In one of my university interviews, I was asked to steelman polytheism.
“Imagine I’m a monotheist and you’re a polytheist”, the tutor said. “You believe in many Gods. I believe in one God. Convince me to change my mind.”
I hadn’t thought much about polytheism before, and on the spot I could only think of three arguments. The two arguments for polytheism that are objectively the best – the argument from diverse religious experiences in relation to particular, non-identical gods, and Tiddy Smith’s argument from common consent – are ones I raised in my interview, and, in my opinion, are more or less golden. They don’t convince me, but they trivially provide some non-trivial evidence for polytheism.
I won’t defend those arguments here. Instead, I’m going to defend my third argument: the Unrestricted Modal Ontological Argument.
Prepare to be convinced, future Pagan.
First, recall the standard MOA:
1. It is possible that a maximally great being exists.
2. If it is possible that a maximally great being exists, then a maximally great being exists in some possible world.
3. If a maximally great being exists in some possible world, then it exists in every possible world.
4. If a maximally great being exists in every possible world, then it exists in the actual world.
5. If a maximally great being exists in the actual world, then a maximally great being exists.
6. Therefore, a maximally great being exists.
The key and controversial premise is (1). If (1) is established to the non-theist’s liking, the argument is persuasive. If not, the argument is a nothing but a nothing burger, with a side of absent fries.
There have been numerous attempts to establish (1) to the non-theist’s liking, including attempts by Pruss, Pruss twice over, Collin, Schmid, Bernstein, Nagasawa, and McIntosh. Suppose you think one of these attempts succeeds, and should establish (1) to the non-theist’s liking. In that case, if you accept the argument’s logical assumptions, the MOA is a winner. There is at least one maximally great being – at least one God.
But wait. There’s more. There are more. If you buy into that argument, and also into the thesis of Unrestricted Composition, you should become a polytheist by the lights of the MOA.
Before I say why, I’ll say a bit about Mereological Universalism and why people endorse it. What is Unrestricted Composition? According to Unrestricted Composition (hereafter, UC), whenever there are two or more concrete objects, there is a further concrete object composed by those objects.
In real terms, this means that for any collection of objects – say, my rucksack and Jimmy Carter’s left leg – there is a further object, which we have no name for, that has my rucksack and the left leg of history’s greatest president as parts. This generalizes to all concrete objects: according to UC, there are objects composed by my face and your face, Donald Trump and the London Eye, and the top half of every trout and the bottom half of every turkey.
“Oh, wellllll, this view is obbbviousssssly false.”
Quiet. Silly person. In the second half of twentieth century, UC was easily the dominant view among philosophers. Today, it still commands a healthy following. As Hud Hudson wrote of philosophers’ attitudes towards UC in 2006:
Among many of our most distinguished analytic metaphysicians (UMC) [Unrestricted Mereological Composition] has an honorary status approaching that of a logical truth; it is the sort of thing you just do not mess with. Go ahead and explain to the initially perplexed the dangers of restricting composition (for example, the threat of ontological vagueness and other horrors), gently help them to recognize that it is just their own parochial interests and narrow values that generate that unfounded skepticism against the fusion of all the extant copies of the Gutenberg Bible and the ruin at Stonehenge, get them to see that they would be wise to restrict their attention (rather than their principles of composition) when it comes to these unfamiliar but perfectly respectable pieces of the world’s furniture, and finally—if you must argue further—showcase the theoretical benefits of UC in modality, in set theory, in whatever . . . but just remember, come what may. . . . Thou shall not restrict composition.
Why is UC so popular? There are a number of reasons, but here are two:
First, the alternatives are problematic. Restrictivism – the common sense view that composition sometimes occurs and sometimes doesn’t – seems to have a vagueness problem; restrictivists typically want to affirm that an object like a motorbike exists, but that the fusion of a motorbike and the Moon does not. But, to capture our common sense, pre-theoretic intuitions about motorbikes, it seems we have to say that it is a vague matter when a motorbike exists, since – if we began removing atoms from the vehicle, one at a time – there is intuitively no non-arbitrary cut-off point, a point at which a motorbike was no longer being composed. But ‘vague existence’ seems conceptually absurd.
A similar problem seems to afflict Organicism, a species of Restrictivism which says that composition only occurs in the case of animals, plants, and other living things. (The most interesting attempt to rescue Organicists from this problem I’ve seen is from William Jaworski, in Structure and the Metaphysics of Mind, Chapters 6 and 7.)
