Justified True Belief

(P1) If God did not exist, then objective moral values would not exist. + (P1) asks what would have to be true for some things to be really wrong. It is not a claim about anyone's behavior. (1) An objective moral value holds whether or not anyone agrees. - To call a value objective is to say it does not depend on anybody's opinion: not yours, not your culture's, not the whole species'. • Two things travel under the phrase moral values, and a world without God gets neither. A value is what is good or bad: kindness is good. A duty is what a person owes: you must not torture the child. - "The torture of children for fun is wrong" is meant as a claim of that kind, not as a report that most people dislike it. If everyone alive were talked out of believing it tomorrow, that would only mean everyone was wrong. - A culture that voted the earth flat would not flatten the earth. Some moral facts are like that: we discover them rather than issue them. (2) On naturalism the world has no ingredient a duty could be made from. - Naturalism is the view that nothing exists beyond the natural order: no minds without brains, no purposes beyond biological function, no values apart from what some creature happens to prefer. On that inventory the universe is particles and forces, and then, late and locally, arrangements of particles with preferences about other arrangements. - Picture a man setting a dog on fire, and describe the scene completely: the chemistry of the accelerant, the nerves firing in the animal, the dopamine in the man. The description finishes with the word "wrong" never having appeared, and no further physical detail will supply it. - A duty has to be made out of something, and nothing in that description is the right sort of stuff. (3) Christian theism grounds goodness in what God is. - God did not consult a list of virtues, and he did not invent one on a whim either. Goodness is his own nature. He is not measured against the standard; he is what the standard is. - Think of an orchestra tuning. Every instrument is in tune or out by reference to the fork, and nobody asks whether the fork itself is in tune with anything further. The fork is what being in tune means. - The claim is old. Augustine argued that everything made is good only so far as it shares in the goodness God simply is. (4) An obligation is the kind of thing that is owed to somebody. - "You must not do that" is a sentence with a speaker behind it, and the speaker is not optional. - Find a note under your wiper reading MOVE THIS CAR and your first question is who wrote it. Learn that nobody did, that wind and wet leaf litter fell that way, and the words stay perfectly legible while the demand evaporates. - Careful philosophers do defend a moral law that is legible and authorless, and one note under a wiper does not refute them. - What the note isolates is the personal side of moral life: being wronged rather than merely damaged, owing an apology, feeling guilt rather than regret. Each is a relation to somebody, and an account with nobody in it must explain all three or do without them. (5) Knowing right from wrong is a different question from what makes it so. - Plenty of atheists are kinder, braver and more honest than plenty of Christians. Everybody knows it, and (P1) says nothing about anyone's character. - Philosophy has two names for the two questions. Moral epistemology asks how a person comes to know right from wrong; moral ontology asks what makes an act right or wrong at all. (P1) is entirely a claim of the second kind. - A man can read a map accurately while insisting nobody drew it, and no amount of good reading settles where it came from. Therefore, if God does not exist, the moral facts we take ourselves to know have nothing underneath them.

(P2) Evil exists. + Two words, and both sides of the argument from evil have already agreed to them. The claim is that some acts are really evil, not that anybody happens to dislike them. (1) The argument from evil already assumes that evil is real. - The problem of evil begins by pointing at something: a camp, a children's cancer ward, a famine that did not have to happen. The pointing is the argument. - If the complaint were only that certain events are not to our taste, nobody would have written a library about it. The force comes entirely from the conviction that these things are wrong, not merely unwelcome. - Scripture does the same pointing, and in the same register: "Woe to those who call evil good and good evil" (Isa 5:20). - So the two sides are not really arguing about this. Two doctors at a bedside can agree completely on the diagnosis and still disagree about what it means for the patient. The agreement here is like that. It is genuine, not a debating concession, and the whole dispute is about what follows from it. (2) The wrongness of some acts is known more surely than any premise denying it. - A jury is shown photographs of a basement where a child was kept for years. Nobody in that room reasons their way to the wrongness of it; it arrives with the facts. - Knowledge like that is properly basic: not worked out from anything more obvious, the way you know you had breakfast without needing an argument for it. Beliefs of that kind stand until something defeats them, and moral experience earns the same trust we give our eyes. - So when a chain of clever premises ends by denying what you know better than any link in it, doubting the chain is the reasonable move. A theory on which the basement was not really wrong is less credible than the thing it denies. Someone may still take the theory, but not because the evidence favors it. (3) Retreating to subjective evil dissolves the objection. - An atheist can always say that evil is a name for what we dislike. The move is available and it is coherent. - What it costs is the argument. Once evil is a report about us, the problem of evil becomes an argument from human distaste, and distaste is not evidence about what exists. - A man who says he hates the cold has told you something about himself. He has told you nothing about the weather, and nothing whatever about whether the weather had a maker. (4) Christian theism agrees that evil is real. - Some worldviews do treat it as an illusion. Strands of Hindu and pantheist thought (pantheism being the view that everything is God) describe the world of suffering as maya, appearance rather than final reality, something the enlightened learn to see through. - Christianity says the opposite and stakes everything on it. The fall, the cross and the last judgment are all unintelligible if evil is a trick of perspective. You do not execute a man to clear up a misunderstanding. - The Bible names evil even where it insists God is at work in it. Joseph tells the brothers who sold him, "you meant evil against me, but God meant it for good" (Gen 50:20). The first half of that sentence is not withdrawn by the second. - "Abhor what is evil; hold fast to what is good" (Rom 12:9). Nobody is commanded to abhor an illusion. Therefore, evil exists, and exists as something more than a name for what we happen to dislike.

(IC) Therefore, objective moral values exist. + Calling an act evil is not a report of distaste but a judgment, and a judgment is either true or false. The step from (P2) is short though not automatic, because everything depends on which sense of the word "evil" is in play. (1) "Evil" is not a natural fact delivered in a harsh tone. - An earthquake can be described completely without moral vocabulary: plates, stored energy, casualty figures. Nothing is left out of the description. - Try the same with a man who drowns his children. "He caused their deaths" is true, and it is plainly not what anyone means. - The gap is the one between "the fire consumed the house" and "he burned the house down." The first is physics. The second is an accusation, and accusations are not longer descriptions. (2) An accusation needs a standard. - To say something ought not to have happened is to measure it against how things ought to be. Remove the second and the first stops meaning anything. - You cannot call a line crooked without a straight edge somewhere. The edge need not be drawn on the same page, but it has to exist, or "crooked" collapses into "I would have drawn it differently." (3) Some philosophers keep the moral language while denying the moral facts. - Picture a philosopher who will tell you the Holocaust was monstrous and mean every word of it, who will argue with anyone who says otherwise, and who nonetheless holds that there is no fact out in the world answering to what he just said. He is not being slippery. This is worked-out philosophy with a large literature behind it. • The family of views is called expressivism: saying "cruelty is wrong" voices a stance rather than reporting a property of the cruelty. The project of earning back the ordinary way we talk, so that a moral claim can be refined, criticized and got wrong by its own internal standards, is called quasi-realism. • The position on the other side holds that some moral claims are simply true of the world, the way "the bridge is unsafe" is true of the bridge. That one is called moral realism, and it is the term the rest of the debate turns on. - One short question separates the two. When the atheist says the Holocaust was evil, is he saying something that would have been true even if every human attitude had been different? - If yes, the values are objective and (IC) follows. If no, the argument from evil has lost its teeth, whatever vocabulary it keeps. (4) The complaint about evil has already granted an objective moral value. - The objection has to treat the evils as really evil, or it does not get moving at all. And treating an act as really evil rather than merely unwelcome is already granting an objective moral value. - A man who sues for breach of contract has conceded that contracts bind. He cannot win the case and then argue that promises are only noises. Therefore, objective moral values exist.

