Types of Translation Errors from Indonesia Language into English in Pharmacy Journal Articles

-The study analyzed the types of errors based on ATA Error Categories version 2021 on the three submitted journal articles translated by lecturers of Clinical Pharmacy and Pharmacy Community of Akademi Farmasi Surabaya. It aimed to find the types of errors often made by pharmacy lecturers in translating the articles from the Indonesian language into English so that the results of this study can help them identify and avoid errors in translation. The study used the descriptive qualitative, using the content analysis method. The objects of this study were three journal articles in the Indonesian language (ST) and English (TT). The data focused on the translation errors based on ATA Error Categories in the words or phrases level. The results showed there were 14 types of errors found, namely: literalness (31.05%), usage (22.22%), punctuation (7.84%), terminology (7.52%), verb form (6.86%), grammar (6.54%), capitalization (3.59%), parts of speech (3.59%), syntax (3.27%), style (2.94%), register (2.29%), omission (1.31%), cohesion (0.65%), and ambiguity (0.33%). The study recommended that translators have sufficient knowledge about the target language mechanics, produce clear renditions of the source text, and maintain the writing quality of the target text.


I. INTRODUCTION
Lecturers are professional educators at the university level who are also scientists.Not only are they obliged to educate, but a lecturer is also required to dedicate themself in terms of research and community service.It is in line with the sound of Law no. 12 of 2012, Article 1 Paragraph 9 concerning obligations of the Tri Dharma of Higher Education to organize Education, Research, and Community Service (Law on Higher Education No. 12, 2012).
Referring to research standard, it is stated that research results pointed at both development of science and technology and an enhancement of the social welfare and the national competitiveness in the form of academic research output (Kementrian Pendidikan Tinggi, Kebudayaan, 2021).Therefore, lecturers are demanded to present the results of the research they have conducted in academic conferences or publish them in proceedings or journals, nationally or internationally.
Submitting papers in journals or conferences implies that authors must follow and obey the author's guidelines.In other words, every journal or proceeding has its writing rule, which may distinguish one another.The text of articles submitted can be in Indonesian or English, based on the author guidelines of the proceedings or the journals.Therefore, to submit their publication in English version, Indonesian lecturers must have the ability either write or translate their articles into good English, either translating manually and utilizing Machine Translation (MT) tools or Computer-Assisted Translation (CAT) tools (often mentioned as Translation Memory/ MT) (Apriliana et al., 2016;Han, 2020;Zaretskaya et al., 2015).In this context, lecturers act as translators involved in communicating messages (Liddicoat, 2016) from one language to another.Liddicoat (2016) further explained that a translator plays a vital role in being an intercultural mediator (Liddicoat, 2015) who transfers meanings and cultural context constructed in one language to readers who do not share the language and culture.Hence, translating activity is essential so that the messages contained in the target language text follow the meanings constructed in the source language.
In line with Catford (1965), a translation is a process of substituting equivalent content from one language (Source Language) into another language (Target Language).Translating is not an action that is done simply by translating the text literally, but the translation contains essential information that the author of source text wants to convey so "the target text must communicate the same meaning as the source text communicates" (Susini, 2020).The exact meanings in this term refer to not only what are provided explicitly in the text but also what lies implicitly through a context (Liddicoat, 2015).Bassnett (2005) stated that translation is not just changing the language but substitution the meaning from the source language (SL) to the target language (TL) by considering the linguistics and culture of the target language.The essence of translation lies in the meaning of the text.It is accordance with what was conveyed by Newmark (1988) that "translation is rendering the meaning of a text into another language in the way that the author intended the text".Therefore, a translation result is expected to be able to convey messages of the source text precisely and accurately.
In the other hand, Mounin (1963) via Newmark(1988) pointed out that the translation cannot simply reproduce or be the original.A translator must have a solid command of both languages to produce a professional translation result that the readers readily understand.Unfortunately, during the translation process, a translator often encounters various linguistic and non-linguistic issues.
As stated by Newmark (1988), several factors need to be considered by the translators, such as the individual style or idiolect of the SL author, the conventional grammatical and lexical usage for this type of text, content items referring specifically to the SL, the typical format of a text, the expectations of the putative readership, what is being described or reported, the independently of the SL text, the expectations of the readership, and the views and prejudices of the translator.Moreover, sound and sense, emphasis (word order) and naturalness (grammar), the figurative and the literal, neatness and comprehensiveness, concision and accuracy are additional tensions in translation.
