Using Trados’s WinAlign Tool to Teach the Translation Equivalence Concept
This article reports on the pedagogical effectiveness of using a translation memory tool, e.g., Trados's WinAlign, to assist in the teaching of the translation equivalence (TE) concept. Conventionally, translation instructors taught the TE concept using the prescriptive method of lecture and discussion of relevant TE theories. As an alternative, this study proposes the use of the descriptive method to teach the TE concept. The author administered an experimental technology-assisted translation project by engaging a group of graduate students in the editing of Trados's WinAlign-produced translation alignments. The editing process allowed students to identify various TE modes, such as strong TE, fair TE, weak TE and very weak TE. At the end of the translation project, students filled out a questionnaire. Although using the Trados WinAlign tool as an aid has demonstrated the pedagogical effectiveness to facilitate student acquisition of the TE concept, we still need to test the results of our research on different groups of students to more convincingly verify its benefits.
Translation equivalence (TE) refers to the equivalent relationships between target language (TL) and source language (SL). This concept has been explored for decades in the translation literature and has been discussed as a crucial subject in teaching translation. It has been found that trainee translators are prone to produce inappropriate translations when they do not have sufficient knowledge of the TE concept. Translation instructors conventionally lecture on the TE theories with supporting examples to teach the TE concept. The author in this study proposes the use of the TM technology, e.g., Trados WinAlign, as an aid to teaching the TE concept. She engaged a group of students in the editing of Trados's WinAlign-produced SL - TL alignments that allowed students to identify various TE modes.
1.1. Research questions
In this experimental project of teaching the TE concept with a visual alignment tool as an aid, the author raised the research hypothesis that this technology-assisted translation method is beneficial to students in three aspects. First, the process of editing a SL - TL alignment provides students with empirical experience and accordingly enhances the effectiveness of learning the TE concept. Second, the variation of the proportions of four TE modes across different text types enables students to learn the dynamic nature of the TE concept. Third, the Trados WinAlign-produced alignment provides visual cues for a better understanding of the TE concept, while student engagement in the editing of alignments reduces learning anxiety.
The above hypotheses led the author to raise some research questions. She expects to find answers from the findings of the students' questionnaires at the end of this project.
The answers to these questions would help the author/researcher understand the participants' thoughts and opinions about the use of the TM technology to assist in teaching the TE concept.
1.2. The structure
Many readers may have no clear idea about the conventional, theoretical study of the TE concept, so the author will introduce it in Section Two of this article. After that, she will introduce Trados's WinAlign, a TM tool, in Section Three. Section Four elaborates on the methodology of this TE project. The components to be discussed involve the subjects, the descriptive method, the teaching procedures, and a questionnaire-based survey. Section Five reports on the findings of the questionnaire-based survey. Section Six discusses the implications of the survey results. The conclusion reiterates the viable function of using the TM technology as an aid to teaching the TE concept and anticipates that this study will bring new insights for the emerging discipline of computer-assisted translation instruction.
II. The translation equivalence concept
The TE concept, a vital component in the discussion of the translation process, has been one of the most problematic and controversial areas in translation studies. Numerous scholars, including Eugene Nida (1964), Roman Jakobson (1959), John C. Catford (1965), Juliane House (1977), Peter Newmark (1988), Vinay and Darbelnet (1995) addressed the subject of TE using either the linguistic approach or the functional approach. Their common approach was to set the rules of TE and then to use samples drawn from texts to support the rules. In other words, the focus of their TE studies gave priority to theory over practice, and to fixed norms over dynamic principles. The following is a summary of their common features.
2.1. The dichotomous forms of TE
The above-cited translation scholars tended to discuss the TE theories in the dichotomous form. Catford made a distinction between textual equivalence and formal equivalence. The former refers to "any TL text or portion of text which is observed on a particular occasion to be the equivalent of a given SL text or portion of text" and the latter is "any TL category (unit, class, structure, element of structure, etc.) which can be said to occupy, as nearly as possible, the same place in the economy of the TL as the given SL category occupied in the SL" (1965: 27). Catford's TE theory is purely linguistic and overlooks the extra-linguistic factors that contribute to the production of functional equivalence between SL and TL texts.