(C) Therefore, God exists. + A deductive argument works like a chain: if every link holds, the conclusion cannot slip out. Here the links are (P1) and (IC), and the conclusion follows whether or not anyone likes where it lands. (1) The form is modus tollens. - Modus tollens means denying the consequent: take an if-then claim, deny its second half, and the first half is denied along with it. If the tide came in, the sandcastle is gone. The sandcastle is standing. So the tide did not come in. - Run it here. (P1): if there were no God, there would be no objective moral values. (IC): there are objective moral values. Therefore there is a God. - A premise in a chain like this does not have to be certain to do its work. It has to be more plausible than its denial, which is the standard a deductive argument in philosophy is normally held to. - The form is deductive. The support for (P1) is not, and that is worth being plain about. What holds (P1) up is a comparative judgment: no account of moral value without God has yet worked as well as the account with him. That is why (P1) is where the case is properly attacked. (2) Either the evils are objective, or the complaint loses its force. - A man who complains that the referee made a terrible call has conceded that there is a game, that it has rules, and that calls can be got wrong. His outrage is not evidence against the existence of rules. It runs on them. - The same fork faces anyone who presses the evils against God, and there is no third prong to it. - Take the first. If the evils are objectively evil, (IC) holds and the chain runs all the way to its conclusion. - Take the second. If the evils are not objectively evil, the accusation goes down with them. What is left is either a report of how the world strikes us or a charge that somebody else's beliefs do not hang together, and neither of those is the argument most people take themselves to be making. (3) Why God permits evil is a different question. - (P1) and (IC) together yield that God exists, and (P2) that evil is real. They yield no account whatever of why he permits it, and none of it is any comfort to a person in the middle of it. - Job never received an explanation. What he received was a voice out of the whirlwind asking him about storehouses of snow and the wild donkey and the way a horse smells battle (Job 38-41). That is an answer, and it is not a reason. (4) What would defeat this. - A non-theistic account of moral value and obligation that actually works, which would take (P1). - A showing that the argument from evil needs no more than evil as a human attitude, or no more than the theist's own standards borrowed for the length of a hypothetical, which would take (IC). - A reason to deny (P2) in the sense it is asserted: that nothing is objectively evil at all, which is a position with a price of its own. Therefore, evil pressed as an objective wrong cannot count against God. If the premises hold, it requires him.

On the moral argument: Adams, Robert Merrihew. Finite and Infinite Goods: A Framework for Ethics. New York: Oxford University Press, 1999. Baggett, David, and Jerry L. Walls. Good God: The Theistic Foundations of Morality. New York: Oxford University Press, 2011. Craig, William Lane. Reasonable Faith: Christian Truth and Apologetics. 3rd ed. Wheaton, IL: Crossway, 2008. Craig, William Lane. Systematic Philosophical Theology, Volume IIb: On God: Excursus on Natural Theology, On the Trinity. Hoboken: Wiley-Blackwell, 2026. Craig, William Lane, and Walter Sinnott-Armstrong. God? A Debate Between a Christian and an Atheist. New York: Oxford University Press, 2003. Evans, C. Stephen. God and Moral Obligation. Oxford: Oxford University Press, 2013. Hare, John E. God's Call: Moral Realism, God's Commands, and Human Autonomy. Grand Rapids: Eerdmans, 2001. Lewis, C. S. The Abolition of Man. London: Oxford University Press, 1943. Lewis, C. S. Mere Christianity. London: Geoffrey Bles, 1952. Moreland, J. P., and William Lane Craig. Philosophical Foundations for a Christian Worldview. 2nd ed. Downers Grove, IL: IVP Academic, 2017. Wolterstorff, Nicholas. Justice: Rights and Wrongs. Princeton, NJ: Princeton University Press, 2008. On moral knowledge and moral language: Clarke-Doane, Justin. "Morality and Mathematics: The Evolutionary Challenge." Ethics 122, no. 2 (2012): 313-340. Cuneo, Terence. The Normative Web: An Argument for Moral Realism. Oxford: Oxford University Press, 2007. Horgan, Terence, and Mark Timmons. "New Wave Moral Realism Meets Moral Twin Earth." Journal of Philosophical Research 16 (1991): 447-465. Moore, G. E. "Proof of an External World." Proceedings of the British Academy 25 (1939): 273-300. Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University Press, 2000. Non-theistic moral realism engaged: Brink, David O. Moral Realism and the Foundations of Ethics. Cambridge: Cambridge University Press, 1989. Harris, Sam. The Moral Landscape: How Science Can Determine Human Values. New York: Free Press, 2010. Shafer-Landau, Russ. Moral Realism: A Defence. Oxford: Clarendon Press, 2003. Wielenberg, Erik J. Robust Ethics: The Metaphysics and Epistemology of Godless Normative Realism. Oxford: Oxford University Press, 2014. On the problem of evil: Howard-Snyder, Daniel, ed. The Evidential Argument from Evil. Bloomington: Indiana University Press, 1996. 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: Antony, Louise M., ed. Philosophers without Gods: Meditations on Atheism and the Secular Life. New York: Oxford University Press, 2007. Draper, Paul. "Pain and Pleasure: An Evidential Problem for Theists." Nous 23, no. 3 (1989): 331-350. Joyce, Richard. The Evolution of Morality. Cambridge, MA: MIT Press, 2006. Joyce, Richard. The Myth of Morality. Cambridge: Cambridge University Press, 2001. Mackie, J. L. Ethics: Inventing Right and Wrong. Harmondsworth: Penguin, 1977. Mackie, J. L. The Miracle of Theism: Arguments For and Against the Existence of God. Oxford: Clarendon Press, 1982. Morriston, Wes. "God and the Ontological Foundation of Morality." Religious Studies 48, no. 1 (2012): 15-34. Nietzsche, Friedrich. The Gay Science. Translated by Walter Kaufmann. New York: Vintage, 1974. Ruse, Michael. "Evolutionary Theory and Christian Ethics." In The Darwinian Paradigm. London: Routledge, 1989. Sartre, Jean-Paul. Existentialism Is a Humanism. Translated by Carol Macomber. New Haven: Yale University Press, 2007. Street, Sharon. "A Darwinian Dilemma for Realist Theories of Value." Philosophical Studies 127, no. 1 (2006): 109-166. Ancient sources: Augustine of Hippo. Confessions. Translated by Henry Chadwick. Oxford: Oxford University Press, 1991. Augustine of Hippo. The City of God against the Pagans. Edited and translated by R. W. Dyson. Cambridge: Cambridge University Press, 1998. Confucius. The Analects. Translated by D. C. Lau. Harmondsworth: Penguin, 1979. The Egyptian Book of the Dead: The Book of Going Forth by Day. Translated by Raymond O. Faulkner. San Francisco: Chronicle Books, 1994. Plato. Euthyphro. In Complete Works, edited by John M. Cooper. Indianapolis: Hackett, 1997. Biblical citations: Genesis 1:27; Genesis 50:20; Job 38-41; Isaiah 5:20; Micah 6:8; Romans 2:14-15; Romans 12:9; Romans 12:21. See also: Natural Theology: The Moral Argument CO / PoE: Intellectual PoE: Logical CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Intellectual PoE: External CO / PoE: Emotional PoE

