Modelagem do ambiente multilíngue de produção de introduções de artigos acadêmicos da área das Ciências da Saúde em inglês e português brasileiro

• Ano de defesa: 2018
• Autor(a) principal: Francieli Silvéria Oliveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: teoria sistêmico-funcional; escrita acadêmica; modelagem do ambiente multilíngue; comparação linguística; Lingüística aplicada; Funcionalismo (Linguística); Comunicação na ciência; Multilinguismo; Tradução e interpretação
• Link de acesso: http://hdl.handle.net/1843/LETR-AX8JFZ

This thesis draws on Systemic-Functional Theory (HALLIDAY; MATTHIESSEN, 2014) to model the multilingual environment in which introductions to research articles in English and in Brazilian Portuguese are produced within the experiential domain Diabetes Mellitus in Health Science. The study examined the SYSTEMIC-FUNCTIONAL PROFILE of introductions with a view to analyzing systemic options in text productionrelated to REGISTER and GENERIC STRUCTURE (MATTHIESSEN, 2015) and comparing relations of intersection and disjunction between the two languages (FIGUEREDO, 2015). Ten research article introductions were compiled, five of them written in English and the remaining five in Brazilian Portuguese. The criteria for the text compilationwere based on the typology and topology of language within the CONTEXT OF CULTURE (URE, 1989; MATTHIESSEN, TERUYA e WU, 2008) and on selected variables of the CONTEXT OF SITUATION (HALLIDAY, 1978). The texts were segmented into CLAUSES inelectronic spreadsheets and manually annotated with categories pertaining to the LEXICOGRAMMATICAL and SEMANTIC SYSTEMS within the INTERPERSONAL, IDEATIONAL and TEXTUAL METAFUNCTIONS. The annotations were processed within the R computational environment (R CORE TEAM, 2017) in order (i) to identify patterns based on the categories selected for each annotated clause, (ii) to perform a clusteranalysis and (iii) to obtain a dendrogram with the results for the two languages. Results showed LEXICOGRAMMATICAL and SEMANTIC patterns for the introductions in English and Brazilian Portuguese regarding their REGISTER and GENERIC STRUCTURE(MATTHIESSEN, 2015). Differences between the two languages were found at the LEXICOGRAMMATICAL STRATUM in the SYSTEMS of PERSON, COMPLEMENT, ADJUNCT 1, ADJUNCT 2, COMMENT ADJUNCT 1 and COMMENT ADJUNCT 2 in the INTERPERSONALMETAFUNCTION, CIRCUMSTANCE 3 in the IDEATIONAL METAFUNCTION and TEXTUAL THEME 2 and IDEATIONAL THEME in the TEXTUAL METAFUNCTION. The REGISTER in English recruited more interpersonal selections within the multilingual environment analyzed, thus being more oriented to INTERPERSONAL METAFUNCTION. The REGISTER in Brazilian Portuguese recruited more ideational selections, thus being more oriented to IDEATIONAL METAFUNCTION. Results presented the GENERIC STRUCTURE Introduction Research problem Aim of the paper in both languages. However, the section Aim ofthe paper was the only one that yielded similar systemic selections in both languages as shown by custer analysis. The results of this thesis contribute to studies on the modeling of multilingual environments of different language systems (BRAGA, 2013) as well as studies of REGISTER modeling (LIMA, 2013).