Translation Skill in Language Learning/ Teaching: EFL Learners’ Point of View

Introduction S T U D I E S O F F O R E I G N L A N G U A G E S / K A L B Ų S T U D I J O S This study seeks to investigate EFL students’ perception of the benefits of translationas a skill for their language learning instead of as a mere tool in the learning process, and the relevance of a translation course in their study program. Ninety-five Indonesian university students of an English Department taking an EFL teacher training program participated in this study by filling in a questionnaire using a four point Likert scale with items consisting of complementary open and closed ended questions. The responses show an overwhelming favorable response towards the merits of translation as a skill for their L2 language learning regardless of their L2 (English) proficiency, which was represented by their GPA. Most participants also believed that a translation course is very relevant to their study although translation is not an intended career path for the program’s graduates. This study further challenges the belief that translation is harmful for learners’ language acqusition and therefore should be avoided in language learning/ teaching.

With the "failure" of the Grammar Translation Method (GTM) in the 19th century, translation has been treated with so much hostility in language teaching that it has even been banned from language classrooms (Carreres, 2006). It is argued that translation prevents the development of learners' automaticity in L2 use, encourages reliance on L1 in L2 production, over-emphasizes accuracy over communication, is a source of language interference, or is simply irrelevant. Recently, however, there has been emerging interest to re-examine the negative claims about translation in language learning/ teaching, an area which Cook (2010) has given the acronym TILT (Translation in Language Teaching/ Learning). In spite of this, almost no study has specifically examined the benefits of translation as a "skill" in this context. Most studies have instead focused more on translation as a tool in language learning, which may not be based on proper translation strategies. Unlike such research, the present study seeks to explore students' views of "translation as a skill" in language teaching/ learn-ing. Such an investigation is paramount in order to give a fair re-evaluation of TILT. This is because "true" translation, even when used as a teaching method or learning tool, cannot be divorced from its nature as a skill of interlingual communication. It is a complex skill which entails different pragmatic considerations of a text and therefore cannot be reduced to a mere change of linguistic forms to transfer meaning, or it will only risk various kinds of faulty rendering. As a preliminary study, this research aims to investigate students' general perceptions of the benefits being skilled in translating for their language learning. Focus has been given to the impacts of translation on L2 writing and speaking proficiency, because the concern tends to be on how translation may be harmful to the acquisition of productive skills in the form of L1 interference. In addition to this, the study also aims to answer two subsidiary questions related to how translation should be applied/ taught in the students' language program and if the students' GPA (grade point average), which reflects their L2 proficiency, influences their beliefs about translation.

Learner's Beliefs about Translation and Its General Benefits for Language Learning
Research on students' perception reveals positive beliefs about translation (Carreres, 2006;Fernández-Guerra, 2014;Liao, 2006). Liao (2006), who examined 351 Taiwanese university students, for instance, reported that participants believed that translation assisted them to acquire L2 (English) in different areas of learning, such as reading, writing, speaking and vocabulary. Further, Carreres' survey (2006) shows Modern Languages students at Cambridge University strongly believed that translation into L2 is a useful device in language learning (4.6 mean out of 5 perfect score). Over 90 % of the students also saw that translation into L2 assisted them to learn vocabulary (100 %), grammar (96 %), writing (96 %) and L2 register (90 %).
In line with students' positive perception, Schaffner (1998, in Dagilienė, 2012), points out that translation may (1) enhance learners' verbal skills, (2) increase L2 vocabulary, (3) improve language style, (4) develop language awareness or understanding of how languages work, (5) help to consolidate their L2 grammar in active use, and (6) check and improve their L2 comprehension. Translation has often also been identified as a means of scaffolding for new L2 learners in their production of L2. According to Renandya 1 (personal communication, 21 February, 2015), translation is "a natural process" -it is "a useful learning tool" for beginners because "L1 is often the only resource that they can fall back on to". However, Renandya also points out that "translation can be problematic if students do this poorly" due to poor L2 proficiency, resulting in grammatically and pragmatically ill forms (ibid). In my view, however, this is a good reason why teaching students how to translate may be helpful. As Dagilienė (2012, p.125) points out, by translating properly students "[need to focus] on identifying differences in structure and vocabulary" and therefore "they have to evolve strategies to deal with them and to negotiate the potential of both languages". In other words, translation skill will help learners to grow in their understanding of how L2 differs from the L1 and this will in turn assist them not to translate literally from L1.

