1 The Issue:
The specific question assigned to us for discussion is this: Grounds for belief in God aside, do the evils in our world make atheistic belief more reasonable than theistic belief? The initial clause in this question is important. For it is one thing to argue that the evils in our world provide such compelling reasons for atheism that the reasons for the existence of God are insufficient to swing the pendulum back in favor of the existence of God, and another thing to argue that, putting aside whatever reasons there may be for believing that God exists, the evils that occur in our world make belief in atheism more reasonable than belief in theism. If we put aside grounds for belief in the existence of God, the likelihood that God exists cannot reasonably be assigned any probability beyond 0.5—where 1 represents God’s existence as certain, and 0 represents certainty that God does not exist. So, if we start from an initial point of God’s existence having a likelihood of 0.5 or less, and restrict ourselves to the evidence generated by the enormous amount of horrendous evil that occurs daily in our world, it should strike anyone that the likelihood of God’s existence can only go downward from 0.5. To reach such a judgment is perfectly consistent with holding that once the reasons supportive of the existence of God are brought into the equation, the likelihood of God’s existence is in fact positive, somewhere between 0.5 and 1. So, we should not confuse arguing that the negative evidence of evil shows God’s existence to be unlikely, even taking into account the positive reasons there are to think that God exists, with arguing that, putting aside the positive reasons for thinking that God exists, the evils that occur in our world make atheistic belief more reasonable than theistic belief. The issue in this discussion is only the latter: Apart from taking into account the positive reasons for thinking that God exists, do the evils that occur in our world make atheistic belief more reasonable than theistic belief? I shall argue that they do.
Before proceeding to argue this point, however, it is important to be clear on what theism is. Theism is the view that there exists an all-powerful, all-knowing, perfectly good being (God). We can call this view restricted theism. It is restricted in that it does not include any claim that is not entailed by it. So, theism itself does not include any of the following claims: God delivered the Ten Commandments to Moses, Jesus was the incarnation of God, Muhammad ascended into heaven. These are claims made in specific theistic religions; thus they are a part of an expanded form of theism: Judaic theism, Christian theism, and Islamic theism. The importance of not taking theism to include the claims held by only one particular religion among the three major theistic religions of the West is that the inclusion would make theism less likely; for if we identify theism with a particular one among the great theistic religions, then the truth of theism itself is made to depend on all the essential beliefs of that particular theistic religion. The other side of this coin is that philosophers who wish to defend theism ought not to suppose that the assumption of theism entitles them to assume any of the special claims associated with their own particular theistic religion. Since most of the philosophers in the Anglo-American tradition who defend theism are adherents of some version of Christian theism, they should beware of confusing the assumption that theism is true with the altogether different, and less likely, assumption that Christian theism is true.
2 The Argument:
Do the evils that occur in our world significantly lower the likelihood of God’s existence? Let’s begin thinking about this problem by considering a simple argument from the existence of some of the evils in our world to the nonexistence of God.
1. There exist horrendous evils that an all-powerful, all-knowing, perfectly good being would have no justifying reason to permit.
2. An all-powerful, all-knowing, perfectly good being would not permit an evil unless he had a justifying reason to permit it.
Therefore,
3. God does not exist.
If theists reject this argument for the nonexistence of God, they must reject either the first premise or the second premise. Most theists accept the second premise, as do nontheists. So, most theists must reject the first premise, holding instead that God has a justifying reason for permitting each and every horrendous evil that occurs. But what would be a justifying reason for God to permit some terrible evil he could prevent? Since an evil is something that by its very nature is bad, God’s justifying reason for permitting it would have to include something else—either some outweighing good that, all things considered, he wishes to realize and cannot realize without permitting that evil, or some equal or worse evil that, all things considered, he wishes to prevent and cannot prevent without permitting that evil. And the question we must ask ourselves is whether it is rational for us to believe that all the terrible evils that occur daily in our world are like that? Is it rational to believe that each evil is such that were an all-powerful, all-knowing being to prevent it, he would have to forfeit some outweighing good?