Eugene A. Nida also made a distinction between two types of equivalence, formal and dynamic equivalence. Formal equivalence "focuses attention on the message itself in both form and content" (Nida, 1964:159). In contrast, "dynamic equivalence is based on the principle of equivalent effect, i.e., that the relationship between receiver and message should aim at being the same as that between the original receivers and the SL message" (Nida, 1964: 159). To illustrate the dynamic equivalence, Nida quoted J. B. Philips' rendering of Romans 16:16, where the idea of "greeting with a holy kiss" was translated as "giving one another a hearty handshake all round" (Bassnett, 1991: 26). In addition, Nida cited one example from a Biblical translation for the Eskimo's audience, where the phrase "Lamb of God" was translated as "Seal of God" because the lamb was not a symbol of innocence in the Eskimo culture. Although Nida illustrated the dynamic TE concept with extra-linguistic, culture-specific factors, his theory remained inadequate because he extracted all of his examples only from Biblical translations, rather than from diverse text types.
Central to House's discussion of TE was the concept of overt vs. covert translations. Overt translation does not need to recreate a second original since it preserves the original flavor in the TL text. Overt translation produces the effect of what Nida identified as the formal equivalence. House argued that "in overt translation, the SL text was not specifically addressed to a TL audience" (1977: 194). In contrast, covert translation implied "the production of a text, which was functionally equivalent to the SL text" (House, 1977: 194). House's theory resembled Nida's theory and remained inadequate because it was restricted to the dichotomous form.
Newmark examined the TE concept from perspectives that swung "between literal and free, faithful and beautiful, exact and natural translation, depending on whether the bias was to be in favor of the author or the reader, the source or the target language of the text" (1988: 45). He clarified that "communicative translation attempts to produce in its readers an effect as close as possible to that produced in the readers of the original" and that "semantic translation attempts to render as closely as the semantic and syntactic structure of the second language allow, the exact contextual meaning of the original" (1988: 39). All the above discussions of the TE concept were rigid and inflexible since they reduced the diversity of translation behavior to clear-cut dichotomous forms.
2.2. The prescriptive study of the TE concept
To modify the inflexibility of TE dichotomy, some scholars have expanded their study frameworks. For example, Mona Baker (1992) offered "a more detailed list of conditions upon which the concept of equivalence can be defined" (http://accurapid.com/journal/14equiv.htm). Putting together the linguistic and the communicative approach, Baker addressed the TE concept at various levels in relation to the translation process. As she observed, the translator first took into consideration equivalence at word level because the translator started analyzing the SL text to seek direct equivalent terms to replace original words as single units in the TL text. Gradually, the translator noted grammatical rules that could vary across languages and added or omitted information in the TL text to seek correspondence in the syntactic structure. Exceeding the above linguistic components, Baker noticed the textual equivalence that seeks contextual correspondence and cohesion between SL and TL texts. She also stressed cohesive ties and coherence of the TL text. Finally, she raised pragmatic equivalence, focusing on the clear presentation of the implied meanings of the source message to make it comprehensible to the TL audience. Although Baker has used a more comprehensive way to address the TE concept, her theory remained limited because her prescriptive study imposed arbitrary regulation without sufficient empirical basis. She discussed the TE norms first and then used examples to support the established norms. In addition, she prescribed some TE norms and then applied them to all text types.
As an alternative, this study neither adopts the prescriptive method of studying TE nor restricts the TE modes to the dichotomous forms. It makes use of the empirical method by identifying varied conditions for the production of TE in the real alignment-editing environment. This study also inserts varying degrees of TE between two poles of TE, ranging from lexicon-level TE, syntax-level TE, through context-level TE to culture/function-level TE. In addition, this study proves that the impersonal norm of TE is not universal to all text types; rather, TE varies with different text types. While conventional TE research moves from theory to practice, this technology-assisted TE research moves from practice to theory.
III. The TRADOS Winalign tool
Trados WinAlign, supported by Trados 6.5 Freelance, is a visual alignment tool., which allows users to create a translation memory through the use of available electronic translation materials from previous projects. The program automatically examines the source- and target-language texts to determine which sentence pairs can be aligned. Alignments in this case mean the result of making each translation unit of the SL text correspond to an equivalent unit of the TL text. The translation unit ranges from one sentence to shorter sequence of sentences or to paragraphs or chapters. The computing capability of probability calculation of equivalences produces the alignments between SL and TL texts. In other words, in Trados WinAlign, an automated structural-semantic recognition check procedure results in the production of SL and TL alignments. On the Trados WinAlign interface, the left column shows the SL text and the right column shows the corresponding TL text.
