One of the objections to William Lane Craig’s suggestion that God is in time ‘since’ creation is that it violates both the immutability of God and the metaphysical simplicity of God, since God must learn tensed propositions over time, and thus his knowledge keeps track of the rollover of propositions as the flow of time proceeds. However, what of somebody who holds a classically Augustinian position which makes reference to God’s knowledge of contingencies such as what Susie freely chooses (in the libertarian sense) at time t1. This knowledge is classically called knowledge by scientia visionis. Does this kind of contingent knowledge, which God would have in virtue of the contingent reality at hand (namely, in this case, what Susie really ‘does’- in the tenseless or perfect sense – freely choose at t1)? If not, how do we differentiate the one instance of knowledge acquisition from another?
Suppose that the best account of logically possible worlds is that they are each maximally specific propositions, rather than sets of maximally consistent propositions (as I’ve previously argued, we may have to say just that given Grim’s Cantorian argument against any set of all true positions). We might then say that God’s knowledge of all and only true positions is satisfied in an unchanging way when he knows by scientia visionis, and yet now if he knows according to tense (and that would entail that tense is an objective feature of reality, which is essentially the ‘A-theory’ postulate). Of course, semantically, Craig’s view satisfies omniscience (at least as long as it isn’t modally constrained). However, the significant difference is that, on Craig’s view, God’s knowledge is mutable. God does not know one maximally specific true proposition, but rather must know all truth intuitively, and yet be stuck with a set of maximally consistent (and ever changing) propositions, which at any time together make up the set of ‘all and only true propositions’. The problem on Craig’s view, it seems, is that God’s knowledge is not immutable. Notice that for God’s knowledge to ‘change’ presupposes that it goes from one state to another different state. That is what happens on Craig’s view, thus violating immutability. The Augustinian view could be accused of also involving a ‘change’ in God’s knowledge of the world too; God’s knowledge of the world prior to creation is merely counterfactual and grounded in potentialities. However, his complete knowledge of the world comes via scientia visionis given creation. This, one might argue, qualifies as a ‘change’ in God’s knowledge (not over time, but still knowledge ‘of’ the world). I think God can know some things ‘about’ the world prior to creation, such as what the telos of the created order would be. It becomes a difficult stretch of language to speak about true propositions about the world absent the existence of the world, and certainly we might just say that God’s knowledge of what a possible world would be like does not qualify as knowledge of the actual world. Thus, we might say that God’s knowledge of the ‘actual’ world is immutable, even if it is informed by the actual world.
Thus, the seminal difference between his view and the classical view is that, although some things need to be known contingently about the world, and thus the world literally informs the maximally specific true proposition which God knows, God does not ‘learn’ in the mutable sense, since his knowledge is unchanging in principle. This, to my mind, satisfies the principle of immutability and successfully safeguards the doctrine of divine simplicity.
Curiously, I think that since Craig accepts the reality of Molinist-type counterfactuals (even while they’re truth values, and existence, are ungrounded) he would probably accept God’s actual counterfactual knowledge of the world absent its creation – including those propositions which would be Molinist-type counterfactuals if the world existed. Although I don’t accept that there are ungrounded true counterfactuals, I think that the admition of them would offer an interesting apologetic to the following argument: https://philosopher145.wordpress.com/2012/02/10/an-argument-against-the-existence-of-the-christian-god-1-updated/ the response to which would come as early as at the fourth premise, where the Molinist would simply object that God happens to know the counterfactual that “Q worlds would not instantiate if creation were to happen” prior to his act of creation. Interestingly, last time I went to this philosopher’s blog, he had a fair argument for Theism posted on there. He seems to have taken it back down… That’s a shame, but hopefully we can look forward to more good philosophy from him either way.