Responding to Feser: The Pitfalls and Prospects of Worldview Comparison for Theistic Philosophy

In contemporary philosophy of religion, the approach of comparing worldviews involves juxtaposing multiple comprehensive “theories of everything” to assess their respective strength, with a particular focus on explanatory comprehensiveness and simplicity. My book “The Best Argument for God” integrates this comparative approach between worldviews with classical metaphysical proofs of God.¹

Dr. Edward Feser, author of “Five Proofs of the Existence of God”, has recently criticized the worldview comparison approach by suggesting it is too unambitious and can lead to erroneous notions of God. Feser’s criticism on this front is worth quoting in full:

On the other hand, we have arguments that proceed a posteriori, but are way too unambitious in their conclusions.  This would include, for example, arguments that treat God’s existence as at best the most probable “hypothesis” among others that might account for such-and-such empirical evidence, or even fail to get to God, strictly speaking, as opposed to a “designer” of some possibly finite sort.  And it would include arguments for survival of death that put the primary emphasis on out-of-body experiences and other phenomena that can at best render a probabilistic judgment.

Thomists tend to put little or no stock in such “god of the gaps” and “soul of the gaps” arguments. At best they are distractions from the more powerful arguments of traditional metaphysics, and thus can make the grounds for natural theology seem weaker than they really are. At worst, they can promote serious misunderstandings of the nature of the soul, of God, and of his relationship to the world.  (For example, they can give the impression that it is at least possible in principle that the world might exist without God, which entails deism at best rather than theism. And they can give the impression that the disembodied soul is a kind of spatially located or even ghost-like thing.)

While Feser definitely raises some valid points, his critique ultimately overlooks the potential of worldview comparison for bolstering the case for classical theism when properly situated. Worldview comparison need not forsake proper metaphysics. In fact, by blending methodologies one can avoid the pitfalls Feser identifies while presenting an overall more convincing case for classical theism.

For example, in “The Best Argument for God” I first deploy a traditional philosophical inquiry into the existence and nature of God, one that is Thomistic in spirit, though not actually any one of Thomas’s arguments exactly. My approach utilizes what I term ‘minimum effective metaphysics’ — that is, defending just enough metaphysical background to launch the argument for God and to deduce the divine attributes. This includes commitments like the act-potency and essence-existence distinction, commitments that Feser defends throughout his work, as well. By way of explanatory principles, I propose a restricted version of the principle of sufficient reason; this, too, is in alignment with Feser.

It is only after I work the traditional approach to God that I aim to further confirm classical theism by bringing it into worldview comparison — that is, by showing how fruitful this particular theistic paradigm is for explaining many other large-scale features of the world, from consciousness to morality and more. Naturally, some may question the necessity of this subsequent step; I suspect Feser himself might wonder about this. After all, didn’t I just argue successfully for the existence of God? Well, I certainly think that I did, but my stance on this issue remains pragmatic, recognizing that no argument, regardless of its strength, will definitively resolve the debate for everyone. As we’ll see in a moment, there are always escape hatches for the skeptic somewhere, however difficult they may seem to squeeze through. Instead of spilling enormous ink attempting to seal off every possible egress, a more powerful and persuasive case, I believe, can be issued through a cumulative effort, hence the move toward worldview comparison after the demonstration has been presented.

That is what my book is up to, anyway. The master argument being that naturalism can explain some but not all of what classical theism can but only when strapped with greater complexity. This thesis allows me to present many arguments relevant to important explanatory targets, all in support of classical theism.

Before modestly defending worldview comparison, I want to highlight where Feser’s criticisms hit their intended target. Firstly, the “hypothesis testing” approach to God almost always invites issues when starting with stipulative understandings of the nature of God or the divine attributes, such as omnipotence or omniscience; that is, without any metaphysical backdrop from which these understandings are inferred, as well as conceptually clarified and constrained. These notions are often just shaped from air, it would seem. No surprise, then, that classical theists, who characterize divine attributes post rigorous metaphysical analysis, frequently find portrayals of God’s omnipotence or omniscience to be defective or misleading among those engaged in worldview comparison. For instance, describing omnipotence merely as the ability to perform any logically possible action — essentially, that God can do anything, because that just seems like what omnipotence might be – immediately runs into issues. Can this omnipotent God swim in a lake of cheese, or commit adultery? If not, then it seems there are logically possible actions God cannot do, and so God is not omnipotent. Now, in fairness, it is worth acknowledging that Richard Swinburne, perhaps the most influential theist making use of worldview comparison, recognizes these issues and has worked tirelessly to refine his characterization of the divine attributes to avoid difficulties such as these. Nonetheless, these challenges are more readily and perhaps inherently addressed, while at the same time providing a more robust understanding of God’s relation to the world, by the "metaphysics first" method that classical theists recommend.

