Classical Theism, First Causes, and Naturalism: A Response to Enric Gel

Recently, Enric F. Gel published a response to my response to his original paper in Religious Studies. I thank Enric wholeheartedly for his (gloriously meme-filled) response, and I also thank Enric for being the awesome person he is. (I’ve really enjoyed our public and private interactions!) In this post, I’ll be responding to Enric’s most recent response.

Some notes before we begin:

• Hereafter—in accord with academic convention—I’ll refer to Enric as “Gel”.
• My responses will follow the order of Gel’s paper, with sections corresponding to Gel’s sections. The titles of my sections will therefore be along the lines of “Gel, “[insert Gel’s section title]””
  • In essence, I’ll be responding to Gel’s paper as it proceeds.
    • I’ll be skipping the first section “In previous papers…”, since it simply summarizes the thus-far-trodden dialectic and sets the stage for Gel’s article.
    • Those not up to date on the dialectic are advised to read the relevant papers; though, absent that, Gel’s section “In previous papers…” will get you reasonably caught up.
• Key terms: a list of the terms I’ll use to abbreviate the relevant papers in the dialectic
  • Gel’s original paper in Religious Studies, “How many and why? A question for Graham Oppy that classical theism can answer” = Gel’s original piece
  • My response in Religious Studies, “Naturalism, classical theism, and first causes” = my original response piece
  • Gel’s rejoinder to my response in Religious Studies, “‘There can only be one.’ A Response to Joseph C. Schmid” = Gel’s recent piece
• Footnotes can be found at the end of each (sub)section in which they appear in sections called “Section footnotes“. I recommend reading the footnotes, as they contain important information! 🙂

Without further ado, let’s dig in!

---

Gel, “Defending an argument for God’s unicity”

Gel, “God’s unicity compromised? IoI” — Or, premise (1) of the argument

The dialectic here is focused on premise (1) of Gel’s argument (as I formulated it in my original response piece), which is an application of the Identity of Indiscernibles (IoI):

I made two points in criticizing (1). The first point was that Gel simply assumed IoI (and hence premise (1)) without any justification. The second point was that a primary motivation of IoI—explicability—doesn’t work. Gel responds to each of these points in turn.

Beginning with the first point, Gel offers in his recent piece a new argument for IoI.[1] It is entirely fair and appropriate to include a new argument for IoI at this juncture in the dialectic. We should at least note, however, that Gel’s argument in his original piece was underdeveloped and unjustified on account of assuming IoI, and so the central point in my original response to Gel’s argument stands. Alas, with an argument for IoI on the table, let’s evaluate it.

Gel writes that “without IoI, our ontology runs the risk of getting chaotically overcrowded very quickly – or at least the possibility of this should force us to remain agnostic as to the number of ordinary objects we encounter in everyday experience.”

But I don’t think this is right. We can do perfectly well without IoI by simply using the standard theoretical virtues here. In particular, when presented with competing hypotheses—say, about the number of ordinary objects we encounter in everyday experience—we can assess these hypotheses on account of their simplicity and explanatory power. If one hypothesis clearly outcompetes the others along these dimensions, we accept that hypothesis. And, indeed, the hypothesis that minimizes the chaotic overcrowding Gel speaks of clearly outcompetes the others along these dimensions, as I will explain below. We can thus make perfect do without IoI. There is no risk of chaotic overcrowding or of agnosticism about the number of ordinary objects we encounter. Standard theory comparison assuages these threats.

Gel continues: “For instance, I have one pencil on my desk. But if I allow it is possible that, were I see one pencil, there are in fact two distinct indiscernible pencils, I’m not sure I can continue to be confident that there is only one pencil on my desk. Consider also that, presumably, if it is possible for there to be two distinct indiscernible objects, it is also possible for there to be three, four, ten, or a million of them. Hence, without IoI or some principle like IoI, we would constantly be in the dark as to how many objects we encounter in everyday experience.”

Again, I don’t think this is right. We can do perfectly well without IoI here by simply applying the standard theoretical virtues to hypotheses about the number of pencils on the table. The hypothesis that there’s only one pencil on the table far outcompetes the other hypotheses in terms of theoretical virtues. First, the hypothesis H₁ that there’s just one pencil explains all the relevant data—the slight forces on the table and the rest of the physical objects in your vicinity, your perceptual experiences as of at least one pencil, and so on. So, among the hypotheses about the number of pencils on your table, H₁ is unsurpassed in its explanatory power. Second, the other hypotheses introduce intolerable explanatory costs in the mystery engendered by their postulation of precise, arbitrary numbers of pencils.[2] So, H₁ is not only unsurpassed in its explanatory power but far surpasses the other hypotheses therein. Third, H₁ is far simpler than the other hypotheses because it minimizes ontological commitment. So, H₁ far surpasses other hypotheses in both explanatory power and simplicity—in which case, we have excellent reason to accept H₁.[3]

Notice that none of this makes use of IoI or any principle like IoI, and notice, further, that the reasoning extends to hypotheses about basically any other object in our ordinary experience. Hence, Gel is mistaken in saying that without IoI or something very much like it, we would be in the dark about how many objects we ordinarily encounter.

Now, one might object that we can simply run essentially the same reasoning for classical theism. In particular, granting that there’s a being of pure esse, wouldn’t the hypothesis that there’s only one such being far outcompete the other hypotheses about their number? Yes! (Or, at least, probably!) But, first, this isn’t an explanation for why there is only one being of pure esse. What we have here is simply good epistemic reason to think that there is only one. Second, the Oppyan naturalist can make exactly the same move—i.e., the Oppyan naturalist can hold that there’s only one first cause on precisely the same grounds of theoretical virtue. The classical theist therefore doesn’t enjoy an advantage over the Oppyan naturalist here.

Another, secondary response to Gel’s point is that we might have good reason for thinking that co-location of ordinary material objects is impossible. Lots of philosophers defend this on grounds independent of IoI.

These responses also take care of Gel’s ensuing discussion involving pine trees and whatnot.

Let’s continue, then, with Gel’s point about the identity of necessary indiscernibles (IonI):

There are at least three things to say in response. First, even if the argument is tweaked in this manner, the central problem remains: the argument assumes a principle without adequate justification, and hence the argument fails to establish its conclusion. Of course, Gel might point to his preceding argument for IoI in support of IonI; but as I’ve explained, that argument fails. (The argument fails by dint of ignoring other ways (e.g., standard theory comparison) to rule out scenarios involving arbitrary numbers of ordinary objects around us. The same is true when we consider not just contingently indiscernible objects but also necessarily indiscernible objects—there are other ways (e.g., standard theory comparison) to rule out scenarios involving chaotic, arbitrary swathes of such objects.)

Second, we should keep in mind that the onus in the dialectical context at hand is on Gel to justify IonI, since Gel is the one offering an argument one of whose premises assumes IonI. Thus, even if there are no reasons to question IonI, what matters for present purposes is whether Gel has offered good reasons to accept IonI. And, as I’ve explained, he hasn’t.

Third, Gel says he can think of no reason to question IonI. But I can think of several. I’ll briefly give one for purposes of space: the existence of two necessarily indiscernible things at least seems coherently conceivable. And if conceivability is defeasible evidence for possibility (as most philosophers seem to think), then—since IonI entails that it is impossible for there to be two necessarily indiscernible things—we have defeasible reason to think IonI is false.

