Dissertation
Dissertation > Language, writing > Linguistics > Translation Studies

TM as a Specialized Corpus for Translation Studies

Author XieBingBing
Tutor TuGuoYuan
School Central South University
Course Translation
Keywords TM corpus-based translation studies specialized corpus non-literary translation studies
CLC H059
Type Master's thesis
Year 2012
Downloads 13
Quotes 0
Download Dissertation

This research aims to reveal possibilities of Translation Memory (TM) serving as a specialized parallel corpus in corpus-based translation studies, especially on non-literary translation. It starts from the influence of information age on translation both in theory and practice by development of electronics tools for translation study experts and translators although later ones have acquired support from them many years earlier than the former ones. By previous studies on TM, we find that the technical and formal features of a TM coincides with those of a corpus, more exactly, a specialized parallel corpus. This implies that studies on a corpus could be also carried out on a TM. In light of this implication, we tend to introduce corpus-based translation study method on studying TM as a translation product. As we have review literatures on TM and corpus-based translation studies, it is found that TM studies confine within technical and pedagogical level while corpus-based translation studies tend to be based on literature corpus or general reference corpus. Thus, a link between TM and corpus-based translation studies could be established as TM and corpora share many similarities. In the light of Laviosa’s concept, we believe that a corpus is a link between translation theory and practice and the situation is the same between TM and corpus-based translation studies. Therefore, a case study is illustrated in this research focusing on a technical textbook TM. In accordance with research purpose, this TM is modified and transferred into a specialized parallel corpus for translation studies. A further processing on this corpus has been done including tokenization and annotation. Corpus study softwares help us calculate statistics in this corpus following our research requirement. When examining the word list, frequency of "shall" in TT is salient comparing with its probable counterparts in ST. We first attempt to explain this phenomenon on macro level with corpus-based translation study methods and relevant linguistic and translation theories followed by a micro level analysis which contains classified samples of parallel concordance of "shall". To each specific type of sample, same methodology is adopted as the analysis of macro level. During the study of "shall", it is also found that its dispersion is unequal with certain regular pattern revealed by Plot function of corpus study software. This phenomenon is explained in terms of translator’s strategy. Based on the case study, it is proved that TM is a possible source for corpus-based translation studies. Due to TM’s feature in translation practice, TM is suitable for the development of non-literary translation corpus with certain text types which could be fairly useful in supporting current results in corpus-based translation studies, further discovering features of non-literary translation language, providing implications on MT, CAT as well as LSP translation teaching.The thesis consists five parts. The first part is introduction providing background information of the research with several research questions and hypotheses. Literature review is in the second part based on the recent decade international papers and monographs revealing the development in two fields, CAT and CTS. It also reviews related work and research on corpora which have been proved serving as a link between translation theories and practices. In the light of this, we attempt to discuss a new application of TM on CTS by its features of a specialized corpus, especially for studying non-literary translation. In the third part we illustrate the possibilities and feasibilities of a TM being adopted into a research-oriented corpus. It also discusses the advantages, disadvantages and remedies of this approach. The forth contains a case study on technical textbook TM applied on CTS. By statistical analysis and theoretical explanation, we reveal features of non-literary translation language, part of them of course. The case study serves as an example for possibly further application of other TMs on CTS. At last, there are implications and conclusions with a trinity model. They indicate that multiple application of this research and require us to do further studies

Related Dissertations
More Dissertations