A imagem do Brasil no jornalismo estrangeiro

AUTOR(ES)
DATA DE PUBLICAÇÃO

2010

RESUMO

This study, part of the GELC Project _ Grupo de Pesquisa em Linguística de Corpus _ aimed at identifying images of Brazil in texts published in US and UK newspapers, that had been translated into Portuguese and printed in Brazilian online news services. An image is a representation of the country or aspects of it. Since there were no methods available to address this issue, another objective of this research was to develop and apply a methodology that enabled the detection of images in written corpora. The methodology draws on concepts from Corpus Linguistics (SINCLAIR, 1987a; BIBER, 1999; BERBER SARDINHA, 2004) and then develops these into original concepts that are specifically related to the issue of (representational) image detection. The notions that the method started off with were lexicogrammatical patterns (SINCLAIR, 1991; PARTINGTON, 1998; HUNSTON e FRANCIS, 2000), Lexical Priming (HOEY, 2005) and Semantic Prosody (SINCLAIR, 1991; LOUW, 1993; STUBBS, 1995; PARTINGTON, 1998). In addition to Corpus Linguistics, other areas were looked at for support such as Sociology and Social Psychology (HALL, 2001; MOSCOVICI, 2007) for developing our working definition of lexicogramatical image, and Anthropology and the work of Brasilianistas about Brazil (BUARQUE DE HOLANDA, 1995; CHAUÍ, 2000; MAXWELL, 2007). Based on these, concepts emerged throughout the development of the methodology which included: key-participant, key-collocate, image index element and imagetic priming. The investigation presented here intends to contribute to a systematic study of lexicogrammatical patterns based on language in use which may reveal the image of Brazil in the foreign press available online. The use of translations is justified as a methodological tool on the grounds that they would ultimately enable the detection of image shifts between the English and Portuguese versions of the reports. The English-Portuguese parallel corpus consisted of 162 texts in each language. Banco de Português v.2, BNC and COCA served as reference corpora. In all, eleven methodological steps were taken in order to achieve the image signaling lexicogrammatical patterns. The results were: (1) the image signaling patterns showed aspects of economy, politics and environment topics; (2) most patterns revealed a new and positive image of Brazil in the foreign source texts written in English; (3) subtle image shifts were found in translated texts

ASSUNTO(S)

linguistica aplicada imagem do brasil padrões lexicogramaticais tradução linguística de corpus image of brazil lexicogrammatical patterns translation corpus linguistics

ACESSO AO ARTIGO

Documentos Relacionados