Justified True Belief

Mapping the Landscape of Good Reasons for the Truth of Christianity

Christian Evidences

Evidence & Arguments for Christian Theism

Divinity of Christ

Biblical Evidence for High-Christology: Jesus is God

TBD

Old Testament Criticism

Evidence for the Reliability of the OT Bible

Common Objections

Objection Analyses to Christian Theism

Problem of Evil

If God is Good...Why Evil?

Intellectual PoE: Logical

(P1) A hidden contradiction requires a necessary truth to expose it. + The problem of evil starts as something anyone can feel, usually at a graveside or a hospital bed: a good and all-powerful God would not have allowed this, and this happened anyway. Its hardest form is not the complaint that evil makes God unlikely, but the charge that "God exists" and "evil exists" cannot both be true, the way one shape cannot be both a triangle and a square. (1) Some pairs of claims rule each other out by definition. - "Bob is a bachelor" and "Bob is married" cannot both be true, and you need nothing else to see it. "Bachelor" already means unmarried, so the second sentence takes back exactly what the first one gave. - Philosophers call that an explicit contradiction: the denial sits right there in the words, and nothing has to be added, investigated, or assumed. - "God exists" and "evil exists" are plainly not like that. Neither sentence contains the negation of the other, so whatever is wrong with holding both, it is not something you can see by reading them. (2) Most alleged contradictions need a hidden premise to do the work. - Think of a man who is rich and thoroughly miserable. "Bob is wealthy" and "Bob is unhappy" make an odd pair, and someone might insist the two of them clash. By themselves they do not, because wealthy people can be miserable and plenty of them are. - To turn that oddity into a genuine contradiction you would have to add a further claim, something like "all wealthy people are happy." With that in place, the pair collapses at once. Without it, there is nothing to collapse. - The charge against Christian belief has exactly this shape. The two claims do sit together awkwardly, and the awkwardness is real, but making it a contradiction takes a third claim that nobody has stated out loud. (3) The hidden premise has to be a necessary truth. - A necessary truth holds in every possible situation, not merely in every situation anyone has run into. Philosophers call each complete way things could have been a possible world: this one is a possible world, and so is the one where you took the other job. A necessary truth holds in all of them, which is why no possible world anywhere contains a married bachelor. - Compare "every swan I have seen is white" with "no swan could possibly be black." The first is a report on somebody's experience, and black swans in Australia finished it off. The second covers every swan there could ever be, and no amount of birdwatching will establish it. - A good rule of thumb will not do the job here. "All wealthy people are happy" would have to be a truth of the second kind for the argument about Bob to work, and so must whatever premise is carried quietly into the argument from evil. (4) The burden falls on whoever claims the contradiction. - In court, the side alleging fraud has to prove it, and the defendant does not have to prove his innocence in order to be entitled to it. Whoever says that two beliefs cannot both be true is the one with something to prove. - The oldest form of the charge is the dilemma reported by Lactantius and credited to Epicurus: if God is willing to prevent evil but unable, he is not omnipotent; if able but unwilling, he is not good. That is a powerful challenge and it deserves an answer. A challenge is not yet a demonstration, though, and the distance between the two is the whole issue. - The reply is deductive, which means its premises work like links in a chain: if each link holds, the conclusion cannot slip out. No link has to be certain. Each has to be more plausible than its denial. Therefore, the charge of contradiction stands or falls on a hidden premise, and that premise has to hold in every possible world.

(P2) Omnipotence does not include control over what a free creature would freely choose. + Omnipotence is the power to do anything that can be done. A string of words that cancels itself describes nothing that can be done, so failing to do it takes nothing away from anyone's power. (1) "Make someone freely choose" describes nothing at all. - To be made to choose is precisely not to choose freely. The phrase takes back with its second word what it granted with its first. - A square circle is not a task so difficult that only God could manage it. It is no task, and there is nothing there to attempt, to succeed at, or to fail. - Think of a master craftsman who can build anything that can be built. He cannot build a sound that is silent, and nobody counts that against his skill, because "a silent sound" names no object he failed to produce. - You might be thinking: why should God be bound by logic in the first place? If he is not, then no charge of contradiction can be brought against him either, and the objection falls apart before the reply to it does. (2) The freedom in question is the kind where the choice is genuinely yours. - Philosophers call it libertarian freedom: your choice is not settled in advance by anything outside your own will, so that in the moment you really could have gone the other way. - The difference shows up in ordinary life. A friend who sits with you in a hospital waiting room because he chose to has given you something. A paid attendant required to sit there has not, however identical the two of them look on the security camera. - Christians have disagreed for centuries about whether human freedom actually works this way, and the disagreement is an honest one on both sides. All that is needed is that libertarian freedom is possible, and nobody has shown it to be incoherent. (3) What a creature would freely do is not God's to decide. - If a choice is genuinely the creature's, then for every situation you could put that creature in, there is a truth about what they would freely do. Philosophers call these truths counterfactuals of creaturely freedom. A counterfactual is a claim about a situation that may never come up: if he were in these circumstances, he would freely do this. - Would your neighbor return a wallet he found on an empty street with two hundred dollars in it? There is an answer, and the answer is his rather than yours. You can put the wallet on the street. You cannot make the answer come out your way and still have it be his. - God knows every one of these truths perfectly. What he does not do is author them, because the moment he authored one it would stop being a free choice and start being his. (4) Some worlds are therefore possible without being ones God can produce. - A world in which every free creature always chooses well is possible in itself. Nothing about it is incoherent. - Whether it is feasible is a separate question, and a different one: not whether the world could exist, but whether God could bring it about, given how creatures would freely act. - Consider a matchmaker who knows in advance exactly who would say yes to whom. Some pairings are perfectly possible and simply not available to him, and not because he lacks skill or effort. The yes belongs to somebody else. Therefore, a world of free creatures who never sin may be genuinely possible and still not be a world God can produce, and that gap is not a gap in his power.

(P3) A perfectly good being may permit an evil for a morally sufficient reason. + Goodness does not require preventing every evil. It requires never permitting one without a morally sufficient reason: a good weighty enough to make permitting the evil the right call, and not available by any cheaper route. (1) Permitting an evil is not the same as doing it. - A parent who shoves a child off a bicycle has done something wrong. A parent who runs alongside, lets go, and lets the child fall has permitted something painful for the sake of something worth having. - Both parents produce the same scraped knee. Nobody thinks they have done the same thing, and no serious system of ethics treats them alike. - The distinction is not a loophole invented for this debate. It runs through medicine, law, and parenting, and it is why a surgeon who cuts you open is not charged with assault. (2) We already grant this reasoning to good people we know. - A mother holds her one-year-old still for a vaccination. She has the power to prevent the pain, she would far rather her child did not feel it, and she permits it anyway. We call her a good mother precisely because she does. - Change nothing but the size of the good at stake and the reasoning is the same. A coach lets a player lose a match, a teacher lets a student fail a first draft, a country lets its citizens make ruinous choices. Each permits a real harm for a real reason. - If permitting any evil at all made an agent bad, every decent person you have ever met would be disqualified along with God. - You might be thinking: the mother has no better option and God does. That is the right question to press, and it is why the kind of good matters. Courage in danger, forgiveness of a real wrong, rescue from a real disaster: each of those is a good whose description already contains the harm, so not even omnipotence can hand it over separately. (3) The reason need not be one we can see. - Your neighbor takes a chainsaw to the healthy-looking oak that shades the whole street, and you decide he is a vandal. What you did not know is that the tree service found rot at the base that would have dropped it across the road in the next storm. Your confidence was real. It was also grounded in not knowing enough. - The claim is modest: not that we know God's reason for any particular evil, and not that any proposed reason is likely, only that a sufficient reason is possible. - Ruling that out is heavier work than it looks. It means surveying every good there could ever be, in every possible world, and confirming that not one of them would justify permitting the evil in question. (4) Some goods are not on the same scale as the harms. - Christian teaching holds that the good on offer, eternal life with God, is not simply larger than earthly suffering but incommensurable with it: not a bigger number on the same scale, but a different scale altogether. - You cannot settle a debt of gratitude with money, however much of it you count out. The currency is wrong rather than insufficient. - Paul makes exactly this comparison: "this light momentary affliction is preparing for us an eternal weight of glory beyond all comparison" (2 Corinthians 4:17). He had been imprisoned, beaten, and shipwrecked three times when he wrote it (2 Corinthians 11:23-25). Therefore, perfect goodness and the permission of evil sit together without contradiction, provided a morally sufficient reason is so much as possible.

