e-Journal of Linguistics


Available online at https://ojs.unud.ac.id/index.php/eol/index

Vol. 16, No. 2, July 2022, pages: 204--2012

Print ISSN: 2541-5514 Online ISSN: 2442-7586

https://doi.org/10.24843/e-jl.2022.v16.i02.p06

Using Translog To Analyze Self-Corrections in Translating Advertisement Texts by Professional Translators

1Clara Brigita Silaban

Universitas Sumatera Utara, Medan, Indonesia, [email protected]

2Rudy Sofyan

Universitas Sumatera Utara, Medan, Indonesia, [email protected]

2Dian Marisha Putri

Universitas Sumatera Utara, Medan, Indonesia, [email protected]

Article info

Received Date: 14 Desember 2021

Accepted Date: 30 Desember 2021

Published Date: 31 July 2022


Keywords:*

Advertisement Text, SelfCorrection, Translation Process, Translog


Abstract*

This study aims to determine the method and type of selfcorrections done by two professional translators in translating texts from English into bahasa Indonesia. This research used a descriptive qualitative method. The data were the recording of translating advertisement texts from English into bahasa Indonesia process using Translog II and Camtasia Studio 8. The findings are (1) Meaning Correction is the most frequent type of self-correction with the frequency of 36.2%; (2) Return Correction is never used by the professional translators; (3) The professional translators apply the multidirectional method in doing self-corrections. Finally, it can be concluded that selfcorrection is the crucial part that the translator should not skip in producing a better quality translation. This research also proves that translators cannot rely entirely on Google Translate as it cannot ultimately deliver the meaning from the source text (ST) into the target text (TT). Consequently, the translation provided by Google Translate needs to be revised.

  • 1.    Introduction

Nowadays, translation plays an essential role because it is both a field of study and a means of communication (Sofyan & Rosa, 2020: 3). Language is one of the communication media, and through language, translation has a relationship with communication. It helps people communicate their minds, even bridges the different cultures worldwide. The translation is a process of transferring meaning from the source language (SL) into the target language (TL); as Catford (1965: 20) says that translation may be defined as the replacement of textual material in one language, i.e., the source language (SL), by equivalent textual material in another language, i.e., the target language (TL). Since the study of translation has been developed, translation is focused not only on the product but also on the process. The translation process tries to access the black box of the translators and see what they do during the translation process (Kussmaul & Trikkonen, 1995: 178).

There have been many pieces of research conducted to see the translation process. In the 1980s, the Think Aloud Protocol (TAP) was introduced to understand the mental process of the translator better. This happened because translation is a cognitive process involving creativity (Kussmaul, 2000). Following TAP, in the 1990s, the keystroke logging program, Translog, was developed to obtain keyboard activities during the translation process. Studies on the translation process are continued about mental processes and typical reading, learning styles, and management resources that can be analyzed through the keylogging program. When translating a text, a translator can't do it without any revision. Robert (2008: 5) states that the term 'revision' refers either to the process of revising one's translation or to the process of revising another's translation. Based on that statement, revision is a process of giving correction to a translation text, in this case, done by the translator him/herself. This process is also called self-correction or self-revision.

Self-corrections or self-revisions are a process that cannot be skipped by translators, whether they are student translators or professional translators (Rosa, 2017). Moreover, many errors should be fixed if they use technology such as Google Translate to help them in the drafting stage. Otherwise, self-correction is not a sign of lousy process translation. Even previous research shows that a professional translator spends more time on self-correction than the student translators (Kunzli, 2006).

Concerning the role of self-correction in the translation process, this article aims to analyze self-corrections done by professional translators. The two respondents are two professional translators, full members of HPI (Association of Indonesian Translators). They were asked to revise English text translated by Google Translate into Indonesian, and the translation process is recorded using Translog. By observing the self-correction process, student translators can improve their skills in translating. Last but not least, self-correction is very important since it can improve the quality of the translation product.

