Against \u27Long\u27 Movement of the Superlative Operator

It is a matter of considerable debate whether degree operators are interpreted in their base position or in some higher position. Kennedy ( 1 997) has shown that degree operators (e.g., the comparative operator) do not interact scopally with quantified expressions. On the other hand, Heim ( 1 999) and Stateva (to appear) have presented evidence that the superlative operator (as in the highest grade) interacts scopally with intensional predicates. This paper argues that despite the fact that the superlative operator seems to interact scopally with intensional predicates, the facts receive a better account under an in situ analysis, rather than a movement analysis, of the operator. This point will be made by (a) looking at examples where a superlative expression is embedded under a propositional attitude verb, and focussing on readings which are neither de re (in the strict sense) nor de dicto (in the strict sense); and (b) examining negative superlatives (e.g., the least high grade) in extensional contexts. Our conclusion will be that Kennedy\u27 s claim that degree operator movement is highly restricted is correct