THE OUGHT-IS GAP: TROUBLE FOR HYBRID SEMANTICS BY MATTHEW S. BEDKE When it comes to the meanings of normative expressions, descriptivist theories and expressivist theories have distinct explanatory virtues. Noting this, and with the hope of not compromising on explanatory resources, hybrid semantic theories refuse to choose. Here, I examine how well the strategy works for Moorean open questions and associated is-ought gaps. Though hybrid theorists typically rely on their expressivist resources for this explanandum, there is a type of open question that unadulterated expressivist theories can handle but hybrid theories cannot – reverse open questions associated with an ought-is gap. Because of this, hybrid theories do not enjoy the best of both worlds, and Moorean considerations favour unadulterated expressivism over any partly descriptivist theory. When it comes to semantic theories of normative expressions two positions dominate the landscape. According to descriptivism, normative expressions purport to describe what things are like, normatively speaking. According to expressivism, the meanings of normative expressions are to be given in terms of the conative attitudes they are used to express. Arguably, the former does a better job than the latter when it comes to explaining surface grammar, inferential roles, truth aptitude, and appar- ent disagreement in fact, whereas the latter does a better job explaining the link between normative judgments and our concerns/motivations, and it dodges metasemantic, 1 metaphysical and epistemic explanatory burdens that weigh on descriptivists. This is a hard choice for those who make it. But why make it? Hybrid semantic theories offer an intoxicating alternative: hold that nor- mative expressions have both descriptive and expressive dimensions to their meanings, and avail yourself of all the explanatory resources on the table. If unadulterated descriptivism has a better explanation than 1 Metasemantic burdens are those borne by explaining how the semantic values of nor- mative terms get fixed, e.g., by associated definite descriptions, by causal regulation, by speaker intentions, or what have you. The Philosophical Quarterly ISSN 0031-8094 doi: 10.1111/j.1467-9213.2012.00082.x © 2012 The Author The Philosophical Quarterly © 2012 The Editors of The Philosophical Quarterly Published by Blackwell Publishing, 9600 Garsington Road, Oxford ox4 2DQ, UK, and 350 Main Street, Malden, MA 02148, USA