Translation? Cognition? 翻譯乎?認知乎?

14 Jul 2015

Source: umagazine, P52-57

The University of Macau’s (UM) Centre for Studies of Translation, Interpreting and Cognition (CSTIC) was inaugurated on 9 November 2014. UM Rector Wei Zhao, Vice Rector for Student Affairs Haydn Chen, Prof William Shi-Yuan Wang from the Chinese University of Hong Kong, and University System of Taiwan Chancellor Dr Ovid J L Tzeng, along with other guests, attended the inauguration ceremony. UM has long offered translation and interpreting training programmes between various language pairs, including Portuguese and Chinese, English and Chinese, and Japanese and Chinese. The training programmes include a minor in translation, MA in Translation Studies, and PhD training in Translation Studies (offered under the PhD in Linguistics programme). The establishment of the CSTIC is expected to bring UM to a new level in teaching and research in the field of translation and interpreting studies.

 front page大相   

Translation Is Everywhere

Translation is as old as language itself. If people who speak different languages want to communicate with one another, they have to rely on the help of a competent translator or interpreter. It is said that Zhuge Liang, the famous statesman, ideologist and strategist during the Three Kingdoms Period in ancient China, was accompanied by a translator during his military campaign aimed to pacify the southern tribes. Another distinguished historical figure, Admiral Zheng He, was also accompanied by a translator during his seven epic voyages to Southeast Asia, the Middle East, and Africa. Since the 1990s, China has experienced rapid economic growth amid the trend of globalisation, and is now one of the major economic powers in the world. With this comes an increasing demand for translators and interpreters in practically every sphere of society, which has in turn fuelled the demand for the training of these professionals. UM’s programmes in translation and interpreting were designed precisely for the purpose of equipping students with bilingual proficiency.


The Rise of Translation Studies

In the 1970s, Translation Studies first rose as a new academic discipline. The discipline was started by European scholars and later spread to other parts of the world, including North America, Asia, and Australia. Translation Studies as an independent discipline has two meanings. First, it means the construction of a theoretical system that studies the process of translation, translated works, and the social and cultural functions of translated works. Thus done, training of translators can be conducted within the framework of the system following a set of principles instead of relying on snippets of subjective experience sharing by individual translators. Second, it means breaking away from tradition. Traditionally, translation Studies was a sub-discipline of Comparative Literary Studies. In China, literary works translated from foreign languages into Chinese are taught in the Chinese department as part of Foreign Literature Studies, which also deals with the studies of the original texts of these translated works. Comparative Literature Studies in the Chinese department is the comparative study of Chinese literature and foreign literature. However, faculty members in the Chinese department generally do not engage in the practice of translation. Instead, they use literary works which have been translated into Chinese by faculty members in the department of foreign languages or professional translators as material for teaching and research. Also, Comparative Literature Studies in the Chinese department does not study the process of translation itself, because to do so one must have a good command of the foreign languages involved. As a result, comparative study of the process of translation and of the quality of translated works falls on the shoulder of the department of foreign languages. Comparative Literature Studies in the department of foreign languages often means comparative studies of foreign literary works and their translated versions in Chinese, or comparative studies of Chinese literary works and their translated versions in foreign languages. Interestingly, such a translated works-oriented two-way approach is also a widely adopted practice in Chinese departments at foreign universities.


