The Paradox of the Stone

One of the most famous problems with omnipotence is the stone paradox. I’m sure most all of us have come across this in form or another. The problem derives from the question: ‘can God create a stone so big that He couldn’t lift it?’ If He can, He is not omnipotent because there is something that He can’t do, namely lifting the stone. If He cannot, then He is not omnipotent because there is something that He cannot do, create the stone. Here is the argument in premise conclusion format:

  1. Either G can create a stone too heavy for G to lift or G cannot create a stone too heavy  for G.
  2. If G can create a stone too heavy for G to lift then G is not omnipotent.
  3. If G cannot create a stone too heavy for G to lift then G is not omnipotent.
  4. Therefore G is not omnipotent.

Note that this paradox only arises if we think that God is able to do anything. There are other restricted definitions of omnipotence that would avoid this paradox. But is this even really a decisive argument against the position that an omnipotent being is capable of doing anything? I am convinced that it is not. In fact that it is a quite poor argument against this understanding of omnipotence.

First we need to think about what is in the range of ‘anything’ in the actions that God could perform. Does ‘anything’ range over just all logically possible things or does it extend to God being able to do logically impossible things? Lets restrict ‘anything’ to mean only logically possible things. Either an unliftable stone is a logically possible thing or a logically impossible thing. If the stone is a possible object then God can make it but it is not threatening to God’s omnipotence that He cannot lift it because it would be impossible to lift an unliftable thing and God’s omnipotence does not allow Him to do anything. If the stone is an impossible object then God could not make it and this would not threaten His omnipotence since an omnipotent being cannot do impossible things. Now let’s say God can do impossible things. It does not matter whether or not the stone is an impossible object. God can create a stone that He cannot lift. But this is not a problem for His omnipotence because God can lift the stone that He cannot lift because His power extends to all things possible and impossible. I think that this is a conclusive objection to this argument. I do think there are other objections to omnipotence that are strong but the paradox of the stone is just not one of them.

Post by Cruz Davis



  1. Cruz,
    If we restrict “anything” to mean only logically possible things, then I become concerned that God is restricted by logic, i.e. his power is limited by logic. This seems like a threat to God’s proposed omnipotence.

    Given the meaning of “anything” to include the logically impossible things, then we enter the realm of intelligibility, in which case the conversation continues on the intelligibility post, would you agree?

    • Sean,

      I do not think restricting the quantifier to only logically possible things is a real restriction. If we think the idea of being able to making impossible things is not only unintelligible but nonsensical, then when we say God is able to perform all actions that are not nonsensical actions are we committing ourselves to nonsensical actions? I just think that the defender of omnipotence as the ability to perform anything logically possible would think that the limit of actions there are or could be are going to all be not nonsensical in their description. So this isn’t a limitation it is just that there isn’t anything else besides the not nonsensical actions.

      I do agree with the later. But I do not think that this hurts the objection to the stone paradox. If we find the ability to do logically impossible things unintelligible then we will just say that the scope of omnipotence would be restricted to all logically possible things, but then we have another way out of the stone paradox. So this is why the ‘paradox’ fails to show what it is meant to show.

      I hope that helps. Let me know if you have anymore questions.

      • Cruz,

        I have to admit I have a really difficult time understanding the first paragraph to your reply, however I think I have grasped it. Therefore, this reply could be totally off. First let me summarize in my own words what I think you said, and hopefully I am not drawing any straw men from it.

        I think you are saying that the defender of omnipotence defined as “the ability to perform anything logical possible” should have no problem with an omnipotent being not being able to perform illogical things. I partly agree with this. However the problem I have with this is that we can (and have) fathomed that omnipotence could be defined as “the ability to perform anything logically or illogically possible”. I think that our ability to fathom this reveals the “restriction” or “shortcoming” in our former definition of omnipotence, namely, that omnipotence as previously defined is not as “omnipotent” as the later definition of omnipotence. I see here that I am making a sort-of ontological argument for the “most omnipotent” definition for omnipotence.

        So long as this “ontological argument for the most omnipotent definition of omnipotence” holds, then a so-called ‘omnipotence’ restricted to the ability to do only logical things will be insufficiently omnipotent.

        I hope this makes sense. If it does, I think it is most important to attack the “ontological argument for the most omnipotent definition of omnipotence”.

  2. Sean,

    Let’s call omnipotence restricted to logical possibility OPL and omnipotence that is not OPI.

    What I was saying is that the defender of OPL is going to think that she is restricting omnipotence in anyway. There are no logically impossible actions that one can perform, so there is nothing that is restricting what kinds of actions an omnipotent being can perform.

    Another thing to note is that we are not talking about different kinds or degrees of omnipotence. These are different definitions of what it means for something to be omnipotent. It is assumed that there is one right or at least a best translation of the term. So we should be careful to not think of OPL and OPI as different version of omnipotence but competing theories of omnipotence.

    I’m not sure if I get the ontological argument bit. Let me get back to you.

    • Summatheoblogica,

      Thanks for defining OPL and OPI, this should help for clarifications.

      Did you mean to say “the definder of OPL is NOT going to think that she is restricting omnipotence in anyway”? I think you meant to?

      For the ontological argument bit, since OPI is conceivably more powerful than OPL (in the sense that OPI = OPL plus the ability to do the illogical), then omnipotence should be defined as OPI is defined.

      However, if omnipotence is defined as OPI is defined, then we enter the realm of intelligibilty.

      So I feel there is a bit of a Catch 22 with regards to the acceptance or rejection of OPI as the true definition of UO:
      It seems impossible to reject OPI as an OPI UO is (as defined above) more powerful than an OPL UO, however if we accept OPI, things become unintelligible.

      • Sean,

        Sorry it has taken so long to get back to you. I hope you enjoyed your holidays.

        Yes by OPL I did intend to put that negation in there.

        You are claiming that an OPI being would be ‘more powerful than and OPL’. The point the defender of OPL will make is that your claim is not an intelligible one. (Or it would at most just be vacuously true). It doesn’t make sense to say that something would be more powerful than an OPL. I think this kind of response is what the defender of OPL would use get out of your problem.

      • @Summa

        If impossibility isn’t a problem for an OPL being, doesn’t it follow that an OPL being could make it the case that there is something more powerful than an OPL being? I can’t imagine anything that an OPL defender could say that couldn’t be replied to with ‘Yes, but OPL being can make the opposite of that the case.’ We can’t analyze this kind of thing without indulging in nonsense.

  3. Riggs,

    I think you mean OPI instead of OPL.

    The point of this post was to show that there are two obvious ways out of the stone paradox. I agree that OPI is not a sensible position. But I think there are attempts at understanding OPI that are worth looking at, which is what the posts on Conee are about.

Comments RSS TrackBack Identifier URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s