Benefits of Translation for Writing
Studies have been conducted on the use of translation in writing (Cohen et al., 2000;Uzawa, 1996;Kobayashi & Rinnert, 1992;Hu, 2003;Lifang, 2008). These studies demonstrate that 1 Willy Renadya, who has extensive teaching in Asia, is a senior lecturer at the Nanyang Technological University, National Institute of Education, Singapore, and specializes in TESOL (Teaching English to Speakers of Other Languages).

Review of Literature
translation benefits EFL learners with lower proficiency to develop more complex ideas and better structures. In spite of this, Kobayashi and Rinnert (1992, p.197) found that the use of translation in their research led to more "lexical choice errors" and "more awkward forms" compared to direct composition in L2. However, the respondents of such studies did not seem to have received proper translation training, hence had not been equipped with skills to render L1 concepts into L2. Because of this, it is unfair that such weakness has been attributed to "translation".
Research also shows that the use of mental translation in writing is common among learners when they are composing in L2 e.g. Cohen et al., 2000;Hu, 2003). This is in line with Leonardi's (2010, pp.62-3) belief that the use of translation (or L1) in L2 teaching is "a natural phenomenon" since L1 and L2 are "constantly and automatically interwoven in the learner's mind at all levels" (e.g. phonology, syntax, lexis and pragmatics). According to Cumming (1989, as cited in Cohen et al., 2000), more advanced learners were also found to use L1 strategies even when they were supposed to write directly inL2. On this account, Cumming (ibid) concludesthat learners' proficiency in L2 did not seem to determine their choice of writing strategies. Hu (2003), on the other hand, suggests that the language by which the knowledge is acquired affects the choice of thinking language in L2 writing. In this way, one would be more likely to think in L1 when dealing with topics studied or acquired in the language. Translation, however, is not without any downside. Lifang's (2008) study, for instance, suggests that the use of translating strategy is not ideal when learners compose their L2 under time pressure (ibid). Such a limitation needs to be judiciously considered in order to develop a sound methodology in implementing translation in language learning.

Benefits of Translation for Speaking
Unlike in writing, studies on the benefits of translation for L2 speaking are very limited. The idea of "using" translation to enhance the oral skill seems to be more controversial compared to that for other productive skills. In spite of this, learners report to have benefited from translation (Fernandes-Guerra, 2014; Liao, 2006). One cutting-edge study is Van Dyk's (2009) research using sight translation, which is normally used to train professional conference interpreters, to help learners speak more "naturally". Seeing that language learners tend to mentally translate but are much more prone to literal translation than professional translators, she believes that they need training using a "professional translation approach" rather than "pedagogical" translation (ibid., p.203). With this reason she advocates the sight translation technique to train students to translate more idiomatically by relying more on the context rather than the source text's form. She found students who were proficient with this method could increasingly speak with less interference from L1 by employing compensatory techniques to overcome their linguistic limitations. This research suggests than translation can help curb unnatural L2 production rather than encouraging it when applied properly in language learning.

Research Questions
This study seeks to answer the following research questions: 1. What are the students' perceptions towards translation as a skill into L2 in language learning/ teaching? 2. What are the students' perceptions of whether or not the Indonesian into English Translation course should be retained in the language program?
3. Does GPA (Grade Point Average) affect their perceptions of translation in language learning/ teaching?

Participants
The participants of this study were intermediate to advanced EFL students of the English Language Education Program, Faculty of Language and Literature, Satya Wacana Christian Uuniversity, Central Java. This program adopts an English-only policy for its courses, except in Translation classes and, only recently, Grammar courses.
A total of 95 students participated out of the targeted population of around 130 students who had taken the Indonesian into English Translation course. This 2 credit course, offered in the third year of the program, is aimed at enabling students to translate more idiomatically from Indonesian into English across different types of texts.
The syllabus is organized according to genres, for instance, translation of news, advertisements, and academic texts. Each lesson (approx. 110 minutes) consists of the following activities: 1. Reading a short Indonesian text (250-350 words) and its English translation and vocabulary exercise to help students understand the translation (15-20minutes).
2. Comparing and contrasting the translation with its Indonesian source text, and analyzing the rendering strategies in terms of their formal and pragmatic aspects, including the consideration of the purpose of communication and the audience (approx. 25 minutes).
3. Language focus to discuss specific linguistic characteristics of the genre (in Indonesian and English) and/ or a common grammatical construction in the English genre (15-20 minutes).
4. Short translation exercise (group work) and discussion of the resulting translation (approx. 45 minutes).
It should be noted that prior to taking this translation into L2 course, students were required to take the English into Indonesian course (translation into L1), in which they were given some theoretical introduction to translation for the first three weeks. This theoretical section covered areas such as the concept of translation as an interlingual type of communication, discrepancies among languages, and various strategies of rendering to negotiate the differences. No L2 speaking activity was involved in the course.

