The Moral Argument

Romans 1 said that creation shows God's eternal power and divine nature. Romans 2 adds a second witness, inside the person rather than overhead. Gentiles who have never been given the law of Moses still do by nature what the law requires; the work of the law is written on their hearts; their conscience bears witness; their thoughts accuse or defend them. Paul does not say this to compliment the nations. He says it as part of an argument that every person, Jew and Gentile alike, stands accountable before God, on the day when God will judge the secrets of men through Christ Jesus (2:16). Moral knowledge is universal because moral accountability is universal.

Amos made the same assumption eight centuries earlier. His oracles against the nations condemn Damascus, Gaza, Tyre, Edom, Ammon and Moab for war crimes: threshing a conquered people with iron sledges, selling whole populations into slavery, ripping open pregnant women to take land. None of these nations had received the law at Sinai, and Amos never suggests they needed it to know better. Only Judah is condemned for rejecting the Law of the LORD (Amos 2:4). Scripture takes for granted that some things are known to be wrong by every people under heaven. C. S. Lewis made the same point in the appendix to The Abolition of Man (1943), where he set side by side moral teachings from ancient Egypt, Babylon, China, India, the Norse, the Greeks and the Hebrews, and showed how much they share: honour toward parents, faithfulness to promises, mercy to the weak, justice in judgment.

Who exactly are the Gentiles of Romans 2:14? Here the commentators divide, and you should know it. Many, including Douglas Moo, read them as non-Christian Gentiles who sometimes do what is right, which shows that they know the law and so cannot plead ignorance. Others, including C. E. B. Cranfield and N. T. Wright, read them as Gentile Christians, in whom the new covenant promise of Jeremiah 31:33, a law written on the heart, has been fulfilled. The moral argument draws mainly on the first reading. But it does not depend on this verse alone: Amos 1 and 2, Genesis 20, and the whole biblical assumption that the nations are judged for what they did, all point the same way.

One more thing must be said before any argument is made, because getting it wrong ruins everything that follows. Romans 2 says that people without the law sometimes do what is right. So the moral argument is not the claim that unbelievers cannot be good, or cannot know what is good. Many atheists are kinder, more honest and more just than many Christians, and the Bible expects some of them to be. The moral argument is about something else: not whether people can know and do good without believing in God, but what goodness and obligation are, and whether they can exist at all if there is no God. That is a question about the foundation of morality, not about the behaviour of unbelievers. A student who says to an atheist, so you have no reason to be good, will have lost the argument and possibly a friend.

The point lands well beyond the West. Across southern Africa people say umuntu ngumuntu ngabantu, a person is a person through other persons: a moral vision of human dignity and mutual obligation that no one learned from a philosophy book. Much of the global south has never doubted that the moral law is real. The question there is often not whether right and wrong exist but who gives the law and who will judge: the ancestors, the community, the spirits, or the God who will judge the secrets of men through Christ Jesus.

C. S. Lewis gave the argument its most widely read modern form in the radio talks he gave on the BBC during the Second World War, later published as Mere Christianity (1952). He begins with an ordinary quarrel. When two people argue about whether one has treated the other fairly, they both appeal to a standard they expect the other to know, and the guilty one does not say that the standard does not exist; he makes excuses. Lewis calls this standard the Law of Human Nature, and he draws two facts from it: people everywhere think they ought to behave in a certain way, and they do not in fact behave that way. Lewis then argues that a law of this kind, one that tells us what we ought to do rather than describing what we do, points to a mind behind the universe that is intensely interested in right conduct. He also told how this reasoning undid his own earlier atheism: his argument against God was that the universe seemed cruel and unjust, and he came to ask where he had got the idea of just and unjust from, since one cannot call a line crooked without some idea of a straight one.

William Lane Craig puts the argument in three lines that are now standard. If God does not exist, objective moral values and duties do not exist. Objective moral values and duties do exist. Therefore God exists. The word objective matters. A moral claim is objective if it is true whether or not anyone believes it: torturing a child for fun would be wrong even if everyone on earth, after some catastrophe of propaganda, came to approve of it. The distinction between values and duties also matters. Values are about what is good or bad; duties are about what we are obliged to do or not to do. Many philosophers think the argument is stronger for duties, because an obligation seems to require someone to whom it is owed.

Others have developed the argument in different directions. Immanuel Kant, in the Critique of Practical Reason (1788), did not offer a proof of God, but argued that the moral life requires us to postulate God, since only God could guarantee that virtue and happiness, which so often come apart in this life, are finally united. John Henry Newman, in his Grammar of Assent (1870), argued from the experience of conscience: we feel responsibility, shame and fear when we do wrong, and these feelings imply one to whom we are responsible. Robert Adams, in Finite and Infinite Goods (1999), developed the most careful modern version: goodness is resemblance to God, who is the Good itself, and moral obligations are constituted by the commands of a loving God to persons with whom he is in relationship. C. Stephen Evans, in God and Moral Obligation (2013), extended Adams's case for obligation in particular.

