English < > Spanish Translation and Interpreting of Proper Names in the European Union Discourse.

Abstract

The main working assumption of this thesis was that mediated and non-mediated names in EU discourse and their verbal patterns should be fully studied as translation or interpreting shifts, and as translationese and interpretese in comparison to non-mediated language, with the help of suitable language technologies, and preferably in two or more languages, to establish whether they respond to a certain degree of construction in any of their textual coordinates. This implies a theoretical approach based on Construction Grammar (Goldberg 1995). The chosen language directions were English and Spanish, studied bidirectionally (English < > Spanish). The language technologies regarded as suitable were intermodal corpora, Named Entity Recognition (NER), corpus-based pattern extraction, online dictionaries, and institutional terminology management systems (TMS). An intermodal EU corpus called PETIMOD was compiled ad hoc for this research. The most relevant results of the study are the following: (1) Translation and interpreting shift in names can be correlated to language directions, mediation modes, and semantic categories. This multifactorial approach stress-tests the universal state granted in the literature to some translating and interpreting behaviours. According to our numbers, the normalisation of names is a language-dependent feature of English > Spanish translation, whereas transformation and simplification are more prominent in Spanish > English interpreting and strongly linked to chrematonyms. (2) Spanish translated verbal patterns with names tends to conform to Spanish norms and exhibit some degree of explicitation. By contrast, English interpreted patterns exhibit less variation, which points to simplification again. (3)