Though translation unit is an important notion for translation studies, for years scholars and researchers have focused on questions such as "what is translation unit" and "what can, or should, be translation unit," while insufficient attention has been paid to the affective factors of translators' choices of translation units. The author of this paper divides translation unit into "analysis unit" and "synthesis unit," sorts out five categories of possible affective factors, selects three factors (readability, style, and content) from the "text" category, and tests them in think-aloud protocols experiments. The experiments show that, measured by syntactic levels, length of analysis units increases as readability increases; measured by counting words, length of analysis units decrease when readability increases or when familiarity of either style or content decreases. On the other hand, limited by short-term memory capacity, synthesis units are not significantly affected by the three factors. The experiments also reveal that syntactic structures have affected translators' choices of translation units.
|