INVESTIGATING DECLARATIVE KNOWLEDGE OF PROFESSIONAL TRANSLATORS

Declarative knowledge is a belief system of translators which underlies the practices of translating. It guides translators in decision-making process in an interlingual communication. Exploited simultaneously and effectively with operative knowledge it results in the production of a quality. In some cases, as evidenced in processoriented studies, there is a gap between the two to the extent that what translators know and believe about translation does not match what they do in practices.


Introduction
There is an interplay between processes of translation, translators as the mediating agents, and products of translation.They are closely related in the sense that when translators are engaged in translation processes, they are obviously producing products.Their competence in the translation process is determined, to a certain degree, by their backgrounds and knowledge about translation.Similarly, the quality of the products, to a large extent, is dependent on how well they can apply their knowledge of translation processes in translation practices.Their linguistic competence in the source language and the target language, their understandings of the subject matter, target readers, and quality of a translation also play a significant role in the performance of translation tasks.
To become a professional in translation, someone is required to possess declarative and operative or procedural knowledge (some experts call it translation competence).De-clarative knowledge refers to the knowledge about underlying concepts of translation, and operative or procedural knowledge refers to ability to produce a quality translation.In many cases, there is a gap between the declarative and operative knowledge.In other words,what is believed is not always applied in practices.Our understanding of the declarative knowledge is paramount important in the attempt to produce professional translators.It is by evidence, however, that research about the knowledge in question is rarely conducted in Indonesia.
Translation is a means of communication.It is "an intercultural verbal activity" (Lvovskaya, 2000: 28) which is needed when there is a communication gap between an author of the source language text and readers of the target language text.This kind of intercultural verbal activity requires the presence of a translator to eliminate the communication barrier.
Translating is a complex activity (Schaffner and Adab, 2000: viii).Richard (1953, as cited in Brislin, 1976: 1) even states that translating 'is probably the most complex type of event yet produced in the evolution of the cosmos'.One may ask what knowledge, skills and competences translators should have in order to perform the task effectively.
Competence is the underlying system of knowledge and skills that enable someone to do particular things.Thus, translation competence can be defined as "the underlying system of knowledge and skills needed to be able to translate" (PACTE, 2000: 100).In a similar vein, Shreve (1997: 120-121) states: Translation competence is a specialized form of communicative competence.It is about knowing about translation and about knowing how to do translation.It is about producing translations that are well formed, referentially accurate with respect to source texts, and socially appropriate in their cultural contexts .
It is widely accepted that translators should have knowledge to enable them to translate.They should have declarative knowledge (knowing what) and procedural knowledge (knowing how) (Schaffner and Adab, 2000;Anderson, 1983, as cited in PACTE, 2000).These two kinds of knowledge underlie competence, which is used as a cover term to encompass skills and expertise elements.
As a specialized form of communicative competence, translation competence must consist of some competences on which it is grounded.In relation to this, Neubert (2000: 6) identifies five qualitative parameters of translation competence: language competence, textual competence, subject competence, cultural competence, and transfer competence.These five qualitative parameters of translation competence are expanded as follows.(a) Language competence.Translators should be competent in source and tar-get languages.They should know the morphological, grammatical and lexical systems of the two languages.In addition, they should be aware of changes in lexical items within the source and target languages generally reflected in dictionaries or other references.(b) Textual competence.It is rarely found that translators work on isolated sentences.
In general, they deal with texts of various types.Therefore, they should be familiar with how sentences are combined into paragraphs, and paragraphs into a text.Depending on the domains of discourse they are translating, translators should be proficient in how the source and target language texts are structured.In short, "they must be sensitised to identify textual features in addition to linguistic ones" (Neubert, 2000: 8).(c) Subject competence.Competency in linguistic systems of the source and target languages and familiarity with the textual features of the source and target language text do not guarantee the production of a quality translation.Familiarity with the subject matter being translated is another important aspect.It should be noted, however, that being competent in the subject matter does not necessarily mean that translators must have highly specialist knowledge, "but they must know the ways and the means of how to access this when they need it" (Neubert, 2000: 9).Familiarity with the subject matter could enhance the comprehension process of the source language text, which in turn could affect the production process of the target language text.It also gives possible solutions to translators with regard to how unfamiliar technical terms should be rendered.