So for instance, the classical theist arrives at omnipotence by first deducing a purely actual reality whose essence just is its existence through metaphysical effect-to-cause reasoning. Their understanding of omnipotence is shaped through the deduction that anything that actually exists, or even could exist, exists only insofar as this fundamental reality is imparting existence to it. Thus, for the classical theist, omnipotence relates to production (i.e., of God being able to produce whatever is inherently producible), not just any old logically possible action. This restricted but non-contrived understanding of omnipotence is less vulnerable to paradox and counterexample (as more restricted notions tend to be) like God not being able to commit adultery. Interestingly enough, Swinburne ultimately gravitates toward the idea of logically possible productions for an understanding of omnipotence himself. The problem is these restrictions to omnipotence can come off as ad-hoc – that is, as mere ways of avoiding difficulty or paradox – if they are not the result of some prior metaphysical understanding of God or the world, as classical theists claim is the case with their account.

The characterization God’s omniscience is also often unhelpful in the worldview comparison approach, with many understanding it as God knowing whatever is logically possible to know, a trait seen as an extension of omnipotence in many cases. While perhaps less misleading than the previous example, the classical theist has issues with this articulation as well, if not so much for what it includes, but for what it leaves out. For example, it seems, by itself, to leave as a matter of brute fact how God knows whatever it is logically possible to know. Does God have some sort of infallible “organ” through which he perfectly scans and comprehends all states of affairs? Moreover, it does not, by itself, suggest any difference in God’s mode of knowledge. Thus, many just take omnipotence to mean that God has a much higher degree of propositional knowledge than us, that God is somehow maxed out with respect to whatever true propositions can be known.

By contrast, the classical theist approaches omniscience through an amalgam of prior metaphysical commitments put to use, of proportionate causality, of essentialism, of the immaterial nature of understanding, and of intentional action. In truncated form: Whatever is in the effect must in some way be contained in the total cause, the essences or forms of things are effects of this world to be explained, and since God qua ultimate cause is immaterial, these forms cannot be contained in God materially, which means they must be contained in an analogously intentional way, which is to say, as *something like* ideas. Moreover, God knows all created essences and states of affairs not by watching, but by bringing them about. God’s knowledge of this world is executive; God knows by doing. Finally, for the classical theist, God’s knowledge is simple, non-propositional, and different from ours not just in degree, but in kind. Whereas all our knowledge is, ultimately, passive, God’s knowledge is entirely active.² Hence why philosopher Barry Miller appropriately classifies God as the limit case instance of knowledge, a being in whom complexity, including that implied by propositional or discursive thought, cannot apply. I am skipping details, of course, but the point for now is just that this understanding of omniscience, as classical theists argue, is entailed by the background metaphysics and subsequent demonstrations producing the result of an absolutely ontologically simple, purely actual reality.

So far, I hope to have provided examples to illustrate the problem Feser is getting at. Problem that, from the classical theistic perspective, are real.

Where, then, do we disagree? On several fronts.

First, Feser’s labelling of the worldview comparison approach as “god of the gaps” reasoning is definitely a point of contention, as I view the approach as a legitimate inference to the best explanation rather than an erroneous argument from ignorance. For there is a clear and significant difference between the unfounded claim, “This is unexplained; therefore, God must be responsible,” and the methodical construction of a theistic hypothesis issued alongside reasons why that hypothesis anticipates certain large-scale features of the world (contingency, consciousness, etc.) better than, say, a naturalistic hypothesis of indifference would. To dismiss this approach as “god of the gaps” is unjust, an action, it should be noted, not typically taken by even the most astute atheistic critics working in the philosophy of religion. Such skeptics often acknowledge the method’s validity but simply contend that the god hypothesis fails compared to the naturalistic hypothesis concerning points of explanation, simplicity, or both.

Next, while I am personally convinced of (several of) the traditional metaphysical demonstrations for the existence of God, it is obviously the case that not everybody is convinced. Why? Well, even though these demonstrations launch from commonsense starting points (i.e., change, contingency, compositeness, etc.), these foundations are not, you know, completely unassailable, nor the metaphysical system meant to explain these features. People indeed deny that change, contingency, or compositeness, etc., are facts of the world, let alone facts standing in need of explanation. The cost of such denials may be high – in some cases, absurdly so – but people are willing to embrace these costs at times, and it would seem some do so for reasons unrelated to the existence of God. For anybody familiar with these debates, this all relates to the “Well, your arguments assume a controversial metaphysics!” objection. While I feel this objection is often overstated and lazily deployed — that is, as means for merely dismissing the argument rather than engaging with it — it is nevertheless frequently made and not totally illegitimate.

