Quantification and the Garden Path Effect Reduction: The Case of Universally Quantified Subjects
18870Authors
nan
Abstract
. This paper investigates the effect of quantification in sentence processing. The experimental results show that temporarily ambiguous sentences that begin with the univer- sally quantified NPs reduced the garden path effect in contrast to the ones that begin with bare NPs. This fact is accounted for by assuming that discourse representation structures are incrementally constructed, and a tripartite structure introduced by the universal quantifier gives a room for temporal ambiguity while a single box associated with a bare NP forces one interpretation and to get the correct interpretation, the single box must be rewritten, which results in the garden path effect. Keywords: temporarily ambiguous sentence, garden path effect, DRT, box merger 1
Keywords
Full Content
Quantification and the Garden Path Effect Reduction: The Case of Universally Quantified Subject ⋆ Akira Ohtani a and Takeo Kurafuji b aFaculty of Informatics, Osaka Gakuin University, 2-36-1 Kishibe-minami, Suita 564-8511, Japan ohtani@ogu.ac.jp bCollege of Law, Ritsumeikan University, 56-1 Kita-machi, Toji-in, Kita-ku, Kyoto 603-8677, Japan kurafuji@fc.ritsumei.ac.jp Abstract. This paper investigates the effect of quantification in sentence processing. The experimental results show that temporarily ambiguous sentences that begin with the univer- sally quantified NPs reduced the garden path effect in contrast to the ones that begin with bare NPs. This fact is accounted for by assuming that discourse representation structures are incrementally constructed, and a tripartite structure introduced by the universal quantifier gives a room for temporal ambiguity while a single box associated with a bare NP forces one interpretation and to get the correct interpretation, the single box must be rewritten, which results in the garden path effect. Keywords: temporarily ambiguous sentence, garden path effect, DRT, box merger 1 Introduction Japanese has several types of temporarily ambiguous sentences (TASs) (Inoue, 2006). This paper focuses on the pattern ‘NP-NOM [VP NP-ACC V] NP-DAT NP-ACC V’ as exemplified in (1). (1) Keekan-ga police.officer-NOM [VP hanni-o criminal-ACC tsukamaeta caught ] otoko-ni man-DAT ree-o thank-ACC itta. said ‘A police officer extended his thanks to a/the man who caught a criminal.’ The first NP keekan-ga ‘police.officer-NOM’ tends to be construed as the subject of the following VP, yielding the reading “a police officer caught a criminal” but this interpretation crashes when another NP otoko-ni ‘man-DAT’ shows up after the V tsukamaeta ‘caught’. To get the correct interpretation, the VP preceding otoko-ni ‘man-DAT’ must be construed as (part of) the relative clause modifying that NP, and the first NP must be construed as the subject of the sentence-final V itta ‘said’. This reinterpretation process is known as the garden path (GP) effect. However, the GP effect of TAS slightly reduces when only a bare NP keekan ‘police.officer’ in (1) is replaced with a universally quantified NP subeteno keekan ‘all police officers’ as in (2). (2) Subeteno all keekan-ga police.officer-NOM [VP hanni-o criminal-ACC tsukamaeta caught ] otoko-ni man-DAT ree-o thank-ACC itta. said ‘All police officers extended their thanks to a/the man who caught a criminal.’ ⋆ We are indebted to three anonymous PACLIC reviewers, Masakatsu Inoue and Michinao F. Matsui for their invalu- able comments on an earlier version of this paper. All remaining inadequacies are our own. This research is partially supported by Osaka Gakuin University and Grant-in-Aid for Scientific Research (C), 18530578 and 21500152 of the Japan Society for the Promotion of Science (JSPS). Copyright 2011 by Akira Ohtani and Takeo Kurafuji 25th Pacific Asia Conference on Language, Information and Computation, pages 41–50 41 This difference suggests that semantic representations play significant roles since the syntactic structure of (1) and (2) are all the same,1 and that the quantificational structure introduced by the universal quantifier makes (2) easy to interpret. In this paper, we claim the relevant semantic representations are built independently of syntac- tic …
Content truncated to 500 words
Labeled Text
. This paper investigates the effect of quantification in sentence processing. The experimental results show that temporarily ambiguous sentences that begin with the universally quantified NPs reduced the garden path effect in contrast to the ones that begin with bare NPs. This fact is accounted for by assuming that discourse representation structures are incrementally constructed, and a tripartite structure introduced by the universal quantifier gives a room for temporal ambiguity while a single box associated with a bare NP forces one interpretation and to get the correct interpretation, the single box must be rewritten, which results in the garden path effect. Keywords: temporarily ambiguous sentence, garden path effect, DRT, box merger 1
Publication Details
- Paper ID:
- 18870
- Year:
- 2011
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan