Abstract Modal operators are unneeded in fundamental inquiries, and so they probably do not carve at the joints. A reduction of modality is therefore demanded. A “Humean” reduction is proposed, according to which necessarily true propositions are the logical consequences of a more‐or‐less arbitrary list of propositions; the list includes mathematical truths, analytic truths, “laws of metaphysics”, “natural kind propositions”, and others. Although the Humean account is similar in some ways to old‐fashioned conventionalism, truth by convention plays no role in it.