Indexicality and Demonstratives

Core Idea

Indexical expressions like 'I,' 'now,' 'here' and demonstratives like 'this' and 'that' have a character—a rule for determining reference based on context—distinct from their content in a specific context. Kaplan's theory distinguishes these dimensions to handle the semantics of directly referential expressions.

Explainer

From your prerequisite study of indexicals and context-sensitivity, you know that expressions like "I," "now," "here," "today," and "this" shift their reference depending on who uses them, when, and where. I utter "I am hungry" — the word "I" refers to me. You utter the same sentence — "I" refers to you. Same word, different reference. This context-dependence is the basic phenomenon. What Kaplan's theory does is give a systematic two-level semantic framework that explains exactly how this works.

The first level is character: the rule or function that, given a context of utterance, determines the content expressed. The character of "I" is something like: *the speaker of the context*. The character of "now" is: *the time of utterance*. The character of "here" is: *the place of utterance*. Characters are stable across all contexts — the character of "I" never changes, which is why you know how to use it correctly. Characters are the linguistic meaning in the fullest sense: what a competent speaker knows when they know what an expression means.

The second level is content: what the expression actually refers to or expresses in a *particular* context. When I say "I" in my utterance, the content is me — Griffin — and that content is constant across possible worlds. This is what makes Kaplan's indexicals directly referential: once the context fixes the reference, the expression contributes just the object itself to the proposition expressed, not a description. The proposition expressed by "I am hungry" (uttered by me) is the singular proposition *<Griffin, hungry>* — true at a possible world if and only if Griffin is hungry there.

Demonstratives like "this" and "that" complicate the picture because they seem to require a directing intention in addition to context. When I say "that statue," I'm not just pointing to a contextually salient object — I'm directing attention to something specific, and different intentions could fix different referents even holding the context fixed. Kaplan distinguished "pure indexicals" (where context alone fixes reference, like "I") from "true demonstratives" (where a directing intention is needed). This distinction matters for explaining cases where demonstratives misfire, succeed despite pointing errors, or pick out objects the speaker didn't intend. Understanding character and content as distinct dimensions is the key to navigating these cases — and to understanding why the same sentence can express different propositions in different mouths.