Philosophy of Religion

Does Limiting God remove His Omnipotence?

Today in class I was in discussion with a classmate of mine about the nature of God’s omnipotence. She made an interesting claim that the God of Aquinas is not all-powerful. I went on to explain that Aquinas’s take on God was that God is indeed all-powerful in that He could instantiate every state of affairs that is logically possible. This implies that God cannot instantiate states of affairs that are logically impossible, i.e., self-contradictions. For example, God cannot make it so that a square-circle, or a married bachelor, or a meat-eating vegetarian exists.

She countered that theologians would say that God’s omnipotence implies that He can instantiate even contradictions and thus, since Aquinas’s God cannot do that, Aquinas’s God is limited and thus not omnipotent.

We can put this into the following form:

  1. If God is omnipotent, then God can actualize any state of affairs (taking any to mean all, including impossible affairs).
  2. God is omnipotent.
  3. Therefore, God can actualize any state of affairs.

Now, if you haven’t caught on, this statement of God’s omnipotence is extremely problematic, and I maintain that anyone who holds onto this, even Theologians, is holding a rationally untenable view. It doesn’t matter to me whether person X is a Christian theologian or not, this view is untenable and here is why.

Suppose God can actualize any state of affairs. That means God can actualize the state of affair where “God both exists and does not exist.” Clearly, no one can actualize that. Not even God. Not because it’s not within God’s power to do that, but because the state of affairs in itself is impossible. No amount of power can bring this about because such a state of affairs cannot obtain no matter what. Ever. My classmate claimed that if God cannot do that, then he is limited to the laws of logic, and if God is limited, then he cannot be omnipotent. The assumption is that if God is limited in anyway, then God cannot be omnipotent. But is that really true? Does any limitation on God mean he is not all powerful? Why even think this? But let’s suppose it’s true. It seems to me that God will always be limited in some fashion. For instance, if God is all-good and he is perfect, then surely God cannot err nor do evil. It’s not possible to do so. He is limited in his ability. This almost sounds like the sophomoric objection that “Look! I can sin and God can’t do that! Ah-ha! I can do something God can’t, therefore he is not all-powerful!” Surely, this if foolish. Moreover, do we really think He is not omnipotent simply because He cannot pick his nose or release gas? I mean, God doesn’t have a body to do those things, so God is limited. Is he thus not all powerful? Clearly this is not the case. Therefore, it seems the God being limited is not an adequate reason to conclude that God is not omnipotent. It just doesn’t follow.

Based on the above definition of omnipotence, God is not omnipotent. But this is not the view of omnipotence that Christian philosophers take God to have, let alone Aquinas. Aquinas brilliantly explains,

It remains therefore, that God is called omnipotent because He can do all things that are possible absolutely; which is the second way of saying a thing is possible. For a thing is said to be possible or impossible absolutely, according to the relation in which the very terms stand to one another, possible if the predicate is not incompatible with the subject, as that Socrates sits; and absolutely impossible when the predicate is altogether incompatible with the subject, as, for instance, that a man is a donkey.

Furthermore, he adds,

Therefore, that which implies being and non-being at the same time is repugnant to the idea of an absolutely possible thing, within the scope of the divine omnipotence. For such cannot come under the divine omnipotence, not because of any defect in the power of God, but because it has not the nature of a feasible or possible thing. Therefore, everything that does not imply a contradiction in terms, is numbered amongst those possible things, in respect of which God is called omnipotent: whereas whatever implies contradiction does not come within the scope of divine omnipotence, because it cannot have the aspect of possibility. Hence it is better to say that such things cannot be done, than that God cannot do them. Nor is this contrary to the word of the angel, saying: “No word shall be impossible with God.” For whatever implies a contradiction cannot be a word, because no intellect can possibly conceive such a thing. (ST I, Q. 25, Art. 3)

I think it’s safe to say that my classmate was gravely mistaken on Aquinas’ take on omnipotence, on the idea that God’s omnipotence entails that God can do anything, and on the idea that any limit on God makes him not omnipotent.


I’m not sure if something similar like this has already been published somewhere academically or on the web. If so, please link it and accept my apologies, and I’ll give credit where it’s due. ¬†I came up with this by just sitting in my room and arguing with my girlfriend’s best friend about her atheism. I was just having a bit of fun with Aquinas’s 5 ways and the KCA (Kalam Cosmological Argument). This argument is a bit of synthesis of Aquinas and the KCA. I like it because it makes a nice air-tight argument when put together. Of course, one could just stick to the KCA or one of the 5 ways by itself.