Let’s now consider Gel’s response to the second point in my criticism of premise (1)—namely, that a primary motivation for IoI, namely explicability, fails. Keep in mind that even if Gel successfully meets this second point, my first point remains unscathed: Gel’s argument rests on an unjustified assumption.

Now, Gel construes my second point as “Schmid’s objection to IoI”, but this could use more precision. My objection targets not IoI but a central reason—explicability—leveled on behalf of IoI. For explication of the objection, see either my original response piece or the quotation thereof in Gel’s recent piece. Gel has two responses to my objection, and I’ll consider each in turn.

Gel’s first response to my objection runs:

I don’t think this is right. I have two responses.

First, Gel is here resisting the need for individuating properties that explain why the properties rectangularity and circularity are distinct. And his resistance, it seems, is based on the fact that the properties in question are self-evidently distinct and hence don’t need a further explanation for their distinction. The problem with this rejoinder, though, is that substances are also self-evidently distinct from one another. To see this, notice that p’s self-evidence to subject S requires that S grasps the natures of the things reported in p. (For instance, that Euclidean triangles have interior angles summing to 180 degrees is not self-evident to my 2-year-old nephew, since he lacks the relevant knowledge about the essences of triangles, degrees, etc. But once one grasps the relevant natures or essences, one thereby sees the relevant truth. This is part of what it is to be self-evident.) So, if S doesn’t grasp the natures of rectangularity and circularity, their distinction won’t be self-evident to S. Gel’s response therefore requires a background grasp of the natures of the realities in question. But here’s the rub: with a background grasp of the natures of indiscernible substances A and B, it is also self-evident that A and B are distinct. For part of the nature of any entity is that it is that particular entity, distinct from other particular entities.[4] Hence, upon grasping the nature of A (say), we would immediately see that it’s distinct from B. But in that case, Gel’s rejoinder here does more harm than good to his case, for his rejoinder also implies that there needn’t be any further explanation for the distinction between substances—in which case, my criticism of the explicability-based motivation for IoI is vindicated.

Second, I’m not sure how Gel is using ‘indiscernibility’ in this passage. As it’s used in the context of the explicability-motivation for IoI (according to which the properties that individuate substances explain why those substances are distinct), x and y are discernible iff and because it is not the case that the set of x’s properties is identical to the set of y’s properties—i.e., iff and because there’s some property that one has which the other lacks. So in saying that rectangularity and circularity are not indiscernible—i.e., that they are discernible—Gel seems to be saying that there are further individuating properties in virtue of which rectangularity and circularity are distinct. But then he’s contradicting his earlier claim in the passage that the explanation for their distinction is not something else. Now, this apparent contradiction gives us reason to think Gel is not using “indiscernibility” in the aforementioned sense. But then he should be clarifying what he means, firstly, and secondly, it seems dialectically improper to use a crucial term differently from how it’s used in the context of the objection at hand.

Here’s Gel’s second response to my objection:

We should clarify at the outset that my objection was not that we need to accept primitive individuation at any and all levels.[5] My objection was simply that primitive individuation seems unavoidable at some level. And in that case, the explicability motivation for IoI—which seeks to avoid primitive individuation altogether—is mistaken. Now, in response to this point, Gel suggests restricting IoI to substances. Setting aside the seemingly ad hoc nature of this restriction, my criticism still stands in another guise: namely, it provides an undercutting defeater for the explicability-based motivation for the restricted IoI. Once we grant that there is primitive individuation (even among properties), we seem to lose any compelling explicability-based reason for thinking that primitive individuation among substances is impossible. (What’s the relevant difference between substances and properties in virtue of which the former, but not the latter, cannot be primitively individuated? For me, no answer presents itself upon reflection.) Of course, one could appeal to some other reason, apart from the explicability-based one I canvassed in my paper; but then one is no longer responding to my objection, since my objection simply concerned the explicability-based motivation for IoI.

Another problem for the restriction of IoI to substances is that the same motivations for thinking IoI holds of substances also motivate thinking IoI holds of properties. Notice that we can run exactly the same ‘chaotic overcrowding’ argument that Gel runs, but this time in terms of the chaotic overcrowding of properties—parodying Gel, “without IoI as applied to properties, our ontology runs the risk of getting chaotically overcrowded with properties very quickly—or at least the possibility of this should force us to remain agnostic as to the number of ordinary properties we encounter in everyday experience.” We can also run the same skeptical scenarios concerning the number of properties we ordinarily encounter as Gel runs concerning the number of objects we ordinarily encounter, etc. So the restriction strategy is inherently unstable—the reasons for thinking IoI holds of substances equally motivate thinking IoI likewise holds of properties. So, if we should accept IoI as applied to substances, we should likewise accept IoI as applied to properties. But as I explained in my original response piece, we should not accept IoI as applied to properties, since that yields a vicious infinite grounding regress. Hence, we should not accept IoI as applied to substances.

Gel also points here to his earlier ‘compelling reasons’ for accepting IoI. But as I’ve explained, Gel’s proffered reasons aren’t ‘compelling’ in the slightest.

I conclude, then, that Gel has failed to assuage the two worries I raised surrounding IoI in my original response piece. Both of my original criticisms stand. Let’s now turn to Gel’s next subsection.

---

Section footnotes

[1] By ‘new’, I mean ‘not in Gel’s original piece’.

[2] In addition to the arbitrary and unexplained precise number of pencils postulated, there is a lot more mystery that such hypotheses engender—i.e., they raise countless unanswered explanatory questions: Where did all those additional pencils come from? Why are they all perfectly co-located, and how are they able to remain perfectly co-located throughout their various changes and interactions with the environment? Why isn’t there interference between them? And so on ad nauseam. (These questions aren’t, of course, unanswerable; but adding answers thereto will drastically increase the complexity of such hypotheses.)

[3] But aren’t there many more ways for there to be more than one pencil and only one way for there to be a single pencil, as Gel notes in his footnote 5? Yes, but those ways are so metaphysically profligate, arbitrary, and unexplanatory that even their disjunction fails to compete with the single pencil hypothesis. What matters isn’t just the number of ways but how much of the probability space is accorded to them (both individually and collectively). (Also: if the fact that there were many more ways for there to be more than one pencil meant that the hypothesis that there’s more than one pencil swamps the single pencil hypothesis in terms of intrinsic probability, this would be disastrous for theism’s intrinsic probability, since there are infinitely many ways for the first cause’s properties to fail to be maximal and only one way for them to be maximal (and so theistic). Consider, e.g., that there’s only one way to be omniscient but infinitely many ways to fail to be omniscient—one could know every fact but one, or but two, or but three, or etc. Similarly for omnipotence: there’s only one way for a first cause to be omnipotent but infinitely many ways to be impotent in some manner—one could have the power to do everything except one task, or two tasks, or three tasks, etc. Mutatis mutandis for other candidate properties of the first cause.) Also, note that we could make all of my points in this section in terms of Bayesianism rather than theoretical virtues. I’ve simply chosen the theoretical virtues construal so as to avoid technicalities that quickly lose readers.

[4] This doesn’t imply that any distinct substances are discernible in the sense required for IoI, since such essential truths (about being that particular substance) needn’t correspond to properties of substances.

