Friday, April 20, 2018

Metaphysical Whack-A-Mole

Here’s an argument for the existence of God:
  • There’s no contradiction in an omnipotent being spontaneously coming into existence.
  • Nothing that was not omnipotent would be able to prevent an omnipotent being coming into existence. (Think of it like a game of metaphysical whack-a-mole.)
  • In a given time period, if there is no contradiction in something happening and there is nothing to prevent it from happening then there is a non-zero probability that it will happen.
  • So in any given time period at the beginning of which there is no omnipotent being, there is a non-zero probability that an omnipotent being spontaneously comes into existence.
  • So over an arbitrarily large time period the probability that an omnipotent being spontaneously comes into existence will be arbitarily close to certainty.
  • There has been enough time that we can be practically certain that an omnipotent being has spontaneously come into existence.
  • Once an omnipotent being exists, it will see to it that it continues to exist, since it is omnipotent and wants to continue to exist.
  • There cannot be more than one omnipotent being, since it follows from their omnipotence that they would be both able and unable to frustrate each other’s intentions, and this is a contradiction.
  • If there is exactly one omnipotent being, then that being is God.
  • So we can be practically certain that God exists.
  • So God exists.

What do you think? Let me know in the comments!


  1. Well, three and six seem like pretty bold assumptions, three being the more interesting one. Is the thought maybe that denying three involves positing brute impossibilities, and that's bad?

    1. Thanks for commenting!

      Three has two readings. On the first reading, we take "probability" to refer to objective chance, and for that we do need to rely on a kind of robust denial of brute impossibilities. On the second reading, we take "probability" to refer to subjective probability, and for that we only need a kind of epistemic modesty about what might and might not happen.

      The plausibility of Five relies largely on an equivocation between these two readings, since the Robust Denial Of Brute Impossibilities is considerably less plausible, but the Epistemic Modesty About Modality doesn't really support the idea that the probabilities are independent. We could sensibly assign a significant subjective probability to the proposition that omnipotent beings popping into existence just isn't the kind of thing that can happen.

      This is the issue I take to be the main weakness in the argument.

      Six is indeed quite bold, but becomes considerably less bold if we grant the premise that the past stretches back infinitely, as many sensible people have thought it does. I think it'd be unfortunate if granting this premise made the argument compelling, but as I say I think the main weakness is in an equivocation between subjective and objective probability.

  2. Point 6 seems dubious without a calculable value for point 4. Point 1 seems unlikely to me, but what do I know.

    1. Thanks for commenting. Those are both fair criticisms.

      Point 6 does mostly rely on the past being infinite, and the probabilities for each disjoint time period being independent, both of which are significant weaknesses of the argument. As I said in the reply to Thomas I think the independence assumption is the main weakness.

      Point 1 is also contestable. One ground for contesting it is that it's contradictory for there to be an omnipotent being at all. Another is that it might be contradictory for there to be an omnipotent being that didn't exist either at all times or outside time.

      In response to that, I'd say the argument would still be interesting if it showed that granting that omnipotence wasn't contradictory we could practically certain that God existed.

      Also, if omnipotence is contradictory, the argument could be modified to demonstrate that the universe is probably dominated by a succession of super-beings, each one more skilled at metaphysical whack-a-mole than the previous one. Although this modified version of the argument might be weaker at point 8.

    2. My instinct for what its worth is that an omnipotent being is intrinsically contradictory, but that doesn't preclude an arbitrarily powerful being. But I would have no problem believing that the probability per unit time for a sufficiently powerful being to pop into existence would give a rate of far less than 1 per 15bn years. If the past is infinite then all bets are off, at least mathematically speaking.


    3. Though, addendum: I think one talks about a merely super-powerful being operating within physical limits, it behoves us to ask how such a beings power would manifest itself. Is it plausible, for example, to imagine it could create stars, or plan out the development of other life forms? Would it, in fact, have any relevance to us on earth? And is this a reformulation of the 'we almost certainly live in a computer simulation' argument?

    4. I don't think it's true that all bets are off if the past is infinite, because if superbeings popping into existence just isn't the sort of thing that can happen, then it won't happen no matter how long you wait.

      If we're talking about Galactus popping into existence via an unlikely quantum fluctuation and then operating within physical limits, I don't expect the argument would work because the beings wouldn't be immortal. The argument relies on the no-God state being less stable than the God state, so we expect to end up in the God state. But it's probably easier for Galactus to be destroyed than for him to be created, and so the no-Galactus state is more stable.

      In view of this kind of thing, the argument is intended to be about supernatural beings rather than ones that obey the laws of physics. But then we have to contend with the possibility that there's literally zero chance of these things ever popping into existence, and so we shouldn't necessarily expect them to show up even given an infinite amount of time.

    5. I don't see that it has very much to do with the simulation argument. I guess it has a similar flavour to Boltzmann brain scepticism in that it appeals to things popping into existence, but the stable state/unstable state issue doesn't arise for Boltzmann brains, and it's an important part of the argument.

    6. I'm skeptical about Boltzmann brains because I fundamentally doubt the popular perception of quantum mechanics as 'random fluctuations', but if these did occur they would be subject to physical constraints. Whereas a simulator running our universe from the outside could potentially reengineer those laws whenever they wanted. The difference is that in the simulator argument they idea is that we will eventually become the gods in another universe, and that one already exists for ours since its initiation, rather than emerging.

    7. I guess the gods I'm picturing in the weaker version of the argument are more like the simulators than I'd thought, then. But The structure of the argument is quite different. The simulation argument says that the most likely explanation of the world being as we find it is that it's a simulation dominated (from the outside) by relative superbeings. Whereas my argument is more or less an a priori argument that if wait long enough you should expect the world to end up dominated by a superbeing.

  3. This comment has been removed by the author.