The explanatory superfluity of non-natural moral facts

I'm interested in the following metaphysical thesis:

Moral non-naturalism: Moral facts do not belong to the natural world, i.e., moral facts are not natural facts.

I accept moral non-naturalism. I accept it despite all the problems it causes (and it causes a lot of problems). My purpose in this post is to urge my fellow non-naturalists to embrace one of the main problems that comes along with our view. Specifically, I'm going to argue that non-naturalism implies that moral facts are explanatorily superfluous with respect to moral attitudes.

Here's my argument:

(1) Natural facts just are facts that are properly subject to scientific investigation—i.e., natural facts just are scientific facts. 

(2) If X plays a contributory role in the best explanation of any natural fact, then X is properly subject to scientific investigation.

(3) If X influences Y, then X plays a contributory role in the best explanation of Y.

(4) Moral attitudes are natural phenomena.

Therefore, if moral facts influence or explain moral attitudes, then moral facts are natural facts. Equivalently: If moral non-naturalism is true, then moral facts do not influence or explain moral attitudes.

The argument is valid. (Here's why: Suppose moral facts influence or explain moral attitudes. Then (by (4)) moral facts influence or explain a class of natural facts. Then (by (3)) moral facts play a contributory role in the best explanation of a class of natural facts. Then (by (2)) moral facts are properly subject to scientific investigation. Then (by (1)) moral facts are natural facts.)

I don't think there ought to be much controversy about (4) and (3). Moral attitudes (moral beliefs, moral seemings, etc.) are a type of psychological phenomena; and psychology is a science; so moral attitudes are natural phenomena, as (4) says. Regarding (3): I know of no counterexamples to it, and I know of a lot of cases that seem to lend confirmatory support to it. For example: If I want to explain a thunderstorm, and it turns out that humidity (say) influenced the thunderstorm in some way, then it seems clear that humidity plays a contributory role in the best explanation of the thunderstorm.

Perhaps (2) will be more controversial, but I think (2) is well supported by the actual practices of scientists, and by the norms that these practices reflect. The idea behind (2) is that science is maximally open: the mission of science is to investigate anything whatsoever that best explains natural facts. Even Zeus, for example, would fall within the purview of scientific investigation, if Zeus influenced or explained anything in the natural world.

(1) is a claim about the boundaries of the natural. The term "natural" has a lot of different meanings in different contexts, and I don't claim that (1) is in line with every such meaning. However, in the context of discussion about moral non-naturalism in particular, (1) is very often endorsed; e.g., Shafer-Landau, Parfit, and Enoch have all relied upon some variant on (1). I think it's clear that (1) captures a principal meaning of "natural" in the metaethical context.

(1) and (2) together imply that Zeus would be a part of the natural world if Zeus influenced or explained anything in the natural world. More generally, (1) and (2) together imply that the non-natural cannot influence or explain the natural. Sometimes people find this to be surprising or even objectionable. I don't think it's particularly objectionable. Further, it seems to me that even if these implications are objectionable to some degree, they're less objectionable than the denial of either (1) or (2).

I think the premises of the above argument are all defensible, and so I think we should all just finally, once and for all, accept the explanatory superfluity of non-natural moral facts: If moral non-naturalism is true, then moral facts do not influence or explain moral attitudes.

This conditional is generally considered to be a problem for non-naturalism (and therefore is sometimes resisted by non-naturalists) for at least two big reasons:

First, the explanatory superfluity of non-natural moral facts saddles non-naturalists with some immediately counterintuitive commitments. Consider: We all believe that torturing for fun is wrong. Why do we believe that? Well, it's tempting to say: We believe that torturing for fun is wrong because torturing for fun really is wrong. But if I'm right, moral non-naturalists can't say that. Whatever the correct explanation of the content of our moral beliefs turns out to be, that explanation does not include any moral facts.

Second, and more importantly, the explanatory superfluity of non-natural moral facts presents non-naturalists with some heavy epistemological difficulties. How could our moral beliefs amount to knowledge of moral facts if our moral beliefs are neither influenced nor explained by moral facts?

Some non-naturalists have tried to avoid these difficulties by rejecting or remaining uncommitted about the explanatory superfluity of non-natural moral facts. That strategy won't work, if I'm correct. In my view, defenders of non-naturalism have to try to overcome our epistemological difficulties, rather than avoid them. We have to accept the explanatory superfluity of non-natural moral facts and find a way to struggle forward.

Comments

[Richard Yetter Chappell]

Hi David, you might find my old post on 'Epiphenomenal Explanations' of interest:
http://www.philosophyetc.net/2011/11/epiphenomenal-explanations.html

Also worth flagging: one (very controversial!) way to deny (4) would be to hold that the *intentional content* of our moral attitudes depends in part on the moral properties themselves, and so our moral attitudes -- in contrast to their behavioural and neurological correlates -- are not wholly natural phenomena.

