The Problem of Evil

© 2011 By Paul Herrick

Issues in Philosophy of Religion

4. The Problem of Evil

Arguments Against the Existence of God

The world contains enormous amounts of suffering. Yet traditional theism claims that this world is the creation of a wholly good, all-powerful, all-knowing, perfectly loving God. Why would a good, loving, all-powerful being create a world containing so much suffering? Aren’t goodness and love opposed to suffering? How could an all-good God even allow one iota of suffering? This question, or puzzle, has traditionally been called the “Problem of Evil” since suffering is an evil, i.e., a bad thing.

Reflection on the problem of evil has given rise to a philosophical argument known as the “argument from evil.” The argument comes in two forms: Deductive versions aim to prove conclusively, beyond a shadow of any reasonable doubt—that God does not exist. Inductive versions aim to establish only that it is very unlikely that God exists, so improbable as to make it unreasonable to believe that God exists.

The Argument From Evil: A Deductive Version

In 1955, Australian philosopher J. L. Mackie (1917-1981) produced a famous version of the deductive argument from evil. First, Mackie argued, the following two propositions are inconsistent:

(A) An omnipotent (“all-powerful”), omniscient (“all-knowing”), and omnibenevolent (“all-good”) God exists.

(B) The world contains suffering or evil.

Why suppose these propositions are inconsistent? Well, an omnibenevolent being would be absolutely opposed to all suffering. An omniscient being would know about any and all suffering. An omnipotent being would have the power to prevent any and all suffering. Therefore, if an omnipotent, omniscient, omnibenevolent God were to exist, it would not allow any suffering whatsoever. Therefore, if God exists, then suffering does not exist, and if suffering exists, then God does not exist. 

Since traditional theism accepts both (A) and (B) as true, it follows that traditional theism is logically inconsistent. Furthermore, argued Mackie, something additional follows. It is clearly beyond dispute that (B) is true: the world obviously contains suffering, lots of suffering. But if (B) is true, then (A) is false, since (A) and (B) are inconsistent. It thus follows that (A) is false: God, as conceived within traditional theism, does not exist.

Simplified Form of the Argument From Evil

Here is a simplified form of the argument from evil:

  1. If an all-good, all-knowing, all-loving, all powerful God exists, then evil (suffering) does not exist.
  2. Evil exists.
  3. Therefore, an all-good, all-knowing, all-loving, all powerful God does not exist.

An argument against the existence of God is called an “atheological argument.” The argument from evil is thus an atheological argument—a reason to suppose there is no such being as God.

A Reply to Mackie

Why suppose that an omnipotent, omnibenevolent, omniscient being (hereafter: a triple o being) would not allow the existence of any evil or suffering? Mackie bases this claim on the following two principles, which we shall name Principles X and Y:

Principle X: A good thing always eliminates evil as far as it can.

Principle Y: There are no limits to what an omnipotent being can do.

Given these two principles, it indeed validly follows that if a triple o being (God) exists, then no evil at all exists. However, Mackie's argument is valid only if Principles X and Y are necessary truths. If they are not—if either is logically contingent and thus might be false—then an omnipotent, wholly good being might permit evil to exist, in which case (A) and (B) are not logically inconsistent. Is Principle X necessarily true?

If a wholly good being happened to have a morally justifying reason to permit an evil that it could prevent, then it might not eliminate all evil as far as it can. Is this possible? In everyday life, we sometimes permit suffering to occur when we have the power to stop it: we do so when we have a morally justifying reason to pemit it. For example, a parent takes a baby into the doctor for a needed vaccination even though the baby cries and is upset. Sometimes parents allow their children to make mistakes and then suffer the consequences of their mistakes, when the parents could easily pay the price of those mistakes (and thus shield their children from consequences). Parents do this when they have a morally justifying reason. One such reason is the value to the children of learning certain lessons in life, including lessons about the consequences of unwise choices. How else does one really learn?   

Whether we know the reason or not, it seems reasonable to suppose that it is at least possible a morally justifying reason exists for God to allow at least some suffering. This at least seems to be a possibility. Principle X, therefore, does not seem necessarily true.

In place of X, a more plausible principle for Mackie would be:

Principle X*: A wholly good being eliminates evil as far as it can, unless it has a morally justifying reason to permit it.

Now we can see that evil will be nonexistent only if it is necessarily true that there is no morally justifying reason for a triple o being to prevent evil. But nothing Mackie suggests rules out the possibility that there is a morally justifying reason for a triple o being to permit an evil that it could prevent. For it at least seems possible that:

J: There is such a morally justifying reason, but it is a reason that is beyond our ken, a reason beyond our understanding, and God permits some evil for that morally sufficient reason.

If nothing we know contradicts J, and if God's existence and the truth of J are logically compatible with the existence of evil, then Mackie’s argument fails.

Back to Top

Reasons to Permit Evil: Theodicies

This brings us to the topic of theodicy. A theodicy is “an account that aims to explain why God would allow suffering.” Theodicies attempt to formulate morally sufficient reasons God might have for allowing some degree of suffering or evil. By articulating a reason why God would allow suffering, a theodicy aims to explain how it is possible that God and evil both exist. A theodicy, if plausible, would give us a reason to reject the claim, presented in the argument from evil, that God’s existence is logically incompatible with the existence of evil. 

Many theodicies have been presented and defended by philosophers down through the centuries, but two  stand out. The “free will theodicy” was first given systematic expression in the writings of St. Augustine (A.D. 354-430). The “soul-making theodicy,” also called the “higher goods theodicy,” first appears in the writings of St. Irenaeus (c.130-c.202). According to the free will theodicy, God allows evil because (i) evil is an unavoidable consequence of the free will given to creatures, and (ii) the existence of this freedom is of such overriding value that preventing the evil would require preventing the freedom, which in turn would mean that the universe loses the value made possible by freedom, which loss would greatly diminish the total amount of value in existence.