  • 2.    Research Methods

This research was conducted using a descriptive qualitative method, which utilizes qualitative data described descriptively in words rather than in numbers. The data were collected using several instruments. Translog II was used to collect the first data by recording all activities on the keyboard; then, the results were presented in the log file form. The second instrument, screen recording Camtasia Studio 8, was used to collect online activities during translation.

The researcher chose two English texts to be translated into bahasa Indonesia. The first text (313 words) was from the Axe page and the second text (279 words) was from the Rexona page. Both texts were taken from Unilever's global company website https://www.unilever.com/. The participant were two professional translators who are full members of HPI (Association of Indonesian Translators), and they were asked to revise bahasa Indonesia texts translated by Google Translate.

The data from Translog II and Camtasia Studio 8 were analyzed qualitatively to find out the types of self-corrections and self-correction methods during the translation process, especially in the post-drafting phase. After they finished their works, the file was saved in an XML document. Then, the file would be opened using the Translog II supervisor to see the linear view of the translation. The linear view allowed the researchers to see the entire process of the translation. First, the researchers identified the methods and types of self-correction used by the professional translators and then classified them. The researchers also counted the type and method of self-corrections most frequently used. Camtasia Studio 8 was used to support the data

that could not be detected on the keyboard, such as types of online resources used by the professional translators, the websites they opened, and their activities during the pauses.

  • 3.    Discussions

Due to the research code ethics, the translators' identification would be kept confidential and respectively. In this research, they were called Translator R and Translator T. They did the translation process in the Translog worksheet, and all keyboard activities were recorded. They spent the different duration of time to finish their works. Translator R spent 30.40 minutes translating the first text and 19 minutes translating the second. Meanwhile, Translator T spent 19.81 minutes translating the first text and 30.10 minutes translating the second text. All keyboard activities were recorded using Translog, and they can be observed with the help of linear view.

  • 3.1    Method of Self-Corrections

Related to the sequences of self-corrections, the professional translators used a similar method. Translator R did self-correction multidirectional. He decided what parts needed to be revised, whether it was started from the first paragraph, second paragraph, or last paragraph. For example, in Text 1, Translator R began revising the first sentence draft before coming to the second text. But suddenly, he edited the title and back to his first revision until the whole text was completely revised. In Text 2, after he corrected the title, he continued revising the first sentence, the second sentence until the last sentence. Still, suddenly he returned to the other part that he did not change at first and corrected it. He also recorrected the part that he already fixed in the first. Like Translator R, Translator T also applied a multidirectional method of self-correction. He did it zig-zag, jumped to the second sentence, to the third sentence, and then back to the first sentence. He did it randomly in both texts, Text 1 and Text 2.

  • 3.2    Types of Self-Corrections

In this research, the type of self-correction followed the classification of self-corrections proposed by Malkiel (2009). In addition, another type of self-correction found by Sofyan (2016), named word addition (WA), was also used to complete Malkiel’s (2009) classification. Based on the data analysis, it was found that there were seven (7) types of self-corrections made by the professional translators, including word substitution (WS), word deletion (WD), meaning correction (M), word addition (WA), capitalization (C), Grammar (G), and spelling correction (S). They did 262 times of self-correction, which were divided into seven types. The only type of self-correction they did not use was return correction (R). The types and frequency of selfcorrection are presented in Table 1.