(C1) Therefore, God's power and goodness do not by themselves exclude the existence of evil. + The hidden premise has to come from somewhere. Without a bridging rule, some further claim that ties the two together, "God exists" and "evil exists" are just two sentences sitting side by side, and the oldest form of the charge supplies two candidates for the job: one taken from God's power, one taken from his goodness. (1) The bridging rule taken from omnipotence fails. - The rule says that necessarily, a God who can do anything can bring about any world he wants, including a world of free creatures who never sin. - Omnipotence has never meant the power to bring off self-cancelling descriptions, and it does not include control over choices that belong to somebody else (P2). - Unlimited power hands nobody the answers that are somebody else's to give, any more than a flawless matchmaker can supply the yes. So the rule is not a necessary truth, and where creatures are genuinely free it is not true at all. (2) The bridging rule taken from goodness fails. - The rule says that necessarily, a perfectly good being prefers every world without evil to every world with evil in it, and eliminates evil as far as it can. - No good person is held to that standard anywhere else, and goodness has never meant preventing every harm regardless of what the harm secures (P3). - A rule that strict convicts the mother at the vaccination clinic, the coach, and the surgeon along with God. That is a sign the rule is wrong rather than a sign they are all monsters. (3) Two candidates are not every candidate. - Denying each of those rules is more plausible than affirming it, which is exactly what a deductive argument asks of its premises (P1). Neither attribute yields the absence of evil, and neither does the pair of them together. - What none of that gives is a proof that no candidate could ever be found. Some third necessary truth might be waiting, and nobody should pretend otherwise. Producing it is the objector's task, since the burden sits with whoever alleges the contradiction (P1). - Closing the gap for good takes a positive step rather than another refutation: describing a single possible situation in which God and evil both exist. Therefore, God's power and goodness do not by themselves exclude evil, and the charge is still missing the necessary truth it needs.

(P4) It is possible that every world of free creatures God could make contains evil. + Two claims are compatible if there is one possible world in which both of them hold. Closing the case therefore takes not another refutation but a single scenario, however unlikely, in which God and evil exist together. (1) The challenge slides between two senses of "could." - The question behind the challenge is a fair one to feel. Why did God not simply make people who always freely choose the good? Such people are possible, God can do anything possible, and his not making them looks like a defect. - But "could have made" is doing two jobs at once. A world where you win a lottery every week for a year is perfectly possible, and it is still not something your employer could arrange for you, however much he might want to, because the winning numbers are not his to set. - What is possible in itself and what a particular agent can bring about come apart whenever part of the outcome belongs to somebody else. With free creatures, that is exactly the situation (P2). (2) A creature might go wrong in every situation God could place it in. - Suppose there is a person such that, in any circumstance where God gives them genuine freedom, they would freely do wrong at least once. Not that they must. That they would. Philosophers call this transworld depravity. - Think of a friend who, however carefully you arrange the evening, ends up saying the one thing that ruins it. Nobody is forcing his mouth open. It is simply how he would freely behave, in this restaurant and in every other restaurant you could book. - If that is true of a person, then no amount of divine rearranging produces a version of the evening where he behaves. God can change the circumstances. He cannot change the answer without changing whose answer it is. (3) It is possible that every creature God could create is like that. - The claim is not that everybody actually suffers from transworld depravity, or even that this is likely. It is that this is possible, which is a statement about what could have been rather than a guess about what is. - A reply that only has to be possible, never true, is called a defense, and this one has a name of its own: the free will defense. Think of the difference between a lawyer showing that somebody else could have done it and proving that somebody else did. The first is the lighter task by far, and it is also the harder one to knock down, because knocking it down means showing the thing is impossible rather than merely improbable (P1). - If it is possible, then there is a world in which God creates free creatures, wants no evil in it, and gets evil anyway. God and evil are both in that world, which is exactly what the charge said could not happen. (4) A world fit for free creatures may have to be a world that can hurt them. - Earthquakes and cancers are nobody's free choice, so an appeal to free choice looks as though it stops short of them. It does not, once the bar is set at possibility rather than plausibility, because there are several candidate accounts of natural evil and only one of them has to be possible. - One candidate is that a stable, law-governed world is what makes choices mean anything. If fire warmed but never burned, nobody could run into a burning house for a child, because there would be nothing to risk. The physics that makes the rescue possible is the physics that makes the fire possible. - Another, offered frankly as a possibility rather than a likelihood, is that some natural evils trace to the free choices of non-human agents. That strikes modern ears as strange, and strangeness is not impossibility. Therefore, it is possible that every world of free creatures available to God contains evil, and that God, wanting free creatures, made one of those worlds anyway.

(P5) It is possible that a world redeemed through Christ is better than any sinless world. + There is a second possibility, and it is distinctly Christian. It stands on its own feet, since it does not need free will to work the way the feasibility argument describes (P4), or to work that way at all. (1) The claim is not that evil is good. - Evil is evil on this account, without qualification and without a silver lining tucked away inside it. The claim is about what God is able to bring out of it. - Joseph says as much to the brothers who sold him into slavery: "you meant evil against me, but God meant it for good" (Genesis 50:20). He does not tell them their act was secretly fine. He tells them two things were happening at once, and only one of them was theirs. - The cross carries the same double description in Christian teaching. It is the worst thing human beings have ever done and the best thing that has ever happened to them, and the New Testament never softens either half. (2) The incarnation and the atonement presuppose a world with evil in it. - A world without sin is a world with no crucifixion, no resurrection, and no God who has personally been through what his creatures go through. Whatever else such a world contains, it does not contain that. - The Exsultet, the proclamation sung at the Easter Vigil, puts the point as a paradox and calls the fall a felix culpa, a happy fault. Not that the fault was good, but that it became the occasion of a good beyond anything the world had before. - Augustine put it in a single sentence: God judged it better to bring good out of evil than to permit no evil at all. (3) Goods of that kind are not weighed against harms on one scale. - A world containing God's own entry into human suffering, and the redemption of everyone who accepts it, may be better than any sinless world, and better not by a wide margin but by a difference in kind. - Think of a father who missed his son's childhood and later offers him a check. The problem is not the size of the check. Some things are not the sort of thing a check can carry, and adding zeroes never gets any closer. - If even one good is like that, then the claim that a good God must prefer any evil-free world to any world with evil is not a necessary truth. It is not even true. (4) This possibility is open to Christians who reject libertarian freedom. - Many Christians hold that human freedom is compatible with God's determining what people choose, a view with a long tradition running through Augustine's later work and the Reformers. They answer the charge along this line rather than through the feasibility of worlds (P4). - The redemption claim asks nothing of them. It requires no counterfactuals of creaturely freedom, no transworld depravity, and no account of what creatures would do, only that a redeemed world may be worth what it cost. - That is why two Christians who argue hard with each other about predestination can give the same answer to this charge, and both of them mean it. Therefore, it is possible that God permits the evil in this world for the sake of a good that no evil-free world could contain.