Notice the strength of the argument. It does not appeal to a gap in science, and no future discovery in biology will remove its force. It starts from something nearly everyone already believes, that some things are really wrong, and it asks what has to be true about the universe for that belief to be true. And if it works, it shows something about God that the cosmological and design arguments do not: that God is good, and that he cares about how we treat one another. That is much closer to the God of Micah 6:8, who has shown you, O man, what is good.

The argument meets two quite different kinds of atheist reply, and they come from opposite directions. The first accepts the argument's first premise and denies the second. J. L. Mackie, in Ethics: Inventing Right and Wrong (1977), argued that objective moral values do not exist. His argument from queerness says that if such values existed, they would be entities or properties utterly unlike anything else in the universe, which somehow have to-be-pursuedness built into them, and we would need a strange faculty to detect them. Since a naturalistic universe contains nothing of that kind, morality is something humans have invented, however useful and however deeply felt. This is called error theory: our moral claims aim at objective truth, but they are all false. Strikingly, in The Miracle of Theism (1982), Mackie conceded that if there were objective values, they would make the existence of God more probable than it would otherwise be. He simply denied that there are any.

The second kind of reply accepts the second premise and denies the first. Erik Wielenberg, in Value and Virtue in a Godless Universe (2005) and Robust Ethics (2014), argues that there are objective moral truths, that they are necessary truths like the truths of mathematics, and that they need no foundation outside themselves. That pain is bad and that cruelty for its own sake is wrong are, on his view, brute ethical facts: they are not grounded in anything further, and they do not need to be, any more than the truth that two and two make four needs God to command it. Derek Parfit, in On What Matters (2011), defended a related view, as do David Enoch in Taking Morality Seriously (2011) and Russ Shafer-Landau in Moral Realism: A Defence (2003). Other atheist philosophers, sometimes called naturalist moral realists, such as Richard Boyd and Peter Railton, argue that moral facts are natural facts about what makes human beings and communities flourish.

Put those two replies side by side and the shape of the argument comes into view. Mackie, the error theorist, agrees with the theist that objective morality would point beyond the natural world, and so rejects objective morality. Wielenberg, the robust realist, agrees with the theist that objective morality is real, and so rejects the claim that it needs God. Each atheist is denying a different premise. A student who answers Mackie with an argument against Wielenberg, or the reverse, has not understood the conversation.

The theist's reply to Mackie is that error theory asks too much. If torturing a child for fun is not really wrong, then nothing is, and most people, including most atheists, are more certain that it is wrong than they are of any philosophical premise Mackie uses to deny it. The theist's reply to Wielenberg is harder and more interesting. Are moral facts really like mathematical facts? A duty seems to make a claim on a person; a sum does not. Brute moral facts, existing necessarily, with nothing about them that explains why they bind a particular kind of creature, can look as strange as Mackie said they would. Wielenberg has answered this at length, and this is one of the most active debates in current philosophy. Know that the theist does not win it by default.

The oldest objection comes from Plato. In the dialogue Euthyphro, Socrates meets a self-styled religious expert, Euthyphro, outside the court in Athens; he is about to prosecute his own father, confident that he knows what piety requires. Socrates asks him a question that has outlived Athens: is the holy loved by the gods because it is holy, or is it holy because it is loved by the gods? Turned toward the moral argument, the question becomes: is something good because God commands it, or does God command it because it is good? If the first, morality looks arbitrary: had God commanded cruelty, cruelty would have been good. If the second, goodness is a standard independent of God, which he too must follow, and then we do not need God to ground it.

Both horns have been taken by believers. The first, sometimes called voluntarism, holds that God's will simply is the standard. William of Ockham in the fourteenth century is often read as holding that God could have commanded even hatred of himself, and it would then have been right. Within Islam, the Ash'arite school, following al-Ash'ari in the tenth century, held that acts are good or evil only because God commands or forbids them, against the Mu'tazilites, who held that reason can know good and evil and that God commands what is good because it is good. A Christian talking with a Muslim friend should know that this argument has been had in their tradition too, at length.

Most Christian philosophers today reject both horns and propose a third option. Goodness is neither a standard above God nor an arbitrary decision of his will; it is grounded in God's own nature, which is essentially and unchangeably good. God commands what he commands because he is good, and his commands flow from his character, so they could not have been cruel, any more than God could lie (Titus 1:2, Hebrews 6:18). William Alston set out this reply in 1990, and Robert Adams built his theory on it: the Good is God himself, and other things are good insofar as they resemble him. Psalm 119:68 says it in one line: You are good, and You do what is good. Jesus put it more sharply: No one is good except God alone (Mark 10:18).

Critics have answered that this simply moves the dilemma back. Is God's nature good because it meets some standard of goodness? Then the standard is independent after all. Or is God's nature called good simply because it is God's? Then to say God is good is empty, like saying God is God. The theistic reply, developed by Adams, is that the identity of the Good with God is discovered, not stipulated, just as the identity of water with H2O was discovered: we come to know goodness first in its effects, as justice, mercy and love, and then learn that its source is God, so that God is good is not a tautology but a substantive truth. Philosophers such as Wes Morriston have pressed further objections, asking for instance why a necessarily existing God's particular nature should be the standard rather than some other. The fair summary is that the Euthyphro dilemma is not the knock-down objection it is often presented as, and that the third option is coherent and widely held, and also that it is still argued over.

