Contemporary discussions of knowledge how typically focus on the question whether or not knowing how to do ϕ consists in propositional knowledge, and divide the field between intellectualists (who think that it does) and anti-intellectualists (who think that it does not, and that it consists instead in the possession of the ability to ϕ). This way of framing the issue is said to derive from Gilbert Ryle. I argue that this is a misreading of Ryle, whose primary interest in discussing knowledge how was not epistemological but rather action-theoretical, whose argument against intellectualism has for this reason been misunderstood and underestimated (by Jason Stanley, among others), and whose positive view aims to chart a middle course between ...