Bilingual Parallel Corpora And Student Translators' Translation Competence

During the middle of the 1980s,a Swedish scholar in Translation Studies(TS)investigated "translationese" with electronic corpora of the Swedish language,and at the same time,scholars in Computer Science found the great potential of corpora for Statistical and Example-based Machine Translation.Since then the corpus approach has played a more and more important role in studies such as the research on Translation Universals,features of translated language,the style of translators,translator training and education,Interpreting Studies as well as Machine Translation,bringing about a radical change in TS.Up till now,studies on corpus-based/-driven translator training and education have mainly concentrated on the issues concerning corpora's functions as a kind of teaching aid or their implementation in teaching or training situations.Almost all these studies are theoretical explication or experimental pilot studies with less than 8 subjects in the control group or the experimental group.Specifically,some researchers,based on translation theories or pedagogy,have proposed the theoretical framework as well as the concrete principles of incorporating corpora into translation teaching or translator training.Other researchers have conducted a couple of experiemental pilot studies exploring the translation quality produced by only several translators consulting corpora.Still others have explored the learning experience of student translators when they read the parallel concordance lines.However,it is rarely studied in what aspects corpora can influence the translation competence and thinking of translators when they consult this reference tool.The present dissertation attempts to study the translation competence of student translators consulting bilingual parallel corpora(BPC).Specifically,it explores whether student translators having access to BPC can produce translations of higher quality than those without access to BPC;on the other hand,it studies whether there are any patterns that can be identified in the mental activities of student translators when they consult BPC.The dissertation hence addresses two sets of research questions.The first set of research questions focuses on the display of the translation competence of student translators when they consult BPC:1.Are BPC beneficial to the improvement of the translation competence of student translators when they conduct translations in different directions?2.Are BPC beneficial to the improvement of the translation competence of student translators when they conduct translation of typical texts of Language for General Purposes(LGP)and Language for Specific Purposes(LSP)?3.When taking into account the level of L1 and L2 proficiency,are BPC beneficial to the improvement of the translation competence of student translators when they conduct translations in different directions and of different text types?The second set of research questions centres on the mental activities of student translators when they consult BPC:4.What influences do BPC have on the mental activities of student translators trying to comprehend the Source Text(ST)and produce the Target Text(TT)?5.Can any patterns be identified in the mental processes of student translators who consult BPC to tackle translation difficulties and make appropriate translation decisions?6.What are the problems and difficulties for student translators when they consult BPC?7.How can BPC be profitably exploited to produce the most positive effect on the mental activities of student translators in their translation-decision-making process?In the present experimental study,the researcher selected as the subjects two classes of third-year English majors from School of Foreign Studies,Anhui University,and on the basis of a self-designed Computer-aided Translation Teaching Platform(CATTP)as well as a constellation of bought,DIY and freely accessible online BPC,gathered data by experiments of letting student translators translate within a time limit of two periods and conducting retrospective interviews just after they finish translating.The same data collection process was repeated four times over a whole academic year.Then the software tools such as Excel,GoldWave and SPSS were used to process the collected data.By a combination of quantitative and qualitative analyses,the researcher explored the translation competence of student translators consulting BPC,that it,the quality of their translation when student translators had access to BPC as well as the characteristics of their mental activities when they consulted BPC.In other words,the researcher treated the consulting of BPC as the independent variable,and explored the dependent variable,that is,the translation competence of student translators as well as the characteristics of the mental activities of student translators when the experimental group of student translators consulted BPC in their translating process while the control group did not consult BPC.Moreover,the researcher put forward a translation-process-based theoretical model of translation competence,holding that the abstract concept of translation competence can be decomposed into "the ability to understand ST" and"the ability to produce TT",which correspond with two operationalised indicators:"conveying in TT the original ST information"(a six-point rating scale)and"expressing in TT the original ST style"(a four-point rating scale).These two indicators can be utilised to assess the translation competence of student translators.After the quantitative and qualitative analyses,the research findings of the present study can be summarised as follows:I.With regard to the display of the translation competence of student translators1.In both English-Chinese and Chinese-English translation,BPC are only useful in several aspects of the student translators' translations such as the translation of frequently used expressions and special terms as well as the production of collocations that conform to the norms of Target Language(TL).2.In English-Chinese LGP translation,BPC has shown some beneficial effects on the improvement of the translation of student