Processos verbais na representação da fala em textos ficcionais: um estudo de romances em inglês e suas traduções e retraduções para o português brasileiro

• Ano de defesa: 2018
• Autor(a) principal: Thais Torres Guimaraes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Textos ficcionais; Tradução; Linguística Sistêmico Funcional; Representação da fala; Funcionalismo (Linguística); Tradução e interpretação; Ficção Traduções para o português
• Link de acesso: http://hdl.handle.net/1843/LETR-B37NUN

This thesis, affiliated to the Translations Studies, draws on Systemic-Functional Linguistics (SFL) (HALLIDAY; MATTHIESSEN, 2014) to report on a study of speech representation in fictional texts. It analyzes patterns of VERBAL PROCESSES and VERBAL CLAUSES used in fictional dialogue in original texts in English and their translations and retranslations into Brazilian Portuguese. It also examines Narratological categories (RIMMON-KENAN, 2005) and translation equivalence relations (CARTFORD, 1978; MATTHIESSEN, 2001) based on the SFL theory and on a contextual approach to translation equivalence (HALLIDAY, 1992). This study aims to empirically support Paganos (2017) findings: a) the so-called Bermans (1990) retranslation hypothesis, which considers that retranslations are closer to the originals than first translations in terms of linguistic patterns; b) differences in the reconstruction of VERBAL PROCESS for speech representation in first translations and in retranslations; c) patterns of use of verbs for speech representation related to context and metacontext; d) literary genres impact on VERBAL CLAUSES choices in original and translated texts. The corpus of this research comprises samples of six fictional texts, originals in English and their first translations and retranslations in Brazilian Portuguese, with three texts representative of novels and three of detective novels. The methodology included: extracting ten excerpts of parts with dialogues approximately 300 words from each novel (original, first translation and retranslation); segmenting those excerpts into CLAUSES; identifying the VERBAL CLAUSES; coping, pasting and aligning these CLAUSES in electronic spreadsheets with tabs for each text; annotating the CLAUSES with categories pertaining to the SFL and Narratology; and then extracting data using the software and programming environment R (R Core Team, 2018). This research yielded results showing that not all texts are in accordance the so-called retranslation hypothesis, then it cannot be confirmed for this corpus (a). The results pointed to metacontextual differences in first translations and retranslations: the later selected the general member type of verb, VERBAL PROCESSES realizing PROJECTION and the semantic function proposition more frequently than the former (b). The different contexts of the two linguistic systems, English and Brazilian Portuguese, were showed by the predominance of the verb say rather than other verbs that represent speech, and the predominance of the general member type of verb and of CIRCUMSTANCES following that type of verb in English, when contrasted with Brazilian Portuguese (c). Regarding literary genre, the results were not conclusive as there is no clear linguistic separation between them text with the same label shows different VERBAL PROCESS patterns (d). Therefore, this research shows similar tendencies Pagano (2017) found and contributes showing that linguistic system components may elucidate context and metacontext attributes of translated texts.