(C2) Therefore, the existence of evil is logically compatible with the existence of God. + The existence of evil does not rule out the existence of God. Two claims are strictly incompatible only if there is no possible world containing both of them (P1), and two such worlds have now been described, either one of them enough on its own: one in which free creatures would go wrong in any world God could make (P4), and one in which the good of redemption outruns what it cost (P5). Neither has to be true. Neither has to be likely. They have to be possible, and nothing offered against them has shown that they are not. That conclusion is narrow, and its narrowness is the point. It does not show that God's existence is probable given the suffering we actually see. It does not explain why any particular child got leukemia. It offers no comfort to anyone in the middle of grief, and it was never built to. What it does settle is the charge of irrationality. A Christian who looks at a cancer ward and still believes in a good God is not holding two beliefs that cannot both be true, in the way someone would be who believed a shape was both round and square. He may still be wrong, and the questions that remain are about evidence, probability, and how any of this is to be lived with. Those are harder questions in some ways, and they have to be argued on their own terms. See also: CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Intellectual PoE: External CO / PoE: Emotional PoE CO / PoE: Proof of God from Evil

On the logical problem of evil: Howard-Snyder, Daniel, ed. The Evidential Argument from Evil. Bloomington, IN: Indiana University Press, 1996. Mackie, J. L. "Evil and Omnipotence." Mind 64, no. 254 (1955): 200-212. Mackie, J. L. The Miracle of Theism: Arguments For and Against the Existence of God. Oxford: Clarendon Press, 1982. Mitchell, Basil. "Theology and Falsification." In New Essays in Philosophical Theology, edited by Antony Flew and Alasdair MacIntyre. London: SCM Press, 1955. Pike, Nelson. "Hume on Evil." Philosophical Review 72, no. 2 (1963): 180-197. Plantinga, Alvin. God and Other Minds: A Study of the Rational Justification of Belief in God. Ithaca, NY: Cornell University Press, 1967. Plantinga, Alvin. God, Freedom, and Evil. Grand Rapids: Eerdmans, 1977. Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon Press, 1974. Plantinga, Alvin. "Self-Profile." In Alvin Plantinga, Profiles 5, edited by James E. Tomberlin and Peter van Inwagen. Dordrecht: D. Reidel, 1985. Rowe, William L. "The Problem of Evil and Some Varieties of Atheism." American Philosophical Quarterly 16, no. 4 (1979): 335-341. On freedom, foreknowledge, and what creatures would do: Adams, Robert Merrihew. "Middle Knowledge and the Problem of Evil." American Philosophical Quarterly 14, no. 2 (1977): 109-117. Bergmann, Michael. "Might Counterfactuals, Transworld Untrustworthiness, and Plantinga's Free Will Defense." Faith and Philosophy 16, no. 3 (1999): 336-351. Craig, William Lane. The Only Wise God: The Compatibility of Divine Foreknowledge and Human Freedom. Grand Rapids: Baker, 1987. Craig, William Lane. Systematic Philosophical Theology, Volume IIb: On God: Excursus on Natural Theology, On the Trinity. Hoboken: Wiley-Blackwell, 2026. Flint, Thomas P. Divine Providence: The Molinist Account. Ithaca, NY: Cornell University Press, 1998. Hasker, William. God, Time, and Knowledge. Ithaca, NY: Cornell University Press, 1989. Helm, Paul. The Providence of God. Downers Grove, IL: InterVarsity Press, 1994. Howard-Snyder, Daniel, and John O'Leary-Hawthorne. "Transworld Sanctity and Plantinga's Free Will Defense." International Journal for Philosophy of Religion 44, no. 1 (1998): 1-21. Pawl, Timothy, and Kevin Timpe. "Incompatibilism, Sin, and Free Will in Heaven." Faith and Philosophy 26, no. 4 (2009): 396-417. On theodicy, redemption, and horrendous evil: Adams, Marilyn McCord. Horrendous Evils and the Goodness of God. Ithaca, NY: Cornell University Press, 1999. Augustine of Hippo. The City of God. Translated by Henry Bettenson. Harmondsworth: Penguin, 1972. Augustine of Hippo. Enchiridion on Faith, Hope, and Love. Translated by Albert C. Outler. Philadelphia: Westminster Press, 1955. Augustine of Hippo. On Free Choice of the Will. Translated by Thomas Williams. Indianapolis: Hackett, 1993. Hick, John. Evil and the God of Love. London: Macmillan, 1966. Lewis, C. S. The Problem of Pain. London: Geoffrey Bles, 1940. Plantinga, Alvin. "Supralapsarianism, or 'O Felix Culpa.'" In Christian Faith and the Problem of Evil, edited by Peter van Inwagen. Grand Rapids: Eerdmans, 2004. Stump, Eleonore. Wandering in Darkness: Narrative and the Problem of Suffering. Oxford: Clarendon Press, 2010. On the limits of human judgment about God's reasons: Bergmann, Michael, and Michael Rea. "In Defence of Sceptical Theism: A Reply to Almeida and Oppy." Australasian Journal of Philosophy 83, no. 2 (2005): 241-251. Wykstra, Stephen J. "The Humean Obstacle to Evidential Arguments from Suffering: On Avoiding the Evils of 'Appearance.'" International Journal for Philosophy of Religion 16, no. 2 (1984): 73-93. Critic sources engaged: Almeida, Michael J., and Graham Oppy. "Sceptical Theism and Evidential Arguments from Evil." Australasian Journal of Philosophy 81, no. 4 (2003): 496-516. Flew, Antony. "Theology and Falsification." In New Essays in Philosophical Theology, edited by Antony Flew and Alasdair MacIntyre, 96-99. London: SCM Press, 1955. Hume, David. Dialogues Concerning Natural Religion. Edited by Norman Kemp Smith. Indianapolis: Bobbs-Merrill, 1980. Morriston, Wes. "Is God 'Significantly Free'?" Faith and Philosophy 2, no. 3 (1985): 257-264. Oppy, Graham. Arguing about Gods. Cambridge: Cambridge University Press, 2006. Sterba, James P. Is a Good God Logically Possible? Cham: Palgrave Macmillan, 2019. Biblical citations: Genesis 50:20; 1 Samuel 23:10-13; Isaiah 1:17; Micah 6:8; Matthew 11:21; Matthew 25:31-46; Matthew 26:53; Romans 3:8; Romans 8:20-22; Romans 8:28-29; 1 Corinthians 15:14-15; 2 Corinthians 4:17; 2 Corinthians 11:23-25; Titus 1:2; Hebrews 6:18 Ancient sources: Epicurus, as reported in Lactantius, De Ira Dei 13. Exsultet (the Easter Proclamation), sung at the Easter Vigil. Latin text in the Roman Missal.
+ If God can prevent evil and will not, he is not good; if he wants to and cannot, he is not omnipotent. Either way the God of Christian belief does not exist, and evil is plainly here.
1. The dilemma quietly assumes a third thing it never argues for. A dilemma works by offering two branches, traditionally called horns, and forcing you onto one of them. This one offers "wants to prevent evil and cannot" and "can prevent evil and will not," and then treats the second branch as equivalent to "is not good." That equivalence is the whole argument, and it is the one thing never defended. A good agent who permits an evil for a sufficient reason walks straight between the horns. He can prevent it, he does not, and he is good (P3). Unless the dilemma rules that possibility out, it is not really a dilemma. It is a list of two options presented as though it were a list of all of them. 2. Ruling that possibility out is the hard part. To close the gap, the objector needs a necessary truth along the lines of "no morally sufficient reason for permitting evil is even possible." Notice how much stronger that is than what anyone usually says: - Not "we cannot think of one." - Not "the reasons proposed so far seem thin." - That no such reason could exist, in any possible world, under any arrangement whatever. That is an enormous claim, and nothing less than that would make the dilemma bite. It is the same claim every version of the charge has to establish (P1). The dilemma reaches us through Lactantius, writing in the early fourth century, who credits it to Epicurus. Whether Epicurus actually framed it is disputed among historians, and nothing turns on the answer. Hume gave it its modern life, putting it in the mouth of his skeptic Philo and remarking that the old questions stood unanswered. Neither the ancient version nor the modern one so much as attempts the premise that would close the gap. 3. The Christian answer to this dilemma is sixteen centuries old. Augustine worked it out in the late fourth century, and his answer runs roughly as follows: - Free will is itself a great good, given by God, and not a regrettable side effect of something else he wanted. - A world of creatures who genuinely love God and one another is better than a world of creatures who cannot love at all, because love that cannot be withheld is not love. - Genuine freedom carries the real possibility of moral evil. Remove the possibility and you have removed the freedom, not protected it. The challenge was already old and in wide circulation when Augustine answered it, and his answer has been on the table ever since. It is not a recent discovery that Christians somehow failed to notice. 4. Once possibility is conceded, the dilemma turns into a different argument. Grant only that some morally sufficient reason is possible, and the dilemma no longer shows that theism, belief in an all-powerful and perfectly good God, is self-contradictory. What is left is a real and serious complaint: that no proposed reason seems adequate to the suffering we actually see. That complaint deserves to be taken seriously, and many people who press the dilemma mean it rather than the logical version. But it is an argument about probability, and it has to be argued on those terms rather than smuggled in under a charge of contradiction. See also: CO / PoE: Intellectual PoE: Probabilistic
+ God is omnipotent, so he could have made free creatures who always freely choose good. Nothing about such a creature is contradictory, and refusing to make them is either a limit on his power or a defect in his goodness.
