Defending the MCA from an Old Criticism

Only propositions are necessary, say Hume and Russell. If God exists, He does not have necessary existence or, more specifically, logically necessary existence. [1] In order for a proposition to be necessary, this view states that only analytical propositions meet the bill, e.g. A=A, either A or ~A, etc.

The above view has a powerful defeater. You will recall that in order for the MCA to work, it is postulated that the sum total of contingent entities C possibly has an external cause. This external cause must be a necessary entity N. Given that N is possible, it follows (in conjunction with S5) that N exists.

In order to reject the MCA, the skeptic has to say that N is not even possible, which entails:

1. Necessarily, every cause is contingent.

Is (1) analytically true? Obviously not. Yet on Hume's hypothesis, (1) should therefore be rejected. Given that the negation of (1) entails that N is possible and therefore actually exists, the Humean is forced to say that N both exists and does not exist. Of course, this is a violation of one of the analytical propositions that Hume does take to be necessary - namely, the law of non-contradiction.

In short, the Humean hypothesis is self-defeating and should not lead the subject to conclude that N is impossible. Moreover, if N is not impossible [2], it follows that N is possible, which is one area where the MCA has its strength.

[1] On this view, God may still have temporally necessary existence.

[2] Other objections will likewise have to be disposed of.