Mackie’s queerness argument (QA), in its metaphysical part, goes something like this. Putative facts about our moral obligations supposedly involve objective values that carry normativity – they give us reason to do things. But any such facts would be metaphysically strange, so it’s hard to countenance the existence of such facts. (Epistemologically, knowing such facts would require some unlikely special moral intuition.)
But this argument’s place in the dialectic between realism and antirealism (about objective moral facts) is itself a bit strange, since it seems that proponents of QA are incoherent when they advance QA, and that a standard realist response to QA is also incoherent.
One way that realists respond to QA is to ally moral facts with other normative facts, e.g., epistemic facts: if we’re willing to say that some person ought to (for good reasons) believe that P, we might as well say that she ought to perform act A. This might seem compelling. However, in response the antirealist can just go global, and say that all normativity, of whatever sort, is metaphysically odd. Our physical world simply doesn’t contain anything that, of itself, gives us reasons, of any sort.
So the next step, made by for example, Shafer-Landau (Moral Realism, 207-208 [on one interpretation of what he’s saying]), is to point out that logic itself is normative. We say that if I am willing to grant that A>B and that A, then I should infer that B. Correlatively, if you are a global antirealist about normativity, then you have to deny the normativity of logic and argumentation. In which case you have to deny that QA gives you or anyone else, including realists, any reason to accept its conclusion. Call this the ‘normativity objection.’
So, Shafer-Landau concludes that QA “cannot rationally compel the allegiance of its detractors.” But that’s not the whole truth of the matter, and here’s where the real strangeness of QA comes in. It is true that for (global) antirealists, QA provides no reasons for believing anything about morality. But for realists about inferential norms and matters epistemic, arguments, like QA, do provide reasons to accept their conclusions. In this case, the realist’s commitment to normativity renders the normativity objection incoherent. There’s something of a paradox here: reasonably, the realist holds that moral normativity is no more odd than other kinds of normativity, but that very claim is inconsistent with the normativity objection’s insistence that QA gives us no reason to doubt the reality of moral facts.
Of course, Shafer-Landau is right that antirealists can’t claim that you should accept the inference. But the point under consideration now is that if you’re a normativity realist, you don’t need to be convinced of that claim anyway.
Now turn to the position of the global antirealist. If she truly denies all normativity (on grounds of its metaphysical oddness), then by her own lights she has no reason to put forth the QA in the first place, as Shafer-Landau points out. Put differently, at the end of a paper in which global antirealism is defended via some version of the QA, the honest author will claim: “But if I’m right, none of this gives you any reason to believe what I said.” Thus the antirealist’s commitment to the metaphysical nonsense of normativity pushes her away from advocating her position via argumentation. So the antirealist has no reason to put forth QA, or for that matter any argument, in the first place. (At least that’s true of the global antirealist, though it’s hard to see why one wouldn’t be a global antirealist if the motivation for one’s antirealism is QA.) Putting forth a queerness argument for antirealism seems incoherent.
If the foregoing is right, then there seems to be something of a paradox. Antirealists are incoherent in putting forth QA. Indeed, putting forth QA as a reason for belief about anything undermines QA itself. The other side of this coin is that realists should be concerned about QA (assuming its soundness), since they can’t argue against it by associating moral normativity with logical and epistemological normativity, for doing so pushes them towards accepting QA as a reason to doubt normativity…which then undermines QA’s own reasons for doubting normativity. So, again, it’s hard to see even the possibility of debate on this question, since both realist normativity objections to, and antirealist advocacy of, QA are incoherent. In which case, one would have to choose between realism and antirealism on some other grounds besides QA and the ubiquity of extra-moral normativity.