The Problem of Evil holds that some facts about human suffering are evidence contrary to (or in contradiction with) classical theism. Philosophers distinguish between two kinds of evil (or suffering): moral evil — the sort caused by human actions (rapes, genocides, etc) — and natural evil — the sort that are not caused by human actions (the destruction caused by storms, earthquakes, etc). It is a matter of common philosophical consensus that the latter is more difficult for the theist to explain away than the former. In the case of the former, theists typically cite the fact that God has granted humans libertarian free-will and that it is the free choices humans make — which God can neither predict nor interfere with — that explains the existence of moral evil. In the case of the latter, it is difficult to see how theists can use the free-will defense (as this argument has come to be called) without appealing to either victim blaming (van Inwagen) or demons (Plantinga).
Nonetheless, in this post, I will consider an argument for the position that the existence of moral evil cannot be explained away using the free-will defense. The choices humans make are typically intelligible to each other; but, as I will argue, this can only be so if those choices are in some sense predictable. I will then leave the theist with a dilemma: either human actions are more intelligible to other humans than they are to God, in which case God’s omniscience is in trouble, or human actions are less intelligible to other humans than they are to God, in which case the free-will defense will be in trouble.
In section VIII of the Enquiry Concerning Human Understanding, David Hume considers several arguments opposing libertarian free-will. One of Hume’s arguments concerns the intelligibility of our social lives. We place expectations on each other and feel justified in holding each other to those expectations. If someone visits a far away country and returns with stories about humans behaving in ways that are too good to be true, we have reason to be skeptical. We have a mutual understanding of each other’s motivations and we find it reasonable that a good friend will not stab us unprovoked. Furthermore, Hume argues, history is only possible because humans in one place behave similarly to humans in other places. Ancient humans were not a different species from present day humans. Hume’s argument can be extended further: the only reason that the social sciences are possible is because human behavior is describable using something that at least approaches scientific laws (even if, as many philosophers of science argue, there are no precise laws in the special sciences).
None of this would be possible if humans were not predictable. The only reason that an expectation is formed within me concerning what my friends will do (or will not do) is because my experience with those friends will be repeated (in at least some sense). The reason we doubt stories concerning the too-good-to-be true individuals who supposedly occupy faraway lands is because our former experience with other humans allows us to form particular expectations about what humans will and will not do. We find our social lives to be intelligible because, most of the time, humans are not erratic. And when humans do behave erratically, it is not as though it is without cause; large differences in behavior often require large differences in causes. Sociology, economics, psychology, and the other social sciences are only possible because, at least statistically, humans are predictable. Otherwise, no experimental method could make sense of human behavior.
The free-will defense begins by pointing out that if libertarian free-will exists then there is no fact of the matter about what humans will do in the future. If there is no fact of the matter about what humans will do in the future, it is no violation of God’s omniscience for God not to know what humans will do. In comparison, no one would think that God knows triangles have six sides. No one can know that triangles have six sides because it is false that triangles have six sides and it is logically impossible to know false things to be true. Likewise, given libertarian free-will, God cannot know what any particular human will do because there is no fact of the matter about what any given human will do (just as there is no fact that triangles have six sides). There is no such fact to know.
The free-will defense continues by noting that God created humans not so that they could do evil, but so that they could freely choose to do good. A world where humans can only choose to do good, but not choose to do evil, is one in which libertarian free-will does not appear. Furthermore, a world in which libertarian free-will is absent is not as good as a world where libertarian free-will is present. Thus, as a perfectly good being, it would be contrary to God’s nature to create anything other than a world in which libertarian free-will existed. In what follows, I will grant the theist this entire picture: the only thing consistent with theism is the creation of a world in which there is libertarian free-will.
Immediately, this creates an obvious problem: we do not appear to live in a world where there is libertarian free-will. Thus, if we accept the free-will defense, then to the degree that we have scientific evidence contrary to the existence of libertarian free-will, we have evidence contrary to God’s existence. But put this to the side; perhaps the theist has some way of explaining away the existence of such evidence.
It is not at all clear to me what sort of explanation theists could provide to explain away the intelligibility of human actions, but whatever explanation they provide should be one which does not deny our experiences of our social lives or the data of the social sciences. It seems plausible that such an explanation is possible: notice that the predictability of human actions does not entail that every particular action is perfectly predictable. It remains a possibility that some actions are more likely and more reasonable than others without entailing that there is a fact about what any particular human will do in the future. Furthermore, notice that the sorts of things a particular human is likely to do most often relate to antecedent causes. For example, a trusted friend is not likely to stab me unless there is some reason that renders their action intelligible. Thus, that some humans will have a greater tendency to do good deeds (and others a greater tendency to do bad deeds) is not a violation of their free-will.
Now ask the following question: are human actions more or less intelligible to God than they are to other humans? If human actions are more intelligible to God, then God can predict with a greater degree of certainty what any particular human will do in any particular circumstance. Yet if God knows what any particular human is likely to do in any particular circumstance, God already possessed that knowledge when God created that human and, through Her providence, brought about whatever general state of affairs that human confronts. Therefore, God could have chosen to create humans who were less prone to violence, and other misdeeds, but chose not to. The Problem of Evil reasserts itself.
The only other possibility is that human actions are even more mysterious to God than they are to other humans. Yet, as an omniscient being, God’s knowledge is, by definition, superior to ours. To the degree that human actions are more intelligible to me than they are to God, I am more knowledgeable than God. This is a contradiction; thus, if the theist chooses this side of the fork, they must either abandon theism or libertarian free-will.
So I leave theists with a dilemma: either human actions are more intelligible to other humans than they are to God, in which case God’s omniscience is in trouble, or human actions are less intelligible to other humans than they are to God, in which case theists again face the Problem of Evil. I’m not sure this argument is completely convincing; leave me your thoughts in the comments below.