Limited Omniscience is compatible with moral perfection

I have a correction to make after reading Richard Swinburne's Providence and the Problem of Evil (specifically p. 231-232).  An argument that I thought was sound is actually not a good one. As has been explained before, Swinburne believes that God doesn't know the future with absolute certainty. He may know what is probable, but he isn't absolutely sure that it will happen. Michael Martin in his book, Atheism: A Philosophical Justification brings up an objection to this. I summarize his argument on my page:

"Limited Omniscience vs. Moral Perfection

If God doesn't know the future, then he can't know whether or not his past actions were morally correct. Whether or not an action is morally correct depends on, in part, what the consequences are. However, God cannot know the consequences of his actions, therefore He can't know that He is morally perfect. God could still choose all the correct actions. However, this would have no relation to his knowledge. He would do the right thing purely on accident. Because no being can be morally perfect on accident, it follows that a being with "limited omniscience" cannot be morally perfect. Since God is supposed to be such a being, it follows that God cannot exist. [2]"

As Swinburne explains however, this objection has a false premise. What makes an action morally correct is, in part, the probable consequences of the act. The actual consequences only matter insofar as they are probable. Someone can't be blamed for what happens if they have no indication whatsoever that it will happen. Since God will know how probable all outcomes are, he can do the act that maximizes the expected value. The act that he does will be the right one.

So, since someone can't be blamed for the actual consequences of an act if there is no indication that those consequences will happen, then what makes an act right or wrong is the probable consequences. Therefore, the argument has a false premise. So, Martin's argument is unsound even though I once thought it worked. I will change it on my site and explain why the argument doesn't work.

It is important to note that this response to the argument only works if we suppose it is impossible for a being to know more than God does about the world. For example, God doesn't know the future on Swinburne's understanding. We also have to give a reason why God cannot know the future in order for this response to stick. Otherwise, one could say that a morally perfect being must know the future which would be a possible thing to know. If one shows that knowing the future is impossible if humans have libertarian free will, then the concept of moral perfection must not require a being to be morally perfect in order to fulfill it; otherwise moral perfection would be necessarily nonexistent.

So, if one defines "moral perfection" in a way that is possible for a being to fulfill it, then one must not need the requirement that the being who is morally perfect knows the future with absolute certainty. If one needs this requirement, then their definition of moral perfection is inapplicable to any possible being. This doesn't show a problem with the idea of God being morally perfect in any sense, just a problem with the sense of moral perfection that the arguer is using. If one defines a concept in a self-contradictory way, one cannot be surprised when it cannot be applied to any being at all.

And for the record, I still think the argument above it titled The Free Will Argument for the Nonexistence of God still works. However, I think it only disproves a God who can know the future. It doesn't disprove Swinburne's God who doesn't know the future with absolute certainty.