Abstract

Many natural language expressions (e.g. predicates of taste, aesthetic and moral terms, epistemic modals, “know”, gradable adjectives, etc.) are perspectival, in the sense that they require a perspective (point of view, judge, or subject) to be supplied for their semantic interpretation. Although most of the time such predicates are used with the speaker’s perspective as default, the perspective can be shifted by using them from someone else’s point of view (“exocentrically”) or via explicit “for”-phases, quantifiers or attitude and speech verbs. Another phenomenon of interest is perspectival plurality: the existence of readings of sentences with two or more perspectival terms of the same kind in which appeal to two different perspectives is needed. I call the sum of all these phenomena “perspective-shifting” and in this paper I tackle the issue of what is the best way to handle it in a relativist framework, with the explicit aim of providing a unified relativistic account. To this end, I explore two possibilities: i) an intensional approach, according to which all shifters are treated as perspective-shifting operators; ii) an extensional approach according to which all shifters are treated as variadic operators. Neither of these two accounts, however, has taken perspectival plurality into consideration. Relying on previous work, I argue that the best way to account for the latter is to introduce a sequence of parameters for perspectives. I then discuss how the two approaches mentioned fit in this novel framework.