In the case of English-to/from-Chinese translations, the automatically-produced alignments are usually inaccurate and need further editing. Trados WinAlign provides the easy-to-use graphical interface for the editor to undo and redo the inappropriate alignments. The editor could use the function of "split" to split a long TL sentence into two shorter sentences to correspond to two short SL sentences. At the same time, the editor could use the "join" function to combine two or several shorter sentences into one long sentence, so it corresponds to a long SL sentence. In addition, the editor could directly add or modify or delete lexical items on the Trados WinAlign interface, so it is very convenient for student editing.
Since the editing process allows students to check whether a TL sentence appropriately corresponds to a SL sentence at the semantic, syntactic and pragmatic levels, students can learn the TE concept and detect varied modes of TE in the actual involvement. The author assumes that this method makes the students learn the TE concept more effectively and more impressively than the conventional theoretical study. To investigate whether this assumption is correct, the author tested a technology-assisted translation project on a group of students. The next section depicts how this project was undertaken.
This section illuminates some crucial components of technology-assisted TE teaching. These components involve the subjects, the descriptive method, teaching procedures, four TE modes, the mobile nature, and a questionnaire-based survey.
4.1. The subjects
A total of 13 participants in this TE project were graduate students from the track of translation and interpretation at National Kaohsiung First University of Science and Technology (NKFUST) in Taiwan. Of the 13 participants, there were 4 second-year students and 9 first-year students. Nine of them were majors in English and two were majors in business and translation in colleges. Their average age was 26 with 2 students over 30 and 11 students between 23 and 30. They had studied English for more than twelve years and their level of English was advanced. All of them had taken a Translation Theories course before they participated in this technology-assisted translation project. However, this was the first time they had the opportunity to learn the TE concept by using the TM technology as an aid.
4.2. The descriptive method
This study employed the descriptive approach, so it involved students in the learning of the TE concept in the opposite direction from the conventional prescriptive method. This alternative method does not give much theoretical input. Rather, it asks students to use Trados WinAlign, a visual alignment tool, to produce SL - TL alignments and to edit the inaccurate alignments. The focus is to infer rules from the data. The instructor does not offer the students preconceived generalized rules of TE. Students need to detect various modes of TE by themselves. This is the descriptive method of learning the TE concept within an empirical and flexible framework.
4.3. The teaching proceduresThe teaching procedure began with the instructor's introduction of how to use Trados WinAlign to align the available SL and TL texts. Next, the instructor showed students inappropriate TL and SL alignments that were displayed on the Trados WinAlign interface for editing. When the instructor used relevant examples to edit the alignments, she identified varied modes of TE. After that, the instructor used some example sentences extracted from three text types to show the changed distribution of the four TE modes across the text types. This demonstrated the mobile nature of TE.
Finally, students had a hands-on experience with the editing of Trados WinAlign-produced alignments. They had to detect the TE modes by empirically editing inaccurate alignments and also had to observe the changed proportions of the four TE modes using different texts of the same three text types.
4.4. Four modes of TEIn the teaching procedure, the instructor identified four TE modes through the editing of the SL - TL alignments. She defined these TE modes as strong TE, fair TE, weak TE and very weak TE.
4.4.1. Strong TE
By the author's definition, strong TE refers to a pair of SL and TL sentences that carry the same syntactic structure, transmitting the same message, and share the same number of corresponding sentences. Figure 1 shows one example of strong TE.
Figure 1: One Example of Strong TE
Figure 1 indicates that four English sentences correspond to four Chinese sentences. Although the Chinese sentences are more concise than the English ones, they are structurally similar and transmit the same message. We could identify this relation between SL and TL as strong equivalence.4.4.2. Fair TE
Fair TE refers to a pair of SL and TL sentences that carry different syntactic structures, but transmit the same message and have the same number of corresponding sentences. Figure 2 shows one example of fair TE.Figure 2: One Example of Fair TE
Figure 2 indicates that two Chinese sentences correspond to two English sentences, but their structures are not same. The adverbial phrase that marks time, 經過三十餘年的努力,is located at the second clause of the Chinese sentence, but its English translation after thirty years of work is at the end of the English sentence. Despite the slight difference in the structure, the SL and TL sentences transmit the same message. We could identify this as fair equivalence.
