Bruce Reichenbach claims that God could not have put his creatures into a universe without at least the level of natural evils in our universe.⁷⁴ His basic idea is that the universe had to be physical and had to be governed by natural laws, and the operation of natural laws is bound to lead to nasty problems for physical creatures. Against the atheist claim that we can imagine a world with fewer evils, Reichenbach retorts that we may think we can do this, but we can’t imagine such a world in all its precise detail, therefore we can’t know that removing one evil won’t indirectly lead to equally bad or worse evils.

God wanted to make a world in which moral agents—persons with free will—could live. The world, says Reichenbach, is either entirely miraculous, or entirely governed by natural law, or some mixture of the two. If it’s entirely miraculous, there can be no moral agents, because there can’t be cause and effect. If the world is a mixture of the two, this would alter “rational predictability” on which we rely, and therefore is also out of the question. So God had no option but to make a wholly physical world subject to invariable natural laws. Reichenbach goes on to point out that there are hidden consequences to any changes one might make in such a world, and it is practically impossible for us to trace out all the ramifications of any change. If, for example, there is nothing which kills people, then overpopulation may result. Since we don’t know all the repercussions of any change, we can’t be sure that any change would lead to a better outcome than what we have. Only God could know that. (Notice that Reichenbach is bound to oppose the common theist view that God has on some occasions miraculously intervened.)

To keep it brief, let’s suppose that God had to create a universe with exactly the physical laws of our universe. Why couldn’t God have intervened occasionally to delete some of the worst evils afflicting humankind? Why couldn’t he, thousands of years ago, have got rid of the smallpox bacillus or changed the structure of the Earth’s crust so that there would be no more earthquakes or tsunamis? This falls under Reichenbach’s category of a mixture of natural law and miraculous intervention, but it would not “make rational prediction and rational action impossible.” Rational prediction and rational action would be exactly as possible as they are today.

I suppose Reichenbach might respond that, thousands of years in the future, human science might conceivably develop to the point where scientists could know that the Earth’s crust ‘should have’ been such as to generate earthquakes and tsunamis, or could determine how many and what type of malignant micro-organisms ‘should have’ been thrown up by evolution. This might then provoke a kind of ideological crisis of confidence in natural laws. But nothing disastrous would come of this. People would be able to handle it. They might conjecture that God had been helping us out, and they would be right.

We can’t help noticing that what Reichenbach declares to be impossible for God to do is precisely the kind of thing that the vast majority of theists assume that God has actually done. So the chief difference between our universe and the hypothetical one is that in the hypothetical universe, when theists tell us that God is looking out for us, they would be telling the truth.

Reichenbach evidently accepts the current scientific consensus on cosmology and evolution. He therefore accepts that the world we now have is essentially the outcome of the Big Bang, followed by billions of years of events governed by physical laws, without any divine intervention (p. 113). He seems to assume that God had to rely on the blind undirected processes of physical nature to bring about the origin of life and consciousness. Here, as so often with modern theistic arguments, we keep on tripping over the tacit abandonment of God’s omnipotence.

Objection #2: Evil Outcomes from Non-Evil Decisions

Advocates of the Free Will Defense often fall into the habit of talking as if human choices leading to evil outcomes are morally wrong choices, and if all humans behaved morally perfectly, these evil outcomes would not happen. Because people have free will, they can make good or evil choices: they choose to make evil choices, and evil results.

Yes, there are many cases where people commit evil actions, with evil consequences. If this were not frequently the case, we would have to reconsider our classification of actions as good or evil. But is it always true that evil results of human actions are results of morally wrong human actions?⁷⁵ I will mention two kinds of instances where behavior that is not morally wrong leads to evil outcomes.

Many human evils are the outcome of actions arising from mistakes due to acceptance of false theories because of limited knowledge. People cannot rightly be held morally culpable for acting for the best according to their limited knowledge. For example, a recurring major evil in human cultures is the persecution, often the killing, of minorities perceived as ‘different’. While no doubt some people take part in these pogroms and witch-hunts from evil motives, other people may participate from non-evil motives combined with a false theory of the world.

