The Role(s) of Native Speakers in L2 Translation

The Role(s) of Native Speakers in L2 Translation David Mraček ABSTRACT: Translating into a non-mother tongue (L2 translation) has received increasing attention from translation scholars over the past two decades in response to the growing proportion of this direction in translation markets in most parts of the world. One of the aspects of L2 translation that remains a relatively uncharted territory is the role of native speakers. Although they are normally involved in relatively few translations from a language of limited diffusion into a major language directly as translators, native speakers need not be entirely absent from L2 translation as it has been suggested that they can assume diverse roles in the process and that cooperation with native speakers brings obvious benefits to L2 translators. The present study aims at providing a more complex picture of the native speaker’s role(s) in L2 translation, drawing on the results of a recent project on the qualitative and sociological aspects of L2 translation. By focusing on the questionnaires that the 40 subjects, professionals and advanced translator trainees, submitted before participating in a translation experiment, the study intends to shed more light on the views, preferences and habits of Czech translators regarding their cooperation with native speakers, discussing the possibilities as well as limitations of native-speaker participation in L2 translation. KEYWORDS: consultant, informant, L1 translation, L2 translation, native speaker, revision 1. INTRODUCTION With translation into a non-mother tongue (L2 translation) having become a fact of life in most parts of the world, this direction of translation has also recently been a frequent topic of research in Translation Studies. Since the late 1990s, researchers have focused on diverse aspects of non-native translation. Some have explored L2 translation as a process and product, analysing, among other things, the quality of literary and non-literary L2 translation, stylistic aspects and explicitation, some have dealt with the market and sociological aspects (frequency, rates, customer expectations etc.), and others still have discussed the teaching of L2 translation. Many of those findings have helped prove the untenability of the golden rule proposed by several Translation Studies authorities several decades ago whereby translating into one’s mother tongue is the only natural, accurate and effective direction while L2 translation is an undesired practice (cf. Newmark, 1988; cf. Ladmiral, 1979). Recent research has increased the general understanding of L2 translation, presenting it as a complex issue which deserves serious discussion and further study and, at the same time, suggesting it is a process where success depends on more than one factor. Although L2 translation implies, by definition, an absence of native speakers of the target language in the capacity of translators, they may still be involved in the translation process, albeit in other roles. This aspect of non-native translation, how- OPEN ACCESS 8 STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 ever, remains relatively under-researched and it is the aim of the present paper to shed more light on the role(s) of native speakers in L2 translation. OPEN ACCESS 2. THEORETICAL FRAMEWORK 2.1 L2 TRANSLATION: A WORLDWIDE REALITY Despite the above mentioned recommendations by translation scholars and, in some countries, institutional guidelines obligating professionals to only translate into their mother tongue, L2 translation has become a fact of life more or less throughout the world. L2 translation is a frequent practice mainly in countries using languages of limited diffusion, as suggested by recent translation market surveys involving Slovenian (cf. Hirci, 2005), Croatian (cf. Pavlović, 2007), Czech (cf. Svoboda, 2011), Polish (cf. Pietrzak, 2013) and Slovak (cf. Ličko, 2014). The obvious reason seems to be an insufficient supply of native speakers, as Gerard McAlester notes: “the volume of work exceeds the number of available translators who are major language native speakers” (McAlester, 1992, p. 292). An alternative explanation is offered by Paulina Pietrzak, who says that the “reason for the prevalence of Polish translators among all the translators working into the English language is primarily the price”, adding that commissioning a translation into English to a Polish translator is simply more cost-effective. (Pietrzak, 2013, p. 235) However, even surveys conducted in countries where “major” languages are spoken note a low availability of qualified native speakers of the target language relative to the demand, i.e. mainly the need to cater to the huge quantities of translation into English. This appears to be the case in Germany (cf. Schmitt, 1998), Spain (cf. Kelly et al., 2003) and China (cf. Pym, 2016). 2.2 NATIVE AND NON-NATIVE SPEAKERS AS TRANSLATORS Although L2 translation has become a standard practice in most parts of the world and a popular research topic in contemporary Translation Studies, this does not mean that a substantial amount of translation is not done by native speakers (L1 translation). Scholarly debates, thus, revolve around the merits as well as challenges associated with L1 and L2 translations. Some of the authors who address the topic discuss the types of texts appropriate for either L1 or L2 translation. There seems to be general agreement that of all the text types that are normally translated, expressive texts such as literary or journalistic works seem to be least suitable for translating out of one’s mother tongue, as the expressive and metaphoric use of language would be very likely to pose problems for a translator not fully in control of the lexical richness of the target language (Snell-Hornby, 2000). A more thorough explanation is given by Jiří Levý, who argues that “the greater the role of language in the artistic structure of the text, the more difficult translation becomes” (Levý, 2011, p. 48). Levý’s point is taken further by Xenia Liashuk, who suggests that less replicable (or, indeed, non-replicable) structures typical of most literary genres DAVID MRAčEK 9 call for a translator of great flexibility and creativity based on the broadest possible mastery of the target language, which is naturally more likely to be found in native speakers. (Liashuk, 2014) This does not mean, however, that successful L2 literary translators do not exist. In her ground-breaking attempt to put some of the traditional views on directionality to empirical test, the Slovene scholar Nike Pokorn (2005) analysed nine translations of Slovene prose into English done by native speakers of Slovene, native speakers of English as well as native speakers of neither of the two. Pokorn found, among