argumentative essay on problem of evil

1000-Word Philosophy: An Introductory Anthology

Philosophy, One Thousand Words at a Time

The Problem of Evil

Author: Thomas Metcalf Category: Philosophy of Religion Word Count: 1000

Many people believe in God and understand God to be an omnipotent (all-powerful), omniscient (all-knowing), and morally perfect being. [1]

But the world contains quite a lot of evil or badness: intense suffering, premature death, and moral wickedness.

This inspires some questions: Why would God permit such evil? Is there a good reason why? Or does it occur in part because there is no God to prevent it?

Asking these questions involves engaging with the Problem of Evil . [2]

The concern is whether evil provides a reason to disbelieve in God. There are four things one might say about evil, ranging from that it proves that God does not exist to that it provides no evidence at all against God’s existence.

1. The Incompatibility Problem of Evil

The ‘Incompatibility’ or ‘Logical’ versions of the Problem of Evil claim that evil’s existence is logically incompatible with God’s existence: believing in God and evil is like believing in a five-sided square, a contradiction. [3]

Most philosophers today reject this argument. [4] They think that God could have some sufficient reason to permit some evil: e.g., personal growth requires confronting challenges that inherently involve some evil or bad things. These defenses [5] seem to show that it is not contradictory to believe in God and the existence of evil.

2. The Evidential Problem of Evil

Other philosophers argue that the mere existence of evil does not prove that God does not exist, but that the facts about evil provide good evidence against God’s existence. [6]

There are probably billions of evils such that we do not know why God, if there is a God, would permit them. Many argue that if even one of these instances is gratuitous —i.e., God could have prevented it without thereby sacrificing an equal or greater good and without thereby permitting an equal or worse evil—then God does not exist. [7]

Theists have reason to find an explanation or set of explanations that could plausibly justify all evils. This involves trying to find plausible theodicies or explanations of why God would permit that evil or why that evil is not as evidentially weighty as it might seem. Here’s a summary of two of the best theodicies.

2.1. Free Will

Many theists hold that humans’ having significant free will is a very great good, one that is worth the evil that sometimes arises from it. [8]

This being a plausible explanation of evil depends on justifying these claims:

(a) we have libertarian free will [9] (a belief that is mostly rejected by philosophers [10] );

(b) (e.g.) Stalin’s free will was more valuable than the lives of the millions he killed (against, presumably, their freely-willed choices to remain alive);

(d) even apparently natural disasters and disease, including those that harm nonhuman animals [12] , are all the result (e.g.) of free-willed evil-spirits’ choices. [13]

2.2. “Soul-Making”

Perhaps encountering evil and freely responding to it develops various virtues, such as compassion, generosity, and courage. [14]

For this to explain evil, the theist may need to argue that:

(a) God could not have developed those virtues in us any other equally valuable but less harmful ways (e.g,. by creating humans who are more morally sensitive in the first place and reducing evil accordingly);

(b) all evil can reasonably be expected to contribute to soul-making; and

(c) the compassion Smith develops when she sees Jones suffering justifies God using Jones (or allowing Jones to be used) as a means to the end of producing that compassion. [15]

Given these and other theodicies, we must ask how much evidence evil provides, and weigh that against any evidence for God’s existence. This will obviously be very complicated.

3. Outweighing Evidence?

Theists might argue that there is so much evidence for God’s existence that we are justified in being confident that God has a purpose for all evil. [16]

We cannot consider those arguments here, but we should recall how many billions of instances of severe, inscrutable evils there are in the world. Therefore, for this defense to work, perhaps there must be very strong evidence for God’s existence. Also, a substantial majority of philosophers reject theism, [17] and so seem to believe that there is little good evidence for God’s existence. Therefore, this strategy may depend on appealing to a set of generally-rejected arguments to try to explain evil.

4. Evil Is No Evidence?

Some defenses amount to the response that evil is no evidence against God’s existence at all.

Some argue that we should not expect to understand why God would permit evil, and so we should not be confident in our ability to assess whether some evil is gratuitous. [18] If there is a God, God might have a purpose for all the evil in the world, a purpose that we do not or cannot understand, and so we should not trust our doubt that some evil in the world is justified. [19]

Typically, this inspires the question of whether a similar argument can be made about other beliefs we have, thereby threatening to produce a deep, general skepticism about science, morality, and even arguments for God’s existence. [20] If God works in mysterious ways, how do I assess the likelihood that God has some inscrutable reason for tricking me into (wrongly) thinking that other minds exist, that the past exists, that an external world exists, and that I ought to save a child drowning in a shallow pond? This is perhaps the primary focus of the debate about the Problem of Evil in recent years.

Finally, some philosophers argue that God’s existence is actually compatible with gratuitous evil after all, [21] although most philosophers disagree. [22]

5. Conclusion

If each particular evil is even a little bit of evidence against God’s existence, the billions and billions of them in history might really pile up. For many people, the problem of evil is not merely an abstract puzzle, for it challenges their most profound beliefs about what God is like and whether God even exists.

[1] Anselm 1965 [1077-78]: ch. 2.

[2] The Problem of Evil involves engaging arguments from the existence of evil, or types of evils, to the conclusion that God does not exist. So the Problem of Evil is also called The Argument from Evil.

[3] Mackie 1955. “Evidential” versions of the argument, discussed in the next section, typically focus on the totality of evil and can be seen as “Incompatibility” arguments also: the claim is that God’s existence entails that there are no gratuitous or pointless evils—evils God could have prevented it without thereby sacrificing an equal or greater good and without thereby permitting an equal or worse evil—but that there are such gratuitous or pointless evils, which is a logical contradiction.

[4] Rowe 1979: 335.

[5] A “defense” is an attempt to explain why God and evil are not incompatible. Defenses are closely related to theodicies (two of which are presented below) which attempt to explain why God permits evil. Defenses and theodicies are different: defenses hold that there is some possible explanation, even if we’re not sure what it is, while theodicies attempt to supply that actual explanation.

[6] Rowe 1979; Draper 1989; Tooley 2014: § 3.2.1.

[7] Howard-Snyder and Howard-Snyder 1999.

[8] Plantinga 1977: 29-59.

[9] For an explanation of what libertarian free will is, see Jonah Nagashima’s Free Will and Free Choice . Libertarians about free will (a view of which has no relation to the political position of the same name) believe that free choices are choices that are not causally determined by the past and the laws of nature (or anything else), and so they believe that determinism is false, yet that such choices are not ultimately random because we are the ultimate source of our choices.

The other broad definition of free will is that of compatibilist free will. On this theory of free will, we can be determined to do what we do, yet our actions can still be done from free will if, e.g., we are doing what we want to do and acting on our own desires. This view of free will seems to allow that God could cause us to not act in horribly evil ways, and that we freely choose to never engage in these evils, and so the free will defense is not available to compatibilists.

[10] Bourget and Chalmers 2014.

[11] So, e.g., Stalin might freely make the choice to kill someone, but whether the effect of that choice—that is, whether someone is actually killed—seems to be another matter. So, a question is whether, if there is a God, God could allow us to freely make decisions (which is assumed to be a great good), but prevent the very bad effects that result from some of them, and God be justified preventing those very bad effects.

[12] Rowe 1979: 337.

[13] Plantinga 1977: 58.

[14] Hick 2007: 253-61.

[15] cf. Kant 1987 [1785]: 4:429; Trakakis 2008.

[16] cf. Rowe 1979: 338.

[17] Bourget and Chalmers 2014.

[18] Howard-Snyder and Howard-Snyder 1999: 115.

[19] Wykstra 1998.

[20] Draper 1998: 188; Russell 1998: 196-98. The general response to the Problem of Evil that we are not likely to know whether any evil is gratuitous or pointless is known as “Skeptical Theism,” since skeptics deny that we have a type of knowledge. A concern about skeptical theism is whether the motivations for it lead to or justify other types of skepticism.

[21] van Inwagen 2000; Kraay 2010. van Inwagen’s argument is complex and depends on the (controversial) claim that it can be permissible to allow some unjustified evils, e.g., that it could be permissible to allow someone to remain imprisoned for at least slightly longer than any just imprisonment because sometimes arbitrary lines must be drawn. From there, he appeals to something like a “little by little” argument (based on concerns about vagueness: see Darren Hibb’s Vagueness ). that if a little unjustified evil can be permissibly allowed, then a tiny bit more can be permissibly allowed, so then a little more can be allowed, leading to the conclusion that any unjustified evils can be allowed.

[22] Howard-Snyder and Howard-Snyder 1999; Trakakis 2003.

Anselm. (1965 [1077-78]). St. Anselm’s Proslogion . Tr. M. J. Charlesworth. Notre Dame: University of Notre Dame Press.

Bourget, David and David J. Chalmers. (2014). “What Do Philosophers Believe?” Philosophical Studies , forthcoming.

Draper, Paul. (1989). “Pain and Pleasure: An Evidential Problem for Theists.” Noûs 23: 331-50.

———. (1998). “The Skeptical Theist.” In Daniel Howard-Snyder (ed.), The Evidential Argument from Evil . Bloomington and Indianapolis: Indiana University Press, 175-92.

Hick, John. (2007). Evil and the God of Love . New York: Palgrave Macmillan.

Howard-Snyder, Daniel and Frances Howard-Snyder. (1999). “Is Theism Compatible with Gratuitous Evil?” American Philosophical Quarterly 36 (2): 115-30.

Kant, Immanuel. Groundwork of the Metaphysics of Morals . In Kant, Immanuel. Practical Philosophy . Ed. Mary J. Gregor. Cambridge, U.K.: Cambridge UP, 1999.

Kraay, Klaas J. “Theism, Possible Worlds, and the Multiverse.” Philosophical Studies 147 (2010), pp. 255-68.

Mackie, J. L. (1955). “Evil and Omnipotence.” Mind 64 (254): 200-12.

Plantinga, Alvin. (1977). God, Freedom, and Evil . Grand Rapids, MI: Wm. B. Eerdmans.

Rowe, William. (1979). “The Problem of Evil and Some Varieties of Atheism.” American Philosophical Quarterly 16(4): 335-41.

Russell, Bruce. (1998). “Defenseless.” In Daniel Howard-Snyder (ed.), The Evidential Argument from Evil . Bloomington and Indianapolis: Indiana University Press: 193-205.

Tooley, Michael. (2014). “The Problem of Evil.” In Edward N. Zalta (ed .), The Stanford Encyclopedia of Philosophy (Spring 2014 edition), URL = <http://plato.stanford.edu/archives/spr2014/entries/evil/>.

Trakakis, Nick. (2008). “Theodicy: The Solution to the Problem of Evil, or Part of the Problem?” Sophia 47: 161-91.

———. (2003). “God, Gratuitous Evil, and van Inwagen’s Attempt to Reconcile the Two.” Ars Disputandi 3 (1): 1-10.

van Inwagen, Peter. (2000). “The Argument from Particular Horrendous Evils.” Proceedings of the American Catholic Philosophical Association 74: 65–80.

Wykstra, Stephen John. (1998). “Rowe’s Noseeum Argument from Evil.” In Daniel Howard-Snyder (ed.), The Evidential Argument from Evil . Bloomington and Indianapolis: Indiana University Press: 126-50.

Related Essays

Attributes of God by Bailie Peterson

The Problem of No Best World by Kirk Lougheed

Divine Hiddenness by David Bayless

Hell and Universalism by A.G. Holdier

Because God Says So: On Divine Command Theory by Spencer Case

Possibility and Necessity: An Introduction to Modality by Andre Leo Rusavuk

Nietzsche and the Death of God by Justin Remhof

Design Arguments for the Existence of God by Thomas Metcalf

Free Will and Free Choice by Jonah Nagashima

Bayesianism by Thomas Metcalf

Vagueness by Darren Hibb

Revision History

This essay, posted 8/16/2020, is a revised version of an essay originally posted 4/7/2014.

PDF Download

Download this essay in PDF .

About the Author

Tom Metcalf is an associate professor at Spring Hill College in Mobile, AL. He received his PhD in philosophy from the University of Colorado, Boulder. He specializes in ethics, metaethics, epistemology, and the philosophy of religion. Tom has two cats whose names are Hesperus and Phosphorus. shc.academia.edu/ThomasMetcalf

Follow 1000-Word Philosophy on Facebook , Twitter and subscribe to receive email notifications of new essays at the bottom of 1000WordPhilosophy.com

8 The Problem of Evil

Peter Van Inwagen is the John Cardinal O'Hara Professor of Philosophy at the University of Notre Dame. He has delivered the Maurice Lectures at King's College, London, the Wilde Lectures at Oxford University, the Stewart Lectures at Princeton University, and the Gifford Lectures at St Andrews University. His books include: An Essay on Free Will (OUP, 1983), Material Beings, Metaphysics (2002 [2nd ed.]), God, Knowledge, and Mystery, Ontology, Identity, and Modality, and The Problem of Evil (2006). He is at work on a book called Being: A Study in Ontology. He was elected to the American Academy of Arts and Sciences in 2005, and was president of the Central Division of the American Philosophical Association, 2008–2009.

Published: 02 September 2009
Cite Icon Cite
Permissions Icon Permissions

There are many ways to understand the phrase “the problem of evil.” This article conceives this phrase as a label for a certain purely intellectual problem—as opposed to an emotional, spiritual, pastoral, or theological problem (and as opposed to a good many other possible categories of problem as well). The fact that there is much evil in the world (that is to say, the fact that many bad things happen) can be the basis for an argument for the nonexistence of God (that is, of an omnipotent and morally perfect God). But this article takes these qualifications to be redundant: It takes the phrases “a less than omnipotent God” and “a God who sometimes does wrong” to be self-contradictory, like “a round square” or “a perfectly transparent object that casts a shadow.”) Here is a simple formulation of this argument: If God existed, he would be all-powerful and morally perfect. An all-powerful and morally perfect being would not allow evil to exist.

1. Introductory Remarks: The Problem of Evil and the Argument from Evil

There are many ways to understand the phrase “the problem of evil.” In this chapter, I understand this phrase as a label for a certain purely intellectual problem—as opposed to an emotional, spiritual, pastoral, or theological problem (and as opposed to a good many other possible categories of problem as well). The fact that there is much evil in the world (that is to say, the fact that many bad things happen) can be the basis for an argument for the nonexistence of God (that is, of an omnipotent and morally perfect God. But I take these qualifications to be redundant: I take the phrases “a less than omnipotent God” and “a God who sometimes does wrong” to be self-contradictory, like “a round square” or “a perfectly transparent object that casts a shadow.”) Here is a simple formulation of this argument:

If God existed, he would be all-powerful and morally perfect. An all-powerful and morally perfect being would not allow evil to exist. But we observe evil. Hence, God does not exist.

Let us call this argument “the argument from evil”—glossing over the fact that there are many arguments for the nonexistence of God that could be described as arguments from evil. The intellectual problem I call the problem of evil can be framed as a series of closely related questions addressed to theists: How would you respond to the argument from evil? Why hasn't it converted you to atheism (for surely you've long known about it)? Is your only response the response of faith—something like, “Evil is a mystery. We must simply trust God and believe that there is some good reason for the evils of the world”? Or can you reply to the argument? Can you explain how, in your view, the argument can be anything less than an unanswerable demonstration of the truth of atheism?

These questions present theists with a purely intellectual challenge. I believe this intellectual challenge can be met. I believe it can be met by critical examination of the argument. I believe critical examination of the argument shows that it is indeed something less than an unanswerable demonstration of the truth of atheism. I attempt just such a critical examination in this chapter. In this chapter, we shall examine this argument, hold it up to critical scrutiny.

2. The “Moral Insensitivity” Charge

Before we examine the argument from evil, however, we must consider the charge that to examine it, to treat it as if it was, as it were, just another philosophical argument whose virtues and defects could be weighed by impartial reason, is a sign of moral insensitivity—or downright wickedness. One might suppose that no argument was exempt from critical examination. But it is frequently asserted, and with considerable vehemence, that it is extremely wicked to examine the argument from evil with a critical eye. Here, for example, is a famous passage from John Stuart Mill's Three Essays on Religion :

We now pass to the moral attributes of the DeityThis question bears a very different aspect to us from what it bears to those teachers of Natural Theology who are encumbered with the necessity of admitting the omnipotence of the Creator. We have not to attempt the impossible problem of reconciling infinite benevolence and justice with infinite power in the Creator of a world such as this. The attempt to do so not only involves absolute contradiction in an intellectual point of view but exhibits to excess the revolting spectacle of a jesuitical defense of moral enormities. (1875, 183)

I cannot resist quoting, in connection with this passage from Mill, a poem that occurs in Kingsley Amis's ( 1966 ) novel The Anti-death League (it is the work of one of the characters). 1 This poem puts a little flesh on the bones of Mill's abstract Victorian prose. It contains several specific allusions to just those arguments Mill describes as jesuitical defenses of moral enormities. Its literary effect depends essentially on putting these arguments, or allusions to them, into the mouth of God.

To a Baby Born without Limbs This is just to show you who's boss around here. It'll keep you on your toes, so to speak. Make you put your best foot forward, so to speak, And give you something to turn your hand to, so to speak. You can face up to it like a man, Or snivel and blubber like a baby. That's up to you. Nothing to do with Me. If you take it in the right spirit, You can have a bloody marvelous life, With the great rewards courage brings, And the beauty of accepting your lot . And think how much good it'll do your Mum and Dad, And your Grans and Gramps and the rest of the shower, To be stopped being complacent. Make sure they baptize you, though, In case some murdering bastard Decides to put you away quick, Which would send you straight to limb-o , ha ha ha. But just a word in your ear, if you've got one. Mind you, do take this in the right spirit, And keep a civil tongue in your head about Me. Because if you don't , I've got plenty of other stuff up My sleeve, Such as leukemia and polio (Which, incidentally, you're welcome to any time, Whatever spirit you take this in). I've given you one love-pat, right? You don't want another. So watch it, Jack.

I am afraid I must accuse Mill (and the many other authors who have expressed similar sentiments) of intellectual dishonesty.

Philosophy is hard . Thinking clearly for an extended period is hard. It is easier to pour scorn on those who disagree with you than actually to address their arguments. And of all the kinds of scorn that can be poured on someone's views, moral scorn is the safest and most pleasant (most pleasant to the one doing the pouring). It is the safest kind because, if you want to pour moral scorn on someone's views, you can pretty much take it for granted that most people will regard what you have said as unanswerable; you can take it as certain that everyone who is predisposed to agree with you will believe you have made an unanswerable point. You can pretty much take it for granted that your audience will dismiss any attempt your opponent in debate makes at an answer as a “rationalization”—that great contribution of modern depth psychology to intellectual complacency and laziness. Moral scorn is the most pleasant kind of scorn to deploy against those who disagree with you because a display of self-righteousness—moral posturing—is a pleasant action whatever the circumstances, and it's nice to have an excuse for it. No one can tell me Mill wasn't enjoying himself when he wrote the words “exhibits to excess the revolting spectacle of a jesuitical defense of moral enormities.” (Perhaps he was enjoying himself so much that his attention was diverted from the question, What would it be to exhibit a revolting spectacle in moderation?)

To people who employ the argument from evil and attempt to deflect critical examination of this argument by that sort of moral posturing, I can only say, Come off it. These people are, in point of principle, in exactly the same position as those defenders of law and order who, if you express a suspicion that a man accused of abducting and molesting a child has been framed by the police, tell you with evident disgust that molesting a child is a monstrous crime and that you're defending a child molester.

3. God's Omnipotence, His Moral Perfection, and His Knowledge of Evil

Having defended the moral propriety of critically examining the argument from evil, I will now do just that. The argument presupposes, and rightly, that two features God is supposed to have are “nonnegotiable”: that he is omnipotent and morally perfect. That he is omnipotent means that he can do anything —provided his doing it doesn't involve an intrinsic impossibility. (Thus, even an omnipotent being can't draw a round square. And God, although he is omnipotent, is unable to lie, for his lying is as much an intrinsic impossibility as a round square.) To say that God is morally perfect is to say that he never does anything morally wrong—that he could not possibly do anything morally wrong. If omnipotence and moral perfection are nonnegotiable components of the idea of God, this fact has the following two logical consequences. (1) If the universe was made by an intelligent being, and if that being is less than omnipotent (and if there's no other being who is omnipotent), the atheists are right: God does not exist. (2) If the universe was made by an omnipotent being, and if that being has done even one morally wrong thing (and if there isn't another omnipotent being, one who never does anything morally wrong), the atheists are right: God does not exist. If, therefore, the Creator of the universe lacked either omnipotence or moral perfection, and if he claimed to be God, he would be either an impostor (if he claimed to be omnipotent and morally perfect) or confused (if he admitted that he was less than omnipotent or less than morally perfect and still claimed to be God).

One premise of the simple version of the argument set out above—that an all-powerful and morally perfect being would not allow evil to exist—might well be false if the all-powerful and wholly good being were ignorant, and not culpably ignorant, of the existence of evil. But this is not a difficulty for the proponent of the simple argument, for God, if he exists, is omniscient. The proponent of the simple argument could, in fact, defend his premise by an appeal to far weaker theses about the extent of God's knowledge than “God is omniscient.” If the simple argument presents an effective prima facie case for the conclusion that there is no omnipotent and morally perfect being who is omnisicent, it presents an equally effective prima facie case for the conclusion that there is no omnipotent and morally perfect being who has even as much knowledge of what goes on in the world as we human beings have. The full panoply of omniscience, so to speak, does not really enter into the initial stages of a presentation and discussion of an argument from evil. Omniscience, omniscience in the full sense of the word, will become important only when we come to examine responses to the argument from evil that involve free will (see Section 9).

How shall we organize our critical examination of the argument from evil? I propose that we imagine in some detail a debate about the existence of God, and that we try to determine how effective a debating point the reality of evil would be for the party to the debate who was trying to show that there was no God.

4. A Description of an Ideal Debate about the Existence of God

Let us imagine that we are about to watch part of a debate between an atheist (“Atheist”) and a theist (“Theist”) about whether there is a God. This debate is being carried on before an audience of agnostics. As we enter the debating hall (the debate has evidently been going on for some time), Atheist has the floor. She is trying to convince the agnostics to abandon their agnosticism and become atheists like herself. Theist is not, not in this part of the debate anyway, trying to convert the agnostics to theism. At present, he is trying to convince the agnostics of only one thing: that Atheist's arguments should not convert them to atheism. (By an odd coincidence, we have arrived just at the moment at which Atheist is beginning to set out the argument from evil.) I mean these fictional characters to be ideal types, ideal representatives of the categories “atheist,” “theist,” and “agnostic”: they are all highly intelligent, rational, and factually well informed; they are indefatigable speakers and listeners, and their attention never wanders from the point at issue. The agnostics, in particular, are moved by a passionate desire for truth. They want to get the question of the existence of God settled , and they don't at all care which way it gets settled. Their only desire is—if this should be possible—to leave the hall with a correct belief about the existence of God, a belief they have good reason to regard as correct. (They recognize, however, that this may very well not be possible, in which case they intend to remain agnostics.) Our two debaters, be it noted, are not interested in changing each other's beliefs. Each is interested in the effects his or her arguments will have on the beliefs of the agnostics and not at all in the effects those arguments will have on the beliefs of the other debater. One important consequence of this is that neither debater will bother to consider the question, Will my opponent accept this premise? Each will consider only the question, Will the agnostics accept this premise?

Can Atheist use the argument from evil to convert these ideal “theologically neutral” agnostics to atheism—in the face of Theist's best efforts to block her attempt to convince them of the truth of atheism? Our examination of the argument from evil will be presented as an attempt to answer this question.

