"What Caused God?" Is Actually a Good Objection

... Sort of, Anyway

Many theists are quick to dismiss the “What caused God?” objection as a kindergarten-level confusion peddled by new atheist types who are, as is often the case, fantastically ignorant on theological matters. Baby’s first theology, you might say.

However, the typical theistic response—namely, to say that God is uncaused because God is necessary—isn’t really a great rejoinder. After all, there are a fair number of atheists (how many, I don’t know, but at least a few smart ones) who argue that the physical universe—or some aspect of it—is necessary. In fact, a counter you see in the literature is something like this: “If it’s good for the theist, it’s good for the naturalist. You say God is uncaused because God is necessary? Fine, I say the universe (or some stage of it) is a necessary reality and therefore uncaused. So, no need for God.”

You can probably guess where I’m going with this. If the theist wants to actually address this concern, they need to bring more to the table than vague appeals to necessity. Being necessary doesn’t automatically mean something is self-explained or uncaused. There are plenty of necessary realities that still have deeper explanations. For instance, imagine God, because of His supreme goodness, must create the best of all possible worlds. In that case, the resulting state of affairs would be necessary—it had to exist—but it would still be causally dependent on God. (By the way, I don’t think this scenario is true; it’s just an example to show how something can be necessary, in some sense or other, and yet not truly ontologically independent—it’s still dependent on something deeper and more fundamental. And if you don’t like that example, just look at mathematical or moral truths. Lots of people take those to be necessary, yet they still call for deeper explanation.)

What we’re really looking for (or at least, what I think everybody should be looking for) is true ontological independence—something that doesn’t rely on anything else, in any way, to exist or to have the characteristics it does. Sure, whatever is truly ontologically independent would also have to be necessary, but necessity by itself isn’t enough to establish this. Things can be necessary and still caused.

Ultimately, I don’t think the "what caused God" objection is successful, but it often highlights that theists may not have fully appreciated the best motivations for theism—or, by extension, developed the best conception of God because of that. What I mean is what I set out to accomplish in The Best Argument for God: we need to think about God as a theoretical entity designed to fulfill a specific role—namely, being the ultimate autonomous (not brute!) fact. That is, the sort of thing where, if we could fully grasp its nature, no mystery would remain as to why it exists (necessarily, even) and why it has the features and characteristics it does.

As I suggest in The Best Argument for God—and elsewhere—this requires first getting clear on all the features that are contingency-implying. In other words, we need to identify the kinds of things that definitely cannot be the sort of theoretical entity we’re after. Here, I think the Perennial Philosophy tradition gets it right (Plato, Aristotle, Aquinas, etc.). Historically, it identified a range of attributes as contingency-implying: mutability, compositeness, qualitative finitude (or, in more contemporary terms, arbitrary limitation), and so on.

There are—and remain—good metaphysical reasons to think that anything changeable is contingent, anything composite is contingent, and anything qualitatively finite is contingent. This means our metaphysical investigation must go deeper than simply distinguishing between (the mere modally) "contingent" and "necessary." We need to explore why things are contingent in the first place and, relatedly, how to identify the kinds of entities that could never serve as the ultimate autonomous fact.

The related move to God, then, is something like this: stripping away all these contingency-implying features and positing, as a theoretical entity, something entirely bereft of them—something that serves as their ultimate, necessary condition or explanation. This would mean something immutable and eternal, non-composite (or absolutely simple), and qualitatively unrestricted (or infinite, perfect).¹ Or in other words, the unchanged changer, the unactualized actualizer, the uncomposed composer, the undirected director, etc.

But the key—and rather simple—point for our purposes today is this: unless the theist commits to a conception of God that truly escapes these contingency-implying features—that transcends the categories effectively screaming, “Look at me, I’m caused!”—they do remain vulnerable to the "what caused God?" objection. In other words, the theist must arrive at a conception of God that is inherently uncausable. This is achieved by maintaining a principled yet restricted causal or explanatory commitment, one that applies only to specific types or categories of things and ultimately terminates in something that entirely escapes those categories.

(An additional advantage of this approach is that when all of this is worked out within a particular metaphysical system, one can conceptually derive the traditional suite of divine attributes, which provides a principled basis for why this ultimate reality is appropriately called God.)

1. For what it’s worth, one could approach this either through inference to the best explanation or by presenting a more robust deductive argument. For those who’ve read my book, they’ll know I endorse both approaches and think these methodologies can be mutually reinforcing.