+ Objective moral values can be grounded without God. Well-being gives morality a subject matter science can study, evolution explains where the moral sense came from, rational agency grounds duty, and moral facts may simply be natural facts under another name.

1. The well-being account has to import the value it cannot derive. - The account says moral truths are facts about the flourishing of conscious creatures. That is real research and it produces real results. But the claim that flourishing is what matters is not itself a finding; it is a moral premise placed at the front, and no amount of neuroscience delivers it. - A brain scanner can show you, in detail, that a man is in agony. Nothing on the screen tells you that you must not put him there. The instrument reads states; it does not read obligations. - The best reply grants all of that and says it costs nothing. Medicine cannot derive the value of health from chemistry either, and every discipline has to start somewhere, so let morality start from the claim that the worst possible misery for everyone is bad and then get on with the work. - The parallel breaks at the one place that matters. Medicine issues hypothetical demands, which bind you only given an aim you have already taken up: if you want the fever down, take this. Decline the aim, walk out of the office, and nothing whatever follows. - Moral demands do not behave that way. They present themselves as categorical, binding you whatever your aims happen to be, and "cruelty was never one of my goals" is not a defense against them. A starting point adopted inside a practice cannot reach the man standing outside the practice, and he is the one who needs reaching. 2. No pile of facts about what is adds up to what ought to be. - That observation is old enough to have a name. The is-ought gap says that facts about how the world stands never by themselves yield a conclusion about how anyone should act. You can know everything about the physics, biology and psychology of an act of cruelty and still not have reached its wrongness. - Think of a complete blueprint of a building. It will tell you every dimension, every material and every load. It will never tell you whether the building should have been put there. - Take the appeal to rational agency, which says morality is built into the structure of rational choice. Grant the structure entirely. Why that structure obliges a creature who happens to have it is a further question, and the further question is not about structures. - Take the evolutionary story, which explains why we hold the moral beliefs we hold. Explaining why a belief is held is a different job from showing that it is true, and a belief that helped our ancestors survive need not be an accurate one. - The appeal to well-being lands in the same place. Somewhere in each of these accounts a premise about what ought to be gets in through the side door, and from that moment the account is helping itself to the very thing in dispute. 3. Identifying moral facts with natural facts relocates the gap rather than closing it. - One naturalism does not try to cross the gap at all. It says moral properties simply are natural properties, discovered the way chemists discovered that water is H2O. Nobody derives "water" from "H2O"; the two words turned out to name one thing. - Wrongness, on that view, might just be the frustration of the interests of creatures that can feel, and no derivation is needed anywhere. It is an ingenious move and it has been argued in the journals for decades. - Then try the thought experiment that has been run against it for just as long. Imagine a second planet whose people say "wrong" with all the force we do, argue about it, and go to war over it, but whose word is fastened to a different natural property from ours. - If moral words worked like "water," those people and we would simply mean different things, and there would be no more disagreement between us than between two chemists describing two different liquids. Yet each side would say the other was mistaken, and it would be a real fight. Moral words are not fastened to the world the way "water" and "gold" are. - The binding force goes missing under the identification as well. If wrongness just is the frustration of interests, then "you should not frustrate interests" says only that frustrating interests frustrates interests. True enough, and it asks nothing of anybody. 4. Obligations to strangers and enemies are the hardest cases. - Kin altruism and reciprocal altruism are well evidenced and explain a great deal. We look after our own, which is kin altruism, and we look after those who may one day look after us, which is the reciprocal kind. Reputation stretches the account further still, since a man known for keeping faith gets treated better by people he has never met. - What none of that reaches is the obligation we take ourselves to have toward someone who can never repay us, cannot damage our standing, and actively wishes us harm. "Overcome evil with good" (Rom 12:21) is not a survival strategy. - Theistic morality grounds that obligation directly. Every person bears the image of the same God (Gen 1:27), and that is a fact about the person rather than a fact about his usefulness to us. 5. Atheist philosophers have said the same thing themselves. - Michael Ruse argues that morality is a biological adaptation, and that taken as a set of objective claims about something beyond us it is an illusion our genes produced because cooperators left more descendants. - J. L. Mackie opened a book on ethics by denying that there are any objective values, and granted that such values, if they existed, would sit far more comfortably in a universe with a god in it than in one without. - Jean-Paul Sartre put it as a loss rather than a relief. With no infinite and perfect consciousness to think it, there is no good settled in advance, and a man is left inventing value with nothing to appeal to. - None of the three was conceding anything to theism. Each was following his own view where it led, which is why (P1) is a shared piece of analysis that the two sides read in opposite directions rather than a theistic invention.

+ Evil is not objective but subjective, a name for what we dislike or what our culture forbids. Cultures have differed on slavery, infanticide and polygamy, which shows the label is constructed rather than discovered.