Instrumentation
In order to collect data, the study administered a questionnaire with eight main statements in Likert Scale and five complementary questions in the form of multiple response sets and open-ended questions (see Appendix). A four point Likert Scale was employed for the main questions to avoid non-committal responses. The first five questions were used to answer the first research question and probe into students' perceptions on the advantages of translation as a skill in students' language learning, especially in relation to their L2 production (speaking and writing skills). On the other hand, statements 6 to 8 were designed to answer the first subsidiary research question, and hence to elicit the respondents' views on how translation skill should be taught/ applied in their language program.

Data Collection and Analysis
This data collection was conducted in classes around three weeks after the translation course was finished and the students had received their grades. This was done so that participants could freely express their opinions. All of the students were given around twenty minutes to fill in the form without specific briefing, as the introduction of the questionnaire was designed in such a way as to enable independent completion. They were requested to write their name and contact details only if they were willing to participate in case there was any interview. After the data was collected and entered into STATA, it was calculated into percentages and tested using ANOVA and Linear regression. This statistical test was conducted to find out if the students' GPA affected their perceptions of translation in language learning/ teaching. The GPA would be used as an approximate indicator of students' English proficiency because most of the participants did not have a TOEFL or IELTS score. Although GPA is not a specific measure of English competence, it is an arguably reliable indicator of language aptitude in this context because English language skill courses made up most of the courses taken by the third year (old curriculum) and over 95 % of the program's courses were conducted in English. In addition to this, the grades in the program's GPA were also obtained from continuous evaluation, which improves its reliability as a benchmark of language competence (as opposed to standard proficiency tests, which are only based on a one-time exam).

Findings and Discussion
Students' Perceptions of the Advantages of Translation Skill Towards Their Learning of English as An L2

Linguistic Awareness
As seen in Figure 1, all but one respondent felt that learning to translate into L2 was helpful to them in developing an awareness of differences between L1 and L2 (Statement 1). As many as 62.1 % of the respondents ticked "agree" while 36.8 % selected "strongly agree", which shows a strong confirmation of the positive view of respondents. This finding is not surprising since translation training requires students to compare and contrast L1 and L2 continuously as they learn to reformulate meaning from one language form to the other. This aspect is vital because in order to translate well, one needs contrastive knowledge which may be defined as "conscious[ness] of the degree to which languages coincide and differ" (Faber, 1998, p.10).
Further, when asked what areas of differences they had become more aware of from learning to translate into L2, most respondents gave more than one answer, typically grammar and diction idiomatic expressions (see Table 1). Discourse awareness was also identified with a significant frequency (26.3 %) out of the total occurrences of the selected choices or by over half of the respondents (57 students). This finding supports Carreres' (2006) report that translation was seen to assist acquisition of grammar, vocabulary and knowledge of regis-

Figure 1
Learning to translate into English helps me to become more aware of the differences between Indonesian and English Disagree ter. These areas of language correspond to those covered when analyzing translated texts in the Indonesian to English course. It is important to note, however, that the respondents may have gained the grammatical advantage not only from the language focus session, but also from the feedback given to them on their translated work. Unlike in L2 writing courses, such feedback may be more effective to correct L1 interference because students can directly contrast L2 forms with those in L1 that they have translated in specific contexts from the source text. Frequent encounters with particular grammatical items (e.g. use of English articles 'a', 'an', and 'the') in the translation exercises also seems to have encouraged learning of grammar. This study also demonstrated the students' positive beliefs towards the role of translation skills in L2 writing (Statement 2 & 3). As seen in Figure 2a, almost all of the students (96.8 %) either agreed or strongly agreed that learning to translate into English is important because they tend to translate ideas mentally when writing in the language. This finding implies that mental translation is an important resource for the students irrespective of their English proficiency as represented by their GPA (Grade Point Average). It is important, however, to note that the context of this research is in a foreign language setting (EFL), in which L2 exposure is limited and day to day communication is normally done in L2 outside classes. In such a context, more knowledge about the world would be acquired in L1.

