image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 1 ESTUDO EXPERIMENTAL DA INFLUÊNCIA DOS SISTEMAS DE MEMÓRIA DE TRADUÇÃO NO SUCESSO DE TRADUÇÕES ESCRITAS POR TRADUTORES NOVATOS EXPERIMENTAL STUDY OF THE INFLUENCE OF TRANSLATION MEMORY SYSTEMS ON THE SUCCESS OF WRITTEN TRANSLATIONS BY NOVICE TRANSLATORS ESTUDIO EXPERIMENTAL DE LA INFLUENCIA DE LOS SISTEMAS DE MEMORIA DE TRADUCCIÓN EN EL ÉXITO DE LAS TRADUCCIONES ESCRITAS DE TRADUCTORES PRINCIPIANTES Elizaveta TRIFONOVA1Ekaterina BABINTSEVA2Elena KARTSEVA3RESUMO:O objetivo da pesquisa é realizar um experimento para estudar o impacto dos sistemas de memória de tradução (em termos de uso de bases terminológicas) na eficácia de futuros tradutores. O artigo fundamenta a relevância do uso de sistemas de memória de tradução, apresenta várias áreas de pesquisa sobre automação de tradução, considera as possibilidades e perspectivas para o desenvolvimento de sistemas CAT. O artigo apresenta os resultados de um estudo experimental que mostraria o quanto tradutores iniciantes dependem de sistemas de memória de tradução e se eles possuem o vocabulário no nível adequado, o que é típico de textos sociopolíticos. Os resultados obtidos confirmaram a hipótese de que ao traduzir o texto original por meio de uma ferramenta CAT usando uma base terminológica com unidades lexicais traduzidas erroneamente, a qualidade da tradução de tradutores novatos diminuirá e aumentará sem usar uma base terminológica com unidades lexicais traduzidas.PALAVRAS-CHAVE:Ferramenta CAT. Futuros tradutores. Bases de terminologia. Formação de tradutores. Pesquisa experimental.RESUMEN:El propósito de la investigación es realizar un experimento para estudiar el impacto de los sistemas de memoria de traducción en la efectividad de los futuros traductores. El artículo fundamenta la relevancia del uso de los sistemas de memoria de traducción, presenta varias áreas de investigación sobre la automatización de la traducción. El artículo presenta los resultados de un estudio experimental que mostraría cuánto confían los traductores novatos en los sistemas de memoria de traducción y si tienen el vocabulario al nivel adecuado, que es típico de los textos sociopolíticos. Los resultados obtenidos confirmaron la hipótesis de que al traducir el texto original por medio de una herramienta TAO utilizando 1Universidade da Amizade dos Povos da Rússia (RUDN), Moscou – Rússia. Professor Senior. ORCID: https://orcid.org/0000-0002-0636-8426. E-mail: trifonova_ed@pfur.ru 2Universidade da Amizade dos Povos da Rússia (RUDN), Moscou – Rússia. PhD em História. ORCID: https://orcid.org/0000-0002-1197-0892. E-mail: babintseva-ea@rudn.ru 3Universidade da Amizade dos Povos da Rússia (RUDN), Moscou – Rússia. PhD em História. ORCID: https://orcid.org/0000-0003-2595-3775. E-mail: kartseva-eyu@udn.ru
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 2 una base terminológica con unidades léxicas traducidas erróneamente, la calidad de la traducción de los traductores novatos disminuirá, y aumentará sin utilizar una base terminológica con unidades léxicas mal traducidas. unidades léxicas traducidas. PALABRAS CLAVE:Herramienta CAT. Futuros traductores. Bases terminológicas. Formación de traductores. Investigación experimental. ABSTRACT: The purpose of the research is to conduct an experiment to study the impact of translation memory systems (in terms of the use of terminology bases) on the effectiveness of future translators. The article substantiates the relevance of the use of translation memory systems, presents various areas of research on translation automation, considers the possibilities and prospects for the development of CAT systems. Article presents the results of an experimental study that would show how much novice translators rely on translation memory systems and whether they have the vocabulary at the proper level, which is typical for socio-political texts. The obtained results confirmed the hypothesis that when translating the original text by means of a CAT tool using a terminological base with erroneously translated lexical units, the quality of the translation of novice translators will decrease, and it will increase without using a terminological base with erroneously translated lexical units. KEYWORDS: CAT-tool. Future translators. Terminology bases. Translator training. Experimental research. Introdução A relevância do estudo deve-se à crescente importância das TIC no mercado moderno de serviços de tradução. As ideias tradicionais sobre a tradução como atividade profissional estão mudando, e o modelo profissional da atividade profissional de um tradutor torna-se mais complexo com elas. O progresso científico e tecnológico certamente afeta as atividades de tradução. Cada vez mais serviços de tradução são fornecidos e oferecidos via Internet. Clientes e tradutores não podem se reunir pessoalmente para coordenar um pedido. Essa forma de traduzir é conveniente para freelancers, pois eles podem oferecer serviços e procurar novos clientes (DOHERTY, 2016). As tecnologias mais recentes geralmente determinam os tipos de textos que precisam ser traduzidos. Além da demanda por tradução de textos em indústrias tradicionais, por exemplo, econômicas ou técnicas, novas áreas estão constantemente surgindo nas quais o volume de traduções está aumentando (BUNDGAARD; CHRISTENSEN; SCHJOLDAGER, 2016). Um tradutor moderno deixa de ser um tradutor universal, domina uma determinada indústria ou várias indústrias nas quais se especializa. Deve constantemente repor seu conhecimento no campo escolhido – estudar não apenas a nova terminologia, mas também o
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 3 software moderno utilizado no processo de tradução, que é constantemente atualizado e aprimorado (PYM, 2013). O tradutor deve ser competente em muitos aspectos (POPOVA; BOBROVA; MALTZAGOV, 2021), e a capacidade de trabalhar com software especializado é extremamente importante na realização de atividades de tradução (KUROCHKINA; MAKAZIEVA; MARKOVA, 2021). Um lugar especial entre as TIC na atividade profissional de um tradutor é ocupado por ferramentas CAT, que simplificam e aceleram muito o processo de tradução (KARPIŃSKA, 2017).Revisão de literatura Procedemos com o fato de que existem (KRÜGER, 2016; ZHANG; YANG, 2012) duas áreas de pesquisa sobre automação de tradução: 1) aplicada (implementação industrial da automação de tradução de texto); e 2) teórico (modelagem da atividade da fala humana como um dos métodos de sua pesquisa; desenvolvimento de formalismos matemáticos para descrições linguísticas; busca de algoritmos para processamento de objetos linguísticos; pesquisa da relação entre o pensamento humano e as máquinas, etc.). Referindo-se às peculiaridades do uso de programas de CAT no trabalho de um tradutor, pesquisadores (PYM, 2011) consideraram as vantagens que um tradutor recebe ao buscar ajuda de tais ferramentas. Em primeiro lugar, as habilidades e competências para o uso de programas CAT já se tornaram um requisito obrigatório para a participação de um tradutor em grandes projetos de tradução (ABU DAYYEH, 2020). Apesar dos sucessos no desenvolvimento de programas de tradução (OLOHAN, 2011), a tradução totalmente automatizada de textos com temas extensos ainda é uma tarefa impossível, e a tradução automática de textos com foco estreito e vocabulário limitado é realizada com participação humana obrigatória (WANG, 2020). O artigo (BOWKER, 2015) também observa que hoje os resultados da tradução automatizada ainda estão longe de ser perfeitos, e a correção profissional de erros no texto traduzido é absolutamente necessária. O tradutor passa muito tempo trabalhando na tradução de forma independente e executando-a manualmente (SHCHEBUNYAEVA et al., 2021). Enquanto isso, trabalhar no mesmo campo da tradução (por exemplo, tradução de textos médicos ou documentos legais) implica uma certa generalidade de termos e expressões estáveis que muitas vezes são encontradas em textos (KOPYTINA et al., 2021). Trabalhando manualmente, o tradutor deve