  1. Whatever begins to exist has a cause.
  2. The universe began to exist.
  3. Therefore, the universe has a cause. [1, 2]
  4. Either the universe’s existence was self-caused or it was not self-caused.
  5. The universe’s existence was not self-caused.
  6. If the universe’s existence was not self-caused, then it was caused to exist by something else.
  7. Therefore, the universe’s existence was caused by something else. [5, 6]
  8. If the universe’s existence was caused by something else, then there cannot be an infinite regress of causes of the universe’s existence.
  9. Therefore, there cannot be an infinite regress of causes of the universe. [7, 8]
  10. If there cannot be an infinite regress of causes of the universe, then the ultimate cause of the universe must be uncaused.
  11. Therefore, the ultimate cause of the universe must be uncaused. [9, 10]
  12. Time, matter, and space came into existence with the universe.
  13. If time, matter, and space came into existence with the universe, then the cause of the universe must be timeless, immaterial, and spaceless.
  14. Therefore, the cause must also be something timeless, immaterial, and not located in space. [12, 13]

I simply won’t go into too much detail defending the premises of the Kalam argument (premises 1 and 2) since it’s pretty familiar to most philosophers, and I’m assuming you’re here reading this far because you like philosophy. If not, then please see this post.

Premise five holds true because it’s absurd to think that something can cause itself to begin to exist. For something to cause itself to begin to exist is to presume that that something already exists beforehand to do the causing. So it exists and it doesn’t exist (since it has to begin to exist) and is thus a contradiction.

Premise six is an obvious truth. If it didn’t cause itself to exist then something else must have.

Premise eight is defended elsewhere in another post of mine. See the part on an infinite regress of causes found under the defense of premise 1.

But what about premise 10? Couldn’t one simply argue its falsity because of idea that there could be a multiverse (MV) and thus time, matter, and space didn’t really come into existence with the universe? One could object that. Besides the fact that there isn’t really any concrete evidence for the MV, I think even if we grant its existence, additional argumentation would establish that the MV would be caused by something uncaused.

Premise twelve is true given contemporary big bang cosmology. Also, William Lane Craig has defended this premise in detail else where. Cf. his book Creation Out of Nothing.

And the consequent in premise thirteen follows since this cause is prior to the existence of time, space and matter. Otherwise the cause would have had to come into existence with the universe, thus lending the cause to exist (prior to the universe) and not exist (since it had to come into being with the universe), which is absurd.

Lastly, premise fourteen follows necessarily from 12 and 13 via modus ponens.

Now, here’s the rest of the argument in which I try to show that even assuming that there’s a MV, it must have been caused by something uncaused.

15. Everything that’s caused must be caused by another.

16.¬†Suppose there’s a MV.

17. If there’s a MV, then another universe, U2 (not the band ha), in the MV caused our universe, U1, to exist.

18. U2, U3, and so on, are either self-caused, eternal, or caused.

19. The universes are not self-caused or eternal.

20. Therefore the universes of the MV are caused.

21. There cannot be an infinite regress of causes [R, 9]

22. Therefore, the ultimate cause of the universes must be uncaused.

23. The MV is simply the mere collection of the universes.

24. Therefore, the ultimate cause of the MV must be uncaused.

25. If there’s a multiverse, then the ultimate cause of the MV must be uncaused. [>I, 15-24]

This argument may be a worthless and failed attempt, it may be decent, or it may be good. Not sure. I had fun thinking about it and throwing it together.

In chapter 14 of The One and the Many: A Contemporary Thomistic Metaphysics, Norris Clarke presents a few arguments for the the ultimate source of all being, which he takes to be God. Here is my summary of the arguments he presents in the chapter. In part 1 of this series, I’m going to try to summarize and explain Clarke’s first argument from a caused being to the infinite source or cause of all being.

Argument 1: From any conditional being to one infinite source of all being

  1. There must exist at least one self-sufficient or uncaused being.

Clarke has readers look directly to our everyday experience. We notice that everything that we encounter is contingent, that is to say that it could have been the case that the stuff we see didn’t exist. In Clarke’s own words they are “not self-sufficient for their own existence but depend on other beings outside of them for their [existence], either to bring them into existence or to maintain them in existence, or both” (216). To take his example, look at ourselves. We depend on our parents for existence, and also the oxygen that we take in and the nutrients we receive from the food we consume. Our existence is dependent on these factors (and many others).

Clarke takes an objection, and a common one at that. Can there be an infinite regress of causes? So being A is caused to exist by being B, and when we get to being B we see that it too is caused to exist by being C, and so on and so forth. Can it be that this chain stretches out towards infinity? Clarke rightly answers no. Here’s why: suppose we have a an infinite amount of boxcars that are all connected to each other and it’s in motion. We ask ourselves, “What caused boxcar A to move?” Well it’s the pushing of boxcar B. Similarly we asked about what caused boxcar B to move, and we say C. We keep asking, and asking, and asking but never arrive at a point in which we can say it’s motion began. We are left with no reason to suppose why or how it began it’s motion. But of course if there is no reason or cause of it’s motion, then the boxcar would be motionless since these boxcars have nothing in its being which would cause it to move.

