The problem of evil concerns the question of whether it is possible to reconcile the existence of "evils" in the world (wickedness, death, suffering) with the existence of a perfectly good, omnipotent God. The argument from evil is an argument that purports to show that these cannot be reconciled, and, therefore, since evils do exist, there cannot exist a God who is both perfectly good and omnipotent.
The argument from evil can succeed only if we have some idea what it means to say that God is omnipotent or perfectly good. I will discuss the property of omnipotence in the next lecture. Filling in the substance of the property of goodness requires the use of ethical theory. Every argument from evil is going to rely on assumptions about ethics. If these assumptions are dubious, and especially, if they contradict tenets normally associated with theism, then the argument will be ineffective.
There are two problems with relying on ethical theory in natural theology. First, we are clearly fallible in constructing ethical theories. There are, and have been for thousands of years, a fairly large number of competing ethical theories and little consensus has emerged. If we find that our ethical theory is in conflict with our natural theology, it may be the ethical theory that has to change. Second, we find, within ethical theory, very few ethical principles that admit of absolutely no exceptions. Even if God's act of creating the world violates certain prima facie duties, it may be that God's action is merely an exception to the corresponding rules and not in fact a wrongful action.
Thinking about God's goodness brings us back to another issue we have discussed before: the issue of nominalism versus realism. If we are nominalists, then to say that God is good is to say that God's character and motivation resemble those of a paradigmatically good human being. In contrast, if we are realists, then to say that God is good is simply to attribute to God the very same property possessed by good human beings, the very property that makes good human beings good. This property of goodness may express itself very differently, depending on the nature of the bearer.
I am going to assume, for the sake of these lectures, that we embrace some form of realism. Realism commits us to the principle of analogy, namely, the principle that the very same property can express itself very differenty in beings of different natures. Since God and human creatures have radically different natures, God being infinite and necessary, human creatures being finite and wholly contingent, God's goodness can amount in practice to something quite different from human goodness. This makes it much more difficult to argue a posteriori from God's actions to the state of God's character. It also makes it difficult to reason from hypotheses about God's character to specific actions or inactions.
Nonetheless, we are not left in a state of complete agnosticism about God's moral qualities. First of all, we may be able to find a priori, metaphysical arguments for or against God's goodness, that are independent of an evaluation of God's actual behavior. Second, we can reason more confidently about God's actions whenever God voluntarily takes on a role that bears great similarity to the social roles of human beings. For example, if God deigns to use human language, God's speech acts are very similar to our own, and moral principles about truth-telling would presumably apply to God's speech as well as to our own. Similarly, if God enters into a practice of treaty-making or promise-keeping, the moral norms governing those practices would apply to God's actions. In contrast, God's act of creating and sustaining the cosmos is radically dissimilar to any human action, and, consequently, we should be very diffident about applying human norms to these actions.
It would seem that the decrease or elimination of evil is always a good thing, and the decrease or elimination of good is always a bad thing. This complementarity has led many philosophers and theologians (most notably Augustine of Hippo) to adopt the "privative theory of evil". According to this theory, every evil just is the absence of some corresponding good. This is why the elimination of some evil consists in the introduction of the corresponding good, and the elimination of some good constitutes the introduction of the corresponding evil.
It is possible that the privative theory of evil is false, but if it were false, this would leave us with no explanation for the complementarity of good and evil. Similarly, it is conceivable that both cold and heat exist, even though neither one is the privation of the other, but this would leave unexplained why an increase in heat is always a decrease in coldness, and why an increase in coldness is always a decrease in heat. We could adopt, I suppose, a privative theory of goodness, according to which every good consists in the absence of some corresponding evil, but this seems intuitively less plausible. Good things are typically positive in a way that bad things are not: life vs. death, knowledge vs. ignorance, health vs. disease, love vs. indifference.