(d) Cultural competence.There is a misunderstanding among those outside of trans-lation or even novice translators that cultural competence is required only in translation of literary texts.If we accept the idea that the production process of the target language text, whether it is academic, technical or literary, is culturally bound, cultural competence is inevitably needed.In addition, if we look at the role of a translator as "agent for affecting a symbiosis of the source culture and target culture at the linguistic level" (Mohanty, 1994: 28), it becomes apparent that translators must know source and target cultures.They have to be "biculturally competent" (Witte, 1994: 71).(e) Transfer competence.Transfer competence refers to "tactics and strategies of converting L1 texts into L2 texts" (Neubert, 2000: 10) While bilinguals may have the four competences described above, it is the transfer competence that distinguishes translators from other communicators (Neubert, 1994: 412).Neubert (2000: 10) argues that translation competence "is where translators are judged".He states further: Whatever they may boast about their knowledge, their amazing individual competences, their language skills and their multifarious erudition or their indepth specialists expertise, even their profound understanding of two or more cultures, all these competen-ces are feathers in the translators' cap.But if this excellent equipment is not match by the unique transfer competence to produce an adequate replica of an original they have failed.It is not enough to know about translating, it has to be done.
The range of knowledge, skills and competences expected of translators may seem overwhelming and is sometimes not recognized in the level of compensation for the work.The requirements described previously, how-ever, are necessary for translators to accomplish their primary task of producing a target language version of the source language text (Danks and Griffin in Danks at.al , 1997: 164).
It has been mentioned above that bilingual competencies underlie translation competence.It is probably the reason why some translation scholars consider bilingualism as an important aspect in any account of the development process of translation competence.Harris and Sherwood (1978) introduced the concept of natural translation, positing that bilinguals 'naturally' acquire an ability to translate in line with the development of their competencies in two languages.Toury (1984: 189-190) suggests that "bilinguals have an innate translation competence comprising bilingual and interlingual ability, as well as transfer competence" and considers bilingual competence as the foundation of translation competence .In addition, Toury sees that competence in two languages intersects and the point of intersection is the transfer competence, that is the ability to transfer texts.However, he does not believe that translation abilities are a necessary derivative of bilingualism (cited in Shreve, 1997: 121).
Lorscher (1986,1995) sees that natural translation is "a result of a translation ability evidenced by bilinguals communicating in real mediating situations" (cited in Shreve, 1997: 122).He distinguishes natural translation competence from the translation competence of second la-nguage learners, which is formed through didactic or formal instructional settings rather than real communication situations.In addition, he claims that a translation ability, which is acquired naturally is sense oriented while the translation skill of the second language learners is sign oriented.In Loscher's view, professional translation is "a developed form of natural translation" (Shreve, 1997: 122).
Harris and Sherwood's concept of natural translation, Toury's idea about the intersec-tion of competencies in both languages which results in translation competence, and Loscher's view of the different nature of translation competence acquired by bilinguals and second language learners are generally accepted.It is also widely recognized that bilingual competence is an ideal foundation of translation competence.However, the statement that professional translation evolves from natural translation is still debatable.A growing number of professional translations (which can be considered competent) are produced daily by professional translators who acquired their translation competence through didactic rather than real communication situations.
Different scholars suggest different ideas about how translation competence develops.Despite the differences, however, there is agreement in the literature that variability is inherent in translation.Shreve (1997: 125) states that "there is little evidence that professional translators translate identically".In a similar vein, Seguinot (1997: 104) states: Translators and people who study translation know that different text types require different approaches, and that different people can translate the same text in different ways.It is also clear that different levels of competence, familiarity with the material to be translated, as well as different interpretations of the nature of the assignment will lead to differences in processes and results .Seguinot (1997 : 126-127) identifies two general factors resulting in the variation of translation performance.The first factor is the variation in individual cognitive styles and the second factor is the variation in translation acquisition history.Shreve (1997 : 108-109) posits that variability in translation may result from the skill level of the translator and the use of different translation strategies .
The following main objectives are set up for the study: 1) to explore their knowledge and beliefs about the translation process and approach, 2) to reveal ways in which translators deal with problems in translation process, 3) to find out translators' beliefs about translation competencies, 4) ways they identify target readers, and 5) to find out criteria translators use to define quality of a translation.

