Like my last post, Julian Sanchez tackles Feser and Volokh's arguments about legislating morality, and makes points similar to mine, but does so in a much clearer and pithier way.
[Feser and Volokh are] right in a sense, of course. But the point is so obviously right that a modicum of interpretive charity might lead one to wonder whether such an elementary point could really have escaped people who use the phrases like "legislating morality"—many of whom seem to be adults capable of tying their shoes and feeding themselves.
I've never been a huge fan of the phrase—partly, I suppose, for that very reason— but I'm fairly sympathetic to what I take to be its clear meaning, if we assume those who deploy it aren't attempting to identify their own positions with amorality or immorality. Rather, the phrase and its cognates seem to typically get used as a sort of shorthand for two related claims about the proper scope of either government action or public-reason justification.
The first claim is something along the lines of Mill's Harm Principle, or, more broadly, the familiar distinction in political philosophy between morality or justice construed as "what we owe to each other" (in Scanlon's phrase) and the broader realm of ethics, in the sense of what makes a good life. I think this is often what people mean when they attack attempts to regulate private sexual behavior: They're saying that even if certain conduct may be morally objectionable in the broadest sense, it is a matter of private morality—between a person and his conscience or favorite deity—rather than public justice. One can quibble about where that boundary lies, but I think most folks in a liberal society will concede that there exist all sorts of obligations that are ethical or moral in a broad sense that aren't fit subjects of regulation.
The other deployment you'll often see—though it's often linked to the first—has to do with what sorts of justifications should be admissible in discourse governed by norms of public reason. Liberal democracies tend not to accept arguments founded solely on sectarian claims that will be persuasive only to members of some particular religious group. Volokh's certainly right that the fact-value gap remains, and that there's no purely logical deductive proof that assault is bad while gay sex is OK. But if he wants us to conclude from this that, so far as public justification is concerned, the claim that we should ban X because it will cause lots of unnecessary deaths is on all fours with the claim that we should ban X because the Sacred Lemur appeared in a vision and said so... well, it becomes hard to take him terribly seriously.