Prof. Robert Koons has recently made his very important paper on classical theism available open-source. I really enjoyed it and I highly recommend reading it to gain an idea of classical theism as a whole. That said, while reading, I noticed I could extract arguments for an ultimate polytheism out of it. I will be referencing the pages here as required.
This post is inspired by Steven Dillon, who gave me some hope that Analytic Philosophy of Religion can accommodate our shared ideas about Polytheist Platonism. Thank you Steven!
I
“First, I will need the concept of a logical moment. If agent A causes some effect E, then we can identify two logical moments, even if the action and the effect are temporally simultaneous. If agent A causes E, then agent A acts at one logical moment M1, and the effect is first in actuality at a posterior moment M2. The relation of priority/posteriority between logical moments is transitive and asymmetric. No logical moment is prior to itself, and no moment is prior to any moment that is prior to it.”1 (Pg 11)
The interesting thing about this proposal is that it can apply to an eternal cause of an eternal effect, which effectively covers even temporal causation since, as Koons argues (Pg 3-4), cause and effect are temporally simultaneous.
It brings up the question, what does the cause actually give the effect? For classical theists, what is shared is "Being", the "what" of the thing. The cause cannot give what it doesn't have. This is the proportionality of cause and effect, as Koons also argues (Pg 3). But the thing is, being temporally simultaneous (or even eternally simultaneous), the cause C1 can't seem to cause the existence of the effect E1. On the temporal side, there is the possibility of infinite regress of temporal causes that are in effect, simultaneous, without any ground for what is shared between all of them. Eternally, not much changes, because there must be something to receive the cause's activity.
The alternative is that the cause does two things, it existentiates the possibility of the entity, and also moves the potential existence of that entity to actuality. These are two separate acts. This is not the same as saying that the cause has to existentiate the "whatness" of the effect. The cause must already have that. Instead, this is saying that it must existentiate the very possibility of the separate entity that has that "whatness". It must existentiate the subject that will have those predicates. But I argue that this is not possible, due to another principle of causation that Classical Theists hold to, namely the principle that states that "the cause must be superior to its effect with respect to what it causes". See Proclus’ Proposition 7 in his “Elements of Theology”:
“Every productive cause is superior to that which it produces.”2
You can think of this more in terms of exemplification. The cause of motion in an effected unit must exemplify the attribute of "motion" more than the effect, insofar as what is being given this attribute lacks it in some way. Given the Simultaneity of cause and effect, it is not that the cause first has this attribute, and then loses it immediately it is transferred to the effect. It is instead that in the very moment of simultaneity of cause and effect, the cause must exemplify in a greater way what it gives the effect. The effect receives it immediately, but it is not thereby equal to the cause, otherwise, there would be no way to distinguish cause and effect, because the arrow of causation can go either way, the exemplification of that attribute being equal between them. It is therefore that the cause is greater than the effect in what it gives the effect. We can apply this to all instances of causation Classical theists consider valid. With this in mind, there is something that is shared equally by all entities, even the classical theist “God” insofar as they are in the causal series: The “attribute” of “unitness”. Each member of a causal chain can be counted. But we do not just mean in the sense of they are part of a series that can be mapped to the number line. We mean that we can count each as one. In a causal series A→B→C→D…, each member of this series is one, a unit, prior to being in this series, i.e. they have to exist as units themselves before they can be placed in a series. That is, their unitness precedes their relation. Arguably, this is essential to existence, as Proclus argues in Proposition 1:
PROP. 1. Every manifold in some way participates unity.
For suppose a manifold in no way participating unity. Neither this manifold as a whole nor any of its several parts will be one; each part will itself be a manifold of parts, and so to infinity; and of this infinity of parts each, once more, will be infinitely manifold; for a manifold which in no way participates any unity, neither as a whole nor in respect of its parts severally, will be infinite in every way and in respect of every part. For each part of the manifold- take which you will-must be either one or not-one; and if not- one, then either many or nothing. But if each part be nothing, the whole is nothing; if many, it is made up of an infinity of infinites. This is impossible: for, on the one hand, nothing which is is made up of an infinity of infinites (since the infinite cannot be exceeded, yet the single part is exceeded by the sum); on the other hand, nothing can be made up of parts which are nothing. Every manifold, therefore, in some way participates unity.3
Indeed this is the most essential thing to existence, that each thing be one thing, otherwise, it is nothing at all, not even something conceivable in principle, insofar as it must be a thing, one thing, some thing, this thing, to be anything at all conceivable. Something might have more or less parts, n parts, less integral parts, flimsy parts, etc, but each of this is somehow one unqualifiedly. No one is less “countable” this way. The star and the God each count “one”.
