One might complain that we can't actually separate value from desires. That is, our only acquaintance with value is through our desires - we have no direct access to the 'objective values'. So we have no reason to think there is any such thing, above and beyond our desires themselves.
But this objection may be too strong, for it seems to apply just as well to truth! "Our only acquaintance with truth is through our beliefs - we have no direct access to the 'objective truth'. So we have no reason to think there is any such thing, above and beyond our beliefs themselves."
We might appeal to the possibility of learning, and argue that there must be something objectively real that we are learning about. We often judge that our past beliefs were mistaken. This might seem to require objective truth (to provide a standard for such judgments). Similar claims could be made of our desires: something desired can turn out to be bad for us. However, such judgments can only be made in light of other desires (e.g. one might deem smoking bad after realising that it conflicts with one's desire for good health). So we need not appeal to 'objective values' after all. But the same holds of truth: we can only judge a past belief false in light of our new beliefs. So wouldn't this argument lead us to also reject objective truth?
This is a problem for me, because I want to say that truth is objective [belief-independent], yet value is subjective [desire-dependent]. But the close analogy between belief/truth and desire/value makes it difficult to justify treating the two pairs differently.
A big problem I have with the 'objective value' picture is that it leaves 'value' a complete mystery. What is, or constitutes, value? How does it exist, and in what form? What makes something valuable but something else not - what difference in their makeup explains this difference in worth? It seems that none of these questions have meaningful answers, unless we use something like the desire-based reduction I've previously advocated.
Perhaps here we finally have a disanalogy with truth. Perhaps objective truth is understandable in a way that objective value is not. Truth is just how things really are. It's an intuitive concept that we can easily grasp. It requires no further explanation. Perhaps a value-objectivist could say that value is just "what things are really worth". But although there may be no logical basis for judging this explanation any worse, I simply can't make intuitive sense of it. I think I understand objective existence (though I might be misleading myself). I'm sure I don't understand objective worth. But I'm not sure whether that's a good enough reason to treat them differently.
I should note that we don't actually require mind-independence (of truth or value) in order to normatively evaluate beliefs and desires. As I've said before:
We can evaluate a desire against how well it would help us fulfill all our other desires. So we can reconcile DF with the idea that desires aim for wellbeing - it's just that this 'aim' presupposes some set of (other) desires to evaluate our wellbeing against.
So we can have a sort of quasi-objectivity even if value is mind-dependent. Does the same hold for truth? It might. Suppose we held 'truth' to simply be coherence with our other beliefs. Then we could still evaluate our beliefs in terms of how well they cohere with all our other beliefs. The odd ones out could be deemed 'false', just as a conflicting desire will be called 'bad'. In neither case do we need to appeal to a mind-independent reality for underlying structure. Coherence within our superficial mental structures is enough.
This is troubling - I'm actually half-tempted to dismiss the correspondance theory of truth after all!
The benefit of positing an external world (of mind-independent truths) is that it can help explain why we have the experiences and beliefs that we do. Is there any corresponding benefit to positing the existence of external values? This I'm not sure about. I think we can have naturalistic normative explanations (e.g. "Bob failed the exam because he didn't study as much as he should have"), i.e. ones which can be 'reduced' to purely descriptive facts. I can't imagine how any irreducibly normative facts could ever figure in worldly explanations, however.
So I think this provides some grounds for rejecting non-naturalistic theories of 'objective value', at least. Could one have naturalistic value grounded in mind-independent natural facts? It's not something I know anything about, but I guess it must be possible. If such a theory - with genuine explanatory power - was offered, then the analogy with truth might make it very hard for me to reject it. (But then, I probably wouldn't want to reject it - I don't find subjectivism all that appealing; it's just that I haven't yet come across any plausible [i.e. naturalistic] alternatives!)
What do you think? Should we accept both objective truth and value, or reject both, or do we have grounds for discriminating between the two?