Benefit in Writing
It is therefore no surprise that students showed an overwhelming support towards the statement that translation skills into English help them to write more naturally (see Figure 2b). All of the respondents responded positively to this item with almost one third of them (28.4 %) expressing strong agreement. While this perceptual data cannot be treated as "direct evidence" for the merit of translation skills in language learning, this finding reveals very strong support from language learners, which we cannot ignore, of the advantage of the interlingual skills in L2 writing.

Benefit in Speaking
While the use of translation as a method to teach L2 speaking

Figure 2a
I think learning how to translate into English is important because I tend to translate mentally when writing in the language

Figure 2b
The skills to translate into English help me to write more naturally in the language has been much avoided since it is thought to inhibit automaticity in language production, there was a strikingly positive response towards the suggested importance of translation as a skill in speaking ( Statement 4 & 5). Figure 3a shows that 87.4 % of the respondents agreed or strongly agreed that the skills helped them in expressing themselves orally in L2 because they tend to use mental translation for these types of activities. It is interesting that students with higher GPAs also reported the tendency of mental translation in speaking and perceived a need to study translation to assist them in the area. This might have been due to students' EFL setting in which the use of English is very limited and Indonesian tended to be the dominant means for cognitive processing.

Figure 3a
Learning to translate into L2 is important for me due to the tendency of mental translation in L2 speaking

Figure 3b
Translation skills into L2 help me to speak more naturally in the language and expressions, they might still have had to rely on their L1 to express more complex ideas and knowledge they had acquired in L1.
It is also very revealing that the data strongly supports the statement that translation skills help them to speak more naturally. As shown by Figure 3b, as many as 77.9 % of the respondents expressed their agreement or strong agreement towards this possible role of translation. This finding echoes that of the same question related to L2 writing, although the confirmation is not as strong, as there is still a small but significant number of students (21.1 %) who responded negatively. In spite of this, the positive views of the far larger number of respondents indicate that translation skills may indeed have more positive influence in learner's oral language production than what is commonly assumed. This finding corroborates Van Dyk's study (2009) on the efficacy of professional translation training to improve learners' L2 speaking performance.

Students' Perception of the Use of L1 Dafting in L2 Writing
As shown in Figure 4, the respondents were divided in their answers to the question about whether drafting in L1 before translating it into English should be allowed in English (L2) writing classes (Statement 6). There is a slight but not significant tendency to be against the statement with one answer missing. As many as 47.4 % either agreed or strongly agreed with the statement, while 51.6 % disagreed to it. This shows that while most students believed that translation is helpful for their language learning, they did not necessarily approve of its application as a "direct" or "explicit strategy" in writing. Table 2 shows reasons why the 45 students (47.4 %) supported L1 drafting in writing classes. They normally ticked two of the suggested choices. The most popular reason was that it enabled them to keep going in writing in L2 (33 students). The next most popular reasons were that it assisted them to think more clearly (28 students) and that it helped them to write more complex ideas (27 students). Despite this, only 19 students ticked the option "The strategy should be allowed because I tend to mentally translate". On the other hand, this low percentage does not necessarily contradict the earlier finding that most students agreed or strongly agreed that translation skill is important because they tend to mentally translate (see Figure 2a). While it may have been the case that they often did mental translation, when responding to the question they may have been more interested in the perceived practical effects of using translation in their L2 writing.
Similarly, as shown by Table 3, the 49 students who had a negative view towards L1 drafting tended to select two reasons for opposing the strategy. As many as 41 students believed the use of L1 drafting takes time, which suggests a concern about efficiency. Not far from this, 33 participants ticked "Students should be encouraged to think and write directly in L2". There was apparently a perceived ten-

Figure 4
I think writing in Indonesian first then translating it into English should be allowed to some extent in our writing classes Table 2 Reasons for supporting L1 drafting

Percent Percent
The strategy sometimes helps me to keep going in writing in L2 33 30 The strategy may help me to think more clearly 28 25.5 The strategy may help me to write more complex ideas 27 24.5 The strategy should be allowed be-  Dichotomy group tabulated at value 1.N= number of students/ responses sion between the need to develop fluency by directly processing meaning in L2 and the use of L1 in translation, which also reflects a common reservation towards translation language learning. In the first place, there is no common approach which integrates proper translation in language learning, and therefore there is no strong evidence of how such practice may benefit learners. A few students (14) even revealed stronger ambivalence by ticking the more negative response that the use of L1 is a hindrance in learning L2.This may have been due to the fact that the participants, who were prospective EFL teachers, have been mainly exposed to theories which favor L2 monolingual approaches, which often suppress the use of L1.

