The Kalam argument is an altered form of the cosmological argument. It is an argument that intends to circumvent the infinite regress problem contained within the traditional version by altering the premises.
William Lane Craig's version:
- Everything that begins to exist has a cause.
- The universe began to exist.
- Therefore, the universe must have a cause.
The distinction between this and the traditional cosmological argument is that it distinguishes effects in general from those that have a beginning. This qualification leaves open an interesting possibility that some things in the universe might exist that never began to exist. But Craig is not that sloppy, so before we jump on this observation we need to address the kalam argument's second premise and its support.
The kalam argument's second premise—"The universe began to exist"—is a claim that seems more of a presupposition than a fact, but watch how it is supported:
- An actual infinite cannot exist.
- A beginningless series of events is an actual infinite
- Therefore, the universe cannot have existed infinitely in the past, as that would be a beginingless series of events.
The important term here is, of course, "actual infinite." Wikipedia has the following to say about actual infinities:
- Actual infinity is the notion that all (natural, real etc.) numbers can be enumerated in any sense sufficiently definite for them to form a set together. Hence, in the philosophy of mathematics, the abstraction of actual infinity is the acceptance of infinite entities, such as the set of all natural numbers or an arbitrary sequence of rational numbers, as given objects.
The mathematical meaning of the term 'actual' in 'actual infinity' is synonymous with 'definite', not to be mistaken for physically existing. The question of whether natural or real numbers form definite sets is therefore independent of the question of whether infinite things exist actually in nature.
In order to say something "comes" into existence, then we must be able to distinguish between two sets of affairs in which that something exists, and a state of affairs in which it does not exist.
X = The universe
S1 = not X, or a state of affairs in which the Universe does not exist
S2 = X, or a state of affairs in which the Universe exists
We also must show that there must be a distinction made between these two states, as if there was no difference (x & not x) then there would be a logical contradiction (Law of non-contradiction)
Frame of reference and causality
If we use time as a frame of reference, where Moment 1 would be S1, and Moment 2 is S2, we can meaningfully say that they are true at different moments. The problem with this is the objective of Hawking that there was no "before" the universe - that the big bang brought into existence time and space. Thus, time cannot be a meaningful way to distinguish between S1 and S2. The only other causality is atemporal causality, if there is such a thing, where cause and effect are simultaneous. However, this leads to a straight contradiction as both S1 and S2 are true at the same time.
Problems with actual infinites
Mathematicians such as Georg Cantor and Michael Dummett have argued that actual infinites can, in fact, exist. This is a problem within mathematics, not a solved problem that the kalam argument can use without protest. As Arnold T. Guminski has elsewhere argued, the application of "Cantorian set theory to the real world…does not generate counterintuitive absurdities." In the mathematical sense, Craig's premise that an "actual infinite cannot exist" does not appear to be the case unequivocally. The question is whether it is in this sense that Craig really intends to use it or not.
In this sense, the entity in question seems to be time, or more correctly space-time, which the following:
- An actual infinite cannot exist.
- A beginningless series of events is an actual infinite
designed to support the kalam argument's second premise;
- 2. The universe began to exist.
seems designed to support. So the real question is whether time itself is infinite, which seems to be what Craig and other proponents of the kalam argument seem to be answering as "no."
One of Craig's examples of an actual infinite is a bookshelf with no end, but the same point is made. The idea is that there must be a point where the books end, otherwise they would fall, either over or down (to where, I wonder).
Scientific research about the universe's origins seem to point to a beginning to the universe in its current form, but not necessarily to the beginning of matter itself, and thus not of time either. Considerable debate exists over this question among scientists, so it is premature to declare that space-time is, by default, a thing with a beginning. The events at the singularity itself, as predicted by the big bang theory, are not understood by current cosmologists. Before a certain point, we cannot say what happened, let alone any possible events that took place before the singularity. This having been admitted, ignorance is not a justification for an insertion of supernatural causation.
