The Application Of Fuzziology To Translation Quality Assessment

Translation quality assessment (TQA) is an indispensable part of translation studies, and with the advent of translation activities, translation qualities have always been discussed, evaluated and criticized. However, compared with the study of translation theory and translation history, TQA study is always in a less favorable position. By introducing the theory of fuzzioiogy, this dissertation aims at setting up a fuzziology-based TQA model, attempts to offer a qualitative and quantitative combined TQA approach, and intends to prove that the fuzziology-based evaluative approach could avoid over-subjectivity and inconsistent criteria in TQA.Fuzzioiogy is the study about the theories and methodologies that describe and deal with the problem of fuzziness. Fuzziness, referring to the indeterminacy of belongingness and unclear referential boundary, widely exists in nature and human life. The study of fuzzy phenomenon in phonetics, grammar and meaning proves that fuzziness is an innate feature of language. Translation, as an activity between two languages undoubtedly possesses fuzzy features. In translation assessment, translation theorists and critics seldom claim that one translation is "perfect," and another is "absolutely bad." In most cases, the quality of a translation is between perfection and total failure. This special feature of the quality of translation accidentally coincides with the theory of fuzzioiogy, that is, the borderline between a good translation and a bad translation is somewhat fuzzy. The discovery of the fuzzy features in language, translation and the quality of translation provides theoretical bases for the application of fuzzioiogy to TQA. Discussions on the competence of translators and the complexity of translation further prove the feasibility of this approach.A literature review on the foreign and domestic translation criticism and TQA reflects that traditional TQA approaches are largely qualitative. The limitations of qualitative evaluation are that it leads to subjective judgments and inconsistency of criteria in TQA easily, while over-subjectivity and inconsistency of criteria will both result in the failure of making an objective assessment on translations. Although thereare some tentative studies on quantitative TQA models, most of such models are confined to certain texts, that is to say, each evaluative model might only be applicable to a type of texts. For instance, the functional TQA model might be effective in the assessment of scientific texts, yet may not be able to give a comprehensive assessment to literary works. After inspecting on the relationship and intersection between fuzziology and translation, this dissertation makes an initiative attempt in the establishment of a fuzziology-based TQA model, and actualizes the possibility of making quantitative assessment on different types of texts by introducing the concepts of DOM, fuzzy set, and linguistic variables from fuzziology. In the meanwhile, this model adopts Gu Zhengkun's "Pluralistic and Complementary System for Translation Criteria," where one or several relatively abstract criteria can be set, and gach criterion can be governed by some concrete parameters. In this way, incomprehensive evaluation caused by one-dimensional criterion can be avoided. To objectify the evaluation, different weight values can also be allocated and given to the concrete parameters.Compared with other TQA models, the fuzziology-based TQA model has the characteristics of being open, comprehensive and direct. The model is open because it is widely applicable. To evaluate different text types, we just need to change the parameters and alter the weight values. The model being comprehensive lies in the fact that we can compute the "value" of a translation by applying fuzziological approach, so that it is more persuasive to say that one version is better than another.' What we mean by saying the model is direct is that we can directly perceive the merits and demerits of different versions with the "value" of each version. The DOM value of each individual parameter can also reflect the goodness of a version on different levels.To further demonstrate the practicability and effectiveness of the fuzziology-based TQA model, this dissertation also makes a case study on the assessment of a Tang poem, and applies this model to assess the three different versions of Du Fu's Deng Gao. By adopting this qualitative and quantitative combined approach, we can not only judge the goodness of each version, but can alsoexplain "why."In conclusion, the author claims that it will not only be feasible to apply fuzziology to TQA study, but will be rather effective as well.