A translation is not an easy process because every language contains its characteristics and culture (Darmayanti, 2020).To have a good product, a translator must overcome the translation process challenges.However, the issues that the translators often face can cause translation errors.However, there is a difference between the terms mistake and error.Goff-Kfouri (2004) stated that a mistake is typically viewed as a flaw in performance and is unsystematic.Still, an error is a gap in a student's knowledge and can be identified systematically (Soltani et al., 2020).
According to Koller (1979) via Dewi (2017), a translation error is inequality between source text (ST) and target text (TT) or the inadequacy of TT if it is viewed from the meaning equivalence of ST and TT.Williams (1989) explained more on major errors rather than minor errors.The examples of major errors as he mentioned are significant omission, inapprehensible jargon, or translation results that totally have dissimilar ideas with the ST.Moreover, Vinay & Darbelnet (1995) stated that translation errors occur when the translator ignores minor variations in the meaning of words considered interchangeable.However, errors can be detected in nearly any translation (Dewi, 2017).It is crucial to understand and classify them so that they can more easily identify the errors that may occur, particularly in the initial draft.
The study applied translation assessment from ATA (American Translation Association) Error Categories Version 2021 (Explanation of Error Categories -American Translators Association (ATA), n.d.) as claimed that assessment can give feedback and quality control to get measurable improvements (Doyle, 2003).Furthermore, Doyle described the benefits of ATA Framework: ATA's framework provides a ready-made, standardized, time-tested, and professionally recognized model for conducting theory-based, systematic, coherent, and consistent evaluations of student translations (Doyle, 2003).Meanwhile, ATA Error Categories Version 2021 assesses translation errors which are divided into three sections, namely target language mechanics, meaning transfer, and writing ability.The section 1 (target language mechanics) refers to errors that interfere one or more rules of the correct written forms of TT, such as grammar, spelling, and punctuation.The meaning transfer (section 2) refers to errors which impact on the TT readers' understanding of ST ideas/ meaning.The last section (writing quality) refers to errors that semantically correct but pragmatically still sound incorrect.
Some prior researches have analyzed translation errors.Wahyuningsih (2021) examined translation errors conducted by 5th semester students at STBA LIA who had translated a guideline of Handling Covid-19 using parameters proposed by Mossop (2014) and Rafsanjani (2020) analyzed the data using translation theories from Sager (1983).The differences between these two researches and this study were the parameters used to assess the translation errors.Nur Najibah Koman et al. ( 2019) classified translation errors using the ATA's Framework for Standardized Error Making, while this study used the newest ATA Error Categories Version 2021.Research conducted by Dewi (2017) analyzed translation and language errors using ATA Framework and LBI Bandscale.The errors analysis theory and the participants in Dewi (2017) and in this study were distinct so the results and findings were also dissimilar.In addition, Soltani et al. (2020) identified translation errors in several literary genres in the English-Persian pair of languages, namely prose, poetry, non-fiction, play text, and media text.The language pairs and the objects used in Soltani et al. (2020) were different, the results surely would be distinctive.
Moreover, some researches have focused on studying translation errors in students' translation practices.Research conducted by Kusumawati (2017) conducted research on the errors analysis of a translation text from English into Indonesia carried out by students of math education semester VI 2016/ 2017.Achmad Muhlisian (2018) in his article entitled analysis the misuse of the Japanese word in Indonesian-Japanese translations on students scientific works.Another case study of errors analysis on students translations of a textbook of Civil Engineering were conducted by Halim & A.R. (2016).The results of these previous research indicated differences with this research in terms of languages pair and the objects of the study.
Thereby, errors in translating the text of Indonesian journal articles into English often become obstacles for lecturers in translating their journal articles from source language to target language.Therefore, this study aimed to identify the types of translation errors and the frequency of translation errors occurrence in the three submitted journal articles by lecturers of Clinical Pharmacy and Pharmacy Community of Akademi Farmasi Surabaya in the hope that the results of this study can help translators identify and avoid errors in translation.

II. METHODS
The study used a descriptive qualitative using the content analysis method.This study's subjects were clinical pharmacy lecturers and the pharmacy community of Akademi Farmasi Surabaya that had submitted journal articles in English (TT) and had the ST in the Indonesian language.The objects of this study were three journal articles in the Indonesian language (Source Text) and in English (Target Language).
The data focused on the translation errors based on ATA Error Categories in the words or phrases level.The data was collected by comparing ST and TT to find the translation errors, then recorded in the table.The next step was identifying and categorizing the data based on ATA Error Categories.After classifying the data, it was analyzed further to identify the translation errors and calculated the frequency of translation errors occurrence.