The strongest published version of this objection is J. L. Mackie's, from 1955, and it is sharper than the popular version. Mackie grants that free will is valuable. His point is that God faced no trade-off at all: if it is possible for a person to freely choose the good on one occasion, then it is possible for a person to freely choose the good on every occasion, and an omnipotent being could have made people like that. Refusing to is either a limit on power or a defect in goodness. 1. One word, "could," is doing two different jobs. Mackie is right that a world of free creatures who never sin is logically possible, and that is not in dispute. The question he passes over is whether such a world is one God could bring about, given that the choices in it are made by the creatures rather than by him. Those are two different questions, and the objection treats them as one (P2). Compare a trial. It is entirely possible that every juror independently reaches the correct verdict. It does not follow that the judge can arrange for that outcome, and if he does arrange it, they are no longer jurors reaching a verdict. 2. The reply needed is far weaker than the objection assumes. The objection usually assumes the answer has to be that every possible creature in every possible world would sin. That would be a wild claim, and nobody makes it. What the free will defense actually claims is this: - It is possible that every creature God could create is such that, in any circumstance where they are genuinely free, they would go wrong at least once. - That is offered as a possibility and nothing more. Whether it is true, and whether it is likely, are questions the reply never has to answer. - Possibility is enough because of what was charged. If God and evil cannot both exist, then describing one coherent situation in which they do ends the matter (P4). 3. The objection rests on a claim about every possible world. For the objection to do logical work, it has to hold that necessarily, an omnipotent God can bring about any logically possible world containing free creatures. Philosophers call a claim of that sort modal, because it is about what must be and what could be rather than about how things happen to stand. It is precisely what the believer denies, and precisely what the objection assumes. Notice the asymmetry. The objector needs a universal claim to be necessarily true. The defender needs a single scenario to be barely possible. Those are not comparable burdens, and the heavier one belongs to whoever alleges the contradiction (P1). 4. The charge of strict inconsistency was given up by its own side. Mackie himself, writing twenty-seven years later, conceded that the problem of evil does not, after all, show that the central doctrines of theism are logically inconsistent with one another. He did not become a theist. He argued instead that libertarian freedom is of doubtful coherence and doubtful worth, and that the evidence still tells against God, which is a different case on different ground. The concession was not his alone. William Rowe, one of the most careful atheist philosophers of religion of the last century, opened his 1979 paper on the problem of evil by setting the strict inconsistency charge aside: "No one, I think, has succeeded in establishing such an extravagant claim." He then built his own argument on probability instead, and that is where the serious debate has been ever since. Attempts to revive the logical version continue, and the most serious recent one is James Sterba's (D4).
+ Even if free will explains moral evil, it cannot explain natural evil. Earthquakes, cancers, and animal suffering are nobody's free choice, yet they cause most of the world's suffering.
1. The bar is possibility, not likelihood. The objection asks for an explanation that seems likely, which is a fair thing to ask in general. It is not what a charge of contradiction requires an answer to. Two words are doing all the work here, and they are worth keeping apart: - A defense is a logically possible account of how God and evil could both exist. It need not be true, or even probable. - A theodicy is a proposed true account of why God actually permits the evil there is. It aims much higher, and it is what most people actually want. Against a charge that God and evil cannot both exist, a defense is exactly the right size of reply. Anything beyond that is generous rather than required. 2. A stable natural order may be required for free choice to mean anything. Consider what a world without natural harm would have to be like: - Fire that warmed but never burned. - Water that carried boats but never drowned. - Cliffs you could step off and float down from. In such a world nobody could choose to pull a child out of a burning building, because nothing burns. Nobody could risk anything, protect anyone, or act with courage, because there would be nothing to be courageous about. The regularities that let a person predict the consequences of an action are the same regularities that produce earthquakes and let cells go wrong. The sharpest reply is that God could keep the regularities and quietly suspend them whenever they were about to hurt somebody. Perhaps he could. But then the regularities are no longer what anyone is relying on, and a world where consequences arrive only when they happen to be harmless is a world where nothing you do ever puts anything at stake. Whether that trade-off is real is genuinely debated. That it is possible is enough to answer a charge of contradiction. 3. One proposal traces natural evil to non-human free agents. Some defenders hold that natural evils could come from the free choices of non-human moral agents. This strikes most modern readers as strange, and its own proponents have said as much in print rather than pretending otherwise. Strangeness is not the issue, though. For the objection to succeed it has to show the proposal is impossible, not merely unfashionable, and nobody has done that. It is hard to see what would. 4. The objection has to close every other candidate too. Several candidates are already on the table: - That natural evils serve the formation of character that could not develop in a painless world. - That the natural order suffers real consequences from the moral fall of free creatures. Paul writes that creation was subjected to futility and groans for release (Romans 8:20-22). How much that passage settles is debated among interpreters, but Christians have read a connection there from early on. - That a world capable of producing and supporting embodied creatures at all is necessarily a world with tectonic plates, cell division, and everything that can go wrong with both. Any one of these, if merely possible, defeats the logical form of the objection. What the objection genuinely establishes is something narrower and more painful: that the amount and the distribution of natural suffering look nothing like what you would expect from a good God. That is a claim about probability and belongs in that debate. See also: CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Animal Suffering
+ James Sterba's revival of the logical version says a good God would prevent, rather than permit, the horrendous consequences of immoral actions whenever that can easily be done. Such evils plainly occur, so no such God exists.
Sterba's 2019 book is the most serious recent attempt to restore the logical version, and it deserves a direct answer. His argument runs from two moral requirements he takes to hold without exception: - Prevent, rather than permit, the significant and especially the horrendous evil consequences of immoral actions, whenever that can be done without violating anyone's rights. - Do not permit those consequences in order to give victims goods or wrongdoers opportunities, when there are countless other ways of providing both. To God he applies the traditional principle that one may never do evil so that good may come (Romans 3:8). What makes the argument worth engaging is that it arrives with an answer to the free will defense already prepared: God need not touch anybody's will. He can leave every immoral choice free and intact and block only the external consequence, the blow landing, the bullet arriving, the camp gate closing. Since he does not, no good God exists. 1. The principle Sterba invokes governs doing evil, not permitting it. "Never do evil that good may come" is a rule about your own acts. It has never been read as "never permit an evil you could prevent," because that reading would condemn every parent, surgeon, and legislator who has ever lived (P3). Sterba needs the two to be morally equivalent in God's case, and he asserts that parallel more than he argues for it. The distinction he wants dropped is one that moral philosophy has held onto, for good reasons, everywhere else it applies. 2. Blocking the worst consequences takes the goods away with the evils. This is where Sterba is strongest and has to be met head on. So grant him the whole request. God leaves the will alone and stands as the wall that the worst acts run into. Now consider what such a world contains, and what it has lost: - Nobody can be murdered, tortured, or enslaved, so nobody can be rescued from any of those things either. - No wrong of that size is ever forgiven, because none is ever done. - Courage, protection, and mercy all need a real victim and a real wound, and there are none left to have. Then there is the sliding scale. Block the horrendous, and whatever is next worst becomes the worst thing that happens, and the argument applies to that with exactly the same force. Run it again and again and you arrive at a world where no choice of yours can seriously harm anyone. That is also a world where no choice of yours seriously matters. 3. The requirements would have to be necessary truths. For the logical version, Sterba's moral requirements cannot merely be correct. They have to hold in every possible world without exception, the way "no bachelor is married" does (P1). They are substantive moral principles about which careful people disagree, and Sterba himself allows that the principle he builds on admits exceptions. A rule with acknowledged exceptions is not the kind of premise a contradiction can be built out of. It can support a strong evidential case, which is a different argument carrying a different burden. Notice too what his own proposal leaves standing. He does not ask God to prevent immoral choices, only their worst consequences, so the world he recommends still has wicked acts in it and a good God over them. On his own account, then, a good God may permit some evil. Once that much is granted the question becomes how much, and how much is a question for evidence rather than for a claim about every possible world. 4. A world without those evils may not be one God could bring about. Sterba's argument assumes that if a world without the relevant evils is logically possible, God can produce it. That assumption is exactly what the feasibility reply denies (P2, P4). Sterba's God is an agent with unlimited options. The God of the free will defense has unlimited power and a limited menu, because part of what fills the menu is what free creatures would do, and that part is not his to write. 5. The verdict on compensation assumes a scale the Christian denies. The argument treats horrendous evils as beyond any possible compensating good. That verdict assumes the goods on offer are the sort of thing that could be weighed against suffering and found lighter. The Christian claim is that they are not on that scale at all (P3, P5). A person who has come through unimaginable suffering into eternal communion with God is not someone whose ledger has finally been balanced. It is not a ledger. If even one good of that kind is possible, the exceptionless requirements have an exception. Sterba has defended the argument vigorously since 2019 and the exchange is still running. Its real force is as a probability claim about horrendous evils in particular, which is a different and harder question. See also: CO / PoE: Intellectual PoE: Probabilistic