Nihilism – the most popular view among internet people but the least popular view among professional philosophers – says that composition never occurs; there are no tables, no chairs, no trees, no homo sapiens.
A big objection to Nihilism – aside from its counter-intuitiveness, which UC shares – is that if you accept that you and I are not simple, part-less substances, like atoms or souls, then you are committed to the eyebrow-raising Buddhist view that you and I do not exist. But since we do, the objection goes, Nihilism cannot be true. But if Restrictivism and Nihilism are out, then, if you accept that there are facts of the matter about when composition does or does not occur, the only option left is UC.
UC, moreover, faces neither of these problems. In fact, one of the biggest motivations for UC is that it doesn’t face the vagueness problem. There is no vague boundary between when two or more objects do and do not compose a further object because they always, in fact, do.
A second motivation for UC is that it captures (an unifies) almost all of our ordinary notions of parthood in one delicious and simple theory. Consider the following claims:
(1) The atom is part of the universe.
(2) The toy is part of the collection.
(3) The leg is part of the table.
(4) The foot is part of the man.
(5) The player is part of the team.
According to Nihilism, all of these sensible-sounding claims are false. According to some forms of Restrictivism, only (3) and (4) could be true. And according to some of the most popular Restrictivist theories, Organicism and some versions of Hylomorphism, not even (3) could be true. The only claim of the five that could be true in any robust sense is (4).
UC, by contrast, allows all five claims to be true at once, and true in exactly the same sense. Thus, UC has an advantage over rival views.
You will object, as, from eternity, I knew you would, by saying that while UC makes nearly all of our common-sense beliefs of the form <the Xs compose a Y> to come out true, it equally makes all of our common-sense beliefs of the form <the Xs do not compose a Y> to come out false. For example, UC implies there are objects composed by my face and your face, Donald Trump and the London Eye, and the top half of every trout and the bottom half of every turkey. This, the objection goes, disqualifies UC.
On this, two points.
First, UC-proponents have an explanation for why claiming that these scattered, gerrymandered objects exist is counterintuitive. In our evolutionary past (and today), there has been no use for such objects. We’ve never had any practical or theoretical use for them, so we’ve never bothered to name them, or form concepts about them. This is why, when UC-proponents claim that trout-turkey exist, their claim strikes us as bizarre.
Still, their claims can be made to feel less bizarre with a thought experiment: suppose we see a scattered assortment of sticks and stones on the ground. I pick up a stone and ask you, “”Is this stone part of anything”. You would probably be inclined to say “no”. But suppose we invent a game involving the sticks and stones, a game in which each stick and stone plays a role. We call the game ‘Stickle Stone’. Suppose, after a few hours of playing it, a dastardly hiker decides to pick up one of the sticks and walk away with it. We would probably say something like: “Hey man, that’s part of our Stickle Stone kit. Use a different stick!”
Obviously, nothing metaphysical has changed about the sticks and stones when we gave a collective name to them. And yet, now we have a practical and theoretical use for the scattered collection of stones and sticks, it feels totally natural to use the language of parthood. UC proponents have a fairly intuitive explanation for this: before, we had no practical or theoretical use for recognizing these disparate objects as part of a greater whole; now we do, recognizing them as such feels perfectly kosher.
Second, and relatedly, many philosophers – after reflecting on cases like these – have stopped finding the claim that these objects exist counterintuitive at all. As Judith Jarvis Thomson wrote in 1983, “one only has to live with fusions for a while to come to love them.” Maybe you can learn to love them too.
Suppose you buy into UC, and also think the standard MOA is persuasive. In that case, consider now the following argument:
1*. It is possible that a maximally great collection of Gods exists.
2*. If it is possible that a maximally great collection of Gods exists, then a maximally great collection of Gods exists in some possible world.
3*. If a maximally great collection of Gods exists in some possible world, then it exists in every possible world.
4*. If a maximally great collection of Gods exists in every possible world, then it exists in the actual world.
5*. If a maximally great collection of Gods exists in the actual world, then a maximally great collection of Gods exists.
6*. Therefore, a maximally great collection of Gods exists.
7*. If a maximally great collection of Gods exists, polytheism is true.
8*. Therefore, polytheism is true.
Not genuflecting to Zeus yet? Here are three objections you might be thinking of, with replies to those objections.
