Indexicals and Demonstratives
Words like 'I', 'here', and 'now' shift their reference with context, forcing a two-dimensional semantics that separates context-fixed character from utterance content.
Definition
An indexical is an expression whose reference depends systematically on features of the context of utterance, such as the speaker, time, or place; Kaplan models its meaning as a character that yields a content once a context is fixed.
Scope
This topic covers the logic and semantics of context-sensitive expressions: pure indexicals such as 'I', 'today', and 'here', and demonstratives such as 'this' and 'that'. It treats Kaplan's distinction between character (a rule from context to content) and content (the proposition expressed), the logic of demonstratives and the direct-reference thesis for indexicals, and the related problem of the essential indexical and de se attitudes that resist capture in purely third-personal terms.
Core questions
- How can a single expression like 'I' have a fixed meaning yet a shifting reference?
- Are indexicals directly referential, contributing only their referent to content?
- What is the relation between character, content, and circumstance of evaluation?
- Why are first-person, essential-indexical beliefs not reducible to objective descriptions?
Key concepts
- character and content
- context vs. circumstance of evaluation
- direct reference
- the essential indexical
- de se attitudes
- pure indexicals vs. demonstratives
Key theories
- Character and content (Kaplan)
- Kaplan distinguishes an indexical's character — a context-invariant rule determining reference from context — from its content, the directly referential object it contributes, yielding a two-dimensional logic of demonstratives.
- The essential indexical
- Perry argues that beliefs essentially involving 'I' or 'now' cannot be replaced by any objective, non-indexical description without losing their power to explain action, motivating de se content beyond ordinary propositions.
History
Reichenbach analyzed indexicals as token-reflexive in 1947, but the modern treatment stems from Kaplan's 'Demonstratives' (circulated from the 1970s, published 1989), which introduced the character/content framework. Perry's 1979 essential-indexical problem and Lewis's de se account reshaped the philosophy of mind and language around self-locating content.
Debates
- How to accommodate de se content
- Whether first-person, self-locating belief requires a special category of de se content (centred propositions, properties self-ascribed) beyond ordinary propositions, and how this interacts with Kaplan's content/character distinction.
Key figures
- David Kaplan
- John Perry
- David Lewis
- Hans Reichenbach
Related topics
Seminal works
- kaplan1989
- perry1979
Frequently asked questions
- What is the difference between character and content?
- Character is the standing, context-independent meaning of an indexical — for 'I', the rule 'refers to the speaker'. Content is what it contributes in a particular use — the speaker herself. The same character ('I') yields different contents in different mouths, which is how a fixed meaning produces a shifting reference.