Table 1 Types and Frequency of Self-Corrections

No

Types of Self-Correction

Translator R

Translator T

Total

%

T1

T2

T1

T2

1

Word Substitution (WS)

16

5

21

19

61

23.2

2

Meaning Correction (M)

15

13

39

28

95

36.2

3

Word Deletion (WD)

12

13

13

15

53

20.2

4

Word Addition (WA)

4

2

5

1

12

4.5

5

Capitalization (C)

4

5

8

5

22

8.3

6

Grammar (G)

4

1

6

6

17

6.4

7

Spelling (S)

1

-

1

-

2

0.8

8

Return (R)

-

-

-

-

-

-

Total

56

39

93

74

262

100%

95

167

Table 1 shows that meaning correction (M) is the type of self-corrections most frequently used by the translators at 36.2%. The other types of self-corrections include WS 61 times (23.2%), WD 53 times (20.2%), WA 12 times (4.5%), C 22 times (8.3%), G 17 times (6.4%), and S 2 times (0.8%). Return correction is the only type that the translators do not use. In both texts, the translators often corrected in terms of meaning correction. Translator R did 28 times out of the total 95 times of his self-corrections, while Translator T did 67 times out of the total 167 times of self-corrections. This happened because the text was a kind of advertisement, and the translator should correct the meaning to make it more acceptable in TT and readable. This was especially done by translators when they added new words that did not exist in the ST to make the translation more natural. It can be seen through the example below:

Source Text : deodorants which go on completely clear, eliminating the white marks aerosols and sticks can leave on dark clothes.

First Draft : deodoran yang benar-benar bening, menghilangkan noda putih yang dapat ditinggalkan aerosol dan stik pada pakaian gelap.

Final Draft : deodoran yang benar-benar bening menghilangkan noda putih akibat aerosol dan stik yang tertinggal pada pakaian berwarna gelap.

From that data, we can see that the translator corrected the meaning by adding new words. First, he added ‘akibat’ to his final draft in order to give the cause of ‘white marks’ or noda putih. He also added the word ‘berwarna’ to emphasize the meaning of ‘dark cloth’. By adding those words, the translation he made was more natural.

Furthermore, the translators corrected the meaning due to the unclear or awkward meaning made by Google Translate. One of the criteria of a good translation is the meaning from the ST can be conveyed to the TT accurately, so when the reader reads it, it does not feel like they read a translation product. While in the translation provided by Google Translate, there were some meaning distortion and inappropriateness in the TT. That is why the translators decided to replace it with the most appropriate meaning. This can be seen when the translator replaced the meaning. For instance, he changed the word ‘meraih’ in ‘reach Axe’ became ‘memilih.’ The word ‘memilih’ was more appropriate for the context of this sentence. This also happened when the translator changed the meaning ‘smell their most attractive’ translated by Google Translate

‘mencium aroma paling menari mereka’ became ‘wangi sepanjang masa’. The translation made by Google Translate cannot be well understood in the TT, that is why he decided to change it, and his translation became easier to understand. Other meaning corrections recorded by the log include ‘dan ini tentang hubungan’ became ‘ditentukan oleh ikatan.’, ‘untuk menyadarkan’ became ‘ingin menyadarkan’, ‘orang-orang’ became ‘para pria’, ‘atau’ became ‘ataupun’ ‘dalam dua hal’ became ‘dalam kedua kondisi tersebut.’.

The second most frequent type of self-corrections made by the translators is Substitution. They did Substitution 61 times or 23.2%. This occurs because many words in the translation made by Google Translate are replaced. The words may be acceptable in TT, but they chose the most appropriate one. Word substitution can be seen in the example below;

Source Text  : We’ve helped guys look, feel, and smell their most attractive

First Draft   : Kami telah membantu para pria melihat, merasakan, dan mencium

aroma paling menarik mereka.

Final Draft : Kami telah membantu para pria untuk tampil, merasa, dan beraroma sangat menarik.

The word substitution can be found when the translator replaced the word ‘melihat’ became ‘tampil’ translated from ‘look’. This was done because ‘melihat’ is less appropriate in the target text. Although ‘melihat’ and ‘tampil’ are acceptable in TT, the translator should fit the meaning with the context of the sentence. In this sentence context, ‘look’ refers to the guy’s appearance, not as a verb to gaze or stare at something. For this reason, the translators chose ‘tampil’ to deliver the meaning of ‘look’. Another example included when the translators chose ‘kami’ instead of ‘kita’, translated from the SL ‘we’, ‘bereveolusi’ became ‘berubah’, equivalent with the ST ‘evolved’. The translators also made self-preference by choosing ‘para lelaki’ to translate the ST ‘guys’, which was translated ‘para pria’ by Google Translate and sometimes ‘kaum adam’.