Perhaps it will make the issue before us a bit more concrete if we focus on some examples of terrible evils, rather than just terrible evils in the abstract. Here are two examples:
A fawn is horribly burned in a forest fire caused by lightning. It lies on the forest floor suffering terribly for five days before death relieves it of its suffering.
A five-year-old girl is brutally beaten, raped, and strangled in Flint, Michigan, on New Year’s Day in 1986.
The theist must believe that for each of these evils there is some greater good to which it leads, a good that an all-powerful being simply could not realize without permitting that evil. But is what the theist believes about these two evils really so? Is there really some great good that an all-powerful being could bring about only by permitting that fawn to be badly burned and to suffer intensely for five long days before death relieves its torment? And is there really some great good that an all-powerful being could bring about only by permitting that little five-year-old girl in Flint, Michigan, to be savagely beaten, raped, and strangled? And even if it should somehow be so in these two cases, is it true that all the instances of intense human and animal suffering occurring daily in our world lead to greater goods in such a way that even an all-powerful, all-knowing being could not have achieved any of those goods without permitting the instances of suffering that supposedly lead to them? In light of our knowledge of the scale of human and animal suffering occurring daily in our world, the idea that none of those instances of suffering could have been prevented by an all-powerful being without the loss of a greater good must strike us as an extraordinary idea, quite beyond our belief. And if it does strike us in this way, the first premise of the argument we are considering—that there exist horrendous evils that an all-powerful, all-knowing, perfectly good being would have no justifying reason to permit—is bound to strike us as plausible, as something quite likely to be true. But since the second premise is generally agreed to be true, we should then reason that it is likely that our conclusion is true, that God does not exist.
It is important here to understand two points about the argument just presented. First, the argument is not, nor is it meant to be, a proof that God does not exist. To be a proof of its conclusion, an argument must be such that its conclusion logically follows from its premises and its premises are known with certainty to be true. The argument we are considering meets the first condition, but not the second. The conclusion follows deductively from the two premises, but its first premise is not known with certainty to be true. The claim is only that the first premise is one that we are rationally justified in believing to be true. And since our confidence in the truth of the conclusion should not exceed our confidence in the premises from which it follows, the claim is only that the premises provide sufficient rational support for that conclusion. Second, the truth of the first premise does not logically depend on any claim about the two examples, of the fawn and the five-year-old girl. The examples are meant to illustrate the profound difficulty in really believing that an all-powerful, all-knowing being is incapable of achieving his noble ends without having to permit such horrendous, undeserved suffering. But if there were only a few such examples as these, perhaps it would not be unreasonable to believe that somehow even an infinitely intelligent, all-powerful being could not achieve his good ends without permitting them. But, of course, our world is not like that. It is the enormous amount of apparently pointless, horrendous suffering occurring daily in our world that grounds the claim in the first premise that there are pointless evils in our world, evils that an all-powerful being could have prevented without forfeiting some outweighing good. But, again, it is not being asserted that the existence of pointless evils is known with certainty, only that it is quite likely that pointless evils occur.
3 Evaluating Two Responses:
3.1 First response
Having looked at a particular argument from evil against theistic belief, we can now consider and critically evaluate two theistic responses to this argument. The first response that the theist may put forth goes something like this:
“The first point I want to make is that thus far we have been given no reason at all to think that premise 1 is true. For all you have pointed out is that we don’t know what the good is that justifies God in permitting any of these horrendous evils, like the fawn’s suffering or the little girl’s suffering. But to argue from the fact that we don’t know what the good is that justifies God in permitting a certain evil to the conclusion that there is no such good is to engage in a fallacious argument from ignorance: we don’t know of any justifying good, therefore there isn’t any. So, you haven’t really given any good reason at all to think that there are terrible evils for which there are no God-justifying goods. All you have shown, if you have shown anything, is that if these evils do serve some God-justifying goods, we don’t know what they are. And the interesting question to ask about our ignorance of these justifying goods is this: Given that God’s mind infinitely transcends ours, is it really at all likely that the goods for the sake of which he permits much horrendous suffering will be goods we comprehend? After all, isn’t God in relation to us like good, loving parents in relation to their small child? Such parents may permit their very young child to suffer a painful surgical procedure for a good the child simply cannot comprehend. So, too, we should expect that if God exists, he may permit many instances of human or animal suffering so as to realize goods that our minds simply cannot comprehend. And if that is so, the fact that we don’t know the goods that justify God in permitting much horrendous suffering cannot really be a reason for thinking he doesn’t exist. For it is just what we should expect to be true if he does exist.”
