More than a century ago, G. E. Moore famously attempted to refute all versions of moral naturalism by offering the open question argument (OQA) followed by the “naturalistic fallacy” charge (NF).1 Although there is consensus that this extended inference fails to undermine all varieties of moral naturalism, OQA is often vindicated as an argument against analytical moral naturalism. By contrast, NF usually finds no takers at all. ln this paper we argue that analytical naturalism of the sort recently proposed by Frank Jackson and Michael Smith does after all rest on a mistake — though perhaps not the NF. Analytical moral naturalism is roughly the doctrine that some moral predicates and sentences are a priori equivalent to predicates and sentences framed in non-moral terms (Jackson 2003: 558). Given moral naturalism, it is at least possible that there are some such a priori or conceptual equivalences. But a properly construed OQA challenges this reductive strategy by showing that it is open to doubt on a priori grounds. We further contend that, in the dialectical context created by our OQA, a "digging in the heeis" defense of the strategy would beg the question. ll...