5. Atheist's Initial Statement of the Argument from Evil; Theist Begins His Reply by Making a Point about Reasons

Atheist, as I have said, is beginning to present the argument from evil to the audience of agnostics. Here is her initial formulation of the argument:

Since God is morally perfect, he must desire that no evil exist—the nonexistence of evil must be what he wants . And an omnipotent being can achieve or bring about whatever he wants—or at least whatever he wants that is intrinsically possible, and the nonexistence of evil is obviously intrinsically possible. So if there were an omnipotent, morally perfect being who knew about these evils—well, they wouldn't have arisen in the first place, for he'd have prevented their occurrence. Or if, for some reason, he didn't do that, he'd certainly remove them the instant they began to exist. But we observe evils, and very long-lasting ones. So we must conclude that God does not exist.

What shall Theist say in reply? I think he should begin with an obvious point about the relations between what one wants, what one can do, and what one will, in the event, do:

I grant that, in some sense of the word, the nonexistence of evil must be what a perfectly good being wants . But we often don't bring about states of affairs we can bring about and want to bring about. Suppose, for example, that Alice's mother is dying in great pain and that Alice yearns desperately for her mother to die—today and not next week or next month. And suppose it would be easy for Alice to arrange this—she is perhaps a doctor or a nurse and has easy access to pharmacological resources that would enable her to achieve this end. Does it follow that she will act on this ability she has? It does not, for Alice might have reasons for not doing what she can do. (She might, for example, think it would be morally wrong to poison her mother; or she might fear being prosecuted for murder.) The conclusion that evil does not exist does not, therefore, follow logically from the premises that the nonexistence of evil is what God wants and that he is able to bring about the object of his desire—since, for all logic can tell us, God might have reasons for allowing evil to exist that, in his mind, outweigh the desirability of the nonexistence of evil.

But Theist must say a great deal more than this, for, if we gave her her head, Atheist could make a pretty good prima facie case for two conclusions: that a morally perfect creator would take pains to prevent the suffering of his creatures, and that the suffering of creatures could not be a necessary means to any end for an omnipotent being. Theist must, therefore, say something about God's reasons for allowing evil, something to make it plausible to believe there might be such reasons. Before I allow him to do this, however, I will introduce some terminology that will help us to understand the general strategy I am going to have him follow in his discussion of God's reasons for allowing evil to exist.

6. A Distinction: “Theodicy” and “Defense”

Suppose that I believe in God and that I think I know what God's reasons for allowing evil to exist are and that I tell them to you. Then I have presented you with what is called a theodicy, from the Greek words for “God” and “justice.” Thus, Milton, in Paradise Lost , tells us that the purpose of the poem is to “justify the ways of God to men”—“justify” meaning “exhibit as just.” (Here I use “theodicy” in Alvin Plantinga's sense. Other writers have used the word in other senses.) If I could present a theodicy, and if the audience to whom I presented it found it convincing, I'd have an effective reply to the argument from evil, at least as regards that particular audience. But suppose that, although I believe in God, I don't claim to know what God's reasons for allowing evil are. Is there any way for someone in my position to reply to the argument from evil? There is. Consider this analogy.

Your friend Clarissa, a single mother, left her two very young children alone in her flat for several hours very late last night. Your Aunt Harriet, a maiden lady of strong moral principles, learns of this and declares that Clarissa is unfit to raise children. You spring to your friend's defense: “Now, Aunt Harriet, don't go jumping to conclusions. There's probably a perfectly good explanation. Maybe Billy or Annie took ill, and she decided to go over to St Luke's for help. You know she hasn't got a phone or a car and no one in that neighborhood of hers would come to the door at two o'clock in the morning.” If you tell your Aunt Harriet a story like this, you don't claim to know what Clarissa's reasons for leaving her children alone really were. And you're not claiming to have said anything that shows that Clarissa really is a good mother. You're claiming only to show that the fact Aunt Harriet has adduced doesn't prove Clarissa isn't a good mother; what you're trying to establish is that for all you or Aunt Harriet know, she had some good reason for what she did. And you're not trying to establish only that there is some remote possibility that she had a good reason. No lawyer would try to raise doubts in the minds of the members of a jury by pointing out to them that for all they knew his client had an identical twin, of whom all record had been lost, and who was the person who had actually committed the crime his client was charged with. That may be a possibility—I suppose it is a possibility—but it is too remote a possibility to raise real doubts in anyone's mind. What you're trying to convince Aunt Harriet of is that there is, as we say, a very real possibility that Clarissa had a good reason for leaving her children alone, and your attempt to convince her of this consists in your presenting her with an example of what such a reason might be.

Critical responses to the argument from evil—at least responses by philoso phers—usually take just this form. A philosopher who responds to the argument from evil typically does so by telling a story, a story in which God allows evil to exist. This story will, of course, represent God as having reasons for allowing the existence of evil, reasons that, if the rest of the story were true, would be good ones. Such a story philosophers call a defense . A defense and a theodicy will not necessarily differ in content. A's defense may, indeed, be verbally identical with B's theodicy. The difference between a theodicy and a defense is simply that a theodicy is put forward as true, while nothing more is claimed for a defense than that it represents a real possibility—or a real possibility given that God exists. If I offer a story about God and evil as a defense, I hope for the following reaction from my audience: “Given that God exists, the rest of the story might well be true. I can't see any reason to rule it out.” The logical point of this should be clear. If the audience of agnostics reacts to a story about God and evil in this way, then, assuming Atheist's argument is valid, they must reach the conclusion Theist wants them to reach: that, for all they know, one of Atheist's premises is false. And if they reach that conclusion, they will, for the moment, remain agnostics.

Some people, if they are familiar with the usual conduct of debates about the argument from evil, may be puzzled by my bringing the notion “a very real possibility” into my fictional debate at this early point. It has become something of a custom for critics of the argument from evil first to discuss the so-called logical problem of evil, the problem of finding a defense that contains no internal logical contradiction; when the critics have dealt with this problem to their own satisfaction, as they always do, they go on to discuss the so-called evidential (or probabilistic) problem of evil, the problem of finding a defense that (among certain other desirable features) represents, in my phrase, a real possibility. A counsel for the defense who followed a parallel strategy in a court of law would first try to convince the jury that his client's innocence was logically consistent with the evidence by telling a story involving twins separated at birth, operatic coincidences, and mental telepathy; only after he had convinced the jury by this method that his client's innocence was logically consistent with the evidence would he go on to try to raise real doubts in the jurors' minds about his client's guilt.

I find this division of the problem artificial and unhelpful and will not allow it to dictate the form of my discussion of the argument from evil. I am, as it were, jumping right into the evidential problem (so-called; I won't use the term) without any consideration of the logical problem. Or none as such, none under the rubric “the logical problem of evil.” Those who know the history of the discussions of the argument from evil in the 1950s and 1960s will see that many of the points I make, or have my creatures Atheist and Theist make, were first made in discussions of the logical problem.

All right. Theist's response will take the form of an attempt to present one or more defenses, and his hope will be that the response of the audience of agnostics to this defense, or these defenses, will be, “Given that God exists, the rest of the story might well be true. I can't see any reason to rule it out.” What form could a plausible defense take?

One point is clear: a defense cannot simply take the form of a story about how God brings some great good out of the evils of the world, a good that outweighs those evils. At the very least, a defense will have to include the proposition that God was unable to bring about the greater good without allowing the evils we observe (or some other evils as bad or worse). And to find a story that can plausibly be said to have this feature is no trivial undertaking. The reason for this lies in God's omnipotence. A human being can often be excused for allowing, or even causing, a certain evil if that evil was a necessary means, or an unavoidable consequence thereof, to some good that outweighed it—or if it was a necessary means to the prevention of some greater evil. The eighteenth-century surgeon who operated without anesthetic caused unimaginable pain to his patients, but we do not condemn him because (at least if he knew what he was about) the pain was an unavoidable consequence of the means necessary to a good that outweighed it: saving the patient's life, for example. But we should condemn a present-day surgeon who had anesthetics available and who nevertheless operated without using them—even if his operation saved the patient's life and thus resulted in a good that outweighed the horrible pain the patient suffered.

7. Theist's Reply Continues; The Initial Statement of the Free-will Defense

There seems to me to be only one defense that has any hope of succeeding, and that is the so-called free-will defense. 2 I am going to imagine Theist putting forward a very simple form of this defense; I will go on to ask what Atheist might say in response:

God made the world and it was very good. An indispensable part of its goodness was the existence of rational beings: self-aware beings capable of abstract thought and love and having the power of free choice between contemplated alternative courses of action. This last feature of rational beings, free choice or free will, is a good. But even an omnipotent being is unable to control the exercise of free choice, for a choice that was controlled would ipso facto not be free. In other words, if I have a free choice between x and y , even God cannot ensure that I choose x . To ask God to give me a free choice between x and y and to see to it that I choose x instead of y is to ask God to bring about the intrinsically impossible; it is like asking him to create a round square or a material body with no shape. Having this power of free choice, some or all human beings misused it and produced a certain amount of evil. But free will is a sufficiently great good that its existence outweighs the evils that have resulted and will result from its abuse; and God foresaw this.

Theist's presentation of the free-will defense immediately suggests several objections. Here are two that would immediately occur to most people:

How could anyone possibly believe that the evils of this world are outweighed by the good inherent in our having free will? Perhaps free will is a good and would outweigh, in Theist's words, “a certain amount of evil,” but it seems impossible to believe that it can outweigh the amount of physical suffering (to say nothing of other sorts of evil) that actually exists. Not all evils are the result of human free will. Consider, for example, the Lisbon earthquake or the almost inconceivable misery and loss of life produced by the hurricane that ravaged Honduras in 1997. Such events are not the result of any act of human will, free or unfree.

In my view, the simple form of the free-will defense I have put into Theist's mouth is unable to deal with either of these objections. The simple form of the free-will defense can deal with at best the existence of some evil—as opposed to the vast amount of evil we actually observe—and the evil with which it can deal is only the evil that results from the acts of human beings. I believe, however, that more sophisticated forms of the free-will defense do have interesting things to say about the vast amount of evil in the world and about the suffering caused by earthquakes and hurricanes and other natural phenomena. Before I discuss these “more sophisticated” forms of the free-will defense, however, I want to examine two objections that have been brought against the free-will defense that are so fundamental that, if they were valid, they would refute any elaboration of the defense, however sophisticated. These objections have to do with free will. I am not going to include them in my dialogue between Atheist and Theist, for the simple reason that, in my view, anyway, they have not got very much force, and I do not want to be accused of fictional character assassination; my Atheist has more interesting arguments at her disposal. But I cannot ignore these arguments: the first has been historically important and the second turns on a point that is likely to occur to most readers.

8. An Objection to the Free-will Defense: God Can Control the Exercise of Free Choice

The first of the two arguments is essentially this: the free-will defense fails because free will and determinism are compatible; God could, therefore, create a world whose inhabitants are free to do evil but do only good.

This might seem a surprising argument. Why should anyone believe that free will and determinism were compatible?

Well, many very able philosophers have believed this, and for reasons unrelated to theological questions. Philosophers of the stature of Thomas Hobbes, David Hume, and John Stuart Mill have held that free will and determinism are perfectly compatible: that there could be a world in which the past determined a unique future and whose inhabitants were nonetheless free agents. Philosophers who accept this thesis are called “compatibilists.” It is not hard to see that if the compatibilists are right about the nature of free will, the free-will defense fails. If free will and determinism are compatible, an omnipotent being can , contrary to a central thesis of the free-will defense, create a person who has a free choice between x and y and ensure that that person choose x rather than y .

Those philosophers who accept the compatibility of free will and determinism defend their thesis as follows: being free is being free to do what one wants to do. Prisoners in a jail, for example, are unfree because they want to leave and can't. The man who desperately wants to stop smoking but can't is unfree for the same reason—even though the barrier that stands between him and a life without nicotine is psychological, and not a physical thing like a wall or a door. The very words “free will” testify to the rightness of this analysis, for one's will is simply what one wants, and a free will is just exactly an unimpeded will. Given this account of free will, a Creator who wants to give me a free choice between x and y has only to arrange matters in such a way that the following two “if” statements are both true: if I were to want x , I'd be able to achieve that desire, and if I were to want y , I'd be able to achieve that desire. And a Creator who wants to ensure that I choose x rather than y has only to implant in me a fairly robust desire for x and see to it that I have no desire at all for y . And these two things are obviously compatible. Suppose, for example, that there was a Creator who had placed a woman in a garden and had commanded her not to eat of the fruit of a certain tree. Could he so arrange matters that she have a free choice between eating of the fruit of that tree and not eating of it—and also ensure that she not eat of it? Certainly. To provide her with a free choice between the two alternatives, he need only see to it that two things are true: first, that if she wanted to eat of the fruit of that tree, no barrier (such as an unclimbable fence or paralysis of the limbs or a neurotic fear of trees) would stand in the way of her acting on that desire, and, second, that if she wanted not to eat of the fruit, nothing would force her to act contrary to that desire. And to ensure that she not eat of the fruit, he need only see to it that not eating of the fruit be what she desires (and that she have no other desire in conflict with this desire). An omnipotent and omniscient being could therefore bring it about that every creature with free will always freely did what was right.

Having thus shown a proposition central to the free-will defense to be false, the critic can make the consequences of its falsity explicit in a few words. If a morally perfect being could bring it about that every creature with free will always freely did what was right, there would of necessity be no creaturely abuse of free will, and evil could not possibly have entered the world through the creaturely abuse of free will. The so-called free-will defense is thus not a defense at all, for it is an impossible story.

We have before us, then, an argument for the conclusion that the story called the free-will defense is an impossible story. But how plausible is the account of free will on which the argument rests? Not very, I think. It certainly yields some odd conclusions. Consider the lower social orders in Brave New World , the “deltas” and “epsilons.” These unfortunate people have their deepest desires chosen for them by others, by the “alphas” who make up the highest social stratum. What the deltas and epsilons primarily desire is to do what the alphas tell them. This is their primary desire because it has been implanted in them by prenatal and postnatal conditioning. (If Huxley were writing today, he might have added genetic engineering to the alphas' list of resources for determining the desires of their slaves.) It would be hard to think of beings who better fitted the description “lacks free will” than the deltas and epsilons of Brave New World . And yet, if the compatibilists' account of free will is right, the deltas and epsilons are exemplars of beings with free will. Each of them is always doing exactly what he wants, after all, and who among us is in that fortunate position? What he wants is to do as he is told by those appointed over him, of course, but the compatibilists' account of free will says nothing about the content of a free agent's desires: it requires only that there be no barrier to acting on them. The compatibilists' account of free will is, therefore, if not evidently false, at least highly implausible—for it has the highly implausible consequence that the deltas and epsilons are free agents. And an opponent of the free-will defense cannot show that that story fails to represent a “real possibility” by deducing its falsity from a highly implausible theory.

9. A Second Objection to the Free-will Defense: Free Will Is Incompatible with God's Omniscience

I turn now to the second argument for the conclusion that any form of the free-will defense must fail: the free-will defense, of course, entails that human beings have free will; but the existence of a being who knows the future is incompatible with free will, and an omnisicent being knows the future, and omniscience belongs to the concept of God; hence, the so-called free-will defense is not a possible story—and is therefore not a defense at all.

Most theists, I think, would reply to this argument by trying to show that divine omniscience and human free will were compatible, for that is what most theists believe. But I find the arguments, which I will not discuss, for the incompatibility of omniscience and freedom, if not indisputably correct, at least pretty convincing, and I will therefore not reply in that way. (And I think that the attempt of Augustine and Boethius and Aquinas to solve the problem by contending that God is outside time—that he is not merely everlasting but altogether nontemporal—is a failure. I don't mean to say that I reject the proposition that God is outside time; I mean that I think his being outside time doesn't solve the problem.) I will instead reply to the argument by engaging in some permissible tinkering with the concept of omniscience. At any rate, I believe it to be permissible for reasons I shall try to make clear.

In what follows, I am going to suppose that God is everlasting but temporal, that he is not “outside time.” I make this assumption because I do not know how to write coherently and in detail about a nontemporal being's knowledge of (what is to us) the future. Now consider these two propositions:

X will freely do A at t . Y, a being whose beliefs cannot be mistaken, believes now that X will do A at t .

These two propositions are consistent with each other or they are not. If they are consistent, there is no problem of omniscience and freedom. Suppose, then, that they are inconsistent, and suppose free will is possible. (If free will isn't possible, the free-will defense is self-contradictory for that reason alone.) Then it is impossible for a being whose beliefs cannot be mistaken to have beliefs about what anyone will freely do in the future. Hence, if free will exists it is impossible for any being to be omniscient. Now, if the existence of free will implies that there cannot be an omniscient being, it might seem, by that very fact, to imply that there cannot be an omnipotent being. For if it is intrinsically impossible for any being now to know what someone will freely do tomorrow or next year, it is intrinsically impossible for any being now to find out what someone will freely do tomorrow or next year; and a being who can do anything can find out anything. But this inference is invalid, for an omnipotent being is, as it were, excused from the requirement that it be able to do the intrinsically impossible. This suggests a solution to the problem of free will and divine omniscience: why should we not qualify the concept of omniscience in a way similar to the way the concept of omnipotence is qualified? Why not say that even an omniscient being is unable to know certain things—those such that its knowing them would be an intrinsically impossible state of affairs. Or we might say this: an omnipotent being is also omnisicent if it knows everything it is able to know. If we say, first, that the omnipotent God is omniscient in the sense that he knows everything that, in his omnipotence, he is able to know, and, second, that he does not know what the future free acts of any agent will be, we do not contradict ourselves—owing to the fact that (now) finding out what the future free acts of an agent will be is an intrinsically impossible action.

I must admit that this solution to the problem of free will and divine foreknowledge raises a further problem for theists: Are not most theists committed (for example, in virtue of the stories told about God's actions in the Bible) to the proposition that God at least sometimes foreknows the free actions of creatures? This is a very important question. In my view, the answer is no, at least as regards the Bible. But a discussion of this important question is not possible within the scope of this chapter.

10. Atheist Contends That the Free-will Defense Cannot Account for the Amount and the Kinds of Evil We Observe

I conclude that neither an appeal to the supposed compatibility of free will and determinism nor an appeal to the supposed incompatibility of free will and omniscience can undermine the free-will defense.

Let us return to Atheist, who, as I said, has better arguments at her disposal than those considered in sections 8 and 9. What shall she say in response to the free-will defense? What she should do, I think, is to concede a certain limited power to the free-will defense and to go on to maintain that this power is essen tially limited. Her best course is to concede that the free-will defense shows there might be, for all anyone can say, a certain amount of evil, a certain amount of pain and suffering, in a world created by an all-powerful and morally perfect being, and to conduct her argument in terms of the amounts and the kinds of evil that we actually observe. Her best course is to argue for the conclusion that neither the simple version of the free-will defense I have had Theist present nor any elaboration of it can constitute a plausible account of the evil, the bad things, that actually exist. I have mentioned two points about the evil we observe in the world that would probably occur to most people immediately upon hearing Theist's initial statement of the free-will defense: that the amount of suffering (and other evils) is enormous and must outweigh whatever goodness is inherent in the reality of free will; that some evils are not caused by human beings and cannot therefore be ascribed to the creaturely abuse of free will. I will now ascribe to Atheist a rather lengthy speech that takes up these two points—and a third, perhaps less obvious.

I will concede that the free-will defense shows that the mere existence of some evil or other cannot be used to prove the nonexistence of God. If we lived in a world in which everyone, or most people, suffered in certain relatively minor ways, and if each instance of suffering could be traced to the wrong or foolish acts of human beings, you would be making a good point when you tell these estimable agnostics that, for all they know, these wrong or foolish acts are free acts, that even an omnipotent being cannot determine the outcome of a free choice, and that the existence of free choice is a good thing, sufficiently good to outweigh the bad consequences of its occasional abuse. But the evil we actually observe in the world is not at all like that. First, the sheer amount of evil in the world is overwhelming. The existence of free will may be worth some evil, but it certainly isn't worth the amount we actually observe. Second, there are lots of evils that can't be traced to the human will, free or unfree. Earthquakes and tornados and genetic defects andwell, one hardly knows where to stop. These two points are familiar ones in discussions of the argument from evil. I want also to make a third point, which, although fairly well-known, is not quite so familiar as these. Let us consider certain particular very bad events—“horrors” I will call them. Here are some examples of what I call horrors: a school bus full of children is crushed by a landslide; a good woman's life is gradually destroyed by the progress of Huntington's Chorea; a baby is born without limbs. Some horrors are consequences of human choices and some are not (consider, for example, William Rowe's [ 1979 ] case of a fawn that dies in agony in a forest fire before there were any human beings). But whether a particular horror is connected with human choices or not, it is evident that God could have prevented the horror without sacrificing any great good or allowing some even greater horror. Now a moment ago I mentioned the enormous amount of evil in the world, and it is certainly true that there is in some sense an enormous amount of evil in the world. But the word “amount” at least suggests that evil is quantifiable, like distance or weight. That may be false or unintelligible, but if it is true, even in a rough-and-ready sort of way, it shows that horrors raise a problem for the theist that is distinct from the problem raised by the enormous amount of evil. If evil can be, even roughly, quantified, as talk about amounts seems to imply, it might be that there was more evil in a world in which there were thousands of millions of relatively minor episodes of suffering (broken ribs, for example) than in a world in which there were a few horrors. But an omnipotent and omniscient creator could be called to moral account for creating a world in which there was even one horror. And the reason is obvious: that horror could have been “left out” of creation without the sacrifice of any great good or the permission of some even greater horror. And leaving it out is exactly what a morally perfect being would do; such good things as might depend causally on the horror could, given the being's omnipotence and omniscience, be secured by (if the word is not morally offensive in this context) more “economical” means. Thus, the sheer amount of evil (which might be distributed in a fairly uniform way) is not the only fact about evil Theist needs to take into account. He must also take into account what we might call (again with some risk of using morally offensive language) high local concentrations of evil—that is, horrors. And it is hard to see how the free-will defense, however elaborated, could provide any resources for dealing with horrors. I will, finally, call your attention to the fact that the case of “Rowe's fawn,” which I briefly described a moment ago, is a particularly difficult case for Theist. True, however sentimental we may be about animals, we must admit that the death of a fawn in a forest fire is not much of a horror compared with, say, a living child's being thrown into a furnace as a sacrifice to Baal. The degree of horror involved in the event is not what creates the special difficulty for theists in this case. What creates the difficulty is rather the complete causal isolation of the fawn's sufferings from the existence and activities of human beings. No appeal to considerations in any way involving human free will can possibly be relevant to the problem with which this case confronts Theist, the difficulty of explaining why an omnipotent and morally perfect being would allow such a thing to happen.

11. Theist Elaborates the Free-will Defense: Evil Results from a Primordial Estrangement of Humanity from God

This is Atheist's response to the free-will defense. How is Theist to reply? If I were he (and in some sense I am), I would reply as follows.