The culture which produced the moral imperative to kill all those believed to be witches had to be a culture steeped in ignorance. But given that ignorance, for which individuals in that culture cannot personally be blamed, a person might honestly be alarmed at the threat posed by witches, and conclude that harsh measures are necessary to protect the community from witches. If illnesses and accidents are largely due to the malevolent magical practices of a few individuals, if the individuals responsible can be identified, and if there is no way to dissuade them from their malign behavior, then there would be a good case for catching and killing them. Some people might therefore have supported the killing of witches from morally impeccable motives combined with false beliefs.

I suppose the theist might retort that limited knowledge is ‘an evil’, and that therefore actions guided by mistaken theories are ‘evil’, and so this is no exception to the rule that evil outcomes can be traced back to evil acts. But this would be to confuse two distinct issues. If mistaken beliefs can be characterized as ‘evil’, this does not gainsay the elementary principle that actions motivated by wholly good intentions in conjunction with erroneous beliefs are morally blameless. What is often taken for granted by proponents of the Free Will Defense is precisely that human acts leading to evil outcomes are morally wrong acts, that if everyone behaved perfectly morally, evil outcomes would evaporate. And this is not true, because people can act morally, though mistakenly, with appalling results. Although there is something bad about limited knowledge, virtuously motivated action guided by a false but sincerely held theory, which is the best available to a person in a given situation (including that person’s limited intellect), cannot be morally wrong.

My second type of evil outcomes from morally unobjectionable actions is situations where persons are so circumstanced that any of the alternative courses of action they select will lead to evil outcomes. In such cases, though an evil outcome can be traced back to human actions, it would be the case that there existed no alternative actions which would have not given rise to an evil outcome.

To illustrate this possibility, I will first mention a simple imaginary example which I don’t claim to be typical. Suppose that a community of one hundred individuals is placed in a situation where no more than half of them can survive, but if the attempt is made for all to survive, they will all starve to death. Then, whatever decision is made, its making will involve free will and will lead to a very bad outcome. Looking at the particular decision and the outcome, we may be able to trace the evil outcome back to that decision. And yet, it could be the case that any alternative decision would have led to equally bad or worse outcomes—that the decision taken was the single possible option that led to the least evil of all feasible outcomes.

That is an unlikely example, chosen as a simple illustration, but it seems obvious to me that the same principle applies many times over, in far more complex examples where it is practically impossible to identify all the specific options and the outcomes which would have ensued.

One of the great evils in human history is war. Wars arise because whole communities, or the leaders of these communities, have differences which they decide to settle by fighting. It is an aspect of human nature that individuals tend to identify with their own group and readily have suspicions of other groups.

Now, it may be said, if everyone were perfectly moral, no one would go to war. But is this plausible? It implies that the only moral course is to be a total pacifist, and possibly an anarchist too. This line has not been generally popular with theists, but still, the tiny minority of theists who have taken such a position (some Mennonites and Quakers, perhaps) could be right. If so, however, this emphasizes an associated point: that knowing the right course of action may be almost impossibly difficult, and someone acting for the best according to his lights may be tragically mistaken. War is an evil, but in some circumstances reluctance to go to war can lead to worse outcomes than readiness to go to war.

Someone might claim that right is always on only one side in any war. All that is necessary, then, to avoid all war, is for no one to take up arms in an unjust war. But if right is always on one side, it is asking too much of people that they are always able to know which side is in the right. The issues in many wars are complex, and it is a natural fact about humans that individuals tend to identify with their own group, to be suspicious of other groups, and to see things from the point of view of their own community’s culture. This fact derives from more fundamental facts: the necessity to conduct policy discussions in simplified, symbolic terms, which in turn arises from the need to economize by making decisions on the basis of incomplete information.

You should easily be able to construct other examples, for instance human practices arising from the genetic endowment mandating sexual appetite, sexual attachment, sexual jealousy, desire for sexual novelty, and sexual rivalry.

Objection #3: Free Will and a Guarantee of Goodness

If God’s omnipotent, couldn’t he have given people free will and at the same time guaranteed that they would in fact always make the ‘right’ choices?