other things, that some translators, despite being native speakers of only one of the languages involved in their translation work, have excellent skills in both, enabling them to produce native-sounding literary translations.1 Pokorn’s analysis also found that some of the translators who were native speakers of the target language (i.e. L1 translators) produced stylistically less than perfect translations, a finding which is hardly surprising given the abundance of Translation Studies literature containing examples of stylistic infelicities in L1 literary translation. When considering which text types are more appropriate for L2 translation, Snell-Hornby (2000) suggests informative texts, which are highly conventionalised with regard to their verbal and non-verbal elements (e.g. instructions for use, public announcements, commercial correspondence and scientific reports) and operative texts such as tourist brochures. These text types lend themselves more easily to L2 translation, particularly if they are intended for international audiences, i.e. where translators are expected to use International English or McEnglish, a lingua franca lacking the lexical richness and idiomaticity required in the translation of expressive texts. McAlester goes even further, arguing that a non-native translator may not only be as competent to translate such texts as a native speaker but sometimes even better prepared for such tasks as she “may write in a simpler language with fewer culturespecific references, and thus one more suited to an audience for whom English is a second language” (McAlester, 1992, p. 293). Competence is understood here as the ability to transmit the message in a comprehensible language, sufficiently correct not to cause amusement on the part of the reader, in other words to produce “translations that are reliable and respectable” (ibid., p. 297). Xenia Liashuk, employing František Miko’s style-defining categories (Miko, 1970), argues that the more factual/informative and content-oriented (and less expressive/emotional and receiver-oriented) a text is, the more its translation depends on successfully mastering the style perceived as natural in the given genre. The translation of informative texts is, therefore, more approachable for a L2 translator as long as she is able to develop an active repertory of contextual equivalents, by being in permanent contact with texts of the given genre in both working languages. (Liashuk, 2014) 1 It must, however, be admitted that Pokorn’s L2 translators were not typical native speakers of Slovene. The personal history of some of the translators suggests that English was not a foreign language but a second language for them. As Roman Ličko points out, this is a crucial distinction as “there is a significant difference between translating a literary text into a second language and rendering it into one’s foreign tongue” (Ličko, 2014, p. 46). OPEN ACCESS 10 OPEN ACCESS STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 Personal experience as well as empirical research seem to suggest, then, that the feasibility of L2 translation is very much dependent on the type and genre of the source text and, in some contexts, the nature of the target-text recipients, rather than merely on the translator’s language competence. The conclusion seems to be that where native-speaker translators are in undersupply, some of the work can be done reasonably well by their non-native peers. 2.3 NATIVE SPEAKERS AS CONSULTANTS AND REVISERS As has been pointed out several times above, a significant number of translators have to be ready to accept assignments involving translation into a non-native language. However, L2 translators need not work alone. The common practice — throughout history as well as today — is for L2 translators to cooperate with native speakers, who may assume diverse roles in the process, as empirical evidence shows. When inquiring about the particular stage at which native speakers are involved in L2 translation from Slovak into English, Ličko (2014) found that 64.7% of the respondents who cooperate with a native speaker do so in the editing stage only, i.e. for proofreading the target text to eliminate errors of grammar, vocabulary and style. As many as 21.6% of the respondents rely on the assistance of a native speaker during the pre-editing stage, i.e. throughout the translation proper. Others pointed out that cooperation is not necessarily limited to either of the two stages, saying that they sometimes enlisted the services of native speakers for both proofreading and continuous advice. A study by Elke St. John (2003) investigated how native-speaker informants can be beneficial for the different stages of L2 translation during translator training. In a video-recorded experiment, one Japanese student and one Chinese student were asked to translate a text with an informative and operative function into English; each of the translators was assigned a native speaker of English as an informant. Two main areas were identified where the native speakers offered help: grammar (specific items reported by St. John’s respondents included prepositions, articles — especially where the translator’s L1 lacks them — word order, active vs. passive voice) and use of language (e.g. lexical choices, collocations). Although in the experiment the native speakers offered a surprisingly great deal of advice, St. John admits that with more specialist and terminology-laden texts, a native speaker with no knowledge of the topic involved would probably be of much less help to the L2 translator. St. John also points to further limitations of native-speaker informants, noting that their insights may not be absolutely reliable and that they may lack the metalanguage necessary to explain some of their intuitions. St. John thus concludes that native speakers are “best dealt with as one of a number of information sources, with some attention being given to what each can best offer the translator”. (St. John, 2003, p. 11) An even more detailed account of the native speaker’s possible involvement in L2 translation — and a remarkable insight into the psychology of the L2 translator relying on native-speaker assistance — is provided by Pedersen (2000). Sharing his own best practice as a L2 translator of Danish texts into English, Pedersen notes that DAVID MRAčEK 11 after the first drafting, the text goes through three stages: renewed comparison with the original, revision by a native speaker, and the final editing performed by the L2 translator. The native reviser is mainly expected to produce formulations that are stylistically more acceptable to native speakers than the translator’s own. An interesting remark is made in relation to the use of the native informant as a supplement to the dictionary: “If I feel reasonably sure that