How do Dutch adolescents translate Latin into coherent Dutch? A Journey into the Unknown

Abstract This article discusses translation processes of proficient students who translate Latin fables into Dutch in secondary school. The participants performed two tasks on a computer. They translated a Latin fable and edited a Dutch translation of another Latin fable while their activities were monitored by eye-tracker, screencast and keystroke logging. Immediately after the tasks the participants were invited to view their eye-tracking film and retrace their thoughts at the time of translating (stimulated recall). The article focuses on the stimulated recall interviews, and more specifically on the role of revision in the translation process. It presents a case study of one participant, as well as the strategies participants described to have used in tackling two specific translation problems. Data suggest that proficient participants rely on text comprehension rather than morphological knowledge to solve translation problems. The research shows three key elements as indicators for successful translation process resulting in a coherent target text: (1) a wide variety of problem-solving strategies and the ability to switch strategies, (2) the availability and use of metalanguage to verbalise the chosen strategy, and (3) revision of the target text.