Friday, 19 September 2014

Strange Notions of Contingency

I recently read an article on Strange Notions by philosopher and theologian, Dr. Peter Kreeft about what he calls “the contingency argument” (CA). It runs as follows:
Many consider the argument for God from contingency to be one of the strongest. The basic form is simple: 
1.  If something exists, there must exist what it takes for that thing to exist. 
2.  The universe—the collection of beings in space and time—exists.  
3. Therefore, there must exist what it takes for the universe to exist. 
4. What it takes for the universe to exist cannot exist within the universe or be bounded by space and time. 
 5. Therefore, what it takes for the universe to exist must transcend both space and time. 
Suppose you deny the first premise. Then if X exists, there need not exist what it takes for X to exist. But "what it takes for X to exist" means the immediate condition(s) for X's existence. You mean that X exists only if Y. Without Y, there can be no X. So the denial of premise 1 amounts to this: X exists; X can only exist if Y exists; and Y does not exist. This is absurd. So there must exist what it takes for the universe to exist.”


He then goes on to argue that this mysterious “what” is God.

Ignoring the latter half of his argument, which treats this final leap of identification, I think that there is an obvious problem that Dr. Kreeft ignores – specifically the fact that the notion of “what it takes” for something to exist is quite complicated. In this article, I shall explain where Dr. Kreeft slips up, and then attempt to indicate the way towards solving the problem.

Before we start, let’s clarify a few terms I use. Firstly, when I say “world”, I mean “the sum total of all states of affairs in an exhaustive set of states of affairs that could exist together”. “Actual world”, refers to the set of actual states of affairs. This is to be distinguished from the term “universe”, which I use to indicate the set of states of affairs that is the existence of the all the stuff that isn’t ‘God’ under the Christian conception (i.e. that which stands in relation to God in such a way that Dr. Kreeft believes demonstrates the existence of God Himself).

Now for the argument.

States of affairs can be either necessary or possible. Necessary states of affairs are the case in all possible worlds, and cannot not exist in any possible world. Possible states of affairs are the case in at least one possible world.

Suppose a state of affairs (e.g. the existence of X) is intrinsically ontologically necessary: to say that its existence presupposes 'what it takes for it to exist' is merely to say that its existence presupposes its being what it is. This means that, if the existence of the universe is necessary, there need not be some kind of 'outside' condition upon which its existence is contingent.

This is the first point where Dr. Kreeft falls down. Prima facie, the existence of the universe is neither clearly necessary, nor clearly contingent. Dr. Kreeft, however, implicitly assumes that the universe is contingent when he assumes that the “the immediate condition(s)” for its existence must be distinct from itself.

However, even if we grant Dr. Kreeft his assumption of contingency (which, as we shall see later, we might have good reason to do), there are still problems.

I will assume (and this is a generally uncontroversial assumption) that a state of affairs which is neither necessary nor impossible is contingent. It can be so in two ways: a) it can be contingent upon either some other condition (‘dependent’ contingency), or b) it can be a brute contingency (that is, its non-existence is possible in any given possible world, but there is no other condition upon which its existence is dependent). If (b) is the case, (1) is just false with regards to it. Indeed, it is only if (a) is true of the actual world that the argument might work.

As we have just shown, there is no reason to assume premise (1) in such a way that the CA leads us to a picture of the actual world in which God exists. There are too many other possibilities for the modal status of the universe. That said, we can possibly argue down the number of these possibilities.

Firstly, we might be willing to reject the notion of the universe as necessary as counter-intuitive. If the universe as it is in the actual world is necessary, then the states of affairs contingent upon it (i.e. the stuff that happens ‘in’ the universe) are effectively necessary too. This is because a necessary universe would underpin the same set of contingent states of affairs in any possible world. As such, the set of existing states of affairs contingent upon the universe in general would be the same in any possible world. This level of ontological determinism seems odd at the least.

We can extend this to construct a plausible argument as to why (a) should be wrong. Let us suppose that (a) is the case. In which case, the existence of the actual universe implies the existence of some condition upon which its existence is determinate. The existence of this condition can either be necessary or contingent. If this condition is necessary, then it is true in all possible worlds. Thus the existence of the actual universe, the existence of which its existence entails, would also be true in all possible worlds. In which case, the actual universe is necessary – which seems counter-intuitive.

If the condition is a brute contingency, then the CA doesn’t face a problem on the fronts mentioned in this piece. However, we might ask why we would think this. Suppose in any given world a contingent object has the possibility of being a brute contingency (as opposed to a mere contingency) of 50% (there are two options, with no reason I can suppose for one option, prima facie, to be the case over another). This means that any contingent universe has a 50% possibility of being a brute contingency, and an equal chance of being a dependent contingency. Given that we have rejected the notion of a necessary universe as counter-intuitive, this means that there is a 50% chance of the actual world being such that the CA might work.


What this means is that the CA could plausibly be a successful argument. However, more ontological groundwork is needed in order to establish this. I have attempted to cover some of this groundwork, although a supporter of the CA would still need to demonstrate the dependent contingency of the universe.

No comments:

Post a Comment