he or she will come up with an acceptable English solution, there is no need to spend a lot of time digging it out of dictionaries myself.” (Pedersen, 2000, p. 112) Pedersen’s revisers are sometimes given the source text and sometimes not, depending on the nature of the text and the reviser’s competence in Danish. Apparently, the role of the native speaker here is more one of a proofreader and/or informant than reviser proper if revision is understood — as is normally the case in translation practice and research — as checking the translation against the source text for accuracy and completeness. In a remark with much pedagogical potential, Pedersen notes that not all changes proposed by a native speaker should be automatically accepted, because “accepting a suggestion which is good English may mean getting too far away from the source language text” (Pedersen, 2000, p. 112). Pedersen goes on to point out, however, that even if the translator has to reject the native informant’s suggestion for lack of accuracy, she may be inspired by it to improve the text, making it read more natural and idiomatic than her original formulation. On the other hand, native speakers tend to have their own preferences, which, stylistically acceptable as they are, may not necessarily be better than the L2 translator’s own writing. Summing up the possibilities and limitations of native-speaker participation in L2 translation, Pedersen concludes that “working with an informant thus does not free translators from the obligation to exercise their own judgment”. (Pedersen, ibid.) Apart from native speakers’ personal limitations, there are purely practical issues that can stand in the way of an efficient use of native speakers. Using the example of native speakers of English in the Slovak translation market, Roman Ličko points to several practical difficulties associated with the involvement of native speakers in L2 translation in a minor language community. First, the participation of a native speaker in a translation assignment means additional financial costs to be incurred by the customer, a fact which most customers seem to be unaware of. Second, while the demand for native speakers is relatively high, their services are difficult to come by as there are few native speakers available in the country and they are very likely to continue to be absent from most L2 translations in the foreseeable future (Ličko, 2014). As is evident from the above overview of recent research and discussions regarding directionality in translation, native speakers can contribute to the translation process in a number of ways. If they cannot be used directly as translators — as seems to be the case in an increasing number of countries and settings — they can assist a non-native translator in various capacities related to linguistic as well as nonlinguistic consultation at different stages of the process. However, their involvement carries several potential risks that should be considered in practice and which are worth further inquiry. OPEN ACCESS 12 STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 3. METHODS OPEN ACCESS 3.1 RESEARCH CONTEXT The present study is part of a large-scale research project, conducted at Charles University’s Institute of Translation Studies in 2016–2018. Taking up the calls for further research to help destigmatize L2 translation (Pavlović, 2007), the project aims at investigating translation into a non-mother tongue, including textual and sociological aspects, between Czech, a language of limited diffusion, and English and French, two major languages frequently demanded in the Czech translation market and in various other areas of employment. What follows is a brief description of the project’s methodology. The research population comprised 40 translators, native speakers of Czech. Twenty of them were professionals who graduated from the Institute’s programme in translation eight to fifteen years ago and have had an uninterrupted translating career since, the other half being students in the second year of their master’s programme in translation. Half of the professionals and students translate between Czech and English, the other half between Czech and French. Data was gathered in several ways, combining quantitative and qualitative methods. The first source of data were translations produced under experimental settings, whereby each subject produced a total of four translations (a promotional text and a legal one were first translated from L2 into L1 and later a comparable pair of texts was translated from L1 to L2). A product-oriented qualitative analysis of the translations was conducted by a panel of assessors, with two to three native speakers evaluating each direction of translation. To better understand the complexity of the translation process and sociology, a series of questionnaires was administered to each respondent. A pre-translation questionnaire inquired about their general attitudes and habits regarding L1 and L2 translation, providing valuable psycho-sociological data on the personality of the translators themselves, a topic still relatively under-researched in L2 translation research. Post-translation questionnaires were designed to explore in more detail the difficulties faced by the translators as well as the strategies and resources employed to overcome the challenges. 3.2 DATA COLLECTION Since the present study is concerned with the ways in which native speakers of (not only) English and French assist native Czech translators in the process of L2 translation in the Czech market, I turned to the reflections provided by Czech translator trainees and professionals as the primary source of data. Therefore, the pre-translation questionnaires described above were analysed to shed more light on the views and experience of Czech translators regarding native speakers and their input as consultants and revisers. Although the subjects participated in the translation experiment only with one of their foreign working languages (translating between Czech and either English or French), they were invited, after filling out the primary ques- DAVID MRAčEK 13 tionnaire concerning the language of the experiment, to share their experience and views on a second foreign language which they use as translators (if they have one). Therefore, whenever quantitative data is discussed, my analysis of the results will inevitably include two sets of data: 40 respondents answering about their primary working foreign language, that is either English or French, with 24 of them providing an extra set of answers concerning their second foreign language, i.e. English (12), Spanish (5), German (4), French (2), Russian (1). A total of four items in the pre-task questionnaire dealt, either directly or indirectly, with the topics under discussion here. Two of the