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 4 gastar tempo retraindo os elementos do texto que os sistemas CAT são capazes de otimizar, reduzindo significativamente o tempo necessário para tradução (STARLANDER, 2013). Foi determinado (BAAR, 2012; LIU, 2017) que os sistemas CAT adquirirão um desenvolvimento ainda maior no futuro, receberão mais, e poderão ser usados com ainda maior eficiência. O estudo (AL-RUMAIH, 2021) observa que os futuros tradutores devem poder utilizar sistemas de memória de tradução, para os quais são introduzidos cursos especiais em programas de educação profissional que introduzem os alunos às ferramentas necessárias. Segundo pesquisadores (MARCHENKOVA; KOLOBKOVA, 2020), é aconselhável oferecer um curso sobre o domínio de tecnologias de tradução moderna para estudantes seniores, uma vez que eles já devem ter desenvolvido habilidades de tradução e habilidades para trabalhar com o sistema CAT. É aconselhável melhorar essas habilidades e competências e a formação de informações e competência computacional através da implementação de exercícios, a principal tarefa deve ser traduzir textos autênticos usando programas CAT (GRIDASOVA et al., 2021). Formatos desatualizados de trabalho com texto impresso estão sendo substituídos por uma versão mais moderna da tradução em formato eletrônico (YAO, 2017). Os sistemas CAT não atuam como um fim em si mesmos de treinamento e atividade profissional, mas como um meio de implementar esses tipos de atividades em um nível de alta qualidade (SIN-WAI, 2010). As tecnologias de tradução no curso de dominá-las pelos alunos devem ser posicionadas como um dos muitos, embora muito importantes, componentes de competência profissional. No entanto, a capacidade de usar ferramentas CAT não é suficiente para realizar uma tradução de alta qualidade. O tradutor deve conhecer o vocabulário e os termos específicos para uma determinada indústria, o que melhorará significativamente a qualidade do resultado final. A edição de um texto traduzido por meio de um programa CAT ficará muito mais complicada na ausência do conhecimento necessário, uma vez que o tradutor ficará um pouco desorientado (MAFULAH et al., 2018). Nesse sentido, consideramos necessário realizar um experimento que mostre o quanto os tradutores novatos dependem de sistemas de memória de tradução e se eles têm o vocabulário típico dos textos sociopolíticos no nível adequado. De acordo com pesquisas científicas (JUAN; YAHAYA, 2019), os alunos em tradução focam principalmente na estrutura superficial do texto original, enquanto tradutores profissionais, antes de mais nada, analisam o significado da mensagem original e tentam recriar a essência da mensagem original.
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 5 É por isso que o objetivo do estudo é realizar um experimento para estudar o impacto dos sistemas de memória de tradução (em termos de uso de bases terminológicas) sobre a eficácia de futuros tradutores. Objetivos de pesquisa - Selecionar o texto original e criar uma base de terminologia contendo erros; - Organizar a tradução do texto original e dividir os participantes do experimento para que um grupo (experimental) utilize a base de terminologia proposta, e o outro (controle) não; - Determinar os princípios pelos quais serão analisados os textos de tradução e analisar os textos de tradução com base nesses princípios; - Realizar o processamento quantitativo dos resultados da pesquisa obtida e analisar e interpretar os resultados obtidos; - Formular conclusões e perspectivas de novas pesquisas com base nos resultados obtidos. Hipótese da pesquisa: a qualidade da tradução dos tradutores iniciantes diminuirá ao traduzir o texto original por meio de uma ferramenta CAT utilizando uma base terminológica com unidades lexicais erroneamente traduzidas, e aumentará sem essa mesma base. O artigo consiste em uma introdução, uma revisão de literatura, métodos, resultados, discussão e conclusão. Métodos Um estudo experimental sobre a influência dos sistemas de memória de tradução (em termos de uso de bases terminológicas) sobre a qualidade das traduções por tradutores novatos foi realizado na Universidade RUDN em 2021. No âmbito de três etapas: preparatória, durante a qual foi formulada uma hipótese e um plano de experiência; o principal, que consistia na implementação prática do experimento; e o final, que previa a análise e interpretação dos dados experimentais. Para testar as hipóteses, foram selecionados 26 alunos do 1º ano do programa de mestrado do Departamento de Tradução da Faculdade de Línguas Estrangeiras, divididos em dois grupos experimentais (EG-1 – 12 pessoas e EG-2 – 14 pessoas), incluindo seis meninos e 20 meninas de 21 a 22 anos.
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 6 O estudo experimental foi realizado enquanto ensinava o básico de trabalhar com ferramentas CAT no âmbito da disciplina "Prática de Tradução". Os alunos se familiarizam com o básico de trabalhar em sistemas de memória de tradução: eles traduzem textos do inglês para o russo e vice-versa, usando um módulo de tradução automática, bases de memória de tradução e bases de terminologia. O material de pesquisa foi de 26 textos de tradução de artigos sobre temas sociopolíticos, traduzidos utilizando ferramentas CAT por tradutores iniciantes – estudantes-membros de grupos experimentais. Um trecho de um artigo em inglês sobre temas sociopolíticos da revista online britânica "The Independent" foi emprestado como o texto original. A escolha da fonte primária deve-se ao fato de os alunos conhecerem as peculiaridades da tradução de textos sociopolíticos no âmbito das aulas de prática de tradução durante seus estudos na universidade. Traduzem notícias políticas e econômicas, ampliam seus conhecimentos sobre a estrutura política de diferentes países e adquirem conhecimentos de fundo necessários para a tradução. O volume do texto é de 21 frases, 473 palavras, 3.288 caracteres impressos (com espaços). O tempo para completar a tradução é de 80 minutos. Os alunos usaram o sistema de memória de tradução baseado em nuvem intitulado Memsource durante a tradução, bem como um banco de dados de terminologia compilado especificamente para o experimento, que continha erros na tradução de 28 unidades léxicas. Todos os alunos traduziram o texto proposto para o russo. Os alunos do primeiro grupo experimental (EG-1) utilizaram a base de terminologia com erros durante a tradução, enquanto os alunos do segundo grupo experimental (EG-2) eram autoconfiantes. Ao verificar as traduções feitas pelos alunos, contamos com um sistema de avaliação que nos permitiu levar em conta as características de nossa pesquisa: 1) Erros devidos aos quais o conteúdo do texto original é significativamente distorcido e/ou a opção de tradução errada é selecionada, o que é oferecido no banco de dados de terminologia (1 ponto de penalidade é concedido); 2) Erros decorrentes dos quais o conteúdo do texto original é parcialmente perdido e/ou a tradução proposta no banco de dados é parcialmente selecionada (0,5 pontos de penalidade são concedidos); 3) Erros que não afetam significativamente ou quase não afetam o conteúdo do texto original (são concedidos 0,1 pontos de penalidade). A confiabilidade das diferenças nos resultados dos alunos de diferentes grupos experimentais foi avaliada pelo teste t para amostras não relacionadas.