According to the soul-making theodicy, God allows evil because (a) suffering is an unavoidable consequence of the many challenges God built into the world, challenges that make the world like an obstacle course, but (b) these challenges exist in order to give human beings an opportinity to develop moral qualities in response to the obstacles, and the development of these moral qualities (such as courage, compassion, justice, perserverence, and love) is of such overriding value that preventing the evil would require preventing the opportunities for moral development and the value it brings into being, but (c) the loss of this value would greatly diminish the total amount of moral value in existence.

Replies to Theodicy

In reply to the free will theodicy, J. L. Mackie has asked: If God is omnipotent, then why can’t God simply make free creatures who always and only do good? In response to the soul making theodicy, some philosophers have asked: If God is omnipotent, then why can’t God simply build the moral qualities into people, without making the world such an obstacle course, without making people suffer through all the hardships of the present world? Of course, theists have proposed answers to these questions; the back and forth goes on. And what a fascinating and illuminating conversation it is.

Back to Top

The Argument From Evil: Inductive Version

The philosopher William Rowe has given an influential version of the inductive argument that may be summarized as follows. Suppose we grant the theist that freedom is necessary for real love, and suppose we grant that suffering, obstacles, and hardships are necessary prerequisites for moral development. This problem remains: Why does the world contain so much suffering and hardship? Why so many obstacles? Why would a God allow this huge quantity ofsuffering? Think of the Holocaust. Think of the millions of deaths from cancer. It’s the quantity that is the problem. Perhaps some instances of suffering and the misuse of freedom are necessary evils, justified ultimately by the overriding good they lead to in the end. But the vast quantity of suffering our world contains seems to go far beyond necessity, far beyond what is needed to attain some overriding higher good in the end. Couldn't an omnipotent God have brought about as much value with far less suffering in the process? In short, isn't much of the world's evil unnecessary, superfluous, gratuitous? In more formal terms, Rowe’s argument can be put this way:

  1. There seem to exist instances of horrible suffering which an omnipotent, omniscient, wholly good being could have prevented without thereby losing some greater good and without thereby permitting some horrible evil equally bad or worse.
  2. An omnipotent, omniscient, wholly good being would prevent the occurrence of any suffering unless it could not do so without thereby losing some greater good or permitting some evil equally bad or worse.
  3. So, there does not exist an omnipotent, omniscient, wholly good being.i

But why should the theist accept premise 1? Rowe gives the following reasoning:

Suppose in some distant forest lightning strikes a dead tree, resulting in a forest fire. In the fire, a fawn is trapped, horribly burned, and lies in terrible agony for several days before death relieves its suffering. So far as we can see, the fawn’s intense suffering is pointless, for there does not appear to be any greater good such that the prevention of the fawn’s suffering would require either the loss of that good or the occurrence of an evil equally bad or worse. Nor does there seem to be an equally bad or worse evil so connected to the fawn’s suffering that it would have had to occur had the fawn’s suffering been prevented.ii

However, Rowe admits that this doesn’t prove premise (1) with certainty, for:

We are often surprised by how things we thought to be unconnected turn out to be intimately connected. Perhaps, for all we know, there is some familiar good outweighing the fawn’s suffering to which that suffering is connected in a way we do not see. Furthermore, there may be unfamiliar goods, goods we haven’t dreamed of, to which the fawn’s suffering is inextricably connected. Indeed, it would seem to require something like omniscience on our part before we should lay claim to knowing that there is no greater good [justifying the fawn’s suffering].iii

Nevertheless, Rowe argues, even if an investigation showed that the fawn’s suffering is not pointless, it seems quite unlikely that all instances of intense suffering occurring daily in our world are intimately related to the occurrence of a greater good or to the prevention of evils at least as bad. So, Rowe concludes, it is reasonable to believe premise (1) of his argument.

Thus, since no good that we know of justifies God in permitting horrific evils, Rowe concludes that it is very probable that no good justifies God permitting the quantity of suffering that we see around us. From this Rowe concludes it is very probable God, defined as a triple o being, does not exist.

Wykstra's Reply

The philosopher Stephen Wykstra has an interesting reply to Rowe. First, Wykstra argues, Rowe seems to be reasoning from:

  1. We cannot see any reason God might have to permit some of the horrible evils around us; in other words, we are not aware of any such reason.


  2. Therefore, there is very probably no reason for a god to permit these evils.

However, Wykstra continues, in general we may legitimately argue from “we see no x” to “there is no x” only when x has “reasonable seeability;” that is, when x is such that if it exists, we could reasonably expect to see it or detect it in some way. For example, from “we see no elephant in the living room,” we can argue to “there is no elephant in here.” This is because elephants have reasonable seeability. However, we cannot argue from “we see no flea” to "there is no flea in this room,” for fleas have “low seeability.”

Next, if God has morally sufficient reasons to allow evil, they would have low seeability, Wykstra argues, for if these reasons do exist, we should not expect to be aware of them. The reason we should not expect to be aware of them, if they exist, is that our cognitve capacities and our moral imaginations are so limited compared to what God's would be. Just as we would not expect a little child to understand all the reasons his parents might have for moving the family to another state, we should not expect humans to be aware of all the reasons God might have for permitting certain evils. But this gives us reason to doubt the first premise of Rowe's argument. So argues Wykstra.


i See William L. Rowe, "Ruminations About Evil," in James E. Tomberlin, ed., Philosophical Perspectives, 5. Philosophy of Religion (Atascadero California: Ridgeview Publishing Company, 1991) ), pp. 69-88.

ii Ibid., p. 70.

iii Ibid., p. 71.

Back to Top