An Ontological Proof of Moral Realism

Ten years ago, I proved the truth of moral realism. It was in the article, “An Ontological Proof of Moral Realism” (Social Philosophy & Policy 30 (2013): 259-79). Following is a summary.

Background: One version of the “ontological argument” for God’s existence claims that, given that it is possible that God exists, God must actually exist. This argument is very bad. However, there is an ontological argument for moral realism that is good — that is, an argument to the effect that, given that moral realism might be true, it is actually true.

1. The Issue

Moral realism: The thesis that there are some objective moral reasons.

That is, there are some reasons for action that are categorical (not based on the agent’s desires), non-selfish (not based on the agent’s interests), and objective (independent of the attitudes of observers towards the action).

2. The Anti-Torture Argument

I promise you’ll get an argument for moral realism, later. But first, I’d like to give an argument against torturing babies just for fun.

1.     There is some reason to think moral realism is true.

Why think this? Well, many people have given arguments for moral realism that are not completely stupid. Most philosophers agree that moral realism is the common sense view. It seems to have been the view of the great majority of ethicists in history, and ethicists are presumably the experts on the subject, if anyone is. Moral realism remains the majority view today among philosophers (https://philpapers.org/surveys/results.pl). And there is no conclusive argument against it.

Premise 1 doesn’t say that moral realism is true, or even probable; it merely claims that there is at least some reason, however small, to believe it. The above remarks suffice to show that. Moving on …

2.     If moral realism is true, then recreational baby-torture is objectively wrong.

Recreational baby-torture is a paradigm of the sort of thing that would be objectively wrong, according to moral realists; surely if anything is objectively wrong, that is. All moral realists seem to agree with this, and there are no arguments to the contrary.

Of course, (2) doesn’t tell us whether moral realism is in fact true or not. So even nihilists could (and should) totally agree with (2).

3.     If recreational baby-torture is objectively wrong, that’s a reason to avoid it.

This is true by definition, just from the meaning of “wrong”. Again, even anti-realists agree with this conditional, though of course they reject the antecedent.

4.     If there’s some reason to believe that P, and P would provide a reason to do A, then there’s some reason to do A.

I call this the “Probabilistic Reasons Principle”. I note that it works for all kinds of reasons. Some examples:

·       a. An epistemic reason: Assume that, if John said that Q, that would be a reason to believe Q. Further assume that you have at least some reason to believe that John said that Q. In that case, you in fact have at least some reason to believe Q. (Of course, it might be only a small reason, it could be outweighed by other reasons, etc.)

·       b. A prudential reason: Assume that, if lottery ticket #x was a winner, then you’d have a reason to buy it. Assume also that you have at least some reason to think that ticket #x is a winner. Then you in fact have at least some reason to buy ticket #x.

·       c. A moral reason: Assume that, if you knew that there were people in the forest, you’d have a reason to avoid shooting your gun into the forest. Also assume that you have some reason to think that there are people in the forest. In that case, you in fact have a reason to avoid shooting into the forest.

I take all of these examples to be intuitive. They serve to illustrate the PRP, which I take to be close to self-evident.

5.     Conclusion: You have some reason to avoid torturing babies for fun.

This follows from 1-4. (From 1+2, you have a reason to think recreational baby-torture is objectively wrong, barring defeaters (which in fact you don’t have). If it were objectively wrong, that would be a reason to avoid it. So, per the Probabilistic Reasons Principle, you in fact have a reason to avoid it.)

3. For Moral Realism

Section 2 establishes a reason to avoid baby-torture. What kind of reason is this? Answer: It is an objective, moral reason. Two arguments:

a. First Argument

What is the reason to avoid baby-torture established in sec. 2? The reason is that baby-torture might be objectively wrong. This reason is obviously a moral reason, and not an appeal to self-interest or desire. It also does not depend on the attitudes of observers toward baby-torture, so it’s not a subjective or culturally relative reason.