Translation Relies on the Brain’s Cognitive Function

Throughout history, translation has served a very practical purpose in teaching foreign languages to non-native speakers. You want to learn a foreign language? Okay, let me teach you some new words and grammar rules, and then let us do translation exercises in both directions over and over again. The more translation exercises you do, the more familiar you will become with the language. The more difficult the content you can translate, the more you have progressed in the foreign language you are learning. But the question is: what role do such translation exercises between two languages play in helping students grasp a language of which they had no knowledge to begin with? We have to understand that translation is not like taking out a pen to do a piece of homework; it is a process that takes place inside the brain. In other words, the brain is where the translation is really done. It does not matter who the translator or sponsor is, what the intention or purpose of translation is (be it for revolution, democracy, science, spread of the gospel, or for the appreciation of literature), what the ideological orientation is, or what the religious beliefs are. The conversion from the source language to the target language must go through the brain. There is simply no other way to accomplish this process. Translation is a mental activity, or, to be exact, a cognitive process where linguistic signals (whether alphabetic or phonetic) are received by one’s sense organs (such as the eyes or the ears), decoded by the brain, and then sent out again, also in the form of linguistic signals, but in a different language. Because this process of conversion involves dealing with the differences with respect to a number of factors, including linguistic contexts, cultures, and ideologies, there is necessarily a process of ‘integration’ between when the signals are received and when the converted signals are transmitted, and this is what ‘translation’ means in the strictest sense of the word.

To study the cognitive process of translation, one must adopt the methodology of empirical sciences; that is, to systematically collect, observe and analyse samples and then to systematically interpret the results. In mature sciences such as physics, chemistry, biology and engineering, there are well-developed theoretical and methodological formulas in each of the three steps mentioned above. In disciplines such as psychology and linguistics, where the object of study concerns humans, there are also well-developed theoretical and methodological formulas. This is especially true of experimental psychology. Moreover, as a result of the rapid development in economic, scientific and technological realms over the past three decades, high-tech apparatuses are being used more widely than ever, greatly benefitting researchers. For instance, high-performance computers, software that facilitates research (such as statistical analysis programme, keylogger, etc), eye trackers, EEG, MEG, PET, MRI, and the like were thought to be used by scientists only, but they are now used in Translation Studies.

The input of one language and the output of another in the process of translation (especially in interpreting) involves the various functions of the brain, especially the cognitive function. Therefore, it is impossible to study the process of translation without using these tools. Scholars of the humanities use laboratory equipment to solve problems in their own fields (eg: using fMRI to scan the brain of a reader while reading literary works), while scholars in science and technology use laboratory equipment to solve problems related to the humanities (eg, Natural Language Processing, NLP). The infusion of the two seems to be the trend and a sign of the progress of society. Of course, one area where there is great potential to be tapped by scholars of Translation Studies, in the absence of their own specialised laboratory equipment, is the construction of corpus, which could consist of texts, recordings of interpretation, or transcripts of such recordings.

UM participants in National Interpreting Contest 2012

Using Corpus in Translation Studies

Traditional corpus technologies come from corpus linguistics and computational linguistics. The original purpose was to collect a large body of natural language data in order to find grammar rules and language usage rules, especially morpheme- and phrase-related rules. Theoretically, this approach deliberately defies the theory-oriented approach to the studies of languages (eg: it does not believe in the existence of universal grammar rules, and maintains instead that any language is a collection of set structures.) Technically, a computer is generally able to search for forms, morphemes, and phrases of a limited length. However, when it comes to searching for sentences or linguistic units above the sentence level, the task becomes more difficult for a computer, and this is when human annotations become necessary. Researchers making manual annotations is an unavoidable part of using corpuses in Translation Studies. This is because under most circumstances when we are studying the texts (source text and translated text), we are not looking at morphemes or words but mini-discourse units (MDU) composed of textual sentences. And this is a distinctive characteristic of using corpus in Translation Studies. As for what kind of corpus should be constructed, that depends on the purpose of the research.

For instance, translation and interpreting are different in that the former is not subject to any time constraints (we don’t know how much time the translator spent producing the translation), while the latter (especially simultaneous interpreting) must be done under time pressure. What this means in the creation of corpuses is that if we are creating a corpus for the purpose of studying translation, then we must find a minimum of two translated texts for each source text, not only because that is the only way to identify cognitive patterns—the common characteristics exhibited by different translators, but also because this can rule out the role of the length of time spent producing the translated text in contributing to such commonality. A corpus for studying interpreting, on the other hand, does not have to meet this ‘one-to-many’ requirement. The same time constraints render the process basically the same regardless of who the interpreter is. In other words, after identifying the factors that pertain to individual interpreters, a one-to-one study is usually sufficient to reveal the commonality of the process. Of course, some other important questions, such as how we might conduct quantitative and qualitative analysis of the patterns automatically generated by the available corpuses, are too complex and involve too many theoretical and methodological issues to be covered within the scope of this article. Suffice it to say that with neuroscience advancing at an astounding pace today, there has never been a better time to study the cognitive process of translation and interpreting.