1. The denial takes the problem of evil down with it. - An atheist who says evil is subjective can no longer say the Holocaust was objectively evil. He can say only that he is against it. That is the immediate cost, and it is not a small one. - Being against something is not evidence about whether God exists. It is autobiography. - The two positions cannot be held at once: evil as an objective fact when it is aimed at God, and evil as a preference when its objectivity is questioned. 2. Disagreement about a thing is not evidence that the thing is unreal. - Cultures have differed about the shape of the earth, the cause of disease and the age of the universe. Nobody concludes that geography, medicine and cosmology are matters of taste. - Disagreement shows that some people were wrong. That is the opposite of what the objection needs, because being wrong requires something to be wrong about. - A good deal of moral disagreement turns out, on inspection, to be disagreement about ordinary facts. Salem hanged women for witchcraft because it believed they were causing real harm by supernatural means. The moral rule underneath, that people who kill their neighbors should be stopped, was never in dispute at all. Correct the factual belief and the moral difference disappears. 3. The variation sits at the edges of a shared core. - Virtually every society has condemned murder of the innocent, betrayal of trust, theft and cruelty to children. What varies is who counts as innocent, what counts as theft, and where the boundary of the community falls. - The overlap is documented rather than assumed. In Spell 125 of the Egyptian Book of the Dead, written down more than three thousand years ago, the dead man stands before forty-two divine judges and declares that he has not killed, has not stolen, has not lied, has not caused anyone pain. Confucius states the same rule from the other direction in the Analects: do not impose on others what you would not want for yourself. - Anthropologists dispute how much that shows, and they are right to. Condemning "murder" comes very close to condemning wrongful killing, which every society will do by definition, so a common core stated that thinly proves almost nothing on its own. - The core is not that thin. Duties to aging parents, the wrongness of treachery toward a guest, an obligation of mercy to the weak: not one of those is true by definition, and all of them turn up across Babylonian, Egyptian, Norse, Hindu, Chinese, Greek and Jewish sources alike. - The harder cases carry the rest. A tribe that spares its own children and kills its neighbors' has not invented a rival morality. It has drawn a smaller circle around the same rule, and somebody who widens the circle is doing exactly what the abolitionists did, which everyone treats as a correction rather than a change of fashion. 4. Moral reform only makes sense if there was something to get right. - The abolition of slavery is normally described as progress. On a purely subjective account it was a change of fashion, and Wilberforce was not discovering anything but merely preferring differently, with no more standing than the slavers he opposed. - Notice what "progress" has to mean on that account. It cannot mean that the abolitionists saw something the slave traders had missed, because there was nothing there to be seen. It can only mean that taste moved, the way hemlines move. - Almost nobody believes that, including people who defend the subjective account in the seminar room. When they are themselves robbed or betrayed, the language they reach for is not the language of preference.

+ Raising the problem of evil commits the atheist to nothing. The argument is a reductio: it grants the Christian's own belief in objective evil for the sake of argument, then shows that belief is incompatible with God.

The strongest form of this says the atheist asserts nothing at all. He takes the theist's own commitments, that God is all-powerful and all-good and that some things are genuinely evil, and shows the set cannot all be true. No moral claim is made in his own voice, so nothing in it can be turned back on him. 1. A reductio does exactly what it says and no more. - A reductio ad absurdum assumes an opponent's position, for the sake of argument only, to show that it leads to a contradiction. It is a perfectly respectable move, and there is nothing shabby about it. - You use one whenever you take somebody at his word to show his word will not hold. "Suppose you really were in Denver at eight, as you say. Then you cannot have been at the restaurant at eight fifteen." That argument tells you the story is broken. It does not tell you where the man actually was. - Read that way, the conclusion is that Christian theism is unstable on its own assumptions, and nothing whatever about whether God exists. "Your position has a problem" and "there is no God" are different claims, and the second is the one usually announced. 2. The borrowed set has to include the grounding. - Borrowing selectively is normally fine. A reductio takes whatever part of a position it needs, and it is no objection at all that the rest was left on the shelf. The trouble here is which part this one takes. - Christian teaching does not carry "some things are objectively evil" as a free-standing item. It holds that such things are evil because they run contrary to the character of the God who made and owns their victims. - Take the objectivity and leave the grounding behind, then, and what is being reduced is no longer the Christian's position at all. It is a hybrid nobody holds. - Take the whole set instead, and (P1) comes along with it. Run the borrowed objective evil through the borrowed (P1) and it yields God, which means the reductio has derived the reverse of what it wanted out of premises it helped itself to. It is like quoting a will to establish that the estate exists while denying the document has any authority. - What does survive is the internal charge: that the theist's own commitments about God's goodness and power sit badly with the evils he admits. That charge is serious, and it is answered on its own terms in the logical and probabilistic forms of the problem of evil. 3. The conditional version still runs. - Grant the reductio framing completely. The chain restates to match it: if there is objective evil, as the objection assumes for the sake of argument, then there is God. - Assumptions carry their consequences with them. Assume for the sake of argument that the fire was arson, and you have assumed an arsonist. You cannot borrow the assumption and then decline what follows from it. - So two doors are left open, and only two. The objection can deny that the evils are objectively evil, which is the error theorist's route (D12). Or it can deny that objectively evil acts imply objective moral values, which is the ground (IC) covers. - Retreating to a hypothetical opens no third door. It relabels the same two. 4. Staying in reductio mode costs the objection most of its force. - In debates and in print the argument from evil arrives with moral conviction. The examples are chosen for their horror, the tone is outrage, and the conclusion drawn is that God does not exist. - The sloppiness of popularizers proves nothing, and it would be cheap to rest on it. What matters is that the persuasive weight of those presentations comes from the reader agreeing that these things are monstrous, and agreeing with something is not the same as entertaining it for a paragraph. - So the result depends on which objector is speaking. Against one who asserts that the evils are objectively evil, (C) follows. - Against one who asserts nothing, what follows is a conditional and a bill. He keeps the internal charge, which has answers of its own. He gives up the evidential weight of the evils themselves, which was where the argument from evil got its power in the first place. That is a smaller argument than the one he started with, and both sides should say so. See also: CO / PoE: Intellectual PoE: Logical

+ Objective moral values can exist as abstract necessary truths, the way numbers do, grounded in the structure of reality rather than in any person. Moral realism does not require God.

