Please use this identifier to cite or link to this item: doi:10.22028/D291-24832
Title: The absorption principle and E-type anaphora
Author(s): Gawron, Jean Mark
Nerbonne, John
Peters, Stanley
Language: English
Year of Publication: 1991
OPUS Source: Kaiserslautern ; Saarbrücken : DFKI, 1991
SWD key words: Künstliche Intelligenz
Semantik
Natürliche Sprache
Anapher <Rhetorik>
DDC notations: 004 Computer science, internet
Publikation type: Report
Abstract: The Absorption Principle is a principle of situation theory which restricts the kinds of parametric information which is available. In particular it rules out abstraction over variable occurrences in parametric restrictions (unless the parameter itself is included). In "Anaphora and Quantification in Situation Semantics", Gawron and Peters showed that the Absorption Principle has intuitively correct consequences in applications to quantificational and anaphoric semantics, but Sem, Saebo, Verne and Vestre (1990) point out cases of incorrect consequences. The present paper provides an analysis of the problematic cases in which the Absorption Principle is maintained. A key part of the analysis is the postulation that anaphors may have quantified NPs as antecedents, a position which has been vigorously advocated by Evans (1980). As a consequence, anaphors of this type are called E-Type'. We argue that the pronoun it' in the following discourse must be analyzed as E-Type: Tom has exactly one car. It is red. We provide an analysis of E-Type anaphora with the following properties: (i) the type of the anaphor is derived from the conservative scope of its antecedent; (ii) its semantics is provided by a choice function; and (iii) there is a pragmatic condition that the choice function not be controlled either by speaker or hearer in the discourse. We demonstrate how this accounts for a wide range of facts, including apparently varying quantificational force.
Link to this record: urn:nbn:de:bsz:291-scidok-35810
hdl:20.500.11880/24888
http://dx.doi.org/10.22028/D291-24832
Series name: Research report / Deutsches Forschungszentrum für Künstliche Intelligenz [ISSN 0946-008x]
Series volume: 91-12
Date of registration: 18-May-2011
Faculty: SE - Sonstige Einrichtungen
Department: SE - DFKI Deutsches Forschungszentrum für Künstliche Intelligenz
Collections:SciDok - Der Wissenschaftsserver der Universität des Saarlandes

Files for this record:
File Description SizeFormat 
RR_91_12.pdf15,62 MBAdobe PDFView/Open


Items in SciDok are protected by copyright, with all rights reserved, unless otherwise indicated.