Is Irreducible Normativity Impossibly Queer?

I argue that Jonas Olson’s argument from irreducible normativity is not a secure basis for an argument for error theory (section 1) and that a better basis is provided by the argument from supervenience, which has more bite against non-naturalist moral realism than Olson is willing to allow (section 2). I suggest there may be a view which can allow for the existence of irreducibly normative facts while remaining unaffected by the kinds of arguments that work against non-naturalist realism. This view is expressivism. Interestingly, James Dreier has recently suggested that expressivism may not escape these arguments. I very briefly outline (but do not pursue) possible response strategies for expressivists (section 3). I close by discussing Olson’s argument against expressivism. Olson suggests, somewhat surprisingly, that expressivism is a bad fit with a plausible evolutionary explanation of our moral thought. I argue that Olson’s argument does not succeed (section 4).