Uma investigação da tradução de colocações a partir de um corpus paralelo de textos jornalísticos de cunho político
Carregando...
Data
2020-11-12
Autores
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Esta pesquisa buscou verificar as traduções das colocações empregadas no Corpus de Aprendizes de Tradução 2 (CAT 2), por meio de um modelo descritivo de análise estatística de dados. O CAT 2 é uma extensão do projeto CAT 1 e ambos fazem parte do projeto guarda-chuva “Compilação de Materiais Didáticos e Dicionários Especializados de Colocações Baseados em Corpora (COMADEC)”, iniciado em 2012, na Universidade Estadual Paulista “Júlio de Mesquita Filho” (UNESP), Campus de São José do Rio Preto. O estudo se pautou nas teorias da Linguística de Corpus (MCENERY; WILSON, 1996; MCENERY; HARDIE, 2012; SINCLAIR, 2004; TOGNINI-BONELLI, 2001) e dos Estudos Fraseológicos (ORENHA-OTTAIANO, 2004, 2015; MARTELLI, 2007; TAGNIN, 2013), com foco nas colocações que corroboram com reflexões para as questões pedagógicas do léxico e da tradução baseada em corpus. Os participantes da investigação são alunos do curso de Bacharelado em Letras com Habilitação para Tradutor, do Centro Universitário Sagrado Coração (UNISAGRADO), de Bauru. Os partícipes deveriam apresentar níveis de proficiência em língua inglesa entre B2 e C2. Para isso, o pesquisador ofereceu um teste escrito (ALLAN, 2004) a fim de classificá-los entre os referidos níveis. O CAT 2 contém dois subcorpora, sendo um de 3 mil palavras, com seis textos originais em Português brasileiro, composto por seis textos jornalísticos com extensão de 500 a 800 palavras e, outro, com seus respectivos textos traduzidos para o inglês, com 95.221 mil palavras em língua inglesa. Para avaliar as colocações no CAT 2 (subcorpus de TOs e subcorpus de TTs), a metodologia contou com as ferramentas Keywords, Word Sketch e Concordance do gerenciador de corpus Sketch Engine (KILGARRIFF et al., 2014) e o corpus de referência English Web 2013 (SUCHOMEL; POMIKÁLEK, 2012; JAKUBÍČEK et al., 2013) alocado nessa mesma plataforma e que serviu para comprovar se os exemplos de colocações levantadas no CAT 2 para a discussão de dados eram empregadas ou não em língua inglesa, com vistas à apontar ainda possíveis dificuldades tradutórias. Depois de observadas as colocações, propomos modelos de atividades baseados nos problemas tradutórios encontrados por meio da Online English Collocations Platform e, em seguida, outra atividade com base no English Web 2013 e na ferramenta SKELL. A análise de dados ocorreu a partir de seis exemplos de colocações extraídas do CAT 2, com o intuito de descrever as escolhas tradutórias utilizadas pelos aprendizes com relação às unidades fraseológicas selecionadas. Como resultado, o presente estudo comprovou que ao buscarem correspondentes para colocação “fazer a segurança” em inglês, os estudantes empregaram os equivalentes do/make security, sendo que não encontramos ocorrências combinadas entre essas partes de léxico no corpus de referência. Pelo contrário, o English Web 2013 apresentou somente as palavras como colocados de security. Assim, observou-se a viabilidade de incentivar mais trabalhos com base nesse modelo de corpus, uma vez que dificuldades são comuns nesse grupo, muitas vezes, devido à ingenuidade do tradutor aprendiz. Portanto, ficou ainda comprovado que o modelo de exploração linguística empregado pode contribuir para a melhoria da competência fraseológica do aluno de tradução, servindo de indício para a elaboração de atividades e materiais pedagógicos com base em corpus para fins tradutórios.