Theists will generally deny that this is possible. But how can it be impossible? According to classical theism, God himself cannot possibly do anything evil. Theists also claim that God has free will. If God combines free will with a guarantee against ever committing evil, then it cannot be impossible to combine free will with a guarantee against ever committing evil.

God, if he decided to create other beings with free will, would also create them, in his own image, with a guarantee against their ever committing evil. The theist who says that God has free will (and they nearly all do) cannot claim that free will and a guarantee against committing evil are metaphysically incompatible, and will therefore find it hard to deny that God could have created humans with a guarantee against their ever committing evil.

A theist might respond to this that humans are not God, but something less. Still, the ‘something less’ is not possession of free will, so human free will doesn’t explain why humans couldn’t have been made in such a fashion that they never made a morally wrong choice. What is the quality that God possesses, making it unthinkable for him to do any evil, that could not have been conferred on humans when God created them? Whatever it may be, it is not free will.

Theists often claim that it is God’s nature that rules out the possibility of his doing anything wrong. ⁷⁶ In that case, God should have created humans with a nature that ruled out the possibility of their doing anything wrong. If the reply is that they were so created, but that then they succumbed to temptation, and their nature turned evil because of this ‘fall’, then one day God might succumb to temptation and develop an evil nature, or perhaps this happened long ago (and God defined as necessarily all-good no longer exists). But if God could not possibly succumb to temptation, then humans (or some other type of intelligent creatures—nothing is gained by quibbling about the zoological category ‘human’) could have been made so that they could not possibly succumb to temptation. I see no escape from the conclusion that if there were some quality of humans which made it impossible for them not to be guaranteed against making evil choices, that quality had to be something other than free will. Therefore, whatever reason God might have had for making humans in such a way that they would be likely to cause a lot of evil, it was certainly not that he wanted to let them have free will.

Leaving aside the counter-example of God himself, it has seemed obvious to many thinkers, both theists and atheists, that there is no problem about God giving his creatures free will and ensuring that they make only right choices. This position is suggested by compatibilism (which I explained above). It’s also suggested by the view that God knows the future, foreseeing all outcomes from the beginning. The view that God foreknows everything that will happen may go along with determinism or with indeterminism.

The idea that God might give people free will and ensure that they always made right choices does not involve God giving anyone an irresistible psychological compulsion, as Van Inwagen strangely seems to suppose. Everyone’s decisions between good and evil could be finely balanced, spontaneous, even whimsical, and absolutely lacking in any compulsiveness. But God who, in making the universe, sees every single event in its entire history, in all its detail, as one vast (though to God’s eye not especially vast) tapestry, taken in at one glance, would just will into existence that universe—that one there, so to speak (if we imagine God consulting an infinitely long list of every possible universe)—one precisely specified universe in which it happens to turn out that no one chooses evil.

If determinism is correct, then there is no Free Will Defense. This holds if there is no such thing as free will, and it also holds if there is free will and determinism. If both human free will and metaphysical determinism are true, then God could both give people free will and set things up so that they always make ‘good’ choices. Therefore, anyone advancing the Free Will Defense has to reject compatibilism—the view that both free will and determinism are true. A proponent of the Free Will Defense has to accept free will and reject determinism.

But even if there is no determinism, if God foresees everything then there is still no Free Will Defense. The view that God is outside time (or outside ‘our’ time), and sees the whole spacetime world as a simultaneous block with time as akin to one spatial dimension, would entail that God, in making the world, intends every detail of it, including those that are not deterministically related to other details.⁷⁷

Objection #4: Different Characters

Free will implies 1. that people have genuine choices, and 2. that the results of their choices are not the inevitable outcomes of earlier circumstances. I’m going to assume here that these two conditions are all that it takes to give us free will.⁷⁸

Free will does not, then, mean that anyone is likely to do just anything. A little thought shows that free will could not mean this, if free will really exists, because it’s an observable fact that people vary in their dispositions. The existence of free will does not imply that someone who hates the taste of fish is just as likely as anyone else to order sashimi.

For simplicity, I’m going to call everything in a person’s make-up that influences the probability of how that person will behave that person’s ‘character’. We know that people’s characters are very largely influenced by their genes: whether someone is shy or outgoing, aggressive or submissive, stable or excitable, good or bad at math or music, is largely a matter of genetics. There are also character-forming influences from upbringing, from the broader culture, and from accidents of a person’s life history.

If people have free will, then people of widely varying inherited dispositions have free will. Some people become sadistic killers, and whether someone becomes a sadistic killer involves that person’s free will. But not all individuals are equally likely to become sadistic killers. Many people simply never find anything appealing in such a course. Genetic variability ensures that some individuals are more prone to become sadistic killers than others. For instance, men are more likely to become sadistic killers than women, and men with some specific genes are more likely to become sadistic killers than other men.

Whatever determines people’s characters influences the kinds of actions that people carry out, and this fact is fully compatible with free will. It follows that in deciding to create humans with specific characters, God has a big influence on the range of likely human actions.

We should not be distracted by considerations alien to omnipotence and omniscience. For example, someone might point out that the genes of an individual especially prone to become a sadistic killer are a recombination of genes that, in other permutations, would be highly beneficial to the population. This is true. But God did not have to use natural selection, he did not have to use genes, he did not even have to give his free-willed creatures physical bodies. And assuming he had decided on genes, he could still have intervened piecemeal to modify the outcome. And even if he had intervened on billions of occasions, this would have been just as easy for him as not intervening—that follows from his omnipotence.

Objection #5: Different Circumstances

The probability that someone of a given character, with free will, will choose one option rather than another is influenced by the circumstances in which that person finds herself.

Take a person A, with wants of a certain kind, these wants being influenced by her character. Whether A can get something she wants by harming person B is a matter of A’s circumstances. For example, among some animals, rape never occurs, because it’s impossible for a male to overpower a female, or because the male sexual appetite is only switched on by a signal of female receptivity.

Combining Objections 5 and 6, we can say that a person with free will is still probabilistically influenced by his character and circumstances. Given human history, it appears that the characters and circumstances of humans, for which God (if there be a God) bears responsibility, have been conducive to acts with evil consequences.

Here we must be aware of a possible equivocation. We all know fine people, people of admirable virtue. It is not outlandish, given God’s omnipotence, to imagine that all humans could have been raised to that level. But according to one strand of Christian thinking, derived from Paul, all humans with the sole exception of Christ are totally depraved and fully deserve to roast in endless Hellfire. If theists claim that universal moral excellence by ordinary standards would still mean universal total depravity and thus horrendously evil outcomes, then, first, this does not look like a persuasive factual claim, and second, it would be wrong of God to make impossibly severe demands on beings he has created. On such a hypothesis, God would be morally contemptible and therefore not the God of classical theism.

Objection #6: Persuasive Intervention by God

There are various ways in which person A can interact with person B, and influence B’s behavior without taking away B’s free will. Most notably, A can supply information, A can persuade, and A can advise. It does not matter here whether A is God or someone else.

Suppose I shout to someone ‘That’s a bad idea’, they reconsider what they were about to do, change their minds, and don’t do it. Surely that person’s free will is intact, yet I have prevented some amount of evil (on the supposition that my judgment of the intended action was correct). The God of classical theism is a person who could talk to people, either literally, or telepathically, or by more indirect means. In talking to them, God could make his identity and his advice quite unambiguous, just as one human can, when offering advice to another. This would not cancel people’s free will. There might be people who would hear the advice and still go against it, but surely very few. At any rate, it’s a palpable fact that God does not generally talk to people in this way. Although I have been speaking of ‘advice’, often the mere provision of information would be sufficient—and then God wouldn’t even have to disclose his identity.

Here the theist might say that God does advise us, for example through the Ten Commandments. Yet, first, we must not make the mistake of supposing that the evil which results from people’s actions results only when they break the moral rules of some religious tradition or other, and never results when they comply with those moral rules. The ancient Hebrews did not see any contradiction between the Ten Commandments and stoning witches to death or slaughtering Canaanite babies. Was this a mistake on their part? If so, it was a mistake shared by the authors of the Torah, who were indispensable in transmitting the Ten Commandments. And, second, God could advise people more directly and more convincingly.