1. The strongest version is not about Forms in a separate world. - Platonism about morality, in its older shape, holds that moral properties exist as abstract objects: things that are real but are not anywhere and are made of nothing, the way the number seven is real without sitting in a drawer somewhere. • They are also necessary, which means there is no way things could have gone in which seven failed to be prime. • And they are causally inert, a technical way of saying they cannot push, speak, intend or require anything. The number seven has never asked a favor of anybody. - Most philosophers who hold the position now claim less than that, and the weaker claim is the harder one to answer. They say the moral truths are necessary and that they hold because of natural facts, without collapsing into them. - Wrongness, on that account, is not a ghostly extra ingredient hovering beside the cruelty. It is what the cruelty amounts to, and the cruelty does all of the causing. Put that way, the complaint that abstract objects cannot push anything has nothing left to push against. - The difficulty that survives is about obligation rather than causation. Being wronged is a relation between persons, and when a child is beaten, nobody says a principle was violated and leaves it there. We say that something was done to her. - A necessary truth is not the sort of thing a wrong can be against, nor the sort of thing that can be owed anything. That difficulty is what has driven much of the recent work on moral obligation, theistic and otherwise. 2. The grounding question comes straight back one level up. - Suppose the moral facts are simply there. Why these facts and not others? What makes this particular set authoritative over concrete persons who never chose it? - Rival sets are easy to imagine, and that is the trouble. One on which loyalty to your own people outranks everything else. One on which the suffering of strangers carries no weight at all. Each is a coherent set, and if nothing selects among the candidates, then calling ours the authoritative one is an assertion rather than an explanation. - Think of being handed a rulebook with no publisher on the cover, for a game you never agreed to join, and being told that the rules bind you because they are the rules. - Two answers are available. Either there is no further explanation, which is a substantial concession, or there is something deeper, which reopens the question the view was meant to close. 3. Theism answers four questions at once that free-standing moral facts answer one at a time. - Why the moral facts obtain: they follow from the character of a necessary being. - Why they bind us: they come from the one to whom we are accountable. - Why we can recognize them: we are made in his image (Gen 1:27), with the work of the law written on our hearts (Rom 2:14-15). - Why they apply to everyone: they come from the source of everyone. - One account covering four questions is worth more than four accounts covering one each, in the same way that a single suspect who fits the fingerprints, the motive and the timeline beats three separate coincidences. 4. Grant that this is an inference, not a proof. - Non-theistic moral realism is a live position held by careful philosophers, and the difficulties just raised do not refute it. - What would settle it in their favor is an account of how a fact that is nobody's lays a claim on a person. Until that arrives, the difficulty stands where it is. - The move from objective moral values to a personal God is comparative: an inference to the better explanation among the live options, which makes (P1) more plausible than its denial rather than beyond dispute.

+ The first premise already presupposes a theistic account of morality, so granting it concedes the conclusion in advance. The argument then walks back to what was smuggled in at the start.

1. Agreeing to an if-then claim is not agreeing that either half is true. - Philosophers call an if-then claim a conditional. "If the milk was left out overnight, it has spoiled" is one. You can accept it completely while holding that the milk was in the refrigerator all night and is perfectly fine. Accepting the link commits you to nothing about the two halves. - (P1) is a conditional of exactly that kind: if God did not exist, objective moral values would not exist. Someone who accepts it has agreed to a link. He has not agreed that God exists, and he has not agreed that objective moral values exist. - An atheist can hold (P1) with perfect consistency and run it the other way: no God, therefore no objective moral values. Serious atheist philosophers have done exactly that, and said so in print (D1). If accepting the conditional conceded theism, their position would be incoherent, and it plainly is not. 2. The premise is defended from moral experience, not from theism. - The case for (P1) runs from features of moral life: that moral claims look objective, that obligation binds rather than merely recommends, and that moral categories resist translation into natural ones. - Take the second of those. Read a report of a child sold into a brothel, and the reaction is not "I personally disapprove of this." It is that something has been done which must not be done, by anyone, ever. Whatever else that experience is, it is not a report of a preference, and noticing so does not require believing in God. - None of those starting points assumes God exists. They are things a committed atheist can grant and many do, which is why (P1) appears in the literature written from both sides. 3. Applied evenly, the charge disqualifies the argument from evil too. - To beg the question is to assume, somewhere in your premises, the very thing you set out to prove. It is a real fault and worth watching for. - Now look at the argument from evil with the same eye. It asserts that if God existed there would be no gratuitous evil, meaning evil that serves no purpose good enough to justify permitting it. It adds that gratuitous evil does exist. It concludes against God. That is the same structure as (P1) and (IC): a conditional, plus a contested second claim. - If stating a conditional that leads somewhere unwelcome counted as begging the question, both arguments would be guilty of it, and the two sides could do nothing but sit and look at each other. The honest response to a valid argument is to attack a premise. 4. A circular argument leaves nothing to push against. - "The Bible is true because it says so" is circular, and an opponent has nothing to take hold of: every handle is attached to the thing he is trying to move. That is the practical test to apply. - (P1) is nothing like that. (D1), (D4), (D7), (D11) and the whole non-theistic moral realism literature are direct attacks on it, and they are exactly the right kind of objection. Books and journal articles are still being written on both sides.

+ Is an act good because God commands it, or does God command it because it is good? If the first, morality is arbitrary and he could as easily have commanded cruelty; if the second, goodness stands above God and does not need him.

1. The dilemma leaves out a third option. - The question is old enough to have a name. Plato set it out in the Euthyphro, where Socrates asks a confident young man whether the holy is loved by the gods because it is holy or holy because the gods love it, and the young man cannot answer. Ever since, the two branches have been called the horns of the Euthyphro dilemma. - Neither horn. Goodness is not a list God consults, and not a list he invents. It is what he is, and his commands express his character rather than his preferences. - The tuning fork again: every instrument is in tune by reference to it, and the fork is not in tune with something further. Asking whether God is good is asking whether the fork is in tune. - This is not a modern patch. Augustine argued it, and the mainstream of Christian moral thought has held it since. 2. The arbitrariness horn assumes God could have been otherwise. - Could God have commanded cruelty? Only if his nature could have been different, and on the historic Christian view it could not. His goodness is necessary, not a setting that happened to land where it did. - "What if God had been evil" is closer to "what if three had been even" than to "what if the speed limit were higher." It looks like a question and does not survive inspection. 3. The independence horn hands goodness back to nobody. - If goodness stands above God, it stands above everyone, and we are back at the note under the wiper with no author: real words, legible, and binding on no one. - That horn does not defeat theism so much as relocate the problem, and it is the position (D4) takes up directly. 4. Grant that the dilemma bites some theistic views. - Views that make God's bare will the whole story, with no appeal to his nature standing behind it, do face the first horn squarely, and the objection is fair against them. - The account most often defended now is not that one. It grounds what is good in God's character and what is obligatory in his commands, so a command is never a bare decree. It is that character reaching a creature who owes him something. Kindness is good because of what God is; you are required to be kind because he has said so, and he could not have said otherwise and remained himself. - Christians disagree about how the two halves fit together, and (P1) through (C) do not require settling it. What every side holds is that the good is not arbitrary, and that is precisely what the first horn attacks.

+ Some moral facts are simply necessary and need no maker, the way it needs no maker that two and two are four. Objective values can be brute features of reality with no God behind them.

