Moore often is credited with implying the view that the meaning of evaluative or normative concepts is distinct from the criteria invoked to justify evaluative or normative judgments. A second view, to the effect that definitions cannot be evaluative or moral assertions, is attributed to him less frequently. In this paper, I shall argue that, while these views seem to be implied by much of what Moore says in Principia Ethica, Moore was not himself uniformly successful in observing their prohibitions. In particular, I shall argue that his account of ‘Right’ in Principia involved the very confusions which he seems to imply others should avoid. Against this backdrop, however, his subsequent treatment of ‘right’ in his Ethics, as well as his retrospective remarks about the relationships between ‘good’ and ‘right’ in his “A Reply to My Critics”, can be interpreted as both predictable and necessary. If the argument developed in this paper is sound, the explanation of Moore's abandonment of his earlier account of ‘right’ is not, as he says, merely because that account is “paradoxical”, but lies, instead, in a latent inconsistency between his Principia account of this predicate and other principles implied in that work.