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 7 Resultados Os resultados da seção experimental do EG-1 são mostrados na Tabela 1. Tabela 1– Os resultados do corte experimental realizado pelo EG-1 utilizando a base de terminologia com erros (em pontos de penalidade) EstudanteErros tipo 1Erros tipo 2Erros tipo 3Número total de pontos de penalidade 18.01.50.19.629.00.50.29.7312.01.50.315.846.00.00.26.255.01.00.26.266.00.50.26.774.02.00.26.287.01.50.18.699.01.00.310.3108.01.50.19.6115.00.50.25.7127.00.50.17.6Valor médio7.171.00.178.52Fonte: Preparado pelos autores Da tabela 1, o valor médio do ponto de penalidade é de 8,52 no EG-1, enquanto o valor médio dos erros do primeiro tipo, que incluem uma distorção significativa do conteúdo ou a escolha de uma tradução de uma determinada palavra ou frase proposta pela base terminologia, é de 7,17. A taxa de erro do primeiro tipo varia de 4,0 a 12,0 pontos de penalidade. A pontuação alta para erros do primeiro tipo no EG-1 é explicada pelo fato de que os alunos foram guiados principalmente pela base e não pelo seu conhecimento e contexto. A pontuação média de penalidade por erros do segundo tipo é de 1,0, o que pode ser explicado pelo fato de que as unidades léxicas, o conteúdo que foi ligeiramente alterado, eram geralmente familiares aos alunos. A maioria dos erros do segundo tipo foram cometidos pelo aluno 7 (a pontuação da penalidade é 2.0), o que pode ser explicado pelo fato de algumas unidades léxicas serem desconhecidas para ele, ao contrário de outros assuntos. A pontuação média de penalidade para erros do terceiro tipo, que inclui aqueles que não afetam o conteúdo da mensagem é de 0,17 no EG-1 (com o indicador mais alto sendo 0,3 pontos de penalidade).
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 8 A Tabela 2 apresenta os resultados de uma seção experimental transversal dos alunos do EG-2, que não utilizaram a base terminologia com erros. Tabela 2– Os resultados de um corte experimental realizado pelo EG-2 sem o uso de uma base de terminologia com erros (em pontos de penalidade) EstudanteErros tipo 1Erros tipo 2Erros tipo 3Número total de pontos de penalidade 13.02.00.05.024.01.00.15.135.02.00.17.144.02.00.26.252.01.50.13.664.00.50.34.877.00.50.17.682.02.00.14.194.00.50.34.8107.01.50.18.61110.01.00.111.1126.02.50.28.7135.01.50.16.6142.02.50.24.7Valor médio4.641.80.26.29Fonte: Preparado pelos autores De acordo com os dados apresentados na Tabela 2, o escore total da penalidade é de 6,29, inferior ao mesmo indicador do EG-1. Os indicadores gerais para o grupo variam de 3,6 a 11,1 pontos de penalidade. Os alunos do EG-2 também cometeram erros do primeiro tipo, mas houve menos deles do que erros do segundo e terceiro tipos. A taxa média de erro do primeiro tipo é de 4,64, que é significativamente menor que o mesmo indicador no EG-1. Um dos alunos tem a maior pontuação de penalidade – 10, o que provavelmente indica que as unidades léxicas eram desconhecidas para ele e, portanto, causaram certas dificuldades na tradução. Três alunos cometeram apenas dois erros do primeiro tipo, que é o melhor resultado do grupo. Curiosamente, o índice de erros do segundo tipo no EG-2 é quase duas vezes maior que os resultados no EG-1: é de 1,8, enquanto o número máximo de pontos de penalidade para erros desse tipo é de 2,5.
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 9 Os resultados da análise estatística da confiabilidade das diferenças utilizando o teste t para o EG-1 e EG-2 confirmaram resultados maiores no EG-2 ao cometer erros do primeiro tipo (tEmp= 2,9; p≤0,01)Após uma análise detalhada das traduções dos dois grupos experimentais, concluímos que o EG-1, que teve a oportunidade de usar a base de terminologia com erros, cometeu um maior número de erros do primeiro tipo, enquanto os do segundo e terceiro tipos não foram significativos em um número relativo aos do primeiro tipo. O EG-2, que não tinha base terminológica, cometeu mais erros do segundo tipo, o que pode ser explicado pela escolha de sinônimos estilísticos que não tiveram sucesso para o contexto proposto. Discussão Os resultados do EG-1 (uso da base terminologia com erros), apresentados na Tabela 1, nos levaram a concluir que é mais provável que os alunos que usaram uma base de termo mal escrito ao traduzir se baseassem na opção sugerida nessa base em vez de analisar criticamente os recursos fornecidos, analisando cuidadosamente o significado da mensagem original, e tentando reproduzir sua essência. Por sua vez, após analisar os resultados do EG-2 (não utilizar a base de terminologia com erros) apresentada na Tabela 2, estávamos convencidos de que esse grupo lidava melhor com a tradução. Os alunos se basearam em seus conhecimentos, criticaram o texto de tradução, analisaram seu profundo significado, verificaram e editaram cuidadosamente o texto de tradução e, portanto, receberam um resultado melhor. Isso é consistente com os resultados da pesquisa (MARCHENKOVA; KOLOBKOVA, 2020; POPOVA; BOBROVA; MALTZAGOV, 2021), segundo o qual os alunos, ao começarem a traduzir textos de vários assuntos em programas de CAT, muitas vezes confiam excessivamente na máquina e não prestam atenção a alguns erros cometidos durante a tradução. Para excluir tais situações, os pesquisadores propõem melhorar as habilidades e competências de tradução e a formação de informações e competências computacionais ao utilizar sistemas de tradução automática, realizando um sistema de exercícios, a principal tarefa deve ser traduzir textos autênticos usando programas de tradução automatizadas. Os sistemas de tradução automática não atuam como um fim em si mesmos de treinamento e atividade profissional, mas como um meio de implementar esses tipos de atividades em um nível de alta qualidade (AFANASIEV et al., 2021). De acordo com os pesquisadores (DOLZHENKOV et al., 2021; IVANOVA; LONTSKAYA; PAKHOMOVA,
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 10 2021), e estamos totalmente de acordo com isso, as tecnologias de tradução no curso de dominá-las pelos alunos devem ser posicionadas como um dos muitos, embora muito importantes, componentes de competência profissional. Acreditamos que, a partir de diversos estudos, o desenvolvimento de um sistema de exercícios para ensino de alunos de tradução deve ser baseado na estrutura psicológica da tradução, que, sendo um tipo de atividade, é formada por diversas operações (habilidades) e ações (habilidades) (BAAR, 2012). Tal estrutura de atividade de tradução, segundo pesquisadores (MAFULAH et al., 2018; STARLANDER, 2013), torna necessário incluir a formação de competência como a posse de ferramentas de automação de tradução no conteúdo do treinamento, o que inclui: - A capacidade de usar a ferramenta de forma eficiente e rápida, integrar diversos programas para ajudar na tradução, edição, processamento de terminologia, layout, pesquisa em documentos (por exemplo, ferramentas de processamento de texto, verificações ortográficas e gramatical, bancos de dados de tradução, bancos de dados de terminologia, programas de reconhecimento de voz); - A capacidade de criar e gerenciar bancos de dados; - A capacidade de adaptar e dominar novas ferramentas, especialmente na tradução de material multimídia e audiovisual; - A capacidade de preparar e realizar a tradução em diversos formatos e para diversos ambientes técnicos; - Conhecimento das possibilidades e limitações da máquina e da tradução automatizada. Conclusão O objetivo do uso de sistemas CAT é otimizar o trabalho do tradutor, reduzir o tempo necessário para a tradução, coordenar o uso de um glossário terminológico durante um projeto longo e evitar a realização de nova tradução de material semelhante ao trabalhar com textos técnicos e oficiais de estilo de negócios (ou seja, tais textos compõem uma camada significativa do trabalho total dos tradutores). Contudo, o experimento provou que, ao usar a base de terminologia, tradutores novatos tendem a confiar muito e ativamente nela sem realizar uma análise crítica dos recursos oferecidos. Isso, por sua vez, indica que os sistemas de memória de tradução podem prejudicar as atividades profissionais dos tradutores iniciantes e, portanto, uma organização especial de treinamento em tecnologias de tradução moderna é importante, o que levaria em conta os
