The effect of self-repair on judged quality of consecutive interpreting: attending to content, form and delivery

This paper investigates the correlations between self-repair and subjective assessments of student interpreters’ performance in consecutive interpreting (CI). Twelve interpretations from an interpreting contest in China are transcribed, with the self-repairs identified and annotated based on Levelt’s classification (1983), including both overt and covert repairs. In addition to the final scores awarded at the contest, different methods and raters are used to assess the comprising aspects of an overall quality, namely content, form and delivery. Statistical analysis shows that: (1) overt repairs have a strong positive correlation with content, and moderate negative correlations with form and delivery; (2) form and delivery are negatively correlated with covert repairs, in terms of the frequencies of repetitions and pauses, and the mean length of pauses; (3) the judges’ overall assessments are more closely correlated with content than self-repairs. Finally, pedagogical implications for CI training are discussed, as are suggestions for future research.