David Sobel has an interesting post up at the revamped PEA Soup blog on 'Normative Stance Independence and Pleasure'. He suggests that if pleasure is best understood in attitudinal terms (as per Parfit's hedonic likings) then this undermines Normative Stance Independence, the view that "normative facts are not made true by anyone’s conative or cognitive stance" or "by virtue of their ratification from within any given actual or hypothetical perspective."
But does it? The distinction between stance-dependence and -independence is a slippery beast. Even if pleasure could be said to involve "taking a stance" towards a base sensation by liking it, it's not so clear that the stance is what does the heavy lifting in explaining why pleasure is good. More plausibly, I think, pleasure is good just because of how it feels, objectively speaking. Again, this normative explanation remains untouched, it seems to me, no matter if the phenomenology of pleasure turns out to be inextricably tied up with the attitude of liking. It could still be the objective phenomenology, rather than the "stance" per se, that matters.
(In support of this point, I take it that if knowledge, for example, has intrinsic value then this is uncontroversially objective or 'stance-independent' in nature, regardless of the fact that knowledge is (or involves) a cognitive state, and so might be considered part of the agent's "stance" in some sense. So, why not the same for pleasure?)