In his ‘Eternity’ (in Quinn and Taliaferro (eds.), A Companion to the Philosophy of Religion (Oxford: Blackwell, 1997), pp. 257-63), Brian Leftow canvasses some of the more prominent arguments for God being in time. Some of these are (quoting Leftow and retaining his labeling):
(b) God is alive. Lives are events. Events must occur in time. So God is in time.
(g) God acts. Actions are events. Again, events must occur in time.
(h) God is a cause. Either causal relations link only events, or they also link agents to events (“agent causality”). If the first, then as events occur only in time, God is in time. If the second, the agent’s action is dated at the time of the effect. So if God has any temporal effects as an agent cause, He is in time. (pp. 260-61)
Leftow notes that all of these arguments assume that:
(E) Events must occur in time.
He thinks that all of the arguments for this assumption are bad arguments. I think he is wrong about one of them.
Leftow observes that one argument for (E) is the following (again quoting Leftow and retaining his labeling):
(4) Necessarily, any event which occurs before or after another event occurs in time, and
(5) Necessarily, any event occurs before or after another event. (p. 262)
Leftow claims that (5) is false. He writes, ‘The mereological sum of all events is an event. So is the existing of all of spacetime. No event occurs before or after either’ (p. 262).
But it is simply not obvious that the mereological sum of all events is an event. On the contrary, there are good reasons for supposing this not to be the case: it seems that it is part of the nature of events that they be able to serve as causes, and to be a cause, an event must have a sort of unity.
Now some mereological sums do exhibit the sort of unity required of causes. Take for instance the mereological sum of events constituting the stock market crash of 1929. It seems reasonable to say that the stock market crash, which is a mereological sum of events, caused a number of things, many of which were horrifically bad. So, it seems, the stock market crash of 1929 is itself an event because it has the requisite sort of unity to serve as a cause.
But not all mereological sums of events are like this. Take the mereological sum of the event of England losing to France last Sunday in Euro 2004 and Neil Armstrong walking on the moon. There seems to be no sort of unity there, and, it seems, the mereological sum of these two disparate events can’t serve as a cause. And so this mereological sum is not itself an event.
Now what is the mereological sum of all events like? Is it like the mereological sum of events constituting the stock market crash of 1929 or is it like the mereological sum of England losing to France and Neil Armstrong walking on the moon?
I suppose I tend to think that it lacks the sort of unity that the stock market crash of 1929 has. But I don’t think I have to argue my case. Rather, I think I have to be convinced that it does have the right sort of unity. That’s so because there’s a well known compositionality fallacy lurking here:
If a is F, b is F, and c is F, it does not follow that the mereological sum of a, b and c is F.
As one instance: let E1, . . ., En denote all the events. It does not follow that the mereological sum of E1, . . . En is an event.
Now I’m prepared to countenance the possibility that:
If E1, . . ., En denote all the events, the mereological sum of E1, . . . En is an event.
is a true instance of a false principle. But I need to be convinced that it is. What special considerations come into play here to make it true? I can’t think of any.
So, provisionally, I think that the argument from (4) and (5) to (E) is a good one. And so the arguments (b), (g) and (h) remain undefeated for God’s being in time.