+ A "defense" shows at most that we cannot see an incoherence, which is not the same as showing a scenario is possible. And bare possibility explains nothing: what is needed is a theodicy, a true account of why God actually permits suffering.
1. Showing possibility is how consistency has always been shown. Grant the distinction first, because it is a real one. Failing to find a contradiction in a description is not the same as the description being genuinely possible, and philosophers who work carefully on what is and is not possible are right to press the difference. Now look at what that standard demands of both sides. The way to show that two claims are consistent is to describe a situation in which both of them hold, and whoever answers that the situation is itself impossible has just taken on the job of showing it. The standard cannot be run in one direction only. The objection needs a world of free creatures who always choose well to be both possible and available to God, and it has no better warrant for that than the reply has for its scenario. Being able to imagine something is good evidence that it is possible, though not proof. When imagination misleads us, the reason is usually that the thing has a hidden nature nobody knew about. You can picture a liquid that fills the rivers and behaves in every way like water and is not H2O, and there could be no such liquid, because water simply is H2O, which was a discovery rather than a definition. So the question for the objector is what the hidden nature would be in this case, the one that makes a morally sufficient reason for permitting evil impossible. Nobody has said. 2. What a reply has to accomplish depends on what was claimed. The charge was that God and evil are strictly incompatible, that no possible world contains both. One such world, described coherently, refutes that completely, with nothing left over. Suppose someone claims that no bachelor could own a house. Showing them one possible house-owning bachelor settles it, and they cannot then complain that you have failed to explain property markets. Demanding a theodicy in answer to a charge of contradiction is changing the subject after the point has been conceded. 3. The complaint is right about what a defense does not do. A defense does not explain any actual evil. It does not tell a grieving parent why. That is a real limitation, and Christians should say so plainly rather than gloss over it. Anyone who has been handed a possible-worlds argument in a hospital corridor knows exactly what is wrong with it. The mistake is not in noticing that. It is in concluding that the reasoning has therefore failed at the job it was doing. 4. Christianity offers theodicies as well as defenses. The tradition is not short of proposed accounts of why God actually permits evil: - Augustine's free will theodicy: evil is the misuse of a good gift, not a thing God made. - The soul-making tradition running back to Irenaeus: this world is a place where persons are formed, not a resort where they are kept comfortable. - The redemptive line: that God permits evil because he intends to defeat it from the inside, at his own cost, and to bring out of it a good that no evil-free world contains (P5). These are the arguments that engage the questions about evidence and about grief, and Christians have always offered both kinds of reply. 5. The complaint that possibility "explains nothing" has already shifted the ground. The moment the objection asks for an explanation of the evil we observe rather than a demonstration of consistency, it has moved from logic to evidence. That is a legitimate move, and it makes for a stronger argument. It is also a concession. Most philosophers who once pressed the logical version have made it, and the substantive debate has been on evidential ground for decades. See also: CO / PoE: Emotional PoE CO / PoE: Intellectual PoE: Probabilistic Philosophy / Logic: Deductive Arguments
+ The torture of a child, or a genocide, is so appalling that no possible good could outweigh it. God's existence is incompatible with horrors of that kind and degree.
1. These evils are exactly as terrible as the objection says. Nobody should answer this briskly. The examples are real, the moral revulsion behind them is correct, and a person who is not disturbed by them has misunderstood something worse than a philosophical argument. The question is not whether the horror is real. It is what would have to be shown to turn that horror into a proof. 2. Narrowing the charge to horrors raises the burden rather than lowering it. To run the logical version on horrendous evils in particular, the objector needs a premise like "necessarily, no morally sufficient reason exists for permitting evils of this kind or degree." Every word of that has to hold in every possible world. That is a heavier burden than the one that already failed, not a lighter one. The original premise had only to cover evil in general. This one has to survey every possible good, in every possible world, and confirm that not one of them would justify permitting these particular evils. 3. Seeing no justifying good is not seeing that there is none. A visitor watching his first chess match sees a grandmaster hand over his queen, and he knows a blunder when he sees one. Five moves later the sacrifice wins the game. He was not being stupid. He was being confident on the basis of what he could see, which happened to be one move out of six. The analogy does not prove God has a reason. What it shows is why the leap from "I can see no justifying good" to "no justifying good is possible" is the step that has to be defended, and it is not defended by restating how bad the evil is. 4. The Christian answer to horrors is not a calculation. The distinctively Christian reply is not that the sums come out positive if you count carefully enough. Philosophers who work on horrendous evils distinguish two things a good can do to an evil. It can balance the evil off, sitting elsewhere on the scale and outweighing it. Or it can defeat the evil, taking it up into a larger whole so that its meaning changes from the inside. A scar from a rescue is not an injury that got outweighed by something else. It is an injury that became something else. What Christianity claims is defeat rather than balance. God does not stand outside the horror running a ledger. He entered it, was tortured to death by an occupying power, and rose. None of that explains any particular horror. It does mean that the good on offer, union with a God who has been through this, is not the kind of thing that could be outweighed by suffering, because it is not on that scale (P5). If that is so much as possible, the narrowed charge fails for the same reason the original one did. It needs a claim about every possible world that nobody has established. 5. As a probability argument, this is among the strongest cases against theism there is. It should not be filed away as though answering the logical form had disposed of it. Showing that a good God and a genocide are not strictly contradictory does nothing at all to show that genocides are what you would expect if a good God existed. See also: CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Emotional PoE
+ Christians say heaven is full of free people who never sin, so a world of free creatures who always choose rightly is not just possible but actual. The free will defense refutes itself.