What are we to make of this response by the theist? Are we really just arguing from ignorance? Perhaps we can come to see that we are not by first distinguishing between goods we know about (goods within our ken) and goods beyond our ken. Consider the suffering of the five-year-old girl as she was brutally beaten, raped, and strangled on New Year’s Day in 1986 in Flint, Michigan. I believe that no good we know about justifies God in permitting that suffering. By “goods we know about” I mean goods that we have some cognitive grasp of, even though we may have no knowledge at all that they have occurred or ever will occur. For example, consider the good of the little girl experiencing complete felicity in the everlasting presence of God. Theists consider this an enormous personal good, perhaps the greatest personal good possible for the little girl. So, even though we don’t have a very clear grasp of what this great good involves, and even though we don’t know that such a good will ever be actualized, I include the good of her experiencing complete felicity in the everlasting presence of God among the goods we know about. Of course, if some good we know about does justify God in permitting her suffering, that good must have already been actualized or be actualized at some point in the future. But the notion of a good we know about extends to many future goods and to goods that never have and never will occur. And what we have good reason to believe is that none of the goods we know about justifies God in permitting the horrendous suffering of that little girl. For with respect to each such good that we consider, we have reason to believe either that it isn’t good enough to justify God in permitting that evil, or that it could likely be actualized by God without his having to permit the horrendous suffering of that little girl, or that some equal or better good could likely be actualized by God without his having to permit the horrendous suffering of that little girl.
Of course, even granting that we know of many great goods and have reason to think that none of these goods justifies God in permitting the little girl’s suffering, there still remains the possibility that some good we cannot even conceive does so. And it is here that the theist may appeal to the analogy between the good parent and God. For we cannot deny that some good the child’s mind cannot even conceive may justify the parents in permitting the child to suffer. And, by analogy, won’t the same be true of God in relation to us as his children? Indeed, since the disparity between his mind and ours may greatly exceed that of the good parents’ minds and the mind of their child, isn’t it likely that the goods that justify him in permitting us to suffer will often be beyond our comprehension? But against this argument from analogy, two points need to be made.
First, although arguments from analogy are rather weak, the analogy in question has some merit if drawn between a good parent and a good deity of considerable, but nevertheless finite, power and knowledge. For, like the good parent, a deity with great but finite powers may reasonably believe that he cannot realize some important future good for some of his creatures without permitting a present evil to befall them. And there may be occasions when, like the good parent, the finite deity is simply unable to prevent a dreadful evil befalling his creatures even though there is no good at all served by it. But the theistic God has unlimited power and knowledge. A good parent may be unable to prevent some suffering that her child undergoes, or even the child’s death from some painful disease. Can we seriously think that an infinitely powerful, all-knowing deity was powerless to prevent the horror of Auschwitz? A good parent may see that she cannot realize some important future good for her child without permitting some present evil to befall the child. Can we seriously think that there is some far-off future good for the victims of Auschwitz, a good that a deity of infinite power and knowledge judged to be worth the horror of Auschwitz, and was powerless to achieve without permitting that horror? Perhaps we can if we turn from reason to faith. But the infinite distance between the God of traditional theism and the good mother with the sick child doesn’t, in my judgment, provide human reason with good grounds for thinking that such a being would be powerless to prevent many of the countless, seemingly pointless horrors in our world without losing some goods so distant from us that even the mere conception of them must elude our grasp.