III. RESULT AND DISCUSSION
The ATA Error Categories Version 2021 was used to determine the types of errors in the translations.Each type of translation errors based on three sections (target language mechanics, meaning transfer, and writing quality) of ATA Error Categories Version 2021 has been found in the articles, namely literalness, usage punctuation, terminology, verb form, grammar, capitalization, part of speech, syntax, style, register, omission, cohesion, and ambiguity.The frequency of translation errors occurrence found in the translated articles was summarized in the Table 1.The most frequent type of error found in this study is literalness, which accounts for 31.05%.Most of the sentences were translated word-forword, ignoring how words are strung together in the target language.As a result, the renditions were unidiomatic and challenging to be understood.The following is an example of a literal error found in the study.
ST: Pada karakteristik pengobatan yang diperoleh pasien mayoritas pengobatan dengan terapi tunggal.TT: On the characteristics of the treatment obtained by patients was the majority of treatment with single therapy.CT: The data on the characteristics of patient treatments showed that most of the participants received singledrug therapy.
Each word in the ST was translated separately, without considering the syntactical structures acceptable in the target language.As a result, the TT is incomprehensible.
The second most frequent type of error is usage, which appeared 68 times (22.22%).Usage errors in the articles occurred due to incorrect or missing articles, wrong choice or lack of prepositions, and incorrect collocations.For instance, the phrase "penderita hipertensi" was translated into "hypertension sufferers", which sounds unidiomatic in English.The more idiomatic translation is "people with hypertension".Below is an example of usage errors due to an incorrect preposition."From" should be used with "to" or "until" to indicate the beginning and ending point of an event in time.Punctuation errors appeared 24 times (7.84%) in the translated articles.The study found that the errors were caused by the incorrect use of commas, hyphens, and slashes.For instance, a punctuation error was discovered in the phrase "self reminder card"; a hyphen should be used as the prefix self is always hyphenated.The following is an example of punctuation errors due to the lack of a comma where it should be given to separate the two clauses.

ST: Sebelum uji validasi dilakukan uji
pre-test terhadap 30 responden … TT: Before the validation test pre-test was conducted … CT: Before the validation test, pre-test was conducted … Another type of error observed from the translated articles is terminology errors (7.52%), occurring due to the selection of words or phrases with inaccurate meanings used to translate the source text.For instance, "tahapan alih bahasa" was translated into "language transfer stages".The term "alih bahasa" has an equivalent term in English, that is "translation".Translation and language transfer have different meanings; thus, the correct rendition should be "translation process".Another example of terminology errors is the phrase "jenis kelamin", translated "gender".The English word with the source text's more correct and appropriate meaning is "sex" since it talks about male and female respondents.

VERB FORM
Comprising 6.54% of the total frequencies, grammar errors were found in the articles due to the violation of grammatical rules of the target language, such as lack of agreement between subject and verb, incorrect verb tenses, incorrect uses of singular or plural nouns, etc.An example of grammar errors due to the lack of agreement between subject and verb can be seen below.The subject and the verb in the following example do not agree with one another because the subject is plural, but the verb is singular.Despite the fact that the source text did not use capitalization correctly to write a proper name, the translator should have improved it.Thus, in the translation, the first letter of each word should be capitalized.
Part of speech errors appeared 11 times (3.59%), occurring due to incorrect grammatical forms.For example, "pasien hipertensi" was translated into "hypertension patients".Instead of a noun, the premodifier should be an adjective because it functions to describe a noun.Thus, the better rendition would be "hypertensive patients".Another example is the translation of the phrase "keberhasilan pengobatan hipertensi" into "the successful of hypertension treatment".The preposition "of" should be preceded by a noun or a noun phrase.However, the word "successful" is an adjective and hence cannot be put before the preposition "of".
The errors due to syntactic deviations appeared ten times (3.27%).In this study, syntax errors occurred because of improper modification, lack of parallelism, run-on structure, and unnatural word order.An example of syntax errors is illustrated below.

