Aquinas’s Proof for God in Plain Language: Objections and Replies

In my previous article—which you should read first—I aimed to present Aquinas’s De Ente argument for the existence of God in plain language. Here, I consider objections.

Once again, my aim is to keep this fairly—as in mostly, but not entirely—simple. I won’t treat all objections, and I won’t treat the objections I do consider in enormous detail—just enough, I hope, to get the general idea on the table and gesture toward how the complaint can be adequately resolved.

Note: Ethan Muse and I plan to present and defend Aquinas's De Ente in a future epsiode of Summa Contra Everbody, our newly launched podcast. If you haven’t already, you might want to subscribe!

Objection 1) Aquinas’s argument involves a “controversial” metaphysic—constituent ontology, existence as a first-level feature, etc.

Answer: This is true. For Aquinas’s argument for God to run, at least some of his background scholastic metaphysical system must be in place. Obviously, for the purposes of this post, that is not something I could hope to substantively defend here.

So, just two relatively brief points.

First, it should be noted that the commitments required to run the rudimentary core of Aquinas’s De Ente argument are minimal and fairly commonsensical.1 They involve, first, the idea that what a thing is is really different from the fact that it is—that these principles of what-ness and is-ness are really distinct, at least where familiar, everyday objects, like my basset hound Pumpkin, are concerned.

The second commitment is that reality is intelligible: if something about a thing cannot be explained, or made sense of, internally, then something—or some collection of things—outside the thing can do the explaining.

To my mind, these are not only safe assumptions—namely, that existence is not “built in” to familiar, everyday objects and that their existence is not simply a brute fact—but assumptions that should be afforded something like default status.2

My second broader point is that there is no getting around doing metaphysics, and no metaphysical system is free of controversy. The benefit of Aquinas’s approach is that his argument for God naturally emerges from a metaphysical system that is independently motivated because of the critical work it does in making sense of the broadest features of reality—how things have the character they do, how they change yet persist over time, the similarities and differences among existing things (the one and the many), what causality consists in, how the laws of nature are grounded, etc., etc.

Not only that, but once deployed in cosmological reasoning and the philosophy of God, the resulting benefits themselves provide reason to commend the system, insofar as one achieves a robust yet conceptually disciplined understanding of God—one that can serve as the right sort of theoretical entity to explain everything else while also explaining, rather than causing, itself.3

Finally, let me say this. As we will hopefully see by the end of this article, this objection is, at once, perhaps the least and the most interesting. It is the least interesting whenever it is lobbed as a general, not-very-specific complaint—a sort of lazy dismissal of an argument or system of thought simply because it is controversial.

On the other hand, it is the most interesting objection because I do think that, once the metaphysical system is in place, all the other usual objections can be quite easily dispatched. The real work of objecting must therefore be done at the level of metaphysical-system comparison.

Objection 2) Aquinas’s argument commits the fallacy of composition. Just because every individual thing whose essence is distinct from its existence has a cause, that doesn’t mean the collection of all those things does.

Answer: Again, two responses.

First, more generally, not every inference from what is true of the parts to what is true of the whole commits a fallacy of composition. This seems obvious enough: if all the bricks are red, then the wall is red too. And it is very, very plausible that, if we are considering all and only those things whose essences are really distinct from their existence, then we still need something to account for why that entire collection exists at all, since nothing within the collection is able to provide such an account.

Nevertheless, it is quite clear that Aquinas’s argument does not reason in this way in the first place. Aquinas’s cosmological argument does not employ a strategy of aggregation or reason from all such beings taken collectively. It simply runs from just one entity whose essence is really distinct from its existence—that’s it.

In other words, from just one being whose essence is really distinct from its existence, Aquinas derives, as its necessary condition, a being whose essence is really the same as its existence. From there, once it is seen that there can be only one such being, it is then inferred that all other beings ultimately derive their existence from it. But again, no worry about a fallacy of composition lurks in any of this.

Objection 3) Aquinas’s argument assumes the principle of sufficient reason—which is controversial. Sorry—CONTROVERSIAL!!!

Answer: Two responses, yet again.

First—yes, it does. At least it assumes a rather modest and highly plausible version of PSR. That is, if something is not accounted for by the essence or nature of a thing, then an outside cause accounts for it.

Aquinas is simply committed to the idea—as plausible as anything—that things that do not account for their own existence or character are not just there for no rhyme or reason whatsoever. Give that up, and I think all rationality and knowledge are lost—skepticism consumes us.

(Of course, many other positive arguments can be given for this form of PSR, but I won’t rehearse those now—see my book for more. The point is, I hardly see this as a “cost” to Aquinas. It would be a cost to deny it!)

Second, there is a way of running an argument for God from Aquinas’s commitments that does not require PSR but actually delivers it as an eventual result. For those interested in seeing how that works, see my Millerian Cosmological Argument for God.

Objection 4) Perhaps an infinite regress can satisfy his principle of explanation after all.