So it’s an objective moral reason, by the definition given in sec. 1.

b. Second Argument

The premises of the anti-torture argument (1-4 in sec. 2) do not depend on self-interest, desire, or the attitudes of observers towards baby-torture. The conclusion, (5), follows from those premises; therefore, it does not depend on self-interest, desire, or the attitudes of observers either.

(5) states a reason for action; since it doesn’t depend on self-interest, desire, or the attitudes of observers, that reason for action is an objective moral reason, in the sense given in sec. 1.

Therefore, at least one objective moral reason exists, so moral realism is true.

4. The Theory-of-Reasons Objection

In the paper, I discussed a few objections. But only one of them is interesting to me, so I will only discuss one objection here. I call it the “Theory of Reasons Objection”. This objection claims that there is an exception to the Probabilistic Reasons Principle: the principle fails just in this one class of cases, namely, cases of uncertainty about the correct theory of reasons.

That is, in the special case where you have some reason to think that a certain theory of reasons is correct, and that theory says that you have a reason to do A, this does not mean that you in fact have any reason at all to do A.

Why might someone think this? Two points:

a. Maybe theories about reasons don’t themselves give us reasons for doing anything. The belief that you have a reason to do A does not itself provide a reason to do A; that (one might claim) would be a conceptual confusion. The belief that you have a reason to do A just reports that something else is a reason to do A; the belief isn’t itself the reason (or even one of the reasons) to do A.

b. In general, it is possible to have some reason to believe a theory that is in fact false. So it should be possible to have some reason to believe a false theory of reasons. This false theory of reasons might say that you have most reason to do A. But since by stipulation this theory is false, that means you do not in fact have a reason to do A (at least, not merely due to this false theory’s saying so; maybe there is some other, true theory of reasons that says you have reason to do A, but let’s leave that aside).

Now, how would this undermine my argument? Well, the debate about moral realism is a debate in the theory of reasons. Moral realism is a theory about what kinds of reasons for action we have, and philosophers are uncertain about whether this theory or another (anti-realist) theory of reasons is true. According to the theory-of-reasons objection, if moral realism is in fact false, then we have no reason at all to do the things that moral realism says we have reason to do.

Reply

I think facts about reasons, in themselves, do give us reasons for doing things. That’s not a conceptual confusion.

Example 1: A perfectly honest expert credibly tells you that there is a lot of evidence for P. He won’t tell you what that evidence is, though. How should you adjust your credence in P?

Obviously, you should raise it. This shows that the fact that something else is a reason to believe P is itself a reason to believe P. It’s not just that the ordinary evidence for P is a reason to believe P; the general fact that some such evidence exists is also a reason to believe P.

Example 2: God gives you a button and tells you that if you push it, your interests will be overall promoted, though he won’t tell you how. What should you do?

Obviously, it would be prudent to push the button. But no particular interest of yours explains this, since you don’t know what interest of yours will be promoted. It’s prudent to push the button because the general belief that pushing would be prudent is itself a prudential reason to push.

Example 3: As before, except God tells you that there is some unspecified reason of some unspecified type to push the button.

In this case, you have reason to push the button, just from God’s telling you that. Here, the general fact that you have some reason or other to push the button is itself a reason to push the button.

That rebuts point (a) above.

About point (b): Well, a theory of reasons, T, could be false, in that the things that T says constitute our reasons for action are not (always) reasons for action. That’s compatible with saying that we have another reason for doing the things that T says we should do – this other reason being the fact that we have some evidence for T. (T presumably doesn’t itself claim that the reason to do A is merely that we have a reason to believe T.)

So it’s perfectly consistent to hold that we can have a reason to believe a false theory of reasons, and that this would give us a reason for doing the things that this theory says we should do.

Conclusion

We have at least some idea of the things that we ought to be doing if moral realism is true. E.g., we should avoid torturing babies. Since moral realism may be true, we have reason to do those things (e.g., to avoid baby-torture). But this reason itself counts as a non-selfish, categorical, objective reason. So, given that moral realism may be true, it is true.