This research aimed to verify the translations of the collocations used in the Translation Learner Corpus 2 (CAT 2), using a descriptive model of statistical data analysis. CAT 2 is a continuation of the CAT 1 project, and both are part of the project called The Compilation of Corpus-based Didactic Materials and Specialized Dictionaries of Collocations (COMADEC), started in 2012, at São Paulo State University “Júlio de Mesquita Filho” (UNESP), in São José do Rio Preto. The study was based on the theories of Corpus Linguistics (MCENERY; WILSON, 1996; MCENERY; HARDIE, 2012; SINCLAIR, 2004; TOGNINI-BONELLI, 2001) and Phraseological Studies (ORENHA-OTTAIANO, 2004, 2015; MARTELLI, 2007; TAGNIN, 2013), focusing on collocations, which corroborate reflections on the pedagogical issues of the lexicon and corpus-based translation. The research participants are students of B.A. in Translation, from the University Center Sagrado Coração (UNISAGRADO), in Bauru. Participants should have levels of proficiency in English between B2 and C2. For this purpose, the researcher offered a written test (ALLAN, 2004), in order to classify them among those levels. CAT 2 contains two subcorpora, one of them with 3,000 words, with six original texts in Brazilian Portuguese, consisting of six journalistic texts ranging from 500 to 800 words, and the other one with their respective texts translated into English, with 95,221 words in English. To assess CAT 2 collocations (TOs subcorpus and TTS subcorpus), the methodology used the Keywords, Word Sketch and Concordance tools of the Sketch Engine corpus manager (KILGARRIFF et al., 2014), and the corpus of reference English Web 2013 (SUCHOMEL; POMIKÁLEK, 2012; JAKUBÍČEK et al., 2013) allocated on the same platform and served to prove whether the examples of collocations found in CAT 2 for the discussion of data were used or not in English, in order to point out possible translation difficulties. After observing the collocations, we propose activity models based on the translation problems found through the Online English Collocations Platform and then another activity based on the English Web 2013 and the SKELL tool. The data analysis occurred from six examples of collocations extracted from CAT 2 to describe the learners’ translation choices in relation to the selected phraseological units. As a result, the present study proved that when searching for correspondents for “fazer a segurança” in English, students used the equivalents do/make security, and we did not find any combined occurrences between these parts of the lexicon in the reference corpus. On the contrary, English Web 2013 presented only the words as collocated of security. Thus, the feasibility of encouraging more work based on this corpus model was observed since collocational difficulties are common in this group, often due to the translation learner’s naivety. Therefore, it was also proven that the linguistic exploration model employed could contribute to the improvement of the phraseological competence of the translation learner, serving as an indication for the elaboration of educational activities and materials based on corpus for translation purposes.
This research aimed to verify the translations of the collocations used in the Translation Learner Corpus 2 (CAT 2), using a descriptive model of statistical data analysis. CAT 2 is a continuation of the CAT 1 project, and both are part of the project called The Compilation of Corpus-based Didactic Materials and Specialized Dictionaries of Collocations (COMADEC), started in 2012, at São Paulo State University “Júlio de Mesquita Filho” (UNESP), in São José do Rio Preto. The study was based on the theories of Corpus Linguistics (MCENERY; WILSON, 1996; MCENERY; HARDIE, 2012; SINCLAIR, 2004; TOGNINI-BONELLI, 2001) and Phraseological Studies (ORENHA-OTTAIANO, 2004, 2015; MARTELLI, 2007; TAGNIN, 2013), focusing on collocations, which corroborate reflections on the pedagogical issues of the lexicon and corpus-based translation. The research participants are students of B.A. in Translation, from the University Center Sagrado Coração (UNISAGRADO), in Bauru. Participants should have levels of proficiency in English between B2 and C2. For this purpose, the researcher offered a written test (ALLAN, 2004), in order to classify them among those levels. CAT 2 contains two subcorpora, one of them with 3,000 words, with six original texts in Brazilian Portuguese, consisting of six journalistic texts ranging from 500 to 800 words, and the other one with their respective texts translated into English, with 95,221 words in English. To assess CAT 2 collocations (TOs subcorpus and TTS subcorpus), the methodology used the Keywords, Word Sketch and Concordance tools of the Sketch Engine corpus manager (KILGARRIFF et al., 2014), and the corpus of reference English Web 2013 (SUCHOMEL; POMIKÁLEK, 2012; JAKUBÍČEK et al., 2013) allocated on the same platform and served to prove whether the examples of collocations found in CAT 2 for the discussion of data were used or not in English, in order to point out possible translation difficulties. After observing the collocations, we propose activity models based on the translation problems found through the Online English Collocations Platform and then another activity based on the English Web 2013 and the SKELL tool. The data analysis occurred from six examples of collocations extracted from CAT 2 to describe the learners’ translation choices in relation to the selected phraseological units. As a result, the present study proved that when searching for correspondents for “fazer a segurança” in English, students used the equivalents do/make security, and we did not find any combined occurrences between these parts of the lexicon in the reference corpus. On the contrary, English Web 2013 presented only the words as collocated of security. Thus, the feasibility of encouraging more work based on this corpus model was observed since collocational difficulties are common in this group, often due to the translation learner’s naivety. Therefore, it was also proven that the linguistic exploration model employed could contribute to the improvement of the phraseological competence of the translation learner, serving as an indication for the elaboration of educational activities and materials based on corpus for translation purposes.
Descrição
Palavras-chave
Pedagogia do léxico e da tradução, Corpus de aprendizes, Fraseologia, Colocações, Português brasileiro, Língua inglesa, Lexicon and translation pedagogy, Learner corpus, Phraseology, Collocations, Brazilian portuguese, English language