[Roger Turner]

I have questions about 2 and 3.

In 2, what do you mean by 'contributory'? Do you mean 'causal'?

In 3, what do you mean by 'influences'? Do you mean 'causally influences'? Also, the question I have about 'contributory' in 2 i have about 'contributory' in 3.

Why think that non-natural moral facts, even if they influence us, *cause* us to have certain attitudes (given that these attitudes are natural phenomena)? Is this because something's being influential is the same as something's being *causally* influential--that is, that some influential thing is also a cause? This doesn't seem obviously true to me.

I think that the fact that 3+2=5 influences how I think about other math problems. And it also influences how i think about certain philosophical matters (e.g., necessary truths). But I don't think the fact that 3+2=5 causes me to do any of these things. And that's at least partly because I don't think facts can cause anything.

What do you think about that?

[B]

I'm uncomfortable with (3).

Hume influenced Kant's thought. He certainly played a contributory role in an explanation of Kant's thought--say, a psychological, intellectual, or biographical one. Presumably, these are good explanations. But is any of them the best explanation?

That seems controversial--much more controversial than I think you want (3) to be. Plausibly, the goodness of an explanation is relative to context, and there's no "best" explanation simpliciter. Or, if it is, maybe it's the explanation in terms of fundamental microphysics. But did Hume play a contributory role in that explanation? Things explain under descriptions, so while the set of microphysical entities identical to Hume certainly played a role, it doesn't follow that Hume himself did.

Note that if this is a real problem, you probably can't get out of it by weakening (3) to "...then X plays a contributory role in a good explanation of Y," since then the corresponding version of (2) would be implausible.

[Mert]

All moral attitudes are natural, psychological, phenomenon in some sense, but a subset of them, namely the morally correct attitudes, are in the following sense not natural. The explanation of why they are the morally correct attitudes is not amenable to scientific inquiry so I would deny (2). Deliberation, reasoning, reflection are psychological processes or events but the best explanation of a subset of them, namely the ones that result in morally correct attitude, involves moral facts. They play a contributory role in the best explanation of the moral attitude but are not thereby natural facts.

Part of the best explanation for why we believe that torturing for fun is wrong as opposed to right, is that there are no considerations that count in favor of it being right. The fact that there are no reasons to believe that torturing for fun is right, is a moral fact that plays a contributory role in the best explanation of our moral belief that torturing for fun is wrong. If this moral fact were not the case, i.e. if there were reason to believe that torturing for fun is right, then we would not believe that torturing for fun is wrong.

[Phil H]

I think you should cut science out of this argument. It doesn't do any explanatory work here, and your (1) is at least controversial. I think it's wrong: I can think of scientific facts which aren't natural (e.g. the results of scientific modelling) and I can think of natural facts which aren't scientific (e.g. Venus is in Orion).

So I'd dump (1) and replace "scientific" with "empirical" in (2). I don't think it changes the force of your argument. I think the argument's right, and inevitable once you give away the farm with (2). My view is that (2) begs the question, and if you want to maintain any kind of non-naturalism at all, you have to deny it. It seems reasonably easy to deny: anything from souls to moral causes to first causes might lead you to deny (2).

If you don't deny it then I wouldn't allow you even the conclusion you draw, "moral facts do not influence or explain..." because I wouldn't allow that there is any such thing as a moral "fact" of the type you're imagining. One key feature of facts is that they can be checked; your non-interacting morals can't be checked by any natural/empirical means, therefore I wouldn't accept calling them facts.

[Derrick Murphy]

How might you respond to the following challenge to the conjunction of (1) & (2): Theists (for example, Leibniz and more recently Richard Swinburne) have traditionally presented arguments for God's existence which conclude that God's existence, power and intentions together constitute the best explanation for the fact that there is a contingent reality. (While theists have traditionally asserted that God created finite supernatural entities as well as natural entities, let's stipulate that the whole of contingent reality is the natural world and its contents.) Also, it seems reasonable to conclude that the fact that there is a natural world is itself a natural fact. By (2) and (1), if we conclude that God is part of the best explanation for the existence of contingent reality, we would have to conclude that God is a natural entity. However, God is supposed to be a paradigmatic non-natural entity, a supernatural entity.

One might think that this is the Zeus case in different garb, but there's a key difference here. The Greeks didn't think their gods created the natural world, but were simply personifications of aspects of the natural world. In a sense, they thought their gods were natural entities. The God of Theism is supposed to be something much more fundamental and something much more transcendent than Zeus and the other Greek gods, something that would have to be a very different sort of thing from the natural world. As a result, it seems that either (1) or (2) should be rejected.

[David Killoren]

Hi everyone,

I really appreciate all of your comments. These are great! I was planning to give a careful reply to everyone this weekend, but I have run out of time. I will post some responses early this week.

Thanks very much to all of you for reading and responding to my post!

David