Reflections on the problem of evil (Part 3) – Logical version

Posted: July 18, 2011 in Philosophy, Religion
Tags: , ,

by Dima Zhyvov

This post is a continuation of the series on the problem of evil. Please see Part 1, Part 2, if you have not read them yet.

Logical version of the problem of evil

According to the logical problem of evil the following two statements are logically incompatible:

(1)   An omnipotent, wholly good God exists.

(2)   Evil exists.

Although the logical version of the problem of evil may be presented in different ways, essentially they all attempt to show that something like (1) and (2) are logically incompatible. If successful, the argument demonstrates that there is a logical inconsistency in what Christians believe. Now notice that, at face value, there is nothing explicitly contradictory in the joint affirmation of God and evil. There is no apparent contradiction between them. If the atheist thinks that they are implicitly contradictory, then he must be assuming some hidden premises that would serve to bring out the contradiction and make it explicit. But what are these hidden assumptions? There seem to be two:

(3)   If God is omnipotent, then he can create any world that he desires.

(4)   If God is wholly good, then he prefers a world without evil over a world with evil.

Now the contradiction is made explicit. Adding (3) and (4) to (1) and (2) creates a contradictory set of propositions, meaning it is impossible for all statements in (1) – (4) to be true. If evil exists, then either God does not exist, or He is not omnipotent, or He is not perfectly good.

The atheist reasons that since God is omnipotent, he could create a world containing free creatures who always freely choose to do the right thing. Such a world would be a sinless world, free of all human, moral evils. Similarly, being omnipotent, God might as well create a world in which no natural evils would exist. It would be a world devoid of all pain and misery. Now notice what the atheist is not saying. He is not saying that people would be mere puppets in such a world. Rather he is saying that there exists a possible world in which all people would always freely choose to do what is right. As a Christian theist, I must admit to the possibility of such a world, or else I would be implicitly committing myself to an unwelcome idea that sin is somehow necessary and cannot be avoided. This logically leads to a denial of people’s moral responsibility, since such responsibility must imply a genuine possibility of choice, in our case the possibility of doing what is right, whenever a moral decision is made. If this were not so, then in what sense, according to Christian faith, can a sinner be culpable before God and accountable for his actions? So, certainly, we can conceive of a world in which everyone freely chooses every single time to do the right thing. But the argument does not stop there, the atheist then reasons that since God is all-powerful, he must be able to actualize such state of affairs.

Furthermore, since God is also wholly good, the objector continues, he would, of course, prefer a pain-and-suffering-free world to any world infected with evil. The idea here is that if God were presented with a choice of creating an evil-free world and a world with evil in it like in the actual world we live in, then, surely, he would choose the former over the latter. Otherwise, God would himself be evil to prefer a world where his creatures suffer and experience pain when he could have given them happiness and bliss. David Hume, a famous Scottish philosopher, summarised the logical version of the internal problem of evil well when he asked concerning God, “Is he willing to prevent evil, but not able? Then he is impotent. Is he able, but not willing? Then he is malevolent. Is he both able and willing? Whence then is evil?”[1]

Now what are we to make of this atheological argument? Certainly, on the surface, the argument appears to be irrefutable. But is it? What response, if any, can a theist make?

I think it would be helpful at this point, following Alvin Plantinga, to distinguish between a “defence” and a “theodicy”. As Plantinga employs these terms, a theodicy aims to give an account of why God actually permits the evils in the world. It attempts to answer Hume’s last question “Whence then is evil?” directly. By contrast a defence provides no such account but seeks merely to undercut the atheist’s claim or to show that atheists have failed to demonstrate that evil is incompatible with God’s existence. Thus, a successful defence would defeat the atheological argument from evil, while still leaving us in the dark as to why God permits evil and suffering in the world. In other words, a successful defence will have undermined the logic of Hume in asking the first two questions and the conclusions that he draws.

Let us examine premises (3) and (4) more closely to see whether they are necessarily true or not. Remember, the proponent of the logical version of the problem of evil has assumed an enormous burden of proof. For her argument to be successful, she must show that it is impossible for (3) and (4) to be false. In other words, she is committed to a claim that there is no possible world in which (3) and (4) are false; they must be true in every possible world. Has this been demonstrated by atheists? Hardly. Consider (3):

(3) If God is omnipotent, then he can create any world that he desires.