On the first point, great plagues which have killed millions of people and caused millions to suffer horrible agonies, would not have been curtailed by everyone obeying the Ten Commandments, or the entire Torah, but could have been curtailed if people had been given reliable information about transmission of infection, along with appropriate public health measures. (They could have also been prevented if God had not created infectious viruses and bacteria in the first place, or not permitted fallen angels to create these micro-organisms, but that was covered under Objection 1. At this point we’re confining our discussion to evil resulting from human behavior.)

If for some reason having God talk to people is unacceptable, God could have planted a prestigious book in a human culture, containing helpful practical advice. Instead of Leviticus, crammed with pointless idiotic restrictions on what to eat and how to behave, there could have been a book filled with insights into bacteriology and other factual matters, along with rules of thumb for healthy living.

Consider the great reductions in human suffering that have been made possible by the invention of anesthetics. Is there any reason why God could not have imparted some of the technology of anesthesia to people thousands of years before they were able to develop this knowledge for themselves, without any help from the Almighty? This would not have hurt free will, and would have been easy for God to accomplish (because of omnipotence, it would have been exactly as easy for him to accomplish it as to refrain from accomplishing it).

It would not have been any abrogation of people’s free will to inform them of the facts, and stop, or reduce, the killing of witches. The same applies to the killing of the Jews down the centuries in Europe. If God had imparted to the Christians of Europe a few key facts, for example that the gospel accounts of the railroading of Christ by Jewish leaders are purely legendary, and that the activities of the Jews as merchants and moneylenders did not deduct from, but added to, the real incomes of Christians,⁷⁹ then the recurring massacres of Jews, like the recurring massacres of supposed witches, would not have occurred, or would have occurred on a lesser scale.

Objection #7: Coercive Intervention by God

Proponents of the Free Will Defense say that almighty God wanted to give humankind free will, and therefore ‘had to’ accept that humans would use their free will to cause horrendous evils. I have argued that there were many things the hypothetical God could have done, non-coercively and person-to-person, to reduce the amount of evil, without taking away anyone’s free will.

But surely, you may think, I must accept that if God wanted to let humans keep their free will, he had to refrain from coercively intervening to prevent evil, either by directly modifying people’s thinking, or by sabotaging their evil plans. I accept nothing of the kind.

If a bank security guard says to a robber, engaged in robbing a bank: ‘Drop that gun or I will shoot you’ (or a more pungent expression carrying this essential message), is the security guard taking away the bank robber’s free will? (And if so, one might be tempted to jocularly add, is the guard expanding the free wills of the other people in the bank at the time, as well as the free wills of all the bank’s depositors and stockholders?) If you think that security guards may stop bank robberies without abolishing human free will, why would you suppose that God cannot conduct analogous coercive operations?

If we’re to think about the whole notion of what it might mean to deny people free will or to take away their free will, we need to look at three different aspects of free will: the exercise of free will, the scope for free will, and the capacity for free will. When people speak of ‘taking away someone’s free will’, it is often unclear which of these is meant. And the Free Will Defense trades on this unclarity.

The exercise of free will is executing a chosen course of action. If on one occasion someone forcibly prevents me doing what I want to do, then I am prevented from exercising my free will in a specific way. My capacity for free will remains intact.

A person’s capacity for free will is only of value because of exercises of free will, but preventing a particular exercise of free will does not detract from a person’s capacity for free will. That is, a person prevented from doing a particular thing still ‘has free will’, just as much as if they were not prevented. Scope for free will refers to the general conditions which make it possible to contemplate various courses of action. If I am locked up in a cell, my scope for free will is much less than if I am at large, but my capacity for free will is no less. If I die before the age of ten, as most children did throughout all of human history up to the industrial revolution, then my scope for free will is restricted in a different way.

It is a fact of life that specific exercises of our free will are continually being blocked off by circumstances, including the actions of other people. If God made the world, then God has arranged things this way. If it’s a question of God ‘taking away our free will’ in the sense that circumstances for which the omnipotent God must be responsible prevent us from doing certain things we would like to do, then God is taking away our free will every second of every day.