Another frequent type of self-corrections done by the translators was Word Deletion (WD) as many as 53 times. The translators did the deletion due to some reason. First, deleted words were needless, and although they were deleted, the meaning from the ST into the TT still can be kept. For example, he deleted the word ‘namun’ (ST: but) in the phrase ‘namun terlepas dari itu’, as seen in the data below.

Source Text  : But despite that, many guys don’t feel comfortable being themselves.

First Draft    : Namun terlepas dari situ, banyak pria merasa tidak nyaman menjadi diri

sendiri.

Final Draft : Terlepas dari itu, banyak pria merasa tidak nyaman menjadi diri sendiri.

In that sentence, the word ‘namun’ was deleted, but it did not change the sentence's meaning. The use of ‘namun’ was not too important, which is why the translator decided to start his sentence with ‘terlepas’. Another example was when the translator deleted the phrase ‘di seluruh dunia’ (SL: across the globe). Those words were not needed because the following phrases represented their meaning. It can be found in the phrase ‘di lebih dari 90 negara’ (SL: in over 90 countries). His final draft became ‘para pria di lebih dari 90 negara memilih Axe’, translated from ‘Men across the globe in over 90 countries reach for Axe’. It was translated by Google Translate became ‘Pria di seluruh dunia di lebih dari 90 negara meraih Ax”.

The second reason for Word Deletion was because of the other type of corrections. It can be seen in the data below:

Source Text  : We’ve helped guys look, feel, and smell their most attractive

First Draft    : Kami telah membantu para pria melihat, merasakan, dan mencium

aroma paling menarik mereka

Final Draft : Kami telah membantu para pria untuk tampil, merasa, dan beraroma sangat menarik.

First, the translator deleted ‘mencium’ translated by Google Translate from ‘smell’. This deletion happened because he changed the structure, in this case, part of speech, where ‘aroma’ functioned as a noun became ‘beraroma’ worked as a verb. After he changed the structure, the word ‘mencium’ was no longer needed because it had already been replaced by ‘beraroma’. From the data above, the translator also deleted the word ‘mereka’ used in both the source text and the first draft provided by Google Translate. Nevertheless, the deletion did not change the meaning. Word deletion can also happen to the affix level. For instance, ‘merasakan’ (SL: feel) was changed to ‘merasa’. The suffix ‘kan’ was deleted to keep the meaning of the translation product.

Following the Word Substitution, the translators' next type of self-correction frequently used was Capitalization. The Capitalization was done when the translators did word addition or word deletion, as shown in the example below:

Source Text  : Men across the globe in over 90 countries reach for Axe

First Draft    : Pria di seluruh dunia di lebih dari 90 negara meraih Ax

Final Draft : Para pria di lebih dari 90 negara memilih Axe

The capitalization happened to the word ‘para’ in the data above. This was done because he added the word ‘para’ to translate ‘guys’, which was translated ‘pria’ by Google Translate. This addition certainly changed the capital letter ‘P’ for the word ‘para’ since it started the sentence. The capitalization also can be found when the translator did word deletion. For instance, he deleted the word ‘namun’ in the phrase ‘terlepas dari itu’. When the translator deleted the word ‘namun’, the sentence will certainly be started with the word ‘terlepas,' and ‘T’ should be written in a capital letter.