questions were related to native-speaker consultations: — Do you consult native speakers when translating out of your native language? Choose one answer: always/most of the time — sometimes — never.2 — Explain what issues are consulted on (vocabulary, terminology and realia were offered by way of example).3 The other two questions inquired about the subjects’ experience regarding revision and/or proofreading by (not only) native speakers: — Are your translations out of your native language revised/proofread? Choose one answer: always/most of the time — sometimes — never — I do not know.4 — Explain who the reviser/proofreader is (a native speaker or a Czech colleague), whether the revision/proofreading applies to the entire text or difficult sections only, whether you are able to see the revised/proofread text, and — supposing you have a say in the revision process — to what extent you accept the changes proposed.5 The following section of the paper will discuss findings yielded from the above items of the questionnaire. Answers to some of the sub-questions can be translated into quantitative data; however, the primary ambition of the study is to analyse and discuss the respondents’ more extended answers in which they shared their experience and views on the subject. 2 3 4 5 The exact wording of the question in Czech is: “Využíváte při překladu do cizího jazyka v této jazykové kombinaci konzultace s rodilým mluvčím? Vyberte jednu z možností: vždy nebo většinou — někdy — nikdy.” The exact wording of the question in Czech is: “Rozveďte odpověď na otázku č. 12 (např. nejčastěji pro ověřování lexika, terminologie, reálií…).” The exact wording of the question in Czech is: “Procházejí vaše překlady do cizího jazyka v této jazykové kombinaci jazykovou revizí/korekturou? Vyberte jednu z možností: vždy nebo většinou — někdy — nikdy — nevím.” The exact wording of the question in Czech is: “Rozveďte odpověď na otázku č. 10 (kdo revizi/korekturu provádí; zda je to rodilý mluvčí, nebo jiný Čech; zda se revize/korektura týká celého textu, nebo jen obtížných míst; zda máte vždy možnost zrevidovaný překlad vidět; pokud máte možnost se k revizi/korektuře vyjádřit, do jaké míry změny akceptujete…).” OPEN ACCESS 14 STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 4. FINDINGS OPEN ACCESS 4.1 CONSULTATIONS The first part of the section devoted to presenting empirical data will discuss (a) the frequency with which the Czech participants of a project on L2 translation seek advice from native-speaker informants when translating out of their mother tongue, and (b) the areas, linguistic or otherwise, that are the object of the consultations. 4.1.1 FREQUENCY Consultation is here understood to cover any type of situation where a Czech translator doing a translation into a non-mother tongue approaches a native speaker of the target language, during the translation process (i.e. excluding the editing stage), to discuss a diversity of issues relating to the translation being done. The respondents were first asked how often they consulted a native speaker during L2 translation, having the following options: always/most of the time — sometimes — never. In summary, of the 40 respondents, most (21) reported that they sought native-speaker consultation “sometimes”, 11 “never”, while only 8 said “always/most of the time”. No significant difference in the distribution of answers was noted between students and professionals. Similar data was obtained from the answers relating to the respondents’ second foreign language (24 respondents), with a majority (13) reporting that they “sometimes” consulted a native speaker, 7 respondents saying “never”, and only 4 saying “always/most of the time”. The quantitative data is presented in Charts 1 and 2 below. How often do you consult native speakers during L2 translation? r = 40 (FL1) Professionals r = 20 Students r = 20 11 10 6 5 4 4 Always/most of the time Sometimes Never Chart 1: Frequency of consultations (FL1: answers given by respondents about their first foreign language). DAVID MRAčEK 15 How often do you consult native speakers during L2 translation? r = 24 (FL2) Professionals r = 15 Students r = 9 9 4 2 4 3 2 Always/most of the time Sometimes Never Chart 2: Frequency of consultations (FL2: answers given by respondents about their second foreign language). 4.1.2 ISSUES SUBJECT TO CONSULTATIONS The present section aims to provide a detailed typology of the various issues that are the subject of consultations demanded of native speakers by the respondents participating in the survey. The respondents were not restricted in terms of the length of their answers and, as mentioned above, vocabulary, terminology and realia were offered by way of example. The area reported by most respondents (24 respondents: 17 FL1, 7 FL2) was vocabulary, i.e. seeking native-speaker advice on the appropriateness of lexical choices. Some of the respondents gave more specific answers, mentioning that they contacted native speakers in order to check the use of collocations (3) and idioms (4). Another broad category frequently mentioned (12 respondents: 9 FL1, 3 FL2) was style, with one respondent explicitly saying that she asked a native speaker to judge the stylistic effect of marked linguistic features proposed by her in her L2 translation. Broad and rather vague as the categories of vocabulary and style are, the predominance of these issues in the respondents’ answers seems to be in line with observations and findings reported by some of the authors writing on the subject, including Elke St. John (see section 2.3 above). These process-related data can also be viewed from the perspective of product quality as some authors have suggested that vocabulary and style are a more frequent source of errors than grammar. For instance, Miroslav Pošta (1999) makes a general observation that the majority of deficiencies in translation are not qualitative errors (errors proper, e.g. grammar). Rather, they are quantitative errors, i.e. features not conforming to prevalent usage; also, they are products of decision-making, being “alternatives chosen by the translator in preference to other options. In other words, this is the realm of stylistics.” (Pošta, 1999, p. 91). In an attempt to present hard data on language mistakes in L2 translation, David Mraček (2017) explored the frequency and typology of grammar and style errors in OPEN ACCESS 16 OPEN ACCESS STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 Czech-English translations of promotional and legal texts, using the same population of participants as the present study, and established that the rate of style errors made by L2 translators, irrespective of their proficiency level (no significant difference was found between translator novices and professionals), was substantially higher than the rate