Answer: Clearly, it cannot.

Given the PSR Aquinas holds, stacking beings whose essences are really distinct from their existence is never going to explain how they ever became—or presently are—actual, as in really existent, in the first place.

Think again of the analogy I gave in the original post. It doesn’t matter how many white Lego blocks you have or how you arrange them: if what we need to explain is a purple tower, then you are going to need something outside the collection of purely white Lego blocks to do it.

The same goes for beings that actually exist. If what we need to explain is how they are actually included in reality, then we are going to need something beyond beings whose natures do not demand that they exist to account for why they do.

I have to be honest here: this seems as obvious to me as anything. There is nothing so special about infinity that an infinite collection of derivative actualities somehow becomes underived. Adding more recipients of existence—even infinitely many—never supplies a source of existence.

P.S. There are more technical arguments for why this sort of grounding chain cannot regress infinitely, which I explore in my Review of Metaphysics article. But this more basic point is, I think, perfectly sufficient for now.

Objection 5) The thesis of existential inertia—namely, that once something is brought into existence, it will continue existing unless and until something extinguishes it—undermines Aquinas’s argument.

Answer: It does not.

Simply put, there is no theoretical room for existential inertia within Aquinas’s system.

To see this—and here I am going to assume some familiarity with Aquinas’s background metaphysics—consider that the potency of familiar, everyday objects must be actualized by a source of actuality. This follows from the real distinction together with a basic principle of causality. Said differently, actuality does not and cannot subsist in some potency, as though it could just “hang there” like any other property inhering in a subject.4 Obviously, apart from a thing’s existence, there is no subject for existence to inhere in.

This means that existence, while remaining a first-level feature of concrete individuals, cannot inhere in individuals in the way other properties do—like, say, a color property. So, suppose someone suggests that, since we can color a chair red without anyone needing continually to dye it for the chair to remain red, perhaps existence works the same way. They would be mistaken. The reason the color can persist in the chair is, quite obviously, that the chair exists. There is already an existing subject for the color to persist in. But a thing itself persists only because that thing has existence. It therefore cannot be the case that a thing’s existence persists because the thing persists.

Recall that on Aquinas’s metaphysic, a thing—and all its properties and powers—has whatever actuality it has only by virtue of its existence, including any supposed property or power of “continuing to exist.” So, if the inertialist tries to explain a thing’s continued existence by appealing to some feature—a real property or power, say, an “inertial” principle of persistence—they have already presupposed the existence of the thing in order for it to possess that feature in the first place.5 But to appeal to something that presupposes the thing’s existence in order to explain why it continues to exist is viciously circular.

Alternatively, one might simply declare the thing’s continued existence to be brute. But then the issue is not that existential inertia is a plausible thesis within Aquinas’s system. The issue is that one is really just rejecting Aquinas’s system—in particular, his explanatory principle. EI is no longer doing the work.

The same applies if one simply wants to say that EI is viable within an alternative system. Perhaps. But again, EI is not by itself doing the work. What is doing the work is something like: “I like this metaphysical system better than Aquinas’s.” And that is clearly a different sort of objection, one related more directly to the first complaint about Aquinas’s general metaphysical scheme.

Finally, one might worry that the Thomist is simply begging the question against EI. But this is not so. Thomists—certainly not Aquinas—do not merely assume at the outset that EI is false. Rather, Aquinas offers arguments for the real distinction between essence and existence—arguments we have not explored here—and, once that distinction and the relevant causal principle are in place, the falsity of EI follows as an implication.

P.S. I offer a more technical refutation of EI in my Millerian Argument, based on the relation between essence and existence as incomplete entities within finite individuals. I won’t get into that here, but I mention it for anyone interested in a refutation of EI that does not require Aquinas’s principle of causality.

Objection 6) Aquinas’s notion of God as absolutely simple requires us to say that God’s power is identical to God’s knowledge, which is identical to God’s existence, and so forth. But this “identity of the divine attributes” thesis is clearly incoherent. So, Aquinas’s approach to God must have gone wrong somewhere.

Answer: It has not gone wrong anywhere.

However, we must distinguish between two ways in which the charge of incoherence might be understood.

The first is easier to address. Here, the claim is that, in being identical with his attributes, God must be identical to various abstract objects. God is identical to power, knowledge, goodness, existence, and so forth—but power, knowledge, and goodness are abstract objects. Therefore, God is somehow identical to a collection of abstract objects—or perhaps to just one abstract object, as the case may be—which is clearly absurd.

The reason this charge is hardly a serious one is that Aquinas rejects precisely this sort of Platonic understanding of predication. The objection assumes a theory according to which predicates correspond to abstract objects that concrete things exemplify, and that it is this exemplification relation that makes those predicates true of them.

But this is not Aquinas’s view. What makes various real—as opposed to merely relational—predicates true of a thing is always something concrete in the thing itself: ordinarily, some constituent form or, more fundamentally, the thing’s essence. This is simply part of Aquinas’s scholastic, or moderate, realism.

