Natural Language Semantic Theory
Jump to navigation
Jump to search
A Natural Language Semantic Theory is a Theory of how to represent the Semantic Structure of an Utterance.
- Context:
- It can include:
- Expressive Adequacy: How well does the theory allow linguistic meanings to be expressed correctly?
- Grammatical Compatibility: How cleanly does the semantic representation link to other kinds of grammatical information (most notably syntax)?
- Computational Tractability: How simple is it to process meanings; to check semantic equivalence efficiently; and to straightforwardly express relationships between semantic representations?
- Underspecifiability: Does it allow for leaving semantic distinctions unresolved (underspecification), and still allow flexible, monotonic resolution of these underspecifications?
- It can include:
- Example(s):
- First-Order Logic: I have a computer and a laptop. ⇒ Have(Speaker, computer) ^ Have(Speaker, computer)
- See: Natural Language Syntactic Theory.
References
2005
- (Copestake et al., 2005) ⇒ Ann Copestake, D. Flickinger, C. Pollard, and I. A. Sag. (2005). “Minimal Recursion Semantics: an Introduction. Research on Language and Computation 3.4: 281-332. (paper.pdf)