When we worry about the problem of evil, we tend to worry most about particularly horrendous evils. But consider the atheological inference from the occurrence of an evil to the claim that there would be no good enough reason for God to permit this evil. Is this inference any stronger in the case of a horrendous evil (e.g., genocide) than in the case of a trivial evil (e.g., uncomfortable pebble in shoe)?
Here is one possibility (which I got from Trent): Say that an evil is in principle unjustifiable iff it is not permissible to allow the evil no matter how great a good would be lost or how great an evil caused by preventing it. Obviously trivial evils are poor candidates for being in principle unjustifiable. So if the inference is: “E is in principle unjustified, and hence God has no good enough reason to permitted”, then the argument is indeed better off in the case of horrendous evils. Maybe there could be in principle unjustifiable evils. For instance: “everyone always suffering horrendously despite everyone being innocent”. But none of the evils we observe are in principle unjustifiable. For instance, let S be the mereological sum of all the human sufferings. Then, S is not in principle unjustifiable, because it would be permissible to allow S if the only way of preventing S would result in a hundred planets full of aliens each suffering in ways comparable to S.
Apart from in principle unjustifiable evils, then, is the atheological inference from horrendous evils better than that from trivial evils? I don’t know. On the one hand, allowing a horrendous evil requires a much more potent justificatory reason. On the other hand, many horrendous evils do very obviously bring along with them the opportunities for very great goods–exercises of courage, compassion, patience, forgiveness, etc. Granted, many will query if the value of the opportunities would be sufficient to justify God’s allowing the horrendous evil, and intuitions will differ, but at least we can, typically, point to a number of uncontroversial goods of quite high magnitude. On the other hand, with the trivial evils it can often be harder to point to any goods (e.g., consider the uncomfortable pebble in one’s shoe).
So maybe it’s not harder to say in the case of the horrendous evil “There is a justification, but we don’t know what it is” than in the case of a trivial evil. But if so, then why is it that when we worry about the problem of evil, we are more apt to worry about greater evils than smaller ones?
I suppose one possibility is that one might think that there is a good G served by having a large number of minor discomforts, but one might also think that no one minor discomfort is such that its loss would cause any measurable loss of G. So maybe God is justified in allowing a large number of minor discomforts, without God needing a reason for allowing each particular one. (Dean Zimmerman and Peter van Inwagen (and others, I suspect) have made arguments along these lines.) But reasoning like this seems repugnant in the case of horrendous evils. While it is easy to say: “I didn’t prevent you from having the pebble in your shoe because a good would be lost if I always prevented pebbles in shoes, and I had to draw the line somewhere”, to say the same thing about genocides seems morally insensitive.
But even though it seems morally insensitive, something like that line might be right. Suppose that some country keeps on burning down villages, including all the innocent villagers, near their border. You can prevent any particular such act. However, you also know that if you prevent too many such acts, then the country’s powerful ally will nuke you. Then you have to prevent some of the horrendous evils, but with respect to others you just have to say: “We had to draw the line somewhere.” If that’s right, then maybe the vagueness line works for horrendous evils, and not just trivial ones.
But even if the vagueness line doesn’t work for horrendous evils (and I am myself sceptical of the line), the horrendous evils may not need it. For the reason we seemed to have needed the vagueness line for trivial evils was the worry that any particular trivial evil contributes insignificantly to the good that justifies having many trivial evils. But any particular horrendous evil produces significant goods. Of course, as I said, there is a question of judgment whether these goods are sufficient to justify permitting the evil, but I still don’t see that saying that either these goods, or others that we are not aware of, justify the horrendous evil is any more problematic than saying that some good justifies having a lot of trivial evils.
I, too, have the intuition that the problem of evil is more of a problem when the evils are horrendous. But I don’t know that the intuition is correct.