Barry Miller’s Cosmological Argument for God: No PSR, No Problem

Barry Miller once published an extremely creative—and, I think, convincing—argument for the existence of God.

Since enough people have asked—and because I now have an article forthcoming in Nova et Vetera on the subject—I thought I’d finally get around to explaining what the Millerian cosmological argument is. For those interested in the full article, I’ve just received permission to upload the pre-print, so I’ll announce it here once it’s available. For now, this piece will simply provide a general overview of what this argument is about and how it differs from some of the traditional philosophical approaches to God.

OK, so. The Millerian cosmological argument—a title I’m proud to have officially coined (super original, I know)—is based on the work of Fr. Barry Miller, whose work I first became interested in when researching the doctrine of divine simplicity. Several people I hold in high regard told me that Miller was a genius on the matter, ably defending the doctrine against the toughest objections and miles ahead of the debate back in the ’90s—all of which I quickly found to be extremely true. Miller immediately struck me as a genius—a very quirky genius, but definitely someone who could just see the world metaphysically and somehow find deeply satisfying solutions to otherwise hugely pesky problems. I devoured everything of his that I could find, including his lifelong work on contingency argumentation. From what I can tell, Miller developed his argument for God over decades, so that by the time From Existence to God was published, it was not only highly refined but ready to take on the most pressing objections from all corners of philosophy.

Now, while everyone I know who’s read Miller consistently praises him—and while he received a good amount of engagement while actively publishing—I was astonished to see just how little attention his work continues to receive. Only one book has really been published on his work—a relatively slim but exceptional volume by Elmar Kremer. Beyond that, his content, though seemingly highly respected by anyone familiar with it—friend or adversary—has sunk into relative obscurity. I found this a terrible shame, of course, but I also suspect it has a lot to do with his quirky analytic style. Miller is a technician, and his approach is unforgiving. He spends little time holding the reader’s hand, expects you to be up to speed on all sorts of deeply nuanced issues, and often makes unusual terminological choices. All of this makes him somewhat inaccessible, even to professional philosophers. Additionally, because he blends both Thomistic and analytic traditions, he’s just… different. He doesn’t follow any familiar path that might make him more accessible and predictable. Instead, he really goes his own way.

For instance, in From Existence to God, he develops his own theory of property instances, motivated largely through the philosophy of language and the idea that language mirrors ontology—all controversial stuff, but positions he defends ably. Still, I don’t think many readers were willing to go along with him on all these points, which he requires to run his argument for God’s existence. While I was willing to follow Miller on these issues, my contribution to developing his thought was to deploy it within a more general Thomistic ontology, which doesn’t rely on Miller’s more controversial positions in the philosophy of language. (Of course, Thomistic ontology is itself controversial, but what can I say? I think it’s true!)

Anyway, let’s turn to the argument—really, all this is just to say that I think you should read Miller. He’s definitely an investment, but once you come to terms with his work, the payoff is enormous. In this sense, I think he’s quite a bit like Bernard Lonergan—tough to get into, but once you do, there’s so much you’ll pull out.

Perhaps the best way to understand the Millerian cosmological argument is by contrasting it with certain contemporary cosmological arguments (see the post linked at the bottom for a primer on cosmological arguments in general). Right away, the major difference is that the Millerian argument requires no upfront causal or explanatory principle, like the principle of sufficient reason. Why is this? Because the Millerian argument aims to show that familiar, everyday particulars are actually contradictory structures when considered in se (Miller demonstrates this through his theory of property instances, while I show it through Aquinas’s constituent ontology of essence and existence) and, therefore, cannot possibly exist—unless… Unless what? Unless they are always and everywhere referred to an extrinsic cause. (In technical terms, for Miller, there is no non-contradictory, non-elliptical construal of atomic sentences of the form “A exists.” Hey, somebody should put that on a t-shirt!)

Drawing out the contradictory implication is subtle, and I’ll leave the details to my article, but it essentially involves a problem of mutual simultaneous construction between (to use a Fregean expression) incomplete entities from which everyday objects are metaphysically constituted, ultimately producing the (absurd) result that certain objects must be composed of parts yet cannot be composed of parts.¹ It’s technical, but I think it’s demonstrable—again, you can read the article for details once it’s available.

The advantage of this approach, of course, is that it doesn’t rely on controversial explanatory principles. While I believe the principle of sufficient reason is true (I defend it in my book), I’ve always been interested in seeing if there were cosmological arguments that could genuinely run without it—and I believe this is one. This means, of course, that “brute facts”—certain things just existing for no rhyme or reason—are no longer an option, as contradictory structures cannot exist at all, even “brutely.” (In fact, the Millerian argument could itself be used as an argument for PSR—pretty neat, I think!)

Another major advantage is that the Miller argument can proceed from just one familiar, everyday particular—like that bunny rabbit in your backyard. It doesn’t require any aggregation strategy or plural reference. You don’t have to smush a bunch of facts together, in other words, or appeal to some cause or explanation for the universe as a whole (which itself invites debates over whether the universe is One Big Object or merely a conceptual aggregation). While I think those moves are perfectly legitimate, they’re common points of skeptical pushback, so if they can be avoided entirely, all the better. And the Miller argument does successfully avoid them.

The final advantage relates to issues of regress, and here again, the Millerian argument offers something powerful. By showing that familiar, everyday objects are deeply dependent, it places them within what Aquinas would call an essentially ordered causal series—or what contemporary philosophers refer to as metaphysical grounding. As I argue in my article Is Grounding Essentially Ordered Causation?, a causal series like this is necessarily either terminating or non-explanatory. This ultimately means that if it’s a causal series at all (since all causality must be explanatory to some extent), it must necessarily terminate.

Naturally, I contend that this is a causal series (undeniably so) and therefore must necessarily terminate; however, it cannot terminate in the same sort of thing that produced the contradictory result, or else the problem just re-emerges. Ultimately, it must terminate in something absolutely, ontologically simple—in my development, in a being whose essence just is its existence: God.

I take it these are three quite significant advantages, as anyone familiar with the history and current literature on cosmological reasoning will surely admit. But, of course, with pros there are always cons, aren’t there? So, what are the drawbacks of the Millerian cosmological argument?

For my development, the main drawback is that it requires a broadly Aristotelian-Thomistic constituent ontology to work. I suspect this will be the biggest point of pushback—and indeed, that’s precisely where I’d push back if I were playing skeptic myself. That said, I find the Thomistic metaphysical picture to be extremely theoretically fruitful—I think it’s true, as you know—and it’s been quite ably defended in recent years by some brilliant thinkers, so I’m not too concerned about this commitment.

The other issues with Miller’s approach are ones I sought to avoid in my development. As already hinted, philosophers who engaged with his work while Miller was alive challenged his commitment to the idea that language mirrors ontology (for example, that predicate and subject truly mirror property and object)—a dusty mirror, Miller would admit, but the only mirror we have. Miller defends this position, but it seemed to be a sticking point where those who engaged with his work often diverged from him. My argument does not rely on this assumption, nor do I think it’s required to motivate the ontology from which my argument proceeds.

1. Metaphysical parts, that is.