Abstract
I argue that level‐incoherence is epistemically valuable in a specific set of epistemic environments: those in which it is easy to acquire justified false beliefs about normative requirements of epistemic rationality. I argue that in these environments level‐incoherence is the rationally dominant strategy. Nevertheless, level‐incoherent combinations exhibit a distinctive tension, and this tension has been thought by many to indicate that level‐incoherence is always irrational. Although this idea has proved resilient, I argue that it is incorrect. I evaluate three candidate explanations for the distinctive tension exhibited by level‐incoherent combinations, only one of which is the traditional view (which I call the ‘Prohibition View’) that epistemic level‐incoherence is prohibited by epistemic rationality. I argue instead for the ‘Inquiry View’, according to which level‐incoherence is not rationally criticisable but is a reason to undertake further inquiry.