Further, it is pertinent to note that traditional arguments for God’s existence, particularly those presented in Feser’s “Five Proofs,” hinge upon some variant of the Principle of Sufficient Reason (PSR). Feser defends his preferred formulation of the PSR in the fifth chapter of his book, in development of the Leibnizian argument. However, despite the distinct starting points of his other arguments, they share a common theme: the definite need to discount brute facts. Said differently, all Feser’s arguments require acceptance of the PSR for their ultimate success. Thus, if one rejects the PSR, even of the more restricted Thomistic sort, one can reject all the arguments Feser advances; that is, they can just take aboard the idea that things reducing from potency to act is a matter of brute fact, or that composite entities exhibiting mutual interdependencies is a matter of brute fact, etc. Of course, I think a PSR of sufficient strength to range over all such phenomena is true, and I mount a cumulative case in support of this thesis in my book. My intention is not to defend the PSR here. Rather, I’m just highlighting a critical juncture where the skeptic can, and frequently does, object; a single point where suspicion can be focused to opt out of the full suite of Feser’s arguments.

My, I guess, bigger, point is this. Does one want to have their entire case for God rest upon any one commitment that, while commonsense for many, is not completely immune to reasonable skepticism?³ For Feser, does he want to have philosophically motivated belief in God effectively come down to accepting PSR? Even if one believes, as I do, that PSR is robustly — indeed, self-evidently — true, surely, they must acknowledge that not everybody thinks this. As it happens, I am aware of intelligent skeptics that happily admit if one accepts the PSR, they should be a theist. But they do not. Why put all your natural theological eggs in one argumentative basket?⁴ Why not advance the sort of argument that Feser does, then supplement it with further considerations from the worldview comparison method to bolster confidence?

Feser wants to issue the most convincing – if possible, compelling – case for classical theism. Me too! Hence, my desire to synthesize approaches to natural theology, rather than divorce them and forcing people to pick just one.  

To my mind, a productive marriage is entirely possible between these philosophical approaches to God. Simply run the traditional metaphysical arguments for God to articulate the classical theistic conception of God and God’s relation to the world. Let the traditional demonstration form the “God hypothesis” while serving as a rigorous argument in itself. Then, bring it into the worldview comparison game, to show just how fruitful classical theism is across the wide range of explanatory targets – from consciousness to rationality, morality, even suffering and evil. All the better if one brings along other old-school metaphysical commitments, like the diffusiveness principle, theory of the transcendentals, etc., to ramp up expectations concerning the type of world God would create while providing a suite of resources from which classical theism can robustly explain various dimensions of reality, i.e., the moral dimension. Finally, one can argue, as I do at length in my book, that the classical theistic theory is impressively virtuous along the most relevant understandings of simplicity (ontological, theoretical, etc.); in fact, unbeatably so.

The advantages of natural theological procedure— that is, leading with the traditional approach and confirming with the contemporary — seem obvious to me, particularly with respect to persuasive force. For those wavering on metaphysical proofs for whatever reason, including the comparative approach can be what tips the scales and builds confidence that God – as classical theists understand God – exists.

While I agree with Feser on many issues philosophic (unsurprising, given that we are both Catholic classical theists), this is one area where we diverge. Indeed, it is the synthesis of traditional metaphysical arguments with the worldview comparison approach that creates the most compelling, multifaceted case for classical theism, one that transcends the inherent limitations of each. To neglect this synthesis is to present a classical theistic argument that, though still formidable, fails to harness the full potential of philosophical inquiry available to the theists.

- Pat

PS - In case you haven’t seen this yet.

1. I should note that the sort of worldview comparison I prefer is one that focuses on large-scale features of reality that everybody has access to; not so much the findings of contemporary science. This preference is not just based on accessibility to the data to be explained but the inherent stability thereof. For it is seriously unlikely that large-scale features like there is stability and order or there is consciousness or there is suffering are going to be “overturned” anytime soon, whereas the findings of contemporary science always could be. Admittedly, I violate this preference when discussing physical fine-tuning in my book, but that section is embedded within a larger context arguing that God better explains the general order and stability of the world, which everyone is aware of.

2. One might think some of our knowledge, like how I know that I am currently tying, is active, and that would be right. However, that knowledge depends on knowing I have hands, which was something I came to learn. Thus, even our active knowledge, upon final analysis, depends upon something we came to know previously, but passively.

3. Note: If one wants to say it is not reasonable to at least question the PSR, they not only have intelligent atheists to deal with but intelligent theists as well, the likes of which include James Ross. Hardly a goon!

4. Of course, I maintain that making legitimate use of inference to the best explanation would itself seem to require a commitment to the PSR. But that’s a conversation for another time.