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 11 resultados obtidos e visando o desenvolvimento da atitude crítica dos alunos aos softwares mencionados e aos recursos fornecidos. No entanto, deve-se levar em conta que, devido ao pequeno número de disciplinas neste experimento, as conclusões não podem ser definitivas e, portanto, devem ser testadas em uma amostra de 150 alunos no futuro. AGRADECIMENTOS: Este artigo foi apoiado pelo Programa de Liderança Acadêmica Estratégica da Universidade RUDN.REFERÊNCIAS ABU DAYYEH, I. Use and evaluation of computer-aided translation tools (CAT) on the word level from the perspective of Palestinian translators and translation trainees. Arab World English Journal For Translation and Literary Studies, v. 4, n. 1, p. 111–130, 2020. Disponível em: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3554133. Acesso em: 3 fev. 2021. AFANASIEV, I. V. et al. The use of cloud resources and services in distance learning of students in the context of restrictions caused by the pandemic. Revista Tempos e Espaços em Educação, v. 14, n. 33, p. e16103, 2021. Disponível em: https://dialnet.unirioja.es/servlet/articulo?codigo=8077569. Acesso em: 30 mar. 2021. AL-RUMAIH, L. A. The integration of computer-aided translation tools in translator-training programs in Saudi Universities: Toward a more visible state. Arab World English Journal for Translation & Literary Studies, v. 5, n. 1, p. 336-362, 2021. Disponível em: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3802984. Acesso em: 12 fev. 2021. BAAR, D. Embedding technology in translation teaching: Evaluative considerations for courseware integration. Computer Assisted Language Learning, v. 1, n. 16, p. 1-14, 2012. Disponível em: https://www.tandfonline.com/doi/abs/10.1080/09588221.2012.658406. Acesso em: 30 mar. 2021. BOWKER, L. General issues of translation technology. In: SINWAI, C. (ed.). The Routledge encylopedia of translation studies. New York: The Routledge, 2015. BUNDGAARD, K.; CHRISTENSEN, T. P.; SCHJOLDAGER, A. Translator-computer interaction in action — An observational process study of computer-aided translation. The Journal of Specialised Translation, v. 25, p. 106-130, 2016. Disponível em: https://vbn.aau.dk/ws/files/261592776/Bundgaard_Christensen_and_Schjoldager_2016_Translator_computer_interaction_in_action_an_observational_process_study_of_computer_aided_translation.pdf. Acesso em: 23 mar. 2021. DOHERTY, S. The impact of translation technologies on the process and product of translation. International Journal of Communication, v. 10, p. 947–969, 2016. Disponível em: https://ijoc.org/index.php/ijoc/article/view/3499/1573. Acesso em: 18 fev. 2021.
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 12 DOLZHENKOV, V. N. et al. Aplicação de podcasting no ensino de habilidades profissionais de língua estrangeira para alunos [Application of podcasting in teaching spoken professionalforeign language skills to students]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16025. Acesso em: 7 mar. 2021. GRIDASOVA, A.V. et al. Formação de habilidades "suaves" no espaço educacional da universidade ao ensinar línguas estrangeiras [Formation of "soft" skills in the educational university space while teaching foreign languages]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, e021101, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15659. Acesso em: 28 fev. 2021. IVANOVA, A. S.; LONTSKAYA, M. A.; PAKHOMOVA, I. N. Ensinando leitura e escuta em russo como língua estrangeira (A1-B1) aulas usando tecnologia digital [Teaching reading and listening in Russian as a foreign language (a1-b1) classes using digital technology]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, p. e021080, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15625. Acesso em: 18 fev. 2021. JUAN, L.; YAHAYA, N. B. Research on application of computer aided translation to translation teaching. International Journal of Academic Research in Progressive Education and Development, v. 8, n. 4, p. 795–804, 2019. Disponível em: https://hrmars.com/papers_submitted/6722/Research_on_Application_of_Computer_Aided_Translation_to_Translation_Teaching.pdf. Acesso em: 20 fev. 2021. KARPIŃSKA, P. Computer aided translation – Possibilities, limitations, and changes in the field of professional translation. Journal of Education Culture and Society, v. 2, p. 133-142, 2017. Disponível em: https://www.ceeol.com/search/article-detail?id=570930. Acesso em: 30 mar. 2021. KOPYTINA, N. N. et al. Lingvoculturology: A importância das informações de base cultural no estudo da semântica das unidades de linguagem [Lingvoculturology: The importance of cultural backgroundinformation in the study of language units’ semantics]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16031. Acesso em: 12 fev. 2021. KRÜGER, R. Contextualising computer-assisted translation tools and modelling their usability. Trans-Kom - Journal of Translation and Technical Communication Research, v. 9, n. 1, p. 114-148, 2016. Disponível em: http://www.trans-kom.eu/bd09nr01/trans-kom_09_01_08_Krueger_CAT.20160705.pdf. Acesso em: 27 mar. 2021. KUROCHKINA, E.; MAKAZIEVA, Z. D.; MARKOVA, T. S. O desenvolvimento da competência comunicativa intercultural de futuros tradutores que estudam alemão após o inglês [The development of intercultural communicative competence of future translators studying German after English]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, e021077, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15622. Acesso em: 9 mar. 2021. LIU, J. Research on course design of computer aided-translation from the perspective of language service technology. Journal of Nanchang Normal University, v. 1, p. 67-70, 2017.
image/svg+xmlEstudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatosRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 13 Disponível em: https://www.cnki.com.cn/Article/CJFDTotal-JXJY201701016.htm. Acesso em: 15 fev. 2021. MAFULAH, S. et al.CAT tool on teaching translation for EFL students. International Journal of Engineering & Technology, v. 7, n. 3-5, p. 98-100, 2018. Disponível em: https://www.sciencepubco.com/index.php/ijet/article/view/21660/10338. Acesso em: 23 fev. 2021. MARCHENKOVA, T.; KOLOBKOVA, A. Design and implementation of a CAT training course. In:RUGELJ, J.; LAPINA, M. (eds.). SLET-2020: International Scientific Conference on Innovative Approaches to the Application of Digital Technologies in Education. Aachen: Ceur Workshop Proceedings, 2020. OLOHAN, M. Translators and translation technology: The dance of agency. Translation Studies, v. 4, n. 3, p. 345-357, 2011. Disponível em: https://www.tandfonline.com/doi/abs/10.1080/14781700.2011.589656. Acesso em: 19 fev. 2021. POPOVA, E. N.; BOBROVA, S. E.; MALTZAGOV, I. D. Desenvolvimento da competência e cultura da utilização de ferramentas de tradução automática em futuros tradutores (a exemplo de dominar o uso de ferramentas gatos) [Development of the competency and culture of usingautomatic translation tools in future translators (on theexample of mastering the use of cat tools)]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Disponível em: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16024. Acesso em: 5 mar. 2021. PYM, A. What technology does to translating. The International Journal of Translation & Interpreting Research, v. 3, n. 1, p. 1–9, 2011. Disponível em: https://search.informit.org/doi/abs/10.3316/INFORMIT.409249876138991. Acesso em: 5 mar. 2021. PYM, A. Translation skill-sets in a machine-translation age. Meta, v. 58, n. 3, p. 487–503, 2013. Disponível em: https://www.erudit.org/en/journals/meta/1900-v1-n1-meta01406/1025047ar/abstract/. Acesso em: 25 mar. 2021. SHCHEBUNYAEVA, N. G. et al. Innovative approaches to teaching foreign languages at the university. Propósitos y Representaciones, v. 9, n. S2, e1107, 2021. Disponível em: https://files.eric.ed.gov/fulltext/EJ1299347.pdf. Acesso em:14 fev. 2021.SIN-WAI, C. A new curriculum for the teaching of translation technology: The teaching of a translation project course as a case study. Journal of Translation Studies, v. 13, n. 1-2, p. 83–154, 2010. Disponível em: https://www.airitilibrary.com/Publication/alDetailedMesh?docid=P20180306001-201012-201803060012-201803060012-83-154. Acesso em: 28 fev. 2021. STARLANDER, M. Training translation students to evaluate CAT tools using eagles: A case study. Translating and the Computer, v. 35, p. 2-10, 2013. Disponível em: https://archive-ouverte.unige.ch/unige:35622. Acesso em: 16 mar. 2021. WANG, X. A study on project-based computer aided translation courses from perspective of language services. International Core Journal of Engineering, v. 6, n. 7, p. 261–266, 2020.