The strongest published form of this is Erik Wielenberg's. On his account moral facts are non-natural, meaning they do not reduce to facts about particles, brains or behavior. They are necessary. And they are sui generis, a Latin phrase meaning "of its own kind": they belong to no wider category and rest on nothing further. Philosophers call a fact like that a brute fact, one that simply is, with no explanation standing behind it. That is not an evasion. It is the most serious version of the objection and it should be met at that strength. 1. The position is coherent. - Caricaturing it wastes everyone's time. It is not the claim that morality is made up, nor that it reduces to feelings. It is full-blooded moral realism that declines the further question. - The parallel with mathematics is doing real work. Nobody thinks the necessary truths of arithmetic require an explanation in terms of something more basic, and the suggestion that moral truths might be similar is respectable. 2. Brute necessity is a bill this view has to pay twice. - Every worldview stops somewhere and calls something basic. That is not a defect in itself, and theism does it too. The question is only what gets to be basic. - On this view two things are basic and neither one explains the other: a rich structure of specific moral facts, and a physical world that eventually produced conscious creatures. Yet the facts turn out to be about exactly the kind of creature that world happened to produce. - Imagine finding a key lying in one field and a lock lying in another, each of them uncaused, neither having anything to do with the other, and the key turns the lock. Nothing about that is impossible. It is simply a second thing to explain, and theism has one necessary being from whose nature the moral facts follow, which is one starting point rather than two that happen to match. - The reply is that necessity removes the coincidence, and it is a good reply. If the moral facts hold in every possible world, there was never a world in which they might have failed to fit us, and a match that could not have failed is not a lucky one. Put the objection as luck and it is answered: the key and the lock could not have missed each other. - What that leaves untouched is the specificity. Necessity is not explanation, and saying a thing could not have been otherwise never says why it is so. - So the view is still positing a large set of necessary connections between natural features and moral ones. Nothing selects which connections obtain, and every one of them has to be accepted as basic. Theism replaces the whole set with a being. 3. Obligation still lacks an addressee. - The difficulty raised in (D4) survives intact here. A necessary truth is not a claim on anyone. That two and two make four is necessarily true and has never obliged anybody to do anything, and it is not obvious why "cruelty is wrong," held in the same way, should be any different. - So the wrongness may be perfectly objective and the accountability still be left unaccounted for. Accountability is a large part of what people mean when they call something evil rather than merely regrettable, and "that was a terrible thing" and "you will answer for that" are not the same sentence. 4. The fit between the moral facts and our knowledge of them needs explaining. - If moral facts do no causal work and our faculties were shaped by survival pressure, it is not obvious why the two should line up. - Wielenberg answers with a third-factor account, and it deserves stating properly. One and the same feature both makes an act wrong and, by a separate route, causes us to believe it wrong. Our capacity to suffer is what makes torturing us wrong, and that same capacity is why we recoil from torture. Nothing has to be coordinated, because a common factor stands behind both, the way one cold front explains the frost on the windshield and the ache in an old injury at the same time. - That is a serious reply and the exchange over it is not finished. What it does not explain is the relation it leans on. - Why should the capacity to suffer be the feature that makes an act wrong, rather than one of the thousand other features any act has? On this view that connection is itself brute, so the correlation has been moved a step back rather than accounted for. - The theistic answer is more direct, and it is the one (D4) named: creatures made in the image of the being whose nature the facts express (Gen 1:27). That is a genuine explanatory advantage rather than a refutation.

+ Evolution shaped our moral intuitions to promote survival, not to track moral truth. Our confidence that some things are really evil is exactly what we would expect whether or not it is true.

The sharpest published version is Sharon Street's dilemma, and it has two horns. Either the forces that shaped our evaluative attitudes had nothing to do with which evaluative claims are true, in which case our moral beliefs are almost certainly off target and moral knowledge collapses. Or those forces did have something to do with it, and then the realist owes an account of how natural selection came to track a moral order it was never aiming at. Street argues that the second horn is worse science than the simple alternative: our evaluative tendencies were selected because they linked circumstances to useful responses, with no truth-tracking involved anywhere. 1. The objection takes evil away from the atheist as well. - This is an evolutionary debunking argument: not the claim that our moral beliefs are false, but the claim that their origin gives us no reason to think they are true. A compass that always points at the nearest iron deposit is not lying to you, but it is not telling you where north is either. - Aimed that way, it is aimed at moral realism as such, not at theism. If it works, our conviction that the Holocaust was objectively evil is debunked along with everything else. - An atheist who runs it has dismantled the argument from evil in the same motion, since that argument needs the evils to be really evil rather than merely felt to be. - It is a coherent position and serious philosophers hold it. The sharpest of them do not end in nihilism either. Value gets rebuilt out of our own attitudes, and moral talk carries on much as before. - What the rebuilt version cannot do is the work the argument from evil needs. It would have come out differently had our attitudes come out differently, and that is exactly the fork (IC) puts to every view of that kind. 2. On theism the fit is not a coincidence. - The objection's force comes from a gap between faculties shaped for survival and truths they were never shaped to track. - Theism closes that gap directly. Creatures made in God's image, with the work of the law written on their hearts (Rom 2:14-15), are built to perceive a moral order that reflects the being who made both the order and them. - Think of a thermometer that reads the room correctly. If somebody calibrated it, there is nothing to explain. If it was assembled at random out of a parts bin, its accuracy is a puzzle. The naturalist has no calibration story available to him, which is why this objection lands harder on his own position than on the theist's. 3. Pressed all the way, it takes more than morality with it. - Our faculties evolved for survival across the board, not only where morality is concerned. No ancestor was ever eaten for failing at topology, and none left more descendants for grasping quantum field theory, yet we take those results to be true. - The published answer is that the cases differ. Believing there is a tiger in the grass when there is in fact a tiger in the grass pays immediately, so ordinary perception had a standing reason to track the truth. Moral facts, on the realist's own account, do no causal work at all, so tracking them pays nothing extra. That is where the debunker says his argument stops, and he says it for a reason rather than for convenience. - The line is harder to hold than it looks. Mathematics and logic earn no direct survival payoff either, and numbers do no causal work whatever. - So a naturalist who believes there really are numbers faces the same question about how creatures like us came to know anything about them. He owes a reason why the answer he gives there is unavailable in the moral case, because "it stops at morality" is the thing that needs arguing. 4. Grant the force of it. - The evolutionary account of where the moral sense came from is well supported, and no honest reply pretends otherwise. - What is disputed is the inference from origin to unreliability. How a belief arose and whether it is true are different questions, and an argument from the first to the second needs its own defense.

+ At most the argument gets an abstract moral lawgiver, not the God of Christianity. Nothing in it delivers the Trinity, the incarnation, or a being who answers prayer.

