Education, Language and Sociology Research ISSN 2690-3644 (Print) ISSN 2690-3652 (Online) Vol. 4, No. 3, 2023 www.scholink.org/ojs/index.php/elsr 132 Original Paper Research on Cognitive Translation Process: An Overview Xu Yuemeng1 & Liu Yongjie2 1 Institute of Modern History, Chinese Academy of Social Sciences—Beijing, China 2 Research Institute for Global Value Chains, University of International Business and Economics— Beijing, China Received: July 10, 2023 Accepted: July 31, 2023 Online Published: August 19, 2023 doi:10.22158/elsr.v4n3p132 URL: http://dx.doi.org/10.22158/elsr.v4n3p132 Abstract Translation process is conceived as a cognitive activity and has been received less attention in early research due to the difficulty of lacking direct access to translators’ thinking process. But researchers still made attempts to describe this process. And hence this article reviews the previous research on cognitive translation process, in particular the research model. The article concludes that a descriptive approach to the investigation of the translation process should be adopted rather than a prescriptive approach. Keywords cognitive translation process, research models, descriptive approaches 1. Introduction Traditional research focuses on two domains of translation practice: translation as a product and translator’s competence. The research on translation as a product has been largely conducted from linguistic perspective by comparison with source language texts, for which a considerable amount of different and highly theoretical models has been formulated with equivalence being their primary concern and neglecting other text features such as cultural elements. As stated by Peter Newmark (1988): The source language text consists of words, that is all that is there on the page. Finally, all you have is words to translate and you have to account for each of them somewhere in your target language text. (p. 193) Though Newmark (1988) suggests to ‘translate words that are more or less linguistically, referentially, culturally and subjectively influenced in meaning’ (p. 193) rather than isolated words, he is in an attempt to achieve equivalence at lexical level. As for Eugene Nida (1964), he was in pursuit of ‘dynamic equivalence’ (p. 159) through his extensive translation practice of Bible. He argued that reader’s response to the translated text should be taken into www.scholink.org/ojs/index.php/elsr Education, Language and Sociology Research Vol. 4, No. 3, 2023 133 Published by SCHOLINK INC. consideration of the translation process, in that Nida realized that people from different cultures might respond differently to the same text. What Nida tried to achieve was that the translated texts should appeal to and produce some kind of emotional response on the part of the reader as that of the reader of the source language and culture to the source texts. The approach adopted by Nida, however, was still product-oriented though having taken cultural elements into consideration. Nida’s approach is not the only case; in fact, most of product-oriented models, if not all, are more prescriptive rather than descriptive. Such models draw heavily attention on linguistic elements in translation, neglecting the subjectivity of translators in the translation processes. The product-oriented study focuses on the analysis of the translated texts, which indeed reflects the translators’ thinking activities to some extent, but far from enough because the processes are much more complicated than the translated texts could reveal. Therefore, other approaches should be employed to explore the translators’ thinking processes. Apart from the product, translation competence is another focus of translation research, as translation is regularly considered as a professional practice requiring specific competence, i.e., a translator must possess certain expertise to fulfill the translation task and reach a successful outcome. Therefore, as a consequence of translation theory being product- and competence-oriented, the research on translation process has hardly drawn any attention before the 1980s. In other words, the Translation Studies have been only focusing on one aspect of the translation proper, because the concept of translation entails at least two aspects: the products of translating and the processes of translating. Therefore, previous study on translation can be said to be incomplete. Such a situation was not changed until 1980 when Ericsson and Simon (1980) first introduced into Translation Studies the empirical approach of verbal reports, which was borrowed from the discipline of cognitive psychology. The verbal report is a data collection method, with which researchers can finally probe into the translation processes of translators that was considered to be inaccessible to direct observations which is due the very nature of human brain, the mysterious ‘black box’. Though the method is not without flaws—the method is challenged by many researchers who consider the data collected from verbal reports are subjective and incomplete, and hence are invalid, however, the verbal reports in fact can provide enough accurate information for analyzing and describing the translation processes. Hence, this part of the paper intends to give an overview of previous studies on translation processes. 2. An Overview of Previous Studies on Cognitive Translation Processes Tapping into the cognitive translation processes has been a difficult task to accomplish, as it is constrained by such intrinsic difficulties as any kind of cognitive processing being not amendable to direction observation. Therefore, researchers have put forward models to investigate and analyze the translation process. These models explain the cognitive processes from theoretical point of view in terms of analyzing the products, i.e., the translated texts instead of actual translation processes, therefore less convincing. In order to better understand cognitive translation processes, researchers turn from theoretical studies to empirical studies, which up till now have proved to be an effective approach. This www.scholink.org/ojs/index.php/elsr Education, Language and Sociology Research Vol. 4, No. 3, 2023 134 Published by SCHOLINK INC. part is in an attempt to discuss the status quo of research on cognitive translation processes. To begin with, a detailed introduction to theoretical studies on translation process, including two models proposed by Seleskovitch and Lederer (1968; 1975; 1981) and Bell (1991), was outlined, pointing out the achievements they have made. The models put forward by such scholars, however, are purely theoretical and speculative, lacking empirical and concrete evidence to support themselves, and therefore less persuasive, which led to the introduction of empirical approaches to the study of translation process. Pioneering research on cognitive translation process was conducted by Seleskovitch and Lederer, employing the interpretive theory of translation (ITT). The ITT divided the translation/interpreting process into three interrelated phases: understanding, deserialization and re-expression. According to ITT, understanding, as the beginning of translation process, requires both linguistic and contextual/cultural knowledge, for linguistic knowledge itself doesn’t suffice to decode the message contained in source texts. The translator or interpreter needs to comprehend the ‘sense’ in its totality of source texts that is resulted from the interdependence of all linguistic and non-linguistic elements of the understanding process. The ‘sense’ translator or interpreter acquires during the understanding is the non-verbal synthesis, which requires the intermediate phase of deserialization. For ITT, the deserialization is an essential part, as the re-expression is achieved through deverbalized meaning rather than on the basis of linguistic form. Re-expression, as the reverse process of understanding, is similar to the process of natural language producing, in which the speaker expresses the end product of non-observable thinking activities through linguistic forms. For ITT, the translator or interpreter performs the role of ‘non-observable thinking activities’ in the context of translation through understanding and deserialization, as a speaker does in the process of natural language producing, and then formulates linguistic expressions of the intended meaning of the sender of the source texts through re-expression. At the end of the whole process, the end product, i.e., the translated texts, is produced and will be delivered to readers of the target language. Though ITT outlines a sketch for translation processes, it is a theoretical model based on linguistic theory, lacking of hard evidence. Bell (1991) argues that the study of translation is best served by the construction of models of the process of translating, for which he proposed an integrated model of translation process by drawing on linguistic and psychological knowledge. By challenging the notion that laid the ground work for early attempts to formulate a theory of translation, Bell argues that the process, during which a translation is produced, and the translator, who performs the act of translating, should be included in the description of a translation theory. Since the translation process is carried out by translators, knowledge and skills of translators, translator competence, should be dealt with first before the investigation of the translation process, because it is the question that needs to be answered—‘What is it that translators need to know and be able to do in order to translate?’ According to Bell, a translator should be familiar with the knowledge shared by people of both target and source languages, and that knowledge should at least contain five aspects: target language (TL) knowledge; text-type knowledge; source language (SL) www.scholink.org/ojs/index.php/elsr Education, Language and Sociology Research Vol. 4, No. 3, 2023 135 Published by SCHOLINK INC. knowledge; subject area (‘real world’) knowledge; and contrastive knowledge (Johnson & Whitelock, 1987, p. 137). Lack of knowledge means that the translator would be unable to perform the translation. The concept of information processing is central to Bell’s model, for Bell divides translation process into analysis and synthesis, which require both short-term and long-term memory of the translator to collaborate in decoding the text in source language and encoding the text into the target language. The analysis/synthesis is achieved through operations at syntactic, semantic and pragmatic levels, which co- occur with the five stages during the translation process—parsing, expression, development, ideation and planning. Decoding process begins with text processing which is preferable by concentrating on the clause, and proceeds in both a bottom-up and a top-down manner. As is suggested by Bell, a visual recognition system is required to distinguish words from other linguistic forms in the source text, and prepares the source text, through identifying distinctive features of letters, for syntactic processing of the clause. At this stage, the syntactic structure of the clause will be analyzed to acquire the lexical meaning, and when the syntactic analyzer fails to pass through the frequent lexis store and the frequent structure store, it will recourse to the parser for analyzing difficult clause structure, or the lexical search mechanism for tackling the linguistic items that it is not familiar with. When the syntactic analyzer finishes decoding the source text, the process can proceed to the next stage, the semantic analysis. The aim of the semantic analysis is to acquire the meaning of the clause through each syntactic component, and to retrieve ‘transitivity’ relations that underlie the syntactic structure of the clause. During the semantic analysis, the semantic analyzer tries to figure out ‘what the clause it about; what it represents; logical relationships between participants and processes; ideational meaning; semantic sense; and propositional content’ (Bell, 1991, p. 53). Once the logical form of the clause is identified, the task is handed over to the final phase—pragmatic analysis. In pragmatic analysis, the pragmatic processor has two tasks to fulfill—they are isolating the thematic structure and providing a register analysis. The thematic structure is concerned with the distribution of information and whether it is in a marked or unmarked order. The register analysis of the clause is performed in terms of tenor of discourse, mode of discourse and domain of discourse. Tenor and mode provide style information while domain provides purpose, suggesting a speech act. The information from thematic structure and register analysis form a semantic representation supported by the idea organizer and the planner. Synthesis is a reverse process of the analysis. The semantic representation of the source text is reprocessed by the pragmatic processor of the target language to find their equivalent style and ways to express the speech act in the target language. And then semantic processor restores the semantic representation received from previous stage to logical forms and passes on to syntactic processor, which searches for suitable lexical items to encode in the target language through the writing system. Finally, it gives rise to a target text. www.scholink.org/ojs/index.php/elsr Education, Language and Sociology Research Vol. 4, No. 3, 2023 136 Published by SCHOLINK INC. 3. Conclusion The models of translation processes that are proposed by researchers so far are product-oriented, therefore theoretical and speculative rather than empirical and concentrated on idealizations rather than on actually occurring data. Researchers speculate the thinking activities of translators during the translation processes based on the text analysis, such as comparing the source texts and the translated texts to identify the inappropriate translations of the translators, which can provide the researchers certain information about the processes. Therefore, the results from such research cannot fully reflect the actual translation processes, and hence more rigid and robust approaches were needed. As proposed by Bassnett-McGuine, a descriptive approach to the investigation of the translation process should be adopted rather than a prescriptive approach. The ‘descriptive approach’ means that the research on translation processes should be conducted via empirical approaches rather than theoretical approaches. References Bassnett, S. (2002). Translation studies (3rd ed.). London and New York: Routledge. Bell, R. T. (1991). Translation and translating: Theory and practice. London and New York: Longman Group UK Limited. Lederer, M. (1981). La traduction simultanée. Paris: Minard. Newmark, P. (1988). A textbook of translation. New York and London: Prentice Hall International (UK) Ltd. Nida, E. A. (1964). Towards a Science of Translating. Shanghai: Shanghai Foreign Language Education Press. https://doi.org/10.1163/9789004495746 Seleskovitch, D. (1968). L’interprète dans les conferences internatinales: Problèmes de langage et de communication. Paris: Minard. Seleskovitch, D. (1975). Langage, langues et mémoire: Etude de la prise de notes en interprétation consécutive. Paris: Minard. https://doi.org/10.1163/9789004495746