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA e Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 14 Disponível em: http://www.icj-e.org/download/ICJE-6-7-261-266.pdf. Acesso em: 20 mar. 2021. YAO, S. Application of computer-aided translation in English teaching. International Journal of Emerging Technologies in Learning, v. 12, n. 8, p. 105–117, 2017. Disponível em: https://www.learntechlib.org/p/182485/. Acesso em: 18 mar. 2021. ZHANG, Z.; YANG, Y. Exploration and application of MTI CAT practical ability. Shanghai Journal of Translators, v. 2, p. 44-47, 2012. Como referenciar este artigo TRIFONOVA, E.; BABINTSEVA, E.; KARSEVA, E. Estudo experimental da influência dos sistemas de memória de tradução no sucesso de traduções escritas por tradutores novatos. Rev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, mar. 2022. e-ISSN: 2447-3529. DOI: https://doi.org/10.29051/el.v8iesp.1.16939 Submetido em: 28/12/2021 Revisões requeridas em: 19/01/2022 Aprovado em: 25/02/2022 Publicado em: 30/03/2022
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 1 EXPERIMENTAL STUDY OF THE INFLUENCE OF TRANSLATION MEMORY SYSTEMS ON THE SUCCESS OF WRITTEN TRANSLATIONS BY NOVICE TRANSLATORS ESTUDO EXPERIMENTAL DA INFLUÊNCIA DOS SISTEMAS DE MEMÓRIA DE TRADUÇÃO NO SUCESSO DE TRADUÇÕES ESCRITAS POR TRADUTORES NOVATOS ESTUDIO EXPERIMENTAL DE LA INFLUENCIA DE LOS SISTEMAS DE MEMORIA DE TRADUCCIÓN EN EL ÉXITO DE LAS TRADUCCIONES ESCRITAS DE TRADUCTORES PRINCIPIANTES Elizaveta TRIFONOVA1Ekaterina BABINTSEVA2Elena KARTSEVA3ABSTRACT: The purpose of the research is to conduct an experiment to study the impact of translation memory systems (in terms of the use of terminology bases) on the effectiveness of future translators. The article substantiates the relevance of the use of translation memory systems, presents various areas of research on translation automation, considers the possibilities and prospects for the development of CAT systems. Article presents the results of an experimental study that would show how much novice translators rely on translation memory systems and whether they have the vocabulary at the proper level, which is typical for socio-political texts. The obtained results confirmed the hypothesis that when translating the original text by means of a CAT tool using a terminological base with erroneously translated lexical units, the quality of the translation of novice translators will decrease, and it will increase without using a terminological base with erroneously translated lexical units. KEYWORDS: CAT-tool. Future translators. Terminology bases. Translator training. Experimental research. RESUMO:O objetivo da pesquisa é realizar um experimento para estudar o impacto dos sistemas de memória de tradução (em termos de uso de bases terminológicas) na eficácia de futuros tradutores. O artigo fundamenta a relevância do uso de sistemas de memória de tradução, apresenta várias áreas de pesquisa sobre automação de tradução, considera as possibilidades e perspectivas para o desenvolvimento de sistemas CAT. O artigo apresenta os resultados de um estudo experimental que mostraria o quanto tradutores iniciantes dependem de sistemas de memória de tradução e se eles possuem o vocabulário no nível adequado, o que é típico de textos sociopolíticos. Os resultados obtidos confirmaram a hipótese de que ao 1Рeoples’ Friendship University of Russia (RUDN), Moscow – Russia. Senior Lecturer. ORCID: https://orcid.org/0000-0002-0636-8426. E-mail: trifonova_ed@pfur.ru 2Рeoples’ Friendship University of Russia (RUDN), Moscow – Russia. PhD in History. ORCID: https://orcid.org/0000-0002-1197-0892. E-mail: babintseva-ea@rudn.ru 3Рeoples’ Friendship Universityof Russia (RUDN), Moscow – Russia. PhD in History. ORCID: https://orcid.org/0000-0003-2595-3775. E-mail: kartseva-eyu@udn.ru
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 2 traduzir o texto original por meio de uma ferramenta CAT usando uma base terminológica com unidades lexicais traduzidas erroneamente, a qualidade da tradução de tradutores novatos diminuirá e aumentará sem usar uma base terminológica com unidades lexicais traduzidas.PALAVRAS-CHAVE:Ferramenta CAT. Futuros tradutores. Bases de terminologia. Formação de tradutores. Pesquisa experimental.RESUMEN:El propósito de la investigación es realizar un experimento para estudiar el impacto de los sistemas de memoria de traducción en la efectividad de los futuros traductores. El artículo fundamenta la relevancia del uso de los sistemas de memoria de traducción, presenta varias áreas de investigación sobre la automatización de la traducción. El artículo presenta los resultados de un estudio experimental que mostraría cuánto confían los traductores novatos en los sistemas de memoria de traducción y si tienen el vocabulario al nivel adecuado, que es típico de los textos sociopolíticos. Los resultados obtenidos confirmaron la hipótesis de que al traducir el texto original por medio de una herramienta TAO utilizando una base terminológica con unidades léxicas traducidas erróneamente, la calidad de la traducción de los traductores novatos disminuirá, y aumentará sin utilizar una base terminológica con unidades léxicas mal traducidas. unidades léxicas traducidas. PALABRAS CLAVE:Herramienta CAT. Futuros traductores. Bases terminológicas. Formación de traductores. Investigación experimental. Introduction The relevance of the study is due to the increasing importance of ICT in the modern market of translation services. Traditional ideas about translation as a professional activity are changing, and the professional model of a translator's professional activity becomes more complex with them. Scientific and technological progress certainly affects translation activities. More and more translation services are provided and offered via the Internet. Customers and translators may not meet in person to coordinate an order. This way of translating is convenient for freelancers since they can offer services and look for new clients (DOHERTY, 2016). The latest technologies often determine the types of texts that need to be translated. In addition to the demand for translation of texts in traditional industries, for example, economic or technical ones, new areas are constantly emerging in which the volume of translations is increasing (BUNDGAARD; CHRISTENSEN; SCHJOLDAGER, 2016). A modern translator ceases to be a universal translator, masters a certain industry or several industries in which he/she specializes. He/she must constantly replenish his/her knowledge in the chosen field –
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 3 to study not only new terminology but also modern software used in the translation process, which is constantly updated and improved (PYM, 2013). The translator must be competent in many aspects (POPOVA; BOBROVA; MALTZAGOV, 2021), and the ability to work with specialized software is extremely important when carrying out translation activities (KUROCHKINA; MAKAZIEVA; MARKOVA, 2021). A special place among ICTs in the professional activity of a translator is occupied by CAT tools, which greatly simplify and accelerate the translation process (KARPIŃSKA, 2017). Literature review We proceed from the fact that there are (KRÜGER, 2016; ZHANG; YANG, 2012) two areas of research on translation automation: 1) applied (industrial implementation of text translation automation); and 2) theoretical (modeling of human speech activity as one of the methods of its research; development of mathematical formalisms for linguistic descriptions; search for algorithms for processing language objects; research of the relationship between human thinking and machines, etc.). Referring to the peculiarities of using CAT programs in the work of a translator, researchers (PYM, 2011) considered the advantages that a translator receives by seeking help from such tools. First of all, the skills and abilities to use CAT programs have already become a mandatory requirement for the participation of a translator in large translation projects (ABU DAYYEH, 2020). Despite the successes in the development of translation programs (OLOHAN, 2011), fully automated translation of texts with extensive topics is still an impossible task, and machine translation of texts with a narrow focus and limited vocabulary is performed with mandatory human participation (WANG, 2020). The paper (BOWKER, 2015) also notes that today the results of automated translation are still far from perfect, and professional correction of errors in the translated text is absolutely necessary. The translator spends a lot of time working on the translation independently and performing it manually (SHCHEBUNYAEVAet al., 2021). Meanwhile, work in the same field of translation (for example, translation of medical texts or legal documents) implies a certain generality of terms and stable expressions that are often found in texts (KOPYTINA et al., 2021). Working manually, the translator must spend time retranslating those elements of