1. Granted: what is established is theism, not Christianity. - A deductive argument delivers what its premises contain and not one thing more. These premises are about the grounding of moral value, so the conclusion is about a ground of moral value, and that is where it stops. - A tape measure tells you a length and says nothing whatever about weight, and it is not a defective tape measure for that. An instrument is judged by whether it reads correctly, not by whether it reads everything. 2. What it does deliver is not thin. - A being whose own nature is the standard of goodness, not measured by anything outside himself. - A being to whom obligations are owed, which means personal rather than abstract, since only persons are owed things. That is the note Scripture strikes when it puts the demand in the second person: "He has told you, O man, what is good; and what does the LORD require of you but to do justice, and to love kindness, and to walk humbly with your God" (Mic 6:8). - A being whose reach is universal, since the obligations bind everyone. - Set that beside what the other arguments for God's existence reach, and the difference shows. A first cause could be an impersonal condition. A fine-tuner could be perfectly indifferent to what it tuned. Goodness is the one attribute natural theology has the hardest time getting to, and it happens to be the attribute at issue whenever anyone complains about evil in the first place. 3. Arguments of this kind are meant to be combined. - No single argument in natural theology gets the whole of Christian doctrine, and none is offered as though it did. The cosmological arguments reach a cause beyond space and time, the teleological arguments a designer, this one a moral ground. - A jury does not expect one witness to prove the case. It expects the fingerprint, the receipt and the motive to converge, and each is worth having precisely because it is partial. 4. The distinctively Christian claims rest on evidence of a different kind. - The Trinity, the incarnation and the resurrection are historical and revelational claims, argued from texts and events rather than from moral ontology. - No amount of reasoning about the grounding of moral value will tell you whether a particular tomb outside Jerusalem was empty on a particular Sunday. That question is settled by evidence of a completely different sort. - Keeping the two kinds of argument separate is a virtue rather than a gap. An argument that claimed to reach the resurrection from moral values alone would be overreaching, and easy to expose. See also: Natural Theology: The Moral Argument Natural Theology: The Kalam Argument CE / Resurrection: Maximal Data Method

+ Granting all of it, the problem of evil is untouched. You have shown that evil implies some moral ground, not why a good and all-powerful God permits a child's cancer.

1. Granted: none of this explains why God permits what he permits. - (C) says as much outright. There is no account here of God's reasons, and nobody sitting in a hospital corridor at two in the morning should be handed a proof in place of an answer. - Showing that evil entails God, and then treating the evils as though they therefore needed no answer, would be worse than useless. (P1) through (C) settle what evil implies about moral reality. What evil implies about God's goodness is argued at length in the logical, probabilistic and external forms of the problem, and neither job substitutes for the other. 2. What changes is the shape of the dispute, not the difficulty of it. - Before, the evils were offered as a reason to think God does not exist. After, they are a hard question for a position that has independent reason to be held. The suffering is exactly as bad either way, and the question of why it is permitted is exactly as hard. - Think of a defendant with an alibi confirmed on camera. The unexplained fingerprint on the door is still unexplained, and it still bothers everyone in the courtroom, but it is no longer the thing that decides the case. - Every worldview has questions it cannot yet answer. Which set of unanswered questions is more tolerable is a real conversation, and it is a different one from the argument from evil. 3. The same fact cannot be used on both sides of the ledger. - The atheist may press the evils as objectively evil and inherit the grounding problem, or treat them as attitudes and lose the objection's force. What is not available is objectivity when attacking and subjectivity when defending. - That is the whole claim, and it is narrow. It does not say the evils are acceptable. It says they cannot be both real enough to accuse and thin enough to cost nothing. See also: CO / PoE: Intellectual PoE: Logical CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Intellectual PoE: External

+ Saying goodness is God's nature only moves the arbitrariness. Either "good" means "resembling God", which makes "God is good" an empty compliment, or his nature meets a standard, and the standard is doing the work.

The published form of this is Wes Morriston's. He argues that a theist who stops at God's nature has explained nothing more than a non-theist who stops at brute necessary moral truths: both say "that is simply how it is" at exactly the same depth, and the theist's stopping point carries the extra embarrassment that calling it good appears to say nothing at all. 1. "God is good" is not a comparison to a yardstick. - The objection assumes that calling a thing good always means holding it up to something else. Most of the time that is right. A good knife cuts, a good argument is valid, and in both cases the standard sits outside the thing being judged. - Some things are not measured; they are the measure. For most of the last century the kilogram was a cylinder of platinum alloy in a vault outside Paris, and asking whether that cylinder weighed a kilogram was not a deep question about the cylinder. It weighed what a kilogram weighs. Everything else in the world was checked against it. - So "God is good" is not the empty report that God resembles himself. It is the claim that the standard is a living character rather than a rule or a list, and that this character is what the words just, faithful and merciful have been tracking the whole time. - The analogy has a limit, and it is exactly the one the objection presses. The cylinder was chosen, and a different cylinder could have been chosen instead. God's nature was not chosen. 2. Necessity is what keeps the answer from being arbitrary. - If God's nature were one option among several that happened to get picked, the arbitrariness charge would land squarely. On the historic Christian view it was never an option. God exists necessarily and is what he is necessarily, so there was no menu and no moment of selection. - Compare a demand to explain why seven is prime. No deeper fact made it prime, and there is no arrangement in which it is not. A stopping point that could not have been otherwise is a different kind of thing from a coin that happened to land heads. 3. The two stopping points are not on a level. - The real force of the objection is comparative: if the theist may stop at God's nature, the atheist may stop at brute moral facts, and nobody has gained anything by it. That is the strongest thing in the objection, and it deserves an answer rather than a definition. - The two stops differ in what they leave over. An obligation is not merely a truth about what would be good. It binds whether or not you care for it, breaking it calls for guilt rather than regret, and guilt is always guilt toward somebody. - A stopping point that is a person accounts for all three of those directly. A stopping point that is a proposition accounts for none of them, and still owes a story about why creatures shaped by survival should perceive it at all. - Put the difference in a courtroom. No man in the dock has ever been asked to answer to the law of gravity, and nobody has ever apologized to a theorem. He answers to the people he wronged and, on the Christian account, to the one who made them. - One starting point that does several jobs is worth more than a separate starting point for each job. That is the same comparison (D4) makes, and it is an inference to the better account rather than a proof. 4. Grant what the objection gets right. - The theist does stop somewhere, and "God's nature" is not a further explanation of goodness underneath goodness. Anyone who presents it as one is overselling. - The stopping point is a person rather than a proposition, and that is exactly the difference obligation needs. Whether it is enough is a live argument, and serious philosophers on both sides are still having it. See also: Natural Theology: The Moral Argument

+ There are no objective moral values, and I say so plainly. Nothing was really evil, so the second premise is false in the sense the argument needs, and I can still press the problem of evil as an inconsistency inside your own beliefs.