To say the cause as a unit causes the unitness of another (thus, its existence) is to say that its unitness is superior to the effect’s unitness. But this is impossible, each is “one”. The numerical unitness - in the way described earlier as their thisness, each counted as “one” thing - is the same for each. Arranged qua unit, there is no way to distinguish cause and effect. The cause does not produce the unitness of the effect, therefore it doesn’t produce the existence of the effect. If the Classical theist God were the first unit in a series of causes, then it faces the problem that it cannot cause the existence of the units, only their substance. The distinction between essence and existence reasserts itself, with existence taking priority.
II
The classical theist has a response to this problem, in that they can appeal to the distinction between unity and units. Indeed, Proclus himself makes this distinction in Prop 6:
“Every multitude is either a multitude of unified multitudes or a multitude of independent Unities.”4
Proclus builds on the previous five propositions that mention “Unity”, “Unities”, and “Unifieds”. Something is “unified” if it has some multitude that “make up” that unity. For instance, the human body is a unit, but it is unit that can be divided. It has parts. This extends to non-bodily or purely intelligible things. Minds have multitudes of thoughts, Souls have parts, an idea is the unity of many related idea-parts that make it intelligibly refer to something that exists. There is a conceptual distinction between the unity of a thing and what that thing unifies. It almost parallels the Thomst distinction between essence and existence in that an essence has many attributes, but the essence needs actualization by something that is already actual, an essence that is actual. This line of reasoning eventually needs to something for which its act of existence and its essence are identical, meaning there is no need for its own essence to be actualized by something else beyond it. Applying this line of reasoning, the classical theist can rebut the previous argument by demonstrating that God is not a mere unified, but Unity itself, which does not need to be unified by anything prior. But there is a subtlety to Proclus’ argument here, in that it applies to each member of the eternal chain of causes in a way that undermines the classical theist argument. There are two related ways to demonstrate this, one leading to another.
III
The first is by considering the relationship between each original unit in the causal series and their essence or essential attributes or properties that they gain from the cause. Is “unitness” among those attributes? We have already demonstrated that it is not among those attributes. It is more proper to say that this “unitness” is simply the “subjectness” of the units in question5. It is around “unitness” that the attributes build themselves, and indeed each attribute or property is itself “one” attribute or property. It is “unitness” all the way down. Together and apart, in the same entity, they participate that entity’s unity. That is, they are unified. But is the entity’s “unitness” itself unified? The answer is “No” because it would lead to the same infinite regress encountered in Proposition 1. The layers of properties and attributes must eventually terminate in the pure unity of the entity as such. This is indeed why this translation of the proposition speaks of Unities as well as Unifieds. These unities are themselves not caused, otherwise they’d be unifieds, leading to that infinite regress. Given that these are pure unities, they do not have “boundaries”; that is, while Unifieds are multitudes, distinguished in various qualitative and quantitative ways, Unities qua Unities have no parts, no multitude, no qualities (qualities are also multiplicities) or quantities, making them ontologically indistinguishable. We seemed to have reached the classical theist synthesis. If they can’t be distinguished ontologically, then (so the argument goes) they are identical. Ultimate Polytheism is impossible, at least for the classical theist, and Koons argues precisely this (Pg 22). How can we proceed?
