Scanlon’s Being Realistic about Reasons is a beautiful book – sleek, sophisticated, and programmatic. One of its key aims is to demystify knowledge of normative and mathematical truths, realistically construed – i.e. construed, roughly, as being true relevantly independent of minds and languages, when interpreted at face-value. In this article, I develop an epistemological problem that Scanlon fails to explicitly address. I argue that his ‘metaphysical pluralism’ can be understood as a response to that problem. However, it resolves the problem only if it undercuts the objectivity of normative and mathematical inquiry.