Research Method 2.1 Criteria for Research Participant Selection
Research participants were 16 Indonesian translators.The sample was confined to translators who translated academic texts.They had to be professional and experienced translators as defined by the publication of their translation.Nevertheless, the 16 translators could be considered a "purposive sample" (Schumacher and McMillan, 1993: 378) who provided sufficient data to allow a closer look at translation process and product.In addition, they are typical (in terms of background factors) of the larger sample surveyed.
The underlying reason for focusing on experienced and professional Indonesian translators who translate academic texts as research participants is three-fold.Firstly, in their working practices, professional translators are generally involved in a decision-making process to solve problems with regard to the transfer of messages of a source language text into a target language.Thus, investigating the working practices of professional translators could provide novice Indonesian translators with useful information about the realization of decision-making process.Secondly, Indonesian translators with these characteristics play an important role in the transfer of science and technology by translating textbooks in the area from foreign languages into Indonesian.Results of the investigation could give useful inputs for publishers and those engaged in translation, teaching and training in the attempt to improve the professional status of translation in Indonesia.In terms of research on transla-tion, other researchers could use results of the study as starting points to investigate further the performance and behavior of professional Indonesian translators.Thirdly, as they are considered 'professionals', it is expected that their translation practice is founded on theoretical knowledge or strategies.

Method of Data Collection
The study investigated research participants' knowledge and beliefs about the translation process, use of particular translation approaches, ways of identifying target readers and criteria used to judge the quality of a good translation.It also examined types of dictionary that research participants believed to be useful when translating academic texts from English into Indonesian.The use of materials such as dictionaries and reference by professional translators can be 'valuable indicators of subtle qualitative aspects of translator performance' (Hatim, 2001: 160).
Considering the objectives of the second phase of study and the nature of data that would be obtained from research participants, it was considered that the use of interview would be appropriate.Cohen and Manion (1980: 292) state, for instance, that interview 'allows for greater depth than is the case with other methods of data collection' .Another advantage of interview is that its high response rate, 'which makes the data more representative than data solicited through a mail questionnaire' (Burns, 1997: 484).
The interview design used in this study had an iterative, flexible and continuous design (Rubin and Rubin, 1995: 43).Instead of preparing a number of questions beforehand, the researcher provided topics and themes related to the objectives of the study, which were further explored in the interview with the research participants.The researcher also followed up responses given by research participants on the questionnaire.Interviews were recorded on a tape recorder.

Translators' Knowledge and Beliefs about Translation Process and Approach
Translators' knowledge and beliefs about the translation process here refer to their understanding and perception of what they usually do to the source language text before and during translation.These two aspects are closely related in the sense that their perception about the translation process rests on their knowledge about it and is believed to have an effect on the ways they approach the task of translation (Baker, 1992;Newmark, 1981).It also examines other aspects relevant to their knowledge and beliefs about the translation process.These include ways in which they solve problems with lexical items, technical terms, and sentence complexity, and perception of the importance of knowledge of translation theory in translation practices, and beliefs about competence that translators should have to perform their task effectively.
The interview began with a personal introduction with the translators.After this personal introduction, the researcher asked them a question about what considerations they took into account before they accepted a translation task.Their responses to this question are summarized as follows.The first thing that they considered was whether they had the ability to do the task.If they thought that they did not have the ability to translate a source language text in question, based on their comprehension of the text, they decided not to accept the task.This often occurred particularly if the text was outside of their subject-area expertise.On the other hand, if they thought that they would be able to translate it into the target language, they considered accepting the task.Their other consideration was whether the commissioner gave them enough time to complete the task.
This question: "What do you know about the translation process?" was then asked to each of the sixteen translators.The translators with academic and/or vocational translation training backgrounds responded that the translation process consisted of the analysis of the source language text, transfer of the messages, and restructuring in the target language.The translators without any experience in academic and/or vocational translation training, on the other hand, generally responded in a way which indicated that they were not familiar with the term and did not have analyzed awareness of steps within the task.One of them, for example, responded: "What do you mean by the translation process?".This question was expanded in order to enable those without any experience in academic and/or vocational translation training to describe their translation process by asking, for example, what they did to the source text after accepting a task.The response of the sixteen translators to this question depicted subsequent stages in the translation process (Bell, 1991;Nida, 1964;Hervey, Higgins & Haywood, 1995;Farghal & Al-Masri, 2000), as summarized below.(a) Stage 1: Reading prior to translating.All translators stated that they read the source text at least once.The purposes of the activity are to find out firstly what the original author wanted to say to his/her readers and secondly to identify difficult lexical items and technical terms and complex sentences.The problems they encountered while reading the original text were generally marked or underlined.Then they tried to find their meanings in dictionaries.One of the translators, for instance, stated: ("Setelah membacanya per kalimat kita berusaha untuk menerjemahkan dalam satu alinea, khan banyak katakata yang sulit, biasanya saya punya kertas tersendiri.Kata-kata yang sulit itu saya urutkan menurut abjad, A, B, C, D, seperti membuat kamus begitu, baru saya melihat di kamus.Sehingga kalau kata-kata itu misalnya muncul lagi pada paragraf atau bab selanjutnya, itu, tinggal mencari.")("After reading the whole texts, i begin to translate each paragraph.As there are  some unfamiliar words, I often make a word list in a separate paper and I look up dictionary for their meanings.It really helps me when I find the some words at the next paragraphs") (b) Stage 2: Translating.As has been mentioned previously, the first stage of the translation process described by the translators includes reading the original text and in some instances identifying difficult lexical items, technical terms, idiomatic expressions and complex sentences.The second stage of the translation process was rendering the source text into the target language sentence by sentence.Table 1 illustrates activities carried out by the translators prior to the act of translating.
(c) Stage 3: Restructuring and revision.The last stage of the translation process was restructuring the translation.Some of the translators referred to restructuring as revision.They stated that the aim of the restructuring was three-fold: 1) to check for an consistent use of technical terms, 2) to ensure that the structure of the translated sentences conformed to Indonesian grammar, and 3) to consider whether long and complex sentences should be rewritten into simple ones for readability purposes (Nababan, 2000;Nord, 2000;Newmark, 1997;Ruuskanen, 1996).

