https://dx.doi.org/10.1207/s15327868ms0104_1">
 

Abstract

Novel linguistic metaphor can be seen as the assignment of attributes to a topic through a vehicle belonging to another domain. The experience evoked by the vehicle is a significant aspect of the meaning of the metaphor, especially for abstract metaphor, which involves more than mere physical similarity. In this article I indicate, through description of a specific model, some possibilities as well as limitations of computer processing directed toward both informative and experiential/affective aspects of metaphor. A background to the discussion is given by other computational treatments of metaphor analysis, as well as by some questions about metaphor originating in other disciplines. The approach on which the present metaphor analysis model is based is consistent with a theory of language comprehension that includes both the intent of the originator and the effect on the recipient of the metaphor. The model addresses the dual problem of (a) determining potentially salient properties of the vehicle concept, and (b) defining extensible symbolic representations of such properties, including affective and other connotations. The nature of the linguistic analysis underlying the model suggests how metaphoric expression of experiential components in abstract metaphor is dependent on the nominalization of actions and attributes. The inverse process of undoing such nominalizations in computer analysis of metaphor constitutes a translation of a metaphor to a "more literal" expression within the metaphor-nonmetaphor dichotomy.

Department

Computer Science

Publication Date

1-1-1986

Journal Title

Metaphor and Symbolic Activity

Publisher

Taylor & Francis

Digital Object Identifier (DOI)

https://dx.doi.org/10.1207/s15327868ms0104_1

Document Type

Article

Comments

This is an Author’s Original Manuscript of an article published by Taylor & Francis in Metaphor and Symbolic Activity in 1986, available online: https://dx.doi.org/10.1207/s15327868ms0104_1

Share

COinS