來源:《澳大新語》, 第52-57頁

澳 門大學翻譯傳譯認知研究中心在2014年11月9日正式掛牌成立,趙偉校長、程海東副校長(學生事務)、香港中文大學王士元教授、台灣聯合大學系統校長曾 志朗院士等等,出席了成立儀式。澳大歷來設有翻譯傳譯課程,含葡-漢/漢-葡、英-漢/漢-英、日-漢/漢-日等語種,包括學士副修、碩士、博士等課程。 翻譯傳譯認知研究中心的設立,為推動翻譯傳譯的教學與研究更上一層樓。



翻譯自人類有語言起,便已存在。講不同語言的人要溝通,只好借助譯員的口頭傳譯或者書面文字轉譯。據說,諸葛亮深入不毛,平定南蠻,就有譯員隨行; 鄭和七下西洋,隨員兼作翻譯。20世紀90年代之後,中國經濟高速起飛,融入全球化的大潮,今天赫然已居世界之重。隨之而來的,是譯事無處不在。貿易、金 融、股市、媒體、涉外行政司法海關等,幾乎每日每時都涉及譯員的工作。對譯員的需求多了,自然就有了培訓譯員的要求。澳大的翻譯傳譯課程就是為了培養學生 的雙語互譯能力而設。



20世紀70年代,翻譯學正式興起成為一門獨立學科。所謂「興起」,是指由歐洲學者發起,再達至北美、亞洲、澳洲等地區。所謂「獨立」,有兩個意 思。其一,要建立起一個關乎翻譯過程、翻譯作品,以及翻譯作品的社會文化功能的理論系統,使之可以有原則性的指導譯員培訓,不再依靠譯者抒發感受的隻字片 語來指導翻譯教學。

其二,要突破傳統。傳統上,翻譯屬於比較文學範疇。在中國,譯成中文的外國文學作品,是在中文系教的,屬於外國文學科的大範疇,也包括對這些已經譯 成中文的外國文學作品的研究。在中文系,比較文學就是中、外文學的對比研究。但是,中文系老師自己基本不做翻譯,靠的是外文系老師和社會上的外語人才已經 做出的外國文學譯本來教書和做研究。另外,中文系的比較文

學研究不涉及翻譯過程本身,因為非通曉外文而不可。這樣一來,涉及翻譯過程、譯文質量的對比研究,就落到外文系身上。於是,外文系的比較文學常常是 拿外國文學作品和它的中文譯本進行對比,以及拿中國文學作品和它的外文譯本進行對比。這種雙向的、基於譯文的中、外文學作品的對比研究,在海外的中文系也 是大行其道,有異曲同工之妙!



古今中外,翻譯活動的一個非常實際的用處,就是用於外語教學。你要學一門外語嗎?好,教你一些生詞,教你一些語法,然後就是反覆地來回翻譯。翻譯得 越多,你對要學的語言就越熟悉;翻譯的東西越高深,證明你的進步越大。問題是,在兩種語言的來回轉譯之中,對學生掌握一門原來完全不熟悉的語言到底起了甚 麼樣的作用?須知,這個過程並不是拿起筆做功課而已,而是通過大腦來完成的。大腦才是完成轉譯的地方。不管是誰來做翻譯,不管譯者受誰贊助,意圖與初衷如 何(做翻譯是為了革命,還是為了民主科學,還是為了傳播福音,還是為了欣賞文學),意識形態傾向如何,宗教信仰如何,源語至譯語的轉換必須經過大腦,別無 他途,是譯者的腦力勞動,或者說是大腦的認知活動。感官(如眼、耳)接收到的言語信號(字元或語音)通過大腦解碼之後,其中的資訊再通過大腦用另一種言語 信號傳遞出去。因為語言系統的轉換、語境、文化、意識形態等等的不同,接收到「資訊」跟再傳遞的「資訊」之間必然有一個「整合」過程,這就是最嚴格意義上 的翻譯。