Thus, to say that God is identical with his goodness is not to say that God is identical with an abstract object called Goodness. It is to say that the concrete reality in virtue of which God is good is nothing other than God himself. Likewise, the concrete reality in virtue of which God is powerful, knowing, and existent is not something added to God or possessed by God. It is God himself.

This, of course, invites the second objection, which goes something like this: But look at this battery—it has power but not knowledge. Power and knowledge are clearly, really distinct. Yet you are insisting that they are identical in God?

The answer to this objection can be spelled out in different ways, but the basic thrust is this. When we say that God’s power is identical to God’s knowledge, and so forth, we are not saying that these perfections as we typically understand and encounter them—say, as participated perfections—are always identical. Clearly, that is false. Rather, we are saying that these perfections in their infinite mode—that is, as essential rather than participated perfections, or in their infinite rather than finite mode—are identical.

In other words, we are saying that something like power, as we commonly experience and understand it, is identical in God to something like knowledge, as we commonly experience and understand it. (Those who have read my previous work will know that I like to cash this out in terms of ordered series and their respective limit cases, but we don’t need to get into all that here.)

Now, the predicates retain their distinct meanings from our side, since power is not what knowledge means, nor is knowledge what goodness means. In other words, the predicates are not semantically equivalent—which is fine, since semantic nonequivalence does not by itself require distinct ontological grounds. Indeed, on Aquinas’s view, the same infinite concrete reality grounds each of these intrinsic predications of God. (In the contemporary literature, one influential way of spelling this out is through divine truthmaker simplicity: God himself is the single truthmaker for these various intrinsic predications.)

And there is not only no reason to think that this could not be the case, but positive reason to think that it must be the case, given Aquinas’s metaphysics of participated versus essential perfections in conjunction with the conclusions of his cosmological argument.

Objection 7) Aquinas’s simple God results in modal collapse. If God is identical with his act, then God is identical with his act of creating—and so the necessity of God’s being transfers to creation, producing the seemingly false result that everything is necessary and nothing contingent.

Answer: Essentially, there are two steps to defusing modal-collapse-style objections.

First, we must distinguish between what is intrinsically predicated of God and what is predicated of God only extrinsically or relationally. The basic idea here is that an intrinsic predication is true in virtue of the way a thing is in and of itself, whereas an extrinsic predication is true at least partly in virtue of the way that thing stands in relation to something outside itself. “John is male” is an intrinsic predication. “John is taller than Sam” is an extrinsic predication—and, importantly, it could become false simply because Sam grows, even if John undergoes no intrinsic change at all. Said differently, extrinsic predications are the sort whose truth can change without the subject of the predication changing intrinsically.

Second, we must note that divine simplicity commits us only to the idea that God is really identical with the concrete reality grounding all true intrinsic predications of him. Said differently, God is identical with whatever in reality makes those intrinsic predications true of him—namely, the divine essence, which is God himself—but not, obviously, with extrinsic or relational predications whose truth depends partly upon creatures.

Once this is in place, we can see where the route to modal collapse itself collapses. “God creates this world” (or “God knows this world,” “God intends this world,” or whatever else) is indeed a true predication of God, but its truth includes an extrinsic reference to a reality beyond God—namely, the contingent world in question. Likewise, “God’s act of creating this world,” insofar as it includes reference to this particular effect, is an extrinsic—relational or Cambridge—predication. It therefore does not name some additional intrinsic reality in God with which God must be identical. Thus, had God not created this world, there need not have been any corresponding intrinsic difference in God. (This does, of course, require the classical theist to deny that every difference in effect requires some corresponding intrinsic difference in the cause. But there are good reasons—on libertarian-freedom grounds and elsewhere—to reject that principle anyway.)

Once all this is understood, we can see why the necessity of God does not transfer to the contingent world merely because “God creates this world” is truly predicated of him. If by “God’s act of creating this world” we mean an act specified partly by its contingent effect, then the predication refers to something beyond God and is therefore extrinsic—not something with which God must be identical. If, on the other hand, by “God’s act of creation” we mean simply the intrinsic act in virtue of which creation comes about, then yes, that act is identical with God. But it no longer includes reference to this world—or to anything outside God—and thus poses no threat of modal collapse. That same intrinsic act could have had a different contingent terminus, or no created terminus at all.

For more on this style of objection, see the following guest article by Tim Pawl.

Conclusion

We have considered some of the most common and strongest objections to Aquinas’s approach to God. None of them succeeds. At most, they show that the real debate is perhaps to be had at the level of the background metaphysics itself.

Is the system Aquinas uses to run his argument for God well motivated? This is an important question—one that should be asked and answered. Obviously, I think it is. And perhaps, in a future article, I can tell you how I like to defend it.

Nevertheless, for our purposes, what should be clear is that, once that system is in place, the resources for handling the sorts of objections we have been discussing are already firmly in place as well. Many of these objections, as it were, simply come too late.