TT:
Instrument items are said to be reliable and can be accepted … CT: Instrument items are considered to be reliable and acceptable … The construction within the above sentence is not parallel because different forms of words are used.Therefore, to improve the readability and effectiveness of the sentence, the translator should have used a parallel construction.
Style errors were also observed nine times (2.94%), occurring due to the lack of using passive voice and shortest expressions, as they are a preferred style in writing scientific papers.For example, the phrase "dapat mengendalikan" was translated into "can control", which is a pretty long phrase.The expression "is able" can be replaced with the modal "can".Thus, a better translation should be "can control".Another example of style errors can be seen below.

ST: Pemilihan sampel dalam penelitian ini menggunakan metode purposive sampling … TT:
The sample selection in this study used the purposive sampling method … CT: Purposive sampling method was used to select the sample of the study ...
A clause above contains a style error due to an unsuitable grammatical structure.Passive voice is preferred in writing scientific papers, particularly the research method section.Thus, the more appropriate rendition should have been written in passive voice.
Occurring seven times (2.29%), register errors in this study were caused by using less formal words, which are inappropriate if used in a scientific paper.The following is an example of registry errors.
ST: tidak menyadari sebagai penderita hipertensi … TT: unaware that they got hypertension … CT: unaware that they had hypertension … The verb "got" is generally considered informal and used more commonly in speaking.Thus, the more formal verb "had" should be used instead.
The omission also appeared four times (1.31%) due to some elements left untranslated in TT.As a result, the meaning of ST was not transferred completely into TT, so there was a distortion of meaning.An example of omission errors is shown in the following.

ST: … diketahui kuesioner HK-LS versi
bahasa Indonesia valid dan reliable digunakan untuk populasi target.TT: … that all HK-LS statement items are valid.CT: … that the Indonesian version of all HK-LS statement items are valid and reliable used in target population.
The phrase of "versi bahasa Indonesia", "reliable", and "digunakan untuk populasi target" were not translated into TT.Not only causes a distortion of meaning in TT, the omission of these words and phrases possibly changes the author's tone.
Cohesion was also found two times (0.65%) because of inappropriate or missing conjunctions.Thus, the text's readability is low, making it difficult to read and understand.The following is an example of cohesion errors.
ST: Sedangkan data tidak berpasangan (independen) … TT: While if the data were unpaired (independent) … CT: Meanwhile, if the data were not paired (independent) … Cohesion functions as links among some text parts and guides the readers to follow and understand the text easily.For instance, the conjunction "while" can be replaced with the conjunction "meanwhile" and then followed with a comma (,) for the punctuation.Thus, using punctuation comma after the conjunction can assist the reader for better understanding.
The minor error was ambiguity which was found once (0.33%).Ambiguity refers to ST or TT, leading to more than one interpretation.Ambiguity can occur at various levels.In this ambiguity, an error example appeared at the sentence level.The ambiguity error is displayed in the following.
ST: Sampel penelitian ini yaitu seluruh pasien dengan diagnosis Hipertensi mendapatkan terapi obat anti hipertensi yang sesuai dengan kriteria penelitian antara lain: TT: The sample of this study was all patients diagnosed hypertension and got antihypertensive drug therapy which was in accordance with the research criteria, namely: CT: The sample of this study was all patients diagnosed with hypertension, given antihypertensive drug therapy, and suited with the research criteria, namely: The ST text allows more than one interpretation.The phrase "which was in accordance with the research criteria" causes an ambiguity.The phrase possibly refers to "all patients diagnosed with hypertension" or "antihypertensive drug therapy".Therefore, the better translation is "suited with the research criteria".