Students' Perceptions of Whether or Not the L1-L2 Translation Course Should Be Retained
Students also gave positive responses to the question if the Indonesian (L1) into English (L2) translation course should be retained in the program (Statement 8). This item is very significant because the language program was considering the possibility of dropping the subject and incorporating it in some other courses. Against this idea, almost all respondents believed that the translation course should be retained in the program, hence including students who were critical of translation as an explicit strategy in writing courses (S6). When asked for their reason(s), the students said that the training should be retained because (1) translation may improve their L2 proficiency in general, (2) it may improve their writing in L2, or because (3) it is an important skill. This finding is consistent with their answers on the perceived benefits of translation as a skill, especially Statement 1 to 3, which elicited positive responses from almost all of the participants. This means there is a strong perceived advantage of translation as a skill among the participants, although they had different views regarding the manner of its application as a strategy in writing.

The Role of GPA in Students' Perceptions of Translation in Language Learning/Teaching
This factor analysis will only examine the first five items of the questionnaire (S1-5), which are concerned about students' perception of the advantages of translation skill towards their learning of English as an L2. Statements 6-7 are excluded because they make the Cronbach's Alpha low. 2 The coefficient, which indicates the internal consistency of the questionnaire, is an important measure for a valid quantitative examination, although it does not determine the validity of the questionnaire items in a qualitative analysis. With the exclusion, the value of Cronbach's Alpha increased to 0.718, which is higher than the minimum acceptable value of 0.7 to validly measure the independent variable (perception). However, two subsequent tests, first, using ANOVA and, second, using Linear Regression, did not detect GPA to be a factor that significantly affected the participants' responses towards items 1-5. The significance value of the GPA variable was 0.332 by ANOVA, and 0.721 by Linear Regression.
There was a possibility that students' GPA was affected by other factors (gender, university year and the parallel class the students attended for the translation course). However, including these controlling variables into the multiple linear regressions also did not change the result (see Table 6). GPA (measured by the actual values of the variable of GPA and GPA group) still did not have a discernible effect on students' perception. The significance values of GPA are 0.851 and 0.976 (well above the threshold value of 0.05), implying that GPA is not significantly associated with perception. Similarly, we also found that the significance values of GPA group are above the threshold value of 0.05.  Table 4 Results of ANOVA testone-way perception GPA group  Table 6 Results of multiple regression analysis Note: the first and third columns use actual GPA value as the main independent variable. The second and fourth columns use GPA group as the main independent variable. Column 1 and 2 add ln_availability as the independent variable. The numbers in brackets are significance values with * p<0.1, ** p<0.05, and *** p<0.01 respectively.
With respect to the participants' responses to S1-S5, this finding means that the participants, regardless of their level of English competence, believed that English is helpful for their language learning. This contradicts the common assumption that translation only benefits students with low L2 competence. As mentioned earlier, although more competent EFL students do not seem to rely on translation to express themselves when dealing with more familiar language routines and topics, they may have to resort to translation to express more complex ideas or when dealing with knowledge they have acquired in L1.
This study reveals a very positive perception of the respondents towards translation as a skill in language learning. The students, regardless of their GPA, believed that it enhances their language learning in different areas of language and productive skills. Contrary to what is commonly assumed, most of the students believed that translation skill may also assist them in speaking in L2. Despite some mixed feelings towards the use of translation strategy in their language learning/ teaching, there was an overwhelming support for the L1-L2 translation course to be kept in their language program. This is a prominent finding because it suggests the students' strong desire to be able to better learn translation even though it is not an intended career path for the program's graduates. Presumably, this is because they believe that translation may support their language learning. More research is necessary to confirm this conclusion and to find out if a proper independent translation training, given in the right dose, may indeed enhance students' foreign language learning regardless of their L2 proficiency. Apart from this, this study overall demonstrates a very significant perceived importance of translation as a skill in language learning. This further challenges the idea that translation is "dangerous" for L2 acquisition.