Moral error theory, the position J. L. Mackie set out, holds that moral statements really do try to describe objective features of the world and that all of them fail. "Torture is wrong" is not true-for-us and not a disguised expression of feeling; it is simply false, the way "Zeus is angry" is false. An error theorist can go on keeping his promises, loving his children and pressing the problem of evil as an internal charge against theism. 1. Against an error theorist the modus tollens does not run. - (IC) is exactly what he denies. Someone who holds that nothing is objectively evil has denied the premise the chain needs, and (C) does not follow against him. Saying otherwise would be a trick, and he should be told so plainly. - What he has done is pay the price rather than dodge it. The question is what the price buys and what it costs, and both are larger than they first look. 2. The cost is the argument he arrived with. - The problem of evil in the form nearly everyone meets asks the reader to look at a burned child and feel the weight of it. On error theory there is no weight to feel. There is a powerful reaction, and alongside it the settled belief that the reaction reports nothing whatever about the world. - What remains is the internal charge that theism's own commitments do not hang together. That charge is serious and it has answers of its own. - What it cannot be is a moral reason for him to disbelieve. On his own account no wrong was done that God failed to prevent, so the accusation everyone else is making is never his to make. 3. The premises that get him there are less certain than the thing they deny. - The usual route is the argument from queerness: objective values would be entities unlike anything else in the world, and the faculty for detecting them would be unlike any other faculty, so the reasonable conclusion is that there are none. - Set that beside what it is being used to overturn. Which is more certain, that a philosopher's list of the kinds of things that exist is complete, or that setting a child on fire for entertainment is really wrong? Most people will not trade the second for the first, and many who find the queerness argument interesting will not either. - Strangeness is a weak argument against existence in any case. Numbers, logical necessity and consciousness are all strange, and every one of them is far easier to describe than to eliminate. - The queerness argument also does not stop where the error theorist wants it to stop. That you ought to proportion your belief to the evidence is no fact about particles either, and it binds you whether or not you happen to care about truth. • Oughts of that kind, about what a person should believe rather than what he should do, are called epistemic. They are every bit as queer as the moral ones. - So take queerness seriously enough to discard the moral oughts and the epistemic oughts go out with them. That leaves the error theorist holding that his own reason for believing error theory is not really a reason at all. - A few philosophers accept exactly that, and say so. It is a heavy thing to carry. 4. The most careful error theorists treat the loss as real. - They do not shrug and move on. Some propose keeping moral practice running after the truth claims are withdrawn, by treating moral talk as a fiction we go on using deliberately, knowing it is not true, the way a man keeps using a map he knows is out of date because it is better than nothing. - That is honest work and it is also an admission. Something worth keeping was lost, and a substitute had to be built for it. - None of that refutes the position. It does show the choice is not between a costly theism and a naturalism that costs nothing. See also: Natural Theology: The Moral Argument CO / PoE: Intellectual PoE: Logical

+ The serious arguments from evil use no moral premise. They compare how likely the observed pattern of pain and pleasure is if a good God exists against how likely it is if nothing cares, and that runs whether or not anything is objectively evil.

The version worth answering is Paul Draper's. He sets theism beside the hypothesis that whatever produced us is simply indifferent to us, asks which better predicts the pattern of pain and pleasure that biology actually shows, and argues that indifference predicts it better. The word "evil" need not appear anywhere in it. 1. An argument with no moral premise is not touched by (P1) through (C). - The chain runs on objective evil from beginning to end. An argument that never asserts objective evil is not answered by it, and claiming otherwise would be a bait and switch. - So this route is open, and a well-read objector should take it. What it changes is the shape of the objection, and the change is not free. 2. The comparison still needs a judgment about what God would do. - To say the pattern is unlikely if God exists, you have to say what a perfectly good and all-powerful being would probably bring about. That is not a claim that the suffering is objectively evil. It is still a claim about what goodness would produce. - A juror who says an honest man would have kept the receipts is making exactly that kind of claim, and nobody thinks he has thereby accused anyone of anything. He has said what honesty looks like from the outside. - The objector may borrow the standard from the theist and use it hypothetically, which is perfectly legitimate, and which puts him back where (D3) leaves him. He asserts nothing, and so concludes at most that the theist's own picture fits the data badly. 3. The conclusion changes with the form. - A probabilistic comparison does not show that theism is false. At most it shows that one body of evidence favors a rival hypothesis, and evidence favoring a rival can be outweighed by other evidence. - Everything then rests on a probability nobody is placed to write down. To say the pattern of suffering is improbable given God is to judge what a perfect being would likely arrange, and that judgment takes a survey of the goods open to him and of which goods are tied to which costs. - A toddler who cannot see why the surgeon is hurting his mother has not thereby acquired evidence that the surgeon is cruel. The gap being appealed to here is a good deal wider than that one. - The name for that reply is skeptical theism, and its skepticism is narrow. It is not doubt about God's goodness. It is doubt that a finite mind can survey the possible goods well enough to run the comparison at all. - Draper's answer is that theism simply fails to lead us to expect the pattern, so no such survey is required. The exchange over that is live rather than finished. - All of which is why the full reply is a case rather than a sentence. The reasons for thinking God exists have to be set against it, and the predictions attributed to theism have to be checked one at a time. 4. The moral vocabulary is hard to keep out for long. - The datum is chosen because it matters. The same argument built on the distribution of blue-eyed animals across species would be a curiosity, not an objection, and nobody would write a book about it. - That does not defeat the argument, since the objector can say pain matters to the sufferer without saying anything is objectively wrong. It does explain why the cold version is rarely the one anybody actually runs, and why the version most readers meet is the one (P1) through (C) address. See also: CO / PoE: Intellectual PoE: Probabilistic CO / PoE: Animal Suffering

+ Christianity's own tradition says evil is not a thing but a privation, an absence of a good that ought to be present. If that is right, "evil exists" is false by your own account, and the second premise collapses.

1. The privation account answers what evil is, not whether there is any. - Privatio boni, Latin for "the privation of good", is Augustine's answer to a different question altogether: whether evil is a thing God made. His answer is that it is not a substance at all but a corruption of something good, the way rot is not an ingredient somebody added to the fruit but the fruit's own goodness failing. - Blindness is the standard illustration, and it is exact. Blindness is the absence of sight, and no eye doctor has ever concluded from that description that his blind patients are imagining it. Absences can be perfectly real, perfectly detectable and perfectly terrible. - So the account never says there are no evils. It says what an evil turns out to be when you look closely at one. 2. A privation is measured against how a thing ought to be. - Call blindness an absence and you have already said what an eye is for. The word works only because there is a standard the eye is falling short of, and nobody has to invent that standard for the diagnosis to be correct. - Cruelty is the same shape. To call it a privation is to say a man failed to be what a man owes his neighbor, which is a judgment about how things ought to stand between them. - So the account does not compete with (IC). It leans on it. An absence of a good that ought to be present has an objective moral standard built into the definition, and the objection has handed over the very thing (IC) asserts. 3. (P2) needs only that some acts and states are really evil. - Read the premise back with the privation account in hand: some things are really as they ought not to be. Nothing in that requires evil to be a substance or an ingredient. - Augustine, who worked the account out, also wrote the ancient world's most sustained descriptions of sin, judgment and a wrecked human will. He was not describing an illusion and did not think he was. 4. Nothing in the chain depends on the account being right. - Christians have disagreed about privation. Some hold that evil is better understood as a positive disorder of the will than as a mere lack, and (P1) through (C) stand either way. - What every side affirms is the thing (P2) asserts: cruelty is not merely unwelcome, it is really wrong. That is common ground across the tradition, and it is all the premise needs.