4.4.3. Weak TE
Weak TE refers to the corresponding units of TL and SL sentences that carry different word order, have a different number of sentences, but present a similar message. Figure 3 shows one example of weak TE.
Figure 3: One Example of Weak TE
Figure 3 indicates that three Chinese sentences correspond to two English sentences. Their structures are different. The Chinese phrase當伴手 (laden with gifts) is not put at the end of the Chinese sentence, but is placed at the end of the translated English sentence. In spite of these differences, the Chinese sentences and the translated English sentences transmit the same message about the products of I-Mei Company and customers' purchasing experience. We could identify this relation as weak equivalence.
4.4.4. Very weak TE
Very weak TE refers to the cases in which, while carrying out a similar textual function, neither SL and TL passages are equivalent in the number of corresponding sentences, word order, syntactic structures, or even in the context. Figure 4 shows one example of very weak TE.
Figure 4: One Example of Very Weak TE
Figure 4 reveals that three Chinese sentences correspond to two English sentences. Their sentence structures are different. For example, the clause 藉由贊助與活動推動, located in the middle of the Chinese sentence, is at the end of the translated English sentence. In addition, the whole Chinese sentence 同時對於文化資產的保護更是基金會工作重點之一 is removed from the English translation. The message concerning the prizes awarded by Cultural Construction Council of Administrative Office is missing in the English translation. Under the circumstances, we define the relation as very weak equivalence.
4.5. The mobile nature of TE
Having identified the four TE modes, the instructor used different text types to test the mobile nature of TE. Different text types show different patterns of TE mode distribution. For example, we extracted an average of 15 sentences respectively from one informative text type "Instructions for the Use of Feria 3D" (a user's manual for a hair-coloring product), one operative text type "The Web-Page of Caesar Hotel" (an advertisement), and one expressive text type Lord of Rings (a novel). A comparison between each of these text types leads to a finding that the informative text type bears the highest percentage of fair TE (66%) while the operative text type has the lowest percentage of strong TE (7%). In addition, the operative text type has the highest percentage of very weak TE (40%) whereas the lowest percentage of very weak TE lies with the informative text type (0%). The hidden reason for this is that the advertisement (the operative text type) contains metaphorical descriptions, so that its translation displays a higher percentage of very weak TE than the informative text type (0%) and the expressive text type (6%).
In contrast, a user's manual (informative text type) provides the literal description so that its translation shows a higher percentage of strong TE (34%) than the operative text type (7%) and the expressive text type (27%). We summarize the statistics of the different proportions of four TE modes in Table 5.
Table 5: Proportions of Four TE Modes in Three Text types
The above table indicates that not one text type presents only one TE mode and the TE modes vary with different text types. In addition, we found that the Chinese source text of the hotel advertisement described the strengths of the hotel to create a powerful appeal to the Chinese audience. However, its English translation, probably for the reason of the translator's limited linguistic competence or for the purpose of adaptation, has deleted some of the original message and has toned down the metaphorical description. As a result, the hotel advertisement presents a higher percentage of very weak TE than other two text types.
At the same time, we found that the proportions of four TE modes underwent a change when we compared different passages which we extracted from the same text type. For example, two texts of the same operative type show different patterns of the four TE modes. This is because one advertisement is strongly factual but the other advertisement likely pursues an emotional appeal through the use of some rhetorical devices or semiotic strategies. Due to the changed strategies of description used in the same text type, the distribution of the four TE modes within a text type is subject to a modification. To justify this mobile feature, we could use different passages or excerpts from the same text types for a comparison.
In the second test, we extracted 15 sentences from a different informative text type, "User Guide for the Air Cleaner," and from a different operative text type, "ASUS Web-page." From the same expressive text type, Lord of Rings, we quoted new 15 sentences. Table 6 shows the statistics of the distribution of the four TE modes in these three text types.
Table 6: Proportions of Four TE Modes in Three Text Types
Table 6 shows different distribution percentages of the four TE modes from those of Table 5. The operative text type (The ASUS Web-Page) has the highest percentage of strong TE and the highest percentage of very weak TE. Furthermore, Table 6 shows that one informative text type (User's Guide for the Air Cleaner) has a lower percentage of strong TE than the same informative text type (Instructions for the Use of Feria 3D) in Table 5. The new passage extracted from the same novel shows a lower percentage of strong TE than the passage in Table 5. However, the new passage has a higher percentage of weak TE. This phenomenon clearly revealed that different passages or excerpts from the same text type would lead to different distribution patterns of the four TE modes. A comparison of Tables 5 and 6 allows students to detect the dynamic nature of TE.
