I have started reading Alexander Miller's "An Introduction to Contemporary Metaethics". The second chapter deals with Moore's the Open-Question argument against naturalism and various reformulations of the argument which avoid the charge of question-begging.
They present some reformulations of the argument which (according to the SEP) are supposed to be seen as a kind of inference to the best explanation. The idea being that to a fully competent speaker, it still seems as though the question is still an open one. For example, some competent speaker who has reflected will not see the question "Is pleasure good?" and "Is pleasure pleasure" as the same kind of question. And this seems to hold for all natural properties like pleasure. The best explanation for this, then, is that the question is in fact open.
It seems as though these arguments are taking a very science-based approach. I.e. they're using an inference to the best explanation from the 'phenomena' of competent speakers seeing the question "is N (natural property) good?" as consistently an open one.
Couldn't a committed naturalist simply argue that these people are mistaken and that the question is, in fact, not open? Aren't metaethical theories like analytic naturalism aiming to establish the most rational way to think about 'goodness'. These arguments make it look like they are simply descriptive linguistic theories. Why do we have to assume that the way that some apparently competent speaker acts and thinks is actually rational and competent? The book treats this argument as a weak and ad hoc explanation but why does the argument have to avoid ad hocness at all? Why do we have to be committed to the idea that our usual use of language is correct?
And furthermore, if the aim of metaethics is to be a kind of linguistic science about language, why aren't philosophers going out and collecting data on how many competent speakers consider the question "is N good" an open question?
I suppose my question boils down to:
I'm not sure whether these metaethical theories are supposed to be prescriptive or descriptive. (Prescriptive in the sense that they tell us about how we should reason about ethics, instead of just telling us how we ordinarily do reason about it). I can't see where the descriptive elements end and where the prescriptive ones begin. What's the whole point of this endeavour?