of grammar errors. The text genre, however, was found to be a crucial factor: the L2 translations of a legal text contained an average of 5.9 style errors as opposed to 3.5 grammar errors; by contrast, in the translations of a promotional text, the ratio rose to twice as many mistakes in style as in grammar (10 style mistakes vs. 4.5 grammar mistakes). It appears, then, that non-native translators do have much less control over items of style, including the use of vocabulary, idioms and collocations, than over grammar, which results — during the translation process — in their heightened need for native-speaker assistance in this area of language. Interestingly, grammar itself was only cited by 4 respondents of the present study (3 FL1, 1 FL2); one specific example was given by a translator who often translates from Czech into French, saying that she explicitly asked the native speaker to check the correct use of articles and punctuation. Other areas consulted by Czech L2 translators will now be discussed. Terminology, though belonging to the realm of vocabulary from a linguistic point of view, is here treated as a separate issue as it constitutes one of the major translation problems in almost any translation assignment. Thus, it was offered in the questionnaire as an example independent of vocabulary, and was listed alongside vocabulary by the respondents (15 respondents: 10 FL1, 5 FL2). A total of 9 respondents (6 FL1, 3 FL2) cited realia, or features relating to life and institutions. One respondent mentioned domain-specific knowledge as, presumably, the native speaker whom she co-operates with specializes in the particular domain of the translation. Finally, although the questionnaire inquired about L2 translation, a total of 4 respondents (3 first-foreign-language subjects, 1 second-foreign-language subject) explicitly stated that they used their native-speaker consultants when translating into their mother tongue, checking comprehension in the source (i.e. foreign) language. There is, however, every reason to believe that a much higher number of the respondents use their native-speaker collaborators in this way and they would no doubt have indicated the fact had they been explicitly invited to do so. All the areas of native-speaker consultation that were identified in the respondents’ answers are presented in descending order in Table 1 below. To conclude the discussion of issues for which non-native translators prefer to seek assistance from native speakers, several remarkable points should be noted as they provide precious insight into the psychology and sociology of L2 translation. One respondent (a professional) noted that she rarely used native-speaker assistance since she mostly found sufficient the online resources available such as dictionaries, search engines and translator forums. One of the translators (a student) who reported that she never consulted native speakers said she preferred to consult her non-native colleagues; another person (also a student) said she had never been in a situation that would necessitate native-speaker assistance. Another interesting answer concerning a more general approach to using information resources in L2 translation — and one that can be thought to apply to more than one translator — DAVID MRAčEK Area consulted lexical choices, incl. collocations and idioms terminology style realia / culture / life and institutions grammar comprehension in L1T domain-specific knowledge 17 No. of answers FL1 respondents 17 10 9 6 3 3 1 No. of answers FL2 respondents 7 5 3 3 1 1 0 table 1: Types of translation-related issues where consultation is sought by non-native translators (FL1: answers given by respondents about their first foreign language; FL2: answers given by respondents about their second foreign language). came from a translation novice, who said she only consulted native speakers over issues that she had been “racking her brains” over for some time, having found no answer in any of the “ordinary” sources of information. Two interesting points were raised concerning terminology. Two respondents said that they did their own terminology research, which was sufficient most of the time. Another respondent noted that she preferred to check the use of English terminology with a Czech-speaking expert in the given field, while the other mentioned that sometimes she approached the customer with terminological issues, pointing out that customers are very often the best source of verification where terminology is concerned. It is hardly surprising that both these points were made by professional translators, and they seem to reflect a more experienced and, perhaps, responsible attitude to terminology, one of the crucial aspects of translation. Finally, the findings seem to confirm the special status of English as far as the availability of (mainly online) resources is concerned. More than one respondent expressed their satisfaction with how many issues relating to the English language can be consulted online, suggesting that the significance of native speakers of English as consultants may be diminishing, or at least changing: according to one respondent, with so many sources available for English, native speakers are used not as informants on particular items (which would more often be the case with other languages) but rather as a fresh pair of eyes to confirm that the text reads smoothly and is comprehensible. The other observation made by respondents who have more than one foreign working language with respect to the availability of resources was that nowadays, English vocabulary, idioms, terminology as well as cultural references are much easier to find online than with other languages. One professional translator pointed out that with Spanish, for instance, L2 translators had to consult native speakers on a broader spectrum of issues than with English. Obviously, the number of linguistic resources available (reference books, dictionaries and, of particular relevance nowadays, online sources) differs considerably among languages and there are lowresource languages and high-resource languages (recently, this dichotomy has been used in machine translation studies to refer to the amount of parallel data that MT OPEN ACCESS 18 OPEN ACCESS STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 can utilize). It would thus seem only natural to acknowledge the special status of English as a “super-high-resource language”. This may have numerous implications for translation practice, research and training, one of them being the need to question the reliability, quality and authoritativeness of the resources that translators use. Interestingly, none of my respondents said anything about possible risks associated with searching information online. 