Modal dualism is the claim that there is a space of epistemically possible worlds that exceeds the space of metaphysically possible worlds. In the present paper, I argue that modal dualism is false. I do so via an argument that differs from most previous arguments against modal dualism in that it does not rely on controversial semantic or epistemological assumptions like descriptivism, internalism or modal rationalism. The point of my argument is, instead, that modal dualism is internally inconsistent because it is the result of the failure to draw the proper conclusions from the abandonment of these semantic and epistemological assumptions. The relevant inconsistency manifests itself in the modal dualist notion of epistemic modality. For this notion to play the explanatory roles assigned to it by modal dualists, it has to be a notion of a genuinely epistemic kind of modality. But the only way to make sense of such a notion is via just those semantic or epistemological assumptions that the modal dualist is keen to avoid. Anyone willing to abandon these assumptions has to go further and detach the epistemic from the modal altogether.
| INTRODUCTIONOnce there was a golden triangle among meaning, modality, and rationality (see e.g., Chalmers, 2006b, 55). As long as this golden triangle were intact, necessity, analyticity, and apriority were regarded as equivalent. The idea is appealing: When do we need no empirical evidence to know a sentence to be true (apriority)? Iff it is true regardless of how the world is. When is a sentence true regardless of how the world is (necessity)? Iff it is true just in virtue of the meanings of its components. When is a sentence true just in virtue of the meanings of its components