Abstraction and Natural Language Semantics

According to the traditional view, a word prototypically denotes a class of objects sharing similar features, i.e. it results from an abstraction based on the detection of common properties in perceived entities. I explore here another idea: words result from abstraction of common premises in the rules governing our actions. I first argue that taking ‘inference', instead of ‘reference', as the basic issue in semantics does matter. I then discuss two phenomena that are, in my opinion, particularly difficult to analyse within the scope of traditional semantic theories: systematic polysemy and plurals. I conclude by a discussion of my approach, and by a summary of its main features.

Data and Resources

Additional Info

Field Value
Source ISSN: 0261-0523
Author Kayser, Daniel
Maintainer CCSD
Last Updated May 7, 2026, 21:13 (UTC)
Created May 7, 2026, 21:13 (UTC)
Identifier hal-00091608
Language en
contributor Laboratoire d'Informatique de Paris-Nord (LIPN) ; Université Paris 13 (UP13)-Institut Galilée-Université Sorbonne Paris Cité (USPC)-Centre National de la Recherche Scientifique (CNRS)
creator Kayser, Daniel
date 2003-05-07T00:00:00
harvest_object_id 44c7731b-bbec-48a4-98bd-df5180fd8bdd
harvest_source_id 3374d638-d20b-4672-ba96-a23232d55657
harvest_source_title test moissonnage SELUNE
metadata_modified 2024-11-29T00:00:00
set_spec type:ART