Augustine's problem

Augustine raises this problem: What was God doing for the infinite amount of time prior to creation? Why didn't God create the world earlier?[note 1] Augustine reports the joke answer that God was busy creating a hell for people who ask such questions, and then goes on to give his famous answer that God created the universe and time simultaneously.

Augustine's answer is a good one. The start of time is a non-arbitrary answer to the question of when to create the universe. However, Augustine's answer can only be adopted if God has an atemporal existence. So if Augustine's answer is the only one or the best one, we have an argument that God has an atemporal existence.

But could someone who takes God to be only a temporal being—say, an open theist—give an answer to Augustine's problem? If God is a temporal being, then time has infinite age, as God then does (the suggestion that God is a temporal being that has finite age is incompatible with divine eternity, even if it is compatible with the claim that God exists at all times). Hence no answer that depends on time's having a beginning will do.

One way to see a problem is to imagine God deliberating annually, say a million years before creation. God has good reason to create that year—after all it, is good that there be a created world. Maybe he has good reason not to create as well (maybe creation entails that there is imperfection; at least, creation makes reality less simple). But in any case, there intuitively should be some moderate probability, say 1/2, that he would create that year. But he doesn't. That's fine: he also had a probability of 1/2 that he wouldn't. But likewise he doesn't next year. He had probability 1/4 of creating in neither year. That's fine: events of that probability aren't very surprising. But keep on running this. He doesn't create in any of a 1000 years. That's much less likely. The probability of that is 2⁻¹⁰⁰⁰. So, it seems, on the assumption that God is in time and there is an infinite past, we have very good reason to expect that God would have created the world earlier than he did—no matter when he created it!

There are two difficulties with this line of thought. The first is that numerical probabilities can't be assigned too divine deliberation. That's fine: they are still heuristic and highlight the extreme unlikeliness of the scenario of God waiting for an infinite amount of time before creating. The second is that it presupposes a particular model of how God deliberates whether to create, namely that he continually deliberates whether to create there and then.

Can we solve Augustine's problem if, instead, we accept a model on which God from eternity (i.e., at every past time) decides on a particular time t at which to create? Well, if God is changeable, that still leaves open the question of why God didn't change his mind—why God kept on waiting, even though he had reason to change his mind (namely, the reason that creation could come earlier if he changed his mind). If he had probability one in a million of changing his mind in any given year, we'd expect that over a million years he'd have changed his mind, and a again an intuitive argument like before can be run. Maybe, though, a changing God can unchangeably determine his will (by making a promise to himself, maybe?), and at every time he always already had done so, given that he knew there would never be any new information becoming available? Or maybe God, while in time, is unchangaeble, and hence his decisions cannot change? Or maybe God, from eternity, efficaciously willed that creation should occur at t₀—God's efficacious willing need not be contemporaneous with what is willed.

So there are some things that can be said about the changeability subproblem. Our best model right now for a God-in-time story is that God from eternity has unchangeably decided that creation would come at t₀. It is tempting to say "And then God waited until t₀." But waiting is what you do when you have little else worth doing, and God is ever infinitely active in his intratrinitarian communion. So we shouldn't say that. Let's focus, however, on a different subproblem. Why did God eternally choose t₀, rather than say t₀−1 year, for the date of creation? Absent distinctions coming from notable events within creation, all times are, presumably, exactly alike. I suppose two answers are available. First, that it is a reasonless divine choice. Second, that there is a reason, in that there is a particular incommensurable value associated with each possible time for creation. I think these are tenable answers, but it must be noted that neither is uncontroversial. The subproblem of why God chose this time rather than another is a hard one.

There is another subproblem, however, related to the changeability subproblem. Take an answer to the changeability subproblem. Consider the proposition p_n which says that in year n B.C., God was decided that he would create at t₀. What explains p_n? Presumably it's p_n+1. I.e., God was decided in year n B.C. because he was decided in year n+1 B.C. But this generates a vicious regress.

All of this suggests that Augustine's answer is the best answer to Augustine's problem. And we have reason to reject views on which God is a temporal being.

This leaves two kinds of views. The first are ones on which God is solely atemporal (except in virtue of an Inarnation) and the second are ones on which God exists atemporally, but with creation comes to be omnitemporally present as well, as an aspect of his omnipresence. I do not know if either view is compatible with the A-theory. Certainly, I don't think presentists can make sense of atemporal divine existence. So theists (at least ones who, like Christians and most relevant scientists, think that the universe has finite age) shouldn't be presentists.