[5] I’m not claiming that Gel attributes this claim to me; I’m simply clarifying my objection. I make this clarification because someone might think, upon reading Gel’s response here, that my objection rests on this claim. I’m simply clarifying that it doesn’t! 🙂

---

Gel, “Distinguishing beings of pure esse” — Or, premise (2) of the argument

In this section, Gel discusses my three criticisms of premise (2) of his argument. Recall the argument as formulated in my original response piece:

My three criticisms of premise (2) were:

• (a) Gel’s proffered justification for premise (2) fails;
• (b) There are candidate differentiating features that might obtain between beings of pure esse; and
• (c) Gel’s proffered justification for premise (2) overlooks the distinction between God’s being identical to existence as such and God’s being identical to his own act of existence.

Let’s proceed through Gel’s rejoinders to each of these criticisms.

Criticism (a): Gel’s proffered justification for premise (2) fails

At the outset, Gel seems to grant that his original justification for premise (2) is sketchy at best, as he grants the fairness of my criticism concerning his ‘outside of’ locution.

Gel notes (in footnote 13) that he was mainly relying on Kerr’s formulation of the De Ente argument. He also cites Kerr’s principle that whatever is distinct from esse tantum is either (i) subject to esse tantum or (ii) nothing. It’s at least worth noting, though, that what Gel said doesn’t bear much resemblance to Kerr’s principle (e.g., Gel said nothing about being subject to esse tantum; Gel talked about being “outside of” rather than distinct from, and if we replace Gel’s “outside of” with “distinct from”, his claim is false; etc.). It’s also worth noting that Gel did not specify which portions of his argument were from Kerr. In particular, he did not cite Kerr when talking about this principle in his original piece. In any case, the important point of this paragraph is that Kerr’s principle won’t help Gel’s justification for premise (2). For if there were more than one esse tantum or being of pure esse, this principle implies neither that any of these beings would be subject to one another nor that any of them would be nothing.[5.5]

Gel then proceeds to offer a new justification for premise (2).[6] It is entirely fair and appropriate to include a new justification at this juncture in the dialectic. We should at least note, however, that we’re no longer talking about Gel’s justification in his original piece, and so my original response to Gel’s justification—to wit, that it doesn’t work—stands. Alas, with a new justification proffered, let’s evaluate it.

Gel’s offers three closely related new justifications for premise (2) (following the lines defended in Feser (2017, pp. 121–122)). Here’s the first:

Setting aside the lack of justification for premise (1) (as that was the topic of the previous section), there are two problems with this justification.

First, Gel illicitly assumes that A having F and B lacking F entails that A is pure being plus feature F. But this is simply untrue—or, at least, it’s entirely unmotivated. In principle, it could easily be the case that A’s pure esse just is A’s F-ness, while B’s pure esse just is B’s G-ness, where F-ness isn’t G-ness. This is kinda similar to how, for the classical theist, God’s pure esse just is God’s omnipotence. It’s not as though God is pure esse plus omnipotence; God’s pure esse just is his omnipotence. The same can be said in the case of A and F—it’s not as though A is pure being plus F-ness; A (and A’s pure esse) just is A’s F-ness. And B is differentiated from A by dint of B (and B’s pure esse) being identical to G-ness, which is distinct from F-ness. Here, A and B are discernible: F-ness is a feature A has (and, indeed, is identical to) that B lacks, while G-ness here is a feature B has (and, indeed, is identical to) that A lacks, and such features differentiate them.[7] Furthermore, each of A and B is pure esse. But then A’s being pure esse plus feature F doesn’t follow from the facts that A has F, B lacks F, and A and B are each beings of pure esse. I’m not claiming that this sort of scenario is metaphysically possible; my point is simply that Gel needs to rule it out for his argument to succeed, and yet he has done no such thing.

Second, an equally (im)plausible parody argument can be run which shows that any being of pure esse couldn’t be trinitarian. In order for there to be more than one divine person, they’d have to be differentiated by some feature that one has which the others lack; but then that divine person wouldn’t be pure esse but would be pure esse plus that feature. But then the divine person couldn’t be God, since God is pure esse. (Furthermore, the divine person would be a composite of pure esse and that feature, and hence there would be something composite intrinsic to God, which is incompatible with DDS.) (NB: I respond below to what Gel says regarding trinitarian parodies.)

Here’s Gel’s second related new justification:

So, I don’t know what “possessing the fullness of being” even means. Alas, insofar as it means anything, the justification here fails for the same reason as above, since A’s possessing the fullness of being could—for all Gel has shown—partly consist in A’s being identical to its F-ness while B’s possessing the fullness of being could partly consist in B’s being identical to its G-ness. Here, there’s nothing superadded to A’s or B’s fullness of being, and hence Gel is mistaken to say that a differentiating feature between two things that are the “fullness of being” would amount to something superadded to their “fullness of being”.

Here’s Gel’s third related new justification:

But Gel is mistaken in saying that “F could not be an essential property of A, since in such a case B would exhibit F as well”. This illicitly assumes that A and B share the same essence or essential properties. But nowhere does Gel justify this. For all Gel’s argument shows, it could be that there are two beings of pure esse, each of which is identical to its own unshared essence. We need not suppose that they share one and the same essence; they may simply have different essences. Perhaps one of them has an essence included in which is the essential attribute of being trinitarian (or omnipotent, or omniscient, or impassible), whereas the other has an essence included in which is the essential attribute of being unitarian (or quasi-omnipotent, or quasi-omniscient, or diff-impassible—cf. my footnote 7). In this case, each being is identical to its own essence (and, we can suppose, its own existence, in which case each being is a being of pure esse), and yet there is no single essence that the beings share in common.

Now, you might object that for each of them to be divine, or for each of them to be pure esse, they must at least minimally share some essential attribute, such as divinity or pure-esse-hood. But why? Why not instead suppose that they each simply have their own (e.g.) trope of divinity or pure-esse-hood (along lines similar to Dr. Joshua Sijuwade’s classical theistic, trope-theoretic model of God)? Or why not take a nominalist approach to such beings, such that there are no such things as divine properties that any such being possesses (and, a fortiori, no properties they share in common)?[8] Or perhaps each being only possesses divinity or pure-esse-hood analogously to the other being(s), such that we block the need for some essential feature that they share in common but nevertheless preserve our ability to truthfully and literally (albeity analogously) predicate ‘pure esse’ and ‘divine’ of each of them. (This is precisely how classical theists allow us to truthfully and literally predicate (say) ‘power’ and ‘knowledge’ of both God and creatures without committing us to features that God and creatures share in common!) To be sure, Gel might not accept these possibilities; but that’s utterly irrelevant. What matters is whether Gel’s argument for uniqueness has given us any reason to rule each of these out, i.e., to think they’re all false. His argument has done no such thing, and so—as it stands—the argument fails.

I conclude, then, that my first criticism of premise (2) stands. Not only does Gel’s original proffered justification for premise (2) fail (which Gel seems to grant), but his three new justifications for premise (2) succumb to the same fate.

