The Moral Argument's Core Mistake

moral properties are properties, and any general theory of properties should cover them if they are real

Roses are red, violets are blue, both are things flowers, and these claims are true.

Say this to a love interest, and he/she/they will immediately wife you up; say this to a philosopher, and he/she/they will start going off about the nature of properties, and what makes property ascriptions like ‘the rose is red’ and ‘these violets are blue’ the case.

A number of theories of properties have been proposed. According to Universalism, violets are blue because they instantiate the universal of blueness.

According to Trope Theory, what makes it true that a violet is blue is that it contains a blueness trope — a theoretical entity which, it’s said, is concretely located in spacetime and confers colour on the violet it’s part of. And the reason multiple violets can be blue in a bunch is that, even though different violets don’t share the same tropes, their tropes nevertheless resemble each other, which is enough to make ‘violets are blue’ the case.

According to Resemblance Nominalism, which eschews universals and tropes, what makes ‘violets are blue’ the case is that they resemble each other. More precisely, what makes it true that two separate violets are blue is that they belong to the same class of resembling particulars. (Resemblance nominalism is called 'nominalism’ because it denies that there are universals and tropes, not because it denies that there are properties. It just gives a different account of what properties are.)

Class Nominalism says you don’t even need resemblance. The reason blue objects belong to the same class — and are, as a result, blue — has nothing to do with their having same universal or with their resembling each other; the reason they belong to same class is just that — in some primitive, unanalysable sense — it is natural for them to be grouped together. If you ask why it’s natural, the Class Nominalist will punch you in the face and tell you there isn’t an answer, putting you in the natural class of people who just got their shit rocked for asking too many questions.

If these all seem crazy to you, there are other, obscurer views on the market. If one or more of these views didn’t even make sense to you… you are not alone.

The point is, there are plenty of general theories about properties, which were cooked up to answer the Problem of Properties: the problem, that is, of what makes it the case that roses are red and violets are blue.

Moral realists believe in a particular species of property: the stance-independent moral property. If moral realism is true, then, for example, killing postmen for sport has the property of being wrong, and wellbeing has the property of being good, regardless of what anyone believes or feels about the matter. If everyone thought wellbeing was bad, moral realists would say they’re all wrong, just like if everyone thought that Mars is made of rubber they’d be wrong.

Assume moral realism is true. According to the Moral Argument for Theism, God’s nature best explains why moral realism is true; and if, per impossible, there were no God, nothing would be stance-independently good or bad and nothing would be wrong or right. Thus, if we want to be moral realists, we had better be theists as well.

(Note: some versions of the moral argument say that God’s desires or commands — as opposed to his nature — ground some or all moral facts. By definition, this is a version of moral subjectivism, since it makes moral properties constitutively depend on a stance taken by a subject (namely God). Here, I’m just going to focus on the version of the Moral Argument that assumes moral realism.)

Enlightened debates over the Moral Argument typically run as follows¹:

THEIST: Some actions — genocide, fornication, *tightens fists, squints eyes, breathes heavily*, transsexualism — are immoral no matter what anyone thinks. What explains this fact? Without God, there is nothing to ground these moral truths.

SKEPTIC: Respectfully, have you read the Philosophical Literature™? There is an easy way to explain why *tightens fists, squints eyes, breathes heavily* transsexualism is immoral on atheism. Atheists can just posit that there is an abstract universal — badness — that is instantiated by… that.

THEIST: *Googles objection to Universalism*. The reason that doesn’t work is… is… is… hold on… is… sorry, loading… is… oh, yeah, the reason that doesn’t work is *reads off specific philosophical objection to Universalism as a general theory of properties*.

SKEPTIC: *replies to objection, as if the dispute fundamentally hinges on Universalism as a general theory of properties*.

I think this dialectic is broken. Moral properties are just a type of property, and most theists grant that not all property ascriptions are made true by being grounded in God’s nature. (For example, you don’t often hear theists say that a rose’s property of being red is made true by God’s nature. Even if God is the reason why the red rose exists in the first place, his nature is not the truth-maker for ‘the rose is red’ on almost anyone’s view.)

As a result, with respect to some non-moral properties, theists are going to have the Problem of Properties on their hands. Unless they want to say that all properties are grounded in God’s nature (which… come on), they’re going to have to tell some story about why their more general theory of properties — whatever it is — doesn’t work for moral properties specifically.

In Good as Usual: Anti-Exceptionalist Essays on Values, Norms, and Actions, philosophy god-king Timothy Williamson defends what he calls the ‘null hypothesis’ about the normative:

The null hypothesis about the normative is that it is unexceptional. On the antiexceptionalist view, a general semantic theory which works for non-normative language will also work for normative language; a general epistemological theory which works for non-normative knowledge and belief will also work for normative knowledge and belief; a general metaphysical theory which works for nonnormatively expressed states of affairs, properties, and relations will also work for normatively expressed states of affairs, properties, and relations. This is not at all to assume that the non-normative is uniform in any of those respects, just that if such a non-gerrymandered generalization covers the non-normative, it will cover the normative too.

There is no suggestion that normative anti-exceptionalism is obviously correct. Rather, it is the null hypothesis in the modest methodological sense that it is the default view: the burden of proof is on its opponents. If a semantic, epistemological, or metaphysical hypothesis has withstood the worst the non-normative in all its variety could throw at it, we should not abandon it for the normative without good reason to do so.²

I think this presumption is correct. Suppose, e.g., that the theist likes Resemblance Nominalism as a solution to the Problem of Properties. In that case, why not just say that murder and genocide are both stance-independently bad because they belong to the same class of resembling particulars? Or if they like Trope Theory, why not say that genocide is bad because genocide always involves a badness trope?

Maybe these theories of properties are unattractive. In that case, replace them with something better. The point is, unless all properties are grounded in God (?), the theist will have to give some story about why roses are red. If they do, the burden of proof is on them to show that moral properties are exceptional, overturning the anti-exceptionalist presumption that they are not.

There is a general trend in metaethics to discuss the metaphysics of moral properties in isolation from the metaphysics of properties in general. This can lead to sloppy thinking, where metaethicists treat moral properties — for no reason — as though they are special metaphysical snowflakes that need to be accounted for specially, untethered from whatever general theory accounts for properties generally. But unless Moral Arguers have an argument for treating moral properties as needing special treatment, then they have just raised a specific instance of the Problem of Properties, which, while puzzling, is equally a puzzle for everyone.

---

1. See this recent post by

Michael Huemer

in reply to William Lane Craig. Huemer is a Platonist, and he was writing in defence of Wielenberg, another Platonist, against Craig’s nominalist shenanigans, so it makes sense that he’d include a defence of Platonism. My point is just that secular moral realists — if they don’t like Platonism — are under no dialectical pressure from Craig to hitch their view to that wagon.

2. For another philosophy of religion x Good as Usual crossover, see this essay by

Brad Skow

.