But suppose we do reason from the good-parent analogy to the behavior of an all-powerful, all-knowing, infinitely good deity. I think we shall see that the good-parent analogy leads in a different direction from what its proposers desire. We know that when a good, loving parent permits her child to suffer severely in the present for some outweighing good which the child cannot comprehend, the loving parent then makes every effort to be present to their children’s consciousness during their period of suffering, giving special assurances of her love, concern, and care. For the child may believe that the parent could prevent her present suffering. So the parent will be particularly careful to give her child special assurances of her love and concern during this period of permitted suffering for a distant good the child does not understand. And indeed, what we know about good, loving parents, especially when they permit their children to suffer intensely for goods the children cannot comprehend, is that they are almost always present to their children’s consciousness during the period of their suffering, giving special assurances of their love and care. So, on the basis of the good-parent analogy, we should infer that it is likely that God, too, will almost always be consciously present to humans, if not other animals, when he permits them to suffer for goods they cannot comprehend, giving special assurances of his love for them. But since countless numbers of human beings undergo prolonged, horrendous suffering without being consciously aware of God’s presence or any special assurances of his love and comfort, we can reasonably infer either that God does not exist or that the good-parent analogy is unable to help us understand why God permits all the horrendous suffering that occurs daily in our world.
Our conclusion about the theist’s first response is this. The argument in support of premise 1 is not an argument from ignorance. It is an argument from our knowledge of many goods and our reasonable judgment that none of them justifies God in permitting instances of horrendous evil. It is also an argument from our knowledge of what a being of infinite power, intelligence, and goodness would be disposed to do and would be capable of doing. Of course, there remains the logical possibility both that some goods incomprehensible to us justify God in permitting all these horrendous evils that occur daily in our world and that some further goods incomprehensible to us justify God in not being consciously present to so many who endure these horrendous evils. So, we cannot prove that premise 1 is true. Nevertheless, the first response of the theist should, I believe, be judged insufficient to defeat our reasons for thinking that premise 1 is probably true.
Before turning to the theist’s second response, we should note that some theists will protest the conclusion that we’ve come to about the first response. Here is what such a theist may say:
“Your distinction between goods we know about and goods beyond our ken is well-taken. Moreover, you are right to insist that your argument is not a flagrant example of an argument from ignorance. But there is one quite important point you have failed to establish. It is crucial to your argument that we should expect to know the goods for the sake of which God permits much terrible suffering or, failing such knowledge, be particularly aware of God’s presence and his love for us during the period of intense suffering for goods we cannot comprehend. For if we have no good reason to expect to know these goods, or to experience God’s presence and love during our suffering, then the fact that we don’t know them and don’t experience God’s presence and love won’t really count against the existence of God. And my point is that God may have good reasons (unknown to us) for not revealing these goods to us. And he also may have good reasons (unknown to us) for not disclosing himself and his love during the period when many suffer terribly for goods they cannot comprehend. How are you able to show that this point of mine is just a mere logical possibility and not the way things really are? I think you need to treat more seriously than you do the distinct possibility that God’s reasons for permitting so much horrendous suffering, and his reasons for not being consciously present to us during our suffering, involve goods that are presently incomprehensible to us.”
The theist here raises an important point. Using the theist’s own good-parent analogy, I argued that there is reason to think that when we don’t know the goods for the sake of which God permits some horrendous suffering, it is probable that, like the good parent, he would provide us, his children, with special assurances of his love and concern. Since many endure horrendous suffering without any such special assurances, I suggested that we have further reason to doubt God’s existence. And the theist’s only reply can be that there are still further unknown goods that justify God in not being consciously present to us when we endure terrible suffering for the sake of goods beyond our ken. And I’ve allowed that we cannot prove that this isn’t so. It clearly remains a logical possibility. I’ve said, however, that we can conclude that premise 1 is probably true. But the theist says that I’m not justified in concluding that premise 1 is probably true unless I give a reason for thinking it likely that there are no unknown goods that justify God in permitting much horrendous suffering or no unknown goods that justify God in not being present to us when we endure suffering for the sake of unknown goods. The theist may grant me that no goods we know of play this justifying role. But before allowing me to conclude that it is probable that premise 1 is true and, therefore, probable that God does not exist, the theist says I must also provide some grounds for thinking that no unknown goods play that justifying role.
