What you are taught in a philosophy degree, I was disappointed to discover, is not the deep philosophical truths of the world, but rather how to argue about philosophy. This is often a matter of memorising stock arguments and counterarguments.
If you want to get good at cutting through bullshit then you should practice noting when people are proving too much. Unpicking complex multi-step arguments is, well, complex. But what’s very easy is taking a slightly dodgy conclusion someone presents, and then showing that it proves something else obviously wrong.
Once you’re bored of discussing ethics and trying to hash out the relative merits of Kantianism and Act Utilitarianism, and as you grow increasingly frustrated by the fact that they all seem to have the same problems, the reasonable next step is to go meta and ask what’s even going on when we talk about ethics? This is meta-ethics: rather than asking why murder is wrong, you ask what it even means to say that murder is wrong.
Enter non-cognitivism, which is the meta-ethical theory that when people make ethical claims they’re not actually expressing propositions that can be true or false, but rather something more like feelings. So when someone literally says “murder is bad”, according to non-cognitivism what they actually mean is “boo murder, murder sucks”. It does convey something, it just doesn’t convey anything fundamental about the universe.
The canonical argument in favour is the delightfully named Argument From Queerness, which is that ethical realism is pretty weird. It sounds simple and it is.
So the sort of culturally default moral realist position is that some things (murder, say) are actually wrong in a meaningful sense, and the accompanying claims (‘Murder is wrong’) are true objective facts about the world. The Argument from Queerness is that these are very strange kinds of facts. How do we come to know them? How might we prove them? Could they ever give us the ability to make empirical predictions? Whatever kinds of fact they are, they’re very unusual, and very complicated. If we can do without them, we ought to. Non-cognitivism seems to present a much simpler explanation of what’s going on. But is it too simple?
The canonical argument against non-cognitivism is “The Embedding Problem”. Basically, non-cognitivism can translate a simple claim like “Murder is bad” into some sort of emotive thing like “boo murder”. But what happens when you embed those simple moral claims into complex moral statements? What does “Most people don’t understand that murder is bad” mean? What about “Once you realise that murder is wrong you only need to think a bit before you realise factory farming is also evil”?
You can make moral claims with pretty fruity internal logic. Surely it wouldn’t be possible to have these kinds of sophisticated compositions if the thing at the bottom wasn’t actually something concrete and truth-functional.
One of Daniel Dennet’s greatest contributions to philosophy was the observation that when somebody says ‘surely’, it’s more often than not a giveaway that they aren’t actually sure, and they can’t give a better argument, and they’d appreciate it if you just took their word for it.
So I invite you to ponder for a second, is the ease with which we embed simple moral claims into complex arguments a good argument for moral claims being actually meaningful at their root?
In particular, would this be proving too much? I think the wrong way to approach objections like The Embedding Problem is to actually work out a full translation procedure. The right way is to say “well, even if I can’t write out a full translation procedure, ethical realism still seems really weird: is this really a good objection in the first place?” and see if you can dismiss the whole argument.
I would like to stress that I’m not asking whether moral claims are meaningful, just whether this property is enough to make us confident that moral claims must be pointing at something deep and important, and not mere expressions of feeling.
So, does the Embedding Problem show that non-cognitivism is false, and ethical claims do reflect something deep and objective?
Well, yes and no. Honestly, interpreting the claim “murder is wrong” as “boo murder” is a bit simplistic. But notice, firstly, that it is exactly as difficult to reduce aesthetic claims, and secondly, that it is painfully easy to make moral claims into aesthetic claims. For instance:
“If lying is wrong rude then so is misleading truth-telling.”
“(P1) If tormenting the cat is bad inappropriate, getting your little brother to do it
is bad inappropriate
(P2) Tormenting the cat is bad
inappropriate.
Ergo, getting your little brother to torment the cat is
bad inappropriate.”
I take the argument from queerness seriously. It sounds so very simplistic, but a lot of philosophy really is just junk, and sometimes the simple arguments are right. At the very least, given how weird the whole idea of deep objective morality is, the burden of proof is on the asserter. If you want to seriously defend realist morality, you need to be able to explain what’s special about it.
I also used to find it very troubling that the line between moral claims and aesthetic claims is so fuzzy. Whether or not aesthetics are deep and platonic, whether or not fashion is immortal and inviolable, I cannot say. But I do suspect that in the cosmic pecking order ethics and aesthetics are on the same rung.
The more I’ve thought about it, the more I find this actually both compelling and satisfying. I wanted to believe that morality was fundamental and important and mandated by some greater force, divine, logical, or otherwise. But when I dug into why I wanted to believe that, the reasons collapsed under scrutiny. More on that in due course.