The free-will defense, in the simple form in which I've stated it, suggests—though it does not entail—that God created human beings with free will, and then just left them to their own devices. It suggests that the evils of the world are the more or less unrelated consequences of uncounted millions of largely unrelated abuses of free will by human beings. Let me propose a sort of plot to be added to the bare and abstract story called the free-will defense. Consider the story of creation and rebellion and the expulsion from paradise we find in the first three chapters of Genesis. Could this story be true—I mean literally true, true in every detail? Well, no. It contradicts what science has discovered about human evolution and the history of the physical universe. And that is hardly surprising, for it long antedates these discoveries. The story is a reworking—with much original material—by a Hebrew author or authors of elements found in many ancient Middle Eastern mythologies. Like Virgil's Aeneid , it is a literary refashioning of materials that were originally mythical and legendary, and it retains a strong flavor of myth. It is possible, nevertheless, that the first three chapters of Genesis are a mythicoliterary representation of actual events of human prehistory. The following is consistent with what we know of human prehistory. Our current knowledge of human evolution, in fact, presents us with no particular reason to believe this story is false: For millions of years, perhaps for thousands of millions of years, God guided the course of evolution so as eventually to produce certain very clever primates, the immediate predecessors of Homo sapiens . At some time in the past few hundred thousand years, the whole population of our prehuman ancestors formed a small breeding community—a few thousand or a few hundred or even a few score. That is to say, there was a time when every ancestor of modern human beings who was then alive was a member of this tiny, geographically tightly knit group of primates. In the fullness of time, God took the members of this breeding group and miraculously raised them to rationality. That is, he gave them the gifts of language, abstract thought, and disinterested love—and, of course, the gift of free will. Perhaps we cannot understand all his reasons for giving human beings free will, but here is one very important one we can understand: He gave them the gift of free will because free will is necessary for love. Love, and not only erotic love, implies free will. The essential connection between love and free will is beautifully illustrated in Ruth's declaration to her mother-in-law, Naomi: And Ruth said, Entreat me not to leave thee, or to return from following after thee: for whither thou goest, I will go; and where thou lodgest, I will lodge: thy people shall be my people and thy God my God: where thou diest, will I die, and there will I be buried; the Lord do so to me, and more also, if aught but death part thee and me. (Ruth 1: 16, 17)

It is also illustrated by the vow Mr. van Inwagen, the author of my fictional being, made when he was married:

I, Peter, take thee, Elisabeth, to my wedded wife, to have and to hold from this day forward, for better for worse, for richer for poorer, in sickness and in health, to love and to cherish, till death us do part, according to God's holy ordinance; and thereto I plight thee my troth.

God not only raised these primates to rationality—not only made of them what we call human beings—but also took them into a kind of mystical union with himself, the sort of union Christians hope for in Heaven and call the Beatific Vision. Being in union with God, these new human beings, these primates who had become human beings at a certain point in their lives, lived together in the harmony of perfect love and also possessed what theologians used to call preternatural powers—something like what people who believe in them today call paranormal abilities. Because they lived in the harmony of perfect love, none of them did any harm to the others. Because of their preternatural powers, they were able somehow to protect themselves from wild beasts (which they were able to tame with a look), from disease (which they were able to cure with a touch), and from random, destructive natural events (like earthquakes), which they knew about in advance and were able to avoid. There was thus no evil in their world. And it was God's intention that they should never become decrepit with age or die, as their primate forbears had. But, somehow, in some way that must be mysterious to us, they were not content with this paradisal state. They abused the gift of free will and separated themselves from their union with God.

The result was horrific: not only did they no longer enjoy the Beatific Vision, but they now faced destruction by the random forces of nature, and became subject once more to old age and natural death. Nevertheless, they were too proud to end their rebellion. As the generations passed, they drifted further and further from God—into the worship of invented gods (a worship that sometimes involved human sacrifice), inter-tribal warfare (complete with the gleeful torture of prisoners of war), private murder, slavery, and rape. On one level, they realized, or some of them realized, that something was horribly wrong, but they were unable to do anything about it. After they had separated themselves from God, they were, as an engineer might say, “not operating under design conditions.” A certain frame of mind became dominant among them, a frame of mind latent in the genes they had inherited from a million or more generations of ancestors. I mean the frame of mind that places one's own desires and perceived welfare above everything else, and that accords to the welfare of one's relatives and the other members of one's tribe a subordinate privileged status, and assigns no status at all to the welfare of anyone else. And this frame of mind was now married to rationality, to the power of abstract thought; the progeny of this marriage were continuing resentment against those whose actions interfere with the fulfillment of one's desires, hatreds cherished in the heart, and the desire for revenge. The inherited genes that produced these baleful effects had been harmless as long as human beings had still had constantly before their minds a representation of perfect love in the Beatific Vision. In the state of separation from God, and conjoined with rationality, they formed the genetic substrate of what is called original or birth sin: an inborn ten dency to do evil against which all human efforts are vain. We, or most of us, have some sort of perception of the distinction between good and evil, but, however we struggle, in the end we give in and do evil. In all cultures there are moral codes (more similar than some would have us believe), and the members of every tribe and nation stand condemned not only by alien moral codes but by their own. The only human beings who consistently do right in their own eyes, whose consciences are always clear, are those who, like the Nazis, have given themselves over entirely to evil, those who say, in some twisted and self-deceptive way what Milton has his Satan say explicitly and clearly: “Evil, be thou my Good.”

When human beings had become like this, God looked out over a ruined world. It would have been just for him to leave human beings in the ruin they had made of themselves and their world. But God is more than a God of justice. He is, indeed, more than a God of mercy—a God who was merely merciful might simply have brought the story of humanity to an end at that point, like a man who shoots a horse with a broken leg. But God, as I have said, is more than a God of mercy: he is a God of love. He therefore neither left humanity to its own devices nor mercifully destroyed it. Rather, he set in motion a rescue operation. He put into operation a plan designed to restore separated humanity to union with himself. This defense will not specify the nature of this plan of atonement. The three Abrahamic religions, Judaism, Christianity, and Islam, tell three different stories about the nature of this plan, and I do not propose to favor one of them over another in telling a story that, after all, I do not maintain is true. This much must be said, however: the plan has the following feature, and any plan with the object of restoring separated humanity to union with God would have to have this feature: its object is to bring it about that human beings once more love God. And, since love essentially involves free will, love is not something that can be imposed from the outside, by an act of sheer power. Human beings must choose freely to be reunited with God and to love him, and this is something they are unable to do of their own efforts. They must therefore cooperate with God. As is the case with many rescue operations, the rescuer and those whom he is rescuing must cooperate. For human beings to cooperate with God in this rescue operation, they must know that they need to be rescued. They must know what it means to be separated from him. And what it means to be separated from God is to live in a world of horrors. If God simply “canceled” all the horrors of this world by an endless series of miracles, he would thereby frustrate his own plan of reconciliation. If he did that, we should be content with our lot and should see no reason to cooperate with him. Here is an analogy. Suppose Dorothy suffers from angina, and that what she needs to do is to stop smoking and lose weight. Suppose her doctor knows of a drug that will stop the pain but will do nothing to cure the condition. Should the doctor prescribe the drug for her, in the full knowledge that if the pain is alleviated, there is no chance she will stop smoking and lose weight? Well, perhaps the answer is yes, if that's what Dorothy insists on. The doctor is Dorothy's fellow adult and fellow citizen, after all. Perhaps it would be insufferably paternalistic to refuse to alleviate Dorothy's pain in order to provide her with a motivation to do what is to her own advantage. If one were of an especially libertarian cast of mind, one might even say that someone who did that was “playing God.” It is far from clear, however, whether there is anything wrong with God's behaving as if he were God. It is at least very plausible to suppose that it is morally permissible for God to allow human beings to suffer if the result of suppressing the suffering would be to deprive them of a very great good, one that far outweighed the suffering. But God does shield us from much evil, from a great proportion of the sufferings that would have resulted from our rebellion if he did nothing. If he did not shield us from much evil, all human history would be at least this bad: every human society would be on the moral level of Nazi Germany—or worse, if there is a “worse.” But, however much evil God shields us from, he must leave a vast amount of evil “in place” if he is not to deceive us about what separation from him means—and, in so deceiving us, to remove our only motivation for cooperating with him in the working out of his plan for divine-human reconciliation. The amount he has left us with is so vast and so horrible that we cannot really comprehend it, especially if we are middle-class Europeans or Americans. Nevertheless, it could have been much worse. The inhabitants of a world in which human beings had separated themselves from God and he had then simply left them to their own devices would regard our world as a comparative paradise. All this evil, however, will come to an end. There will come a time after which, for all eternity, there will be no more unmerited suffering. Every evil done by the wicked to the innocent will have been avenged, and every tear will have been wiped away. If there is still suffering, it will be merited: the suffering of those who refuse to cooperate with God in his great rescue operation and are allowed by him to exist forever in a state of elected ruin—those who, in a word, are in Hell.

One aspect of this story needs to be brought out more clearly than it has been. If the story is true, much of the evil in the world is due to chance. There is generally no explanation of why this evil happened to that person. What there is is an explanation of why evils happen to people without any reason. And the explanation is: that is part of what our being separated from God means: it means our being the playthings of chance. It means not only living in a world in which innocent children die horribly, it means living in a world in which each innocent child who dies horribly dies horribly for no reason at all. It means living in a world in which the wicked, through sheer luck, often prosper. Anyone who does not want to live in such a world, a world in which we are the playthings of chance, had better accept God's offer of a way out of that world.

I will call this story the expanded free-will defense. I mean it to include the “simple” free-will defense as a part. Thus, it is a feature of the expanded free-will defense that even an omnipotent being, having raised our remote ancestors to rationality and having given them the gift of free will, which included a free choice between remaining united with him in bonds of love and turning away from him to follow the devices and desires of their own hearts, was not able to ensure that they have done the former—although we may be confident he did everything omnipotence could do to raise the probability of their doing the former. But, before there were human beings, God knew that, however much evil might result from the elected separation from himself, and consequent self-ruin, of his human creatures—if it should occur—the gift of free will would be, so to speak, worth it. For the existence of an eternity of love depends on this gift, and that eternity outweighs the horrors of the very long but, in the most literal sense, temporary period of divine-human estrangement.

Here, then, is a defense, the expanded free-will defense. I contend that the expanded free-will defense is a possible story (internally consistent, at least as far as we can see); that, given that there is a God, the rest of the story might well be true; that it includes evil in the amount and of the kinds we find in the actual world, including what is sometimes called natural evil, such as the suffering caused by the Lisbon earthquake. (Natural evil, according to the expanded free-will defense, is a special case of the evil that results from the abuse of free will; the fact that human beings are subject to destruction by earthquakes is a consequence of a primordial abuse of free will.) I concede that it does not help us with cases like “Rowe's fawn”—cases of suffering that occurred before there were human beings or that are for some other reason causally unconnected with human choice. But I claim to have presented a defense that accounts for all actual human suffering.

That was a long speech on the part of Theist. I now return to speaking in propria persona. I have had Theist tell a story, a story he calls the expanded free-will defense. You may want to ask whether I believe this story I have put into the mouth of my creature. Well, I believe parts of it and I don't disbelieve any of it. (Even those parts I believe do not, for the most part, belong to my faith; they are merely some of my religious opinions.) I am not at all sure about “preternatural powers,” for example, or about the proposition that God shields us from much of the evil that would have been a “natural” consequence of our estrangement from him. But what I believe and don't believe is not really much to the point. The story I have told is, I remind you, only supposed to be a defense. Theist does not put forward the expanded free-will defense as a theodicy, as a statement of the real truth of the matter concerning the coexistence of God and evil. Nor would I, if I told it in circumstances like Theist's. Theist contends only, I contend only, that the story is—given that God exists—true for all anyone knows. And I certainly don't see any very compelling reason to reject any of it. In particular, I don't see any reason to reject the thesis that God raised a small population of our ancestors to rationality by a specific action on, say, June 13, 116,027 bc , or on some such particular date. It is not a discovery of evolutionary biology that there are no miraculous events in our evolutionary history. It could not be, any more than it could be a discovery of meteorology that the weather at Dunkirk during those fateful days in 1940 was not due to a specific and local divine action. It could , of course, be a discovery of evolutionary biology that the genesis of rationality was not a sudden, local event. But no such discovery has been made. If someone, for some reason, put forward the theory that extraterrestrial beings visited the earth, and by some prodigy of genetic engineering, raised some population of our primate ancestors to rationality in a single generation (something like this happened in the movie 2001: A Space Odyssey ), this theory could not be refuted by any facts known to physical anthropology.

12. Atheist Turns to the Consideration of a Particular Horrible Evil

How might Atheist respond to the expanded free-will defense, given that this defense is, as I argued, consistent with what science has discovered about human prehistory? If I were in her position, I would respond to Theist in some such words as these:

You, Theist, may have told a story that accounts for the enormous amount of evil in the world, and for the fact that much evil is not caused by human beings. But I don't think you appreciate the force of the argument from horrors (so to call it), and I think I can make the agnostics, at any rate, see this. Let me state the argument from horrors a little more systematically; let me lay out its premises explicitly, and you can tell me which of its premises you deny. There are many horrors, vastly many, from which no discernible good results—and certainly no good, discernible or not, that an omnipotent being couldn't have got without the horror; in fact, without any suffering at all. Here is a true story. A man came upon a young woman in an isolated place. He overpowered her, chopped off her arms at the elbows with an axe, raped her, and left her to die. Somehow she managed to drag herself on the stumps of her arms to the side of a road, where she was discovered. She lived, but she experienced indescribable suffering, and although she is alive, she must live the rest of her life without arms and with the memory of what she had been forced to endure. No discernible good came of this, and it is wholly unreasonable to believe that any good could have come of it that an omnipotent being couldn't have achieved without employing the raped and mutilated woman's horrible suffering as a means to it. And even if this is wrong and some good came into being with which the woman's suffering was so intimately connected that even an omnipotent being couldn't have got the good without the suffering, it wouldn't follow that that good outweighed the suffering. (It would certainly have to be a very great good to do that.) I will now draw on these reflections to construct a version of the argument from evil, a version that, unlike the version I presented earlier, refers not to all the evils of the world, but just to this one event. (The argument is modeled on the central argument of William Rowe's “The Problem of Evil and Some Varieties of Atheism” [ 1979 ].) I will refer to the events in the story I have told collectively as “the Mutilation.” I argue: (1) If the Mutilation had not occurred, if it had been, so to speak, simply left out of the world, the world would be no worse than it is. (It would seem, in fact, that the world would be significantly better if the Mutilation had been left out of it, but my argument doesn't require that premise.) (2) The Mutilation in fact occurred and was a horror. (3) If a morally perfect creator could have left a certain horror out of the world he created, and if the world he created would have been no worse if that horror had been left out of it than it would have been if it had included that horror, then the morally perfect creator would have left the horror out of the world he created—or at any rate, he would have left it out if he had been able to. (4) If an omnipotent being created the world, he was able to leave the Mutilation out of the world (and was able to do so in a way that would have left the world otherwise much as it is). There is, therefore, no omnipotent and morally perfect creator.

You, Theist, must deny at least one of the four premises of this argument; or at any rate, you must show that serious doubts can be raised about at least one of them. But which?

So speaks Atheist. How might Theist reply? Atheist has said that her argument was modeled on an argument of William Rowe's. If Theist models his reply on the replies made by most of the theists who have written on Rowe's argument, he will attack the first premise (see, for example, Wykstra 1996 ). He will try to show that, for all anyone knows, the world (considered under the aspect of eternity) is a better place for containing the Mutilation. He will try to show that for all anyone knows, God has brought, or will at some future time bring, some great good out of the Mutilation, a good that outweighs it, or else has employed the Mutilation as a means to preventing some even greater evil; and he will argue that, for all anyone knows, the great good achieved or the great evil prevented could not have been, respectively, achieved or prevented, even by an omnipotent being, otherwise than by some means that essentially involved the Mutilation (or something else as bad or worse).

13. Theist Discusses the Relation of the Expanded Free-will Defense to the Question Whether an Omnipotent and Morally Perfect Being Would Eliminate Every Particular Horror from the World

I am not going to have Theist reply to Atheist's argument in this way. I find (1) fairly plausible, even if I am not as sure as Atheist is (or as sure as most atheists who have discussed the issue seem to be) that (1) is true. I am going to represent Theist as employing another line of attack on Atheist's response to his expanded free-will defense. I am going to represent him as denying premise (3), or, more precisely, as trying to show that the expanded free-will defense casts considerable doubt on premise (3). And here is his reply:

Why should we accept premise (3) of Atheist's argument? I have had a look at Rowe's defense of the corresponding premise of his argument, the entirety of which I will quote: “[This premise] seems to express a belief that accords with our basic moral principles, principles shared both by theists and non-theists.” ( 1979 , 337) But what are these “basic moral principles, shared both by theists and non-theists”? Rowe does not say, but I believe there is really just one moral principle it would be plausible to appeal to in defense of premise (3). It might be stated like this. If one is in a position to prevent some evil, one should not allow that evil to occur—not unless allowing it to occur would result in some good that would outweigh it or preventing it would result in some other evil at least as bad.

Is this principle true?

I think not. (I can, in fact, think of several obvious objections to it. But most of these objections would apply only to the case of human agents, and I shall therefore not mention them.) Consider this example. Suppose you are an official who has the power to release anyone from prison at any time. Blodgett has been sentenced to ten years in prison for felonious assault. His sentence is nearing its end, and he petitions you to release him from prison a day early. Should you? Well, the principle says so. A day spent in prison is an evil—if you don't think so, I invite you to spend a day in prison. Let's suppose that the only good that results from putting criminals in prison is the deterrence of crime. (This assumption is made to simplify the argument. That it is false introduces no real defect into the argument.) Obviously, nine years, 364 days spent in prison is not going to have a significantly different power to deter felonious assault from ten years spent in prison. So: no good will be secured by visiting on Blodgett that last day in prison, and that last day spent in prison is an evil. The principle tells you, the official, to let him out a day early. This much, I think, is enough to show that the principle is wrong, for you have no such obligation. But the principle is in more trouble than this simple criticism suggests.

It would seem that if a threatened punishment of n days in prison has a certain power to deter felonious assault, a threatened punishment of n − 1 days spent in prison will have a power to deter felonious assault that is not significantly less. Consider the power to deter felonious assault that belongs to a threatened punishment of 1,023 days in prison. Consider the power to deter felonious assault that belongs to a threatened punishment of 1,022 days in prison. There is, surely, no significant difference. Consider the power to deter felonious assault that belongs to a threatened punishment of 98 days in prison. Consider the power to deter felonious assault that belongs to a threatened punishment of 97 days in prison. There is, surely, no significant difference. Consider the power to deter felonious assault that belongs to a threatened punishment of one day in prison. Consider the power to deter felonious assault that belongs to a threatened punishment of no time in prison at all. There is, surely, no significant difference. (In this last case, of course, this is because the threat of one day in prison would have essentially no power to deter felonious assault.)

A moment's reflection shows that if this is true, as it seems to be, then the moral principle entails that Blodgett ought to spend no time in prison at all. For suppose Blodgett had lodged his appeal to have his sentence reduced by a day not shortly before he was to be released but before he had entered prison at all. He lodges this appeal with you, the official who accepts the moral principle. For the reason I have set out, you must grant his appeal. Now suppose that when it has been granted, clever Blodgett lodges a second appeal: that his sentence be reduced to ten years minus two days. This second appeal you will also be obliged to grant, for there is no difference between ten years less a day and ten years less two days as regards the power to deter felonious assault. I am sure you can see where this is going. Provided only that Blodgett has the time and the energy to lodge 3,648 successive appeals for a one-day reduction of his sentence, he will escape prison altogether.

This result is, I take it, a reductio ad absurdum of the moral principle. As the practical wisdom has it (and this is no compromise between practical considerations and strict morality; it is strict morality), You have to draw a line somewhere. And this means an arbitrary line. The principle fails precisely because it forbids the drawing of morally arbitrary lines. There is nothing wrong, or nothing that can be determined a priori to be wrong, with a legislature's setting ten years in prison as the minimum punishment for felonious assault—and this despite the fact that ten years in prison, considered as a precise span of days , is an arbitrary punishment.

The moral principle is therefore false—or possesses whatever defect is the analogue in the realm of moral principles of falsity in the realm of factual statements. What are the consequences of its falsity, of its failure to be an ac ceptable moral principle, for the “argument from horrors”? Let us return to the expanded free-will defense. This story accounts for the existence of horrors—that is, that there are horrors is a part of the story. The story explains why there are such things as horrors (at least, it explains why there are postlapsarian horrors) although it says nothing about any particular horror. And to explain why there are horrors is not to meet the argument from horrors.

A general account of the existence of horrors does not constitute a reply to the argument from horrors because it does not tell us which premise of the argument to deny. Let us examine this point in detail. According to the expanded free-will defense, God prevents the occurrence of many of the horrors that would naturally have resulted from our separation from him. But he cannot, so to speak, prevent all of them, for that would frustrate his plan for reuniting human beings with himself. And if he prevents only some horrors, how shall he decide which ones to prevent? Where shall he draw the line—the line between threatened horrors that are prevented and threatened horrors that are allowed to occur? I suggest that wherever he draws the line, it will be an arbitrary line. That this must be so is easily seen by thinking about the Mutilation. If God had added that particular horror to his list of horrors to be prevented, and that one alone, the world, considered as a whole, would not have been a significantly less horrible place, and the general realization of human beings that they live in a world of horrors would not have been significantly different from what it is. The existence of that general realization is just the factor in his plan for humanity that (according to the expanded free-will defense) provides his general reason for allowing horrors to occur. Therefore, preventing the Mutilation would in no way have interfered with his plan for the restoration of our species. If the expanded free-will defense is a true story, God has made a choice about where to draw the line, the line between the actual horrors of history, the horrors that are real , and the horrors that are mere averted possibilities, might-have-beens. The Mutilation falls on the “actual horrors of history” side of the line. And this fact shows that the line is an arbitrary one, for if he had drawn it so as to exclude the Mutilation from reality (and left it otherwise the same) he would have lost no good thereby and he would have allowed no greater evil. He had no reason for drawing the line where he did. But then what justifies him in drawing the line where he did? What justifies him in including the Mutilation in reality when he could have excluded it without losing any good thereby? Has the victim of the Mutilation not got a moral case against him? He could have saved her and he did not, and he does not even claim to have achieved some good by not saving her. It would seem that God is in the dock, in C. S. Lewis's words; if he is, then I, Theist, am playing the part of his barrister, and you, the Agnostics, are the jury. I offer the following obvious consideration in defense of my client: there was no nonarbitrary line to be drawn. Wherever God drew the line, there would have been countless horrors left in the world—his plan requires the actual existence of countless horrors—and the victim or victims of any of those horrors could bring the same charge against him that we have imagined the victim of the Mutilation bringing against him.

But I see Atheist stirring in protest; she is planning to tell you that, given the terms of the expanded free-will defense, God should have allowed the minimum number of horrors consistent with his project of reconciliation, and that it is obvious he has not done this. She is going to tell you that there is a nonarbitrary line for God to draw, and that it is the line that has the minimum number of horrors on the “actuality” side. But there is no such line to be drawn. There is no minimum number of horrors consistent with God's plan of reconciliation, for the prevention of any one particular horror could not possibly have any effect on God's plan. For any n , if the existence of n horrors is consistent with God's plan, the existence of n −1 horrors will be equally consistent with God's plan. To ask what the minimum number of horrors consistent with God's plan is is like asking, What is the minimum number of raindrops that could have fallen on England in the nineteenth century that is consistent with England's having been a fertile country in the nineteenth century? Here is a simple analogy of proportion: a given evil is to the openness of human beings to the idea that human life is horrible and that no human efforts will ever alter this fact as a given raindrop is to the fertility of England.

And this is why God did not prevent the Mutilation—insofar as there is a “why.” He had to draw an arbitrary line and he drew it. And that's all there is to be said. This, of course, is cold comfort to the victim. Or, since we are merely telling a story, it would be better to say: if this story were true and known to be true, knowing its truth would be cold comfort to the victim. But the purpose of the story is not to comfort anyone. It is not to give an example of a possible story that would comfort anyone if it were true and that person knew it to be true. If a child dies on the operating table in what was supposed to be a routine operation and a board of medical inquiry finds that the death was due to some factor the surgeon could not have anticipated and that the surgeon was not at fault, that finding will be of no comfort to the child's parents. But it is not the purpose of a board of medical inquiry to comfort anyone; the purpose of a board of medical inquiry is, by examining the facts of the matter, to determine whether anyone was at fault. And it is not my purpose in offering a defense to provide even hypothetical comfort to anyone. It is to determine whether the existence of horrors entails that God is at fault—or, rather, since by definition God is never at fault, to determine whether the existence of horrors entails that an omnipotent creator would be at fault.

