I've come to learn a lot more about meta-ethics since I wrote this fumbling post way back in August of 2009. That long ramble is still fairly close to my current thoughts, but now I know a lot more of the conventional views and terminology -- and let's face it, even though jargon can be obscure and irritating, properly used it can help clarify and focus our thoughts, by giving us a name by which to refer to a very complicated idea -- and as of late, I've been stimulated to ponder it much more deeply by four events: The discussion over Sam Harris' The Moral Landscape, which I suppose I will have to read now; a series of posts from Russell Blackford about meta-ethics, partially as a follow-up to discussions with Harris; my discovery of Daniel Fincke's blog Camels with Hammers, and his unconventional take on meta-ethics; and a conversation with Bjørn Østman in which he called me out on sloppily referring to objective morality in another post without proper caveats. So my thinking is evolving.
Despite the fact that as recently as a week ago I declared that I "certainly do not" subscribe to moral error theory, I am beginning to think that on some level it is the most accurate approach to meta-ethics. Although I think it is utterly irrelevant in practice, if we want there to be (as Blackford puts it) "more metaphysical grunt" behind morality, if we want to assert that people are compelled to behave morally as opposed to just naming what is moral and what is not... if that's important to us, well, you can't have that, so in that sense if that is what most of us are trying to say when we make moral statements, then that is necessarily an error, and moral error theory must be correct.
I am not sure if in practice the lack of a true metaphysical imperative is really all that important, though. I might also describe myself as a moral realist in the sense that I think what is meant by morality is non-arbitrary, describable at least in principle, and -- here's where I got into trouble tossing around that word "objective" so flippantly -- once we have all that, I think that for many moral propositions we can confidently assign it an unambiguous truth value.
The only thing that is missing is some ghostly metaphysical demand that we "ought" to behave that way, but I don't think this is actually a problem. As a result of natural selection, virtually all of us (excepting true sociopaths and those whose meta-ethical foundations have been severely distorted by bogus ideologies and dogmas) buy into the core subjective "ought": namely that we ought to try and promote those universal values which seem to be shared among all human cultures.
The elusive metaphysical imperative is unnecessary, because we already have the necessary imperative built in; indeed, one could argue that it's a pretty fundamental part of being human.
Which is not to say that humans naturally behave morally, heavens no. But we all naturally have an inclination towards fairness, towards avoiding undue suffering, etc., and given that, our big brains are (or ought to be) able to puzzle out some more advanced moral concepts in order to support that underlying imperative.
So I am asserting: that morality is supported by a built-in imperative universal to virtually all humans; that it is non-arbitrary; and that for at least some moral statements, we can confidently state their truth or falsity in a way that firmly transcends culture. That's 99% of the way to moral realism, I think, and so I am hesitant to disclaim the label. Call it moral quasi-realism perhaps?
Now there is a parallel here to my thoughts on the Problem of Induction. In both cases I am making an assertion ("inductive reasoning is effective", "people ought to behave in a way that promotes universally shared values") which I believe to be ultimately unsupportable, but I am not too worried about it because I think these assertions are shared by pretty much everybody; and I think that the rest of what we want to get (objective reality in one case, moral truth in the other) flows naturally from that.
There is one important difference, however: In the case of inductive reasoning, while I believe the assertion is unsupportable without resorting to a defective circular argument, I also happen to believe the assertion is objectively true, no matter what humans might have to say about it. If the world blew up tomorrow, I'm pretty confident that inductive reasoning would continue to be a valid way of uncovering reality (assuming there were some other beings around to do the reasoning, but of course even if there weren't, it would still be true that it would work). Objective reality does not depend on the existence of humans or other sapient beings. I cannot prove this, nor can I really offer any evidence without resorting to circular reasoning; but I am extremely confident it is true nonetheless.
In the case of morality, however, even though I think it is non-arbitrary and that moral statements can sometimes be classed as true or false, all of that depends on humans (or sapience, or at the very least sentience). Morality is non-arbitrary only in reference to who we are, and if we didn't exist, or if what it means to be human fundamentally changed in certain key ways, those assertions would no longer stand up.
So while my belief in objective reality depends on an unsupportable assertion, I believe that assertion is objectively true. It's objective all the way to the core. In contrast, my belief in "objective" morality not only depends on an unsupportable assertion, but that assertion depends on some subjective traits of humanity. My meta-ethics has a crunchy objective shell with a chewy subjective middle. Bite into it, and I suppose I subscribe to moral error theory; but for all outward intents and purposes I am a moral realist.