The Effect of Source Text Structure in L2-to-L1 Translation

초록

Research in translation has long suggested that the source text structure exerts a strong influence on the translated output, but previous works have mainly examined the source text in isolation in the absence of a structural alternative. It remains unclear whether the strong effect of the source text persists when another target language structure is available, and the present study investigates the extent to which bilinguals are influenced by the syntactic structure of the source text as opposed to that of a previously encountered translation. Forty-five Korean learners of English participated in two experiments that used a structural priming paradigm to examine whether their syntactic choices in translation were more influenced by the structure of English source sentences or that of the Korean translations they had previously read. In Experiment 1, attributive and predicative adjective structures were placed in systematic conflict between source text and translated text, employing a between-subjects design crossing prime structure with text presentation format (consecutive vs. simultaneous). In Experiment 2, nominalized structures in English were paired with Korean translations using either parallel nominalizations or finite clause structures in a within-subjects design. Across both experiments, participants consistently followed the source text structure in their translations (94.0% and 92.1%, respectively) regardless of text presentation format or second language proficiency. These findings suggest that source text structure operates as a robust processing anchor in translation, with previously encountered translations exerting no detectable influence under the conditions tested.

키워드