This is the weekly Q & A blog post by our Research Professor in Philosophy, Dr. William Lane Craig.
As I am transcribing your latest Defenders lectures on the problem of evil, I was hoping someone would ask the question, but I don't think it has been asked. So, maybe you can attend to it next week?
Instead of arguing that "even though some evils look gratuitous, they really aren't" (i.e. we can't discern what greater good will come out of any evil — this greater good could occur centuries later in another country), why not come up with an argument that says, yes, gratuitous evil does exist (since it seems more obvious than not that it does exist), but that that somehow doesn't refute God's existence?
Specifically, have you read Kirk MacGregor's response to the problem of evil, and what are your thoughts? I am referring to this: Kirk R. MacGregor, "The Existence and Irrelevance of Gratuitous Evil", Philosophia Christi, Vol. 14, No. 1, 2012. He refutes the notion that there is no gratuitous evil, and he says "it does not follow from either logical improvability or statistical inscrutability of gratuitous evil that gratuitous evil does not exist." He goes on to say belief in the existence of gratuitous evil is properly basic (among other things) so why fight it? To agree that gratuitous evil exists "seems a far more fruitful path for Christian theists to take in terms of apologetic tactics, since it neither places upon theistic shoulders the overwhelming burden of proof of showing that none of the world's evils is gratuitous ... nor does it ask hearers of theistic arguments to abandon their cognitive insight that gratuitous evil exists."
But what he does argue is that if God exists and has created a universe then gratuitous evil exists (he gives several arguments for this — this is what he does in the second half of his paper). In fact, he goes as far as to say gratuitous evil necessarily exists if God were to create a universe ("All such evils are, in and of themselves, gratuitous or pointless; their only reason d'etre is the logically unavoidable privation of ontological necessity exhibited by created entities" — I take it here he is targeting natural evils (earthquakes, tsunamis, etc.) — he gets into moral evils and the issue of libertarian free will in paragraphs after this).
Anyway, that's my question. In short, why shoulder the heavy burden of trying to prove gratuitous evil does NOT exist? Instead, assert that it exists, but show that it is not incompatible with God's existence (or go further, like MacGregor does, and show that "if God exists then, gratuitous evil exists").
Dr. William Lane Craig’s Response
You’re referring to Part 33 of our Excursus on Natural Theology, where I discuss the so-called evidential or probabilistic version of the problem of evil. I haven't read Kirk's article, though I'm puzzled that a Molinist should think that there is gratuitous (pointless or unjustified) evil. I've read Peter van Inwagen's defense of the reality of gratuitous evil, but he denies middle knowledge. I’m not opposed to the claim in principle, but I don’t take that route because it concedes too much to the atheist, namely, that gratuitous evil exists.
The key move in the evidential problem of evil is the atheist’s inference from “Much of the evil in the world appears to be pointless” to “Much of the evil in the world really is pointless.” In view of our inherent cognitive limitations, which I describe in my lecture, I’m very sceptical that the atheist can justify that inference. As I emphasized, my response does not "place upon theistic shoulders the overwhelming burden of proof of showing that none of the world's evils is gratuitous." Rather, it is wholly the atheist's burden to prove that because some evils appear pointless, they really are pointless. So, as I explained in response to a questioner, there is no "heavy burden of trying to prove gratuitous evil does NOT exist." The theist has nothing to prove here (hence, this response is often called “skeptical theism”).
As a result of reading Stephen Wykstra’s contribution to a forthcoming book on the problem of evil, in which I defend a Molinist perspective, I have actually come to see, however, that we should not argue, as I did, that "even though some evils look gratuitous, they really aren't". But the reason is because that statement also concedes too much to the atheist! It is not really true, as I had conceded, that some evils look gratuitous.
Wykstra’s claim seems at first outrageous. But he carefully differentiates between (i) not seeing that evil has a point and (ii) seeing that evil has no point. The scope of the negation is crucial.
Wykstra gives the following illustration. Suppose you're visiting the doctor to get a vaccination and the nurse drops the needle on the floor. She picks it up, examines it, and says, "It appears that there are no bacteria on the needle. Pull up your sleeve." Would you go ahead? No, for while it does not appear that there are bacteria on the needle, you would not say that it appears that there are no bacteria on the needle.
So with respect to some evils in the world, it does not appear that they have a reason; but, in view of our inherent cognitive limitations, it is false that they appear to have no reason.
So the atheist argument turns out to be even more difficult to make than I thought. He cannot show from the fact that evils appear to be pointless that they are pointless (my point), but worse, he cannot even show that they appear to be pointless (Wykstra's point).
The lesson once again: Don't let the atheist shift the burden of proof.
 For example, take the question of whether there is life on other planets. It is true that “It does not appear that there is life on other planets,” but it is false that “It appears that there is not life on other planets.” If the negation has a wide scope—“NOT (there appears to be life on other planets)”—the statement is true, but if the negation has a narrow scope—“It appears that NOT (there is life on other planets)”—, it is false.