Corpus Research on Multiword Discourse Markers for Raising Translation Awareness

Abstract

One of the focuses of the research in linguistics and translation studies is discourse markers ensuring the relations between clause and sentences in the text. The contrastive analysis of multiword discourse markers in English and other languages provides data at the pragmatic level to translators from English into native languages. Awareness of the features and use of discourse markers gives deeper insight into translation in educational processes in university studies. Multiword expressions are important in language generation and processing, contribute to discourse organization and often function as discourse markers. In the present research, we combine the alignment model of the phrase-based statistical machine translation (MT) and manual processing of the data to examine English multiword discourse markers and their translations in Lithuanian and Hebrew. Our research proves that the examined multiword discourse markers have a tendency to remain multiword lexemes in Hebrew translation, but they are one word discourse markers in Lithuanian. Also, there is possible context-based influence guiding the translation to choose a particle or connective integration in Lithuanian or Hebrew translated discourse markers, which envisions future research focusing on the context-based influence. Also, the creation of multilingual resources and making these resources available is necessary for research in multiple languages. The current research has created an extensive multilingual parallel data resource focused on multiword expressions as discourse markers.