Psycholinguistic aspects of simultaneous interpretation

Psycholinguistics studies the processes of speech generation and perception in their correlation with the language system. This science seeks to interpret language as a dynamic system of human speech activity. Since the object of psycholinguistics is speech activity, and the object of translation theory is a special type of speech activity, the tasks of these disciplines largely overlap.

The data of psycholinguistics are used in the study of the features of simultaneous translation and the mechanisms that ensure the simultaneous processes of perception of the original message and the generation of the translation text. Simultaneous translation is one of the most difficult types of translation. This is due to the psycholinguistic features of this speech activity, including the features of the perception of the incoming signal, the functioning of memory at various stages of processing linguistic information, the specifics of generating utterances in the target language, as well as the features of controlling the generated translation solution.

The main difference between simultaneous translation and other types of translation is the parallel perception of the speaker's speech and the generation of speech in the PL. This determines other features of this type of translation and, first of all, a strict time limit: the simultaneous interpreter is given only the period of speech delivery by the speaker. This time is twice less than the time available to the translator for consecutive translation. Therefore, an indispensable condition for the success of the synchronist's work is the presence of a large stock of equivalent pairs of lexical units connected with each other by a sign connection, which allows translating not through analysis and synthesis (i.e. thinking), but through conditioned reflexes, i.e. at the level of "stimulus-reaction".

The simultaneous interpreter is also imposed the pace of translation, which must correspond to the pace of the speaker's speech. In addition, a feature of simultaneous translation is the so-called segment-by-segment nature: a simultaneous translator translates the text in segments as they arrive, whereas in sequential translation, the translator first listens to the entire text. The work of an interpreter in conditions of simultaneous listening and speaking requires constant concentration and continuous speaking. There is a situation of split attention caused by the need to constantly compare two languages and switch from one code to another. A. F. Shiryaev proposed a psycholinguistic model of simultaneous translation, in which translation is considered as an activity consisting of successive interrelated actions. Each action consists of three phases: the orientation phase in the context of the next translation task and the search for a solution, the implementation phase and the control phase. Let's look at these phases in more detail.

The first phase of the action involves orientation in the conditions of the translation task. If, as an example, we consider the situation in relation to the activities of an interpreter at an international congress of political figures, then on the basis of knowledge of the international situation, the history of the preparation and convening of the congress, the goals of the participants, the policy of the country that the speaker represents, and other factors, the translator builds in his mind a probabilistic model of the oratorical speech that he will translate. At the beginning of the speaker's speech, the translator perceives the initial segment of the source text that makes up the conditions for the first translation task, and identifies landmarks that characterize its semantic content and language form. Thus, the translator accumulates information to search for a translation solution. If the translator's pre-setup for the first action was adequate and sufficiently detailed, then the initial segment is identified, and the translator proceeds to choose a solution from a number of prepared ones. If the forecast turned out to be incomplete or not quite accurate, the translator searches for a solution based on the selected landmarks, their previous experience and existing knowledge. With the decision, the first phase of the action ends.

The second phase of the action (the implementation phase) is the completion of the generation of the text segment in the target language in accordance with the decision made. During the second phase, the translator unfolds in the external speech a fragment of a statement in the target language, constructed or selected from a number of prepared actions during the first phase.

The first and second phases of the translation activity are characterized by transience. Their duration is calculated in fractions of a second or several seconds. The third phase of the action (the control phase) is the translator's assessment of the correctness of the translation. This phase is significantly extended in time, since the translator can verify the correctness of the decision made only after listening to several statements of the speaker. In parallel with the search for a solution and the implementation of the first action, the translator is forced to start navigating in the conditions of the next, second translation task. Thus, the first phase of the second translation action began at the moment when the first action was on the verge of its second phase. The search for a translation solution in the second act was carried out later-in parallel with the end of the second phase of the first act. As for the implementation phase of the second action, it followed the implementation phase of the first action. The control phase of the second action followed in turn the implementation phase of the same action, unfolding in parallel with the control phase of the previous action.

The third and subsequent translation actions follow a similar pattern and join with the preceding action in the same way that the second action joins with the first. In the process of simultaneous translation, listening and speaking are two sides of dialectical unity, opposing and "interfering" with each other, but forming a single whole. Based on the existence of a mode of simultaneous perception and speaking in simultaneous translation, the question arises as to what mechanisms ensure this simultaneity.

