I think that the problem of evil provides a pretty good argument against the hypothesis that an omniscient, omnipotent, omnibenevolent God exists. But I’ve been thinking off and on for a while now about various replies to the argument from evil, and more and more I’ve been thinking that the best reply is the many-universes reply. Since that reply isn’t discussed much in the literature on the problem of evil, I thought I’d present it here. 

This isn’t the most formal way to present it, but I’ll present it with a parable. Suppose that God exists, and God is omnipotent and omniscient, and has the desire to be omnibenevolent. So God creates a very nice universe, a universe with no evil. We might at first think that God has fulfilled the criterion of omnibenevolence, but then we recognize that God could do more — God could create another universe that’s also very nice. Agents could exist in that universe that didn’t exist in the first universe, and so there’s an intuitive sense (which is admittedly tricky to make precise mathematically) in which there would be more goodness to reality than there would be were God just to create one universe.

But of course there’s no reason to stop at two — God should create an infinite number of universes. Now, he could just create an infinite number of universes, where in each universe no evil things happen. But in doing so, there would be certain creatures that wouldn’t exist — creatures like us, who exist in a universe with evil, and are essential products of that universe. So God has to decide whether to create our universe as well. What criterion should he use in making this decision? My thought is that he should create all the universes that have more good than evil, and not create the universes that have more evil than good. 

So that’s why an omnipotent omniscient omnibenevolent God would create our universe, even though it has evil — our universe adds (in an intuitive sense, setting aside mathematical technicalities) to the sum total of goodness in the universe, and hence it’s worth creating. 

I’m not saying this argument is perfect, and I’m not saying that I’ve worked out all the kinks in this short blog post, and I’m not saying that all the kinks even can be worked out. But I do think that an argument like this provides the most promising reply to the argument from evil. So if I could write a paper showing that this argument doesn’t work as a reply to the problem of evil, I’d be happy. Perhaps I will work on that…

For more on the many-universe solution to the problem of evil, see for example Donald Turner’s “The Many-Universes Solution to the Problem of Evil” in The Existence of God edited by Gale and Pruss (which I haven’t yet read, but I’m going to) and the discussion by Hud Hudson in his book The Metaphysics of Hyperspace. Also see this paper by Klaas J. Kraay. Also see Section 6 of this paper by physicist Don Page.