Yet this normal human condition, in which we find our scope for free will narrowly restricted and our conceivable exercises of free will blocked off at almost every turn, is spoken of by the proponent of the Free Will Defense as one in which we retain all our free will. It seems, then, that by saying that God leaves us with our free will, the proponent of the Free Will Defense must mean that we retain our capacity for free will. But as we have seen, nobody is suggesting that God should take away our capacity for free will, rather it is suggested that God continue to do what he is doing—preventing us doing what we want to do on innumerable occasions every day—but do this a bit differently, so that we are prevented from doing certain things which lead to a huge amount of appalling evil (and perhaps stop being prevented from doing certain things which would lead to good).

Many of the occasions where God might act to prevent an exercise of free will by one human would incidentally facilitate exercises of free will by other humans. The whole notion that God cannot intervene coercively because this would take away people’s free will is therefore nothing more than a dreadful muddle.

Objection #8: Collectivizing Humans

Although proponents of the Free Will Defense speak of God’s desire to leave us with our free will, it’s clear from an examination of what they say that they are actually thinking of something rather different.

The real principle to which they appeal is that if some free-willed person created by God harms another free-willed person created by God, then God washes his hands of the matter. God is not going to assume the responsibility of protecting any free-willed creature (or any animal) from the harm done by another free-willed creature. Victims have to suffer at the hands of aggressors.

This is not compatible with God’s benevolence towards humans and his omnipotence and omniscience.

Are Atheists Sissies?

Suffering is evil, or alternatively (depending on precisely how we define ‘suffering’ and ‘evil’) excessive suffering is evil. But theists sometimes insinuate that atheists are sissies. They are too sensitive to suffering and should be tougher. Aren’t there other kinds of evil? (See Swinburne 1996, p. 102.)

I agree that suffering is not the only evil. I go further and assert that departures from happiness are not the only evil. There are departures from happiness that do not amount to evil, and there are evils other than departures from happiness (Steele 2005b). Given the way things actually are, it’s even worth purchasing some good things by an increase in the amount of suffering. All this applies to the world as it actually is, not to the immensely better world you or I would have created if we had been omnipotent and omniscient.

Yet suffering or excessive suffering is an evil, and it is one that is clear-cut. One evil other than suffering is unfulfilled potential, the kind of tragic outcome described in Jude the Obscure. I could recast everything I have said here about suffering in terms of unfulfilled potential. But the identification of unfulfilled potential is generally more subtle and more controversial than the identification of suffering. It’s therefore easier to talk about suffering as standing for evil in general. To raise the question of evils other than suffering cannot save the Free Will Defense, since these other evils are just as rife in the world, and could equally well have been avoided or greatly reduced by God, if there were a God.

Theists sometimes say that the existence of extreme pain is justified by its role in preserving and protecting life. No doubt this is true from the standpoint of natural selection alone, but omnipotence and omniscience must answer to a higher standard. It could not be beyond the wit of an omnipotent, omniscient creator to devise a way to have organisms seek their survival and health without suffering. For instance, it could be that when certain kinds of injury or threat of injury appear, the normal brain functions would be over-ridden by a sort of irresistible compulsion, akin to post-hypnotic suggestion, only stronger. This over-ride would make the organism behave in some way most appropriate to its wellbeing. Of course, this would work only roughly and probabilistically—but just the same is true of pain.

As against this, some recent work by philosophers suggests that emotion may be a precondition of conscious thought. This might imply that some suffering must accompany any intensely effective awareness of injury. I think an omnipotent God would be able to get around that, but rather than pursue a lengthy examination of this issue, I will merely point out that it would be possible to have an over-ride of the kind mentioned in at least some cases. In fact, if any design were involved, it would be a simple matter to retain pain in some circumstances but eliminate pain in the many occasions (the majority, perhaps) where pain is perfectly useless.

One response to this might be that God could have done it, but could not have arranged for it to evolve by natural selection. That may be true, but there is no imperative for God to confine his creative role to setting up the natural laws which would cause life to evolve by an undirected process. Any outcome reachable by natural selection over millions of years could have been attained in an instant, by the divine equivalent of a snap of the fingers. Furthermore, God could have arranged for evolution by natural selection and also miraculously intervened piecemeal from time to time—precisely what many theists (when they are speaking in other contexts) insist that he has done.