The first of them is the mechanism of probabilistic forecasting. The essence of this mechanism is that in the process of auditory perception of speech, the translator puts forward hypotheses about a particular semantic or verbal development or completion of the author's intentions. Such hypotheses are put forward on the basis of a subconscious subjective assessment of the probabilities of further development of a given semantic or verbal situation. In the course of translation, the translator develops and clarifies these ideas, constantly puts forward assumptions about the syntactic structures of utterances and their lexical content, and clarifies his hypotheses with the help of orientation in the speaker's speech. Continuous probabilistic forecasting of the speaker's speech at several levels at once facilitates the work of the translator, since it sets him up in advance to solve continuously emerging translation problems, allowing him to identify them by individual signs instead of completely perceiving them and analyzing them.

Steyer in the article "On the mechanism of simultaneous translation" expresses the opinion that the semantic hypothesis can not be the main lever in the mechanism of simultaneous translation. He argues for this position by the fact that translators often have to serve scientific conferences, where issues are discussed that are completely incomprehensible to the translator. However, the author acknowledges that "design" takes place in relation to the structure of the sentence. You can design the structure of a translation phrase, as a rule, at the very beginning, sometimes even from the first word of the original sentence.

The next way to create a synchronization mechanism is speech compression. Speech compression is a method of reducing the syllabic value of a text without causing significant damage to the performance of the speaker's communicative task. Speech compression becomes possible due to the information redundancy of speech. Schweitzer believes that text compression is achieved by omitting redundant elements of the utterance, elements that can be filled in from the non-linguistic situation and the context of communication, and, in addition, by using more compact forms of expression. The phenomenon of compression in simultaneous translation follows from a number of linguistic and psychological patterns of speech message transmission, when the core of the original utterance is retold, which carries essential information for understanding, and the details of the message expressed by the secondary members are lost during recoding. Speech compression is performed using a number of techniques.

The first and most common technique of speech compression is to eliminate the segments of the text that carry information, which is filled up by the extralinguistic situation of communication. The second technique is to eliminate from the text segments that duplicate the content of the previous statements of the speaker.

The third method of speech compression consists in removing from the text the formulas of politeness, deviations from the topic, some epithets and participial phrases. Also, numerical data, dates and proper names may in some cases not carry a significant semantic load (for example, acting as an illustration to a certain thesis put forward by the speaker), and in such cases the translator may sacrifice them in the translation process. In conditions of time scarcity, the omission of such elements of the statement seems quite justified.

The next mechanism that allows the translator to perceive and understand the speech of the speaker during the periods of pronouncing the text on the TL is called by A. F. Shiryaev the mechanism of auditory recognition of speech images.In the conditions of normal speech activity, the perception and understanding of speech occurs on the basis of recognition of words and phrases with hidden articulatory reproduction of speech signals. In simultaneous translation, during the periods when the text is spoken in the target language, the participation of the speech engine analyzer in recognizing speech signals in the source language is extremely limited. The limited possibilities of using the speech motor code to translate the speaker's speech into their own internal language encourages translators to use a new subjective code that is formed during training, usually without the translator realizing this phenomenon. The basis of the functioning of this code is the recognition of words and phrases only on the basis of their auditory images without counter articulation reproduction to oneself, as well as the retention in consciousness of the information received in the form of visual and auditory sensations. Therefore, during simultaneous interpretation, many translators note the appearance of vague visual sensations corresponding to the phenomena described in the speaker's speech. The mechanism of auditory recognition of speech images without their articulatory reproduction is supplemented by the use of micro-pauses in their own speech for the perception of certain fragments of words in the source language. Utterance of speech on the TL with regular micropauses within the speech links makes it possible to perceive individual fragments of words with an unoccupied speech motor analyzer.

The psycholinguistic aspects discussed in the article reveal the principles of simultaneous translation and explain how simultaneous perception and speaking in simultaneous translation becomes possible.

Жанатаев Данат Жанатайұлы Әл-Фараби атындағы Қазақ ұлттық университетінің профессоры, Болат Адия Ержанқызы Әл-Фараби атындағы Қазақ ұлттық университетінің магистранты

Әлеуметтік желілерде бөлісіңіз: