Bilingual language processing of evidentiality in a Turkish-English translation and judgment task: the effect of translator training

This study aimed to examine how two different factors, formal instruction in translation and location in an L2 country, affect the ways in which bilingual speakers process information when using multiple languages concurrently. The focus was on a linguistic property of evidentiality - the marking of the source of information about past events - as witnessed firsthand or obtained from another source (inference or hearsay). This distinction is grammatically mandatory in Turkish and optional in English. To test the possible impact of this distinction on bilingual output and judgments, a translation production task was administered in which two groups, L1 Turkish/L2 English speakers and Turkish-English translators, translated simple declarative sentences that varied in their evidential meaning (firsthand vs. non-firsthand evidence). A follow-up second task was carried out whereby participants made judgments about the levels of speaker certainty on a Likert scale. The results showed that both participant groups were more accurate in translating firsthand versus non-firsthand information in translations from English-to-Turkish than the reverse. However, the frequency of correct translations and participants' judgments on speaker certainty varied based on how they acquired and used their languages. The Complex Adaptive System Principles (CASP) Model for Bilingualism was used to explain the results.