上海外国语大学英语学院口译研究沙龙系列讲座开讲

　　非全日制研究生网讯：3月26日晚7点，上外英语学院口译研究沙龙系列讲座第一讲在侯靖靖老师的主持下拉开序幕，主讲人潘珺博士，硕士毕业于上海外国语大学英语学院，现任香港浸会大学翻译、传译及跨文化研究系副教授。讲座以腾讯会议形式开展，参会人员包括万宏瑜等上外教师、各专业在读研究生博士生、已毕业上外学子等等。

　　讲座内容围绕“基于语料库的口译研究”展开，主要涉及三大板块：口译语料库研究发展历史，该领域未来发展方向，口译语料库的数据挖掘。

　　回顾口译语料库研究发展历史，潘珺博士谈到了建设语料库最基本的三个要素：真实的语料，口译语料库须包含源语和译语的文字材料、相关的音频及视频材料；电子格式，包括Word、Txt、Excel、PDF、XML等计算机可读的格式；建库遵循的标准，例如单模态和多模态语料库建设过程中对语料处理的方式和侧重有所不同。关于基于语料库的口译研究（CIS），有学者将建设口译语料库的过程与“cottage industries”进行了类比(讲座原文转录如下)：

 　　In 2011, most of the corpus has interpreted whether these are domestic forms, how to get back. And, of course, It involves a lot of manual work, most people most researchers who develop those corpora work independently, and is certain kinds of like homemade system, which is a very very vivid, this status quo of corpus- based interpretive studies, almost 10 years ago. Now let's look at the definition. Wild refers to the circuit certain kind of connection with telecommunication network, and especially to the internet, while also refers to the kind of connection. So, when we talk about CIS evolved into a wild card industry, it means that in CIS, we're trying to employ certain kinds of machine work instead of just a manual work, and we also involve more connection among independent researchers.

　　潘珺博士还推荐了两本非常实用的书籍：Breaking Ground in Corpus-based Interpreting Studies 和Making Way in Corpus-based Interpreting Studies.

　　展望口译语料库研究未来发展方向，潘博士介绍了自己团队2017年开始的一项数字人文项目，建设中英政治语料库the Chinese English Political Interpreting Corpus（CEPIC），这是一项开源性数字资源，研究者无需注册即可使用该数据库。

　　潘博士讲述了团队建设语料库的过程和开发的半自动注释体系“POS tagging”，并针对该语料库的使用结合实例进行了介绍和讲解。

　　例如(转录自讲座原文):

    We're trying to compare the usage of pragmatic group of pragmatic markers, if it was interpreted coppers. And then we found that among all of the pragmatic markers used, there's a tendency of overuse of all, pragmatic, markers, we selected for analysis in native English, which is the US and UK sub corpora, as compared to what we call the CIPSE back then, that's the interpreted English. So there is a tendency for original speakers to employ more often. The strategies or the pragmatic markers as compared to interpreters, which is quite interesting. There are some certain kinds of differences. Definitely especially when we talk about, but we can see that for the case of “however”, CIPSE employs more has a higher average frequency as compared to CEPS. So which indicates that interpreters prefer to use however, as compared to native English speakers. This may have something to do with the possibility of programmatic strategy employed by public interpreters working in political settings, and the strategy of mitigation.

　　关于口译数据挖掘和未来发展，潘博士分别从研究者、软件开发者和语料库建设者三个方向进行了展望。

　　讲座的最后一个环节是答疑，由于参会者学术热情非常高昂，答疑时间长达50分钟，潘博士非常详实认真地回答了大家提出的问题。讲座后大家纷纷表示收获很多，对于口译语料库研究有了更全面了解，自己的相关研究方向也受到了极大的启发。