~
Objection #1: maximal greatness presupposes a maximum possible degree of greatness. But a collection of Gods couldn’t have a maximum degree of greatness, because there could always be one more God, making the collection greater by one unit of God.
Reply: Not if the collection is infinite. There is an infinity of Gods.
~
Objection #2: Actual infinites are metaphysically impossible.
Reply: No.
~
Objection #3: It’s impossible for there to be more than one maximally great being. A maximally great being would possess aseity – that is, it would be the only self-existent being. A being without aseity has fewer great-making properties, all else equal, than a being that has aseity. Therefore, if there is a maximally great being, there can’t possibly be others.
Reply: Polytheists can just make the move Yujin Nagasawa makes when he defends the traditional ontological argument, and endorse what I call the Maximal Gods Thesis:
Maximal Gods Thesis: The infinite collection of Gods possesses the maximal consistent set of the great-making properties power, knowledge, aseity, and benevolence.
On this thesis, if maximal aseity on the level of individual Gods would rule out the possibility of there being other such Gods, then the infinite collection of Gods doesn’t have maximal aseity on the level of individual Gods. This doesn’t defeat the argument, though, because the infinite collection of Gods is still greater, overall, than a single God with maximally aseity.
Similarly, if you were worried about the omnipotence paradox for polytheism (“Could one God make a rock so heavy that the other God’s couldn’t lift it? If no, that God isn’t omnipotent; if yes, the other Gods aren’t omnipotent.”), this reply works for that too. All you have to say is that the collection of Gods possesses the maximal consistent set of power. If it would be paradoxical to have more than one fully omnipotent being, then there isn’t more than one fully omnipotent being. Rather, there is an infinity of beings with as much power as it is possible to have without generating the omnipotence paradox.
~
Objection #4: In a somewhat groovy paper by Ross Inman and Alexander Pruss, there two arguments against the possibility of UC given Anselmian theism. One of them, which they call Value, runs as follows:
1. UC is true and God exists. (Assumption for reductio)
2. If x and z have no parts in common, and z has positive intrinsic value, the object composed of x and z is at least as intrinsically valuable as x. (premise)
3. Necessarily, no concrete object that is numerically distinct from God is as intrinsically valuable as God. (T1)
4. God is concrete. (premise)
5. There is a concrete object, z, that has no parts in common with God and that has positive intrinsic value. (premise)
6. There is a concrete object, y, composed of God and z. (1,4,5, CP)
7. y is at least as intrinsically valuable as God. (2,5,6)
8. z is not a part of God. (5)
9. y is numerically distinct from God. (6,8,PP)
10. Contradiction! (3,7,9)
This seems to make trouble for my argument. If we accept the premise that, necessarily, nothing distinct from the Infinite Divine Pantheon can be as intrinsically valuable as the Infinite Divine Pantheon itself, then, it seems, UC will be incompatible with its existence. But UC was what needed to prove its existence. Thus, if we grant this premise, my argument is done.
Reply: It seems like the best move for the polytheist is to reject the claim that <nothing distinct from the Infinite Divine Pantheon can be as intrinsically valuable as the Infinite Divine Pantheon itself> as unmotivated. I’d say more, but I’m running on four hours of sleep and a mug of vegetable soup, and my brain is too tired plan a devastating rebuttal. Instead, I’m going to take an earth-shaking, game-ending nap, the likes of which has never been seen in this world.
(Obviously, were I blogging in my prime, I would raze this argument to the ground.)
Steelmanning Polytheism
"If you accept X (which is false) and Y (which is impressively false, almost scandalously so), then polytheism follows."
Commitment to mereological universalism and the unrestricted composition principle doesn't seem to offer the perfect being theist any reason to affirm that a maximally great collection of "Gods" exists. For polytheism to be true is for it to be the case that there is an x and a y such that (i) x is god (i.e., the predicate “god” applies to x), (ii) y is god (i.e., the predicate “god” applies to y), and (iii) x is not identical to y. In other words, there are at least two gods.
However, the predicate “god” does not apply to a maximally great being; rather, the predicate “God” applies to a maximally great being, and perfect being theists endorse the following account of what it is to be a God: x is a God iff (∃y)(y=God & x=y). This account entails that polytheism is logically impossible.
—
• Trinity and Polytheism by Edward Wierenga
• The Problem With Social Trinitarianism: A Reply to Wierenga by J.E. Brower, see pg. 299
• Social Trinitarianism and Polytheism by Brandon Carey, see pg. 100-101