Binding theory explores how noun phrases relate to each other in sentences. It sets rules for anaphors, pronouns, and referential expressions, explaining when they can refer to the same entity or must refer to different ones.
These principles help us understand why certain sentences sound right or wrong. They're based on syntactic structure, particularly the concept of c-command, but don't account for all factors influencing how we interpret references in language.
Binding Theory and Coreference Constraints
Principles of binding theory
- Binding theory consists of a set of principles governing the referential dependencies between noun phrases (NPs) in a sentence
- Aims to explain the distribution and interpretation of anaphors (reflexives and reciprocals), pronouns, and referential expressions (names)
- Principle A states that anaphors must be bound in their local domain
- Binding is a syntactic relation where an NP (the antecedent) c-commands and is coindexed with another NP
- Local domain is roughly equivalent to the minimal clause or complex NP containing the anaphor
- Principle B asserts that pronouns must be free (not bound) in their local domain
- Pronouns can be bound by an antecedent outside their local domain
- Principle C requires that R-expressions (referring expressions like names) must be free everywhere
- R-expressions cannot be bound by any antecedent in the sentence
Application of binding principles
- Anaphors such as reflexives (himself, herself) and reciprocals (each other) must have a local, c-commanding antecedent
- "John$_i$ likes himself$_i$" is acceptable because "John" c-commands and is coindexed with "himself" in the local domain
- "John$_i$'s mother likes himself$_i$" is unacceptable as "himself" is not bound by a local antecedent
- Pronouns must be free in their local domain but can be bound by a non-local antecedent
- "John$_i$ thinks he$_j$ is smart" is acceptable, where "he" refers to someone else
- "John$_i$ thinks he$_i$ is smart" is also acceptable, with "he" referring to John
- R-expressions cannot be bound by any antecedent in the sentence
- "He$_i$ thinks John$_j$ is smart" is acceptable since "John" is not bound by any antecedent
- "John$_i$ thinks John$_i$ is smart" is unacceptable because the second occurrence of "John" is bound by the first
Syntax and coreference constraints
- C-command is a structural relation between nodes in a syntax tree
- Node A c-commands node B if every branching node dominating A also dominates B, and A does not dominate B
- Binding principles rely on c-command and syntactic domains to determine coreference possibilities
- Anaphors must be c-commanded by their antecedents within their local domain
- Pronouns must not be c-commanded by their antecedents within their local domain
- R-expressions must not be c-commanded by any coindexed NP
- Syntactic structure plays a crucial role in determining the acceptability of coreference relations
Limitations of binding theory
- Binding theory focuses on syntactic constraints and does not fully account for all coreference possibilities
- Some limitations include:
- Does not explain the interpretation of pronouns in discourse across sentence boundaries
- Does not account for the role of pragmatics and context in determining coreference
- Does not explain the distribution of null pronouns (pro) in languages like Spanish and Italian
- Other factors influencing coreference:
- Discourse prominence and accessibility of referents
- Pragmatic factors such as plausibility and world knowledge
- Morphological and semantic features like gender, number, and animacy
- Binding theory provides a foundation for understanding coreference constraints but does not capture all aspects of reference resolution