It is perhaps important to point out that we might easily find ourselves in a moral situation like God's moral situation according to the expanded free-will defense, a situation in which we must draw an arbitrary line and allow some bad thing to happen when we could have prevented it, and in which, moreover, no good whatever comes of our allowing it to happen. In fact, we do find ourselves in this situation. In a welfare state, for example, we use taxation to divert money from its primary economic role in order to spend it to prevent or alleviate various social evils. And how much money, what proportion of the gross national product, shall we—that is, the state—divert for this purpose? Well, not none of it and not all of it (enforcing a tax rate of 100 percent on all earned income and all profits would be the same as not having a money economy at all). And where we draw the line is an arbitrary matter. However much we spend on social services, we shall always be able to find some person or family who would be saved from misery if the state spent (in the right way) a mere $1,000 more than it in fact plans to spend. And the state can always find another $1,000, and can find it without damaging the economy or doing any other sort of harm.

14. Concluding Remarks: Evaluating Theist's Response to the Argument from Evil

So Theist replies to Atheist's argument from horrors. But we may note that Theist has failed to respond to an important point Atheist has made. As he himself conceded, his reply takes account only of postlapsarian horrors. There is still to be considered the matter of prelapsarian horrors, horrors such as Rowe's poor fawn. There were certainly sentient animals long before there were sapient animals, and the paleontological record shows that for much of the long prehuman past, sentient creatures died agonizing deaths in natural disasters. Obviously, the free-will defense cannot be expanded in such a way as to account for these agonizing deaths, for only sapient creatures have free will, and these deaths cannot therefore have resulted from the abuse of free will—unless, as C. S. Lewis has suggested, prehuman animal suffering is ascribed to a corruption of nature by fallen angels ( 1940 , 122–24). Interesting as this suggestion is, I do not propose to endorse it, even as a defense. I confess myself unable to treat this difficult problem adequately within the scope of this chapter. I should have to devote a whole essay to the problem of prelapsarian horrors to say anything of value about it. I must simply declare this topic outside the scope of this chapter. I refer the reader to my essay “The Problem of Evil, the Problem of Air, and the Problem of Silence,” (van Inwagen 1991 ), which contains a defense—not a version of the free-will defense—that purports to account for the sufferings of prehuman animals. I will remark that this defense shares one important feature of the expanded free-will defense. This defense, too, requires God to draw an arbitrary line; it allows God to eliminate much animal suffering that would otherwise have occurred in the course of nature, but it requires him, as it were, to stop eliminating it at some point, even though no good is gained by his stopping at whatever point he does stop at. I would thus say that God could have eliminated the suffering of Rowe's fawn at no cost and did not, and that this fact does not count against his moral perfection—just as the fact that he could have eliminated the Mutilation at no cost and did not does not count against his moral perfection. But the nature of the goods involved in this other defense is a subject I cannot discuss here.

Let me put this question to the readers of this chapter: Has Theist successfully replied to the argument from horrors insofar as those horrors are events that involve human beings ? Well, much depends on what further things Atheist might have to say. Perhaps Atheist has a dialectically effective rejoinder to Theist's reply to the argument from horrors. But one must make an end somewhere. The trouble with real philosophical debates is that they almost never come to a neat and satisfactory conclusion. Philosophy is argument without end. I do think this much: if Atheist has nothing more to say, the Agnostics should render a verdict of “not proven” as regards premise (3) of the argument from horrors and the moral principle on which it is based, namely, that, if it is within one's power to prevent some evil, one should not allow that evil to occur unless allowing it to occur would result in some good that would outweigh it or preventing it would result in some other evil at least as bad.

Let me put a similar question before the readers of this chapter as regards the extended free-will defense and the problem of the vast amount of evil (including the vast amounts of natural evil): Does Theist's presentation of the extended free-will defense constitute a successful reply to Atheist's contention that an omnipotent and morally perfect God would not allow the existence of a world that contains evil in the amount and of the kinds we observe in the world around us insofar as this contention involves only evils that befall human beings ? Again, much depends on what further things Atheist might have to say. My own opinion is this: if Atheist has nothing further to say, an audience of agnostics of the sort I have imagined should concede that for all anyone knows , a world created by an omnipotent and morally perfect God might contain human suffering in the amount and of the kinds we observe. 3

In the novel, there are several minor illiteracies in the poem (e.g., “whose” for “who's” in the first stanza). (The fictional author of the poem, a well-educated man, was trying to hide the fact of his authorship.) I have corrected these, despite the judgment of Martin Amis that the illiteracies are an intended part of the literary effect of the poem (intended, that is, by its real author, Kingsley Amis, not by its fictional author).

Almost all theists who reply to the argument from evil employ some form of the free-will defense. The free-will defense I am going to have Theist employ derives, at a great historical remove, from Saint Augustine. A useful selection of Augustine's writings on free will and the origin of evil (from The City of God and the Enchiridion ) can be found in Melden ( 1955 , 164–77).

For a very different approach to the problem of evil (to the purely intellectual problem considered in this chapter and to many other problems connected with trust in God and the very worst evils present in his creation), see Marilyn McCord Adams, Hor rendous Evils and the Goodness of God ( 1999 ). I find this book unpersuasive (as regards its general tendency and main theses; I think Adams is certainly right about many relatively minor but not unimportant points), but endlessly fascinating. I hope that my friend Marilyn, if she reads the sentence to which this note is appended, will take special notice of the words “seems to me,” and will accept my assurance that their presence in that sentence is not a mere literary reflex.

For another important but very different discussion of the problem of evil, see Eleonore Stump's Stob Lectures, Faith and the Problem of Evil ( 1999 ).

Many recent versions of the free-will defense (including the version developed in the seminal work of Alvin Plantinga) can be found in Pike ( 1964 ), Adams and Adams ( 1990 ), and Peterson ( 1992 ), collections that contain excellent and representative selections from the important philosophical work on the argument from evil that had been published as of their copyright dates.

Three important book-length treatments of the problem of evil, all in the Augustinian (or “free will”) tradition, are Lewis ( 1940 ), Geach (1977), and Swinburne ( 1998 ).

For another version of Theist's argument (in which something like the story here called the expanded free will defense is presented not as a defense but as a theodicy—a “theodicy” in a weaker sense than the word is given in this chapter), see van Inwagen ( 1988 ).

A longer version of the debate between Atheist and Theist concerning the “argument from horrors” is contained in van Inwagen ( 2000 ).

Works Cited

Adams, Marilyn McCord . 1999 . Horrendous Evils and the Goodness of God. Ithaca, N.Y.: Cornell University Press.

Google Scholar

Google Preview

Adams, Marilyn McCord , and Robert Merrihew Adams , eds. 1990 . The Problem of Evil. Oxford: Oxford University Press.

Amis, Kingsley . 1966 . The Anti-death League. London: Victor Gollancz.

Geach, P. T. 1997 . Providence and Evil. Cambridge, England: Cambridge University Press.

Howard-Snyder, Daniel , ed. 1996 . The Evidential Argument from Evil. Bloomington: Indiana University Press.

Lewis, C. S . 1940 . The Problem of Pain. London: Macmillan.

Melden, A. I ., ed. 1955 . Ethical Theories. 2nd edition. Englewood Cliffs, N.J.: Prentice-Hall.

Mill, John Stuart . 1875 . Three Essays on Religion. London: Longmans, Green.

Peterson, Michael L. , ed. 1992 . The Problem of Evil. Notre Dame, Ind.: University of Notre Dame Press.

Pike, Nelson , ed. 1964 . God and Evil. Englewood Cliffs, N.J.: Prentice-Hall.

Rowe, William L. 1979 . “ The Problem of Evil and Some Varieties of Atheism. ” American Philosophical Quarterly 16: 335–41. Reprinted in Adams and Adams, 1990.

Stump, Eleonore . 1999 . Faith and the Problem of Evil. Grand Rapids, Mich.: Stob Lectures Endowment.

Swinburne, Richard . 1998 . Providence and the Problem of Evil. Oxford: Oxford University Press. 10.1093/0198237987.001.0001

Van Inwagen, Peter . 1988 . “ The Magnitude, Duration, and Distribution of Evil: A Theodicy. ” Philosophical Topics 16: 161–87. Reprinted in van Inwagen 1995; in leonore Stump and Michael Murray , eds., The Big Questions: Philosophy of Religion (Oxford: Blackwell, 1999); in Joel Feinberg and Russ Shafer-Landau , eds., Reason and Responsibility , 11th edition (Belmont, Calif.: Wadsworth/Thompson, 2002); and in William Lane Craig , ed., Philosophy of Religion: A Contemporary Reader , forthcoming from Edinburgh University Press.

Van Inwagen, Peter . 1991 . “ The Problem of Evil, the Problem of Air, and the Problem of Silence. ” Philosophical Perspectives, Vol. 5, Philosophy of Religion : 135–65. Reprinted in Howard-Snyder 1996, and in van Inwagen 1995.

Van Inwagen, Peter . 1995 . God, Knowledge, and Mystery: Essays in Philosophical Theology. Ithaca, N.Y.: Cornell University Press.

Van Inwagen, Peter . 2000 . “ The Argument from Particular Horrendous Evils. ” Proceedings of the American Catholic Philosophical Association (annual supplement to the American Catholic Philosophical Quarterly ) 74: 65–80.

Wykstra, Stephen John . 1996. “Rowe's Noseeum Arguments from Evil.” In Howard-Snyder 1996, 126–50.

About Oxford Academic
Publish journals with us
University press partners
What we publish
New features
Open access
Institutional account management
Rights and permissions
Get help with access
Accessibility
Advertising
Media enquiries
Oxford University Press
Oxford Languages
University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

Copyright © 2024 Oxford University Press
Cookie settings
Cookie policy
Privacy policy
Legal notice

This Feature Is Available To Subscribers Only

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Featured Essay The Love of God An essay by Sam Storms Read Now
Faithfulness of God
Saving Grace
Adoption by God

U.S. Edition

Arts & Culture
Bible & Theology
Christian Living
Current Events
Faith & Work
Gospelbound
TGC Podcast
The Carson Center Podcast
The Everyday Pastor
You're Not Crazy
As In Heaven
Churches Planting Churches
Help Me Teach The Bible
Post-Christianity?
Word Of The Week
Past Conference Media
Foundation Documents
Advertise With Us
Regional Chapters
Church Directory
Global Resourcing
Donate to TGC

To All The World

The world is a confusing place right now. We believe that faithful proclamation of the gospel is what our hostile and disoriented world needs. Do you believe that too? Help TGC bring biblical wisdom to the confusing issues across the world by making a gift to our international work.

The Problem of Evil

Other essays.

“The problem of evil” is one of the most discussed objections to the existence of God and is a top reason many unbelievers give for their unbelief. These objectors argue that since there are so many cases of significant pain and suffering in the world that God could easily prevent, the fact that all this evil was not prevented means it is very unlikely (if not impossible) that God exists.

“The problem of evil” appeals to the phenomenon of evil (significant cases of pain and suffering) as evidence against the existence of God. For many, this evidence appears decisive, because if God existed, he would be powerful enough to prevent such evil, and good enough to want to prevent such evil. Since there is evil, no such powerful and good being exists. For the past two millennia Christians have typically urged two points in reply: theodicy and inscrutability. First, God may very well have a good reason for allowing the evil he does allow – a reason compatible with his holy and good character – and the way of theodicy goes on to list a number of these reasons. Second, the fact that unbelievers may not be able to discern or correctly guess at God’s justifying reason for allowing evil is no good reason to think he doesn’t have a reason. Given the infinity of God’s omniscience, the complexity of his providence, the depth of the goods he aims at, and our own substantial cognitive limitations, we shouldn’t expect to guess God’s reasons.

What Is the Problem of Evil?

The so-called “problem of evil” is an argument against the existence of God that reasons along these lines:

A perfectly powerful being can prevent any evil.
A perfectly good being will prevent evil as far as he can.
God is perfectly powerful and good.
So, if a perfectly powerful and good God exists, there will be no evil.
There is evil.
Therefore, God doesn’t exist.

“Evil,” here is understood as any significant case of pain and suffering in the world, whether “moral” (evil willfully caused by human beings such as murder, adultery, theft, rape, etc.) or “natural” evil (harm caused by impersonal forces of nature such as earthquakes, tornadoes, plague, etc.).

Responding to the Problem of Evil

Nonstarters.

A Christian must be truthful and face the question honestly. It will not do to deny that evil exists (#5 above), for evil is the very presumption of the gospel. Nor can we deny that God could prevent evil (#1 above) or that he is perfect in power and goodness (#3). However, we can (and should) question the second premise above – that a perfectly good God must prevent all evil – for it doesn’t necessarily follow from God’s perfect goodness that he will prevent every evil he can prevent. Perhaps God has a good reason for permitting evil rather than preventing it; if so, then his permission of evil is justified and doesn’t militate against his goodness.

The Ways of Theodicy and Inscrutability

Our response the problem of evil, then, may take either of two approaches. We may argue that the second premise above is false and seek to demonstrate that it is false by showing God’s reasons for permitting evil – the way of “theodicy.” Or we could argue that the second premise is unproven because unbelievers can’t rule out God’s having a good reason for permitting evil – the way of “inscrutability.”

The way of theodicy (from the Greek theos , “God,” and dikaios , “just”; hence, a justification of the ways of God in his dealings with men) seeks to demonstrate God’s reasons for permitting evil. The idea is that by allowing evil God attains greater good than possible apart from evil. The way of theodicy shows that premise (2) is false, arguing that God wouldn’t prevent every evil he could prevent.

The way of inscrutability argues, more modestly, that no one knows that premise (2) is true because no one can know enough to conclude that God doesn’t have good reason for permitting evil. We just cannot grasp God’s knowledge, the complexity of his plans, or the deep nature of the good he aims at in providence. And there is no proof that God does not have good reasons for allowing evil, but because he is good we can only assume that he does. Here we don’t have to come up with ‘theodicies’ to defend God against the problem of evil. Rather, the way of inscrutability shows that it is entirely to be expected that creatures like us can’t come up with God’s reasons, given who God is and who we are.

The Way of Theodicy

Two popular theodicies that have no biblical basis ..

Some theodicies that have been offered lack solid biblical grounding. The free will theodicy , for example, argues that moral evil is due to human abuse of free will. The value of free will is a great good: the possibility of morally good choice and of human beings imaging God by way of these choices. But free will has the unfortunate consequence of allowing for the possibility of moral evil. In response to this we might ask, if free will of this sort is so valuable then why doesn’t God have it, and why won’t we have it in heaven?

The natural law theodicy argues that natural evil is due to the laws of nature. The value of laws of nature is a great good: a stable environment needed for making rational choices of any sort. But laws of nature have the unfortunate consequence of allowing for the possibility of natural disasters (e.g., earthquakes, hurricanes, etc.). In response to this we might ask, if a stable environment requires the possibility of natural evil by requiring laws of nature then why isn’t there any natural evil in the pre-fall Garden of Eden or in the new heavens and the new earth?

Four popular theodicies have some biblical basis

By contrast, at least four theodicies have been offered that have some biblical basis. The punishment theodicy argues that suffering is a result of God’s just punishment of evildoers (Gen 3:14-19; Rom 1:24-32, 5:12, 6:23, 8:20-21; Isa 29:5-6; Ezek 38:19; Rev 6:12; 11:13; 16:18). In punishment God aims at the good of displaying his judgment against sin. The soul-building theodicy argues that suffering leads us from self-centeredness to other-centeredness (Heb 12:5-11; Rom 5:3-5; 2Cor 4:17; Jas 1:2-4; 1Pet 1:6-7; cf. Prov 10:13, 13:24; 22:15; 23:13-24, 29:15). In painful providences God aims at the good of displaying his goodness in shaping our character for good. The pain as God’s megaphone theodicy argues that pain is God’s way of getting the attention of unbelievers in a noncoercive way so that they might forget the vanities of earth, consider spiritual things instead, and perhaps even repent of sin (Luke 13:1-5). In pain God aims at the good of displaying his mercy that through such warnings we might be delivered from the wrath to come. The higher-order goods theodicy says that some goods can’t exist apart from the evils to which they are a response. There is no courage without danger, no sympathy without suffering, no forgiveness without sin, no atonement without suffering, no compassion without need, no patience without adversity. God must often allow lots of evils to make these goods a part of his world, given how these goods are defined (Eph 1:3-10; 1Pet 1:18-20).

These theodicies fall under the umbrella of the “greater good theodicy.”

A “greater good theodicy” (GGT) argues that the pain and suﬀering in God’s world play a necessary role in bringing about greater goods that could not be brought about otherwise. The question that remains, then, is just this: does the Bible really teach that God aims at great goods by way of various evils?

Constructing the “Greater Good Theodicy”: a Three-Fold Argument for Three Biblical Themes

Our argument here is that Scripture combines the ways of theodicy and inscrutability . The biblical accounts of Job, Joseph, and Jesus reveal the goodness of God in the midst of evil, weaving together these three themes:

God aims at great goods (either for mankind, or for himself, or both).
God often intends these great goods to come about by way of various evils .
God leaves created persons in the dark (in the dark about which goods are indeed his reasons for the evils, or about how the goods depend on the evils).

Thus, the Bible seems to strongly suggest that the GGT (God’s aiming at great goods by way of various evils) is in fact his modus operandi in providence, his “way of working.” But this GGT is tempered by a good dose of divine inscrutability.

The Case of Job

In the case of Job God aims at a great good: his own vindication – in particular, the vindication of his worthiness to be served for who he is rather than for the earthly goods he supplies (Job 1:11; 2:5). God intends the great good of the vindication of his own name to come to pass by way of various evils . These are a combination of moral evil and natural evil (Job 1:15, 16, 17, 19, 21-22; 2:7, 10; 42:11). God also leaves Job in the dark about what God is doing , for Job has no access to the story’s prologue in chapter 1. And when God speaks to him “out of the whirlwind” he never reveals to Job why he suffered. Instead, Job’s ignorance of the whole spectrum of created reality is exposed (Job 38:4-39:30; 40:6-41:34), and Job confesses his ignorance of both creation and providence (Job 40:3-5; 42:1-6).

The Case of Joseph

In the case of Joseph we find the same. God aims at great goods: saving the broader Mediterranean world from a famine, preserving his people amid such danger, and (ultimately) bringing a Redeemer into the world descended from such Israelites (Matt 1:1-17; Luke 3:23-38). God intends the great good of the preservation of his people from famine to come to pass by way of various evils (Gen 45:5, 7; Psa 105:16-17), including Joseph’s betrayal, being sold into slavery, and suffering unjust accusation and imprisonment (Gen 37, 39). Joseph sees these evils as the means of God’s sovereign providence (Gen 50:20). But God leaves Joseph’s brothers, the Midianite traders, Potiphar’s wife, and the cupbearer in the dark . None of these people knew the role their blameworthy actions would play in preserving God’s people in a time of danger. They had no clue which goods depended on which evils, or that the evils would even work toward any goods at all.

The Case of Jesus

And in the case of Jesus we see the same again. God aims at great goods: the redemption of his people by the atonement of Christ and the glorification of God in the display of his justice, love, grace, mercy, wisdom, and power. God intends the great good of atonement to come to pass by way of various evils : Jewish plots (Matt 26:3-4, 14-15), Satan’s promptings (John 13:21-30), Judas’s betrayal (Matt 26:47-56; 27:3-10; Luke 22:22), Roman injustice (Matt 26:57-68), Pilate’s cowardice (Matt 27:15–26), and the soldiers’ brutality (Matt 27:27-44). But God leaves various created agents (human and demonic) in the dark , for it is clear that the Jewish leaders, Satan, Judas, Pilate, and the soldiers are all ignorant of the role they play in fulfilling the divinely prophesied redemptive purpose by the cross of Christ (Acts 2:23, 3:18, 4:25-29; John 13:18, 17:12, 19:23-24).

Licensing and Limiting the GGT

In each narrative, the first two themes highlight the way of theodicy (God aiming at great goods by way of evils), while the third theme highlights the way of inscrutability (left to ourselves, we cannot discern what God’s reasons are for any case of evil). By way of the first two themes Scripture repeatedly encourages the view that God has a justifying reason for permitting the evils of the world. That is what’s right with the way of theodicy. But Scripture, by way of the third theme, repeatedly discourages the view that we can ever know what that reason is in any particular case of evil. That is what’s right with the way of inscrutability. In contemporary philosophy, these are usually presented as two different ways to solve the problem of evil (theodicy and inscrutability). However, the Bible seems to combine these two ways when it speaks of God’s relation to the evils in the world. That is, it licenses the greater good theodicy as an overall perspective on evil, but wisely limits that perspective in a way that is instructive for both Christians and non-Christians.

Licensing the GGT: God’s Sovereignty over All Evil

God’s sovereignty over natural evil.

It is one thing to acknowledge God’s sovereign and purposeful providence over the moral and natural evils mentioned in the Job, Joseph, and Jesus narratives. It is quite another to claim that God is sovereign over all moral and natural evils. But this is what the Bible repeatedly teaches. This takes us a considerable way towards licensing the GGT as a general approach to the problem of evil. The Bible presents multitudes of examples of God intentionally bringing about natural evils – famine, drought, rampaging wild animals, disease, birth defects such as blindness and deafness, and even death itself – rather than being someone who merely permits nature to ‘do its thing’ on its own. Here are some samples:

Famine (Deut 32:23-24; 2Kgs 8:1; Psa 105:16; Isa 3:1; Ezek 4:16, 5:16-17, 14:13, 14:21; Hos 2:9; Amos 4:6, 9; Hag 2:17)
Drought (Deut 28:22; 1Kgs 8:35; Isa 3:1; Hos 2:3; Amos 4:6-8; Hag 1:11)
Rampaging wild animals (Lev 26:22; Num 21:6; Deut 32:23-24; 2Kgs 17:25; Jer 8:17; Ezek 5:17, 14:15, 14:21, 33:27)
Disease (Lev 26:16, 25; Num 14:12; Deut 28:21-22, 28:27; 2Kgs 15:5; 2Chron 21:14, 26:19-20)
Birth defects such as blindness and deafness (Exod 4:11; John 9:1-3)
Death itself (Deut 32:39; 1Sam 2:6-7)
Ten Egyptian plagues (Exod 7:14-24, 8:1-15, 8:16-19, 8:20-32, 9:1-7, 9:8-12, 9:13-35, 10:1-20, 10:21-29, 11:4-10, 12:12-13, 12:27-30)
‘Impersonal’ forces and objects (Psa 65:9-11, 77:18, 83:13-15, 97:4, 104:4, 104:10-24, 107:25, 29, 135:6-7, 147:8, 147:16-18, 148:7-8, Jonah 1:4, Nah 1:3-4, Zech 7:14, Matt 5:45, Acts 14:17)

God’s Sovereignty over Moral Evil

In addition, and perhaps surprisingly, the Bible presents God as having such meticulous control over the course of human history that a wide range of moral evils – murder, adultery, disobedience to parents, rejecting wise counsel, even human hatred – can be regarded as “of the Lord.” Without erasing or suppressing the intentionality of creatures – and this includes their deliberations, their reasoning, their choosing between alternatives they consider and reflect upon – God’s own intentionality stands above and behind the responsible choices of his creatures. Again, some samples:

Eli’s sons’ disobedience (1Sam 2:23-25)
Samson’s desire for a foreign wife (Jdg 14:1-4)
Absalom, Rehoboam, and Amaziah rejecting wise counsel (2Sam 17:14; 1Kgs 12:15; 2Chron 25:20)
Assassination (2Chron 22:7, 9, 32:21-22)
Adultery (2Sam 12:11-12, 16:22)
Human hatred (Psa 105:23-25; Exod 4:21; Deut 2:30, 32; Josh 11:20; 1Kgs 11:23, 25; 2Chron 21:16-17)

God’s Sovereignty over All Evil

So the Job, Joseph, and Jesus passages are not anomalies, but part and parcel of a more general view the Bible takes on the subject, with respect to both natural and moral evil. Indeed, in addition to this large swath of ‘particular’ texts about individual cases of evil, there are quite a few “universal” texts which seem to trace all calamities, all human decision-making, all events whatsoever, back to the will of God.