4.6. A questionnaire-based survey
Having implemented the Trados WinAlign-assisted project for four 3-hour sessions in sequence in her translation class, the instructor/author conducted a questionnaire-based survey. The author followed the research hypothesis to design a questionnaire that consisted of 15 Yes-No questions and two open-response questions (see Appendix 5). The questions were evenly distributed under three categories: the descriptive study of TE, varied TE modes and their mobile nature, and affective and cognitive impacts. All these questions in the questionnaire attempt to elicit answers as clues to verify the author's hypotheses.
At the end of the Trados WinAlign-assisted translation project, a total of 13 students filled out the questionnaires. The instructor reminded them that all answers were not related to the assessment of their translation performance, so that they could feel free to write their answers frankly.
5.1. Results of the "Yes-No" questionnaire questions
We calculated the total number of responses under each category and the total percentages. Tables 7-10 show the results of student responses to these questions.
Table 7 indicates that student responses to Q4 and Q5 were more positive than the responses to Q1, Q2 and Q3.
Table 8: Statistics on Responses to the Questions Concerning Varied TE Modes and their Mobile Nature
Table 8 shows that all students had one-hundred percent positive responses to Q8, Q9 and Q10. In addition, student responses to Q6 and Q7 were quite positive, reaching the significant percentage of 92.3%.
Table 9 indicates that student responses to Q11 presented the highest percentage, while student responses to Q12 showed the lowest percentage. Table 10 shows the students' overall responses to this TE project based on the calculation of the average percentages.
Table 10: Statistics on Students' Overall Responses to This Project
Table 10 revealed that a majority of students, the significant percentage of 82.05%, accepted the use of the Trados WinAlign tool as an aid in the lerning of the TE concept.
5.2. Results of the "Open-Response" questions in the questionnaire
There were two open-response questions in the questionnaire. We transcribed the written answers in a matrix under the categories of learning experience and expectations concerning the innovative teaching pedagogy. Table 11 summarizes the students' overall positive responses to this survey.
Table 11: Statistics on Student Written Answers to the "Open-response" Questions
Table 11 clearly shows that all students made positive comments on the use of Trados WinAlign to teach the TE concept. They affirmed the pedagogical effectiveness of the integration of technology into the translation class for the purpose of teaching the translation theory.
VI. Discussion and implications
The results of the questionnaire revealed the students' positive attitude toward the implementation of the technology-assisted translation pedagogy. Following is a summary of the significant implications behind the findings of this survey in the areas of "Yes-No" questions and the "Open-Response" questions.
6.1. The technology-assisted approach to the teaching of TE
The results of the second-part questionnaire indicated that more than half of respondents agreed that the innovative Trados WinAlign-assisted approach was satisfactory and effective. However, we noted that few students did not think that the editing of Trados WinAlign-produced alignments was less theoretical and accordingly more convincing. In addition, these students did not think this technology-assisted pedagogy was more objective than the conventional theoretical study. These results implied that the use of technology to teach the TE concept cannot be regarded as a method of enhancing the validity of the TE concept neither can it be accepted as a more objective way of learning the TE concept. However, we cannot overlook the fact that the students had reached a consensus that the Trados WinAlign-assisted pedagogy of teaching the TE concept was more effective than the conventional prescriptive study.
6.2. The cognitive learning of the TE concept
The result of student responses to the third-part questionnaire showed that the overwhelming majority of students (above 92.30%) agreed that the editing of the Trados WinAlign-produced alignments not only helped them learn the various modes of TE but also enabled them to realize their mobile nature. In addition, they noticed that the distribution patterns of the four TE modes varied across different text types. This clearly outlined that the realistic editing environment provided students with a good opportunity to observe the sophisticated TE concept.