Or look at ourselves. Assuming we have free will, when I move my arm to scratch my nose, the movement of my hand is actualized by my arm, my arm by my muscles, and my muscles by neurons firing (or whatever else goes on in this complicated process). Eventually we must get to a point in which there’s simply an unmoved mover. Something to get the process started and going. My arm can’t just be suspending in the air scratching my nose for an infinite amount of time without being caused by something. And if these causes go back infinitely, we have no reason to suppose that my arm would be suspended in the first place.

So to make sense of the motion of the boxcar, my arm, and the existence of the contingent “stuff” we see, there must be at least one uncaused or unmoved being or “thing” to get it all started.

Clarke’s rejection of an infinite regress is stated this way:

Given that being A here and now exists, categorically, not conditionally (i.e., as an “is,” not an “if” statement). Now suppose one tries to explain the actual existence of A thus: A exists only if B, B only if C, C only if D, and so on to infinity, in an endless series of “only if” statements. In this case, since each member depends on the conditions for its existence being fulfilled by another, and these conditions in turn remain endlessly unfulfilled, the entire series remains conditionally (“iffy”) in its existence. Unless one of the members along the line exists unconditionally, categorically, with no more conditions to be fulfilled, then the original existence of A itself becomes only conditional. There could never be a categorical affirmation of anything at all (nothing but “only ifs”). But the original A does exist, categorically and unconditionally, as a fact, not “iffily.” (218-219).

Clarke concludes that “there must exist at least one self-sufficient being” (219).

2. Any being that is self-sufficient for its own existence or is uncaused must be infinite in perfection.

One way Clarke supports this is via reductio ad absurdum:

  1. Suppose there exists a finite self-sufficient being.
  2. This being would have to be the total ultimate source of all attributes within it, including existence.
  3. It cannot be the case that the ultimate source of a perfection would posses this perfection in a limited way, imperfect way since the being in question is the ultimate source of this perfection.
  4. But in virtue of being finite, this finite being possesses these perfections in a limited, imperfect, or less than full way.
  5. Therefore, there does not exist a finite self-sufficient being.

So, if there is a being that is self-sufficient for it’s existence, it follows, then, that this being cannot possess existence or any other perfection in a limited, imperfect, or less than full way, and thus it cannot be finite. This being can be either finite or infinite. Thus it must be infinite in perfection.

3. There can be only one such infinite being.

This argument is represented straightforwardly by another reductio ad absudrum:

  1. Suppose there were two such beings, that is, two self-sufficient infinite beings.
  2. These two beings are distinct from one another and cannot in anyway be each other.
  3. If these two beings are completely distinct from one another, then one of them lacks something the other doesn’t.
  4. Therefore, one self-suffucient infinite being lacks something the other doesn’t.
  5. If each being didn’t lack something the other had, then there would just be one being since the beings would be entirely identical to each other.
  6. If one lacked something the other one doesn’t, then one being lacks a perfection.
  7. A self-sufficient infinite being is one who has all perfections
  8. One of the two self-sufficient infinite being’s lacks a perfection
  9. Thus, there cannot be two self-sufficient infinite beings.

Clarke thus concludes the argument in the following way:

“Either the universe is unintelligible, or there must exist one and only one Infinite Source of all other beings, both of their actual existence and of all the perfections (goodness) within them. It is at once Infinite Being and Infinite Goodness, in which all finite beings participate both for their existence and their goodness” (221). In my next post, I’ll cover Clarke’s second argument.


Anthony Kenny’s and Bertrand Russell on Aquinas’ Christian theism (in particular his Catholicism):

“There was,” writes Russell, “little of the true philosophical spirit in Aquinas : he could not, like Socrates, follow an argument wherever it might lead, since he knew the truth in advance, all declared in the Catholic Faith …

[But] it is not,” replies Kenny, “in fact a serious charge against a philosopher to say that he is looking for good reasons for what he already believes in. Descartes, sitting around his fire, wearing his dress gown, sought reasons for judging that that was what he was doing, and took a long time to find them. Russell himself spent much energy seeking proofs of what he already believed: ‘Principia Mathematica’ takes hundreds of pages to prove that 1 and 1 make 2 … We judge a philosopher by whether his reasonings are sound or unsound, not where he first lighted on his premises or how he first came to believe his conclusions. Hostility to Aquinas on the basis of his official position in Catholicism is thus unjustified, however understandable, even for secular philosophers.”

— Anthony Kenny, agnostic — “A New History of Western Philosophy Volume 2: Medieval Philosophy,” pg 76
In undergraduate classes, I’ve seen atheists discredit Christian philosophers simply because these philosophers seek out reasons to demonstrate the truth of their faith. This shouldn’t matter. If there are sound reasons for one’s belief, it shouldn’t matter that the person tries to seek reasons to hold their belief. If no reason can be found, then the next rational thing should be to let go of that belief.