God’s sovereignty over all calamity (Ecc 7:13-14; Isa 45:7; Lam 3:37-38; Amos 3:6)
God’s sovereignty over all human decision-making (Prov 16:9, 19:21, 20:24, 21:1; Jer 10:23)
God’s sovereignty over all events whatsoever (Psa 115:3; Prov 16:33; Isa 46:9-10; Rom 8:28, 11:36; Eph 1:11)

Limiting the GGT: The Inscrutability of God’s Purposes

Establishing the burden of proof.

Of course, each specific theodicy mentioned earlier has significant limitations. For instance, the Bible frequently discourages the idea that the punishment theodicy can explain all evils in the world (Job 1:1, 1:8, 2:3, 42:7-8; John 9:1-3; Acts 28:1-6). More generally, Christians can never know enough about a person’s situation, or about God’s purposes, to rule in a specific theodicy as being God’s reason for permitting evil in a particular case. In fact, it would be entirely presumptuous to do so. But if he who affirms must prove, then the question in the problem of evil is not whether Christians know enough to “rule in” the applicability of a theodicy on any particular occasion, but whether critics know enough to “rule out” the applicability of any theodicy. But how could a critic reasonably claim to know that there is no reason that would justify God in permitting suffering? How could he know that premise (2) of the original argument is true? For why think that God’s reasons for permitting particular cases of evil are the kinds of things that we would discern by our cognitive capacities, if such reasons were there?

Analogies for our Cognitive Limitations

It is widely recognized that we have cognitive limitations with respect to discerning goods and connections, at least in territories where we lack the relevant expertise, experience, or vantage point. Some examples:

It doesn’t seem to me that there is a perfectly spherical rock on the dark side of the moon right now, but that’s no reason to conclude that such a rock isn’t there.
It didn’t seem to any medievals that the theories of special relativity or quantum mechanics were true, but that was no reason to think they weren’t true.
It didn’t seem to humans in earlier eras that fundamental human rights of one sort or another were in fact fundamental human rights, but that was no reason to think there weren’t any such rights.
It wouldn’t seem to a non-Greek-speaker that spoken Greek sentences have any meaning, but that is no reason to think they don’t have a meaning.
It wouldn’t seem to the musically uninitiated that Beethoven projected the ‘sonata form’ onto the symphony as a whole, giving the entire musical work a fundamental unity it would not otherwise have had. But it wouldn’t follow from their ignorance that Beethoven didn’t have such a purpose, much less that he was unsuccessful in executing it.
It might not seem to my one-month-old son that I have a good reason for him to receive a painful series of shots at the doctor’s office. But it wouldn’t follow from his ignorance that there isn’t a good reason.

God is omniscient, which means he not only knows everything that we are likely to guess at, but every truth whatsoever. This means that God knows things that we cannot even fathom. As the above analogies suggest, this is easily demonstrated for a huge range of cases. If the complexities of an infinite God’s divine plan for the unfolding of the universe does involve God’s recognizing either deep goods, or necessary connections between various evils and the realization of those goods, or both of these things, would our inability to discern these goods or connections give us a reason for thinking they aren’t there? What would be the basis of such confidence? But without such confidence, we have little reason to accept premise (2) of the problem of evil. So we have little reason to accept its conclusion.

Biblical Argument for Divine Inscrutability

The theme of divine inscrutability is not only exceedingly defensible common sense. It also looms large in the Bible, having both pastoral and apologetic implications. It closes the mouths of Christians who would insensitively offer “God’s reasons” to those who suffer (when they don’t know such reasons). And it closes the mouths of critics who would irrationally preclude divine reasons for the suffering. Imagine we were on the scene in the cases of Job (as his friend), Joseph (as his brother), and Jesus (as his tormentor). Would we have been able to guess at God’s purpose for the suffering? Would we not instead have been wholly unaware of any such purpose? Does not a large part of the literary power of the Bible’s narrative, and the spiritual encouragement it offers, rest upon this interplay between the ignorance of the human actors and the wisdom of divine providence?

One of the most extended reflections in the New Testament on the problem of evil – in this case, the evil of Jewish apostasy – is Romans 9-11. Paul’s concluding doxology blends together these twin themes of divine sovereignty over evil and divine inscrutability in the midst of evil:

Oh, the depth of the riches and wisdom and knowledge of God! How unsearchable are his judgments and how inscrutable his ways! ‘For who has known the mind of the Lord, or who has been his counselor?’ ‘Or who has given a gift to him that he might be repaid?’ For from him and through him and to him are all things. To him be glory forever. Amen (Rom 11:33–36).

To the extent that God has not spoken about a particular event in history, his judgments are unsearchable, and his paths are beyond tracing out. But that does not mean there is not a greater good which justifies God’s purposing of that event.

The problem of evil summary notes

OCR Philosophy

This page contains summary revision notes for the Problem of Evil topic. There are two versions of these notes. Click on the A*-A grade tab, or the B-C grade tab, depending on the grade you are trying to get.

Find the full revision page here.

This topic unfortunately cannot be split into neat paragraphs.

There are jigsaw puzzle pieces that you have to learn and figure out which to put into each essay, including whether to do the full AO1 if they are the focus or minor AO1 if not.

Here are the pieces:

The logical problem of evil
The evidential problem of evil
Augustine’s theodicy

Evaluation of Augustine vs the logical problem

Evaluation of Augustine vs the evidential problem

Irenaeus/Hick’s theodicy
Evaluation of Irenaeus/Hick vs the logical problem
Evaluation of Irenaeus/Hick vs the evidential problem
The issue of theodicies vs free will

The logical problem of evil (Mackie)

The logical problem of evil is a deductive argument. It aims to show that the existence of evil is logically inconsistent with the existence of the God of classical theism – God defined as omnipotent, omnibenevolent and omniscient.
The classical logical problem of evil: epicurus
Epicurus (ancient Greek philosopher, one of the first to formulate the problem of evil)
Is God willing but not able to prevent evil? Then he isn’t omnipotent
Is God is able to prevent evil but not willing? Then he isn’t omnibenevolent
If God is both able and willing, then why is there evil?
If God is neither able or willing then why call him God?
The modern logical problem of evil Mackie’s inconsistent triad
P1. An omnipotent God has the power to eliminate evil.
P2. An omnibenevolent God has the motivation to eliminate evil.
P3. Nothing can exist if there is a being with the power and motivation to eliminate it.
C1. Evil, omnipotence and omnibenevolence thus form an inconsistent triad such that God (as classically defined) and evil cannot possibly co-exist.
This is a deductive argument because there is no logical way for the premises to be true yet the conclusion false.
It’s also an a priori argument because it doesn’t reference experience. It is based on the logical analysis of the meaning of the words ‘omnipotent’, ‘omnibenevolent’ and ‘evil’.
The conclusion of the logical problem makes a large claim – that if evil exists, it is impossible that God exists.

The evidential problem of evil

The evidential problem was first put forward by Hume. Hume was clear that he doesn’t think the co-existence of God and evil is impossible. He accepts that it’s technically possible that a perfect God created an imperfect world for reasons consistent with God’s perfection.
Hume points out the a posteriori evidence of evil in the world:
1 – Animal suffering. Why shouldn’t nature be created such that animals feel less pain, or indeed no pain at all? (Roe’s example of the deer dying in a forest fire)
2 – Creatures have limited abilities to ensure their survival and happiness
3 – Why does nature have extremes which make survival and happiness more difficult? Natural evil
4 – Why doesn’t God intervene to prevent individual natural disasters?
P1. We are only justified in believing what the evidence suggests (empiricism).
P2. We only have evidence of imperfection (a world with both good and evil).
C1. We are only justified in believing that imperfection exists.
C2. So, belief in a perfectly powerful and good being is not justified.
So the evidential problem is a posteriori – it is based on our experience of the world and the evil in it. It is also an inductive argument. It claims that belief in God is unjustified on the basis of the evidence of imperfection we experience. Hume’s not claiming to deductively prove that God either doesn’t or cannot exist.
The conclusion of the evidential problem makes a more modest claim – that the existence of evil is sufficient evidence against God’s existence such that belief in God is not justified.

Augustine’s theodicy

Evil exists because we created it and deserve it.
“Evil is either sin or punishment for sin”.
Adam and Eve created original sin – a corruption in human nature giving us an irresistible temptation to sin (moral evil) – which was created by Adam and Eve’s sin against God.
They were also forced to live in a fallen world which is full of evil, as punishment. Adam and Eve chose to create us, and we are born with original sin and in a fallen world (natural evil).
So, evil exists because we created it and deserve it as punishment. So it’s not God’s fault.
In fact, it’s consistent with God’s divine justice to allow evil as punishment for our sinful ways.
Evil is ‘privatio boni’ – an absence of good.
Evil doesn’t actually exist, it has no what philosophers would call ‘positive’ existence. Augustine says it is like blindness – not a thing in itself, just the absence of sight.
Evil is merely the result of our falling away from God’s goodness.

Augustine vs the evidential problem (the scientific critique of the fall & original sin)

Scientific evidence against the fall – we evolved, Genesis cannot be literally true, genetic diversity shows we couldn’t have all come from two ancestors.
Augustine doesn’t understand reproduction – we weren’t all present in Adam’s loins, that’s unscientific nonsense.
So, the doctrine of original sin doesn’t seem to be true.
This casts doubt on Augustine’s attempt to avoid God’s responsibility for evil by blaming humanity for it.
Some, like G. K. Chesterton, attempt to defend the doctrine of original sin from this issue. He said original sin can be seen in ‘the street’, meaning it is empirically verifiable.
Perhaps the Genesis account of the fall isn’t literally true, but original sin still does seems accurate if you look around you at how terrible humans can be.
Augustine’s pear also illustrates this. He told a story how, as a child, he stole a pear, not because he was hungry but just for the fun of sinning. So even children must be born with a desire to sin = original sin.
However, this defence fails because modern sociological evidence shows that humanity has morally improved over time. Stephen Pinker studied the history of violence and showed it has declined in modern society compared to mediaeval and ancient times.
This proves that original sin cannot exist. If we really had an irresistible temptation to sin, we could not have morally improved, yet we have.
Evil actions are better explained by social conditioning, or what Pelagius called being ‘educated in evil’.
Freud could explain Augustine’s ‘pear’ story. Socialisation forces us to feel frustrated over repressing our instincts to the point where rebelling against social control can feel good in itself.
Pelagius travelled from Ireland to Augustine’s part of the world, and noted that people in Rome were much more sinful than he was used to, which proves this sociological point that it is culture and upbringing which causes sinful behaviour. Believing it is nature is really just an excuse.
If there is any selfishness in human nature, evolution is a much simpler and more scientific explanation of it.
Original sin is a false doctrine, so Augustine’s theodicy fails.

Augustine vs the logical problem (Original sin violates moral responsibility & incompatibility with omnibenevolence)

Pelagius argued that if we have original sin and are thus completely unable to avoid doing evil, it would surely be unjust for God to punish us for our sinful behaviour.
It’s not ethical for all humanity to be blamed for the actions of Adam and Eve.
This suggests an indefensible view of moral responsibility – that people can be responsible for actions committed by others which is of special absurdity in this case since the action occurred before they were even born.
Augustine responded that this might seem unfair, but puts it down to the “secret yet just judgement of God”. This means that God’s reasoning is inscrutable – impossible for us to understand – but we should have faith it is just. Augustine points to Psalm 25:10: ‘All the paths of the Lord are mercy and truth,’ and concludes: neither can his grace be unjust, nor his justice cruel”.
Further biblical evidence comes from the book of Job. God allowed terrible evil to befall Job. Job couldn’t see how such evil could be justified. However, God responded to Job by potining our that he has a limited human mind. We might not understand why evil exists, but that doesn’t mean God can’t have a reason for allowing it.
Nonetheless, Augustine would have to accept that a child who dies of cancer somehow deserved it.
This seems incompatible with a loving God.
Even if the child was born with original sin, it’s not it’s fault that it was. So it can’t be justice for God to allow it to suffer needlessly.
So, Augustine’s theodicy is not logically coherent and thus fails to solve the logical problem of evil.

Irenaeus & Hick’s theodicy

God allows evil because it serves the good purpose of soul-making (character development).
A good person is someone who chooses good over evil.
To get into heaven we’ve got to become good people.
So, God has to allow evil to exist to give us a chance to become good people (by choosing good over it).
Irenaeus points to the biblical example of Jonah who disobeyed God, so God sent a thunderstorm and a big fish to eat Jonah for three days (natural evil) – then the fish spat Jonah out, and he decided to obey God (he had become morally improved due to the evil).
Hick: epistemic distance – God has to ‘hide’ himself or not allow us to know for sure that God exists – because if we did know for sure that god existed, then we would just follow his commands out of obedience – but this is not the right moral motivation required for genuine character development – genuine development of personal virtue.

Irenaeus & Hick vs the evidential problem (The issues of purposeless and soul-breaking evil)

The strength of soul-making vs the evidential problem is that there is evidence that encountering and overcoming evil develops a person’s character and virtue.
This is behind the idea of character development in literature. It is also behind the idea that people become spoiled if they have too much luxury and not enough responsibility or difficulty to overcome.
By going through harsh struggles, a person becomes stronger and gains compassion for others. This does seem to be a factual occurrence in life.
Some evil is dysteleological (purposeless).
It has no chance of leading to spiritual development.
For example, a child who dies of cancer.
They are too young to even understand what is happening, let alone learn anything from it.
Most animal suffering is also dysteleological.
The holocaust is an example of evil which is dysteleological, soul-breaking and where the amount of evil outweighs our soul-making requirements. D Z Phillips criticised Hick, questioning whether anyone in their “right mind” could say the holocaust was justified because a few survivors were strengthened by it.
However, this criticism is unsuccessful because it misunderstands Hick’s theodicy, especially his concept of the epistemic distance.
This is the idea that if God made it obvious that he exists, we would just obey him out of obedience – but that would not develop our personal virtue.
If every case of evil we experienced was perfectly aligned with the soul-making requirements of those who suffered from it, we would know there was a God controlling the process – which would break the epistemic distance and ruin our soul-making opportunities.
So, to really enable soul-making, God must create a world where evil appears random.
So Hick isn’t saying the holocaust or every case of evil serves a good purpose.
He’s saying that living in a world full of evil is necessary for soul-making to be possible.
So God puts us in a world with random evil and this gives us the opportunity to grow from it. But the evil is not all carefully designed for our growth.

Irenaeus & Hick vs the logical problem

A strength of soul-making theodicy is its premise that creating us fully developed was logically impossible.
A fully developed soul is one which has chosen good over evil. This requires having made a choice. Therefore, it’s logically impossible for God to create us fully developed.
Most theologians agree omnipotence does not include the power to do the logically impossible. So, a perfect God would create us undeveloped and allow us the freedom to choose good over evil. Evil is needed because it serves this good purpose of soul-making. So, evil isn’t incompatible with God’s existence. Mackie’s logical problem seems defeated.
Dostoyevsky’s character ‘Ivan’ makes a critique of attempts to justify evil
The key detail of Ivan’s argument is his connection between the suffering of innocent children and the gain of heaven for others.
People get into heaven because of, on the back of, the suffering of innocent children.
Ivan says no good person or God would design this method of getting into heaven.
It’s not that the evil is dysteleological, nor that the process of soul-making is not worth it. It’s that the whole process of soul-making is actually not morally acceptable.
If the suffering of a child was the cost of the of the soul-making of others, Ivan’s point is that this is indecent. It’s not moral. Building heaven on a foundation of children suffering is not what Hick’s supposed ‘God of love’ would accept. So, Hick fails to solve the logical problem of evil.
This critique from Ivan is successful because it gets around Hick’s standard defenses of himself. Hick doesn’t say that every case of evil has a soul-making benefit, but that the possibility of soul-making requires a world in which evil, even purposeless evil, is possible. Ivan’s point is that this is not a morally acceptable system and that his own moral virtue compels him to reject it.
Ivan’s discomfort is logical. It doesn’t seem right to accept heaven for himself if the price is the suffering of innocent children.

The problem of evil & the issue of free will (this paragraph can be used in any question)

All theodicies have a similar strength in that they try to link God allowing evil to free will in some way.

God cannot remove evil without removing:

Our deserved punishment (Augustine)
Our free will, since moral evil results from our misuse of free will, and natural evil results from the free will of adam and Eve and the devil – a lack of free will would make our lives pointless (Plantinga & Augustine)
Our opportunities for soul-making as a result from freely choosing good over evil
So, removing evil is actually not something a perfect God would do – because the knock on effects regarding free will would actually leave us even worse off.
Mackie responds that a world where free creatures always choose good is logically possible, so a perfect God would have created it.
However, Plantinga counters that just because such a world is logically possible, it doesn’t follow that God could create it.
God cannot create a world where free creatures only choose good without forcing them to choose good – which undermines their having made a choice.
God can only do logically possible things – it’s not logically possible to ‘make’ people ‘choose’ good – since that would undermine their free will.
Mackie counters by arguing against Plantinga’s understanding of free will.
Mackie argues that the idea of ‘free will’ can only really make sense on a ‘compatibilist’ view of free will.
Plantinga, like most religious philosophers, believes in ‘libertarian free will’ – the ability to have done otherwise than what we did.
Mackie rejects this definition as incoherent.
Mackie argues that our choices must have a cause, and that cause could only be either randomness, external causes or our own character.
We only talk about being ‘responsible’ for our choices that come from our character.
So, the only valid definition of free will is the compatibilist one – we are free when our actions are determined by our character.
Some of our actions are caused by our desires/wants and Mackie thinks we should call those actions ‘free’, even though they were predetermined.
We did not create our character – that was predetermined.
So, a perfect God would have created us to all have good characters and thus all be determined to make good choices.
So, God does not exist.

The logical problem of evil as a version of the problem of evil (Mackie)

Mackie argues that an all-powerful God would be able to stop evil and an all-loving God would be motivated to stop evil.
Logically, something can’t exist if there’s a being with the power and motivation to eliminate it.
So, if evil exists, then God cannot exist.
Evil, omnipotence & omnibenevolence form an inconsistent triad – meaning they cannot all exist.

The evidential problem of evil as a version of the problem of evil (Hume or Rowe)

Hume puts forward an evidential problem of evil.
Hume thinks it is actually logically possible for God and evil to exist.
However, evil is evidence against God’s existence.
The evidence of evil that we see in the world means belief in God is not justified.
There could be a God technically, but the evidence is against it.

Augustine’s theodicy (response to the problem of evil)

Augustine says that God allows evil because we deserve it.
He says “evil is either sin or punishment for sin”.
The ‘fall’ of humanity due to Adam & Eve disobeying God caused original sin – a corruption in human nature giving us an irresistible temptation to sin. This explains moral evil – the evil humans do to each other.
As punishment, God banished us from the garden of eden into this fallen world, which explains natural evil – evil resulting from the workings of the natural world e.g. disease and natural disasters.

Augustine vs the logical problem

Original sin violates moral responsibility & incompatibility with omnibenevolence.
Criticism of Augustine: how can it be fair for us to be punished for the actions of Adam and Eve?
Their disobedience was not our fault, so it cannot be loving for God to punish us for that.
Augustine responds: we aren’t punished for their actions, we are punished because we are sinful beings – because we are born with original sin!

Augustine vs the evidential problem

Criticism of Augustine: there is lots of scientific evidence against the fall – evolution suggests we evolved.
The idea that sin is inherited is unscientific nonsense.
Maybe the fall story didn’t happen – but there is still good evidence for original sin – look at how terrible humans are e.g. Nazis.

Irenaeus & Hick’s theodicy (response to the problem of evil)

God allows evil because it serves the good purpose of soul-making – character development.
To become good people and deserve heaven, we must choose good over evil.
In that case, we need evil in order to become good.
E.g. if you see someone suffering from disease, you might become a better person through being more compassionate.
So, evil is required for us to be able to develop into good people who deserve to go to heaven.

Irenaeus & Hick vs the logical problem

Criticism of Irenaeus/Hick: Why didn’t God just create us good to begin with..?
This whole process of soul-making is unnecessary and evil.

Evaluation of Irenaeus & Hick vs the logical problem

However – Irenaeus/Hick argue God can’t make us good.
A good person is one who has freely chosen good over evil. If God makes us good – then we aren’t really choosing good ourselves, and then we wouldn’t really be good.
So, Soul-making is necessary.

Irenaeus & Hick vs the evidential problem

Criticism of Irenaeus/Hick: there is lots of evil that does not help soul-making.
E.g. a child who dies of cancer – they were too young to understand what was happening, there’s no way that evil helped them become a better person – in fact it prevented that.
Some evil is soul-breaking, causing people to become depressed etc – it’s not soul-making.

Evaluation of Irenaeus & Hick vs the evidential problem

Maybe other people could have learned lessons from the child dying.
E.g. their parents.

Reason and Meaning

Philosophical reflections on life, death, and the meaning of life, summary of “the problem of evil”.

The “ problem of evil ” is thought to be one of the most difficult for theists. Why? Put simply the existence of bad or evil things isn’t hard to explain for non-theists—human beings and the world are imperfect—but they are hard to explain for classical theists.

The Problem – The gods are all-good, powerful, and knowing and yet there is evil. Thus either the gods can’t do away with evil—in which case they’re not all-powerful, or they won’t do away with evil—in which case they’re not all good. We can distinguish between:

a) The logical problem of evil – gods and evil are incompatible or inconsistent; and b)The evidentiary problem of evil – evil counts as evidence against the gods.

Response to the problem – Theists have articulated defenses , but generally can’t advance theodicies (complete explanations for evil.) A defense is easy, you just need to show that it is rational to believe in gods and evil simultaneously. A theodicy is hard, it must show how evil fits into a god’s plan. Most theologians think that the best we can do is to show that evil and the gods are compatible, but they don’t believe they can completely explain evil. In order to defend the rationality of religious belief—to offer a strong defense—philosophers/theologians try to provide reasons for the existence of evil. These include:

1. The ideas that pain/evil is necessary as part of the body’s warning system

PROBLEMS – Sometimes we need warnings but there is no pain (carbon monoxide, obesity, etc.); sometimes the pain doesn’t help us (cancer, etc.); sometimes pain may be debilitating. Furthermore, why would gods create pain? What explains such cruelty?

2. T he idea that evil is necessary so that we may better appreciate the good – ( Logically this implies that we would have no notion of bad without good, or tall without short. Psychologically this implies that we wouldn’t appreciate good things with bad things, pleasure without pain, and happiness without unhappiness. )

PROBLEMS – Even if this is true, why do we need so much evil? We have cancer and heart disease, so do we really need Alzheimer’s and Parkinson’s and Huntington’s and ebola too? And do you really need to know there are bad things to enjoy good things? (If you believe in heaven or paradise where supposedly you are eternally happy, would you need occasional pain there to appreciate its goodness?)

3. The idea that evil is punishment from wrongdoing; we bring it on ourselves

PROBLEMS – This makes sense only if moral character and suffering correlate. But misfortune/evil strikes indiscriminately, as does good fortune. Moreover, do babies deserve misfortune? Do we deserve horrible diseases? Can one ever do enough bad things to deserve say, everlasting punishment?

4. The idea that evil results from free will – Evil results from free will. A world with humans and the evil that results from their free will is better than one without humans even if that world had no evil. War, murder, torture, etc. are worth the price of the positives that derive from human free will.