4.2 REVISION The present section is devoted to revision in L2 translation and aims to discuss: (a) the frequency with which the Czech participants of a project on L2 translation have their translations into a non-mother tongue revised (by native speakers or otherwise), and (b) the personality of the reviser, the extent of revision as well as the respondents’ experience and views regarding revision in L2 translation. 4.2.1 FREQUENCY Revision (here understood to cover proofreading) is another frequent service which native speakers of a target language offer to the translation market and is perhaps more visible than native-speaker assistance during the translation process (such as consultation discussed in section 4.1). It was thus a natural part of the pre-task questionnaire inquiring about the roles of native speakers in L2 translation. The respondents were first asked whether (and how often) their L2 translations were revised, having the following options: always/most of the time — sometimes — never — I do not know. In summary, most of the 40 respondents know whether or not their L2 translations are revised (35), with 21 respondents noting that their translations are revised “always/most of the time”, 12 respondents “sometimes” and 2 respondents “never”. The remaining 5 respondents do not know whether there is a reviser. No significant difference in the distribution of answers was noted between novices and professionals. Similar data was obtained from the answers relating to the respondents’ second foreign language, where 20 of the 24 respondents know whether or not their L2 translations are revised. Of them, 15 reported that their translations were revised “always/most of the time”, 4 respondents “sometimes” and 1 respondent “never”. The quantitative data is summarized in Charts 3 and 4 below. It needs to be borne in mind, however, that given the formulation of the question, these sets of data relate to revision in L2 translation in general, disregarding the language status of the reviser. Whether or not the reviser is a native speaker of the target language will become obvious from the answers to the following set of questions. Next, the respondents were invited to elaborate on the issue of revision and indicate who the reviser is (a native speaker of L2 or a Czech colleague), whether the revision applies to the entire text or difficult sections only, and, finally, whether the respondents are able to see the revised text and, supposing they have a say in the revision process, to what extent they accept the changes proposed. What follows in sections 4.2.2, 4.2.3 and 4.2.4 is a summary of the most interesting points raised in the answers. DAVID MRAčEK 19 How often are your L2 translations revised? r = 40 (FL1) Professionals r = 20 Students r = 20 11 10 6 6 1 Always/most Sometimes of the time 1 Never 2 3 I don't know Chart 3: Frequency of revision (FL1: answers given by respondents about their first foreign language). How often are your L2 translations revised? r = 24 (FL2) Professionals r = 15 Students r = 9 10 5 3 1 Always/most Sometimes of the time 3 1 0 Never 1 I don't know Chart 4: Frequency of revision (FL2: answers given by respondents about their second foreign language). 4.2.2 THE REVISER: NATIVE OR NON-NATIVE? The general conclusion as to the language status of the reviser seems to be that revisers are not always native speakers: of the 27 respondents who were explicit about the personality of the reviser, about a half said that their translations were always revised by native speakers but the rest reported various other scenarios (with revisers being sometimes native, sometimes non-native, mostly native, mostly non-native, or always non-native). The data is summarized in Table 2 below for respondents who gave answers relating to their first foreign working language as well as those reporting about their second foreign language. Much more illuminative of the revision practice than the figures above are some of the comments left by the respondents, suggesting that more than one factor plays a role in whether revision is assigned to a native speaker or not. The difficulty of the OPEN ACCESS 20 Revisers: native vs. non-native OPEN ACCESS Always native Mostly native Sometimes native, sometimes non-native Mostly non-native Always non-native STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 No. of respondents FL1 14 3 5 2 3 No. of respondents FL2 7 1 1 1 2 table 2: Revisers of L2 translations according to native-speaker status (FL1: answers given by respondents about their first foreign language; FL2: answers given by respondents about their second foreign language). translation seems to be a decisive text-internal factor, with one respondent (a student) pointing out that most of her L2 work is revised by Czechs and only more challenging texts undergo native-speaker revision. Most factors, however, are external, having to do with the different actors in the translation process. It was, for instance, mentioned by one respondent (a professional) that her translations assigned by a translation agency were revised by native speakers while her assignments for direct customers were not revised at all. As regards text recipients, one respondent (a professional) pointed out that native-speaker revision only took place with texts intended for native-speaker users, implying that at least some of her Czech-English translations were targeted at lingua franca users. A more general view was expressed by another respondent (a professional), who believes that no native-speaker revision is necessary with in-house texts, texts not intended for publication and/or texts where further editing is expected; in these categories of texts, minor linguistic imperfections are easily tolerated, according to the respondent. Of the other text-external factors, the customer was mentioned several times. For one thing, a respondent (a professional) pointed out that native-speaker revision was only done if required by the customer. On the other hand, according to another respondent (a professional), few customers are willing to pay extra rates for native-speaker revision (an attitude reported also by Ličko’s survey of the Slovak market, cf. section 2 of the present article); price, therefore, seems to be a decisive factor. Furthermore, some of the respondents’ comments on the profile of the reviser shed more light on the revision process itself. One example of good practice is a system of peer-revision by non-native speaker translators, with the respondent (a professional) explaining that editing can be done using the “Comments” or “Track changes” options in MS Word while it is the translator who is responsible for making the final decision. More on the psychology of L2 translators face-to-face with revisions will be said in section 4.2.4. An alternative scenario, realistic with texts intended for multinational