•  
  •  
 

University of Pennsylvania Working Papers in Linguistics

Abstract

We report two experiments examining first/second-person blocking effects on the Chinese long-distance reflexive ziji during on-line processing. Participants read sentences with varying matrix and embedded subjects (Exp1: 1st-person pronoun/3rd-person name; Exp2: 2nd-person pronoun/3rd-person name) and answered comprehension questions probing their interpretations of ziji. Work on English found that structurally inaccessible referents can cause competition at the reflexive, indicated by reading-time slowdowns (Badecker and Straub 2002). In Exp1, the 1st-person blocking condition (3rd-person matrix/1st-person embedded) exhibited slowdowns and a higher-than-expected rate of matrix-subject-interpretations, suggesting 1st-person blocking is not consistently effective. However, the subset of trials with effective blocking (local-antecedent interpretations) revealed no slowdowns. In Exp2, the 2nd-person blocking condition (3rd-person matrix/2nd-person embedded) showed consistent blocking and no significant slowdowns. Our results suggest that referents’ ability to compete depends not only on prominence (Badecker and Straub 2002) but also how it is blocked (person-feature vs. syntactic barrier). Building upon Brunyé et al.’s (2009) finding that 2nd-person pronouns are more effective at triggering perspective-taking than 1st-person pronouns, we suggest that the difference between first- and second- person blocking may be attributable to perspective taking: Identifying with the 2nd-person addressee leads comprehenders to more consistently interpret the reflexive as referring to the local 2nd-person subject, resulting in a consistent blocking effect.

Share

COinS