Criticism (b): There are candidate differentiating features that might obtain between beings of pure esse

Gel grants that this is a fair criticism against the original justification for premise (2). He then argues that the criticism doesn’t work against his new (tripartite) justification for premise (2):

There are two problems with this, however. First, the success of Gel’s reply here relies on the success of Gel’s reply to my first criticism (criticism (a)). But as I argued in the previous sub-section, Gel’s reply to my first criticism fails, and hence his reply here fails. Second, in this new context of Gel’s tripartite justification for premise (2), I view criticism (b) as a way of filling out my first point in criticism (a). As I argued therein, the success of Gel’s new (tripartite) justification presupposes the incompatibility of those other properties with being pure esse,[9] and hence the argument cannot be used to establish as much.

I conclude, then, that my second criticism of premise (2) stands. Onward we march to the third criticism.

Criticism (c): Gel’s proffered justification for premise (2) overlooks the distinction between God’s being identical to existence as such and God’s being identical to God’s own act of existence

Gel responds to my third response beginning with the following:[10]

Note, though, that the same can be said with multiple beings of pure esse—their esse’s are different because they are different. And they may be different either primitively or because they are discernible (say, A is identical to its F-ness while B is identical to its G-ness, where F-ness isn’t G-ness). (Be thankful that I’m using ‘F-ness’ and ‘G-ness’ rather than ‘P-ness’!)

Gel continues:

But there’s a rather clear problem with this: Gel blithely asserts that A’s identity to A’s own existence amounts to A’s identity with existence as such. Just read the passage again—carefully, slowly, with a critical eye. There’s no argument for it—it’s just asserted. Gel simply asserts that A’s essence being identical to existence as such follows from A’s being identical to A’s own act of existence. But this is precisely the question at issue. No headway is made here by blithely asserting the very question at issue. I was precisely demanding a justification for this slide from <identical to one’s own act of existence> to <identical to existence as such>. Gel fails to offer any inkling of a justification for this slide.

I conclude, then, that my third criticism of premise (2) stands. Thus, all three of my criticisms of premise (2) stand. Gel’s responses to each criticism fail.

Up next is Gel’s response to my trinitarian parody of his argument for uniqueness.

---

Section footnotes

[5.5] Edits to this paragraph were made on March 10th, 2024 thanks to some clarifications from Enric Gel! 🙂

[6] Again, by ‘new’, I mean ‘not in Gel’s original piece’.

[7] I don’t need specific candidates for such features in order for my objection here to succeed (since the onus is on Gel to rule out even my abstractly characterized scenario). Alas, here are some options. Perhaps A is pure esse and identical to its omnipotence (F-ness), whereas B is pure esse and identical to its quasi-omnipotence (G-ness), where a being is quasi-omnipotent iff it can do do everything an omnipotent being can do except for some trivial task in some remote possible world. Or perhaps A is pure esse and identical to its omniscience, whereas B is pure esse and identical to its quasi-omniscience (G-ness), where a being is quasi-omniscient iff it knows everything an omniscient being knows except for some trivial truth in some remote possible world. Or perhaps A is pure esse and identical to its own impassibility (F-ness), whereas B is pure esse and identical to its diff-impassibility (G-ness), where a being is diff-impassible iff its conscious state slightly differs in qualitative character from an impassible being’s conscious state (which is pure, infinite bliss/joy/happiness/delight—cf. Mullins’ paper here articulating the classical doctrine of impassibility). (We all know that different types and tokens of joy, happiness, and delight have slightly different phenomenological feels, so there’s nothing extravagant here.) Or perhaps F-ness is trinitarianhood and G-ness is unitarianhood. Or etc. Examples can be multiplied ad nauseam for those who want to nitpick about the particularities of the aforementioned examples. Importantly, I don’t see any good reason for thinking that being pure esse is incompatible with (e.g.) quasi-omnipotence, or quasi-omniscience, or diff-impassibility, or unitarianhood, or what have you. (You can’t claim that pure esse would be unlimited, and yet the quasi- properties entail limits. For (trinitarian) Thomists already grant a certain kind of limit in God (to wit, a limit in the number of persons). And besides, this doesn’t apply to diff-impassibility, unitarianhood, and countless other candidates for G-ness.)

[8] This doesn’t require the truth of nominalism. It just requires that God is different from us when it comes to the relationship between entitative character and property possession—something that classical theists already accept (albeit in a different fashion).

[9] Since the compatibility therewith undergirds the criticism that Gel fails to rule out a scenario wherein pure esse A is identical to its F-ness while pure esse B is identical to its G-ness, etc. See the previous sub-section for this.

[10] For an explication of the objection, see either p. 7 of my original response piece or Gel’s quotation thereof in his recent piece.

---

Gel, “Trinitarian trouble?” — Or, my trinitarian parody of the earlier argument from (1)–(5)

I couldn’t help but comment on Gel’s claim at the beginning of this section that he’s “shown why Schmid’s defeaters fail”. He has shown no such thing—his responses are fraught with problems.

Anyway, recall Gel’s argument for uniqueness as formulated in my original response piece:

Here’s my original trinitarian parody of the argument:

Gel alleges in his recent piece that this argument is invalid. This is false, as I’ll explain below. Indeed, Gel himself—since the publication of his paper—agrees that the argument is valid, as he’s said in his streams and in personal correspondence.

Let’s begin with something Gel says by way of stage-setting:

For some mysterious reason, this ‘unique characteristic’ that each person possesses doesn’t make them pure esse plus their unique, differentiating characteristic. And for some equally mysterious reason, a unique characteristic of a being of pure esse that differentiates it from other such beings would make it pure esse plus its unique, differentiating feature. My point here is simply this: any reason for thinking that there could be differentiating features among divine persons consistent with their being pure esse also seems to be a reason for thinking there could be differentiating features among beings consistent with their being pure esse. Or, equivalently: any reason for thinking there couldn’t be differentiating features among beings of pure esse also seems to be a reason for thinking there couldn’t be differentiating features among divine persons of pure esse. There just doesn’t seem to be a relevant difference between the cases. If you get to posit differentiating features among divine persons consistent with each being pure esse, I simply see no reason for ruling out differentiating features among beings consistent with each being a being of pure esse. This is the heart of the syllogism from (1)–(7) and stands irrespective of the particular ways in which the divine persons are distinct from one another and identical-in-being to pure esse. All that matters is the seeming epistemic symmetry between (i) differentiating features not compromising a divine person’s status as pure esse and (ii) differentiating features not compromising a being’s status as pure esse.

Gel continues:

But this misses the point. Of course the argument for God’s unicity denies only the possibility of any ad extra differentiating feature between distinct beings of pure esse. But nowhere did I deny this. My point was simply that the reason for thinking that there cannot be any ad extra differentiating feature of this kind between beings equally justifies the claim that there also cannot be any ad intra differentiating feature between divine persons. Obviously Gel’s argument itself is simply cast in terms of beings and ad extra differentiation among them and so doesn’t by itself entail anything about ad intra differentiation among divine persons. My point was simply that accepting that argument should also lead one to accept a different argument that disallows ad intra differentiation among divine persons. And the reason is the epistemic symmetry mentioned in my paragraph above (and in my original piece). There just doesn’t seem to be any principled reason for thinking that a differentiating feature compromises a being’s pure esse status while a differentiating feature suddenly doesn’t compromise a divine person’s pure esse status. That is, there doesn’t seem to be a relevant difference between beings and divine persons that could account for why a differentiating feature had by beings would compromise their pure esse status while a differentiating feature had by divine persons wouldn’t compromise their pure esse status. Just as a being with a differentiating feature that’s allegedly pure esse would instead ultimately be pure esse plus that differentiating feature (and so not truly pure esse as such), so too would a divine person with a differentiating feature that’s allegedly pure esse would instead ultimately be pure esse plus that differentiating feature (and so not truly pure esse as such). The reasoning in the ‘plus that differentiating feature’ argument is perfectly general.

Gel goes on to disambiguate my premises based on whether they’re talking about divine persons being the same being of pure esse or different beings of pure esse.[11] But, first, none of this is relevant to the central point about epistemic symmetry. The symmetry discussed above remains regardless of whether they’re the same being of pure esse. Second, I appreciate the disambiguation, since clarity is king; but I thought it was already clear that I had in mind Gel’s (6a), (7a), and (8a) interpretations of (1), (2), and (3) in my parody argument, since every competent contributor to the relevant debate knows that the divine persons aren’t construed as different beings (as in (6b), (7b), and (8b)), and hence I thought it was already clear in my premises and discussion thereof that I had (6a), (7a), and (8a) in mind. And once we have (6a), (7a), and (8a) as the understanding of (1), (2), and (3), the parody is perfectly valid. (And the argument was always perfectly valid, since—as I just pointed out—(1)-(3) were always intended as (6a)-(8a).) So Gel’s criticism that my parody argument is invalid is mistaken. It remains valid, as Gel himself has since agreed.

In connection with (6) (i.e., (1) of my parody), Gel writes in footnote 21:

This won’t work. First, by ‘differentiating feature’ between x and y, I simply mean what I take Gel and Pawl to mean when they talk about a ‘unique characteristic’ in virtue of which the divine persons are numerically distinct from one another. It’s simply something that x has (or is) and which y doesn’t have (or isn’t). So it’s not an option to deny the use of this expression, since it’s essentially the same as what Gel and Pawl (and Trinitarians more generally) already apply to the divine persons. Second, denying application of IoI to divine persons fails to engage my criticism, since (part of) my criticism is precisely that the same motivation for thinking that IoI applies to beings equally motivates thinking IoI applies to persons. It is not an option to simply assert that IoI applies to the former but not the latter; one must address my case that one has no motivation for accepting that IoI applies to the former that doesn’t also motivate accepting that IoI applies to the latter.[12]

Gel claims that in order to validly infer (8a) (i.e., (3)), my (1) and (2) would have to be understood as (6a) and (7a). But, according to Gel, “in order… to be true to [Gel’s original] unicity argument”, my (2) would have to be understood in the sense of ad extra differentiation, i.e., (7b) as opposed to (7a). But again, this just misses the point. I wasn’t trying to be “true to” Gel’s original unicity argument. I was pointing out that the same motivations for thinking Gel’s original premise in terms of ad extra differentiation among beings equally motivate my new premise—i.e., (2) or (7a)—in terms of ad intra differentiation among divine persons. The point was never to follow Gel’s application to ad extra differentiation; the point was to emphasize that the motivations for accepting Gel’s premise about ad extra differentiation equally motivate accepting a different premise about ad intra differentiation, and hence if one accepts the former, one should also accept the latter. In other words, if we should accept Gel’s premise about ad extra differentiation, we should also accept my new premise about ad intra differentiation, given the epistemic symmetry discussed earlier.[13] This is the point.

Now, Gel recognizes something kinda like this point when he writes: “At this point, could someone claim the problem to be that any justification for (7b) will inevitably carry over to (7a), creating a bridge between the unicity argument and the parody argument? Might one say, for instance, that if the Son has his proper characteristic (filiation) in distinction to the Father, then the Son can’t be the same being of pure esse than the Father, but being plus filiation?”

We should note, though, that this isn’t the point I was making (although it’s kinda similar thereto). The point of my criticism isn’t that the justification for (7b) carries over to (7a); as I’ve explained, the point is that the justification for Gel’s original premises (1) and (2) carries over to the new, parody premises (1) and (2) (i.e., (6a) and (7a)). And, in light of my actual point, Gel’s earlier responses about ad intra and ad extra differentiation, as well as about invalidity, entirely fall by the wayside—they simply miss the point. So if Gel wants to salvage his case, he needs his response to the abovequoted point to successfully rebut my actual point. What’s his response? It’s highlighted in blue:

It should be clear at this point why this doesn’t work: the exact same thing can be said about distinct beings of pure esse. Each of them could be identical with their own unique characteristic, which is exactly what Gel is saying here about the divine persons. In fact, if you’ll recall, Gel’s response here, cast in terms of divine persons, is exactly the same response that I gave, cast in terms of beings, in response to Gel’s rejoinder to my criticism (a)—I made precisely the point that A could be identical to its unique characteristic F-ness whereas B could be identical to its unique characteristic G-ness. To put things differently: Gel is here responding to the criticism that the divine persons, in virtue of possessing differentiating characteristics, would have to be pure esse plus that differentiating characteristic. How does he respond? By saying no, because the divine persons are each identical to their respective differentiating characteristics. But literally the same thing can be said about beings of pure esse in response to Gel’s claim that each such being would have to be pure esse plus their differentiating characteristic. In particular, the appropriate response is no, because the beings are each identical to their respective differentiating characteristics. It surprises me that Gel doesn’t ask whether his response here—that the divine persons needn’t be pure esse plus their differentiating characteristic—would equally carry over to beings and thereby undermine his unicity argument. (It does, as I’ve explained.)

Notice also that the dialectic here vindicates my original point about the epistemic symmetry between Gel’s unicity argument and my parody argument: they stand or fall together. A response to the parody argument can equally be co-opted as a response to Gel’s unicity argument. This is precisely what we just witnessed: Gel responds to the parody argument by identifying the divine persons with their respective differentiating characteristics; and yet a parallel response equally afflicts Gel’s original unicity argument: simply identify the beings with their respective differentiating characteristics.[13.5]

Thus, Gel’s responses to my parody argument fail. My argument isn’t invalid; it doesn’t equivocate between ad extra and ad intra differentiation; and Gel’s response to the central point about justification transference between the parody argument and the original unicity argument (wherein Gel identifies the divine persons with their differentiating characteristics) ends up undermining Gel’s unicity argument. I conclude, then, that my original trinitarian parody stands unscathed.

---

Section footnotes

[11] I will assume that you have read Gel’s section here and so know what (6a), (6b), (7a), (7b), (8a), and (8b) are.

[12] I know Gel himself doesn’t necessarily endorse this footnote, since he simply says “maybe the Trinitarian would” take this route. But I think it’s important to address nonetheless, since the route fails.

[13] And this epistemic symmetry is precisely what I conveyed in my original response piece. In setting up my parody, I explicitly said that “the exact same reasons for thinking there could only be one purely actual being of pure esse would equally motivate thinking that there could only be one purely actual divine person of pure esse” (p. 7). Notice I’m saying that the motivation for Gel’s argument cast in terms of beings equally motivates a different argument—the one that I immediately proceed to give—cast in terms of persons.

[13.5] I’m setting aside here the seeming tension between (a) the divine persons being identical to their unique, differentiating characteristics (which are not shared among the persons) and (b) the divine persons being identical to a shared divine essence. For more on potential tensions between Trinitarianism and DDS, see my experimental post here.

---

Gel, “More arguments for God’s unicity but no more ‘IoI-ing’”

In this section (and its two subsections), Gel offers two further arguments for God’s unicity. Below I examine each in turn.

