| Self-repair is self-initiated repair of speech errors made by the speaker.It’s a universal linguistic feature and the main form of repair in conversation.As a complex language conversion activity,interpreting cannot be as accurate and fluent as written language;it’s full of errors and various non-fluent phenomena,inducing self-repair occurrence.This thesis aims to analyze the self-repair in English-Chinese consecutive interpreting,explore the definition,taxonomy and causes of self-repair,to deepen trainee interpreters’ understanding of this phenomenon and help them better discover,analyze and solve problems.This thesis uses the keynote speech by Mark Zuckerberg in 2016 APEC as experimental material.It introduces previous researches on self-repair,proposes a tentative classification of self-repair on the basis of researches by Levelt and Kormos,and gives further illustration on the classification of and reasons behind self-repair with the combination of the practice case,transcription and audio analysis.The thesis concludes with the following findings:(1)There are four types of self-repair in the practice case: repeated information self-repair,different information self-repair,error repair and appropriateness self-repair.(2)Among the four types,repeated information self-repair occurs the most,indicating that the interpreter uses this strategy to gain more time because of failure to come up with accurate expression immediately after self-monitoring of speech errors.(3)Three main reasons behind self-repair are negative interlingual transfer,poor listening and discerning skills.This thesis put forward solutions,hoping to help other trainee interpreters. |