Is (3) necessarily true? Plantinga argues in his free will defence (successfully I think) that (3) is not necessarily true. He reasons that if it is even possible that creatures have libertarian freedom[2], then the assumption made by the objector is not necessarily true, which it must be if the atheist is to show that there is no possibility of the coexistence of God and evil. If libertarian free will is possible, then it is not necessarily true that God’s omnipotence implies that he can create just any possible world that he desires. To understand Plantinga here, we must first grasp just what it is we are talking about when we ascribe omnipotence to God. What does divine omnipotence mean? Does God’s being omnipotent imply that He can do logical impossibilities, such as making a round square, or a married bachelor, or making someone freely do something? Very few thinkers, aside from Descartes and perhaps Martin Luther, have been willing to affirm that the doctrine means that God can do just anything. Such view affirms the incredible proposition that there are no necessary truths. For on this view an omnipotent deity could have brought it about that even logical contradictions be true and tautologies be false, as crazy as this may seem to us. I think this position is incoherent, for it affirms what it denies: is the statement, “There are no necessary truths” itself necessarily true or not? If it is true, then this view is self-refuting. If it is false, then it is possible for there to be necessary truths, the very thing this doctrine seems to deny. Let me warn you that concentrating too much on these things is likely to send your head spinning. As I said very few philosophers or theologians have understood omnipotence in this way. Then how has omnipotence been traditionally understood?  This question deserves a separate article, which would include detailed discussion of possibility, necessity, feasibility, and so on, but briefly, divine omnipotence has been generally understood in terms of God’s ability to actualize certain states of affairs, rather than in terms of raw power[3]. Thus it is not due to the lack of power on God’s part that he cannot make a stone too heavy for him to lift, for the proposition “a stone too heavy for God to lift” describes a logically impossible state of affairs, and thus it describes nothing at all.

So let us go back to our original discussion of premise (3) of the logical version of the argument from evil. Let us raise the question again: is (3) necessarily true? Well, given traditional understanding of divine omnipotence, it is not. How so? Because it implies God’s ability to do at least one logical impossibility, namely, to make someone freely do something. For if God causes a person to make a specific choice, then the choice is no longer free in the libertarian sense. It follows, then, that if God gives people genuine freedom to choose as they like, then it is impossible for him to guarantee what their choices will be. All he can do is to create the circumstances in which free agents are able to exercise their libertarian freedom, and then, so to speak, stand back and let them choose. Now this implies that there are states of affairs which are possible in and of themselves, but which God is incapable of creating. For it is possible that in every world with free creatures in it, some of those creatures freely choose to do evil, in which case it is the creatures themselves that bring about evil, and God can do nothing to prevent them from doing so, apart from refusing to actualize any such worlds. Thus it is possible that every world with free creatures in it which God is able to actualize is a world with sin and evil. Now an atheist might object that it is unlikely that every world which God is able to actualize is a world with sin and evil, but then the objector would confuse the logical problem of evil with the probabilistic one. All one needs to do to refute the logical version of the problem is to show that such an explanation is possible, because if it is even possible that every world feasible for God which contains free creatures in it is a world with sin and evil, then it follows that (3) is not necessarily true, which it must be for the logical argument to be successful.

What about natural evil though? So far Plantinga’s free will defence has undercut the “necessary nature” of premise (3) with respect to moral evil, but what about other evils such as natural disasters? Plantinga points out that these could be the result of demonic activity in the world. Demons can have freedom just like human beings, and it is possible that God could not preclude natural evil without removing the free will of demonic creatures. Now one might think that such solution to the problem of natural evil is utterly ridiculous, but that again would be to confuse the logical version of the problem with the probabilistic one. I concede that attributing all natural evils to demonic beings is highly improbable, but that is strictly irrelevant. All one needs to show at this point is that such an explanation is possible, and that, as a consequence, the objector’s argument that God and evil are logically incompatible fails. So the first assumption made by the objector, namely, that an omnipotent God can create any world that he desires, is just not necessarily true. Therefore, on this ground alone, the argument at hand is invalid.

But I think we can go even further. Let us examine premise (4) and see whether this proposition is necessarily true. Step (4) of the argument reads:

(4)  If God is wholly good, then he prefers a world without evil over a world with evil.

Again, such an assumption is not necessarily true. We can think of various cases when we allow pain and suffering to occur in person’s life in order to bring about some greater good or because we have morally sufficient reason for allowing it. Think of human parents for a moment. Are there times when discipline must be inflicted on the child in order to teach him or her to become a mature, responsible adult? Similarly, God may permit suffering in our lives in order to build us or test us, or to build and test others, or to achieve some other overriding end. At this point, I am not claiming that soul-making or character building is in fact the reason why God permits pain and suffering in the world; that would be a project of theodicy, not of defence. Rather I am only suggesting that it is possible for God to have morally sufficient reasons for permitting pain and suffering. If the atheist thinks it is impossible for God to have such reasons, he needs to show us why. Until and unless he does that, I am within my epistemic rights to think that premise (4), “a wholly good God prefers a world without evil over a world with evil”, is also not necessarily true. The argument is thus doubly invalid.

A crucial assumption on the part of those who propound the logical version of the problem of evil is the notion that God cannot have morally sufficient reasons for permitting the amount and the kinds of evil that exists. But it is not at all clear that this assumption is necessarily true. Consider first the amount of evil in the world. As terrible a place as the world is, there is still on balance a great deal more good in it than evil. In spite of life’s difficulties, people generally agree that life is worth living. Now it is possible, given creaturely freedom, that in any world of free creatures feasible for God, the balance between good and evil would have been no better than in the actual world. That is to say, any world containing less evil might also have had less good. Maybe the actual world, Plantinga speculates, has in it the most good God could get for the least amount of evil. Maybe something similar holds for the kinds of evil in the world as well. It is possible that God has overriding reasons for permitting the world’s most horrific kinds of evil. Similarly, it is possible that if the world had fewer such horrendous evils, then it would also have been lacking in such important, overriding good.

Someone might object that these scenarios seem pretty unlikely. But then the person would be conflating once again the logical problem of evil with the probabilistic problem of evil. To refute the logical version of the internal problem of evil, the theist does not have to present a plausible or likely solution – all he has to do is to suggest a possible one. The point is that if atheists want to show that it is logically impossible for God and evil to exist, then they must prove that God cannot have morally sufficient reasons for permitting the amount and kinds of evils in the world. And this is an enormously heavy burden, which no atheist so far has been able to carry.

But a careful observer might note that there is an important difference between the claim that God and evil are compatible with each other and a claim that no one has shown the two to be incompatible. Suppose Plantinga has successfully undercut all claims of inconsistency with respect to coexistence of God and evil. Does it follow, then, that God and evil are in fact compatible? Not necessarily. The fact that no logical version of the problem of evil to date has been able to prove the incompatibility of God and evil does not entail that the two are compatible. To show that they are compatible, a theist needs to introduce a new premise to (1) and (2), which in conjunction with them would prove that God and evil are consistent. In order to do that, a theist must come up with some possible explanation of the evil in the world that is compatible with God’s existence. Craig suggests the following to be a candidate for such an explanation:

(5)  God could not have created a world that had so much good as the actual world but had less evil, both in terms of quantity and quality; and, moreover, God has morally sufficient reasons for permitting the evil that exists[4].

The “could not” in the above statement should be understood to mean that such a world is not feasible for God to create. That is to say, although such a world is possible in and of itself, God is powerless to actualize it, due to the libertarian freedom of his creatures and God’s inability to bring about logically impossible states of affairs such as making someone freely do something (see the discussion of God’s omnipotence above). So long as this explanation is even possible, it proves that God and the evil in the world are logically compatible.

So in conclusion, I think that the logical version of the problem of evil is unsuccessful, because two of its main assumptions are not necessarily true. No one, after centuries of philosophical discussions, has been able to demonstrate the logical incompatibility of God and evil. The atheist assumes too heavy a burden of proof for him to carry. Moreover, the theist is able to propose a possible explanation of the evil in the world that is compatible with God’s existence, which proves that God and evil are logically consistent.

As a result, it is now almost unanimously agreed that there is nothing like a straightforward contradiction or necessary falsehood in the joint affirmation of God and evil; the existence of evil is not logically incompatible with the existence of an all-powerful, all-knowing, and perfectly good God.

Fair enough, the logical version of the internal problem of evil is bankrupt. The short and neat argument from evil, which was supposed to settle the question of God’s existence once and for all, has failed, but does it follow then that theism is off the hook? Fine, an atheist may admit, there is no logical contradiction in the coexistence of God and evil, but there also isn’t anything logically contradictory in beliefs such as the earth is flat, that a prime minister is a prime number, that my mother-in-law will live to be two hundred years old, etc. The objection here is that there are all kinds of crazy beliefs that aren’t logically contradictory, yet which we, nevertheless, would be irrational to accept. This objection takes us to the probabilistic version of the internal problem of evil. So to this version of the problem we must turn next.

Click here to see part 4 of the article

[1] Hume, Dialogues Concerning Natural Religion, part 10, as cited by Moreland and Craig, Foundations, 536

[2] Theopedia, the online encyclopedia of Biblical Christianity defines libertarian free will thus:

Libertarian free will means that our choices are free from the determination or constraints of human nature and free from any predetermination by God. All “free will theists” hold that libertarian freedom is essential for moral responsibility, for if our choice is determined or caused by anything, including our own desires, they reason, it cannot properly be called a free choice. Libertarian freedom is, therefore, the freedom to act contrary to one’s nature, predisposition and greatest desires. Responsibility, in this view, always means that one could have done otherwise.” See

[3] See The Coherence of Theism, in Philosophical Foundations for a Christian Worldview, Craig & Moreland, p. 528

[4] See The Problem of evil, in Philosophical Foundations for a Christian Worldview, Craig & Moreland, p. 541


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s