Unlike most other recent writers on the subject, Sobel argues that the logical problem of evil - that is, the problem of showing that it is logically possible for God and evil to coexist - is a serious problem which recent treatments have not adequately dealt with. In his 12th chapter, he considers several deductive arguments from evil against the existence of God. In future posts, I will consider the specific arguments that Sobel makes, but here I just want to point out a flaw or limitation in the way Sobel frames his arguments.
Each version of the problem of evil Sobel considers has the following structure:
- Empirical claim about what the world is like.
- Claim that (1) and the existence of a perfect being are each individually possible, but incompossible with one another.
- Claim that there is no perfect being.
Now, classical theists generally hold that it is a necessary truth that God exists and is perfect. Let E be one of Sobel's empirical claims. The following inference is valid in standard modal logics:
- □(God exists)
- ~◊(God exists & E)
In other words, in each of the arguments, the conjunction of the second premise with the claim that God exists necessarily yields a contradiction, so necessary being theists will be committed to the rejection of (2) in each case.
Now, perhaps Sobel does this because he thinks that, earlier in his book, he's shown that necessary being theism isn't one of the better versions of theism. I think he's wrong about that.
At any rate, this is, as I said, merely a technical problem; I think that each of Sobel's arguments has some intuitive pull which is independent of his problematic formalizations, and I will discuss some of them in future posts.
[cross-posted at blog.kennypearce.net]