Suppose we are unsure whether Smith will be in town this evening. It is just as likely, say, that he will be out of town this evening as that he will be in town. Suppose, however, that we do know that if Smith is in town, it is just as likely that he will be at the concert this evening as that he won’t be. Later we discover that he is not at the concert. I conclude that, given this further information (that he is not at the concert), it is now less likely that he’s in town than that he’s out of town, that given our information that he is not at the concert, it is more likely that he is out of town than that he is in town. I do admit, however, that I haven’t done anything to show that he is not actually somewhere else in town. All I’ve established is that he is not at the concert. I acknowledge that it is logically possible that he’s somewhere else in town. Nor do I know for certain that he is not somewhere else in town. All I claim is that it is probable that he is not in town, that it is more likely that he is not in town than that he is in town. Those who want to believe that Smith is in town may say that I’m not justified in concluding that it is probable that he’s out of town unless I give some reason to think that he is not somewhere else in town. For, they may say, all I’ve done is exclude one of the places he will be if he is in town. Similarly, the theist says that if God exists, then either all the horrendous evils we consider serve unknown goods or some of them serve goods we know of. We might even agree that if God exists, it is equally likely that some of the justifying goods will be known to us as that all of the justifying goods will be beyond our ken. After all, when we understand why God may be permitting some terrible evils to occur, those evils will be easier to bear than if we haven’t a clue as to why God is permitting them to occur. Suppose we then consider the goods we know of, and reasonably conclude that none of them justifies God in permitting any of these horrendous evils that abound in our world. The theist may even agree that this is true. I then say that it is probable that God does not exist. The theist says I’m not justified in drawing this conclusion unless I give some reason to think that no unknown goods justify God in permitting all these terrible evils. For, he says, all I’ve done is exclude one sort of good (goods known by us) as God’s justification for permitting any of these terrible evils. Who is right here?
Let’s go back to the claim that it is probable that Smith is not in town this evening. How can we be justified in making that claim if we’ve learned only that he is not at the concert? The reason is this. We originally knew that it was equally likely that he would be out of town as in town. We also agreed that if he is in town, it is equally likely that he will be at the concert as that he won’t be. Once we learn that he is not at the concert, the likelihood that he is out of town must increase, as does the likelihood that he is somewhere else in town. But since it was equally likely that he is out of town as in town, if the likelihood that he is out of town goes up, it then becomes greater than 0.5, with the result that it is probable that he is not in town.
Turn now to the existence of God and the occurrence of horrendous evils. Either God exists or he does not. Suppose for a moment that, like the case of Smith being or not being in town, each of these (God exists, God does not exist) is equally likely on the information we have prior to considering the problem of evil. Consider again the many horrendous evils that we know to occur in our world. Before we examine these evils and consider what sort of goods (known or unknown) might justify God (if he exists) in permitting them, let us suppose that it is as likely that the justifying goods for some of these evils are known to us as that the justifying goods for all of these evils are unknown to us. We then examine the known goods and those horrendous evils and come to the conclusion that no known good justifies God in permitting any of those horrendous evils. That discovery parallels our discovery that Smith is not at the concert. And the result is just the same: it is then more probable than not that God does not exist.
3.2 Second response
The second response which the theist can give to the challenge of the problem of evil is the following:
“It is a mistake to think that the goods for which God permits these horrendous evils are totally beyond our ken. For religious thinkers have developed very plausible theodicies that suggest a variety of goods that may well constitute God’s reasons for permitting many of the horrendous evils that affect human and animal existence. When we seriously consider these theodicies, we can see that we have good reason to think that premise 1 is false. For these theodicies provide us with plausible accounts of what may be God’s justifying reasons for permitting the evils that occur in our world.”
