《经济学人》社论的机器翻译及译后编辑问题探析
An Analysis of Machine Translation and Post-Editing Issues in The Economist’s Editorials
摘要: 本文旨在研究基于LLM的机翻系统在《经济学人》社论英译汉中所出现的典型问题及译后编辑重点。本文首先简要回顾了机器翻译与译后编辑的相关研究,以及新闻社论的文本特点与翻译标准,继而采用个案研究法与对比研究法,以《经济学人》2026年5月刊社论“From cyber-security to biosecurity”的英译汉实践为研究对象,对比DeepSeek-V4-Pro、豆包Doubao-Seed-2.1-Turbo模型的输出译文与人工译后编辑版本,从专业术语、搭配与表达、句式结构、隐喻修辞等维度系统考察了AI翻译在新闻社论中的典型问题。研究发现,两个系统在术语的概念化处理、语序调整、被动语态转换、定语从句逻辑显化、对隐喻修辞的把握等方面存在不足,而译后编辑的核心任务可概括为“去欧化、显逻辑、调语气”。本文指出,在“数智翻译时代”,人类译者应与机器翻译系统保持“竞合”状态,一方面充分利用技术带来的效率提升,另一方面持续强化自身在英汉对比意识、语体敏感度和语境推理能力方面的优势。
Abstract: This paper aims to investigate the typical problems and post-editing priorities of LLM-based machine translation systems in the English-to-Chinese translation of The Economist editorials. It begins with a brief review of relevant studies on machine translation and post-editing, as well as the textual features and translation standards of news editorials. Adopting a case study approach and a comparative research method, the study takes the English-to-Chinese translation practice of the editorial “From cyber-security to biosecurity” published in the May 2026 issue of The Economist as its research object. It compares the output translations of the DeepSeek-V4-Pro and Doubao-Seed-2.1-Turbo models with a human post-edited version, and systematically examines the typical problems of AI translation in news editorials from four dimensions: terminology, collocation and expression, syntactic structure, and metaphorical rhetoric. The findings reveal that both systems exhibit deficiencies in the conceptualization of terminology, word-order adjustment, passive-voice conversion, logical explication of attributive clauses, and the handling of metaphorical rhetoric. The core tasks of post-editing can be summarized as “de-Europeanization, logical explication, and tonal adjustment”. This paper argues that, in the age of digital intelligence translation, human translators should maintain a state of “coopetition” with machine translation systems: on the one hand, they should make full use of the efficiency gains brought by technology; on the other hand, they should continuously strengthen their advantages in English-Chinese contrastive awareness, stylistic sensitivity, and contextual reasoning ability.
文章引用:袁媛. 《经济学人》社论的机器翻译及译后编辑问题探析[J]. 现代语言学, 2026, 14(8): 322-331. https://doi.org/10.12677/ml.2026.148789

参考文献

[1] International Organization for Standardization (2020) ISO 20771:2020 Legal Translation—Requirements.
https://www.iso.org/obp/ui#iso:std:iso:20771:ed-1:v1:en
[2] 王湘玲, 杨艳霞. 国内60年机器翻译研究探索: 基于外语类核心期刊的分析[J]. 湖南大学学报(社会科学版), 2019, 33(4): 90-96.
[3] 罗季美, 李梅. 机器翻译译文错误分析[J]. 中国翻译, 2012, 33(5): 84-89.
[4] 崔启亮, 李闻. 译后编辑错误类型研究: 基于科技文本英汉机器翻译[J]. 中国科技翻译, 2015, 28(4): 19-22.
[5] 包凯. 谷歌翻译汉译英错误类型及纠错方法初探[J]. 中国科技翻译, 2017, 30(4): 20-23.
[6] 李国兵, 张敬. 机器英译物流论文摘要问题探究[J]. 中国科技翻译, 2022, 35(2): 24-27, 39.
[7] 耿芳, 胡健. 人工智能辅助译后编辑新方向——基于ChatGPT的翻译实例研究[J]. 中国外语, 2023, 20(3): 41-47.
[8] International Organization for Standardization (2017) ISO 18587:2017 Translation Services—Post-Editing of Machine Translation Output—Requirements.
https://www.iso.org/obp/ui/en/#iso:std:iso:18587:ed-1:v1:en
[9] 冯全功, 崔启亮. 译后编辑研究: 焦点透析与发展趋势[J]. 上海翻译, 2016(6): 67-74, 89, 94.
[10] 朱慧芬, 赵锦文, 诸逸飞. 在线机器翻译的译后编辑原则研究: 以“八八战略”为例[J]. 中国科技翻译, 2020, 33(2): 24-27.
[11] 陈胜, 田传茂. 在线翻译平台汉英翻译的问题及译后编辑: 以石油地质文献为例[J]. 中国科技翻译, 2021, 34(1): 31-34.
[12] 冯全功, 张慧玉. 全球语言服务行业背景下译后编辑者培养研究[J]. 外语界, 2015(1): 67-73.
[13] 冯全功, 刘明. 译后编辑能力三维模型构建[J]. 外语界, 2018(3): 63-70.
[14] 王律, 王湘玲. ChatGPT时代机器翻译译后编辑能力培养模式研究[J]. 外语电化教学, 2023(4): 16-23+115.
[15] 杨艳霞, 魏向清. 基于认知范畴观的机器翻译译后编辑能力解构与培养研究[J]. 外语教学, 2023, 44(1): 90-96.
[16] Reiss, K. (2000) Translation Criticism: The Potentials and Limitations: Categories and Criteria for Translation Quality Assessment. St Jerome Publishing, 16.
[17] 张美芳. 文本类型理论及其对翻译研究的启示[J]. 中国翻译, 2009, 30(5): 53-60+95.
[18] 连淑能. 英汉对比研究[M]. 北京: 高等教育出版社, 2010: 242-255.