The Biology of Evil

If we accept classical theism, we have to go to extraordinary lengths to explain why there is so much evil. And the explanations are inconclusive, as they can be paralleled by similar arguments defending the existence of a perfectly evil God. A perfectly evil God would want to keep humans alive so that they could endure as much suffering as possible, he would want to keep their hopes up so that he could cruelly dash them, he would want them to have free will, knowing that they would commit atrocities, and he would want some paragons of virtue, so that they could be tormented by the wicked (Russell 1945, p. 590).

If we suppose that there is neither a good God nor an evil God, but just no God (or a God indifferent to human welfare) then there’s no difficulty about explaining the existence of evil and the amount of evil that exists. The Problem of Evil is only a problem for the theist. The evil we observe is pretty much what we might expect on the hypothesis that we have come about by an undesigned process of evolution and if the universe were indifferent to our fate. Instead of stipulating a benevolent and omnipotent God, and then trying to come up with reasons why the omnipotent God is in practice impotent, we can account for the existence of bad things, including very bad things, by pointing to natural facts.

Consider the example of sickle-cell anemia in humans. Humans are susceptible to malaria. For every gene that you and I carry, we carry two versions, known as alleles. We get one of each from each of our parents. If one of those alleles is a sickle-cell allele, it puts some sickle cells into our blood, giving us some very slight health problems but also a lot of immunity to malaria. But if both of the alleles that an individual inherits are sickle-cell alleles, then that person has sickle-cell anemia. Sickle-cell anemia causes bouts of pain and other suffering, and a shorter life. If there is no malaria, sickle-cell anemia gradually disappears (over thousands of years) as it is now gradually disappearing among African Americans. But if malaria is rife, sickle-cell anemia spreads in the population until it reaches a high level (but not one hundred percent, as the person with the best chance of survival is the person who carries only one sickle-cell allele).

If some benevolent and very clever designer were designing people, and wanted to give them some protection against malaria, because he had their welfare at heart, he would not come up with a crazy scheme like this, which purchases some protection against the horrors of malaria by horribly tormenting one child in every four born to parents who both carry the ‘protective’ allele. But this is just the kind of thing that is liable to emerge from the blind, undesigned process of natural selection. This is one extreme version of a tragedy that’s very common: a gene spreads in the population because it gives a reproductive advantage to people who have it as one allele, while it causes serious health problems for the smaller number of people who inherit it as both alleles. Other problems are caused by the genetic mechanisms known as ‘linkage’ and ‘pleiotropy’. Genetic diseases may be favored by natural selection because the genes responsible for them also have beneficial results.

The Free Will Defense Contradicts Religious Orthodoxy

I’ve been looking at the Free Will Defense purely as it relates to the existence of God, without considering other beliefs commonly held by theists. But if we introduce some of the other beliefs to which most theists are committed, we find additional difficulties.

The Bible has many stories of God intervening in human affairs, and if we believe even a tenth of them, then we must conclude that God does not refrain from interfering in human affairs because ‘he wants to leave humans with their free will’. Nothing could be further from his thoughts!

We read that God drowned the Pharoah’s army by miraculously parting the waters of the Red Sea (Exodus 14:27-28). However, before this occurred, God had repeatedly intervened to harden the Pharoah’s heart so that he would refuse to allow the Hebrews to leave Egypt (Exodus 10:1-2, 20, 27; 11:9-10). In other words, Exodus reports that the Pharaoh, left to his own free choice, would have let the Hebrews go, but God miraculously intervened to bend the Pharaoh’s mind, against his prior will, to refuse to let the Hebrews go. God’s motive (Exodus makes clear) was to show off, by drowning the Pharaoh’s armies. The motive of the scribes who concocted this tale (or the fireside raconteurs who gave it to the scribes) was presumably to emphasize God’s untrammeled and fearsome despotism. In Job, we are told that God, because of a bet by Satan (with whom God is on fairly cordial terms) inflicts numerous sufferings upon Job, including the deaths of his wife and children. And don’t forget that a little bit earlier, God had deliberately slaughtered the entire population of the world except for eight people.

