Must God be Unique?
Presenting and analyzing Al-Taftazani's argument
Arguments in the Kalam tradition usually proceeded with the following three steps:
An argument to establish the existence of a necessary being.
An argument to establish that the necessary being in question is a personal agent that is omnipotence, omniscience, etc.
An argument to establish the uniqueness of this necessary being.
These three steps, taken collectively, were understood to constitute an argument for the existence of God. Most people are relatively familiar with the arguments for (1) and (2), but I don’t see many people talk about the arguments for (3). So here I’ll present Al-Taftazani’s argument for the third premise, from the mutual hindrance among gods.
Assume we have two omnipotent beings, O1 and O2. In virtue of their omnipotence, when O1 or O2 wills a possible situation to take place, it must take place. Now assume we have some contingent state of affairs, let’s say whether some person moves. Since whether they move is contingent, it’s both possible that the person moves and that they don’t. That the person moves falls under the scope of O1’s power, which is to say O1 could wish that they move and have it obtain in all possible worlds. In a similar vein, that they don’t move falls under the scope of O2’s power; if O2 wishes it, the person not moving could obtain in all possible worlds.
Let there be a possible world in which O1 wishes for the person to move and could realize it, but O2 wishes for the person not to move and could realize it. There are three distinct outcomes here. First, the person both moves and doesn’t move. This is obviously a contradiction and is therefore ruled out. Second, the person does move. If this were to occur, the will of O1 would obtain at the expense of the will of O2. But if O2 were to will something, it must happen in virtue of their omnipotence, and so O2 must not be omnipotent. Similarly, if the person were not to move, the will of O2 would obtain at the expense of the will of O1 not obtaining, and so O1 must not be omnipotent.
In any case, there can be only one omnipotent being. Since there can be only one, assuming there is at least one personal omnipotent being, there is only one. God is unique.
A few issues could be raised with this argument. First, one could point to the general observation that impossible tasks do not fall under the scope of divine power. We wouldn’t say God fails to be omnipotent because He cannot instantiate a square circle, since it would be impossible to do. Similarly, if two omnipotent beings disagreed, they would simply block each other’s powers. The task in which their volitions contradict becomes impossible for them to realize either way. Since impossibilities do not threaten omnipotence, the tasks in which they disagree do not threaten their omnipotence.
You may say, well, in the case illustrated above, the person will either move or not move, and when they do, they will be in accord with the will of one of the omnipotent beings and not the other, and so the other mustn’t be omnipotent. Note, however, that under the objection being provided here, the state of affairs is consistent with the will of either O1 or O2, but not due to the volition of O1 or O2.
Say that the above objection fails. One may still ask, “Doesn’t Al-Taftazani’s argument just show that if there were multiple omnipotent beings, they must necessarily have compatible wills?” If they necessarily have compatible wills, it would be impossible for them to disagree on what contingent states of affairs obtain, and we would avoid having to pick which one’s will is violated in order for the other’s will to obtain. This seems like a plausible route.1
God’s omnipotence is grounded in His absolute perfection, and from His absolute perfection would arise the will (and in turn the act) to instantiate the perfect order; this would similarly hold for our second omnipotent being. Divine nature seems to delimit divine will. So, for there to be disagreement over some contingent state of affairs, there must be two distinct perfect orders of things. This could only be the case if there are arbitrary features of reality that are irrelevant to its perfection, allowing the perfect orders to differ. Can this be the case? There do seem to be arbitrary features of our reality. Is it seriously important to the perfection of reality that I sit an inch away from my desk rather than two? I’m skeptical that it is, to say the least. If so, you can’t ground the necessity of the compatibility of will between omnipotent beings in their divine nature. In that case, it’s hard to see what could ground the necessity of compatibility of volitions. This objection fails in that case.
An objection one may raise is that if wills necessarily share the same desires and volitions, it’s not clear they’re even distinct. I don’t mention this, in part because I don’t know how I feel about it. If it’s a condition that, for wills to be distinct, it’s possible for them to differ, then one cannot respond to Al-Taftazani’s argument by saying the omnipotent beings necessarily share all volitions, but I don’t really have an argument for such a thesis, though it doesn’t sound like a totally implausible principle.
Maybe an interesting connection with Christian theology here: Trinitarians in fact do hold that the persons of the Trinity share a single (numerically identical) will, and so cannot possibly disagree.