Gel, “From simplicity to unicity”

Gel nicely summarizes the first argument thusly:

This argument doesn’t work. I actually address it in my forthcoming Springer book Existential Inertia and Classical Theistic Proofs. I specifically engage Feser’s articulation of the argument. Feser writes:

“The first argument Aquinas gives in the Summa Theologiae for divine unity appeals to divine simplicity. Because God’s essence and existence are identical—because he just is his existence—his essence is not something that can be shared with anything else, so that in the nature of the case there cannot be more than one God.” (Forthcoming)

There are two troubles with this argument.[14] As the problems appear in my book:

You could, of course, claim that the Father’s essence couldn’t be shared with another being but could be shared with another, numerically distinct reality (e.g., another divine person) that isn’t a different being. But—as before—this response fails to engage the second criticism. The second criticism is that the same motivation for thinking that the Father’s essence couldn’t be shared with another being equally motivates thinking that the Father’s essence couldn’t be shared with another person. If one accepts that the Father’s essence couldn’t be shared with another being on the basis of the Gel-Feser-Aquinas argument, one should also accept that the Father’s essence couldn’t be shared with another person on the basis of the parody argument.[15] It is not an option, then, to merely assert that the essence is shareable in the latter way but not the former way; one must address the epistemic symmetry between the cases.

Gel does say something about a potential trinitarian parody, but what he says suffers from the same problem as his reply in the previous section. Gel writes (in footnote 26):

Again, this just misses the point of the trinitarian parody. The point isn’t whether the wording of the argument would only apply to distinct beings. Everyone grants that. The point is that the justification for thinking that something absolutely simple cannot share an essence with another being would equally justify a different premise in a different argument according to which something absolutely simple cannot share an essence with another person, such that accepting the former argument in terms of beings should lead one to accept the latter argument in terms of persons. It is patently irrelevant to this claim to point out that the wording of the original argument by itself entails nothing about whether a simple essence can be shared among distinct persons. The point at hand is that the reasons for accepting the original premise cast in terms of distinct beings also apply to a new premise cast in terms of distinct persons. So, once again, Gel has failed to avert the trinitarian parody.

Next up is Gel’s second argument for unicity.

---

Section footnotes

[14] The troubles afflict both Feser’s articulation and Gel’s, since they’re the same argument. I include Feser’s articulation for three reasons. First, seeing different articulations of the argument aids understanding. Second, Feser’s articulation is admirably clear and pithy. Third, doing so allows me to shamelessly plug my Springer book.

[15] See the preceding section for more on how the trinitarian parody criticism works.

---

Gel, “From perfection to unicity”

Gel’s second argument for God’s unicity derives from perfection. Gel’s strategy here—following Leftow—is to show either that uniqueness itself is a perfection or that uniqueness is entailed by something that’s a perfection. Since a perfect being has every perfection essentially, it follows that a perfect being would be unique. As Gel points out, this reasoning holds for both classical and non-classical theistic models of God.

Now, at the outset, I want to emphasize that Gel’s argument here—if successful—illustrates the folly of much of internet classical theist rhetoric. It’s a commonly parroted point among internet classical theists that only divine simplicity allows us to secure monotheism or God’s uniqueness (or, equivalently: if divine simplicity is false, then there could in principle be more than one God). Feser (2017, p. 189) makes this claim (without offering any justification thereof—something internet classical theists also never do), and it’s a talking point among countless others. If Gel’s argument here succeeds, hopefully this confused and hopelessly myopic talking point is put to rest.[16]

I won’t examine one-by-one Gel’s sub-arguments in this section for the claim that uniqueness (or something which entails as much) is a perfection.[17] I will here simply note some serious worries that need to be addressed in order for Gel’s sub-arguments to succeed.

First, it’s not immediately clear at this juncture whether claiming that these properties are perfections is dialectically appropriate. For if there could be more than one perfect being, then the properties to which Gel appeals are not perfections after all,[18] and hence to assert that they’re perfections just is to assert, in part, that there couldn’t be more than one perfect being—and yet that’s the very question at issue. Admittedly, this worry gets into really tricky waters in Perfect Being Theology (PBT). For example: when are we entitled to claim that some property P is a perfection? Must we first know that P is even possible to begin with? Or is the intuition that <P is a perfection> itself a reason to think P is possible? While I would be inclined to the latter view (at least with my theist hat on), it’s still not entirely clear how to answer these questions. Furthermore, there’s considerable controversy over them in the PBT literature.

Second, for Gel’s arguments in this section of his paper to succeed, he has to add to the core hypothesis of theism various claims about value (e.g., which properties are perfections), claims about the reliability of our intuitions concerning which properties are necessarily great-making, claims about the proper methodology of PBT (as we saw above), claims about the conditions under which we’re epistemically entitled to infer P is a perfection (as we saw above), and so on. But adding postulates to one’s theory increases its complexity, and hence even if Gel purchases some explanatory power with respect to the number of first causes by adding such postulates, it’s not at all clear whether this offsets the incurred cost of added complexity.

Third, trinitarian parodies of all sorts rear their heads. (Remember, even if Gel’s arguments themselves are cast in terms of unique beings, the point is that the motivations Gel offers equally motivate new arguments cast in terms of unique persons.) Below I survey each argument and offer a trinitarian parody thereto. For each argument and its corresponding parody, there seems to be an epistemic symmetry between them of precisely the kind discussed earlier. I see absolutely no reason for thinking the arguments cast in terms of beings are more acceptable than their near-identical counterparts cast in terms of persons, and the justifications for their key steps are essentially identical. If you accept the former, you should also accept the latter.

• Gel’s first argument:

“[C]onsider that F is a perfection if it is ‘objectively and intrinsically such that something F is more worthy of respect, admiration, honor, or awe than something not F, ceteris paribus’ (Leftow (2012), 178). But it seems that something unique is more worthy of respect, admiration, honor, or awe than something not unique. Hence, being unique seems to be a perfection. But there does not appear to be any incompatibility with being unique and other properties a perfect being ought to have. Hence, we can say that, plausibly, a perfect being would be unique.”

• Trinitarian parody:

Consider that F is a perfection if it is ‘objectively and intrinsically such that something F is more worthy of respect, admiration, honor, or awe than something not F, ceteris paribus’ (Leftow (2012), 178). But it seems that something unique—like a unique divine person (i.e., a divine person who is the only divine person)—is more worthy of respect, admiration, honor, or awe than something not unique—like a non-unique divine person (i.e., a divine person who isn’t the only divine person) . Hence, being unique seems to be a perfection. But there does not appear to be any incompatibility with being unique and other properties a perfect person ought to have. Hence, we can say that, plausibly, a perfect person would be unique. There cannot, in other words, be more than one perfect person.

• Gel’s second argument:

“Consider now that a perfect being would plausibly possess supreme or absolute value. But something is more valuable in the same measure as it is more unique – or at least that seems reasonable enough and congruent with how we measure value. Hence, a perfect being would plausibly be unique.”

• Trinitarian parody:

Consider now that a perfect person would plausibly possess supreme or absolute value. But something is more valuable in the same measure as it is more unique – or at least that seems reasonable enough and congruent with how we measure value. The less unique a perfect person is – i.e., the greater the number of perfect persons – the less value it has. Hence, a perfect person would plausibly be unique. There cannot, in other words, be more than one perfect person.

