Notes and Discussion

Abstract

Since translations can be viewed on the surface level, they are convenient for linguistic examination. Any wrong deviations from the original text are assumed to reveal some lack in language proficiency. As the students translated into their "mother tongue", any mistrans-Brought to you by | Nanyang Technological University Authenticated Download Date | 5/31/15 6:43 AM 126 IRAL, VOL. XXVII/2, MAY 1989 lations would indicate some gap -linguistic, functional or cultural (Catford, 1965) -in their knowledge of the foreign language, English in this case. Indeed, translation is considered a good test of language proficiency for university students (Aarts, 1968, Pickett, 1968. It is also described äs an integrative exercise which can be used to bridge the gaps between Ll and L2 of the students (Kirkwood, 1966, Green, 1970, King, 1973, Rokkan, 1980, Titford, 1983.According to Widdowson, 1979, translation operates on three levels of the reader's understanding: the structural, the semantic and the pragmatic; and "whereas semantic equivalence has to do with the propositional content of the sentences, pragmatic equivalence has to do with the ülocutionary effect of utterances" (Widdowson, 1979, p. 105). A detailed analysis of the types and goals of translation from the viewpoints of both texts and translators is presented in Hartmann, 1980: eh. 5-7.Mistranslations were assumed, then, to reflect some linguistic difficulty for the students. However, the source of this difficulty was not always clear to us. (This problem often troubles many teachers who have to evaluate students' papers.) In this study our purpose was to compare scoring for micro-level mistranslations with macro-level mistranslations, in order to be able to evaluate the students' comprehension of the original text and to compare the reliability of either evaluation method for the language teacher. We constructed a model for mistranslation categories, with two major parts, macro-level and micro-level structures. The macro-level part included Propositional Content, and Communicative Function. The latter included two sub-groups, namely Explicit and Implicit Communicative Functions. The micro-level structures included elements on the level of words, that is to say Vocabulary and Expressions, Parts of Speech/Verb Tense, Reference, Cohesion elements, and Register. In addition we took note of Unacceptable Translations in the Native Language, that is grammatical errors in Arabic.We used the technique of "backtranslation" to arrive at the evaluation of each of the 19 selected units in the students' understanding of the original text. Counts were made of the sum of mistranslations per category (see Table 1).As a separate evaluation, each student's translation was given a grade by three independent judges on a threefold scale: 2 points for correct translation, l point for partially correct translation, and 0 for an entirely wrong translation. Units correctly translated by few students were considered more dif ficult than units correctly translated by many students. As t...