translators,but in Chinese-English LGP translation,BPC do not suggest any positive effects.On the other hand,in both English-Chinese and Chinese-English LSP translation,BPC are only useful in respects of the appropriate translation of some special terms and expressions instead of improving the quality of the overall translation of student translators.3.The translation by student translators with a higher level of L1 proficiency who have access to BPC is no better than that of those who do not have access to BPC in different translation directions as well as translation of different text types;however,in Chinese-English LSP translation,BPC could improve the translation of such kind of student translators in such aspects as the accurate translation of special terms and the idiomatic usage of expressions in LSP,For student translators with a lower level of L1 proficiency,the situation is quite similar.On the other hand,BPC may have some negative effects on student translators with a higher level of L2 proficiency,for they might make a wrong judgment when facing parallel concordance lines of varying translation quality,leading to poor quality of their own translations,For student translators with a lower level of L2 proficiency,consulting BPC could not obviously help to improve their overall Chinese-English translation quality,even though they could find some useful references from BPC in their translating process.?.With respect to the mental activities of student translators4.(1)As far as the comprehension of LGP ST is concerned,BPC can help student translators thoroughly understand the original English text;furthermore,they can even help student translators understand some archaic words or expressions of the Chinese ST by their corresponding English translations;(2)As far as the production of LGP TT is concerned,BPC could provide student translators with more alternative expressions,and contribute to the improvement of their level of L2 proficiency,thus making their translation all the more conform to the L2 norms;(3)As far as the comprehension of LSP ST is concerned,BPC could present student translators with a relatively complete context,and help them understand the usage of the terms and specialised expressions consulted;and(4)As far as the production of LSP TT is concerned,BPC could help student translators translate in a more idiomatic and professional way,particularly conducive to their approximation to translations by the professional translators familiar with related background knowledge.5.The mental activities of student translators consulting BPC suggest a general trend of divergent thinking when they face abundant concordance lines of bilingual data.These bilingual data can activate more nodes(corresponding translation equivalents)in the bilingual mental lexicon of student translators.Then these student translators can make full use of relevant semantic,syntactic and pragmatic information and knowledge already stored in their bilingual mental lexicon to verify or adapt those translation equivalents provided by BPC,and finally produce the expressions appropriate for the TL context.6.During the processes of consulting BPC,student translators encounter four major types of problems:(1)There are always some expressions that cannot be found in the constellation of BPC used in the present study;(2)So many similar bilingual translation examples of the item consulted are crammed into the first dozen webpages drawn from the BPC used in the study,which is considered as being informationally redundant;(3)There is no parallel correspondence at the lexical or phrasal level in those concordance lines of bilingual data,causing inconvenience of consulting to student translators,especially when the bilingual translation examples are long sentences or even paragraphs;and(4)There is no chance for student translators to review the workable translation solutions obtained from the previous experience of consulting BPC in class,especially those translation solutions to the difficult ST expressions,and student translators consequently feel that the translation knowledge acquired in class through consulting BPC cannot be internalized as their own translation knowledge.7.In terms of the effective exploitation of BPC in training and teaching student translators,the present study finds:(1)In class,student translators had better consult BPC to revise their translations only after they have finished their translations without access to BPC-this is called "post-verification"translation pedagogy,which is more effective than the method of letting student translators copy directly the translation equivalents consulted from BPC;and(2)After class,student translators are encouraged to consult BPC on their own and share with their classmates the successful experience of using this reference tool to deal with translation problems—this is called an"experience-sharing" learning method,which could help student translators form a learning community,and increase their experiences of using this reference tool by mutual inspiration.The combination of the above two methods could bring out the full potential of BPC in training the thinking of student translators.Although corpora have been proved to be a powerful research tool in TS and attracted many researchers working to integrate this reference tool into translator training and education,the present study reveals:Consulting BPC could not obviously improve the overall translation quality of student translators.Therefore,the key to training student translators at the initial stage is to cultivate their critical thinking ability and to develop their language competence in both the Source Language(SL)and the Target Language(TL).Correspondingly,translation teachers should not take corpora as the most important element at this stage in their teaching.Corpora are only assisting tools for teaching.They can provide student translators with more alternative translation equivalents,and help them solve translation problems.However,due to the size of the corpora and the quality of their data,they cannot be considered as a "cure-all" for all translation problems ever encountered by student translators in their translating process.