Intro
As far as I can tell, the cosmological argument remains the sole most convincing argument for God’s existence. Whether in William Lane Craig’s Kalam or Leibniz’s contingency form, both pose God as the unmoved mover, a being whose existence is necessary in order for the universe to be explained. Both arguments also make some key assumptions: that things must have a cause outside themselves and, more importantly, that an infinite regress cannot occur. The second assumption means that chains of events cannot go on forever; they must have a stopping point somewhere. I intend to argue in this post that (a) an infinite regress, while unintuitive, is not inconceivable, and (b) even if we accept the unfeasibility of an infinite regress, we shouldn’t necessarily accept the existence of God.
Infinite Regress
Traditionally, cosmological arguments have relied on the notion that something exists at the end of every chain. This is to say that while A may be caused by B, which may be caused by C, which may be caused by D, this process cannot continue to infinity—and something like Z must have a cause as well. Usually, this final cause is referred to as God. This seems like the intuitive answer: it is certainly unsatisfying to have a chain of events without an anchor grounding it at the end. But just because it seemingly lacks explanatory value doesn’t mean it’s wrong. In fact, rarely do we ever find complete explanatory value in the real world. Let’s consider pi. Pi is irrational, meaning it has an infinite number of digits. However, we can still utilize pi—the 1,000th digit in pi has a local explanation based on a rule of the numbers surrounding it. But the chain of pi still has no finite endpoint. The infinite regress does not block understanding or use: one can still compute any desired finite precision of the number. Modern-day algorithms can produce hundreds of new digits for pi at a time. Yet, they will never be able to find that last digit. If we can explain a finite number of digits with a logical rule, why do we need a fixed endpoint? I’d argue that the rejection of the infinite regress stems primarily from a desire to fully explain the universe in a satisfying way. Sure, one could believe that the creation of the universe is based on an infinite chain of events which we have not and will not explain, or we could just say that the universe was created by an omni-omni God. The second option sounds a lot more interesting, which I think is why it is so convincing.
Ontological Simplicity
Say I agree that an infinite regress is impossible. It still doesn’t necessarily follow that the being at the end of the causal chain is God. It doesn’t necessarily follow that a being of any kind is at the end of the chain either. We can look to Ockham’s razor for this. Formulated by English friar William of Ockham, the razor states that if one has two explanations with equal explanatory value, the one that makes the fewest assumptions is the better one. Here are the two live options:
- a necessary personal being with a raft of attributes (omnipotence, omniscience, perfect goodness, etc.) who freely wills the cosmos, or
- a necessary impersonal universe that just happens to exist—for example, a brute law of nature or initial condition that catalyzed the universe.
Both options seem to halt the spiral toward infinity. However, I’d argue that since neither one adds extra conceptual payoff or explanatory value, we should favor the impersonal option. We can already observe natural laws in the universe; what’s to say that there isn’t another—one of necessity—that the universe exists in the first place? On the other hand, a metaphysical, all-powerful, and all-knowing being seems like a bit of a stretch. It is, by definition, outside of our scope of reality. Invoking a God adds commitments that go well beyond what is needed to stop the chain. Therefore, it is unnecessary.
A theist might reply that only a necessary being can explain why a causal universe can exist at all (a version of the Principle of Sufficient Reason). But that is precisely what I am contesting: the PSR need not extend everywhere; local explanations may suffice, and a brute necessity is a coherent alternative to a metaphysical being. Until there is a clear explanatory surplus—something the God hypothesis accounts for but the minimal necessary terminus does not—the inference “no infinite regress, therefore God” is underdetermined.
So Why Are These Arguments So Convincing?
Let’s take a trip back to the argument from design, where a conclusion that the universe was made for us to inhabit it is implied. This belief is false—we evolved to fit the universe, not the other way around. An argument ad absurdum would be to say that the hole around a puddle was constructed to fit the puddle. It’s the opposite: the puddle evolved to fit the hole. Similarly, we know through evolution that our bodily mechanisms, seemingly created by design, simply evolved to fit our surroundings. Underneath this argument is the human-centered assumption that the universe is created for us, not that we are the result of chance. An assumption like this exists in the cosmological argument as well. Rather than positing a necessary being to ground reality, we must accept the brute fact that the universe simply exists. Though less consoling—and forfeiting the hope of a providential God—this may be the more defensible position, however difficult it is to embrace at present.