• Gel’s third argument:

“Consider also that it seems to follow from the notion of a perfect being that it could not have a superior, that nothing could be greater in perfection than it. But there is also a case to be made that ‘there cannot be something wholly distinct from [God] and as great as He is’ (Leftow (2012), 207) – that is, that a perfect being could not have an equal. Indeed, it seems greater to be unmatched in perfection than not to be. As Leftow puts it, ‘[i]t would be greater to be intrinsically such as to be the greatest possible being among commensurable rivals than not to be. No constellation of attributes could confer more perfection than one that made one thus greatest’ (ibid.). Hence, it seems to follow once more that a perfect being would plausibly be unique – it would have no superior and no equal.”

• Trinitarian parody:

Consider also that it seems to follow from the notion of a perfect person that it could not have a superior, that nothing could be greater in perfection than it. But there is also a case to be made that there cannot be something distinct from a perfect person and as great as that person  – that is, that a perfect person could not have an equal. Indeed, it seems greater to be unmatched in perfection than not to be. To modify Leftow, it would be greater to be intrinsically such as to be the greatest possible person among commensurable rivals than not to be. No constellation of attributes could confer more perfection than one that made one thus greatest. Hence, it seems to follow once more that a perfect person would plausibly be unique – it would have no superior and no equal.

• Gel’s fourth argument:

“Finally, consider what Leftow calls the GSA-property (short for ‘God, Source of All’): x has the GSA-property if, for any concrete substance wholly distinct from x, x and only x makes ‘the creating-ex-nihilo sort of causal contribution’ to its continued existence (Leftow (2012), 21). As Leftow argues, the GSA-property is either a perfection or a constituent of other perfections. Why think this? First, consider that ‘[b]eing a potential ultimate source of some proportion of what benefits things is a good property to have’ (Leftow (2012), 22). But being the ultimate source of all that benefits things would be the maximal degree of this good property, and hence, given that ‘a property is a perfection iff it is the maximal degree of a degreed good attribute to have’ (Leftow (2012), 22), being the ultimate source of all that benefits things is a perfection. Now, such a perfection supervenes on the GSA-property – and so, either the GSA-property, by a plausible supervenience principle, is itself a perfection or it is a necessary condition of a perfection. In either case, a perfect-being will have the GSA-property.

Consider also that the GSA-property, together with the ability to freely exercise one’s own power, constitutes the property of having complete control over all other concrete objects. But ‘[i]t is good to have power over other things’ existence … Power over existence is degreed. Complete power over all other concrete things’ existence is its maximum, and so plausibly a perfection’ (Leftow (2012), 22). In this case, the GSA-property is a constituent of another perfection, and so a perfect being would have the GSA-property.

But it seems clear that there could only be one being which had the GSA-property. For suppose there are two distinct gods, Alpha and Omega, which both have the GSA-property. Because of that, Alpha and Omega would simultaneously be causally dependent on each other, which is viciously circular – Alpha will be creating Omega only insofar as Omega will be creating Alpha, but Omega will be creating Alpha only insofar as Alpha will be creating Omega. So, at most only one thing can have the GSA-property (Leftow (2012), 192-193). But if a perfect being would plausibly have the GSA-property, it follows that there could only be one perfect being.”

• Trinitarian parody:

Consider the GSA*-property: x has the GSA*-property iff, for any y distinct from x, x and only x is y’s (continued) source. Plausibly, the GSA*-property is either a perfection or a constituent of other perfections. Why think this? First, consider that ‘[b]eing a potential ultimate source of some proportion of what benefits things is a good property to have’ (Leftow (2012), 22). But x being the ultimate source of everything that benefits anything distinct from x would be the maximal degree of this good property, and hence, given that ‘a property is a perfection iff it is the maximal degree of a degreed good attribute to have’ (Leftow (2012), 22), x being the ultimate source of everything that benefits anything distinct from x is a perfection of x. Now, such a perfection supervenes on the GSA*-property – and so, either the GSA*-property, by a plausible supervenience principle, is itself a perfection or it is a necessary condition of a perfection. In either case, a perfect person will have the GSA*-property (since a perfect person has every perfection essentially, by definition).

Consider also that the GSA*-property, together with the ability to freely exercise one’s own sourcing power, constitutes the property of having complete control over everything else. But it is good to have power over the existence of everything else. Power over existence is degreed. Complete power over the existence of everything else is its maximum, and so plausibly a perfection. In this case, the GSA*-property is a constituent of another perfection, and so a perfect person would have the GSA*-property.

But it seems clear that there could only be one person with the GSA*-property. For suppose there are two distinct persons, Alpha and Omega, both of whom have the GSA*-property. Because of that, Alpha and Omega would simultaneously be sourced in and dependent upon each other, which is viciously circular – Alpha will be explaining Omega’s existence only insofar as Omega will be explaining Alpha’s existence, but Omega will be explaining Alpha’s existence only insofar as Alpha will be explaining Omega’s existence. So, at most only one perfect person can have the GSA*-property. But if a perfect person would plausibly have the GSA*-property (as was established above), it follows that there could only be one perfect person.

(End of parodies)

Gel again tries to address trinitarian parodies at this juncture. He writes:

“Does this reasoning conflict with Trinitarianism? If unicity is a perfection that any perfect being ought to have, some will say, then for a divine person to really be divine (and hence, perfect) it would also have to be unique. And so, the same intuitions would support the conclusion that there can only be one divine person. But at least the traditional account of the Trinity presented above can easily deal with this objection. The ad intra differentiation that takes place within God does not make it so that now we have more than one perfect being, and each divine person is still perfect in being identical to one and the same perfect substance, God.”

Once again, this response just misses the point. The point is not that the ad intra differentiation that takes place within God makes it so that we have more than one perfect being. The trinitarian parodies don’t rest on this claim. That’s just not the point. The point is that the reasons for thinking that there cannot be ad extra differentiation among beings equally justifies thinking that there likewise cannot be ad intra differentiation among persons. No one is claiming that the ad intra differentiation within God makes it so that we have more than one perfect being. This is totally irrelevant to the point of the parodies.

Gel says each divine person is still perfect in being identical to one and the same perfect substance, God. But this simply fails to engage the parodies. The parody arguments (and their epistemic symmetry with the original argument) give us reason to think that the divine persons are not perfect after all if there’s more than one of them. In other words, the epistemically symmetric parody arguments give us reason to think that there cannot be multiple persons that are perfect in being identical to a perfect being. It’s no use just blithely asserting, in response, that the multiple persons are perfect in being identical to a perfect being, since I just gave an argument for thinking that’s false. Gel needs to address the argument, not just flatly assert the falsity of its conclusion. I’ve given reasons for thinking the conclusion is true, and merely asserting the conclusion’s falsity simply fails to engage those reasons.

Gel concludes by saying, “Also, further considerations about perfection could support the case that the one and only perfect being should be, internally speaking, more than one person (see, again, Sijuwade (2021)).”

