Contingency and Knowledge
On the Cosmological Argument and the Rejection of the Infinite
One of the classical arguments in defense of theism is the cosmological argument. As with all arguments in this most ancient of debates, there are a few different varieties of the cosmological argument, but the basic gist goes more or less as follows:
Everything has a cause
An infinite regression of causation is impossible
Therefore there must be some first cause
This first cause is God
This is, of course, a summary, and could be fairly called a “straw-man” of more sophisticated versions of the argument, such as those posited by Aquinas, which I will briefly include for context. Aquinas argues:
There exist contingent things, for which non-existence is possible.
It is impossible for contingent things to always exist, so at some time they did not exist.
Therefore, if all things are contingent, then nothing would exist now.
There exists something rather than nothing.
Therefore, there exists a necessary being.
It is possible that a necessary being has a cause of its necessity in another necessary being.
The derivation of necessity between beings cannot regress to infinity (being an essentially ordered causal series).
Therefore, there exists a being that is necessary of itself, from which all necessity derives.
That being is whom everyone calls God.
But my object of interest is not necessarily the argument. Rather, it is a premise within the argument — specifically, premise (2) in my own little summary, which corresponds with premise (7) in Aquinas’ version.
Most cosmological back-and-forths revolve around the “what caused God?” question. There are various methods of attempting to modify the premises in order to deal with this usual counter (the “Kalam” variation famously attempts to sidestep the issue by modifying premise (1): “everything which begins to exist has a cause”). Generally speaking, the treating of an infinite regression as a kind of reductio ad absurdum is granted.
But why is this the case?
What is logically impossible about an infinite regression? All that is necessary is infinite.
Professor William Lane Craig — who has become a kind of central hub of apologetics — argues that infinite cannot exist because it is an idea, not something observable. But the constraints we imagine prohibiting the possibility of infinite might also be said to be ideas — constructs of our own invention. In the absence of such ideas, the default position might very well be infinite. It is important to point out that we are speaking of “infinite” in very (necessarily) broad terms, since it might not be definable in principle. But this indefinability does not prevent us from recognizing the possibility of there being no “beginning,” no “limits” in chronology which would necessitate some “first,” non-contingent entity, whatever we might wish to call this limitlessness.
Because of this limits of human knowledge, the possibility of infinite is actually impossible to refute. Of course, this does not prove that there is infinite. Bad science is classically marked by unfalsifiable propositions. But here we are not advancing a proposition, but challenging a premise — that infinite is impossible.
How could we possibly know this?
Logically speaking, it is actually somewhat absurd that this premise is allowed.
In fact, let us state the objection in clear, logical terms:
Finitude is not a necessary quality of being.
At the very least, we have no grounds to assert the contrary, that finitude is a necessary quality of being. It might be, but we can’t possibly know one way or another. The universe and its causal chain could go back forever.
(Is that a tautology? Was there time before the universe?)
But allowed the premise often is, because — as it turns out — finitude is a necessary quality of something else: explanation.
Any explanation which requires the invocation of infinite does not explain anything. (Here, again, it is important to remember that we, who are invoking infinite, are not now positing an explanation, but deconstructing one). An explanation has to offer some sort of understanding, which brings order and separation to a previously disordered appearance of things. Infinite is undifferentiated and — presumably — disordered. So while it is logically illegitimate to presume that an infinite regression of causation is false, any theory which purports to explain causality back to the beginning of time must necessarily reject infinite or else it isn’t an explanation.
What this reveals is a hidden premise in the entire cosmological argument: that the universe is ordered in such a way that it is knowable.
Stating the premise directly invokes all sorts of interesting questions about what it means to “know” something — is it really just “justified, true belief? That must be relegated to a separate essay. Let us question the premise: how can we know that the universe is ordered in such a manner? Did someone order it that way? If that is the assumption — as it was for Descartes, who believed his senses were reliable because God does not deceive — then the entire point of bringing up a cosmological argument is superfluous (for its circularity).
If we assume that the entire universe is knowable because we can construct useful rules and laws that help us practically navigate the world — tools like mathematics, and laws of logic, and so forth — on what grounds do we claim, with certainty, that this functional-enough experience we have (and we are all too liable to forget our miscalculations and errors here) describes a general knowability of the universe? How can a mole who knows his tunnel imagine he therefore grasps the sky, the oceans, and the stars? If he believes certain “principles” of tunneling are universal, does that prove a general knowability of the universe?
(Does the hard problem of consciousness itself disprove the presumption of knowability?)
There seems to be a gross problem of generalization, wherein we imagine that because we have come to grasp a few things, we can therefore deduce that… infinite is impossible.
Of course, the fact that we don’t — and maybe can’t — know some things doesn’t mean we can’t know anything. Even the bible — a book held in very high esteem by many proponents of the cosmological argument — tells us that we cannot know the hearts of men, and yet that does not stop the believer from recognizing things he can know (whether God designed them so that they could be knowable, or for some other reason).
Indeed, to believe that if not quite everything is knowable in principle then we can’t know anything at all… this might be a symptom of holding a coherence theory of truth, rather than a correspondence theory of truth. A coherence theory of truth requires certain unquestionable axioms, or else the whole edifice of knowledge comes crashing down and nothing seems knowable. If infinite is lurking somewhere in the background — or if an evil Gödel monster is skulking in the shadows, waiting to steal away the completeness of your mathematics — then one might feel that there is no solid axiom, that coherence is impossible, truth is impossible, and objective knowledge is illusory. This was Hume’s ultimate conclusion, and the motivation for Kant to invent the idea of a priori knowledge — the possibility of knowledge outside of the scary world of physical reality, with its infinites and Gödel monsters.
But it seems more likely that the cosmological apologists simply have knowledge wrong. It isn’t something we build up from first principles at all, but rather something we learn by experience in a messier, organic fashion. Perhaps “true, objective knowledge” of the sort they want really doesn’t exist… but that hasn’t stopped engineers from “close enough”-ing into being thousands of bridges and skyscrapers and other marvels that non-philosophers seem to treat as knowledge-enough to count.
In either case, it seems unlikely that God exists because philosophical explanations don’t get much benefit from infinite.