The Moral Argument

The moral argument infers God from the existence of morality itself: if there are objective moral values and duties — if some things are really wrong, independently of what anyone thinks — then (the argument claims) there must be a God to ground them, because no merely natural fact could confer genuine, categorical authority on a moral demand. Unlike the cosmological and teleological arguments, which start from features of the external world, the moral argument starts from the deliverances of conscience, and it has a distinctive practical appeal: it speaks to the felt absoluteness of obligation. It is also the argument most tightly bound to a counter-argument, the Euthyphro dilemma, which many take to show that morality cannot depend on God after all.

This page sets out Kant's practical version and the modern metaethical version (Adams, Craig), distinguishes the argument from the separate claim that atheists cannot be moral, and then turns to the Euthyphro dilemma and the secular-ethics challenge that are its principal obstacles.

Two forms of the argument

The label covers two quite different arguments that should not be confused.

Kant's practical postulate

Kant did not argue that God is needed to ground moral truth — for him morality is autonomous, binding on any rational agent through the Categorical Imperative regardless of God. His argument is practical: morality commands us to seek the highest good (summum bonum), the union of perfect virtue with the happiness it deserves. But we can neither achieve perfect virtue within a finite life nor guarantee that virtue is rewarded with happiness — nature is indifferent to desert. For the moral enterprise to be rational rather than absurd, we must postulate what would make the highest good attainable: immortality (time enough to perfect virtue) and God (a power that proportions happiness to virtue). God is thus a postulate of practical reason — not proven to exist, but something the moral agent is rationally entitled, indeed required, to believe in to avoid the incoherence of pursuing an impossible end. (Compare the free-will postulate, which Kant treats the same way.)

The metaethical (value-grounding) argument

The version dominant today is metaethical: it argues that God is the best — or the only — explanation of objective morality. Its canonical form (Craig):

1. If God does not exist, objective moral values and duties do not exist.
2. Objective moral values and duties do exist.
3. Therefore God exists.

The work is in premise 1, defended by elimination: if morality is objective (premise 2, taken as a strongly held datum — that the torture of innocents is really wrong, not merely disapproved), what could make it so? Natural facts (pleasure, survival, social contract, evolved sentiment) seem unable to generate categorical, authoritative, prescriptive obligations — an "ought" that binds you whether or not you care. Evolution explains why we feel bound but, if that is the whole story, debunks the objectivity (it would have instilled whatever dispositions aided reproduction). A personal God, by contrast, can be the source of authoritative commands and the standard of goodness. So the best explanation of objective morality is theism.

Divine Command Theory

The metaethical argument is usually backed by a metaethics: Divine Command Theory (DCT), on which moral obligations are (or are constituted by) God's commands — an act is obligatory because God commands it, wrong because God forbids it. DCT gives moral duties exactly the features that puzzle the naturalist: they are prescriptive (commands are intrinsically action-guiding), authoritative (issued by a competent authority with standing over us), and objective (independent of our attitudes). It is the natural partner of the moral argument: if DCT is the right account of obligation, then no God means no obligations.

The classic objection to DCT — and to the moral argument that leans on it — is the Euthyphro dilemma.

The Euthyphro dilemma

In Plato's Euthyphro, Socrates asks: is the pious loved by the gods because it is pious, or pious because it is loved by the gods? Transposed to DCT:

Is an action good because God commands it, or does God command it because it is good?

Both horns appear fatal:

• First horn — good because commanded (the arbitrariness horn). If actions are good solely because God commands them, then morality is arbitrary: had God commanded cruelty, cruelty would be good. "God is good" reduces to the empty "God does what God commands," and God's goodness can no longer be praised (it is true by definition). Worse, it seems possible that gratuitous torture could have been obligatory — a conclusion most find a reductio.
• Second horn — commanded because good (the autonomy horn). If God commands actions because they are already good, then goodness is a standard independent of God, to which God himself defers. Morality then does not depend on God after all — the first premise of the moral argument is false, and an atheist can help herself to the same independent standard.