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 4 the text that CAT systems are able to optimize, significantly reducing the time required for translation (STARLANDER, 2013). It was determined (BAAR, 2012; LIU, 2017) that CAT systems will acquire even greater development in the future, will receive more, and can be used with even greater efficiency. The study (AL-RUMAIH, 2021) notes that future translators should be able to use translation memory systems, for which special courses are introduced in vocational education programs that introduce students to the necessary tools. According to researchers (MARCHENKOVA; KOLOBKOVA, 2020), it is advisable to offer a course on mastering modern translation technologies to senior students, since they must already have developed translation skills and abilities to work with the CAT system. It is advisable to improve these skills and abilities and the formation of information and computer competence through the implementation of exercises, the main task of which should be to translate authentic texts using CAT programs (GRIDASOVAet al., 2021). Outdated formats of working with printed text are being replaced by a more modern version of the translation in electronic format (YAO, 2017). CAT systems do not act as an end in themselves of training and professional activity but as a means of implementing these types of activities at a high-quality level (SIN-WAI, 2010). Translation technologies in the course of mastering them by students should be positioned as one of many, albeit very important, components of professional competence. However, the ability to use CAT tools is not enough to perform a high-quality translation. The translator must know the vocabulary and terms specific to a particular industry, which will significantly improve the quality of the final result. Editing a text translated using a CAT program will become much more complicated in the absence of the necessary knowledge since the translator will be somewhat disoriented (MAFULAHet al., 2018). In this regard, we consider it necessary to conduct an experiment that would show how much novice translators rely on translation memory systems and whether they have the vocabulary that is typical of socio-political texts at the proper level. According to scientific research (JUAN; YAHAYA, 2019), students in translation mainly focus on the surface structure of the original text, while professional translators, first of all, analyze the meaning of the original message and try to recreate the essence of the original message.
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 5 That is why the purpose of the study is to conduct an experiment to study the impact of translation memory systems (in terms of the use of terminology bases) on the effectiveness of future translators. Research objectives - To select the original text and create a terminology base containing errors; - To organize the translation of the original text and divide the participants of the experiment so that one group (experimental) uses the proposed terminology base, and the other (control) does not; - To determine the principles by which the translation texts will be analyzed, and analyze the translation texts based on these principles; - To carry out quantitative processing of the obtained research results and to analyze and interpret the results obtained; - To formulate conclusions and prospects for further research based on the results obtained. Research hypothesis: the quality of the translation of novice translators will decrease when translating the original text by means of a CAT tool using a terminological base with erroneously translated lexical units, and it will increase without using a terminological base with erroneously translated lexical units. The article consists of an introduction, a literature review, methods, results, discussion, and conclusion. Methods An experimental study on the influence of translation memory systems (in terms of the use of terminology bases) on the quality of translations by novice translators was conducted at the RUDN University in 2021. Within the framework of three stages: preparatory, during which a hypothesis was formulated and an experiment plan was developed; the main one, which consisted in the practical implementation of the experiment; and the final one, which provided for the analysis and interpretation of experimental data. To test the hypotheses, we selected 26 1st-year students of the Master's degree program of the Translation Department of the Faculty of Foreign Languages, divided into two
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 6 experimental groups (EG-1 – 12 people and EG-2 – 14 people), including six boys and 20 girls aged 21-22 years. The experimental study was conducted while teaching the basics of working with CAT tools in the framework of the "Translation Practice" discipline. Students get acquainted with the basics of working in translation memory systems: they translate texts from English into Russian and vice versa, using a machine translation module, translation memory bases, and terminology bases. The research material was 26 translation texts of articles on socio-political topics, translated using CAT tools by novice translators – student-members of experimental groups. An excerpt of an article in English on socio-political topics from "The Independent" British online magazine was borrowed as the original text. The choice of the primary source is due to the fact that students get acquainted with the peculiarities of the translation of socio-political texts within the framework of translation practice classes during their studies at the university. They translate political and economic news, expand their knowledge of the political structure of different countries and acquire background knowledge necessary for translation. The volume of the text is 21 sentences, 473 words, 3,288 printed characters (with spaces). The time to complete the translation is 80 minutes. Students used the Memsource cloud-based translation memory system during the translation, as well as a terminology database compiled specifically for the experiment, which contained errors in the translation of 28 lexical units. All students have fully translated the proposed text into Russian. The students of the first experimental group (EG-1) used the terminology base with errors during the translation, while the students of the second experimental group (EG-2) were self-reliant. When checking the translations made by students, we relied on an assessment system that allowed us to take into account the features of our research: 1) Errors due to which the content of the original text is significantly distorted and/or the wrong translation option is selected, which is offered in the terminology database (1 penalty point is awarded); 2) Errors due to which the content of the original text is partially lost and/or the translation proposed in the database is partially selected (0.5 penalty points are awarded); 3) Errors that do not significantly or almost do not affect the content of the original text (0.1 penalty points are awarded). The reliability of the differences in the results of students from different experimental groups was assessed by Student's t-test for unrelated samples.
