Descrição
A corpus-based study involves working with the available linguistic material. In this paper, I address the tendency of translated texts to have some distinctive features (Baker, 2007), such as simplification and normalisation, which may be found in both standard translations and in self-translations. The focus in this work was on normalisation in the self-translated historical novel An Invincible Memory (Viva o Povo Brasileiro) by João Ubaldo Ribeiro (1984, 1989) to describe lexical and collocational normalisation. The novel covers 400 years of Brazilian history, numerous common people characters, and entails a non-literate use of language. The objectives were to provide a quantitative analysis of normalisation in self-translation, to indicate how self-translation operates when the direction of translation is from a less frequently spoken source language, such as Portuguese, to a more frequently spoken target language, such as English and, finally, to explore how lexical and collocational normalisation could be represented statistically in this self-translation. The methodology was in #LancsBox and consisted of the compilation of the texts in the form of corpora, the generation of wordlists including the number of items for a statistical analysis, a type/token ratio (TTR) to identify the first signs of normalisation, and the Key Word in Context (KWIC) tool to determine the lengths of the sentences in both texts. Finally, concordance graphics (GraphColls) were generated to summarise the results. The findings provided evidence of the standardisation of the TTR values (0.111 and 0.077, respectively) in the source text and in the self-translation. In addition, the article the (raw frequency 13,498) was in the top 10 of the most frequent words, thus denoting the overuse of target language patterns to ensure easy readability via a lower lexical range. The study aimed to provide some insights into a more pragmatic approach to João Ubaldo Ribeiro’s work.