DISCUSSION
The current study found that the three most common errors in the articles were literalness, usage, and punctuation, while the least standard error was ambiguity.According to the ATA Framework, literal errors are serious since they affect the meaning transfer, hindering readers in grasping facts or ideas in the source text.Usage errors, however, do not break grammar rules but damage the quality of writing.Meanwhile, grammar errors deal with the mechanics of the target language but do not seriously affect the meaning of the source text.
However, the results of this study do not support the previous research.Using the ATA Framework in her research, Dewi (2003) found that the most common errors committed by the respondents were grammatical and usage errors, while the least frequent errors were due to literalness.Also, another study conducted by Soltani, Nemati, & Yamini (2020) showed contrasting results.The translation errors with the highest frequencies were terminology, omission, and punctuation.
In contrast, the majority of errors in the current study occurred due to literalness, meaning that the source text was translated word for word without taking into account how words are tied together in the target language.Literal translation, however, is not always erroneous and inappropriate.In point of fact, this procedure is useful in translating scientific papers and hence should not be averted entirely.It is in line with the statement of Newmark (1988) "Literal translation is correct and must not be avoided if it secures referential and pragmatic equivalence to the original.Literal translation can be applied if the translator can maintain the accurate meanings of the source text both semantically and pragmatically.
In this study, literal translation causes errors that obscure or change the meanings of the source texts.A possible explanation for these results is the poorly written source texts translated wordfor-word into the target language.Many sentences in the source texts were found to contain missing commas, wrong choice of words, and poor syntax, causing difficulty reading and comprehending.
It can be assumed that literal translation should not be used to translate the source texts with poor qualities.Newmark (1988) argued that a good translator does not opt for word-for-word translation if the source text appears to be incorrect or poorly written.When confronted with this sort of situation, translators should rectify the errors in grammar, word choice, and other linguistic elements in the poorly written source text.It is consistent with what Nida ( 2001) suggested that it is frequently necessary for professional translators to make corrections on poorly written texts.
Usage errors had the second-highest frequency of occurrence in this study.Usage deals with the idiomaticness or naturalness of the renditions that can be achieved by using articles, prepositions, and collocations correctly and appropriately.The study results indicate that the translators are less knowledgeable of idiomatic expressions in the target language.Many renditions sound unnatural and inappropriate due to incorrect or missing articles, wrong choices of prepositions, and incorrect collocations.These results emphasize the importance of using idiomatic language, which aligns with Larson's assertion (1998) that a good translator will always aim at idiomatic translation.Seeking naturalness should always be a translator's goal.
One thing the translators could do to help them identify the correct collocations is read authentic materials as much as possible.In line with the statement of Armstrong ( 2005), good reading is closely linked to the translator's success in searching for the correct collocations.Reading authentic materials may help translators increase exposure to and improve their knowledge of idiomatic expressions in the target language.
The third most frequent errors found in the current study are punctuation.Based on the study's results, punctuation errors occurred due to the incorrect use of hyphens, commas, and slashes.
Indonesian and English have slightly different usage of punctuation marks, particularly hyphens.Hyphens are essential in English since they are used for various purposes.Hyphens are used when two or more words are joined to become a single modifier (e.g., my three-year-old son, the well-known singer).In some cases, hyphens connect a prefix and a root word (e.g., co-owner, self-esteem).
The study results suggest that the translators were careless or less knowledgeable about punctuation rules in the target language.It should become a concern for translators since punctuation has a significant role in writing, especially in conveying a message.It is in accord with what Newmark (1988) stated: punctuation shows a semantic relationship between sentences and clauses.

IV. CONCLUSION
The result of the study showed fourteen types of translation errors based on ATA Framework version 2021.The types of translation errors that occurred were literalness, usage, punctuation, terminology, verb form, grammar, capitalization, part of speech, syntax, style, register, omission, cohesion, and ambiguity.The two most prominent errors found were literalness and usage.Meanwhile, punctuation, terminology, verb form, and grammar occurred around the twenties.Capitalization and part of speech had the same frequency of occurrence.Syntax occurred ten times, and the others were under ten times of occurrence, with the least common error was ambiguity.
Therefore, it is suggested that 1) translators should have precise knowledge both on ST and TL and have the capability to transfer accurate and equivalent meanings semantically and pragmatically; 2) translators should be able to make corrections on poorly written texts; 3) translators should consider the naturalness of the TL in order to get high readability; and 4) translators should be conscientious of punctuation since it has a substantial role in writing.Therefore, the results of this study were hoped to give insights to the translators to be more cautious in translating so the types of errors that appeared in this study can be avoided.
ST: dilakukan pada bulan Desember 2019 sampai Februari 2020 TT: conducted in December 2019 to February 2020 CT: conducted from December 2019 to February 2020

ST:
Kelemahan penelitian ini adalah … TT: The weaknesses of this research was … CT: The weakness of this research was … Capitalization errors occurred 11 times (3.59%).The errors were caused by the incorrect usage of lower and upper case.Below is an example of capitalization errors found in the study.ST: … menurut riset kesehatan dasar nasional tahun 2018, … TT: … according to national basic health research in 2018, … CT: … according to the National Basic Health Research in 2018, …

Table 1 .
Types of Translation Errors Found in the Articles