An Argument Against the Possibility of Multiple Omnipotent Beings and Some Implications for the Trinity and Incarnation

Argument: Suppose there are two omnipotent beings G1 and G2. Now, suppose that G1 wills that Earth gets created and that G2 wills that Earth does not get created. Who “wins”? On the one hand, Earth should get created since G1 wills it, G1 is omnipotent, and the creation of Earth is logically possible. On the other hand, Earth should not get created since G2 wills it, G2 is omnipotent, and Earth not being created is logically possible. The answer therefore seems to be that Earth gets created and Earth does not get created, a flat contradiction. Such a state of affairs, therefore, is impossible. And yet, such a state of affairs seems like it would be possible if there really were two omnipotent beings (since it seems possible that the two beings could will such mutually contradictory states of affairs). Hence, it must not be the case that there are two omnipotent beings. Even more strongly, it must not be possible for there to be two omnipotent beings. And more generally, it must not be possible for there to be any number of omnipotent beings greater than one, since the same paradox can be derived no matter how many more omnipotent beings we keep adding beyond two. The argument can be formalized as follows:

1. If there can be multiple omnipotent beings, then there can be two mutually contradictory (but individually logically possible) states of affairs willed by multiple omnipotent beings.
2. If there can be two mutually contradictory (but individually logically possible) states of affairs willed by multiple omnipotent beings, then there can be a logical contradiction.
3. There cannot be a logical contradiction.
4. Therefore, there cannot be two mutually contradictory (but individually logically possible) states of affairs willed by multiple omnipotent beings (2, 3).
5. Therefore, there cannot be multiple omnipotent beings (1, 4).

From this we can also conclude, given that God is essentially omnipotent, that there cannot exist more than one God.

Objection: In order to assess whether Earth being created is logically possible for G1 to actualize, we must take into account the will of G2. So, we can resolve the paradox by noting that it is only logically possible for G1 to create Earth if it is the case that G2 does not will that Earth not be created. And the omnipotence of a being is not infringed upon by the being in question not being able to bring about logical impossibilities. Mutatis mutandis for G2 bringing it about that Earth is not created with respect to G1’s will. The paradox, therefore, is resolved.

Reply: It seems like what we’re saying is that G1’s power is limited by G2’s power. In other words, G1 can only do something if G2 permits it. But in that case, we are essentially saying that G2 has power over G1. But then, G2 actually has more power than G1; therefore, G1 is not omnipotent after all. G2 is the sole omnipotent being. Parallel reasoning also allows us to derive the conclusion that G1 has power over G2 and that therefore G1 is the sole omnipotent being. So, we continue to be mired in contradictions. Furthermore, this objection leads to a vicious infinite regress: G1 can only do something if G2 permits it. But in the same way, G2 can only do something if G1 permits it. So, G1 can only do x if G2 permits x. But G2 can only permit x if G1 permits that G2 permits x. But G1 can permit that G2 permits x only if G2 permits that G1 permits that G2 permits x. And on and on we go! The objection, therefore, fails.

Objection: The philosopher Richard Swinburne suggests that we can avoid paradox with multiple omnipotent beings if there were some sort of mechanism to ensure that the two (or more) omnipotent beings would not interfere with one another. What might such a mechanism be? Swinburne sketches three possibilities:

There are many different ways in which unity of action can be secured among individuals who might otherwise impede each other's efforts. One of them could take all the decisions and the others simply execute those decisions. Another way is to have a vote on every issue and for each then to carry out the result of the vote. A third way is to have a division of functions. One individual takes decisions on certain kinds of issues, and the others support him in these. Another individual takes decisions on other issues, and the others support him in those, and so on. Which would be the best way for Gods to secure unity of action, to determine a choice between equally good but incompatible actions? The first way would seem an imperfect way of sharing power between Gods, and so one which G₁ would not adopt. The second way, taken strictly, is not a possible way when there are only two individuals, for, unless they agree in advance, votes will always be tied. (Marriage cannot be a democracy.) And where there are more than two individuals, but many alternative actions (such that there is no overriding reason for doing one rather than another), is there any reason to suppose that there will often be a majority in favour of one course of action? Only the third way would seem a viable way of securing unity of action in shared power among Gods.

