Monday, July 3, 2017
A modal argument for God's existence - why it fails
1. It is possible God exists
just trades on the ambiguity between metaphysical and epistemic necessity/possibility? Maybe there is, but I have not spotted one yet given an admittedly brief survey.
Here is a simple modal argument for the existence of God:
1. It is possible God exists
2. If it is possible God exists, God exists at least one possible world
3. If God exists at one possible world he exists at all of them (being a necessary being)
4. Therefore God exists at every possible world
5. Therefore God exists at the actual world
Note first of all that if we have good grounds for supposing God does not exist at the actual world, then the logic of the above argument also requires that we have good grounds for supposing God does not exist at any possible world. That is to say, we have grounds for thinking it is impossible that God exists. And perhaps we do have such grounds - e.g. in the form of vast swathes of evil.
In response, theists may say: 'Ah but it is possible God exists, because I can imagine God existing - I can certainly imagine that the property of maximal greatness is instantiated, say.'
This is to muddle epistemic and metaphysical possibility. Epistemic possibility is what might be the case given our fallibility/ignorance. It's the for-all-we-know variety of possibility.
To illustrate: it is epistemically possible that 12 x 12 does not make 144. That's epistemically possible because I acknowledge my mathematical fallibility - I may have miscalculated. However, that epistemic possibility does not establish that it's metaphysically possible that 12 x 12 does not make 144. It isn't (given that mathematical truths are, while epistemically contingent, also metaphysically necessary). 12 x 12 = 144 at every possible world.
Similarly, that Hesperus is not Phosphorus is epistemically possible (I can imagine it turning out that there's been a major astronomical cock-up or conspiracy regarding the supposed identity of the evening and morning stars). However, as Kripke points out, given we are dealing with two rigid designators here ('Hesperus' and 'Phosphorus'), if Hesperus is Phosphorus, then at no possible world is it false that Hesperus is Phosphorus. So it is not metaphysically possible that Hesperus isn't Phosphorus, despite it being epistemically possible that Hesperus isn't Phosphorus.
In short, if there are good grounds for thinking that x is metaphysically impossible, pointing out that x is epistemically possible does not, in general, provide any sort of rebuttal. E.g. If there's good empirical evidence Hesperus is Phosphorus, then there is in turn good evidence that Hesperus is Phosphorus is true with respect to every possible world, notwithstanding the fact that it's epistemically possible that Hesperus isn't Phosphorus.
Now yes, God's existence is epistemically possible. I can imagine God existing.
But suppose God does not actually exist. Then his existence remains metaphysically impossible - God exists at no possible world.
So the epistemic possibility of God existing is entirely compatible with the metaphysical impossibility of God existing.
So suppose I have good evidence that God does not actually exist - evidence in the form of the evidential problem of evil, say. Then I have good grounds for thinking, not just that God does not exist at the actual world, but that it is metaphysically impossible that God exists - that God does not exist at any possible world.
It won't do to reply - 'Ah, but God's existence clearly is possible, as I can imagine God existing!' For that would be to invoke the wrong sort of possibility.
Indeed, Atheists might consider the above modal argument as a kind of logical amplifier for their conclusion that God does not actually exist. It allows them to move from: as a matter of fact God does not exist, to necessarily, God does not exist! So perhaps we should add it to our arsenal?
The other obvious problem (which I know others have raised) with this modal argument is that it similarly allows one to move from it's possible God does not exist to, God actually does not exist. And most theists acknowledge the possibility that God does not exist. Hence they are, by the same logic, required to move from that possibility - which they acknowledge - of God not existing to the conclusion that as a matter of fact God doesn't exist.
What does Plantinga says is the aim of his argument. He says it is not a 'proof' of theism, but that it does 'establish' the rational acceptability of theism.
Does Plantinga's argument establish the rational acceptability of belief in God?
No. If Plantinga were right, it would then follow that if (as seems equally clear) it is not irrational to believe it's possible God doesn't exist, so we could similarly 'establish', by means of a mirror argument, the rational acceptability of believing that God does not exist.
But Plantinga thinks believing God does not exist is not rationally acceptable! That's the conclusion of his Evolutionary Argument Against Theism - he says the rejection of theism is self-defeating in a way that theism is not.
So, Plantinga, to be consistent, must consider the mirror premise that that it's possible God does not exist to be irrational, despite appearances to the contrary.
In short, the seeming rationality of both premise 1 and this premise may in each case quite easily be shown to illusory, depending on what case can be made for/against rational acceptability of theism.
But then Plantinga's modal argument for the rational acceptability of theism, and the above mirror argument for the rational acceptability of rejecting theism, clearly fail to, as Plantinga puts it, 'establish' the rational acceptability of either position.