The theist’s first response was to argue both that we have given no reason at all for thinking that premise 1 is true and that our ignorance of many goods that God’s mind can comprehend prevents us from being able to establish that premise 1 is probably true. In the second response, the theist proposes to give a good reason for thinking that premise 1 is false. And, of course, to the extent that theodicies do provide a good reason for rejecting premise 1, to that extent the theist will have pointed the way to reconciling the existence of God with the fact that our world contains the horrendous evils that it does. But do these theodicies really succeed in providing a good reason for rejecting premise 1? I believe they do not. But to demonstrate this, we would have to show that these theodicies, taken together, are really unsuccessful in providing what could be God’s reasons for permitting the horrendous evils in the world. Although I believe this can be done, I propose here to take just one of these theodicies, the one most commonly appealed to, and show how it fails to provide a good reason for rejecting premise 1. I refer to the free will theodicy, a theodicy that has played a central role in defense of theism in the theistic religions of the West.
Developed extensively by St. Augustine (AD 354-430), the free will theodicy proposes to explain all the evils in the world as due either directly to evil acts of human free will or to divine punishment for evil acts of human free will. The basic idea is that, rather than create humans so that they behave like automatons, acting rightly of necessity, God created beings who have the power to act well or ill, free either to pursue the good and thereby enjoy God’s eternal blessing or to pursue the bad and thereby experience God’s punishment. As things turned out, many humans used their free will to turn away from God, freely choosing to do ill rather than good, rejecting God’s purpose for their lives. Thus, the evils in the world that are not bad acts of human free will, or their causal effects, are due to God’s own acts of punishment for wrongful exercise of human free will.
The cornerstone of this theodicy is that human free will is a good of such enormous value that God is justified in creating humans with free will even if, as Augustine held, God knew in advance of creating them that certain human beings would use their freedom to do ill rather than good, while knowing that others would use their freedom to do only (or mostly) what is good. So, all the horrendous evils occurring daily in our world are either evil acts of free human beings and their causal effects or divine punishments for those acts. And the implication of this theodicy is that the good of human free will justifies God in permitting all these horrendous acts of evil and their causal effects, as well as the other evils resulting from plagues, floods, hurricanes, etc., that are God’s ways of punishing us for our evil acts.
While this theodicy may explain some of the evil in our world, it cannot account for the massive amount of human suffering that is not due to human acts of free will. Natural disasters (floods, earthquakes, hurricanes, etc.) bring about enormous amounts of human and animal suffering. But it is obvious that such suffering is not proportionate to the abuses of free will by humans. So, we cannot reasonably think that such disasters are God’s way of punishing human free choices to do evil. Second, while being free to do evil may be essential to genuine freedom, no responsible person thinks that the good of human freedom is so great as to require that no steps be taken to prevent some of the more flagrant abuses of free choice that result in massive, undeserved suffering by human and animals. Any moral person who had power to do so would have intervened to prevent the evil free choices that resulted in the torture and death of six million Jews in the Holocaust. We commonly act to restrict egregious abuses of human freedom that result in massive, undeserved human and animal suffering. Any moral being, including God, if he exists, would likely do the same. And since the free will theodicy is representative of the other attempts to justify God’s permission of the horrendous evils in our world, it is reasonably clear that these evils cannot be explained away by appeal to theodicies.
In this essay I have argued that, putting aside whatever reasons there may be to think that the theistic God exists, the facts about evil in our world provide good reason to think that God does not exist. While the argument is only one of probability, it provides a sound basis for an affirmative answer to the question that is the focus of this exchange.
William L. Rowe, in Contemporary Debates in the Philosophy of Religion, ed. Michael L. Peterson and Raymond J. VanArragon (Malden, MA: Blackwell Publishing, 2004).