Self-Repair in Aborted Utterances: A Multimodal Analysis of Simultaneous Interpreting

Abstract

The article presents the results of an empirical research dedicated to the co-occurrence of gestures and self-repairs in simultaneous interpreting. Self-repairs, viewed as a type of disfluencies, are divided into three categories: 1) a disfluency followed by a resolution (“positive” self-repair), 2) a disfluency consisting of repetition of the same lexical unit (“zero” self-repair); 3) an utterance truncated without a restart, which means that such a disfluency is not resolved. The study is based on the assumption that in simultaneous interpreting statistically significant correlation will be observed between various types of self-repair and various types of gestures, such as pragmatic, representational, deictic gestures and adapters. The data was obtained from 18 interpreters who were asked to interpret a popular science lecture from Russian into English. The material was analysed in ELAN and JAMOVI, and quantitative and statistical methods (T-test and ANOVA) were used to check on the hypotheses. The study revealed a strong correlation between self-repairs and gesture usage, which suggests that in moments of disfluency gestures facilitate speech. Concerning the relation between various types of self-repair and functional types of gestures, the hypothesis was not confirmed, although the statistics points to the fact that the simultaneous interpreters tend to use particular types of gestures with certain self-repairs.