The Moral Argument

Why "good" and "evil" feel like discoveries, not inventions.

If you sat someone down — anyone, of any culture, of any age — and told them you had just driven past a man torturing a child in a parking lot, for fun, and walked on, they would not say I cannot evaluate your account without a value system. They would say that is evil. You should have stopped him.

That reaction — the immediate, prephilosophical recognition that some things are actually wrong — is, I want to argue, a fact about the world that needs an explanation. And the explanation that makes the best sense of it is theism.

This is the moral argument. It is older than Christianity, and it has had many versions; the one I find sharpest goes like this.

The argument

1. If objective moral values and duties exist, God exists.
2. Objective moral values and duties do exist.
3. Therefore God exists.

The argument is not subtle. Almost every disagreement with it is about the second premise, and almost everyone who actually rejects the second premise turns out, on closer inspection, to not really reject it — they reject it in theory and operate by it in practice.

On (2): the moral realism premise

The claim of (2) is that some things are wrong whether or not anyone thinks they are wrong. Torturing children for fun is wrong even if the entire human race were brainwashed into approving of it. The Nazis were wrong, and they would have been wrong even if they had won the war and successfully indoctrinated the survivors into thinking the Holocaust was good.

Notice what you are tempted to say in response. You are tempted to say yes, of course. This is moral realism. It is what almost every human being actually believes when they are not actively philosophizing themselves out of it.

It is also what nearly every moral conversation presupposes. When you argue that something is unjust — that a policy is racist, that a war is unjust, that an employer is exploiting workers — you are not asking whether enough people feel this way. You are asserting that the wrongness is real, and that disagreement is mistake.

On (1): why this requires God

This is the contentious step. Why can’t moral facts just be there in the universe the way mathematical facts are there in the universe? Why does moral realism require a God?

The argument has a few moving parts. The clearest version, I think, is this:

Moral duties are obligations to someone. If I have a duty not to torture a child, that duty is owed to somebody — minimally, to the child. But when the obligation is unconditional, and binds even when no one is watching, even when I could get away with it, even when no human party to the act would ever know — that has the shape of an obligation to a third party. The shape of an obligation owed to a person who knows what I have done.

Moral facts have the structure of commands. When we say X is wrong, we do not mean X has the property of being a polygon. We mean X must not be done. Commands require a commander. A moral fact without a moral lawgiver has the form of a sentence with no speaker — and it is unclear how that gets the prescriptive force we mean by moral.

Moral facts assume the dignity of persons. Almost every moral intuition we have is grounded in some sense in the inviolability of the person. But if persons are, on the naturalist view, accidental arrangements of matter — molecules in motion — it is not obvious why they would have the inviolable dignity that grounds the moral intuition. The Christian story, by contrast, has a clean answer: every person is made in the image of God, and the dignity of persons flows from the dignity of the one in whose image they are made.

None of these moves is a knock-down argument. But each is, I think, better explained on theism than on naturalism. The naturalist has to do extra work — extra postulation, extra ad-hoc machinery — to get to the moral realism the rest of us all assume.

The honest naturalist responses

I want to take the naturalist response seriously. There are basically three.

Moral facts are reducible to brain states. Sam Harris has argued this most prominently. The Moral Landscape tries to ground moral facts in measurable facts about flourishing or suffering. The problem is that this only gets you a description of which states are preferred; it does not get you the prescription that you should prefer them. The is-ought gap is not closed; it is just relocated.

Moral facts are evolutionary. They evolved because they conferred survival advantage on the groups that held them. This is almost certainly true at the descriptive level — our moral intuitions have an evolutionary backstory. But this account, taken seriously, undermines the moral realism we started with. If my moral intuitions are there because they helped my ancestors survive, I have no reason to think they are tracking moral reality; I have reason to think they are tracking reproductive success. The argument collapses (2) into a form of error theory: moral facts are not objective, they only feel objective. This is a coherent position; it is also a position most people abandon the moment a child is actually being harmed.

Moral facts are conventions. Wrongness is whatever the community agrees is wrong. This is the position that, taken seriously, makes the Nazis right inside their own community, and so almost no one actually holds it. It survives mostly as a kind of moral relativism that everyone professes in undergraduate seminars and abandons before they leave the building.

What this gets you

The moral argument does not, by itself, get you to the God of the Bible. It gets you to a moral lawgiver — a being whose existence grounds the prescriptive force of moral facts.

But notice that this is the second time we have been led, by a serious argument, to the same kind of conclusion. The contingency argument led us to a necessary being. The fine-tuning argument led us to a cosmic designer. The moral argument leads us to a moral lawgiver. These are pointing at the same person.

For the rest of who that person is, you have to go to history.