The modified-DCT reply

The most influential theist response, Robert Adams's modified divine command theory, threads between the horns by distinguishing goodness (value) from obligation (duty):

• The good is not commanded into being; it is grounded in God's own nature — God is the supreme good, the standard of value, and finite things are good by resembling or participating in that nature. So goodness is neither arbitrary (it is fixed by God's unchanging, essentially loving nature, not by sheer will) nor independent of God (it is God's nature).
• Obligation — what we are required to do — is grounded in the commands of this essentially loving God. Because the commander is essentially good and loving, the arbitrariness worry lapses: such a God cannot command gratuitous cruelty, since that would contradict the very nature that constitutes the good.

The dilemma is thus claimed to be a false one: there is a third option — goodness is neither independent of God nor a product of arbitrary will, but identical with God's nature. The critic presses back: either God's nature is good by some standard (re-opening the second horn at one remove — why is this nature good?) or "God's nature is the standard of goodness" is stipulative and the arbitrariness returns one level up. Whether grounding value in God's nature genuinely escapes the dilemma or merely relocates it is the heart of the religion-and-morality debate.

"Can atheists be moral?" — a distinction

The moral argument is frequently confused with an offensive and irrelevant claim, and the confusion must be cleared. The argument is not that atheists cannot behave morally or know right from wrong. Defenders concede readily that atheists can be — often are — morally exemplary, and can have full epistemic access to moral truths. The claim is ontological / explanatory, not epistemological or behavioural:

• Epistemology (knowing what is right) — atheists have it; conscience does not depend on belief in God.
• Ontology (what makes things right) — this, the argument contends, requires God.

So "atheists can be good" is fully compatible with the moral argument; the theist's claim is that the atheist enjoys objective morality while lacking an account of its foundation — living, as it were, on borrowed metaphysical capital. The atheist's task is therefore not to display moral behaviour but to provide the missing foundation without God.

The secular-ethics challenge

The decisive question for premise 1 is whether objective morality can be grounded without God. The atheist has several well-developed options, and their availability is what keeps the moral argument from being a proof:

• Non-natural moral realism / Platonism (Moore, the "Cornell realists," Parfit, Enoch): objective moral facts exist as sui generis, non-natural features of reality, neither reducible to natural facts nor dependent on God — abstract truths on a par with mathematical ones. If moral facts can be brute necessary truths, premise 1 fails: morality is objective and godless.
• Naturalist realism (Railton, the neo-Aristotelians, Foot's "natural goodness"): moral facts are constituted by natural facts about flourishing, well-being, or rational agency, knowable by ordinary inquiry.
• Kantian constructivism (Korsgaard, Scanlon): objectivity is grounded in the constitutive demands of rational agency or in what no one could reasonably reject, not in any external commander.
• Contractarianism and evolved cooperation: morality is the rational output of mutually beneficial agreement, with an evolutionary backstory explaining moral motivation.

Against all of these stands the debunking worry from the other side: Nietzsche and Mackie argued that if God (or any transcendent ground) goes, objective values go with him — Mackie's "argument from queerness" holds that objective, intrinsically prescriptive values would be metaphysically odd entities unlike anything else in a naturalistic world, so we should be moral error theorists and deny that any moral claims are true. Notably, this is the moral argument's first premise accepted — "no God, no objective morality" — but its conclusion reversed: the error theorist takes the absence of God to show morality is an illusion, where the theist takes the reality of morality to show there is a God. Dostoevsky's "if God does not exist, everything is permitted" (Ivan Karamazov) is the literary form of the shared premise. Which way to run the modus — tollens (no God, so no real morality) or ponens (real morality, so God) — depends entirely on how confident one is in objective morality versus in atheism, and on whether the secular groundings above succeed.

Where the moral argument sits

The moral argument trades the external evidence of the cosmological and teleological arguments for the inner evidence of obligation, and its logic is a standoff over which premise is more certain: the theist runs modus ponens (objective morality exists, so God does); Nietzsche and the error theorist run modus tollens (no God, so objective morality is an illusion) — both granting the shared conditional that morality needs a transcendent ground. The argument's strength is the vividness of moral experience; its weakness is that the shared conditional is exactly what the secular realist denies, offering Platonist, naturalist, or constructivist groundings of objective morality without God. And even granting the conditional, the Euthyphro dilemma threatens to show that morality cannot rest on divine command without becoming arbitrary — a threat the modified DCT of Adams answers by grounding goodness in God's nature, with the debate then turning on whether that move escapes the dilemma or merely relocates it. The full development belongs to religion and morality; as a theistic argument, it is, like its siblings, one weighted strand in a cumulative case.