corporations, was mentioned by the same respondent whereby the customer may be asked to have the target text revised by his/her colleagues in the target-language country. On a more general level, it was also pointed out by a respondent (a professional) that revision is a good thing per se as a second pair of eyes can always see something that the first one missed. DAVID MRAčEK 21 4.2.3 EXTENT OF REVISION When asked about whether the revision (native or otherwise) covers the entire translation or only some portions thereof, most respondents said that revision normally applied to the entire translation (17 of the 22 who explicitly answered the question). However, several others said that they only asked for revision of difficult portions of the text or segments that they were not very sure about. One of these respondents mentioned the time factor (with the reviser only concentrating on challenging sections when a limited amount of time is available), while another pointed out the low availability of qualified native speakers and budgetary constraints as factors which sometimes only allow for partial revision. 4.2.4 THE POST-REVISION STAGE: PRACTICE AND ATTITUDES In order to find more about Czech practice with respect to translation revisions, and in an attempt to shed more light on the psychology of L2 translation, the questionnaire also investigated the respondents’ degree of confidence when dealing with revisions done to their L2 translations. Therefore, the respondents were asked whether they were normally able to see the revised text and how many of the reviser’s interventions they normally accepted. Out of the 22 respondents who made an explicit note regarding how often they were able to see their revised translations, 14 said that they always were. In two cases, the reason was that it was normally the translator who sent the final product to the customer. Interestingly, one of the 14 respondents, a professional, said that she always had access to the revised text because she herself always placed a demand for revision, while another professional, demonstrating a laudable sense of responsibility, pointed out that she never submitted a translation to the customer without seeing the revised version first. A further three respondents said that they were able to see the revised text most of the time, with one of them admitting that she was normally sent the revised text “on demand”. Three respondents said that they were sometimes sent the revision and sometimes not: one pointed out that some of her translations were revised by the recipient’s own reviser in which case she was only able to see the final version once it was published; the other two said the normal practice with translations assigned through agencies was that the translator was sent the revisions at random or only if the interventions were numerous. Translation agencies were also mentioned by the two respondents who said that they knew that their text was revised but received very little feedback and almost never saw the final product. It follows from the above that a considerable number of the respondents are normally able to see their L2 translations after they have been revised. It is then very interesting to see what happens to the text at this stage, i.e. before it reaches the end customer. Although the general conclusion is that L2 translators normally accept the changes introduced by revisers (even though these are not always native, as it turns out from the previous answers, as summarized in Table 2), there is substantial variation among OPEN ACCESS 22 OPEN ACCESS STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 respondents as to their experience, degree of trust and sense of responsibility. Some gave rather general answers; for instance, the relatively frequent “I accept most of the changes” type of answer was given equally often by professionals and students. Two respondents, who indicated that theirs was always a native-speaker reviser, seem to fully trust their collaborators, saying that they “definitely” or “naturally” accept the changes. Another, also a professional collaborating almost invariably with native speakers, said that she accepted ninety-nine percent of the changes as her revisers are fellow translators. However, a few respondents said, in a rather vague manner, that they accepted only some of the changes. Two of these (both students) pointed out that they checked the revised text carefully and reversed some of the reviser’s procedures to eliminate any shifts of meaning that might have been introduced by the native speaker, especially where native speakers worked without the source text. It does seem, then, that it is important to distinguish between revision and proof-reading, a terminological distinction which was left unsolved by the authors of the questionnaire. Another student respondent said that she had to decline some of her reviser’s suggestions because she had to make the text conform to the terminological and other specificities demanded by the customer (as she pointed out, she would normally expect her reviser to only improve the text in terms of style and comprehensibility). Several other remarkable points were raised as to whether or not to accept all changes proposed by native-speaker revisers. First, not every native speaker is qualified enough to do a qualified revision. Second, the translator may have a better domain-specific knowledge — and also a different feel for the language — than the reviser. Finally, and perhaps most importantly, the translator should use her own judgement when deciding which changes to accept and which to reject because, after all, it is the translator who is responsible for the quality of the end product. Perhaps not surprisingly, these remarks were all made by professionals, which may point to the fact that professionals tend to be more aware of their own responsibility for the quality of the product and also of the limitations associated with native speakers; moreover, they seem to be in line with the experience presented by translation scholars (cf. Pedersen in section 2.3). Finally, some of the answers reveal something about the interaction between translators and revisers. For instance, more than one respondent said that they discussed the proposed changes with their revisers, either in person, or via telephone or Skype. One respondent, a student, even has two regular native-speaker collaborators: the first is asked to do a more thorough revision, which is then discussed by the translator and the reviser. The result, a semi-finalized product, is given to another native speaker, who is responsible for checking the overall fluency of the target text. 5. DISCUSSION AND CONCLUSION “All too often, the translator has been envisaged as a more or less solitary individual and there has been a sense in