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 7 Results The results of the experimental section of the EG-1 are shown in Table 1. Table 1The results of the experimental cut performed by the EG-1 using the terminology base with errors (in penalty points) StudentType 1 errorsType 2 errorsType 3 errorsTotal number of penalty points 18.01.50.19.629.00.50.29.7312.01.50.315.846.00.00.26.255.01.00.26.266.00.50.26.774.02.00.26.287.01.50.18.699.01.00.310.3108.01.50.19.6115.00.50.25.7127.00.50.17.6Average value7.171.00.178.52Source: Prepared by the authors As follows from Table 1, the average value of the penalty point is 8.52 in EG-1, while the average value of errors of the first type, which include a significant distortion of the content or the choice of a translation of a certain word or phrase proposed by the terminology base, is 7.17. The error rate of the first type varies from 4.0 to 12.0 penalty points. The high score for errors of the first type in EG-1 is explained by the fact that students were mainly guided by the base rather than by their knowledge and context. The average penalty score for errors of the second type is 1.0, which can be explained by the fact that the lexical units, the content of which was slightly changed, were generally familiar to students. Most of the mistakes of the second type were made by student 7 (the penalty score is 2.0), which can be explained by the fact that some lexical units were unknown to him, unlike other subjects. The average penalty score for errors of the third type, which includes those that do not affect the content of the message is 0.17 in EG-1 (with the highest indicator being 0.3 penalty points).
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 8 Table 2 presents the results of an experimental cross-section of students of the EG-2, which did not use the terminology base with errors. Table 2The results of an experimental cut performed by the EG-2 without the use of a terminology base with errors (in penalty points) StudentType 1 errorsType 2errorsType 3 errorsTotal number of penalty points 13.02.00.05.024.01.00.15.135.02.00.17.144.02.00.26.252.01.50.13.664.00.50.34.877.00.50.17.682.02.00.14.194.00.50.34.8107.01.50.18.61110.01.00.111.1126.02.50.28.7135.01.50.16.6142.02.50.24.7Average value4.641.80.26.29Source: Prepared by the authors According to the data presented in Table 2, the total penalty score is 6.29, which is less than the same indicator in the EG-1. The overall indicators for the group range from 3.6 to 11.1 penalty points. Students in EG-2 also made mistakes of the first type, but there were fewer of them than errors of the second and third types. The average error rate of the first type is 4.64, which is significantly less than the same indicator in EG-1. One of the students has the highest penalty score – 10, which most likely indicates that the lexical units were unknown to him/her, and therefore caused certain difficulties in translation. Three students made only two mistakes of the first type, which is the best result in the group. Interestingly, the index of errors of the second type in EG-2 is almost twice as high as the results in EG-1: it is 1.8, while the maximum number of penalty points for errors of this type is 2.5.
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 9 The results of statistical analysis of the reliability of differences using Student's t-test for the EG-1 and EG-2 confirmed higher results in the EG-2 when making errors of the first type (tEmp= 2.9; p≤0.01) After a detailed analysis of the translations of the two experimental groups, we conclude that EG-1, which had the opportunity to use the terminology base with errors, made a greater number of errors of the first type, while errors of the second and third types were not significant in a number concerning errors of the first type. EG-2, which did not have a terminological base, made more mistakes of the second type, which can be explained by the choice of stylistic synonyms that were unsuccessful for the proposed context. Discussion The results of EG-1 (use of the terminology base with errors), presented in Table 1 led us to conclude that it is more likely that students who used a misspelled termbase when translating relied on the option suggested in that termbase instead of critically analyzing the resources provided, thoughtfully analyzing the meaning of the original message, and trying to reproduce its essence. In turn, after analyzing the results of the EG-2 (did not use the terminology base with errors) presented in Table 2, we were convinced that this group coped better with the translation. Students relied on their knowledge, were critical of the translation text, analyzed its deep meaning, carefully checked and edited the translation text, and therefore received a better result. This is consistent with the research results (MARCHENKOVA; KOLOBKOVA, 2020; POPOVA; BOBROVA; MALTZAGOV, 2021), according to which students, when starting to translate texts of various subjects in CAT programs, often overly trust the machine and do not pay attention to some mistakes made during translation. To exclude such situations, the researchers propose to improve translation skills and abilities and the formation of information and computer competence when using machine translation systems by performing a system of exercises, the main task of which should be to translate authentic texts using automated translation programs. Machine translation systems do not act as an end in themselves of training and professional activity but as a means of implementing these types of activities at a high-quality level (AFANASIEV et al., 2021). According to researchers (DOLZHENKOV et al., 2021; IVANOVA; LONTSKAYA; PAKHOMOVA, 2021), and we fully agree with this, translation
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 10 technologies in the course of mastering them by students should be positioned as one of many, although very important, components of professional competence. We believe that based on several studies, the development of a system of exercises for teaching translation students should be based on the psychological structure of translation, which, being a type of activity, is formed by several operations (abilities) and actions (skills) (BAAR, 2012). Such a structure of translation activity, according to researchers (MAFULAH et al., 2018; STARLANDER, 2013), makes it necessary to include the formation of such competence as the possession of translation automation tools in the content of training, which includes: - The ability to use the tool efficiently and quickly, integrate several programs to help with translation, editing, terminology processing, layout, search in documents (for example, text processing tools, spelling and grammar checks, translation databases, terminology databases, voice recognition programs); - The ability to create and manage databases; - The ability to adapt and master new tools, especially when translating multimedia and audiovisual material; - The ability to prepare and carry out the translation in various formats and for various technical environments; - Knowledge of the possibilities and limitations of the machine and automated translation. Conclusion The purpose of using CAT systems is to optimize the translator's work, reduce the time required for translation, coordinate the use of a terminological glossary during a long project and avoid re-translating similar material when working with technical and official-business style texts (namely, such texts make up a significant layer of the total work of translators). However, the experiment proved that when using the terminology base, novice translators tend to trust it too much and actively rely on it without conducting a critical analysis of the resources offered to them. This, in turn, indicates that translation memory systems can hurt the professional activities of novice translators, and therefore a special organization of training in modern translation technologies is important, which would take
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 11 into account the results obtained and aimed at developing students' critical attitude to the mentioned software and the resources provided. However, it should be taken into account that due to the small number of subjects in this experiment, the conclusions cannot be final, and therefore they should be tested on a sample of 150 students in the future. ACKNOWLEDGMENTS: This paper has been supported by the RUDN University Strategic Academic Leadership Program.REFERENCES ABU DAYYEH, I. Use and evaluation of computer-aided translation tools (CAT) on the word level from the perspective of Palestinian translators and translation trainees. Arab World English Journal For Translation and Literary Studies, v. 4, n. 1, p. 111–130, 2020. Available in: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3554133. Access in: 3 Feb. 2021. AFANASIEV, I. V. et al. The use of cloud resources and services in distance learning of students in the context of restrictions caused by the pandemic. Revista Tempos e Espaços em Educação, v. 14, n. 33, p. e16103, 2021. Available in: https://dialnet.unirioja.es/servlet/articulo?codigo=8077569. Access in: 30 Mar. 2021. AL-RUMAIH, L. A. The integration of computer-aided translation tools in translator-training programs in Saudi Universities:Toward a more visible state. Arab World English Journal for Translation & Literary Studies, v. 5, n. 1, p. 336-362, 2021. Available in: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3802984. Access in: 12 Feb. 2021. BAAR, D. Embedding technology in translation teaching: Evaluative considerations for courseware integration. Computer Assisted Language Learning, v. 1, n. 16, p. 1-14, 2012. Available in: https://www.tandfonline.com/doi/abs/10.1080/09588221.2012.658406. Access in: 30 Mar. 2021. BOWKER, L. General issues of translation technology. In: SINWAI, C. (ed.). The Routledge encylopedia of translation studies. New York: The Routledge, 2015. BUNDGAARD, K.; CHRISTENSEN, T. P.; SCHJOLDAGER, A. Translator-computer interaction in action — An observational process study of computer-aided translation. The Journal of Specialised Translation, v. 25, p. 106-130, 2016. Available in: https://vbn.aau.dk/ws/files/261592776/Bundgaard_Christensen_and_Schjoldager_2016_Translator_computer_interaction_in_action_an_observational_process_study_of_computer_aided_translation.pdf. Access in: 23 Mar. 2021. DOHERTY, S. The impact of translation technologies on the process and product of translation. International Journal of Communication, v. 10, p. 947–969, 2016. Available in: https://ijoc.org/index.php/ijoc/article/view/3499/1573. Access in: 18 Feb. 2021.
