First Steps in Philosophical Taxonomy

Abstract

A.N. Prior once showed that on certain apparently reasonable assumptions, a thesis sometimes associated with the name of Hume to the effect that no set of factual statements can ever entail an evaluative statement (call this principle ‘H’), is quite untenable. We assume only that there is at least one statement of each kind, and that the negation of a factual statement (evaluative statement) is factual (evaluative, respectively) — a principle we may call ‘N'. Now consider the disjunction F V E of some factual with some evaluative statement. Since the disjunction is entailed by F, a factual statement, it must, if principle H is correct, be classified as factual. But by N, ∽ F is also factual, and this together with F V E entails E, thus violating H since E was exhypothesi an evaluative statement.