Do not miss the pastoral point inside this abstract debate. If good were simply whatever God happened to command, then a person could never appeal to God's goodness, as Abraham did when he asked whether the Judge of all the earth would not do what is right (Genesis 18:25), because right would just mean whatever the Judge did. The Bible's prayers, its complaints, its laments and its trust all assume that God's goodness is not arbitrary but is who he is, and that it can be known, at least in part, by those made in his image. That is the third option, lived out on the knees.

The most important recent objection is that evolution explains our moral beliefs without their needing to be true. Our ancestors survived by cooperating, caring for their young, punishing cheats and keeping faith with allies, and natural selection would have favoured those who felt strongly that these things were right. The biologist E. O. Wilson and the philosopher Michael Ruse argued in the 1980s that morality is an adaptation, and that our sense that it is objective is itself part of the adaptation, a useful illusion. Richard Joyce, in The Evolution of Morality (2006), developed the argument with great care: if evolution explains why we have the moral beliefs we have, whether or not they are true, then we have no reason to trust them as tracking objective truth.

Sharon Street, in a much-discussed paper of 2006, A Darwinian Dilemma for Realist Theories of Value, sharpened this into a dilemma for any moral realist. Either evolutionary pressures and the objective moral truths are related, or they are not. If they are not, then it would be a remarkable coincidence if the beliefs evolution gave us happened to match the truths, so we should doubt our moral beliefs. If they are related, the realist owes an account of how, and the simplest account is that our values just are what evolution produced, with no independent truths needed. Street's own view is that moral truths are real but constructed from human attitudes, not discovered.

Christian philosophers have turned this argument around. If naturalism is true and evolution produced our moral beliefs with no reference to their truth, then Street is right and the naturalist moral realist has no reason to trust his moral knowledge. But if God exists and intends us to know the good, he could use evolution, or any other process, to produce moral faculties that are reliable, in the same way that he gives us eyes that reliably see. So the debunking argument is a problem for atheist moral realism, not for theism. Atheist realists reply with what is called a third-factor account: survival and flourishing are themselves connected to moral truths, since pain really is bad and cooperation really is good, so a creature shaped to avoid pain and value cooperation will tend to track moral truth without any God. Enoch and Wielenberg have developed this reply carefully, and it is not easily dismissed.

Be aware that the debunking strategy can also be aimed at belief in God. The cognitive science of religion asks why belief in gods is so widespread, and some researchers propose that human minds have a tendency to detect agents everywhere, a rustle in the grass taken for a predator, which would make belief in invisible agents natural whether or not any exist. Justin Barrett, a Christian and a leading researcher in this field, argued in Why Would Anyone Believe in God? (2004) that this shows belief in God is natural, not that it is false; a God who wanted to be known might well make creatures inclined to look for him. The structure of the debate is the same as the moral one: whether an evolutionary explanation of a belief undermines it depends on whether there is any connection between the process and the truth. The genetic fallacy, judging a belief false because of how it arose, is a fallacy in both directions.

If the moral argument succeeds, it shows that there is a source of objective goodness and obligation that is personal, since obligations are owed to persons, and good, since goodness has its foundation in it. That is far more of the God of the Bible than the cosmological argument reaches. It does not show that this God is triune, that he has become incarnate, or that he forgives. In fact it does something uncomfortable first: it shows us under judgment. Paul's argument in Romans 2 runs from the law written on the heart to the day when God will judge men's secrets. The moral argument, rightly preached, ends not in a proof but in a need. Every honest person who accepts its conclusion has also accepted that he has not lived up to the standard he now knows to be real. That is where the gospel begins.

It also has limits you must state. Both premises are contested by serious philosophers. Mackie's error theory is a coherent position, though most people find it unliveable; Wielenberg's robust realism is a coherent position, and many philosophers hold some version of it. The argument is strongest when you are speaking to someone who is sure that some things are really wrong, as almost everyone is, and who has not thought about what that implies. It is weakest when someone has already decided that morality is a human construction and is willing to live with that.

Now add the third paragraph to your page headed My defence of the faith. Write the moral argument in three lines in your own words. Under it, write the reply you think the people around you would most likely give, whether error theory, robust realism, the Euthyphro dilemma or evolutionary debunking, stated as its defenders would put it. Then write your answer in five sentences or fewer, and then: This argument shows... and: It does not show...

Then go back to what you wrote in the inquiry, the thing you are sure is wrong. Ask one person this week, a colleague, a classmate, a neighbour, whether they agree it is wrong and why they think so. Do not argue. Listen, and write down their reason as fairly as you can. You will find that most people have never been asked, and that their answer tells you which of the replies in this lesson you will actually meet. If they give you a reason that is not in this lesson, write it down: that is the next thing you need to study.