If the privative theory of evil is true, this has some bearing on the problem of evil. It means that we do not need to find a cause of the existence of evil. Only positive existences need causes. We must find a cause for the existence of every good thing, but we need no cause for evil, which is simply the absence of some possible good. It may be that we need to find a causal explanation of each evil, where the causal explanation can consist in the non-existence of a cause of the corresponding good. However, the chain of causal explanations need not lead back to the First Cause. In some cases, the non-existence of certain potential causes may be an inexplicable brute fact.
Thus, we should not ask the question, why did God cause or create evil? Since evil is only a privation, it is not the sort of thing that can be caused or created. However, we can still ask the question, why did God not cause or create more good? Nonetheless, this second question may have no interesting answer. God may have had no reason not to do certain things He did not in fact do. He may simply have chosen freely not to do those things. To demand that God must have a reason for not doing anything He does not do is to assume the Principle of Sufficient Reason, and to assume, in fact, that whatever happens happens by necessity. In contrast, to insist that God must have a reason for doing everything (positive) that He does, is simply to assume that God is rational and purposeful.
If evil is merely a privation, then evil is not caused. If evil were caused, then it would seem reasonable to hold that whatever causes evil must itself be evil. This would force us to attribute evil to the First Cause. However, since evil is never caused, this regress is unsuccessful. A perfectly good being can cause a situation which involves evil, by simply not realizing every potentiality for good in that situation.
Let's suppose that the argument from evil is entirely successful. What would be the upshot? The conclusion of the argument from evil is this: there cannot exist a God who is both omnipotent and wholly good. Note that the argument of evil does not establish the truth of atheism. Even if we assume that omnipotence is an essential part of the definition of God, we are still left with two options: there is no God, or there is a God who is not perfectly good.
There are three possibilities concerning God: eutheism, dystheism and atheism. Eutheism is the thesis that God exists and is wholly good. Dystheism is the thesis that God exists but is not wholly good. Atheism is the thesis that God does not exist. The argument from evil, if it is successful, establishes the falsity of eutheism, leaving us with the two remaining options. If the argument from evil is to be used as an argument for atheism, we need some argument against dystheism.
This leads us to the subject of a priori arguments for God's moral goodness. The atheist needs such arguments, no less than the eutheist does, since unless God could not exist without being morally good, the argument from evil cannot establish God's non-existence. I will mention briefly four traditional arguments for the goodness of God.
God is perfect, and evil is a privation, so God cannot be evil. Moral evil is the absence of some possible good, such as wisdom, love or self-control. God cannot lack any possible good, so God cannot be evil in any respect.
God is identical to the property of goodness itself. There must be such a property, since good things do exist. Since God is the first cause of all good things, God must be good Himself. Since God is absolutely simple, God is identical to His own essence. Hence, God is identical to His own goodness. Since God is the first cause, God's goodness is the universal form of goodness. Therefore, God is identical to the property of goodness itself.
The property of goodness must be perfectly good. Thus, God is perfectly good.
If we suppose the God is perfectly good, this hypothesis can be used to explain how it is that we have moral knowledge. A perfectly good God would both know all moral truths and desire to share that knowledge with all rational creatures. Hence, we can expect that a perfectly good God would have created us with an innate disposition toward reliable knowledge of moral truth. If God were not morally good, either because He lacked moral insight or because He lacked good will, we would have no explanation of how we, His creatures, came to have reliable moral knowledge.
If God were a mixture of good and evil, then the two would have an equal footing in reality. Nonetheless, it seems that there is a fundamental asymmetry between the two, and this asymmetry must have its root in God's character. The simplest explanation of this asymmetry would be to posit that God is wholly good and utterly lacking in evil.
If the argument from evil is to be an argument for atheism, the arguments for God's moral goodness must be stronger than the arguments for God's existence. Otherwise, the argument from evil would give us reason to embrace dystheism, rather than atheism. Although the arguments for God's moral goodness above have some force, they seem considerably weaker than the cosmological or design arguments. Consequently, it would seem that the upshot of the problem of evil should be to lead us toward dystheism and not toward atheism.