Thesis
Compositional lexical networks: a case study of the English spatial adjectives
- Abstract:
- Most words cannot be given a single precise definition, but instead consist of multiple senses related to each other like members of a family. In cognitive approaches to semantics, this kind of category is described by a lexical, a diagram in which nodes represent senses and arrows represent sense connections. However, lexical network theory is not compositional: it does not explain how lexical networks are combined together to yield the meanings of phrases and sentences. The aim of this thesis is to develop lexical network theory in a formal, compositional setting. I argue that a traditional approach to formal semantics based on the simply-typed lambda calculus is not rich enough to implement lexical networks because it is unable to type the arrows which link word senses together. Instead, I propose replacing simple type theory with Martin-Löf Dependent Type Theory, and show how this allows for a fully compositional implementation of lexical networks. The resulting theory is applied to the description of the English spatial adjectives - high, low, tall, long, short, deep, shallow, thick and thin. These adjectives are an ideal starting point for studying the interaction between lexical and compositional semantics, since they have been studied extensively from both points of view. I illustrate how a compositional theory of lexical networks can provide an interface by which the insights of cognitive semantics can be imported into formal semantics, and vice versa.
Actions
Access Document
- Files:
-
-
(Preview, Dissemination version, pdf, 2.7MB, Terms of use)
-
Authors
+ Arts and Humanities Research Council
More from this funder
- Funder identifier:
- http://dx.doi.org/10.13039/501100000267
- Programme:
- AHRC Doctoral Training Partnership Studentship
- DOI:
- Type of award:
- DPhil
- Level of award:
- Doctoral
- Awarding institution:
- University of Oxford
- Language:
-
English
- Keywords:
- Subjects:
- Deposit date:
-
2023-05-17
- ARK identifier:
Terms of use
- Copyright holder:
- Daniel Martin Worthing
- Copyright date:
- 2021
- Rights statement:
- © the Author(s) 2021
If you are the owner of this record, you can report an update to it here: Report update to this record