property’. We need a predicate like this to demarcate the Q-properties from the P-properties. We are confronted with our original problem.
Since his objection here does not rest on the relevant intrinsic representational properties being beyond us causally, the reply that we can know these properties by inference to the best explanation does not directly address the issue.
According to Slote, propositions are to be identified with possibilities, as per Ockham’s razor (Slote 1975, p. 148). One example: ‘[t]he proposition that Helen is white at \(t\) is, on our view, the logical possibility that Helen is white at \(t\), or, perhaps, alternatively, the logical possibility of Helen’s being white at \(t\)’ (Slote 1975, p. 150). The proposition that \(p\) is true is the possibility of \(p\) being realised. A possible world is thus a maximal possibility. Properties are also identified with possibilities (Slote 1975, p. 154). What it is for \(a\) to have \(F\) is for the possibility that \(a\) is \(F\) to be realised. States of affairs, propositions and properties are all identified with primitive possibilities. Even necessary propositions are accounted for; they are possibilities that must be realised (Slote 1975, p. 153).
Although Lewis does not mention Adams in his list of nondescript ersatzers on p. 183 of (1986b), it seems reasonable to include Adams in the second camp.
Consider a similar thought expressed elsewhere by Plantinga: ‘Possible worlds themselves are typically ‘taken as primitive’, as the saying goes: but by way of informal explanation it may be said that a possible world is a way things could have been—a total way’ (Plantinga 1976, p. 139, his italics).
Jubien (1991, pp. 265–66) says Lewis thinks that the phrase ‘the proposition that p’ does not pick out a specific proposition and that the phrase ‘representing that p’ does not refer to a specific representing property had by an element. But this is not entirely correct. Lewis says they can be named but thinks nondescript ersatzers must provide a story about how this is so.
References
Adams, R. M. (1974). Theories of actuality. Noûs, 8(3), 211–231.
Alexander, S. (1920). Space, time, and deity: The Gifford Lectures at Glasgow 1916–1918 (Vol. 1). London: Macmillan.
Bennett, K. (2013). Having a part twice over. Australasian Journal of Philosophy, 91(1), 83–103.
Denby, D. A. (2006). In defence of magical ersatzism. Philosophical Quarterly, 56(223), 161–174.
Forrest, P. (1986). Ways worlds could be. Australasian Journal of Philosophy, 64(1), 15–24.
Hawley, K. (2010). Mereology, modality and magic. Australasian Journal of Philosophy, 88(1), 117–133.
Hymers, M. (1991). Something less than paradise: The magic of modal realism. Australasian Journal of Philosophy, 69(3), 251–263.
Jubien, M. (1991). Could this be magic? Philosophical Review, 100(2), 249–267.
Lewis, D. (1966). An argument for the identity theory. Journal of Philosophy, 63(1), 17–25.
Lewis, D. (1973). Counterfactuals. Cambridge: Harvard University Press.
Lewis, D. (1976). The paradoxes of time travel. American Philosophical Quarterly, 13(2), 145–152.
Lewis, D. (1983). New work for a theory of universals. Australasian Journal of Philosophy, 61(4), 343–377.
Lewis, D. (1986a). Against structural universals. Australasian Journal of Philosophy, 64(1), 25–46.
Lewis, D. (1986b). On the plurality of worlds. Oxford: Basil Blackwell.
Lewis, D. (1991). Parts of classes. Cambridge, MA: B. Blackwell.
Lewis, D. (1999). A world of truthmakers? In D. K. Lewis (Ed.), Papers in metaphysics and epistemology. Cambridge: Cambridge University Press.
Melia, J. (2008). Ersatz possible worlds. In T. Sider, J. Hawthorne, & D. Zimmerman (Eds.), Contemporary debates in metaphysics. Malden, MA: Blackwell.
Nolan, D. (2015). It’s a kind of magic: Lewis, magic and properties. Synthese. doi:10.1007/s11229-014-0565-4.
Plantinga, A. (1974). The nature of necessity. Oxford: Clarendon Press.
Plantinga, A. (1976). Actualism and possible worlds. Theoria, 42(1–3), 139–160.
Slote, M. A. (1975). Metaphysics and essence. Oxford: Basil Blackwell.
Stalnaker, R. C. (1976). Possible worlds. Noûs, 10(1), 65–75.
Stalnaker, R. C. (1984). Inquiry. Cambridge, MA: MIT Press.
van Inwagen, P. (1986). Two concepts of possible worlds. Midwest Studies in Philosophy, 11, 185–213.
Zaragoza, K. (2007). Bring back the magic. Pacific Philosophical Quarterly, 88(3), 391–402.
Acknowledgments
Thanks to Peter Forrest for discussion and Steffi Lewis for permission to publish an excerpt from a letter by David Lewis. I am grateful to the British Academy for a Newton International Fellowship and the John Rylands Research Institute in Manchester, UK for research support.