Editing the inaccurate translation alignments allowed students to develop the ability to solve problems. The process of thinking about how to distinguish varied TE modes has contributed to the cognitive learning of the TE concept. We could describe this kind of learning as the production of situated cognition. Theorists like Brown Collins and Diguid have argued for the importance in the learning of what has become known as situated cognition (Butler-Pascoe & Wiburg, 2003: 177). This claim echoed the underlying theme of social constructivism, raised by Piaget (1929, 1970, 1973), Vygotsky (1978) and Bruner (1990). These theorists proposed that there was a need to design a curriculum in which students were afforded opportunities to construct their own meaning through student-directed interaction rich in resources and problems to solve. The technology-assisted teaching we emphasized in this study relied on the students' empirical experience and the situated learning environment helped them to develop the cognitive ability. In other words, students' actual engagement with the editing of various sets of SL and TL alignments helps them to integrate the knowledge of TE modes into their own developing cognitive structure (Butler-Pascoe & Wiburg, 2003: 177). This view conforms to the function of process learning. In short, this technology-assisted pedagogy relates student learning experience to the actual environment and enhances their cognitive integration.
6.3. Affective contribution
Of the five questions in the fourth part, only less than half of the respondents (37.61%) agreed that the process of editing the inaccurate alignments made them feel less anxious when they had to monitor and detect the TM modes without the teacher's interference. One student even (7.07%) disagreed with this point. This revealed that many students still needed immediate and constant assistance from the instructor in the editing process. In the technology-assisted learning environment, the human element still plays a significant role and we cannot overlook the instructor's role in giving students immediate feedback. With regard to the visual effect of linking lines, more than half of students (70.76%) agreed that they relied on the visual cues to help them grasp the concept of equivalence between SL and TL sentences. According to Buttler-Pascoe & Wiburg (2003), everything students learn must "first come through the senses because the initial stage of cognitive processing requires perception" (167). The perception acquired through the visual cue helps students to easily move onto the phase of TE concept acquisition. In this study, the visual cues of SL and TL alignments, displayed in the window of Trados WinAlign, attracted student attention and then motivated students to edit them to seek varied modes of equivalence. The visual cross-reference between SL and TL pairs enables students to learn TE modes more substantially and impressively. In short, many students have positive responses to the visual impacts of editing the Trados WinAlign-produced alignments.
6.4. The overall positive attitude towards the technology-assisted TE pedagogy
Student reflections on the two open-response questions doubtlessly revealed that students were enormously satisfied with using the Trados WinAlign tool to assist in the leaning of the TE concept. All students were positive about the hands-on experience with the editing task. They all agreed that this method provided students with a clear, easy, effective, and substantial way of learning the TE concept. This ensured that the use of the translation technology could facilitate student acquisition of the TE concept.
Although the use of the Trados WinAlign tool could support the development of student cognitive ability via discerning the TE phenomenon in the editing process, the instructor must design some tasks that initiate the problem-solving process and therefore enhance the pedagogical effectiveness.
Contrasting with the teaching of Trados WinAlign simply for the input of technical know-how, this Trados WinAlign-assisted pedagogy method teaches students about the TE concept. Since this study has confirmed the functional relevance of the translation technology to the teaching of the TE concept in the translation class, other translation instructors can make further attempts at other functions of the Trados WinAlign-assisted translation pedagogy.
Bassnett, Susan (1991). Translation Studies. London and New York: Routledge.
Bruner, J. (1990). Acts of Meaning. Cambridge, MA: Harvard University Press.
Butler-Pascoe, M. E. & Wiburg, K. M. (2003). Technology and Teaching English Language Learners. Boston & New York: Pearson Education, Inc.
Catford, J. C. (1965). A Linguistic Theory of Translation: An Essay in Applied Linguistics. London: Oxford University Press.
House, Juliane (1977). A Model for Translation Quality Assessment. Tubingen: Gunter Narr.
Newmark, Peter (1988). A Textbook of Translation. London & New York: Prentice Hall.
Nida, E. A. (1964) Toward a Science of Translating: With Special Reference to Principles and Procedures Involved in Bible Translating, Leiden: E. J. Brill.
Piaget, J. (1929). The Child's Conception of the World. New York: Harcourt, Brace Jovanovich.
------- (1970). The Science of Education and the Psychology of the Child. New York: Grossman.
------- (1973). To Understand Is to Invent. New York: Grossman.
Vinay, J. P. & Darbelnet, J. (1995). Comparative Stylistics of French and English: A Methodology for Translation (trans. and ed. By J.C. Sager and M. J. Hamel), Amsterdam: Benjamins.
Vygotsky, L. S. (1978). Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press.
Please see some ads as well as other content from TranslationDirectory.com: