Abstract
The scope argument challenges Gricean theories of metaphor by claiming that because metaphorical readings survive under logical and intensional operators, they must belong to what is said. I resist this conclusion. I argue that persistence under embedding shows that composition can proceed once satisfiability conditions at the relevant node are repaired by truth evaluable reflexive constraints that do not encode speaker meaning, so long as those constraints have the right semantic type. Drawing on parallels with scalar enrichment, irony, and other embedded implicatures, I defend a subsentential Gricean model in which metaphorical interpretation arises through local pragmatic inference. Interpretation introduces a reflexive constraint on the relevant constituent which is sufficient for compositional evaluation. The hearer may then infer a more determinate sense as speaker meant. This requires addressing two challenges to Gricean accounts of metaphor, the calculation problem and the composition problem. I propose a type driven inferential model grounded in meaning postulates on which literal meaning remains fixed and pragmatic reasoning supplies reflexive constraints that repair satisfiability and license evaluation, guiding the recovery of speaker meaning. The result is a lean semantics paired with a principled account of metaphorical meaning.
Full text
621 characters
· extracted from
oa-doi-fallback
· click to expand
There is a newer version available for this {{ publicationType }}. View latest version
{{ publication.field_name }}
{{ publication.subfield_name }}
Copyright: © {{ publicationYear }} {{ publication.presentation_authors[0].full_name + (publication.presentation_authors.length > 1 ? ' et al' : '') }}. This is an open access publication distributed under the terms of the CC BY 4.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Check the {{ publicationType | capitalize }} Source for copyright and license information.
Listen on
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.