英译汉同传中的非流利现象研究——以21位口译员的同传实验表现为例

Study on Disfluency in E-C Simultaneous Interpreting—Taking the Simultaneous Transmission Experimental Performance of 21 Interpreters as an Example

非流利现象阻碍语言流畅性,得到口译学界诸多关注。鉴于先前研究仅聚焦个别非流利子类型,该领域亟待更深入系统的探究。因此,本文采用实证法,拟探究英译汉同传中非流利现象的分布特征以及成因。研究数据来自21位学生译员的同传表现,通过Elan 6.1软件进行标注分析。在为期两天的译前准备后,学生译员完成一段300词篇幅的同传任务,内容为新冠疫情期间的英国酒吧。同传结束后,译员参与回顾性访谈,以引导其回忆同传中非流利现象的成因。该研究主要结论如下:1) 在非流利现象的四种子类型中(无声停顿、有声停顿、自我修正及重复),无声停顿出现频次最高,有声停顿和自我修正位居其次,重复出现频次最低。2) 非流利现象的成因归为两类:由高认知负荷导致的非流利和低认知负荷信息中出现的非流利。3) 在高认知负荷下,非流利现象多发生于源语和目标语间的语法及句法差异处,同时,数字、列举以及专有名词等特殊信息类型也易导致非流利现象产生。4) 低认知负荷信息中出现的非流利现象成因包括同传策略、译员生理/心理状态以及个人习惯。

Disfluency hinders the production of fluent speech, and it has received much attention from interpreting scholars. Since most previous research has focused only on one specific subtype of disfluency, a more in-depth and systematic study is needed. Therefore, this study adopts an empirical approach to figure out the characteristics of and motivations for disfluencies in English-Chinese (E-C) Simultaneous Interpreting (SI). The data were collected from the SI performance from English to Chinese of 21 trainee interpreters, managed by the software Elan 6.1. After two days of preparation, the students were asked to interpret a 300-word passage talking about the Britain bars during COVID-19. After the task, retrospective interviews were conducted to stimulate their recall of the reasons for disfluencies in SI production. The major findings are: 1) Among four subtypes of disfluencies (silent pause, filled pause, self-repair and repeat), silent pause occurs most in SI, followed by filled pause and self-repair, while the occurrence of repeat is the least. 2) Motivations for disfluencies are divided into two categories, including disfluencies caused by high cognitive load and disfluencies occurring in normal cognitive load information. 3) Under high cognitive load, disfluencies tend to happen when interpreters encounter grammatical and syntactic differences between the source text and target text, and information such as numbers, enumerations and proper nouns can also cause disfluent output. 4) The triggers for disfluencies when processing normal cognitive load information include SI-specific strategies, the physical/mental state and the idiosyncrasies of interpreters.