(redated post originally published on 16 November 2011)
I have heard in various quarters recently the claim that Lewis’ version of modal realism is (a) just a kind of multi-verse theory; and (b) intrinsically incompatible with theism. A partial discussion of this issue may be found in the pages of Philosophia Christi:
Richard Davis ‘God and Modal Concretism’ Philosophia Christi 10, 1, 2008, 57-74
Graham Oppy ‘Reply to Richard Davis’ Philosophia Christi 11, 2, 2009, 423-36
Richard Davis ‘Oppy and Modal Theistic Proofs’ Philosophia Christi 11, 2, 2009, 437-44
I think that (a) and (b) are both mistaken. But, in order to explain why, we need to have some details of the relevant theories before us.
A. Lewis’ Plurality of Worlds
Here are some of the salient features of the Lewisian view:
1. Individuals are world-bound: no individual exists in more than one world.
2. Worlds have no external relations to one another.
3. (Consequence of 1.) There are no ‘extra worldly’ individuals who are externally related to more than one world.
4. (Consequence of 1.) If there are gods, then gods are world-bound individuals related to exactly one world.
5. There are no ‘island universes’: in each world, there is but one connected spatio-temporal manifold.
6. If there are gods, then spatio-temporal relations are not the sole external relations (since gods are then externally related to spatio-temporal manifolds and yet that external relation is not spatio-temporal). If there are island-universes, then spatio-temporal relations are not the sole external relation (since island universes are externally related to one another, and yet that external relation is not spatio-temporal).
7. The worlds–and their ‘contents’–are truth-makers for (de re) modal claims. ‘I might have Fd’ is true just in case there is a world in which a counterpart of mine Fs.
8. Standard theism receives the following formulation on Lewis view: In every world, there is exactly one all-powerful, all-knowing, perfectly good being that creates the single connected spatio-temporal manifold of that world (if there is one), and these beings are all counterparts of one another. On Lewis’s account of (de re) modal claims, this state of affairs would make-true the sentence ‘Necessarily, God creates whatever spatio-temporal manifold there is’. (Of course, Lewis does not accept standard theism. The point here is just that this is what the view would look like in a Lewisian framework.)
1. Possible worlds may ‘contain’ many universes, i.e. many maximally connected spatio-temporal domains. If the actual world is a multi-verse, then it ‘contains’ many universes. (There is a question about whether to allow connections involving wormholes, or singularities, or the like. I think that this is simply a verbal question about the definitions of ‘maximal connectedness’ and ‘universe’.)
2. Individuals are universe-bound: no individual exists in more than one universe.
3. Universes have no ‘regular’ external relations to one another. (No information can pass through a wormhole or a singularity, if wormholes or singularities are held to ‘connect’ universes. Etc.)
4. Standard theism receives the following formulation: There is an all-powerful, all-knowing, perfectly good being that exists in every world and that creates all of the universes in every one of the worlds (whether there are many, one, or none). (‘Necessarily, God creates whatever universes there are.’)
5. The multi-verse hypothesis is neutral on the logic, semantics, and metaphysics of modality. Defenders of this hypothesis could be: ersatzists, or primitivists, or fictionalists, or (I say) even (modified) Lewisian realists.
C. What modifications are needed?
1. We need to give up the idea that there is just one connected spatio-temporal manifold in each world.
2. On grounds of simplicity, Lewis prefers a view on which there is just one external relation: worlds just are universes, i.e., maximally spatio-temporally related entities. But, he recognises that his view can accommodate further external relations (and would need to do so, if, for example, it was to allow that it is possible that there are gods or island universes).
3. On grounds of simplicity, Lewis is opposed to island universes. However, as just noted, provided that he admits further external relations, his view can perfectly well accommodate island universes.
D. What might be controversial about the fore-going?
It goes without saying that Lewis’s account of (de re) modality is controversial. Famously, Kripke thought that invoking counterparts as truth-makers does not do justice to the thought that (de re) modal claims are about the very individuals to which they refer. But this is not an issue that has anything particularly to do with theism; if it succeeds, it’s equally an objection to the proposition that Lewis theory can accommodate (de re) modal claims about any other individuals (e.g. you or me). When we ask whether we can accommodate standard theism in a Lewisian framework, we only have an interesting question if we set aside global worries about the adequacy of the Lewisian account of (de re) modality.
I imagine some theists might say: no view can be acceptable if it allows that there are universes that are not made by God (where ‘God’ is taken to refer to a being that exists in the actual world). Now, it is true that this claim would be true on the above version of Lewis’ view. But it is also true, on the above version of Lewis’s view, that it is impossible that there is a universe that is not made by God. Does this mean that the above version of Lewis’ view is contradictory? No.
On the version of Lewis’ view in question, there are no ACTUAL universes that are not made by the ACTUAL God — even though there are non-ACTUAL universes that are not made by the ACTUAL God. Our objecting theist should be happy to accept that there are MERELY POSSIBLE universes that are not made by the ACTUAL God — but, by Lewis’s lights, that’s all that you are required to commit yourself to when you accept that there ARE universes that are not made by God.
(redated post originally published on 16 November 2011)