Ways in Which Translators Deal with Problems in Translation Process
All translators stated that when engaged in the translation process they got stuck with two kinds of problems.The first problem related to word meanings and the second problem had to do with the complexity of the sentence structures found in the source text.To solve problems with word meanings, the translators consulted dictionaries.All translators stated that if dictionaries did not provide the closest equivalence, they decided to retain the original words in their translation.An annotation was then given to help the readers understand the meaning of the original words.
With regard to dictionary use (Table 2), the sixteen translators said that they used an English-Indonesian dictionary as the primary dictionary when translating an academic text from English into Indonesian.Of the 16 research participants, 14 translators reported using an English-English dictionary, 6 translators used an Indonesian-Indonesian dictionary, and 5 translators said that they frequently consulted dictionaries of technical terms when translating academic texts.One translator reported using dictionaries of English-Indonesian idiomatic expressions.The translators who used an English-English dictionary while translating an English academic text into Indonesian had two main reasons to do so: firstly, when a word they are looking for is not available in a bilingual dictionary (English-Indonesian), and secondly, when a word they are searching for is available in a bilingual dictionary but none of its meanings are suitable for the context.One of the translators, for instance, stated that he had to use an Oxford Dictionary to understand the word "explicate" as the word cannot be found in an English-Indonesian dictionary.Their underlying reason for using an Indonesian-Indonesian dictionary is to ensure that a word they choose or use was considered to be an Indonesian word and to achieve variety in the use of Indonesian words in their translations.For example, the word 'relestat', as the equivalence of 'real estate', has been widely used, as a result of the policy of the Pusat Pembinaan dan Pengembangan Bahasa (Center for Indonesian Language Development) to eliminate the use of foreign words in Indonesian.In addition, the Center promotes the use of the word 'anggit' as the new equivalence of 'concept.'Indonesian writers in general and Indonesian translators in particular avoid using this new equivalence, as it is not familiar to target readers.Instead, they use the old and popular transliteration: konsep (see Kussmaul, 1995 on how translators employ dictionary to solve problems with word meanings).
All of the translators believed that technical terms were key words that needed to be rendered carefully.They stated further that consistency in using technical terms should be maintained in the target language text, especially if different translators translated chapters of a textbook.This also led them to argue that a text or a textbook should be translated by one translator to avoid inconsistent use of technical terms.If a text or a textbook has to be translated by more than one translator, meetings among the translators, they argued, need to be held regularly to discuss and decide what technical terms should be used.
With regard to problems with sentence complexity one of the research participants (Translator 1) said that he tried to solve the problem before he continued working on other sentences.The other fifteen translators, on the other hand, said that they left the original sentence(s) and continued to translate the other parts of the source text.Then, they went back to the sentences.If they could not solve the problem, they decided to put the original sentence(s) in their translation.
It was also found that the translators consciously considered the readership of their translation in making decisions about the style of target language complexity (Fifteen of the sixteen translators stated that they tended to cut a complex sentence into simple sentences to help readers understand the intended message of the original.Only one translator maintained complex sentences in his translation.He stated: "Saya tidak pernah memotongmotong kalimat, karena takut mengurangi kekuatan gagasan".('I never break long sentences, as I am afraid I will lose their ideas').