This is the sharpest internal objection to the free will defense, and it has been pressed in the philosophical literature for decades. If the people in heaven are both free and impeccable, meaning unable to sin at all, then God evidently can produce free creatures who never go wrong, and the claim that he might have been unable to looks false by Christianity's own lights. 1. Nobody in heaven was created there. Those in heaven did not arrive as finished products. On every orthodox account they got there through a life, with real choices in it, including the choice to accept what was offered them. God did not create confirmed saints. He created people, and some of them became confirmed saints. So the objection has to claim something stronger than it usually states: not that a world of free sinless creatures is possible, but that God could have started with one and skipped the road entirely. That is a different claim, and it does not follow from anything Christians say about heaven. 2. Christians hold more than one account of heavenly freedom. Each of them blocks the objection, and all three are live positions among Christian philosophers today: - On the first, the freedom of the blessed, those already in heaven, is real but shaped by the character they freely formed, so that sin is no longer a live option for them in the way it once was. They are like a lifelong honest man who genuinely cannot bring himself to steal, and whose inability is an achievement rather than a restriction. - On the second, the inability to sin is derived rather than built in. It is granted to those who freely came to want it, so the blessed are answerable for the settled character they now have, in the way a man is answerable for habits he spent thirty years forming. A gift given in response to a free choice is not a substitute for the choice. - On the third, going back to Augustine, the blessed reach a state of non posse peccare, not able to sin, which he explicitly contrasts with the posse non peccare, able not to sin, of the unfallen creature (City of God 22.30). Adam could avoid sinning; the redeemed cannot sin at all. Augustine treats these as two different conditions rather than one, and he says the last is the greater freedom rather than the smaller. Not one of those accounts licenses the step from "heaven exists" to "God could have made heaven directly." 3. A world of people who never chose is not the same world as heaven. Even granting a world of free creatures created already confirmed in goodness, it would not be heaven. Heaven, on the Christian account, is populated by people with histories: people who were tempted, or who suffered, or who were forgiven for something specific. The community is constituted by what its members came through. Think of the difference between a marriage of fifty years and an identical couple created five minutes ago with matching memories installed. The second is not a shortcut to the first. It is a different thing entirely. 4. The redemption route never depended on any of this. Even if all three accounts of heavenly freedom failed, the second possibility would stand untouched. A world redeemed through Christ may be better than any sinless world (P5), and that claim says nothing about counterfactuals, transworld depravity, or the mechanics of heavenly freedom. Heaven is not a counterexample to it. Heaven is what it results in.
+ The defense rests on "counterfactuals of creaturely freedom." If a choice is genuinely undetermined, nothing makes it true what someone would freely do in a situation that never happens, so those statements are not true at all.
This is the grounding objection, pressed by Robert Adams and developed at length by William Hasker, and it is the strongest technical attack on the machinery the free will defense uses. Both are Christian philosophers arguing inside the faith rather than against it, and the objection is not obviously wrong. 1. Everyone treats these statements as true all day long. Consider how ordinary they are: - You believe your friend would refuse a bribe. - A hiring manager believes a candidate would handle a difficult client well. - A parent believes their teenager would call if the car broke down. Those are exactly the statements the objection says are neither true nor false, and almost every practical decision involving another person runs on them. If none of them were ever true, working out how anyone would respond to anything would be as senseless as predicting a coin toss. Nobody reasons that way, including the philosophers pressing this objection. Scripture speaks this way as well. David asks whether Saul will come down to Keilah and whether the men of Keilah will hand him over, and God answers that they will; David leaves, and neither thing happens (1 Samuel 23:10-13). Jesus says that if the miracles worked in Chorazin and Bethsaida had been worked in Tyre and Sidon, those cities would have repented long ago (Matthew 11:21). Both passages are disputed. Defenders of the objection read them as claims about what was likely, or about a settled intention, rather than about what would certainly have been done. Neither passage is decisive on its own, though the plain sense of each is a truth about what free people would have done in a situation that never arrived. 2. The demand that every truth have a truth-maker is itself contested. The objection assumes what philosophers call truth-maker maximalism. A truth-maker is whatever in the world makes a statement true: your keys lying on the counter is the truth-maker for "my keys are on the counter." Maximalism says every true statement must have one, and that is a live dispute among philosophers rather than a settled result. The standard counterexamples are ordinary enough. "There are no unicorns" is true, and there is no unicorn-absence sitting somewhere making it true. Negative truths, truths about the past, and truths about what merely could have been all press the same difficulty, and whoever wants to rule out counterfactuals of creaturely freedom on truth-maker grounds has to explain why these survive. The difficulty reaches truths about God's own free acts as well. Jesus tells Peter that if he asked, his Father would at once send him more than twelve legions of angels (Matthew 26:53). Nothing outside a free divine choice settles what that choice would be either, so the same reasoning cancels that truth along with the creaturely ones. 3. To defeat the defense, the objection has to be necessarily true. This is the decisive point. The defense claims only that it is possible that the relevant counterfactuals are true, and to defeat a possibility claim the objector has to show impossibility (P1). So the grounding objection would have to establish, not that counterfactuals of creaturely freedom lack truth-makers as a matter of fact, but that they could not possibly have them in any possible world. That is a far stronger claim than its defenders typically argue for, and it puts the objection in exactly the position that sank the original charge: needing a necessary truth that has not been supplied. 4. The defense can be rebuilt without those statements at all. Suppose the objection wins outright and no counterfactual of creaturely freedom is ever true. The reply can then be run on weaker statements the objection does not touch: not "if placed in these circumstances, he would go wrong," but "if placed in these circumstances, he might go wrong." That "might" is not a confession of ignorance. It says the going wrong is a live possibility which the circumstances leave open, whether or not anybody happens to know it. That weaker claim carries the whole load. If, for every arrangement God could make, the creatures in it might go wrong, then no arrangement guarantees a sinless world of free creatures, and there is a possible world in which God wants one, does not get one, and creates free creatures anyway. The reformulation has been worked out in detail in print, and the objection the stronger version has to fight leaves it untouched. 5. The redemption route uses none of this machinery. No counterfactual of creaturely freedom appears anywhere in the claim that a world redeemed through Christ may be better than any world without sin (P5). If the grounding objection succeeded completely tomorrow, that possibility would stand exactly where it stands now, and it is enough on its own.
+ The whole reply rests on libertarian free will, which most philosophers reject and many Christians reject too. If freedom just means acting on your own desires without compulsion, God could have determined that free creatures always desire the good.
1. Christians have divided over this for centuries. One tradition, running through much of Augustine's later work and the Reformers, holds that human freedom is compatible with God's determining what people choose. Philosophers call that view compatibilism: on it, you act freely whenever you do what you actually want, even if you never chose to want it, the way a man who stays with his family because he loves them acts freely though he never decided to love them. The other tradition holds that a free choice has to be undetermined by anything outside the agent's own will. It has defenders in every branch of the church, and the Molinist and Arminian streams state it most explicitly. Both views are held by orthodox Christians in good standing, and neither side has to win for the charge of contradiction to fail. 2. The objection needs libertarian freedom to be impossible, not merely false. Even granting that compatibilism is the majority view among philosophers, and it is, a head count establishes nothing at all about possibility. The free will defense claims that it is possible that creatures have libertarian freedom and that every one God could create would go wrong. To defeat that, the objection has to show libertarian freedom is incoherent in every possible world. Libertarian freedom is a minority position, defended by serious philosophers with detailed accounts, and nobody has shown it to be impossible. 3. Undetermined is not the same as random. The strongest form of the objection presses a dilemma. Either a choice follows from the agent's character and desires, in which case God could simply have given everyone a good character, or it does not, in which case it is a twitch, a piece of noise, and worth nothing. Neither horn describes what libertarians actually hold. On their account the choice is not uncaused. It is caused by the agent: not a prior state pushing from behind, and not chance, but a person deciding, with reasons that pull without settling the matter. A juror who weighs the evidence for two days and then votes is neither being shoved nor flipping a coin, and describing her as one or the other misses what she was doing the whole time. Whether that account is finally correct is a live argument with able people on both sides. Whether it is impossible is a different question, and nobody has answered it in the objection's favor. 4. The compatibilist has a different answer, not a missing one. A Christian who rejects libertarian freedom does not thereby lose the reply. He answers along a different line: - God permits evil for the sake of goods that could not otherwise obtain, which requires no account of undetermined choice at all (P3). - A world redeemed through Christ may be better than any sinless world, and that claim is fully available on a compatibilist account of freedom (P5). - The judgment that God has a morally sufficient reason for what he permits is common ground between the two traditions, however they explain the mechanics. 5. Either route defeats a charge of contradiction. Whichever account of freedom turns out to be correct, there is a world where God and evil are both present. The libertarian reaches it through the feasibility of worlds (P4); the compatibilist reaches it through the value of redemption (P5). The objection would have to close both doors, and it argues about only one of them.