Such unity of action could be secured if the first God solemnly vows to the second God in creating him that he will not frustrate any action of his in a certain sphere of activity, and expresses the request that in return the second God should not frustrate any action of his in the other sphere. The vow of the first God would create an obligation on him not to frustrate any action within his allocated sphere of activity. So, although the first God retains his omnipotence, it is, as before, limited by the inability to do evil and in virtue of his promise this limitation will ensure that he does not frustrate the actions of the second God. Conversely, although the power is given to the second God, it comes with a request that it should not be exercised in a certain way. The overall goodness of conformity to that request (not to conform would be not to conform to a reasonable request from the source of his being and power) will ensure that, although omnipotent, the second God cannot frustrate any action of the first God. The sharing of divinity could (logically) only occur subject to some restriction preventing mutual impediment of action. I have presented a highly fallible human judgment as to what the best such mechanism (and so the one which would be adopted) would be (“Could There be More Than One God?”, Faith and Philosophy).

Reply: The primary issue I see with all three suggestions is that none seems to be metaphysically necessary, which is what I think Swinburne would need in order to resolve the paradox that arises with the supposition that there are multiple omnipotent beings. Why, for instance, should it be metaphysically necessary that the Gods organize in such a way that one takes decisions on certain kinds of issues, while others take decisions on other issues? Why should it be metaphysically necessary that one God promises not to interfere with another? If these suggestions are not metaphysically necessary, then it is possible that the Gods all try to take decisions on the same kinds of issues, which would lead to the paradox identified above: mutually contradictory decisions might be willed by different omnipotent beings.

Implications for the debate on the nature of the Trinity and the Incarnation: There are two main models of the Trinity in philosophical theology: social trinitarianism and Latin trinitarianism (alternatively, “anti social trinitarianism”). Philosophers JP Moreland and William Lane Craig describe these two models as follows:

The central commitment of social trinitarianism is that in God there are three distinct centers of self-consciousness, each with its proper intellect and will. The central commitment of anti social trinitarianism is that there is only one God, whose unicity of intellect and will is not compromised by the diversity of persons. Social trinitarianism threatens to veer into tritheism; anti social trinitarianism is in danger of lapsing into unitarianism (Philosophical Foundations for a Christian Worldview, pg. 582).

Given the above argument for the impossibility of multiple omnipotent beings, we seem to have strong reasons for accepting anti social trinitarianism over and against social trinitarianism. For although the Trinity is a plurality of persons and not a plurality of beings, the argument against multiple omnipotent beings derives a paradox from the assertion that there are multiple omnipotent wills. And social trinitarianism implies this very assertion. On the orthodox view of the Trinity, each person in the Trinity is fully God. Thus, each person in the Trinity is omnipotent. And, on social trinitarianism, each person of the Trinity has His own independent will. From these facts, we can derive the same paradox we derived above. Therefore, we have grounds to reject social trinitarianism and instead subscribe to anti social trinitarianism.

The unicity of will in the Godhead that anti social trinitarianism is committed to is also biblically supported by passages such as John 10:30. This passage is cited by some in support of modalism, the heretical view that rejects the doctrine of the Trinity. According to the modalist, the three “persons” of the one God are simply the different “modes” of being of the one and only one person who is God. As an example, a single person can have three different modes of being: being a father, being a son, and being a lawyer. Although there are three modes of being in this case, there is nevertheless only a single person. Since, however, modalism is false, a common understanding of John 10:30 is that Jesus is teaching that He and the Father are one in being (but still distinct persons), one in purpose, and one in will.

This conclusion has relevance to the Incarnation. Since Scripture speaks of Jesus having a will distinct from the Father (Luke 22:42, John 6:38), it must be (given the truth of anti social trinitarianism and our understanding of John 10:30) that in the incarnated Christ, there are two wills: the divine will and a human will. This view of the Incarnation is called dyothelitism (alternatively, dythelitism). This is in contrast to monothelitism, the view that in the incarnated Christ there is only one will. Monothelitism was condemned under an anathema by the Church at the Third Council of Constantinople, and its rejection is also considered a basic point of orthodoxy by most Protestants.

Dythelitism also helps to make sense of Jesus being tempted by Satan in the wilderness in Matthew 4:1-11. Hebrews 2:18 and 4:15 confirm that Jesus was tempted. Now, if Jesus only had a divine will and not also a human will, then it seems that Jesus could not have genuinely been tempted because God cannot be tempted by evil (James 1:13). Thus, assuming the truth of monothelitism, Jesus’ “temptation” in the wilderness is reduced to a show and a sham. By contrast, Jesus’ genuine temptation isn’t problematic at all on dythelitism because Jesus has not only the divine will but also a human will. And a human will can be tempted. Thus, Jesus’ human will was tempted by Satan, even while His divine will was not. Nevertheless, even though Jesus' human will could be and was tempted, it could not actually sin. For an account of this fact, the interested reader can check out a previous post of mine.