Translators' Beliefs about Translation Competencies
All translators stated that translating was a difficult task.They stated that translators needed to have a wide range of knowledge, skills and experience in order to do the task successfully.They believed that a good command of the source language and target languages is the main key to success in translation.They also believed that practical experience in translation plays an important role; they stated that the more they translated the better their translations would be.When asked whether knowledge about translation theory contributes to their translation skills, two kinds of response were obtained.Translators with vo- Translators who obtained academic translation training during their tertiary education, on the other hand, believed that an adequate knowledge of translation theory does not guarantee that better translations will be produced, even though it does help translators to tackle translation problems.

Ways in Which Translators Identify
Target Readers Translation is to be read.This means that the readers for whom a translation is intended need to be identified by translators at the preliminary stage of the translation process.The advantage of identifying target readers at the preliminary stage of the translation process is to enable translators to choose or use the most suitable words, technical terms, and sentence structures for target readers.A study by Ruuskanen (1996) indicates that professional translators generally ask questions to the commissioners of a translation and that some of the questions relate to the identification of target readers.
With regard to whether the sixteen translators interviewed in this study put questions to clients after they accepted a translation task, four translators answered "Yes" while the other twelve translators said "No".When the four translators were asked further about what questions they usually put to the clients, various responses were obtained.As shown in Table 3, the most typical questions asked led to preliminary identification of target readers

Table 3. Questions Frequently Asked by Translators to the Commissioners
Questions frequently asked to the commissioners Translator No. • Who will be reader(s) of this translation?
• When should the translation be completed?
• Should the translation be typed or hand-written?6 • Who will be reader(s) of this translation?7 • Who will be reader(s) of this translation?
• What is the text about?

8
• What is the text about?
• When should the translation be completed?11 and the topic or content of the source text.Less frequently put questions were about the completion time and the format or layout of the translation.
Table 3 shows that three translators (6, 7 and 8) solicited information about who would be the reader(s) of their translation.They told the researcher that information about the readers was very useful to obtain at the preliminary stage of the translation process as it helped them to determine the readability of their translation in terms of the choice and use of appropriate words, technical terms and sentence structures for their readers.Similarly, two translators (6 and 11) reported asking about the deadline of the commission.These two translators claimed that the length of time spent on translating a source text into a target text affected the quality of the outcome.They stated further that the longer they spent on the task, the better their translation would be.In addition, two translators (8 and 11) asked about the subject matter of the source text, commenting that they would not accept a source text to be translated if it was outside their expertise or field of academic preparation.Meanwhile, one translator (Translator 6) asked whether the finished product should be typed or hand-written.
It is apparent that three of the sixteen translators explicitly solicited information regarding target readers' backgrounds.The other thirteen participants, on the other hand, stated that there was no need to obtain such information directly from the clients, arguing that they could identify end users of a translation by reading the content of the source language text.They stated further that in the case of translating a textbook, it was the publisher who provided the information.This indicates that the thirteen translators do consider the question but extrapolate information about target readers in a different way.

Criteria Translators Use to Define
Quality of a Translation This section examines translators' views about criteria for defining a quality translation and whether these criteria determine what emphasis they put on the products of translation.
If they put an emphasis on readability rather than accuracy of a translation, for instance, it is possible that they will tend to distort messages of the original text.On the other hand, if they believe that accuracy is the most important criterion of a quality translation, they may neglect readability aspects.There is a possibility, however, that they put an emphasis on both aspects: accuracy and readability as the salient features of a quality translation.
With regard to what requirements a quality translation should have, a straightforward response was obtained.All translators put accuracy and readability as two salient features of a quality of translation.If this is not possible, they see accuracy as the first priority and readability as the second one.In terms of translating an academic text, they added that the language style used should be formal and the consistency in using technical terms had to be maintained in the target text.

Conclusion
The interview data indicate that all of the translators read the source language text at least once.Their main purpose was to identify difficult lexical items and technical terms, and complex sentences.
The interview also showed that the majority of the translators relied on a bilingual general (English-Indonesian) dictionary when translating an academic text from English into Indonesian.This bilingual general dictionary did not help them much in solving problems with word meanings.It was found, for example, that problems with meanings of lexical items and technical terms were prevalent in the six translations.
Most of the translators involved in this study had an adequate knowledge of translation.They also had an awareness of the importance of target readers to consider in translation process.