要研究翻譯的認知過程,必須採納經驗科學的研究方法。即系統地採樣、系統地觀察與分析採到的樣本、系統地對分析結果進行解釋。成熟的經驗學科如物 理、化學、生物以及各類工程學,在上述三個階段都有成套的理論和方法程式。研究對象跟人有關的心理學和語言學,理論和方法程式也相當完備,尤其是實驗心理 學。另外,過去30年經濟、科學技術的發展帶來的一項碩果,就是高科技產品比以往任何時候都更普及,使研究者們受惠,比如大容量、高效能的計算機、各種便 利研究的軟件(比如統計程式、鍵盤輸入記錄程式等等)、眼動儀(eye-tracker)、腦電波儀(EEG)、腦磁波儀(MEG)、正電子儀 (PET)、磁共振儀(MRI)等等。這些以前以為只有科技研究者才用的工具,翻譯研究者也照樣用。

翻譯(尤其是傳譯)時不同語言的一進一出,完全是人的大腦和認知能力的應用,非用這些工具而不逮。人文學者應用實驗工具解決自己的問題(再如用 fMRI調查讀者閱讀文學作品時的大腦成像),科技學者應用實驗工具解決跟人文有關的問題(譬如自然語言處理),二者之間交叉融合,似乎是學科發展的方 向、時代前進的象徵。當然,翻譯研究者如果沒有自己的實驗設備,大可作為的地方就是建立語料庫。語料庫可以是文本,也可以是口譯錄音及其文字的轉寫。



傳統的語料庫技術是從語料庫語言學和計算語言學那裡傳承下來的。初衷是採集大量的自然語言語料,從中找到語法規則和語言應用的規則,尤其是語素和詞 組的規則。理論上,它是特意要與由理論導向的語言研究相對抗(譬如不相信存在普遍語法規則、認為語言系統是各種特定構式的集合)。技術上,計算機搜索形 態、語素、長度有限的詞組,基本上做得到,而要搜索句式和比句大的語言單位,就有困難,必須要進行人工標註。而研究者要親力親為去做標註,正是利用語料庫 來研究翻譯不可避免的一關,因為我們要從文本(原文和譯文)中去看的東西,絕大多數情況下不是語素和詞,而是以文本句(textual sentence)為單位的「話語小段」(mini-discourse units,MDU)。這是應用語料庫來進行翻譯研究的特質。至於要建甚麼樣的庫,是跟研究目的相關的。

比如,筆譯和傳譯不同,前者基本不受時間的限制,我們不知道譯者花了多少時間把譯本生產出來,而後者(尤其是同聲傳譯)非守時而不可。因此,文本庫 必須是「一對多」,即一個原文對應兩個或以上的譯文,如此,方能找到文本中認知特徵的共性,即不管譯者是誰,都會表現出來的那些特徵。而且,因為是「一對 多」,如果有共性,那就是跟時間因素無關的共同特徵。與之相對,傳譯庫則不必是「一對多」,因為時間因素的限制,任何傳譯員,在即時傳譯的條件下,傳譯過 程應該大致相同,除去個人因素,留下來的,足以反映過程的共性。當然,即使是拿語料庫作工具,取語料庫自動生成的語料模式,如何去定量定性分析,涉及一系 列的理論和方法程式的問題,這裡就不一一贅述了。今天,大腦科學進展神速,翻譯認知過程的研究隨之興起,正是適逢其時。