+ If every evil serves a greater good, then no evil is really pointless and there is no reason to fight any of it: curing a disease might be thwarting God's plan. And calling the fall a "happy fault" makes the Holocaust a good thing.
1. Bringing good out of an evil is not calling the evil good. The two are opposites, and only the first is claimed. Joseph does not tell his brothers that selling him into slavery turned out fine after all. He tells them they meant evil and God meant good, and both halves of that sentence are true at once (Genesis 50:20). Christianity's central image has the same shape. The crucifixion is described as the greatest crime ever committed and the greatest good ever done, in the same books and sometimes in the same paragraph, and the New Testament never resolves the tension by softening the first half. Paul's most quoted line on this gets misquoted constantly. He writes that God works all things together for good for those who love him, and the very next verse says what the good is: being conformed to the image of his Son (Romans 8:28-29). It is a promise that God will bring his people through and make something of them, not a claim that each event was secretly a blessing in disguise. 2. Christians are commanded to fight evil, not to calculate around it. The objection assumes that a believer's job is to guess at God's permissions and act accordingly. It is not. It is to obey commands that are entirely explicit about this: - "Learn to do good; seek justice, correct oppression; bring justice to the fatherless, plead the widow’s cause" (Isaiah 1:17). - "To do justice, and to love kindness, and to walk humbly with your God" (Micah 6:8). - Jesus makes feeding the hungry, clothing the naked, and visiting the sick and the imprisoned the criterion of judgment itself (Matthew 25:31-46). The historical record follows the commands rather than the objection's logic. Hospitals, orphanages, famine relief, and abolition movements grew out of this theology rather than in spite of it. If greater-good reasoning paralyzed people, that is not what the last two thousand years would look like. 3. Permission is not approval. A surgeon permits post-operative pain because the operation was worth it. When the nurse comes round with morphine, she is not thwarting the surgeon. Relieving that pain was part of the point all along. God's permitting an evil says nothing whatever about whether you should relieve it, and in Christian teaching the relief is frequently the very good he permitted it for. The one thing a believer is never licensed to do is cause an evil on the theory that God will redeem it, which Paul rejects explicitly and with some heat (Romans 3:8). 4. The felix culpa phrase means something narrower than the objection hears. The line comes from the Exsultet, the proclamation sung at the Easter Vigil, and it is sung about the fall rather than about any particular atrocity. It says that the fault became the occasion of the incarnation, so that what came out of it was greater than what was there before. It does not say the fault was a good deed, and it has never been used to justify anyone's crime. Applying it to the Holocaust in the way the objection suggests turns it inside out. The Christian claim about the camps is that they were evil without remainder, that God was present in them as one of the victims rather than as their author, and that his answer to them is resurrection rather than explanation. 5. Even at its strongest, this is a moral complaint rather than a logical one. Suppose someone finds all of that unsatisfying. The charge on the table was that God and evil cannot both exist. Distaste for the shape of a reply, however sincere, does not restore a contradiction. It belongs to a different argument, and that argument is worth having on its own terms. See also: CO / PoE: Emotional PoE CO / PoE: Proof of God from Evil
+ Among all the possible people God could have made, surely some would freely have chosen well, and he could have created only those. That every last one would go wrong is a fantastically improbable guess, not a defense.
The strongest form of this objection grants the feasibility point entirely and goes after the numbers instead. There are indefinitely many possible people. For the free will defense to work, every single one of them has to be such that in every circumstance God could place them in, they would go wrong at least once. The odds against that look astronomical, and leaning on a possibility that remote can look like a trick. 1. Only impossibility is on trial. The charge was that God and evil cannot both be true, the way a shape cannot be both round and square (P1). It was not the charge that they probably are not both true, nor that the story a Christian tells to reconcile them is a likely one. Against a charge like that, a scenario with a probability of one in a trillion works exactly as well as a scenario with a probability of one half. If the objection now wants to argue from improbability, it has quietly changed arguments, and the argument it has changed to is the evidential one. 2. God cannot pick the good ones out of a catalogue. The picture behind the objection is a lineup of possible people, each with a fixed record printed on the back, from which God selects the ones who come out clean. That is not the situation. What God has to work with are creaturely essences: the complete individual nature of a possible person, everything that would make her that woman and not her twin sister. Whether a given essence would go wrong in some circumstance or other is not a dial he can turn. What a person would freely do depends on the circumstances they are in, and their circumstances include the other people. Put a man in a world with different neighbors, a different marriage, a different war, and he is facing different choices with different answers. There is no shelf of pre-tested individuals waiting to be lifted off and assembled into a world, because the testing and the assembling are the same operation. Think of a dinner party. You know each of your eight friends well, and you know exactly how each of them behaves one to one. That still does not hand you a seating chart guaranteed to produce a pleasant evening, because what each of them would do depends on who else is at the table. 3. The improbability is not something anyone can calculate. Talk of astronomical odds assumes there is some distribution over possible people and how each of them would behave. Nobody has one. We have no access to that space, no sample from it, and no idea what proportion of it has any property whatever. "Surely some of them would have chosen well" is a hunch about a domain where nobody has ever taken a measurement. It may well be right. It is not the sort of thing that establishes a necessary truth, and a necessary truth is what the charge needs. The intuition also runs the other way once the choices are counted. Transworld depravity does not require a person to be wicked, or even unusual. It requires only that in any life God could give them, they would go wrong at least once, across every decision in that life. Set that against the number of small moral decisions in one ordinary morning, and "never once, in any circumstance, in any life he could be given" starts to look like the extraordinary description. "Somewhere along the way he slips" is the ordinary one. 4. Creating only those who would never go wrong has a price. Suppose God could have created only those who would always freely choose well, and suppose there are some. Then everybody who would have gone wrong is simply never made, which includes, on any honest reckoning, you and nearly everyone you love. That is not a knockdown reply and should not be dressed up as one. It is a reminder that "God should have made only the good ones" is a proposal with a cost, and that the cost is paid by the people left out of it. See also: CO / PoE: Intellectual PoE: Probabilistic
+ God cannot do wrong and is still free and praiseworthy. So freedom does not require the ability to sin, and God could have made creatures free in exactly the way he is.
Wes Morriston put this question to the free will defense in 1985, and the pressure has not let up since. On classical theism God is essentially good: sinning is not something he heroically refrains from, it is something he cannot do. Yet nobody wants to say that God is therefore unfree, or that his goodness is not to his credit. If free and unable to sin is a coherent combination, the assumption at the center of the free will defense is false, because God could have made creatures like that from the start. 1. The objection is right about God. God cannot lie and cannot break a promise, and Scripture states it flatly rather than apologizing for it (Titus 1:2, Hebrews 6:18). The tradition has never treated this as a limitation on him, and it will not do to start treating it as one now, just because that would be convenient. So the reply cannot be that an impeccable being is not really free. It has to be about where the impeccability comes from. 2. God's goodness is nobody's gift to him. God is not good because something outside him made him that way. He is the standard, and his goodness is as underived as his existence, because there was nothing prior to him that could have set it. A creature is in the opposite position. If a created being is unable to sin from its first moment, that inability was put there by its maker, and the old question comes straight back: is a will manufactured to want only the right things the will of an agent making choices, or of an instrument working correctly (P2)? Think of the difference between an honest man and a till that cannot come up short. Both hand over the right amount every time. Only one of them is doing anything. 3. God's freedom may not be the kind of freedom at issue. The free will defense turns on what philosophers call significant freedom: freedom with respect to an action that it would be wrong to perform and right to refrain from, the freedom a student has when the answer key is on the desk and the proctor has stepped out. A being who never once faces such an action can be entirely free without being free in that particular way. On many orthodox accounts God's freedom is exactly like that, a freedom among goods rather than a freedom between good and evil. He chooses whether to create and what to create, and he never faces a live option to do wrong, because there is nothing in him that could want it. If that is right, the parallel breaks at just the place where the objection needs it to hold. What creatures were given is significant freedom, the kind that runs between good and evil. The objection has to show that God's kind, freedom among goods, handed to beings who are not themselves the standard of goodness, would still be the thing that made freedom worth giving. 4. Even granting the objection, the value question is still open. Suppose creatures could be created both free and unable to sin. It would not follow that such a world is the best one available, or that a good God must prefer it to a world where creatures reach that condition through a history he redeems (P5). Christian teaching does say that God eventually produces free creatures who cannot sin. It says he does it at the end of a story rather than instead of one (D7).