IV
This brings us to the second way. This time, instead of reasoning “up” to Unities (or Unity?), we presuppose the Unity and reason downwards, as a thought experiment. Let us consider a causal series beginning from Unity itself. We are presupposing the principle of proportional causation and the dictum that cause is always greater than effect. What does “Unity itself” cause? Now this is tricky because some might say simply “Units” or “Unifieds” (same thing). But this is not necessarily true. Motion in objects primarily causes motion. Things with some essence that is given in the cause primarily give that essence before anything else. Something can be secondarily caused as a consequence of this primary causation, maybe in a subsequent effect later in the series, but this is because there is something primary being caused, and this is the thing shared between cause and effect. So, Unity itself doesn’t primarily cause units (i.e. unifieds), it causes Unities. Specifically, it causes the unities of the unifieds, the unity that is “specific” to them. Before there is the secondary effect that is the unifieds, there is the unity specific to it. But then, we run into a problem. Given what unities are, not being ontologically distinct, nor graded, whatever they unify – the unity specific to some star is no less a unity than the unity specific to the universe as a physical whole – there is still no way to distinguish both the unities from each other and the unities from Unity itself.
V
For the problem of distinguishing the unities from each other, the only way to do this “conventionally” is to distinguish the unifieds themselves. That is, the cause is known through the effect. The unities are known through the unifieds they unify. But what exactly is known here? We are not distinguishing the Unities via themselves, and we are also not distinguishing the Unifieds via themselves. The former is impossible ontologically, as we have seen. The latter would simply be referring to the Unifieds without reference to the unities, and would thus be an abstract distinguishing of attributes in isolation from the unities they inhere in i.e. it would be a mere analysis of concepts qua concepts. What we want is something that does indeed analyse the relationships between the Unifieds, but in a way that reveals or treats them as caught up in some pure unity. Given how these Unifieds entangle among one another, even between units all throughout existence – i.e. there are no true voids in nature, everything is connected to everything else – this search for the “character” of any of these unities is potentially endless in detail. But this endlessness does not mean we cannot recognize the character even within the process of discovery, and we indeed have a very well-known logic for this kind of analysis. It is the logic of the proper name. The proper name is in principle unique to only one entity6. Even if we share the same name in terms of spelling, pronunciation, meaning, etc; in calling that name, you are referring to just one person, not as some instance of a common nature – that is a secondary consideration – but as a one of a kind, unrepeatable unit, an unrepeatable unit that you can endlessly explicate. That is directly mappable to the logic of the Unifieds revealing or showing forth the character of the Unities. To explicate me fully, you are going to have to explicate my relation to the entire universe, past, present, future. This is impossible to do at any one time, but this is true in principle. Yet, I am this one person, knowable all through it. This is, again, directly mappable to the Unities known through Unifieds. Each, in explicating them, implicates the all, and includes them, immediately, insofar as their unity is not separated or defined by any ontological boundaries or bridges that might mediate it. This logic is how there can be a multiplicity of Unities without being an ontological multiplicity of limited units only definable in opposition to each other or being members of a species, even a species of one member, as Koons says of the Thomist Angels (Pg 22).
VI
For the problem of distinguishing the Unities from Unity itself, the indistinguishability of cause and effect in terms of difference of power or exemplification means that the entire structure collapses. Remember, Unity and Unities do not differ in any ontological way. They have no boundaries or explicit multiplicities – qualitative or quantitative – nor is Unity itself any more powerful than Unities for the same reason, as power is itself a multiplicity if it has grades. Unities cannot be parts of Unity itself for this same reason. Being ontologically indistinguishable, one might say Unity itself should have its own “character”, just as the Unities. But the Unities are not distinguished ontologically, so they cannot “display” any character of their supposed cause, especially not “together”. They aren’t “together”, that implies boundaries and ontological distinction. They are simply “each” Unity, our use of “they” being a concession to language and our referral to their character in the Unifieds. The collapse of the causal structure due to this means that there is no “Unity itself” above the “Unities”. There are simply Unities. That is, what there is, is an Ultimate Polytheism7.
VII
What then of the causal series and the particularly the contingency argument? Well, what the classical theist contingency argument leads to is the necessity of a first essential cause, which could be any God, any Unity in their aspect of simply Being. The chain of causation is then the “drama” of divine cosmic arrangement, others “following” the lead of that first God in that particular world, from Being itself to ever most specific structures of being. This is indeed what Proclus describes in his Platonic Theology. The autarchy of these Gods is best expressed in the very possibility of multiple worlds. Indeed we might say that this ultimate polytheism predicts this possibility, and permits critique of cosmologies8, rather than reifying them as ultimate. Best of all, this Polytheism tracks better than Monotheism and Atheism with the majority of the world’s religious experience, which has throughout most of history affirmed multiple ultimate Gods.
