Rousseau's understanding of legislation as the expression of the general will implies a constitutional principle of legislative supremacy. In turn, this should translate to a narrow, mechanical account of adjudication, lest creative judicial interpretation subvert the primacy of legislative power. Yet in his constitutional writings, Rousseau recommends open-textured and vague legislative codes, which he openly admits will require judicial development. Thus he apparently trusts a great deal in judicial discretion. Ostensibly, then, he overlooks the problem of how legislative indeterminacy—and correspondingly, (...) judicial discretion—may undermine the authority of the general will. However, I argue that Rousseau aims to check judicial subversion of legislative supremacy simply by extending his broader social politics—and specifically, his peculiar concept of republican virtue—to the domain of law. His main concern is that the law should not develop as a mystifying expert practice; therefore, he necessarily rejects any understanding of judicial virtue as lying in principled discourse. Instead, he envisages that judicial power will be checked by a more generic sense of republican virtue. In turn this echoes his apprehension of social differentiation and social complexity as sources of domination and hierarchy. (shrink)

Scott Soames argues that consideration of the practice of legal judgement gives us good reason to favor the partial-definition/context-sensitive theory of vagueness against epistemicism. Despite the fact that the value of power-delegation through vagueness is evidenced in practice, Soames says, epistemicism cannot account for it theoretically, while the partial-definition/context-sensitive theory is capable of it. In this paper, I examine the two possible arguments against epistemicism that can be extracted from Soames’s account: (i) an argument based on unknown obligations, and (ii) (...) an argument based on power-delegation through vagueness. The first argument tries to convince us that, as based on epistemicism, the law has already decided the borderline cases, so that judges have obligatory decisions even in such cases: therefore epistemicism is inconsistent with the discretion of judges in borderline cases. I show that even if we sympathize with Soames’s intuitions concerning the legal practice, the argument he offers is not conclusive since it is either invalid, unsound, or paradoxical. The second argument holds that only the gaps which the partial-definition/context-sensitive theory predicts give judges the possibility of lawmaking in borderline cases. However, by categorizing the vague laws as imperfect laws, the judges can claim the right of lawmaking without any need to refer to gaps in the law. By neutralizing these arguments, I argue that epistemicism is able to explain the phenomena just as well as the partial-definition/context-sensitive theory. (shrink)