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 12 DOLZHENKOV, V. N. et al. Aplicação de podcasting no ensino de habilidades profissionais de língua estrangeira para alunos [Application of podcasting in teaching spoken professionalforeign language skills to students]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16025. Access in: 7 Mar. 2021. GRIDASOVA, A.V. et al. Formação de habilidades "suaves" no espaço educacional da universidade ao ensinar línguas estrangeiras [Formation of "soft" skills in the educational university space while teaching foreign languages]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, e021101, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15659. Access in: 28 Feb. 2021. IVANOVA, A. S.; LONTSKAYA, M. A.; PAKHOMOVA, I. N. Ensinando leitura e escuta em russo como língua estrangeira (A1-B1) aulas usando tecnologia digital [Teaching reading and listening in Russian as a foreign language (a1-b1) classes using digital technology]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, p. e021080, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15625. Access in: 18 Feb. 2021. JUAN, L.; YAHAYA, N. B. Research on application of computer aided translation to translation teaching. International Journal of Academic Research in Progressive Education and Development, v. 8, n. 4, p. 795–804, 2019. Available in: https://hrmars.com/papers_submitted/6722/Research_on_Application_of_Computer_Aided_Translation_to_Translation_Teaching.pdf. Access in: 20 Feb. 2021. KARPIŃSKA, P. Computer aided translation – Possibilities, limitations, and changes in the field of professional translation. Journal of Education Culture and Society, v. 2, p. 133-142, 2017. Available in: https://www.ceeol.com/search/article-detail?id=570930. Access in: 30 Mar. 2021. KOPYTINA, N. N. et al. Lingvoculturology: A importância das informações de base cultural no estudo da semântica das unidades de linguagem [Lingvoculturology: The importance of cultural backgroundinformation in the study of language units’ semantics]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16031. Access in: 12 Fev. 2021. KRÜGER, R. Contextualising computer-assisted translation tools and modelling their usability. Trans-Kom - Journal of Translation and Technical Communication Research, v. 9, n. 1, p. 114-148, 2016. Available in: http://www.trans-kom.eu/bd09nr01/trans-kom_09_01_08_Krueger_CAT.20160705.pdf. Access in: 27 Mar. 2021. KUROCHKINA, E.; MAKAZIEVA, Z. D.; MARKOVA, T. S. O desenvolvimento da competência comunicativa intercultural de futuros tradutores que estudam alemão após o inglês [The development of intercultural communicative competence of future translators studying German after English]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 4, e021077, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/15622. Access in: 9 Mar. 2021. LIU, J. Research on course design of computer aided-translation from the perspective of language service technology. Journal of Nanchang Normal University, v. 1, p. 67-70, 2017.
image/svg+xmlExperimental study of the influence of translation memory systems on the success of written translations by novice translatorsRev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 13 Available in: https://www.cnki.com.cn/Article/CJFDTotal-JXJY201701016.htm. Access in: 15 Feb. 2021. MAFULAH, S. et al.CAT tool on teaching translation for EFL students. International Journal of Engineering & Technology, v. 7, n. 3-5, p. 98-100, 2018. Available in: https://www.sciencepubco.com/index.php/ijet/article/view/21660/10338. Access in: 23 Feb. 2021. MARCHENKOVA, T.; KOLOBKOVA, A. Design and implementation of a CAT training course. In:RUGELJ, J.; LAPINA, M. (eds.). SLET-2020: International Scientific Conference on Innovative Approaches to the Application of Digital Technologies in Education. Aachen: Ceur Workshop Proceedings, 2020. OLOHAN, M. Translators and translation technology: The dance of agency. Translation Studies, v. 4, n. 3, p. 345-357, 2011. Available in: https://www.tandfonline.com/doi/abs/10.1080/14781700.2011.589656. Access in: 19 Feb. 2021. POPOVA, E. N.; BOBROVA, S. E.; MALTZAGOV, I. D. Desenvolvimento da competência e cultura da utilização de ferramentas de tradução automática em futuros tradutores (a exemplo de dominar o uso de ferramentas gatos) [Development of the competency and culture of usingautomatic translation tools in future translators (on theexample of mastering the use of cat tools)]. Revista EntreLinguas, Araraquara, v. 7, n. esp. 5, 2021. Available in: https://periodicos.fclar.unesp.br/entrelinguas/article/view/16024. Access in: 5 Mar. 2021. PYM, A. What technology does to translating. The International Journal of Translation & Interpreting Research, v. 3, n. 1, p. 1–9, 2011. Available in: https://search.informit.org/doi/abs/10.3316/INFORMIT.409249876138991. Access in: 5 Mar. 2021. PYM, A. Translation skill-sets in a machine-translation age. Meta, v. 58, n. 3, p. 487–503, 2013. Available in: https://www.erudit.org/en/journals/meta/1900-v1-n1-meta01406/1025047ar/abstract/. Access in: 25 Mar. 2021. SHCHEBUNYAEVA, N. G. et al. Innovative approaches to teaching foreign languages at the university. Propósitos y Representaciones, v. 9, n. S2, e1107, 2021. Available in: https://files.eric.ed.gov/fulltext/EJ1299347.pdf. Access in:14 Feb. 2021.SIN-WAI, C. A new curriculum for the teaching of translation technology: The teaching of a translation project course as a case study. Journal of Translation Studies, v. 13, n. 1-2, p. 83–154, 2010. Available in: https://www.airitilibrary.com/Publication/alDetailedMesh?docid=P20180306001-201012-201803060012-201803060012-83-154. Access in: 28 Feb. 2021. STARLANDER, M. Training translation students to evaluate CAT tools using eagles: A case study. Translating and the Computer, v. 35, p. 2-10, 2013. Available in: https://archive-ouverte.unige.ch/unige:35622. Access in: 16 Mar. 2021.
image/svg+xmlElizaveta TRIFONOVA; Ekaterina BABINTSEVA and Elena KARSEVARev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v8iesp.1.16939 14 WANG, X. A study on project-based computer aided translation courses from perspective of language services. International Core Journal of Engineering, v. 6, n. 7, p. 261–266, 2020. Available in: http://www.icj-e.org/download/ICJE-6-7-261-266.pdf. Access in: 20 mar. 2021. YAO, S. Application of computer-aided translation in English teaching. International Journal of Emerging Technologies in Learning, v. 12, n. 8, p. 105–117, 2017. Available in: https://www.learntechlib.org/p/182485/. Access in: 18 mar. 2021. ZHANG, Z.; YANG, Y. Exploration and application of MTI CAT practical ability. Shanghai Journal of Translators, v. 2, p. 44-47, 2012. How to reference this articleTRIFONOVA, E.; BABINTSEVA, E.; KARSEVA, E. Experimental study of the influence of translation memory systems on the success of written translations by novice translators. Rev. EntreLínguas, Araraquara, v. 8, n. esp. 1, e022026, Mar. 2022. e-ISSN: 2447-3529. DOI: https://doi.org/10.29051/el.v8iesp.1.16939 Submitted: 28/12/2021 Required revisions: 19/01/2022 Approved: 25/02/2022 Published: 30/03/2022