The translators less frequently used the other types of self-correction. They are grammar, word addition, and spelling. In terms of grammar correction, the translators corrected the structure by changing part of speech, such as ‘aroma’ (noun) became ‘beraroma’ (verb); changing noun clause became noun phrase, such as ‘apa yang unik tentang mereka’ became ‘keunikan mereka’ from the ST ‘what’s unique about them’. Another example of grammar correction was when the translator moved the ST object to become the TT's subject. The first draft by Google Translate was ‘Rexona tidak akan mengecewakan anda’ from ST ‘rexona won’t let you down’. Then the translator revised it to become ‘anda pasti tidak akan kecewa dengan Rexona.’. In terms of word addition (WA), the translators added a word that existed in the source text but was not translated by Google Translate. For instance, in the ST, it is written ‘to collect empty aerosol’. The translation provided by Google Translate was ‘mengumpulkan aerosol kosong,' and this was replaced by ‘untuk mengumpulkan kaleng aerosol’ in the translator’s final draft. There was also word addition in translating the word ‘guys’ became ‘para pria’, which was translated by Google Translate with ‘pria’. The last self-correction made by translators was spelling correction. There was only one spelling correction in the word ‘Axe’. Google Translate cannot translate this word correctly, and it resulted in the wrong spelling that was ‘Ax’. This

translation was obviously incorrect since ‘Axe’ is the product's name and should be written as ‘Axe’ too.

  • 3.3    The use of Online Resource in Self-Correction

From the recording of Camtasia Studio 8, it was found that both translators visited some websites to find the most equivalent word that they could use while doing self-corrections. For instance, Translator R opened Google to look for 'stereotip' meaning from ST 'find stereotypically ‘manly’ guys more attractive’ in correcting ‘menganggap pria stereotip jantan lebih menarik’. This was done because the translator wanted to ensure that his translation was equivalent and had the same meaning as the one in TT. In this case, he opened the KKBI website and looked for the 'stereotip' meaning. But after Translator R visited the website, he decided not to correct the translation provided by Google Translate in the first draft because he thought it had already been equivalent. In translating Text 1, Translator R only visited Google once, and he did selfcorrection based on his knowledge and experience. In text 2, Translator R also did not use many online resources, and he only visited Google three times.

Unlike Translator R, Translator T often visited Google to find the most appropriate and acceptable word for his self-correction. For example, he self-corrected ‘melihat’ from ST ‘looked’ became ‘terlihat makin ganteng’. After he corrected the meaning by adding other new words, he opened Google and typed 'Axe terlihat makin ganteng’. He did it because he wanted to see whether his correction was equivalent with ST and acceptable in TT. In addition, Translator T visited Google five times in self-correcting Text 1 and 8 times in self-correcting Text 2. He also opened Google Translate to find the most appropriate vocabulary for his self-corrections. This was done when he looked for the meaning of ‘judge’ from ST ‘for fear of being judged and labelled’. After Translator T found the meaning from Google, he corrected ‘dihakimi’ provided by Google Translate to become ‘dikritik’. Although only a few use of the online resource by professional translators, they still need it sometimes. In addition, this proved that the knowledge and experience of both translators are not doubted in the translation field since they can correct the texts without using many online resources.

  • 4.    Novelties

Based on the analysis, it was found that there were some types of self-correction by Kourouni (2012) applied by professional translators in doing self-correction. This proves that self-correction is one of the crucial parts of the translation process and cannot be missed by translators.

  • 5.    Conclusion

The findings show that self-correction is very important, and by knowing self-correction done by professional translators, student translators can improve their skills to produce a high-quality translation. It also proves that spending much time on self-correction does not mean that translators cannot make better quality translation products. Translog is beneficial to record all activities on the keyboard to see what and how translators can do self-correction. Besides, the use of screen recording such as Camtasia is also important to record the monitor and see what website or online resources visited by the translator to help them in doing self-correction. Finally, it is crucial to note that Google Translate cannot be used as a reference to produce a better quality of translation. It needs to be revised significantly in terms of meaning and word substitution.

References

Bell, R. T. (1991). Translation and translating: Theory and practice. New York: Longman Inc.