Koons, Robert C. ‘Does the God of Classical Theism Exist? 1’. In Classical Theism. Routledge, 2023.
Proclus, and E. R. Dodds. The Elements of Theology. Oxford University Press, 1971.
Proclus, and E. R. Dodds. The Elements of Theology. Oxford University Press, 1971.
Vargas, Antonio. ‘Proclus’ Elements of Metaphysics’. Accessed 23 November 2022. https://www.academia.edu/44841806/Proclus_Elements_of_Metaphysics.
I chose this translation for its use of “Unities” instead of “Henads”. The former communicates the point of the latter far more readily to an unfamiliar audience. Same for “Multitude” instead of “Manifold” as in Dodds’ translation.
Dillon, Steven. Pagan Portals - Polytheism: A Platonic Approach. Alresford: Moon Books, 2022.
“In this who-and-what, proper-name-and-common-noun practice, we see two fundamental aspects of anything’s unity. In proper-name unity, we zero in immediately upon one unique entity, whereas in common-noun unity, we as it were fashion increasingly fine nets in which to catch a smaller and smaller number of entities, until we get down either to one, or to a set whose members are indiscernible according to the criteria of the ‘net’ we’re using. We can’t reduce the proper-name unity to the common-noun unity, or else we break it. A proper name by definition is not supposed to apply to more than one being. In practice, of course, things may have the same name, but that’s not how proper names work in principle. And even when we happen to arrive at a net that sorts a category down to a set with one member, we can’t ensure that there couldn’t be more than one being in the set unless we turn the sortal term (the ‘net’) in effect into a proper name, and then we’ve broken that.8 We can name a lion ‘Lion’ and mean just this peculiar one we’ve met, but the two uses of ‘lion’ no longer function the same way from then on, and we show this in English by using the capital letter.” - E.P. Butler. On the Gods and the Good. Polytheist Leadership Conference, Fishkill, NY, July 12, 2014. https://henadology.wordpress.com/wp-content/uploads/2014/07/on-the-gods-and-the-good-final.pdf
This second way from Unity to Unities is basically Proclus’ argument in Book III, Chapter 1 of his Platonic Theology:
“It is necessary therefore, from the before-mentioned axioms, since there is one unity the principle of the whole of things, and from which every hyparxis derives its subsistence, that this unity should produce from itself, prior to all other things, a multitude characterized by unity, and a number most allied to its cause. For if every other cause constitutes a progeny similar to itself prior to that which is dissimilar, much more must The One untold into light after this manner things posterior to itself, since it is beyond similitude, and The One Itself must produce according to union things which primarily proceed from it. For how can The One give subsistence to its progeny except unically? For nature generates things secondary to itself physically, soul psychically, and intellect intellectually. The one therefore is the cause of the whole of things according to union, and the progression from The One is uniform. But if that which primarily produces all things is The One, and the progression from it is unical, it is certainly necessary that the multitude thence produced should be self-perfect unities, most allied to their producing cause. Farther still, if every monad constitutes a number adapted to itself, as was before demonstrated, by a much greater priority must The One generate a number of this kind. For in the progression of things, that which is produced is frequently dissimilar to its producing cause, through the dominion of difference: for such are the last of things, and which are far distant from their proper principles. But the first number, and which is connascent with The One, is uniform, ineffable, superessential, and perfectly similar to its cause. For in the first causes, neither does difference intervening separate from the generator the things begotten, and transfer them into another order, nor does the motion of the cause effecting a remission of power, produce into dissimilitude and indefiniteness the generation of the whole of things, but the cause of all things being unically raised above all motion and division, has established about itself a divine number, and has united it to its own simplicity. The one therefore prior to beings has given subsistence to the unities of beings.” - Proclus, The Theology of Plato. (Trans. Thomas Taylor). Prometheus Trust, 1995.
The “first number” here is the multiplicity of Unities, or Henads.
This was a key take-away of a recent class on Parmenides, where his ontology was read as affirming an irreducible multiplicity in order to relativize ontology as such, to make the latter open to critique.