Moral Argument

Most confusion in this debate comes from collapsing questions that must be kept apart. Moral epistemology asks how we know what is right; moral ontology asks what makes it right; moral semantics asks what moral language means; moral motivation asks why we should care. The argument is exclusively about ontology.

This matters because the most common objection — 'atheists can be good people' — answers the epistemology and motivation questions, which no serious version of the argument denies. Of course an atheist can know that cruelty is wrong and can refuse to be cruel. The question is what makes it wrong, whoever recognizes it.

The usual support is our strong moral intuitions in clear cases. The torture of children for entertainment is not merely disapproved of in our culture; it is wrong, and it would be wrong in a society that unanimously approved it. Moral progress presupposes this: to say abolition was progress rather than mere change is to say the earlier consensus was mistaken about something.

This premise is defended by many atheist philosophers as well. Erik Wielenberg, Russ Shafer-Landau, and Derek Parfit all defend robust moral realism without God, which is exactly why the argument's weight falls on the second premise.

The main challengers are error theory, which holds that all moral claims are false because nothing has the required properties, and evolutionary debunking arguments, which hold that our moral faculties were shaped for reproductive fitness rather than truth. Sharon Street's debunking argument is the most cited; interestingly it targets secular realism at least as hard as theistic accounts, since the theist can hold that God shaped the faculties to track moral facts.

The version most defended today is not divine command theory simpliciter but a modified divine command theory, associated with Robert Adams and William Alston. Moral values are grounded in God's necessary nature, which is essentially good; moral obligations are constituted by the commands of that essentially good being. Values and duties thus have distinct grounds, which is the key structural move.

The argument for this is comparative. Duties are relational: an obligation is something owed to someone, and it plausibly requires a person with authority to impose it. On naturalism, the difficulty is explaining how impersonal facts generate binding obligations rather than merely describing regularities or preferences. On Platonism, abstract objects are causally inert and it is unclear how the abstract object Goodness obligates anyone.

Wielenberg's non-natural non-theistic realism is the strongest rival and should be engaged directly: he argues that moral facts can be brute necessities requiring no further grounding. The theistic reply is that brute necessary moral facts are more mysterious than a necessary personal being, but this is an argument about explanatory economy, not a refutation.

Plato's question, adapted: is something good because God commands it, or does God command it because it is good? If the first, then goodness is arbitrary and God could have commanded cruelty. If the second, then goodness is independent of God and the argument collapses.

The standard theistic answer takes the third option: God's commands flow necessarily from his own nature, which is the standard of goodness itself. Goodness is neither above God nor arbitrarily invented by God; God is the paradigm. Cruelty is impossible to command not because an external rule forbids it but because it contradicts what God essentially is.

The serious follow-up is whether 'God is good' then becomes empty — a tautology meaning only 'God is God-like'. The reply is that the term picks out a real character with content: faithful, just, self-giving, opposed to cruelty and betrayal. Whether that reply fully succeeds is a live question, and the argument is stronger when presented with the follow-up acknowledged rather than treated as settled by the standard answer.

If successful, the argument supports a personal, necessarily good, authoritative moral ground — which is much closer to classical theism than the conclusion of the cosmological argument is. It does not establish any particular revelation, and it does not resolve difficult ethical cases; a theist and an atheist can share every moral judgment while disagreeing about ontology.

It also carries a cost that should be admitted: the theist must handle the moral difficulties in scripture, since a divine-command framework invites the immediate question of what happens when the commands recorded appear unjust. That is a separate argument, but it is where the conversation almost always goes next.