Carl, M., & Kay, M. (2011). Gazing and typing activities during translation: A comparative study of translation units of professional and student translators. Meta: Translators’ Journal, 56(4), 952-975.

Catford, J. C. (1965). A linguistic theory of translation. Oxford: Oxford University.

Jakobsen, A. L., & Schou, L. (1999). Translog documentation. In G. Hansen (Ed.), Probing the process in translation: Methods and results (pp. 149-184). Copenhagen: Samfundslitteratur.

Kevin. (2017). Dampak swa-koreksi pada proses terjemahan Mandarin-Indonesia seni pertunjukan komedi Tiongkok Xiangsheng (Unpublished Bachelor Thesis). Medan: Universitas Sumatera Utara.

Kourouni, K. (2012). Translating under time constraints in an undergraduate context: A study of students’ products, processes and learning styles (Unpublished Doctoral Thesis). Tarragona: Universitat Rovira I Virgili.

Kunzli, A. (2006). Translation revision - A study of the performance of ten professional translators revising a technical text. In M. Gotti & S. Sarcevic (Eds.), Insights into specialized translation (pp. 195-214). Bern/Frankfurt: Peter Lang.

Malkiel, B. (2009). From Antonia to my Antonia: Tracking self-corrections with Translog. In S. Göpferich, A. L. Jakobsen & I. M. Mees (Eds.), Behind the mind: Methods, models, and results in translation process research (pp. 149-166). Copenhagen: Samfundslitteratur Press.

Mossop, B. (2001). Revising and editing for translators. Manchester, UK; Northampton, MA: St. Jerome.

Muchtar, M. (2013). Translation: Theory, practice, and study. Medan: Bartong Jaya.

Nababan, M. R. (1999). Teori menerjemah bahasa Inggris. Pustaka Pelajar.

Newmark, P. (1988). A textbook of translation (Vol. 66). New York: Prentice-Hall.

Robert, I. (2008). Translation revision procedures: An explorative study. In P. Boulogne (Ed.), Translation and Its Others: Selected Papers of the CETRA Research Seminar in Translation Studies 2007, pp. 1-22.

Rosa, R. N. (2017). An analysis on translation and translating: SFL language metafunctions in the translation of student and professional translators (Unpublished Doctoral Dissertation). Medan: Universitas Sumatera Utara.

Sofyan, R. (2016). Translation process and translation quality: A study of Indonesian student translators (Unpublished Doctoral Dissertation). Medan: Universitas Sumatera Utara

Sofyan, R., & Rosa, R. N. (2015). Using Translog to investigate self-corrections in translation: A pilot study in translation process. Proceedings of the 15th International Conference on Translation (ICT15), pp. 500-508

Sofyan, R., & Rosa, R. N. (2020). Kajian terjemahan: Panduan praktik dan penelitian terjemahan. Tangerang: Mahara Publishing.


Biography of Authors

Clara Brigita Silaban S.S is a fresh graduate from the English Department, Faculty of Cultural Studies, Universitas Sumatera Utara. She is an English teacher in MEC Indonesia, Medan Branch.

Email: [email protected]

Rudy Sofyan was born in Kabanjahe on 13th November 1972. He got his bachelor's degree in 1996 from the English Department, Faculty of Letters, Universitas Sumatera Utara. He finished his master's in 2009 and doctoral degree in the Postgraduate School, Linguistics Department, Universitas Sumatera Utara in 2016. He works as an associate professor of Translation Studies at Universitas Sumatera Utara. He is a full member of the Association of Indonesian Translators.

Email:[email protected]

Dian Marisha Putri is a lecturer at the Faculty of Cultural Studies, Universitas Sumatera Utara. She accomplished her bachelor's degree in English Department Universitas Sumatera Utara. She also gained her master's degree from Linguistics Department Universitas Sumatera Utara.

Email: [email protected]