摘要
本研究基于多维度质量指标,构建英-汉翻译错误类型框架,开展眼动实验,考察英译汉神经网络机器翻译错误类型如何影响译者在译后编辑和人工翻译过程中注意资源分配的情况。研究发现:1)词义错误和衔接与连贯问题会使译者在译后编辑过程中分别投入更多的注意资源至外部查询和原文区;2)风格问题和术语问题是机器翻译和人工翻译共同面临的“困难”;3)译者的英语水平、翻译经验与机器翻译错误类型之间的交互效应不显著,而机器翻译的风格问题对译者分配至译后编辑总兴趣区和译文区注意资源的影响,则取决于译者对待译后编辑的态度。
This study investigates the impact of neural machine translation error types on the allocation of attentional resources in post-editing of machine translation(PEMT)and human translation in the English-to-Chinese language pair.An MQM-derived error typology for English-Chinese translation was proposed and data collected from an eye-tracking experiment were analyzed.The results indicate that word sense errors and cohesion and coherence issues are significantlyassociated with the amount of attentional resources respectively allocated to external consultation and source text during post-editing.Stylistic issues and terminology are shown to be“problematic”for machine translations and human translators alike.In addition,the interaction effects between English proficiency/translation experience and machine translation error types are not found to be statistically significant.However,the impact of stylistic issues on the amount of attentional resources allocated to total area of interest and target text is modulated by translators’attitude toward PEMT.
作者
钱家骏
肖维青
李彦
项霞
QIAN Jiajun;XIAO Weiqing;LI Yan;XIANG Xia(College of Foreign Languages,Shanghai Maritime University,Shanghai 201306,China;School of English Studies,Shanghai International Studies University,Shanghai 200083,China;Schoolof Foreign Languages&Literature,Yunnan Normal University,Kunming 650500,China;College of Science&Technology,Ningbo University,Ningbo 315300,China)
出处
《外语教学与研究》
CSSCI
北大核心
2022年第5期750-761,800,F0003,共14页
Foreign Language Teaching and Research
基金
国家社科基金项目“认知工效学视域下的译后编辑过程实证研究”(21BYY063)的资助。