+ If we cannot tell whether an evil has a justifying reason, we cannot tell anything about God's goodness, and we lose our grounds for interfering with suffering at all. The move that rescues theism destroys moral reasoning.
The position under attack has a name and a serious literature behind it. Skeptical theism holds that our failure to see a justifying reason for an evil is weak evidence that there is none, because the goods there could be, and the connections between them, may lie far beyond anything we can survey. Almeida and Oppy have pressed the standard reply: if that is so, then for all you know, stopping a mugging thwarts an outweighing good you cannot see, and you have no business stopping it. Worse, if God's goodness is compatible with anything we might observe, then observation tells us nothing about him, including anything that would support trusting a revelation. 1. The claim actually in play is far narrower than skeptical theism. What a compatibility argument needs is not that we are generally poor judges of good and evil. It needs one thing only: that a justifying good is still possible. Closing that off would mean surveying every good there could ever be, in every possible world, and confirming that not one of them would justify permitting the evil in question (P3). That is a statement about the reach of our inventory of possible goods. It is not a statement about our ability to tell cruelty from kindness. A person can be an excellent judge of what is happening in front of him and still never have taken a census of the possible. Christian philosophers themselves divide over how much of skeptical theism to accept, and some would rather argue for particular reasons than plead the limits of human sight. That disagreement is real and it is unresolved. The narrow claim in play does not wait on it. 2. Obligations do not come from guessing at God's permissions. The paralysis argument assumes that a believer decides what to do by estimating which outcomes God has planned. Nobody does this, and Christian ethics forbids it. You act on what you are commanded and on the good in front of you (Micah 6:8, Matthew 25:31-46). A premise is hiding in the argument, and it is not one most people hold. The paralysis follows only if the right act is whatever produces the best total outcome, so that any doubt about total outcomes suspends the duty. Take that premise away and the duty simply stands. A surgeon operates on the patient in front of her without first working out what the patient's grandchildren will do. There is a further reason the paralysis never arrives. If God permits an evil for the sake of a good, one of those goods may be precisely somebody's stopping it. Pulling a child out of the road is never known to be thwarting anything, because a rescue is exactly the sort of thing a justifying reason would be made of. 3. The charge needs far more moral vision than the reply does. The objection says human beings are not in a position to judge whether an evil is justified. Notice who needs that judgment most. A charge of contradiction requires knowing that no possible good could justify permitting any evil at all, in any possible world, which is a vastly more ambitious survey than anything the reply asks for. If human moral vision is good enough to establish that, it is more than good enough for the modest claim on the other side. If it is not good enough for either, the charge fails before the reply does. 4. We do not learn God's goodness by reading the world. On the Christian account we do not work out what God's goodness amounts to by inference from the state of the world. We learn it from what he has said and from what he did in Christ, which is why "God is good" comes with obligations, expectations, and predictions attached rather than working as an all-purpose excuse. That is also why the conclusion stays narrow. Consistency is shown, and nothing further. Whether the world we actually see counts as evidence against a good God is a live question, and it is settled with evidence rather than with claims about every possible world. See also: CO / PoE: Intellectual PoE: Probabilistic
+ If God's goodness survives any evil whatever, because there might always be a reason we cannot see, then calling him good says nothing at all. The claim has been qualified to death.
This is Antony Flew's challenge, framed with a parable he took over and sharpened. Two explorers find a clearing in the jungle and one of them insists a gardener tends it. No gardener is ever seen, so the gardener is invisible. Nothing is heard, so he is silent. The dogs never catch a scent, the electric fence never registers a touch, and by the end the gardener differs in no way at all from no gardener. Flew's question to the believer was simple: what would have to happen for you to give the claim up? If the answer is nothing, the claim was never saying anything. 1. "Morally sufficient" is a demanding standard, not a blank check. A sufficient reason is not just any reason. It has to be a good weighty enough that permitting the evil was the right call, and not obtainable by some cheaper route. Most proposed reasons fail that test on the spot: "to teach the parents a lesson" fails, "because he could" fails, "for the sake of a slightly more interesting universe" fails. Saying such a reason is possible is not saying that reasons come cheap. It is saying that nobody is in a position to rule out the whole category (P3). 2. Christian goodness has content that could have been violated. Ask what would count against it and there are answers ready: - A God who lied would not be good, and Scripture says flatly that he cannot (Titus 1:2). - A God who broke his covenant would not be good. - A God who condemned the innocent, ignored the widow and the orphan, or shrugged at injustice would not be the God of Isaiah 1:17 and Micah 6:8. The claim was also staked on events. Paul says that if Christ has not been raised, the Christian message is empty and its preachers are liars about God (1 Corinthians 15:14-15). Whatever else that is, it is not a hypothesis with no way of coming out false. 3. Evidence is allowed to count against the claim. The honest answer is not that nothing could tell against God's goodness. Some things do tell against it, and they are carried alongside other grounds for trusting him. Think of a man in an occupied country who meets a stranger one night, becomes convinced the stranger leads the resistance, and is told to trust him whatever happens. Some nights the stranger is seen smuggling people out. Other nights he is seen in police uniform, handing members of the resistance over. The man does not pretend the second kind of night is neutral. He counts it against, and it costs him something, and he goes on trusting for reasons that have nothing to do with pretending. What he refuses to do is redescribe each betrayal until it stops looking like one. Christianity behaves the same way in the open. It offers substantive accounts of why God permits evil, argues about them, and throws some of them out (D5). A theodicy that made God the author of sin gets rejected, and it gets rejected on grounds internal to the faith. 4. A charge about every possible world is the less testable claim. Put the gardener question to the other side of the table. What would have to happen for someone to withdraw the claim that no possible good could justify permitting any evil? Nothing observable, so far as anyone has said, because the claim ranges over every possible world and no experience reaches that far. That is not a reason to drop the charge. It is a reason to notice that the complaint about claims nothing could refute lands hardest on the most ambitious claim in the room, which is the charge itself. See also: CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Proof of God from Evil

Science

Is Science Compatible with Christian Theism?

TBD

Bad Claims

Common "Bad Information" Challenges to Christianity

Miscellaneous

Objections That Fit No Other Category

World Religions

Critical Analyses of Non-Christian Religions

Philosophical Theology

Analytical Analyses of Christian Systematic Theology

Bibliology

The Doctrine of Scripture

Theology Proper

The Divine Nature and Properties of God

Creation

The Doctrine of Creation

TBD

Anthropology

The Doctrine of Humanity

TBD

Christology

The Doctrine of Christ

TBD

Soteriology

The Doctrine of Salvation

TBD

Ecclesiology

The Doctrine of the Church

TBD

Eschatology

The Doctrine of Last Things

Public Theology

Theological Analyses of Societal Issues

Theology of the Family

Where Faith and Family Intersect

Biographies

Notable Works & Great Quotes from Key Figures
Playlist
Justified True Belief