It follows that even if the Free Will Defense worked (and we have seen that it does not), it is not available to believers in the inerrancy of the Bible, or of the approximate reliability of the Bible, for several stories in the Bible are quite incompatible with God strictly abstaining from interference in human affairs.

Another difficulty stemming from theological convictions arises from belief in Heaven. Christians and Muslims, and most Jews, believe in an eternal afterlife of perfect happiness for at least some sizeable number of humans (usually that group of humans whose opinions conform most closely to those of whichever theologian is speaking).

Now, if some humans are to enjoy eternal bliss, then it cannot be the case that God ‘has to’ allow the possibility of enormous evils to fall upon humans (Mackie 1982, p. 162). The usual Christian or Muslim view of Heaven is that human souls in Heaven will continue to possess free will, but will never encounter any evil, for ‘all eternity’—for gazillions of years to come. So how could it be that, for a few million years, God ‘had to’ allow all sorts of evils to befall humans and other mammals?

Swinburne apparently notices this problem, and suggests that souls in Heaven will be limited in some ways, lacking the fullness of opportunity available to humans today.⁸⁰ Swinburne’s proposal presumably means that other persons such as angels, who also once had the kind of free will that God felt he had better not constrain, permitting the rebellion led by Lucifer, will also eventually have permanently limited capacities if they are not to be destroyed.

But this does not answer the objection I have raised here (nor does Swinburne explicitly offer it as an answer to this objection). Why is appalling suffering something God ‘has to’ fail to prevent for a few million years if he is then going to prevent it for all conscious beings, including humans, for uncountable eons upon eons? How can it be the case both that terrible suffering is the price God had to pay for generating a population of souls endowed with free will and that a population of souls endowed with free will be guaranteed to flourish for endless eons without any suffering?

A theist might say that souls in heaven will not possess free will, though I doubt that many theists would adopt that position. However, that would not circumvent the problem, which would re-emerge in this form: if souls without free will are good enough for all eternity after a certain date, why weren’t they good enough for a finite period before that date? A parallel question would arise if it were claimed that souls in heaven would have ‘some degree of’ free will, but curtailed in some way.

I have here supposed that ‘eternity’ means ‘for ever and ever’, for an infinite future period of time. If instead we suppose that Heaven is outside time, a similar problem arises. If human souls, or those that are saved from Hell, are transported out of time to the timeless Heaven, why wouldn’t it have been better to send them to Heaven as soon as they were created, and cut out all the history of the physical cosmos, with its terrible evils? A natural answer would be that souls have to go through a testing stage before being fit for Heaven, but that is incompatible with omnipotence and omniscience. An omnipotent, omniscient being could never run a test—the concept is senseless. We run tests because our knowledge is limited. The God of classical theism could have created them just as though they had been tested, and again cut out the whole messy business of space, time, and matter-energy.

Giving Up Classical Theism

Since the Free Will Defense to the Argument from Evil is a failure, the God of classical theism does not exist. Theists who understand this are compelled to modify their concept of God. The two most likely ways to do this are either to accept that God is limited in his powers, or to accept that God is not fully a person.

In 1981 Rabbi Harold Kushner, influenced by Process Theology, had a multi-million best-seller with When Bad Things Happen to Good People. Essential to Kushner’s explanation for evil afflicting the innocent was the limited power of God and the fact that some things just happen for no reason. To the book-buying public this did not come as a shock, but to some theologians it was a scandal. Rabbi Yitzchok Kirzner responded to Kushner, with a book restating the more traditional position. Kushner’s book was criticized on similar grounds by Protestants and Catholics.

The apparently orthodox Catholic Peter Geach denied that God feels any obligation to do all he can for the benefit of humans, and Brian Davies has taken this a bit further, maintaining that God is not a morally responsible agent and explicitly questioning whether God is a person at all. Contrary to Peter Parker’s Uncle Ben, Davies holds that with great power comes no responsibility. If God is not a person, or if God does not truly love humans, then the Argument from Evil loses its force. It remains to be seen whether theists who adopt this position are prepared to follow through and strip their pronouncements of all references to God as a loving parent.