PROBLEMS – We can answer that free will is not worth all the misery that ensues from free choice. In addition, we might also wonder why an omnipotent God couldn’t create humans with the freedom to do bad things, but who never do them. Moreover, free will, if it even exists, only accounts for moral evil (evils attributed to free will like murder, rape, etc.) but not physical evil (earthquakes, floods, disease, etc) which have nothing to do with free will. Finally, if evil results from freedom and there is no evil in heaven, then we aren’t free in heaven.

5. The idea that evil is necessary for the development of moral character . In a world without “trials and tribulations” we wouldn’t get to develop our moral characters or make our souls. Such a world wouldn’t elicit generosity, courage, kindness, mercy, perseverance, creativity, etc.

If the moral character development argument is combined with the free will defense then we have given the best account of evil possible. This is not a theodicy—a complete explanation—but a defense—a partial explanation. We could even add that since there is another world evil here is no big deal anyway. That is, all this pain will be insignificant when we all enjoy eternal bliss. Of course, even if we can overcome the problem of evil that doesn’t mean the theistic story is true.

PROBLEMS –At least three basic problems remain in our attempt to reconcile evil and all good, all-knowing and all-powerful gods.

1) Why don’t the gods intervene to prevent extreme cruelty—such as the abuse of an innocent child? The free will defense is implausible here.

2) Why is there so much human suffering? Do we really need all these hurricanes and diseases? Do we really need to develop our characters by, for example, seeing children die or suffering from cancer? And even if we need to occasionally die in childbirth or from cancer, couldn’t we have fewer cases of this evil?

3) Why do non-human animals suffer so much? They don’t have freedom or need to develop their moral characters, yet they suffer. (Darwin was moved by this argument.) If you look at the entire world, and the entire history of the world, does the evidence suggest that it is the product of all good, all-powerful, deities? Or does the evidence suggest the opposite? At the very least, doesn’t evil provide evidence against the existence of such gods? Of course, it does.

(This entry relied heavily on James and Stuart Rachels’ book: Problems from Philosophy .)

Bibliography

Academic tools.

Friends PDF Preview
Author and Citation Info
Back to Top

Kinds and Origins of Evil

Unde malum? What is evil—if it is anything at all—and whence does it arise? Is evil just badness by another name? Is it the inevitable “shadow side” of the good? Or is it more substantial: an active, striving force that is opposed to the good in a Star Wars, Manichean kind of way?

Does evil always originate in the causal powers of nature? Is it sometimes based in the choices of moral agents? Or, perhaps most disturbingly, does evil sometimes have its source in something non-human and impersonal—a malevolent tendency in the universe not just to general winding-down but also to outbreaks of targeted hellishness?

Finally, what is radical evil , and how does it differ from other kinds of evil?

These are some of the key metaphysical questions that philosophers have raised concerning evil. The goal of this entry is to provide a taxonomy of the most prominent answers: the main theories of evil’s kinds and origins on offer in the western philosophical tradition. This is meant to supplement the discussion in the entry on the concept of evil , although Section 1 begins with some conceptual and semantic issues.

Section 2 introduces two key distinctions that are then further developed in Sections 3 and 4. The first distinction has to do with the kinds of evil: insofar as evil is anything at all, is it a deep metaphysical feature of things, or is it always (or at least sometimes) merely an empirical phenomenon? The second distinction has to do with the origins of evil: is evil ever (or always) based in entirely natural phenomena, or does it sometimes (or always) have a moral or supernatural origin?

Sections 5 through 7 consider some puzzling cases of hard-to-categorize evils: systemic evil; symbolic evil; and so-called “radical” evil.

1. Speaking of Evil

2. two distinctions in evil: kinds and origins, 3.1.1 absence theory, 3.1.2 matter theory, 3.1.3 privation theory, 3.1.4 real property theory, 3.2 empirical evil, 4.1 moral evil, 4.2 natural evil, 5. systemic evil, 6. symbolic evil, 7.1 radical evil as violation of the moral law, 7.2. radical evil as choosing evil for its own sake, 7.3. radical evil as repudiation of the moral law, 7.4. radical evil as systemic evil: human beings made superfluous in another way, other internet resources, related entries.

“Evil” and related terms in the Germanic branch of Indo-European have referred, at various points, to suffering and wrongdoing, but also to defecation, latrines, spoiled fruit, diseases, prostitution, and (oddly enough) forks.

The Greek term “ kakos ” may be related to the Proto-Indo-European term “ kakka ”—“defecation”. But only the first two meanings survive in English, and non-ironic uses of the term are relatively rare outside of ceremonial and literary contexts. Indeed, speaking of evil nowadays often feels like an exercise in anachronism—like speaking of wickedness, abomination, uncleanness, and iniquity.

The Oxford English Dictionary explains:

In modern colloquial English it [ evil ]; is little used, such currency as it has being due to literary influence. In quite familiar speech the adjective is commonly superseded by bad ; the noun is somewhat more frequent, but chiefly in the widest senses, the more specific senses being expressed by other words, such as harm , injury , misfortune , disease , etc. (“evil, adj. and n. 1”, under A., abbreviations expanded, OED Online , accessed September 2021)

This trend is found in other modern languages, but not in all. Ruppel (2019) notes that in German-speaking lands “ das Übel ” declined just like “evil” did in England, but was soon replaced by “ das Böse ”, which is still alive and well in Germany.

This slow erasure of “evil” and its cognates from many European languages, which began in the seventeenth century, was due to the rejection of the concept of evil, especially by elites. Doctors, moral philosophers, natural scientists, and even theologians shied away from evil —preferring more tractable notions like badness , harm , and misfortune , or quasi-quantifiable concepts like pain , suffering , trauma , and disutility . Traditional views of ontologically substantive and supernatural evil—something able to possess a body or terrorize a soul—came to be seen as quaint, unscientific, embarrassing (Ruppel 2019).

Philosophers of religion are a half-exception to the rule. They did and do continue to speak of evil, at least when discussing the “problem” thereof. (See the entry on the problem of evil .) If pressed, though, they typically admit that this is because the great framers of the problem—Augustine, Aquinas, Leibniz, Bayle—used the term (in Latin or French), and then proceed to gloss it generically as, in Michael Tooley’s words, “any undesirable states of affairs” (2002 [2019]). Philosophers of religion in the broadly Continental tradition are less likely to assimilate evil to more general or anodyne notions in this way, and more likely to discuss the nature of evil as opposed to the “problem” it raises for theism (e.g., Kearney 2001 and Matuštík 2008).

Despite this widespread squeamishness about “evil” in both scientific culture and common parlance, there are moments when the pull of the ancient lexicon is irresistible—at the very least expressively, in the mode of both condemnation and lament. Premeditated mass shootings aren’t just bad or traumatic; rather, they are something else: here people still reach for “evil” or even “radical evil”. The years-long imprisonment and rape of children by their parents is a misfortune that produces negative utility, to be sure, but the transfixing horror of it seems only to be captured by the invocation of “evil”. The same is true of most instances of genocide, sex-trafficking, torture-slaying, terrorism, serial killing, and slavery: these are one and all bad, harmful, and traumatic activities, but they are also something else—something excessive, mesmerizing, and revolting all at once (see Stone 2009 for a psychologist’s account). In the face of such acts, we—along with our spiritual leaders, newscasters, and politicians—are still willing to speak, preach, and tweet about “pure evil”.

Thus after a school shooting in February 2017, Donald Trump (@realDonaldTrump) tweeted that “we must keep ‘evil’ out of our country”. (Despite the quotation marks, it was clear that he meant evil the entity, and not “evil” the word.) After the Las Vegas mass shooting in October 2017, Trump and many others in leadership referred to the event as “an act of pure evil” (Matuson 2017). Less recently, George W. Bush referred to Iran, Iraq, and North Korea as the “Axis of Evil” in a state of the union address on 29 January 2002 (see Other Internet Resources), and Ronald Reagan repeatedly characterized the Soviet Union as “the evil empire”, famously at a speech on 8 March 1983 to the National Association of Evangelicals (see Other Internet Resources).

But when we do this—when we speak of evil , das Böse , il male nowadays—what is it that we are referring to, and where does it come from ?

Pressed with such questions, many people (philosophers included) revert to the more tractable terms. Of course what we are really talking about (whispering about, thundering about, shaking our heads about) in those moments of condemnation and lament is an extreme instance of suffering or disutility. Of course “evil” is to “bad” what “wicked” is to “immoral”: a conceptual vestige of a pre-scientific, credulous past that we invoke for the sake of solemnity, empathy, or emphasis. A concept that—outside of horror films and fiction—is best analyzed in terms of nature’s frustration of the basic needs of sentient creatures, or as the effects of illness and ill-parenting. Yes, evil happenings have an excessive, egregious quality that makes them notable, even transfixing. But they are not, in the end, sui generis or metaphysically mysterious: neuroscience, medicine, psychology, and law have domesticated evil. Taken to its logical extreme, the doctrine that characterizes this camp would be that all evil is “natural” (a product of various causal processes in nature).

Others prefer to answer the questions about the origins of evil in terms of choice, agency , and will . For people in this camp, evil consists in malevolent intentions, malice with forethought, and self-conscious cruelty that leads to extreme suffering and tribulation. They may allow that there are contributing factors and preconditions, but ultimately hold the agents themselves responsible for evil. Note, however, that the appeal to human free will can also be seen as an effort to domesticate evil—to make it explicable in terms of familiar concepts, to set it on a continuum with other, familiar acts and events. Taken to its logical extreme, the doctrine held by people in this camp is that all evil has “moral” origins—it is a product of choice or agency of some sort.

This debate about the roots of evil plays out not only in philosophy seminar rooms and psychology labs, but also on cable news stations and op-ed pages. People in the second camp tend to the political right, and sometimes even make a show of using “evil” because they think that people in the first camp (who tend to the political left) are uncomfortable with the idea of personal responsibility and blameworthiness.

I said these were the two opposing camps. In truth there is another one—one that used to be very popular but now seems sparsely populated, at least among philosophers. People in this third camp eschew efforts to domesticate evil; for them, what we mean by “evil” is not equivalent to what we mean by “bad” or “wrong” or even “very very very bad” or “very very very wrong”. In other words, evil is not just illness, misfortune, or malevolent choices by another name but rather a positive, substantial rottenness in the universe. It is, or has its origin in, some non-agential force or shadow side of reality—something spooky, imperceptible, but out there (“in them woods”).

The late antique (Plotinus, Proclus, Augustine, Boethius), medieval (Anselm, Ibn Sina, Aquinas), and early modern (Descartes, Leibniz, Bayle, Kant) eras contain sophisticated traditions of reflection on questions about evil—about its being or non-being, its intrinsic features and natural manifestations, and its origins in nature, will, or supernature. Over the course of that centuries-long discussion, two main distinctions emerged.

The first main distinction has to do with the nature or kinds of evil: is evil at bottom just an empirical phenomenon—something that is given in the causal, phenomenal world of our experience? Or is there a deeper, metaphysical aspect to some evils? Note that this is not an exclusive distinction: people who endorse the idea of metaphysical evil typically assume that it also has an empirical character or manifestation.

Suppose, for example, we come across the sort of scene that drove Friedrich Nietzsche mad in Turin: a coachman mercilessly beating his horse (Prideaux 2018). In this version of the case, however, suppose that the coachman’s cruelty is a response to his having been recently diagnosed with terminal cancer. So here there is certainly some empirical evil: the cancerous disruption to the body, the cruelty of the man, the pain of the horse. Some philosophers will say that there is also metaphysical evil: neither the man nor the horse is metaphysically perfect, and so on the Absence Theory considered below ( section 3.1.1 ) both are in that respect evil. We might also regard the body and character of the man as corrupted and conclude that he is metaphysically evil in a “privative” way too (see section 3.1.3 ). Metaphysical evils like these are distinct from the empirical evils of the cancer, the cruelty, and the suffering of the horse, though on many accounts they are the ground of the latter.

The second main distinction has to do with the origins of evil, and tracks the differences between the three camps mentioned above. The first option here is to say that a specific evil arises entirely from natural phenomena for which no one is responsible. In the case of the man and his horse, it is common to think that their metaphysical finitude and incapacities, as well as the cancerous tumor and the canine pain, are “natural” in this way: they seem to be based in facts about the natures, events, and causal laws involved.

An alternative is to say that a specific evil has its origin in moral actions and intentions. Applied to our case, it is common to think that the man’s agency—the choice to beat his vulnerable steed—is the origin of the cruelty and the pain. If there is metaphysical evil here, then it too might have a moral origin: on some religious pictures, for instance, the corrupted human nature that leads to disease, cruelty, and enmity between him and other creatures is a result of free choice on the part of his primordial ancestors. (If there is agency in any non-human creatures—animals or angelic—then it would also fit here.)

The third (and now-quite-unpopular) view about origins says that a specific evil arises ultimately not from nature or from choice but from something that is both supernatural and non-agential (call this a “spooky non-agential” origin). On such views there is a dark force or side of reality that is the ultimate origin of, say, the metaphysical evil in the man’s nature. It may also be the ultimate source of the empirical evil involved.

Three further preliminary notes:

The second main distinction here is often regarded as exclusive with respect to a specific evil, since we are asking about its ultimate origin. If the man’s cancer and ill-treatment of the horse originate entirely in the causal powers of the physical universe, then they are not also based in either free choice or supernatural spookiness (and vice versa). “Typically” here is key, however, since a compatibilist picture of free will (O’Connor & Franklin 2018 [2021]) says that free choices themselves are determined by natural causes. On that view, perhaps, the origin of the man’s cruel act is both moral and natural. Compatibilisms between natural and spooky origins are also conceptually possible.
Although some theorists think that all instances of evil (whether metaphysical or empirical) are grounded in just one of these ultimate origins, most will allow that different evils have different ultimate origins. For instance, someone might coherently think that the man’s cancer has a natural origin, that his cruelty has a moral origin, and yet that the ferociousness of the beating has a spooky or “dark force” origin. Section 4 considers some historical efforts to suggest that all evils are ultimately moral in origin.
Although these conceptual distinctions are fairly clear, there is terminological variation in the historical and contemporary literature with respect to the term “natural evil”. Some philosophers and theologians use “natural evil” to refer not to an origin but to a kind—the kind that is here called “empirical evil”. When the distinctions are maintained, however, it should be clear that they are orthogonal: both metaphysical evil and empirical evil can have natural, moral, or spooky origins (see Figure 1 ).

Figure 1. The Table of Evils, applied to Nietzsche case

The distinctions represented in the Table of Evil are the topics of Sections 3 and 4. Sections 5 through 7 look at three varieties of evil (systemic, symbolic, and radical) whose positions within the Table are more difficult to discern.

3. Kinds of Evil: Metaphysical and Empirical

The first key distinction is concerned with the kinds of evil—with what evil is or consists in, and thus with where and how it manifests. Again, the distinction is not exclusive: someone might hold that there is both metaphysical evil and empirical evil, and that the latter is typically a manifestation of metaphysical evil. Someone else, however, might hold that there is no such thing as metaphysical evil, and that all evils can be accounted for at the empirical, causal level.

3.1 Metaphysical Evil

Many of the traditional kakologists believed in metaphysical evil—i.e., evil that has to do with the way things exist or fail to exist. Typically, metaphysical evil is supposed to be a function of a thing’s nature and characterized by a kind of unintelligiblity. As we have seen, many such theorists also typically assume that metaphysical evil has empirical manifestations.

Both metaphysical and empirical evil have been described in terms of four main theory-templates: Absence, Matter, Privation , and Real Property . These templates are laid out in more detail in the discussion of metaphysical evil in this section, and applied again in section 3.2 ’s discussion of empirical evil.

The Absence Theory of Evil has its origins in the Platonic idea that there are different “degrees of being” corresponding to the number and kinds of capacities a thing has. Roughly speaking, the more numerous and impressive a thing’s capacities, the more real and thus better it is, metaphysically-speaking. A dog cannot stand erect; an ape can. A rock cannot pass through walls; an angel can. These lacks or absences are essential to being the kind of finite creature that a dog or a rock is; miracles aside, a rock is just the kind of thing that cannot pass through walls. All the same, the lack of that ability is an evil.

Absence theorists typically add a “plenum” thesis here: it is fitting, or beautiful, or perhaps even necessary, that all the different degrees of being are exemplified, and thus that every link in the “great chain of being” is occupied. Evil is a function of the way things ought to be or even must be.

Absence Theory was popular across antiquity, but it was particularly attractive to philosophers in monotheistic traditions because it allowed them to say that evil is not a thing —and thus not some thing that a good, all-powerful God created or sustains. Anselm writes in On the Fall of the Devil :

Just as nothing that is not good comes from the Supreme Good, and every good is from the Supreme Good, likewise nothing that is not being [ essentia ] comes from the Supreme Being [ essentia ], and all being is from the Supreme Being. Since the Supreme Good is the Supreme Being, it follows that every being is a good thing and every good thing is a being. Therefore, just as nothing and non-being [ non esse ] are not being [ essentia ], likewise they are not good. So, nothing and non-being are not from He from whom nothing is unless it is good and being. ( De Casu Diaboli , v.1, 235; translation by Sadler [ Other Internet Resources ])

A lingering problem in the theistic context, however, is that Absence Theory entails that finite things have a degree of evil just in virtue of not being at the top of the chain. So even if God does not create evil (because absence is uncreated), God creates beings that are essentially evil.

Plotinus rejects Absence Theory for a related reason: it compels us

to say that there are evils in the higher world too; for there the Soul is inferior to Intellect, and Intellect is lesser than [the One]. ( Enneads I, 8, 13)

For Plotinus, it is not merely being lower than the highest One that makes something evil; rather, evil consists in being so low as to be associated with matter. Indeterminate or “unformed matter” is the final term on the cosmic chain from being to non-being—it is as far away from intelligence as possible, and thus equally far away from goodness. This is the Matter Theory of Evil that was influential in various gnosticisms, early Christian heresies, and late antique platonisms (see O’Meara 2019).

Augustine, as well as many scholastics and early moderns, rejected Matter Theory on religious grounds. God created matter, and so it cannot be bad in itself. Instead of reverting to pure Absence Theory, however, these monotheists developed the Privation Theory of Evil . This can be construed as Absence Theory plus the Aristotelian idea that goodness is relative to a thing’s kind . Individuals of different kinds have different ends dictated by their natures: as long as the dog achieves the ends set out by its nature, it counts as fully good of its canine kind, even if it essentially lacks the good-making capacities (or “realities”) of beings higher on the chain. (For more on Augustine’s version, see King 2019; on Aquinas, see Davies 2019).

Privation Theory thus reduces pressure on monotheism: evil is not a being but rather an absence, and so God did not create it. Moreover, evil is the kind of absence that is not a function of the essential natures of things, and so God cannot be faulted for creating things that are essentially evil. Failing to accomplish the end set out by one’s nature—failing to be the way one ought to be— is a privation, however, and so it is evil.

Friends of Privation Theory offer different accounts of how and why things fail to accomplish their natural end. Most ascribe the failure to something in the individual creature—culpable ignorance, Original Sin, free agency—rather than to God, thereby grounding metaphysical evil in moral evil (on which see section 4.1 below). The extent to which these appeals to privation succeed in getting God off the hook is, naturally, controversial.

Plotinus’s view is effectively a hybrid of Privation Theory and Matter Theory: for him, privation only occurs when some of the matter in something is unformed or unmastered by a form (see Enneads I, 8, 5, 19–26). But even if other Privation Theorists do not view privation in this way, the conception of evil as infinitely distant from reason and intelligibility survives. In other words, Ignorance, Original Sin, and other misuses of freedom are not necessarily a function of irrational matter , but they are still opposed to rational mind . Descartes is illustrative here: for him, the only teleology (and privation) in the world relates to the souls—and in particular the wills—of human beings. Material substance is mechanistic, and aggregates of it (in the forms of animals and plants) can be explained mechanistically, without invoking teleology or kind-relative values (compare Newlands 2019).

A fourth major account of the nature of metaphysical evil takes it to be something more substantive than absence, privation, or unformed matter. Call this the Real Property Theory of Evil : evil is some sort of reality —a determinate feature of certain finite beings (see J. Russell 1977, 1981, and Frankfurter 2006). Some versions of this picture say that evil is ultimately dependent on the good. Other, more Manichaean versions take the two to be coeval and independent, locked in an eternal axiological struggle. Although the relevant “property” here is supposed to have more reality or substance than mere absence or privation, there is still a kind of unintelligibility to it. The evil side of reality, the dark force, or the malevolent will is a kind of black hole for the “natural light” of reason—positively real, but inaccessible to complete explanation.

Empirical evil is a capacious category: it covers bodily pain, damage, and disease as well as the psychological concomitants or effects of these physical phenomena—suffering, terror, depression, mental illness. Traditionally, it has also included social ills such as oppression, poverty, and structural injustice (see Sharpe 1909). Philosophers who believe in metaphysical evil often take them to manifest in empirical evils. Philosophers who reject metaphysical evil, by contrast, take the empirical kind to be fundamental.

Calling pain an evil might raise eyebrows, since clearly some pain is beneficial: it protects us from collisions, diseases, and predators. But the idea is that, whatever instrumental uses it has, pain qua phenomenal quality —the feeling of it—is intrinsically bad (setting aside for these purposes tricky cases like that of the masochist who takes pleasure in experiencing pain). Pain in the horrendous amounts and kinds that we encounter in human history as well as in the “charnel house” of evolutionary history is sickeningly and obviously bad (Murray 2008; Martin & Watkins 2019).

The templates used to characterize metaphysical evil above can be applied to empirical evil, too. An Absence Theory of empirical evil construes it simply as the absence of physical-psychological states of pleasure, health, stability, justice, and even life. Matter Theory regards pain, disease, mental malaise, and social ills as effects of our standing as material beings, vulnerable to the “matter” in our organism breaking down or coming into conflict with other parts of material creation. Privation Theory says that empirical evil is the absence of some such good which ought to exist. There will then be different accounts of why the good ought to exist, and why it doesn’t—some of these will be based in a theory of metaphysical evil.

A Real Property Theory of empirical evil, by contrast, insists that pain and suffering are positive realities and not mere absences (it is thus compatible with Matter Theory, but not with Absence or Privation Theory). As Malebranche notes, this is problematic in theological contexts, since God is supposed to be the ground of all positive beings (see Malebranche 1674–5 [1997: 348 and 392]). In a letter to a mathematician named Arnold Eckhard, G.W. Leibniz discusses this issue, and suggests “with some scruples” that “pain too is a perfection” (1677 [1969: 177]). That is a fairly unusual view, one that is made particularly poignant by the fact that Leibniz himself spent the last few years of his life in immense pain. He apparently had a habit of sleeping in a chair near his writing desk, and it

led to his having an open sore on his right leg. This caused him difficulty in walking; he tried to remedy it, but only by putting blotting paper on it. Later, to reduce the pain and to make the nerves insensitive he had a number of wooden clamps made, and these he screwed onto himself wherever he felt pain. I suspect that by doing this he so damaged his nerves that eventually he could no longer use his feet and had to stay in bed. (Guhrauer 1842 [1966: vol 2, p.336]; quoted in Mates 1989: 29)

Although such pain is a “perfection” and thus a real property, for Leibniz, it still involves a kind of weakness or imperfection in the person who has it, and so God cannot, in the end, exemplify this “perfection”. Rather, God’s possession of maximal pleasure is somehow sufficient to ground the “reality” that is found in both pleasure and pain (Leibniz 1677 [1969: 177]). This seems fishy, and other philosophers argue against Leibniz that if pain is a real property (rather than a mere absence), then God as the “ground of all reality” must indeed exemplify it. Some recent theologians and philosophers even welcome this idea, arguing that a conception of God as grounding empirical evil by suffering it has advantages for projects in theodicy (Hartshorne 1984; Wolterstorff 1988).