Our inability to comprehend the nature of this enigma, however mind-boggling it is, is not sufficient to insert an answer that would require the same explanation. This god of the gaps argument—that because we don't know some supernatural deus ex machine must intervene—is not reasonable. There is no reason to reject, out of hand, that the universe can't be an actual infinite (or, for that matter, that it is incapable of self-cause) no matter how non-sensible it sounds. Without a conclusion as to why an actual infinite can't exist or why time cannot be infinite we cannot accept kalam's second premise.
Furthermore, the disputed fact that an "actual infinite cannot exist" does not prove that there was time before the Universe in which God created the universe. It just proves that there could not be an infinite amount of time between the first moment of the Universe and now.
The kalam argument seems to have been worded specifically to address the refutation of the cosmological argument, as it made the qualification that only things that begin have causes. The kalam arguer will simply state that God didn't begin, and so no regress occurs and no Creator of God is necessary.
However, this answer seems flawed. The essential criticism can be pin-pointed in an arbitrary choice, a form of the fallacy of special pleading, that must be made. As Richard Dawkins put it, the cosmological argument makes "the entirely unwarranted assumption that God himself is immune to the regress." Whether we qualify the first premise to exclude non-beginning things (as the kalam argument does) or not (as the cosmological does), the essential question is why it is more logically defensible to claim that for the rule that everything (or at least things that begin) must have a cause, an exception is made for God but not for the natural universe as a whole? Why does god not begin? It appears to be a wholly arbitrary choice, as in either case the rule must be violated, but with the proposition of God, we have to add something to the theory that adds nothing else to it.
If God not having a beginning is not a problem for Craig and other defenders of this argument, why is it a problem for the natural universe? To answer this, we must look at a further problem. This problem concerns the definition of god used in both arguments. A theologian might reply this counter argument and insist that the decision is not arbitrary, and that god must be allowed to have these attributes that the kalam argument seems to imply. He may say that the argument is an attempt to show the need for there to be a God that has the attributes that we cannot find in the universe. He might say that because we know that everything in the universe needs a cause and that the idea of infinite time is non-sense, there must be this being with these unique attributes. That is, there must be this being that does not begin, has no creator, and is thus able to create the universe. However, this suffers from the same problem from, and is in fact the same as, the ontological argument.
The ontological argument strives to define a god into existence. Essentially, it asks us to imagine the most perfect of all beings, and says that it must exist because existence is better than non-existence, and if this being is truly perfect it must have this attribute as well. The problems with this argument are two-fold; merely thinking or imagining some being does not imply the being has actual existence outside of it being conjured in the imagination.
This choice is not only unnecessary, it is not parsimonious. In order to explain something apparently designed and which cannot create itself, a being is conjured into existence which would require even more unlikely explanation.
The Kalam argument attempts to circumvent the problem of infinite regress but steps right into the problem of special pleading so is no better off.
"Comparing apples and oranges"
In the first premise, Craig declares "everything that begins requires a cause," and goes on to place the universe at the same logical level as its contents.
The first premise refers to every "thing," and the second premise treats the "universe as if it were a member of the set of "things." But since a set should not be considered a member of itself, the cosmological argument is comparing apples and oranges.
See Russell's paradox for issues that arise from allowing a set to be a member of itself.
The Cosmological argument does not prove that the cause was a supernatural cause, or not a natural cause.
Although some other variation of the Kalam argument or Cosmological argument may be internally consistent even if all the terms given are agreed upon by all parties concerned, the argument actually makes no effort to demonstrate anything tangible in nature regarding the manifestation of a God. An example analogous to the Kalam argument would be a geometry proof on some type of polygon. Even though the entire table of proofs is totally internally consistent, it does not demonstrate that the actual polygon exists in nature. An exhaustive effort to prove all the angles of a triangle will always add up to 180 degrees says nothing about whether or not triangles exist.