Setting aside the profoundly dubious nature of the ‘further considerations’ Gel merely adumbrates, this is irrelevant to the point of the parodies. The point of the parodies is that if you accept the original arguments from perfection to the uniqueness of divine beings, then you should also accept symmetric arguments from perfection to the uniqueness of divine persons. Gel’s response here doesn’t engage the truth of this conditional; his response simply denies its consequent. So be it if you have reasons for denying the consequent; that just means you should also deny its antecedent (by modus tollens), and hence you should not accept the original arguments from perfection for the uniqueness of divine beings. Nothing Gel says here addresses the conditional itself (or the parodies and epistemic symmetry that justify it).[19]

---

Section footnotes

[16] I address other common internet classical theist talking points—such as the clearly false claim that DDS best solves the Euthyphro problem—in my two-part video series “Arguments for Classical Theism”. Here’s Part 1 and Part 2.

[17] I’m sympathetic with some of them, in fact. Though, as I will argue below, if we accept them, we should reject trinitarianism.

[18] This just contraposes Gel’s claim that if they are perfections, then there couldn’t be more than one perfect being.

[19] Compare: suppose I argue that if a blue ball could come into existence uncaused, then a red ball could also come into existence uncaused. And suppose my reason for thinking this is that there seems to be no relevant difference between a blue ball and a red ball with respect to coming into existence uncaused. It’s no use responding to the conditional by arguing that a red ball couldn’t come into existence uncaused (i.e., by arguing for a denial of the conditional’s consequent). This simply doesn’t challenge the conditional, and nor does it challenge the reason offered on its behalf.

---

Gel, “Can these arguments be of use to the naturalist?”

In this section, Gel asks… you guessed it… whether these arguments can be used by the naturalist.

Gel begins by noting that theism gains a ceteris paribus point in its favor vis-a-vis naturalism by the former’s ability (and the latter’s inability) to explain the number of first causes. However, there are two things to note here. First, as I argued in my original response—and as Gel’s recent responses have failed to salvage—Gel has not shown that theism has an explanation of the number of first causes, since the arguments that Gel proffers fail. (Or, in the case of the arguments from perfection, it’s not immediately clear that they succeed, and furthermore they create problems for trinitarianism.) Second, Gel’s arguments almost invariably add to the complexity of theism—by, e.g., adding the monumentally controversial principle IoI, or adding claims about which properties are perfections together with controversial claims about the appropriate methodology for PBT, etc. So even if Gel secures a point in theism’s favor, this may very well be offset by incurred complexity costs. Note, moreover, that the onus is on Gel to show that the gain in explanatoriness outweighs the incurred complexity cost. As of yet, he has shown no such thing.

Regarding my point on atemporal wavefunction monism, Gel writes: “Surely, atemporal wavefunction monism is an interesting view on its own right. Still, as a hypothetical example of a purely actual material thing, in the Aristotelian-Thomistic sense of ‘material’ with which I was operating, it is bound to be incoherent.” I find this response odd, since I don’t see any indication in Gel’s original piece that he was employing the Aristotelian-Thomistic sense of ‘material’. For starters, Gel didn’t articulate how he was using ‘material’ anywhere in his article. Moreover, at this stage in the dialectic of his original piece, he was considering what a naturalist who accepts a purely actual first cause would be committed to. And the naturalist—qua naturalist—is not committed to the Aristotelian-Thomistic understanding of ‘material’. At the very least, Gel should have specified in his original piece that by ‘material’, he means the Aristotelian-Thomistic sense thereof. He did no such thing, and so I had to supply my own understanding by employing that conditional phrase (“if we understand ‘material’ and ‘physical’ to be synonymous…”).

---

Gel, “Is this advantage worth the price?”

Gel writes, in footnote 29, that

“Another problem I see is that Schmid relies on the idea that ‘Oppy’s entities are a proper subset of the classical theist’s’ (Schmid (2022), 4). But this does not seem true, since Oppy’s ontology contains something which does not figure in the theist’s – an uncaused necessary initial physical state with a beginning.”

But here Gel seems to confuse including different entities in their ontologies with having different theoretical characterizations of a single entity present in both ontologies. Both classical theism and Oppyan naturalism include the initial singularity. Sure, they characterize it differently—for the Oppyan naturalist, it’s necessary, while for the classical theist, it’s contingent. But the entity itself is included in both ontologies. (Compare: the historically informed atheist and the Christian both believe that Jesus existed—the entity, Jesus, is in both their ontologies—even though their characterizations of Jesus diverge.) So the point about proper subsets and simplicity stands. The classical theist, minimally, has {Necessary(God), Contingent(initial singularity), Contingent(non-initial states)}, whereas the Oppyan naturalist only has {Necessary(initial singularity), Contingent(non-initial states)}. In terms of ontological commitments, naturalism is clearly simpler here.[20]

Gel also writes, in the same footnote:

“While theism posits additional kinds Oppy does without—non-physical, unlimited , perfect—because of this the theist is able to give a more unified account of the kinds Oppy recognizes. For the theist, all that is physical falls under the kinds contingent and caused. For Oppy, some of what is physical falls under the kinds contingent and caused, but other physical things fall under the kinds necessary and uncaused. It seems that the denial of the additional theistic kinds comes at the price of additional naturalistic kinds (or subkinds). This seems to be a multiplication of overall complexity difficult to compare with that of the theist.”

But what matters for purposes of ontological parsimony is what kinds we need to characterize everything. The theist and Oppyan naturalist alike need: necessary, contingent, caused, and uncaused. It’s a draw with respect to these. But, where the naturalist only needs natural, the theist needs that and then some: perfect, unlimited, etc. Sure, there may be different combinations of these kinds that the theist and Oppyan naturalist commit to—e.g., the Oppyan naturalist commits to a combination necessary physical and uncaused physical while the theist doesn’t. But that doesn’t change the fact that the number of kinds to which the naturalist is committed is a proper subset of those of the theist. Gel is therefore mistaken in saying that the naturalist posits additional kinds.

Gel finishes the section by alleging that theism has other advantages over naturalism, but his discussion is brief and punts to other works and arguments for justifications of key claims. As such, there really isn’t much to address here without doing an in-depth examination of those other works.

---

Section footnotes

[20] This is only one dimension of simplicity—to wit, Oppy’s ontological simplicity. Note that Gel doesn’t address my case that naturalism is simpler than theism along the other dimensions of simplicity.

---

Conclusion

I’ve argued that basically all of Gel’s responses to my criticisms of his original piece fail. I won’t, however, retrace all the steps I took in showing this. I’m just too lazy for that at this point, lol. (Or should I say ‘IoI’?)

I’ve really enjoyed interacting with Gel, both in print and our private messages. He is incredibly thoughtful, kind, and helpful, and I deeply appreciate our friendship. While I have been critical of his rejoinders, none of my criticisms are meant to tear down or destroy; they’re meant to serve all of us in our collective pursuit of truth. I thank Gel for serving me in this regard, and I hope to have served him (and you) likewise.

Author: Joe

Addendum: If you’re reading this, I think you’ll enjoy my work on these topics on YouTube (see, e.g., my classical theism playlist) and PhilPapers. I also think you’ll especially enjoy my Springer book Existential Inertia and Classical Theistic Proofs (co-authored with Dr. Daniel Linford), hopefully available within the next month or two. (For once it’s out: If you can’t afford it, ask your local library if they can get ahold of it! [They can probably do an interlibrary loan.] And if that doesn’t work, email me.)

Links: Personal website | PhilPeople profile (with all my papers) | Patreon