Mental illnesses can be classified as “empirical evils” if we assume that they have empirical causes and neural bases (Bhattacharjee 2018). Such illness often generates further empirical evil in the form of inexplicably cruel or destructive behavior to self and others, as will be familiar to anyone who has tried to live (or love someone) with mental illnesses like posttraumatic stress disorder (PTSD), borderline personality disorder, schizophrenia, and so on. It is unclear whether the theologians just mentioned would want to say that this kind of empirical evil, too, must have its ground in the divine by way of exemplification. The theological consequences of such a suggestion seem unattractive, to say the least.

4. Origins of Evil: Moral and Natural

Moral evil is metaphysical or empirical evil that arises out of the acts or intentions of agents: other traditional terms for it include “sin”, “wickedness”, “trespass”, and “iniquity”.

Philosophers in the Abrahamic tradition typically hold that we can be damaged by moral evil even before we have performed any actions whatsoever. Augustine, for example, characterizes Original Sin as a result of a primordial choice that damaged our very nature such that each member of the species is born already worthy of infinite punishment and strongly inclined to engage in further moral evil. So this is a case of metaphysical evil that has its origin in moral evil: bad choices on the part of the forefather led to damaged natures all the way down the spermatic line.

Such metaphysical evil will often manifest in empirical evil (although it need not do so); it also makes it likely that there will be further metaphysical evil. Thus an initial moral choice starts off a kind of snowballing into hellishness. It is clear why ancient and medieval and even some contemporary responses to the “ Unde Malum? ” question focus on how a morally uncorrupted creature (Adam, Eve, or Lucifer) could have started the process in the first place (see, e.g., Anselm De Casu Diaboli , vol.1 and Johnston forthcoming).

Kant appropriates this idea in a modern context, arguing that we are all originally afflicted by a propensity to moral evil ( Hang zum Bösen ), even apart from any specific actions that we perform. This “radical evil”—the metaphysical evil at the root of our nature (“ radix ” = Latin for “root”)—is something for which not Adam but we as individual agents (or perhaps the species itself ) are somehow responsible. And this is reflected even in garden-variety peccadillos: any act that subordinates commitment to the moral law to something else is an expression of radical evil. This makes it clear that people who use “radical evil” to refer to something particularly horrendous or awful (Arendt 1951, Bernstein 2002) are departing from the Kantian concept (see Section 7 below). That said, whether it is intelligible to say that we are culpable for a status that is not temporal is a notoriously open question in Kant-interpretation (see Card 2010, Wood 2010, and other essays in Anderson-Gold & Muchnik 2010).

Although moral evildoing often leads to empirical evil, it needn’t do so, at least on non-utilitarian accounts. Some crimes can be “victimless” at the empirical level.

Once again, the four traditional theoretical templates can be used to further characterize the evil acts and intentions that make an evil “moral”. Absence Theory says that the absence of a good will—either because something lacks a will altogether, or because its will is not good—is evil just by way of being an absence. Privation Theory says that a will is evil when it ought not have the absence of good orientation that it does. Matter Theory says that our embodiment and other engagements with matter explain the misorientation of our will. Real Property Theory insists that a will that is oriented to the bad is a real thing, and that its orientation to the bad is a real feature of it, not just a privation.

Declaring that moral evil is rooted in the will does not fully explain it. We would need a further account of how evil choices are made—of what sort of moral psychology could explain them. For Kant, explanation ceases at some point; evil choices, at bottom, are irrational—surds that we can identify and impute, but never fully explain. “There is no conceivable ground for us, therefore, from which moral evil could first have come in us” (Kant 1793 [AK 6:43; 1998: 64]).

Other philosophers offer partial explanations. Augustine says that although there is no determining cause of the devil’s choice, it is able to be partly “rationalized” in terms of Satan’s self-obsessed delight in his own powers (see King 2019). This is an early version of what is now called a “dispositional” account of evil agency (L. Russell 2014: ch.10; Kamtekar 2019). Early Islamic interpreters of the Qu’ran, by contrast, offer explanations in terms of ignorance: there is something that the supremely evil agent (Iblis) didn’t realize, or an inference that he failed to make, and this is what explains his orientation to the bad (see Germann 2019).

Moral evil is what contemporary people—including philosophers—tend to have in mind when they talk of “evil”. Extreme forms of it are viewed by many philosophers view as “unintelligible”—as defying ordinary explanation in significant and threatening ways. Such extreme and unintelligible moral evils are what many philosophers (though not Kant) are referring to when they speak of “radical” evil (Arendt 1951; L. Russell 2014; see also Section 7 ).

Natural evil refers to metaphysical and/or empirical evils whose origins are “natural”—i.e., grounded in the natures of things and/or the natural laws. The very nature of a horse makes it incapable of language: if that incapacity is a metaphysical evil (as it would be on Absence Theory), then given its origin it also counts as a natural rather than a moral evil. Likewise, cancer, pandemics, earthquakes, meteor strikes, aging and perhaps even death itself are (typically) regarded as natural rather than moral evils. In the tradition there are characterizations of natural evil that use one or more of the four templates discussed earlier: Absence , Privation , malfunctions in Matter , or some other Real Property .

Outside of religious contexts, however, many philosophers (and people generally) will be reluctant to characterize hurricanes, diseases, and meteor strikes as a source of “evil”. The very idea of natural evil seems most at home in theological debates: can the Author of Nature be supremely good, wise, and powerful and yet still create a world that contains so much pain and suffering, not to mention Category Five hurricanes, animal predation, and Alzheimer’s Disease?

There are different kinds of responses to the problem of natural evil in theological traditions: Aesthetic responses say that we don’t presently have the right perspective to see the overall beauty of the natural system, and thus that there really is no natural evil; Soul-making responses say that this present vale of natural evil is justified because it gives us the chance to become virtuous; Skeptical theistic responses say that given our limited faculties we cannot reasonably expect to understand why God would allow natural evils. (For more on these responses see M. Adams 1998, Tooley 2002 [2019]).

A very different kind of response involves recharacterizing at least some natural evils as moral evils. For instance, we might focus on ways in which human activity has set in motion the kinds of environmental, climatic, microbial, and biospheric changes that lead to “natural” disasters, pandemics, famines, and other empirical evils. Or, in a more religious context, we might seek to explain the suffering and misery caused by “nature” by appeal to sin, karmic law, or divine justice.

On some versions of this view, the morally responsible agent can be someone other than the victim: Adam sins, and now the non-human “creation has been growning” ( Romans 8:22, NRSV)—and makes the rest of us groan along with it. Or: we emit the greenhouse gasses, and our descendants three generations later suffer.

Other versions insist that the fault lies with the victim of natural evil himself: Eliphaz the Temanite says to Job

Think now, who that was innocent ever perished? Or where were the upright cut off? As I have seen, those who plow iniquity and sow trouble reap the same. By the breath of God they perish, and by the blast of his anger they are consumed. The roar of the lion, the voice of the fierce lion, and the teeth of the young lions are broken. ( Job 4: 7–10, NRSV)

Eliphaz suggests that if Job continues to suffer, then there must be some explanation for these natural evils in Job’s own past behavior. The risk in this kind of doctrine will be obvious to less ancient sensibilities: to suggest that a victim of natural evil must be morally responsible for it seems like the very definition of adding insult to injury.

Another way to recharacterize natural evil as moral evil is by appealing to the actions of moral agents who are neither divine nor human. This appears to have been Augustine’s position in places; more recently, it has been invoked by Alvin Plantinga as an at least broadly logically possible scenario which could be used in a “defense” against the logical problem of evil (Tooley 2002 [2019]). The scenario says that

[n]atural evil is due to the free actions of nonhuman persons; there is a balance of good over evil with respect to the actions of these nonhuman persons; and it was not within the power of God to create a world that contains a more favorable balance of good over evil with respect to the actions of the nonhuman persons it contains. (Plantinga 1989: 58)

A final way to recharacterize natural evil as moral would simply be to hold God alone morally blameworthy for it, since God created a world in which sentient creatures suffer so terribly. Obviously this would not be a winning strategy if the goal is a successful theodicy. Indeed, some theistic traditions block this route a priori by saying that it is conceptually impossible for God to do wrong: a perfect creator is either unbound by moral principles, or essentially incapable of violating them (see Murphy 2017).

If these efforts to locate a moral basis for natural evil (whether empirical or metaphysical) are unsuccessful, then “natural” remains a distinct category of evil’s origin.

Here again is our Table of Evils, now with more examples that go beyond Nietzsche’s horse case:

Figure 2. The Table of Evils, with general examples

Some types of evil do not seem to fit nicely in the Table of Evils. The three dealt with in what follows are systemic evil, symbolic evil , and radical evil . The nature and the origins of such evils are difficult to discern, and may require an expansion of the Table.

Systemic evil is the kind of evil that exists at the level of systems or groups rather than merely at the level of individuals. Organized structures like governments, corporations, teams, and religious institutions can be evil in this way; so can more loosely-organized systems such as “Academia”, “White Supremacy”, and “Wall Street”. Indeed, Google includes the motto “Don’t Be Evil” in its Company Code of Conduct (for the origins of this, see Chang 2019); disgruntled employees later sued the company alleging that it had violated that pledge (Allyn 2021).

Systemic evil seems empirical rather than metaphysical, but its origins are difficult to identify. It does not appear to be a merely natural phenomenon, or a spooky non-agential one, and yet its supra-agential character makes it seem not entirely moral either.

Hannah Arendt’s early work on totalitarianism (1951) depicts systemic evil as a kind of empirical evil for which no individual or even collection of individuals is fully responsible. The cogs in the machine, as well as the leader or leadership, may be the origin of some of the harms involved, but the evil of the whole structure (on Arendt’s view) is somehow greater than the sum produced by its parts.

Racism is another prominent example of systemic evil. Recent accounts of entrenched racist structures in certain societies (the United States, for instance) suggest that such evil has its origin not merely in actions and intentions but also in omissions and passivity: people who do not explicitly support the systemic evil can count as agents of it. If this is right, then active work against the evil in question—“anti-racist” activity rather than “non-racist” activity, for example—becomes morally required (Kendi 2019). But even such active opposition efforts may not be sufficient to avoid complicity in systemic evil: some theorists argue that even anti-racist people living in racist societies are “tainted” by its evil all the same (Rothstein 2017). That leads to the next kind of problem case.

Symbolic value is a less familiar idea in philosophy than it is in Anthropology, Religious Studies, and the other social sciences that deal with production, exchange, and consumption. The main idea is that an object or act can have far more “symbolic” value to a certain individual than its exchange or monetary value on some market or other, typically because of its causal history. That bauble given to you by a now-deceased friend has far more symbolic value (to you, at least) than the monetary or exchange value that it would fetch at auction. This is because the gift had its origin in the generosity of a friend who has since died, has the ability to invoke his memory, and so on.

Symbolic disvalue works in an analogous way. Products that fetch a certain monetary value on an open market may have significant symbolic disvalue that is not reflected in their price. This disvalue often arises, again, from the product’s provenance: an industrial chicken sandwich is a result of the obscene degradation of animals, workers, and the environment on the part of an industry that pays revolving-door lobbyists to promote food policy that, in turn, keeps production costs and prices artificially low, in part by externalizing most of the harms it causes. In some cases, such disvalue may be significant enough to make purchasing the product wrong, even if doing so does not lead, causally, to any actual harm or rights-infringement (see Chignell 2016).

The term “evil” is typically only applied in cases of symbolic disvalue when there is something excessive about them. Stepping on a flag or consuming a chicken sandwich might have some symbolic disvalue, but it would be strange to say that these are cases of symbolic evil . Having anything to do with soap made from the body fat of people murdered at the Stutthof concentration camp, by contrast, seems downright evil, even if the intended consequences (i.e., getting oneself clean) are good. Indeed, the symbolic evil attached to such a product may “touch” everyone who was part of the society that permitted its manufacture, whether they were directly involved or not. The evil of sustained chattel slavery may have a similar sort of symbolic power, even generations later.

Both the nature and the origins of symbolic evil are hard to characterize; thus it is hard to see where it fits in the Table of Evils. Is there a metaphysical aspect to symbolic evil, or is it fully empirical? Is its origin entirely natural, or moral, or is there something spooky and non-agential about it? A related set of difficulties has to do with the fact that, in some contexts anyway, the transfer of symbolic disvalue operates via the complex logic of “taint”, contagion, or uncleanness. Some philosophers who write about symbolic value argue that the only way to avoid being “touched” by such evil is via active and explicit dissociation —a symbolic “standing with the good” (R. Adams 1999). In other words, even if our abstinence does not make a difference, we must symbolically oppose an evil practice by, say, explicitly signaling opposition and (where possible) refusing to consume or benefit from its results (see Hill 1983; Appiah 1986).

7. Radical Evil: Four Conceptions

Kant is the source of “radical evil”, but his way of using the term is now out of favor. In fact, most people writing on this issue use the term in a precisely non-Kantian way to refer to something spectacularly excessive—an act or event whose badness is deeper and more mysterious than the badness of ordinary states or activities (Bernstein 2002).

In her famous account of the trial of Adolf Eichmann in Jerusalem, Arendt (1963) promoted the idea that even the worst evils can be “banal” and bureaucratical. A decade earlier, however, she conceived of “radical evil” quite differently. In a March 1951 letter to Karl Jaspers she wrote:

We know that the greatest evils or radical evil has nothing to do anymore with humanly understandable, sinful motives. What radical evil is I don’t know, but it seems to me to somehow have to do with the following phenomenon: making human beings as human beings superfluous . (Arendt & Jaspers 1992: 166, emphasis added)

Arendt went on in the letter to insist that “making human beings as human beings superfluous” is not the same as treating them as “mere means to an end”. She thus rejected the Kantian view that radical evil—no matter how benign or awful the effects—has its root (Latin: radix ) in the willful violation of the categorical imperative by individual free agents (see Louden 2010: 98).

Jaspers’s view of radical evil, by contrast, was consistently non-exotic and Kantian: “there is evil because there is freedom. It is only possible for the will alone to be evil ” (Jaspers 1947 [1958: 532], emphasis added). According to one commentator, this focus on human freedom as the root of all evil is “significant and commendable”:

Kant [and following him Jaspers] refuses to cater to our prurient craving for a special account that applies especially to the most extreme cases of evil…. He fears that occupying our imaginations with extreme cases of evil may be merely a way of indulging some of our nastier human traits—rationalizing our resentment and vindictiveness by supplying it with an object that would seem to justify it. (Wood 2010: 157)

The concerns expressed in these passages are common in discussions of the nature and origins of evil. Call them Jaspersian concerns for short. They are second-order concerns about how we should conceive and speak of evil, especially when calling it “radical”. The concerns fall into two broad kinds: (1) concerns about exoticizing wrongdoing with excess-locutions like “evil” and “radical”; and (2) concerns about tainting people and things that are touched by such evil, beyond the straightforward condemnation of the perpetrators (for more on such concerns, see Card 2002, Cole 2006, and Calder 2013 [2018]).

The Kantian conception of radical evil is considered in the next sub-section. Sections 7.2–7.4 survey three other conceptions and look at how each might raise Jaspersian concerns.

Kant argues in Religion within the Bounds of Reason Alone (1793) that free choices against the moral law are unintelligible in the sense that they are irrational ; our propensity to make them is thus an inexplicable mystery at the “root” ( radix ) of our moral psychology. The post-Kantian idealist F.W. J. Schelling, picking up the refrain, rejects the Privation Theory of radical evil in favor of the view that it has its origin in the positive, irrational decision to prefer self-advantage over the moral law, though “just how the decision for good or evil comes to pass in the individual, that is still wrapped in total darkness” (Schelling 1809 [1936: 59]).

On the Kantian view of radical evil, then, there is a normative sense in which it is “unintelligible”: it is an irrational propensity and thus cannot be “understood” in the sense that it cannot be sanctioned by proper reason. Likewise, an immoral choice cannot be fully explained: full explanations appeal to good reasons, and there are no good reasons for wrongdoing.

There is a broader sense of “understand” or “explain”, however, in which such evil is no mystery at all: it is the most depressingly familiar thing in the world. Everyday immoral agents presumably see what they are doing under the aspect of some good or other—good for their company, good for their bonuses, good for their reelection efforts—even if they also know that what they are doing is ultimately wrong. If we interpret the war cry of Milton’s Satan—“Evil be thou my good!” (1667, Paradise Lost , Bk IV, line 110)—as implying that he is taking evil under the guise of the good, then it is not absurd. But it is still not wholly intelligible, either.

The Kantian conception of radical evil, then, says that that the unintelligibility of wrongdoing does not prevent us from assigning blame, holding perpetrators responsible, and refusing to taint the innocent. This explains why Jaspers found the conception so attractive:

To rank the will to happiness, which dominates among men’s motives, above the unconditioned law that shows itself in reason— that is the root of evil, the “propensity” which Kant calls “radical evil”. (Jaspers 1962: 321)

Some philosophers, and more than a few novelists and screenwriters, find the view that agents are always choosing under “the guise of the good” as inadequate to the psychology of extreme malevolence. It may have suited the melioristic conceptions of the Enlightenment, but the well-publicized, mechanized horrors of the recent past allegedly demand a bleaker picture of perpetrator psychology. Serial killers, murderous dictators, torturers, derivatives traders: the idea is that at least some of these malign actors see their own actions as atrocious —as making the world worse rather than better on the whole (and sometimes worse for themselves)—and yet still choose to perform them. They are thus not saying “Evil be thou my good”, at least on the interpretation just offered. Rather, they are self-consciously male-volent: they will the bad under the aspect of the bad . And yet they are not insane—this is what makes their actions especially difficult for the rest of us to understand. This is a second common way in which the term “radical evil” is used. (See Pauer-Studer and Velleman 2015 for a reading of Milton according to which Satan’s war cry is interpreted in this way.)

Augustine reports in his Confessions that during the pears incident he took pleasure in

the theft and sin itself.… Behold, now, let me heart tell Thee what it was seeking there, that I should be gratuitously wanton, having no inducement to evil but the evil itself. It was foul, and I loved it … I loved my own error – not that for which I erred, but the error itself. ( Confessions Bk 2, Chap. IV, paragraph 9 [1876: 30])

It is tempting to say that in “loving the sin itself”, Augustine was in some sense taking it to be good. But he at least gestures here at this second conception of radical evil: some acts can be performed under the aspect of their own abject deformation and rottenness—as ultimately bad even for the agent himself.

As we have seen, the Prussian philosopher who coined the term “radical evil” 1400 years later did not think that it can involve choosing evil for its own sake. In fact, Kant argues that such a “diabolical” conception of evil is incoherent or at least psychologically inapplicable to human beings. Kant’s skepticism, however, did not prevent Dostoyevsky from composing Notes from the Underground (1864) as an extended and somewhat plausible portrait of one man’s effort to choose evil qua evil. And there are more recent accounts of people who confess—in private diaries or braggadocios depositions—that they did what they did because it was evil.

In this context, consider (if you can bear it) Stone 2009’s portraits of the worst serial killers: although some perpetrators report that they take what they are doing to be good—ridding the world of “garbage women”, giving someone his “just deserts”, correctly following the orders of the voices in their head, and so on—others openly admit that what they are doing is bad, wrong, evil, despicable, and so on. In such cases, the perpetrators are not making a series of unsound inferences, or mistaking the bad for the good. Rather, they seem to be engaged in a self-conscious turning away from anything that could be regarded as good by anyone . Radical evil on this conception is sometimes described as a self-conscious turning away from being itself (Eagleton 2009: 16; L. Russell 2014: 23).

If Augustine’s confessions about the pears—or these more spectacular recent confessions—are accurate, then there may be ( contra Kant) a baffling diabolical state that some people can fall into, and that the rest of us cannot entirely fathom. The unintelligibility here thus threatens to raise Jaspersian concerns about exoticization and taint.

A third conception of “radical evil” takes the term to refer to acts or practices (like slavery and genocide) that do not merely violate rational principles but also constitute an effort to repudiate moral rationality altogether, or at least to transcend it. When we steal or lie, there is a sense in which we might be failing to treat others with the respect that they deserve. And as we have seen, Kant does not hesitate to view such irrational choices as stemming from our propensity to “radical evil”. But radical evil on this third conception is different: it involves an intentional refusal to acknowledge that some group of persons has any moral standing at all —the kind of moral standing that would prohibit us from instigating or complying with their humiliation, degradation, or extinction.

Radical evil of this sort, in other words, denies the universal scope—and thus, perhaps, the very existence—of the moral sphere altogether. It is anti-rational rather than merely irrational : this is what threatens to make it incomprehensible or inexplicable in a unique way. This is also presumably what the early Arendt means when she says that radical evil involves making human beings “superfluous”.

By way of analogy: suppose that ordinary moral wrongdoing is like making a bad move in chess, or like cheating by moving one’s pieces in an illegal way when one’s opponent isn’t looking. A radically evil act on the present conception, by contrast, would be like crushing all of the opponent’s pieces and upending the table. The player who does the latter is no longer or perhaps never was a player: she cannot explain her behavior by saying that she was trying to win by making what turned out to be a bad move or a cheat. She cannot explain her behavior in terms of chess at all. Likewise, a radically evil act is supposed to transcend the terms of moral rationality.

The analogy extends only so far, however. That’s because there are still some reasonable explanations left to our non-player—explanations external to the game:

I was hungry, so I crushed all of your pieces and swept them off the board in the hopes that you would suggest lunch.

By contrast, it is not obvious how there can be any credible reason given for doing something that constitutes a denial of reason altogether. That is why radical evil in the repudiative sense threatens to be uniquely unintelligible and troubling. If someone—not a beast or a machine, but a human being—acts in a way that entirely disclaims not just an awareness of moral authority but also the basic rules of moral reason altogether, he effectively places himself in an antelapsarian state, unburdened with the knowledge of good and evil. His act asserts that he has transcended entirely the moral sphere—that what he does cannot be wrong.

Can human beings really perpetrate radical evil in this sense? Richard Ramirez, the “Night Stalker” serial killer in 1980’s Los Angeles, certainly took himself to be doing so. He repeatedly snuck into people’s houses to rape and kill them, sometimes cutting off body parts and taking them with him. After his fourteenth murder, he was caught and then boasted to his captors:

You don’t understand me… you are not capable of it. I am beyond good and evil…I love to kill people. I love to watch them die. I would shoot them in the head and they would wiggle and squirm… I love all that blood. (quoted in Stone 2009: 208).

This idea—that a human being could do something that would enact his own transcendence of moral norms, make other persons superfluous, and establish his own status as somehow “beyond good and evil”—is what concerned Karl Jaspers, and what he was hoping Arendt would resist. In her face-to-face confrontation with the quotidian, bureaucratic evil of Adolf Eichmann, she seems to have changed her mind (though see Cesarani 2007 and Margalit 2019 for a portrait of Eichmann that resists her famous “banality thesis”). But even if Arendt ultimately repudiated the repudiative conception of radical evil, others remain sympathetic to some version of it (e.g., Bernstein 2002 and Motzkin 2019).

“Systemic” or “structural” evil exists at the level of groups, networks, races, and collectives rather than merely at the level of individuals (see section 5 ). The fourth conception of “radical evil” applies the concept to particularly rampant or entrenched evils of the systemic sort.

Although Arendt’s interest was in totalitarian state systems, contemporary philosophers are equally interested in non-state but still super-human systems: corporations, collectives, markets, artificial intelligences. These are now some of the most potent sources of value and disvalue at work in the world. As noted in section 5 , the nature of systemic evil is empirical, but its origin is more difficult to grasp. Corporations have boardrooms and executive suites, and are even treated as persons by some legal systems, but (as investigation after investigation indicates) it is hard to find the heart of their darkness when assigning responsibility. Still-powerful systems of white supremacy are based in structures of which no individual human mind was or is fully aware (Rothstein 2017). Likewise, markets, algorithms, blockchains, and various forms of artificial intelligence arise out of innumerable individual human efforts, but are also designed to make human beings superfluous in a literal and unprecedented way. Some of this is and will be for the good—there are already fewer back-broken menial laborers in the fields or exhaust-guzzling toll collectors on the highways. But some will surely be for the bad, and in ways that are not foreseeable. Radical systemic evil—especially in a technological age—does not seem natural or spooky, but it does not seem fully moral either.

In sum: although natural, moral, and spooky non-agential conceptions of the origins of evil were the focus of most traditional discussions of the origins of evil, it looks like some varieties of radical evil have a different origin altogether—one that is not entirely intelligible. This raises Jaspersian concerns: a world in which (non-Kantian) radical evils are widespread is one in which ascriptions of moral responsibility are increasingly difficult to make.

Adams, Marilyn McCord, 1998, Horrendous Evil and the Goodness of God , Ithaca, NY: Cornell University Press.
Adams, Robert Merrihew, 1999, Finite and Infinite Goods: A Framework for Ethics , New York: Oxford University Press. doi:10.1093/0195153715.001.0001
Appiah, Kwame Anthony, 1986, “Racism and Moral Pollution”, Philosophical Forum , 18(2): 185–202
Allyn, Bobby, 2021, “Ex-Google Workers Sue Company, Saying it Betrayed ‘Don’t be Evil’ Motto”, National Public Radio . [ Allyn 2021 available online ]
Anderson-Gold, Sharon and Pablo Muchnik (eds), 2010, Kant’s Anatomy of Evil , New York: Cambridge University Press. DOI: 10.1017/CBO9780511691898
Anselm, c. 1080–86, De Casu Diaboli . Collected in vol 1 of Dom F. S. Schmitt, O.S.B.S. Anselmi Cantuariensis Archiepiscopi Opera Omnia , 6 vols., Edinburgh: Thomas Nelson and Sons, 1940–1961.
Arendt, Hannah, 1951, Origins of Totalitarianism , New York: Harcourt, Brace.
–––, 1963, Eichmann in Jerusalem: A Report on the Banality of Evil , New York: Viking Press. Originally a five part series in The New Yorker , February-March 1963.
Arendt, Hannah and Karl Jaspers, 1992, Hannah Arendt/Karl Jaspers correspondence, 1926–1969 , Lotte Kohler and Hans Saner (eds), Robert and Rita Kimber (trans), New York: Harcourt Brace Jovanovich.
Augustine, c. 400 [1963] Confessiones . Translated in The Confessions of St. Augustine , J.G. Pilkington (trans.), Edinburgh: T&T Clark Press.
–––, The City of God , in Henry Bettenson (trans.) The City of God , Harmondsworth: Penguin, 1977.
Bernstein, Richard J., 2002, Radical Evil: A Philosophical Interrogation , New York: Polity Press.
Bhattacharjee Yudhijit, 2018, “The Science of Good and Evil”, National Geographic , January edition, pp. 116–142.
Calder, Todd, 2013 [2018], “The Concept of Evil”, The Stanford Encyclopedia of Philosophy (Fall 2018 Edition), Edward N. Zalta (ed.), URL = https://plato.stanford.edu/archives/fall2018/entries/concept-evil/ .
Card, Claudia, 2002, The Atrocity Paradigm: A Theory of Evil , New York: Oxford University Press. doi:10.1093/0195145089.001.0001
–––, 2010, “Kant’s Moral Excluded Middle”, in Anderson-Gold and Muchnik 2010: 74–92. doi:10.1017/CBO9780511691898.005
Cesarani, David, 2007, Becoming Eichmann: Rethinking the Life, Crimes, and Trial of a “Desk Murderer” , London: De Capo Press.
Chang, Wesley, 2019. “On Google and Not Being Evil”, in Chignell 2019b: 450–455. doi: 10.1093/oso/9780199915453.003.0027
Chignell, Andrew P., 2016, “Can We Really Vote with our Forks? Opportunism and the Threshold Chicken”, in Chignell et al. (eds), Philosophy Comes to Dinner , New York: Routledge, pp. 182–202.
–––, 2019a, “Evil, Intelligibility, Radicality: Footnotes to a Correspondence between Hannah Arendt and Karl Jaspers”, in Chignell 2019b: 18–42. doi:10.1093/oso/9780199915453.003.0002
––– (ed.), 2019b, Evil: A History , New York: Oxford University Press. doi:10.1093/oso/9780199915453.001.0001
Cole, Phillip, 2006, The Myth of Evil , Edinburgh: University of Edinburgh Press.
Davies, Brian, 2019, “Evil and Late Medieval Thought”, in Chignell 2019b: 225–251. doi:10.1093/oso/9780199915453.003.0012
Eagleton, Terry, 2009, On Evil , New Haven, CT: Yale University Press.
Frankfurter, David, 2006, Evil Incarnate , Princeton, NJ: Princeton University Press.
Germann, Nadja, 2019, “… but draw not nigh this tree : Evil in Early Islamic Thought”, in Chignell 2019b: 198–224. doi: 10.1093/oso/9780199915453.003.0011
Guhrauer, G.E., 1842 [1966], Gottfried Wilhelm Freiherr v. Leibnitz. Eine Biographie , 2 volumes, Breslau: Hirt. Edition of 1846 reprinted Hildesheim: Olms, 1966.
Hartshorne, Charles, 1984, Omnipotence and other Theological Mistakes , Buffalo, NY: SUNY Press.
Haybron, Daniel, 2002, “Consistency of Character and the Character of Evil”, in Earth’s Abominations: Philosophical Studies of Evil , Daniel M. Haybron (ed.), Amsterdam: Rodopi, pp. 63–78.
Hill, Thomas E., 1983, “Ideals of Human Excellence and Preserving Natural Environments”, Environmental Ethics 5(3): 211–224.
Jaspers, Karl, 1947 [1958], Von der Wahrheit: Philosophische Logik , Erster Band, Munich: R. Piper & Co.. Second edition, 1958.
–––, 1962, Der Philosophische Glaube angesichts der Offenbarung , Munich: R. Piper.
Johnston, Mark, forthcoming, “How did Evil Come into the World?”, Gifford Lectures, Fall 2019.
Kamtekar, Rachana, 2019, “Explaining Evil in Plato, Euripedes, and Seneca”, in Chignell 2019b: 97–128. doi:10.1093/oso/9780199915453.003.0007
Kant, Immanuel, 1793, Die Religion innerhalb der Grenzen der bloßen Vernunft , Königsberg. Collected in AK 6: 1–202. Translated as Religion within the Bounds of Mere Reason , Allen Wood and George di Giovanni (trans), Cambridge: Cambridge University Press, 1998.
–––, [AK] 1902– . Gesammelte Schriften , Berlin: Königlich preussischen Akademie der Wissenschaften, now DeGruyter, quoted by volume:page.
Kearney, Richard, 2001,“Evil, Monstrosity and The Sublime”, Revista Portuguesa de Filosofia , T. 57, Fasc. 3, Desafios do Mal: Do Mistério à Sabedoria, Jul-Sep edition, pp. 485–502.
Kendi, Ibram X., 2019. How to be Anti-Racist , New York: One World Books (Penguin).
King, Peter, 2019, “Augustine on Evil”, in Chignell 2019b: 155–193. doi:10.1093/oso/9780199915453.003.0009
Leibniz, G.W., 1677 [1969], “Letter to Arnold Eckard”, written Summer 1677. Collected in Die philosophischen Schriften von Gottfried Wilhelm Leibniz [G], C.I. Gerhardt (trans. and ed.), Leipzig: Weidmann, 1875; translated in Leibniz 1969: 177–182.
–––, 1969, Philosophical Papers and Letters , second edition, Leroy Loemker (trans./ed.), Dordrecht: Reidel.
Louden, Robert, 2010, “Evil Everywhere: The Ordinariness of Kantian Radical Evil”, in Anderson-Gold and Muchnik 2010: 93–115. doi:10.1017/CBO9780511691898.006
Malebranche, Nicolas, 1674–5 [1997], De la Recherche de la vérité , 2 volumes, Paris: André Pralard. Translated as Search after Truth , Thomas M. Lennon and Paul J. Olscamp (trans. and eds.), New York: Cambridge University Press, 1997.
Margalit, Avishai, 2019, “Evil, Instigators, and Compliers”, in Chignell 2019b: 444–449. doi:10.1093/oso/9780199915453.003.0026
Martin, Eric and Eric Watkins, 2019, “Evil, Natural Science, and Animal Suffering”, in Chignell 2019b: 383–413. doi:10.1093/oso/9780199915453.003.0022
Mates, Benson, 1989, The Philosophy of Leibniz: Metaphysics and Language , New York: Oxford University Press. doi:10.1093/0195059468.001.0001
Matuson, Dana. 2017. “President Trump Responds Differently to NY, LV Attacks: What They’re Saying”, [ Matuson 2017 available online ]
Matuštík, Martin J. 2008. Radical Evil and the Scarcity of Hope , Bloomington, IN: Indiana University Press.
Motzkin, Gabriel, 2019, “Evil after the Holocaust”, in Chignell 2019b: 429–443. doi:10.1093/oso/9780199915453.003.0025
Murphy, Mark, 2017, God’s Own Ethics: Norms of Divine Agency and the Argument from Evil . New York: Oxford.
Murray, Michael, 2008, Nature Red in Tooth and Claw: Theism and the Problem of Animal Suffering , New York: Oxford. doi:10.1093/acprof:oso/9780199237272.001.0001
Newlands, Samuel, 2019, “Evil, Privations, and the Early Moderns”, in Chignell 2019b: 273–305. doi:10.1093/oso/9780199915453.003.0016
O’Connor, Timothy and Christopher Franklin, 2018 [2021] “Free Will”, The Stanford Encyclopedia of Philosophy (Spring 2021 Edition), Edward N. Zalta (ed.), URL = https://plato.stanford.edu/archives/spr2021/entries/freewill/ .
O’Meara, Dominic J., 2019, “Explaining Evil in Late Antiquity: Plotinus and his Critics”, in Chignell 2019b: 129–154. doi:10.1093/oso/9780199915453.003.0008
Plantinga, Alvin, 1989, God, Freedom, and Evil , Grand Rapids, MI: Eerdmans.
Pauer-Studer, Herlinde and J. David Velleman, 2015, Konrad Morgan: Conscience of a Nazi Judge , New York: Palgrave MacMillan.
Plotinus, Enneads , collected c. 270 CE. Translated in Plotinus: The Enneads , Lloyd P. Gerson (trans and ed.), Cambridge: Cambridge University Press, 2017.
Prideaux, Sue, 2018, I am Dynamite! A Life of Friedrich Nietzsche , New York, NY: Penguin.
Rothstein, Richard, 2017, The Color of Law: A Forgotten History of How our Government Segregated America , New York, NY: Liveright Press (W.W. Norton).
Ruppel, Antonia, 2019, “Kakology: A Study of Some Evil Words”, in Chignell 2019b: 44–59. doi:10.1093/oso/9780199915453.001.0001
Russell, Jeffrey Burton, 1977, The Devil , Ithaca, NY: Cornell University Press.
–––, 1981, Satan , Ithaca, NY: Cornell University Press.
Russell, Luke, 2014, Evil: A Philosophical Investigation , New York: Oxford University Press. doi:10.1093/acprof:oso/9780198712480.001.0001
Schelling, F.W.J., 1809 [1936], Philosophische Untersuchungen über das Wesen der menschlichen Freiheit und die damit zusammenhängenden Gegenstände . Translated as Of Human Freedom , James Gutmann (ed. & trans.), Chicago: Open Court.
Sharpe, A., 1909, “Evil”, in The Catholic Encyclopedia , New York: Robert Appleton Company. [ Sharpe 1909 available online ]
Stone, Michael, 2009, The Anatomy of Evil , Amherst, NY: Prometheus Books.
Tooley, Michael, 2002 [2019], “The Problem of Evil”, The Stanford Encyclopedia of Philosophy (Spring 2019 Edition), Edward N. Zalta (ed.), URL = < https://plato.stanford.edu/archives/spr2019/entries/evil/ >
Wolterstorff, Nicholas, 1988, “Suffering Love”, in Philosophy and the Christian Faith , Thomas V. Morris (ed.), Notre Dame, IN: University of Notre Dame Press, 196–237.
Wood, Allen W., 2010, “Kant and the Intelligibility of Evil”, in Anderson-Gold and Muchnik 2010:, 144–172. doi:10.1017/CBO9780511691898.008

How to cite this entry . Preview the PDF version of this entry at the Friends of the SEP Society . Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entry at PhilPapers , with links to its database.

Bush, George W., 2002, State of the Union Address , January 29, 2002.
Reagan, Ronald, 1983, Evil Empire Speech , to the National Association of Evangelicals, March 8, 1983
The Philosophical Lexicon , by Daniel Dennett and Asbjørn Steglich-Peterson
“ Anselm of Canterbury 1033–1109 ”, Internet Encyclopedia of Philosophy .

Acknowledgments

The author wishes to thank participants in a workshop at the Princeton Project in Philosophy and Religion, as well as Brendan Kolb and an anonymous referee, for helpful comments on an earlier draft. Thanks also to Oxford University Press for permission to incorporate some passages from Chignell 2019a and 2019b into this entry.

Accessibility

Support SEP

Mirror sites.

View this site from another server:

Info about mirror sites

Library of Congress Catalog Data: ISSN 1095-5054

IMAGES

Problem of Evil Essay A-Level AQA Philosophy (7172)
Understanding Evil Paper Arguments Free Essay Example
⇉The problem of good and evil Essay Example
Possibility Of Evil Free Essay Example
≫ The Problem of Evil Free Essay Sample on Samploon.com
The Problem of Evil and Suffering

VIDEO

Argumentative Essay
Argumentative Essay Rubric
The Problem of Evil, Theodicy, and Atonement Theories OPEN PANEL
Monologue A discussion about Carbon Footprints and Speaking Task
Leibniz and the Problem of Evil
Not My Problem— EVIL Jean

COMMENTS

The Problem of Evil
1. The Incompatibility Problem of Evil. The 'Incompatibility' or 'Logical' versions of the Problem of Evil claim that evil's existence is logically incompatible with God's existence: believing in God and evil is like believing in a five-sided square, a contradiction.[3] Most philosophers today reject this argument.[4]
The Problem of Evil
The Problem of Evil. First published Mon Sep 16, 2002; substantive revision Tue Mar 3, 2015. The epistemic question posed by evil is whether the world contains undesirable states of affairs that provide the basis for an argument that makes it unreasonable to believe in the existence of God. This discussion is divided into eight sections.
The Problem of Evil
The logical problem makes a large claim, that evil and God cannot possibly co-exist. Defeating the logical problem requires conceiving of some logically possible scenario or reason God could have for allowing evil. The Evidential problem of evil. This is the a posteriori argument that the evidence of evil in the world makes belief in God ...
Leibniz on the Problem of Evil
The standard solution adopted by medieval thinkers was to deny an assumption of the preceding argument, namely, that evil is "something." Evil was claimed not to have any positive reality, but to be a mere "privation" or "lack" of being. ... is the first book-length treatment of Leibniz's work on the problem of evil; the essays in ...
Logical Problem of Evil
This essay examines one form the argument from evil has taken, which is known as "the logical problem of evil." In the second half of the twentieth century, atheologians (that is, persons who try to prove the non-existence of God) commonly claimed that the problem of evil was a problem of logical inconsistency.
PDF THE PROBLEM OF EVIL
the problem of evil 189 Let us call this argument "the argument from evil"—glossing over the fact that there are many arguments for the nonexistence of God that could be described as arguments from evil. The intellectual problem I call the problem of evil can be framed as a series of closely related questions addressed to theists: How would
The Problem of Evil
the problem of evil is not so much an intellectual problem as it is a problem of lived experience; it is an existential problem related to their physical, emotional, and relational well-being. ... The argument is that a greater good occurs if people are given the freedom to choose, even if some choose what is evil, than to have never been given ...
Augustine's Treatment of the Problem of Evil
In fact, we find Augustine asserts the following as to how it is that we can understand evil as unsubstantial and therefore also unreal: "…evil whose origin I sought is not a substance.". "…To You, then evil utterly is not—and not only to You, but to Your whole creation likewise, evil is not…". Moreover, because evil is not, or ...
Problem of evil
The problem of evil is the philosophical question of how to reconcile the existence of evil and suffering with an omnipotent, ... a defense refers to attempts to address the logical argument of evil that says "it is logically impossible ... Economic theorist Thomas Malthus stated in a 1798 essay on the question of population over-crowding, ...
Problem of evil
In this argument and in the problem of evil itself, evil is understood to encompass both moral evil (caused by free human actions) and natural evil (caused by natural phenomena such as disease, earthquakes, and floods). Most thinkers, however, have found this argument too simple, since it does not recognize cases in which eliminating one evil ...
8 The Problem of Evil
Oxford Handbooks. Collection: Oxford Handbooks Online, Oxford Scholarship Online. 1. Introductory Remarks: The Problem of Evil and the Argument from Evil. There are many ways to understand the phrase "the problem of evil.". In this chapter, I understand this phrase as a label for a certain purely intellectual problem—as opposed to an ...
Evidential Problem of Evil, The
The Evidential Problem of Evil. The evidential problem of evil is the problem of determining whether and, if so, to what extent the existence of evil (or certain instances, kinds, quantities, or distributions of evil) constitutes evidence against the existence of God, that is to say, a being perfect in power, knowledge and goodness.Evidential arguments from evil attempt to show that, once we ...
Argumentative Essay: The Problem Of Evil
However, this has led to profound criticism and arguments of God's inexistence. The strongest argument in contradiction to God's existence is the Problem of Evil, presented by J.L Mackie. In this paper, I aim to describe the problem of evil, analyse the objection of the Paradox of Omnipotence and provide rebuttals to this objection.
PDF The problem of evil and the free will defence
The problem of evil and the free will defence. THE ARGUMENT. God is traditionally understood to be perfectly good, omnipotent and omniscient. The existence of evil causes problems for this definition. If God is good, then he has the desire to eliminate evil. If God is omnipotent, then God is able to eliminate evil.
How Should Christians Approach the Problem of Evil?
Christians cannot refute the problem of evil by appealing only to the future, which the skeptic, not presupposing biblical eschatology, recognizes as the fallacy of argumentum ad futuris. 2 Further, Meister's argument misrepresents biblical eschatology, for Scripture nowhere teaches that God will abolish all evil.
The Problem of Evil
The so-called "problem of evil" is an argument against the existence of God that reasons along these lines: A perfectly powerful being can prevent any evil. A perfectly good being will prevent evil as far as he can. God is perfectly powerful and good. So, if a perfectly powerful and good God exists, there will be no evil.
The problem of evil summary notes
The logical problem of evil (Mackie) The logical problem of evil is a deductive argument. It aims to show that the existence of evil is logically inconsistent with the existence of the God of classical theism - God defined as omnipotent, omnibenevolent and omniscient. The classical logical problem of evil: epicurus.
The Problem of Evil
The Problem of Evil. First published Mon Sep 16, 2002; substantive revision Fri Aug 21, 2009. The epistemic question posed by evil is whether the world contains undesirable states of affairs that provide the basis for an argument that makes it unreasonable for anyone to believe in the existence of God. This discussion is divided into nine sections.
Summary of "The Problem of Evil"
4. The idea that evil results from free will - Evil results from free will. A world with humans and the evil that results from their free will is better than one without humans even if that world had no evil. War, murder, torture, etc. are worth the price of the positives that derive from human free will. PROBLEMS - We can answer that free ...
Kinds and Origins of Evil
Philosophers of religion are a half-exception to the rule. They did and do continue to speak of evil, at least when discussing the "problem" thereof. (See the entry on the problem of evil.) If pressed, though, they typically admit that this is because the great framers of the problem—Augustine, Aquinas, Leibniz, Bayle—used the term (in ...
The Problem of Evil: Selected Readings, Second Edition on JSTOR
The philosophical problem of evil is the challenge of reconciling belief in God with evil in the world. The theistic concept of God as supremely powerful, intelligent, and good makes the problem very difficult because such a being, it would seem, would make a much better world than this one. All three great theistic religions—Judaism ...

The Problem of Evil

1. The Incompatibility Problem of Evil

2. The Evidential Problem of Evil

2.1. Free Will

2.2. “Soul-Making”

3. Outweighing Evidence?

4. Evil Is No Evidence?

5. Conclusion

Related Essays

Revision History

PDF Download

About the Author

Follow 1000-Word Philosophy on Facebook , Twitter and subscribe to receive email notifications of new essays at the bottom of 1000WordPhilosophy.com

8 The Problem of Evil

1. Introductory Remarks: The Problem of Evil and the Argument from Evil

2. The “Moral Insensitivity” Charge

3. God's Omnipotence, His Moral Perfection, and His Knowledge of Evil

4. A Description of an Ideal Debate about the Existence of God

5. Atheist's Initial Statement of the Argument from Evil; Theist Begins His Reply by Making a Point about Reasons

6. A Distinction: “Theodicy” and “Defense”

7. Theist's Reply Continues; The Initial Statement of the Free-will Defense

8. An Objection to the Free-will Defense: God Can Control the Exercise of Free Choice

9. A Second Objection to the Free-will Defense: Free Will Is Incompatible with God's Omniscience

10. Atheist Contends That the Free-will Defense Cannot Account for the Amount and the Kinds of Evil We Observe

11. Theist Elaborates the Free-will Defense: Evil Results from a Primordial Estrangement of Humanity from God

12. Atheist Turns to the Consideration of a Particular Horrible Evil

13. Theist Discusses the Relation of the Expanded Free-will Defense to the Question Whether an Omnipotent and Morally Perfect Being Would Eliminate Every Particular Horror from the World

14. Concluding Remarks: Evaluating Theist's Response to the Argument from Evil

Works Cited

This Feature Is Available To Subscribers Only

Most Popular

U.S. Edition

To All The World

The Problem of Evil

What Is the Problem of Evil?

Responding to the Problem of Evil

The Ways of Theodicy and Inscrutability

The Way of Theodicy

Four popular theodicies have some biblical basis

These theodicies fall under the umbrella of the “greater good theodicy.”

Constructing the “Greater Good Theodicy”: a Three-Fold Argument for Three Biblical Themes

The Case of Job

The Case of Joseph

The Case of Jesus

Licensing and Limiting the GGT

Licensing the GGT: God’s Sovereignty over All Evil

God’s Sovereignty over Moral Evil

God’s Sovereignty over All Evil

Limiting the GGT: The Inscrutability of God’s Purposes

Analogies for our Cognitive Limitations

Biblical Argument for Divine Inscrutability

Further Reading

The problem of evil summary notes

The logical problem of evil (Mackie)

The evidential problem of evil

Augustine’s theodicy

Augustine vs the evidential problem (the scientific critique of the fall & original sin)

Augustine vs the logical problem (Original sin violates moral responsibility & incompatibility with omnibenevolence)

Irenaeus & Hick’s theodicy

Irenaeus & Hick vs the evidential problem (The issues of purposeless and soul-breaking evil)

Irenaeus & Hick vs the logical problem

The problem of evil & the issue of free will (this paragraph can be used in any question)

The logical problem of evil as a version of the problem of evil (Mackie)

The evidential problem of evil as a version of the problem of evil (Hume or Rowe)

Augustine’s theodicy (response to the problem of evil)

Augustine vs the logical problem

Augustine vs the evidential problem

Irenaeus & Hick’s theodicy (response to the problem of evil)

Irenaeus & Hick vs the logical problem

Reason and Meaning

Share this:

Leave a Reply

Bibliography

Kinds and Origins of Evil

1. Speaking of Evil

3. Kinds of Evil: Metaphysical and Empirical

3.1 Metaphysical Evil

4. Origins of Evil: Moral and Natural

7. Radical Evil: Four Conceptions

Acknowledgments