which asking for linguistic advice is an admission of inadequacy.” (St. John, 2003, p. 11) The image of the translator depicted in the above quote was shattered by St. John’s own experiment and several other authors (cf. sec- DAVID MRAčEK 23 tion 2 of the present paper). Without making strong claims about the representativeness of my sample, the present paper, too, has revealed that translators can — and often do — enjoy the benefits of collaboration with other individuals, professional or otherwise, who may or may not share the translator’s mother tongue, and can assume a diversity of roles in the translation process. Considering the topic of the present paper, i.e. the role(s) of native speakers in L2 translation, the answers obtained from questionnaires that were administered to a population of 40 Czech professional and novice translators reveal, first and foremost, that the respondents do a substantial amount of networking as they seek advice from native speakers on a variety of issues, with lexical choices clearly leading the list. At the same time, however, the answers suggest a possible future trend whereby L2 translators working into English may be relying less and less on native-speaker informants as English is becoming a “super-high-resource” language in terms of the amount of linguistic material available online. As regards revision, most of the respondents have their L2 translations revised on a regular basis. The quantitative data can be compared to the findings of researchers who have explored the translation markets in countries where other languages of limited diffusion are spoken. For instance, in his survey of the Slovak translation market involving 111 translators who regularly translate from their native Slovak into English as a foreign language, Roman Ličko inquired, inter alia, about the extent of nativespeaker involvement in L2 translation, in any role, i.e. including consultants and revisers. He found that “the majority of Slovak translators of English cooperated with native speakers very little — i.e. either occasionally (38.2%) or even rarely (20.9%)” (Ličko, 2014, p. 50). Similarly, Nataša Pavlović, in her survey involving 193 Croatian translators, found that “only a slim number of respondents have their translation revised by a native speaker on a regular basis” (Pavlović, 2007, p. 88). Both Ličko and Pavlović cite the insufficient number of competent L1 English speakers available in the local market and customers’ general unwillingness to pay for the extra cost of revision as the possible reasons. The more optimistic results shown by the present study may have to do with the fact that, unlike the two surveys conducted in Slovakia and Croatia, the Czech participants were all graduates or near-graduates of a university degree programme in translation, implying a possibly heightened awareness of their own limits regarding L2 translation skills. Furthermore, the answers in which the respondents were able to explain their experience with L2 revision in more detail reveal no small amount of variation regarding what Czech translators think about, on the one hand, native speakers’ roles and skills and, on the other, their own abilities, limits and responsibilities as regards non-native translation. Professionals, in general, came across as more confident but also responsible than students, although the latter also demonstrated a growing awareness of the specific demands of L2 translation, crucial to the development of their self-efficacy. Although native speakers of the target language may be absent from a substantial number of L2 translations, the benefits of their assistance “must be more than obvious to all practicing translators” (Ličko, 2014, p. 51). It is then the task of translator training to ensure that the professionals of tomorrow are presented the full picture OPEN ACCESS 24 OPEN ACCESS STUDIE Z APLIKOVANÉ LINGVISTIKY 1/2019 of the possibilities and limitations of co-operation with native speakers. After all, McAlester (1992, p. 296) notes that students should be encouraged to work collaboratively with native speakers of other languages during translator training, “the aim being not only to get the students to produce as good a first version as possible, but also to train them in using a native-speaker reviser effectively”. However, this is no easy task as native speakers tend to vary in linguistic as well as extralinguistic competence; furthermore, students should be taught to combat the “unfortunate tendency for commissioners of translations to use a reviser independently of the original translator” (ibid.). It is, however, important to admit that in the present article, I have presented a considerable number of remarkable views on the participation of native speakers in L2 translation as seen by non-native translators and the realities of the translation market. It is equally interesting to explore the issue of L2 translation from the perspective of native speakers themselves, focusing on their views which are no doubt informed to a large extent by their experience as readers and revisers of L2 translations. Here are some of the questions that I feel are worth investigating: How difficult is revision of L2 translations for native speakers? How often do they feel that the entire process of L2 translation plus native-speaker revision is cost-effective? And how often do they feel that the final result of this double process really conforms to their own quality expectations? These questions (and many more related to native speakers’ viewpoints) have hardly been addressed by translation theory and research, but the following quote from a statement by William Lise, president of the Japan Association of Translators, may point to some of the possible answers; consider the words in italics: “[…] almost all of Japan’s Japanese-English translation is done by Japanese writing English as a foreign language, which is then the object of heroic damage-repair efforts by foreign rewriters.” (Lise, 1997, p. 27) To bridge this gap in our understanding of native speakers’ participation in L2 translation, the long-term project of which the present study was a part sought to consider the issue from the native speaker’s perspective as well. Data was gathered through interviews with a population of native speakers of English and French, and analysed to find out more about their long-time experience as translators and revisers in the Czech translation market (cf. Duběda et al., 2018). ACKNOWLEDGEMENT The article presents data obtained in an ongoing research project supported by the Czech Science Foundation (GA ČR) under no. 16-03037S, panel P406. Project title: Directionality in translation: qualitative and sociological aspects. REFERENCES Duběda, T., Mraček, D., & Obdržálková V. (2018). Překlad do nemateřského jazyka. Fakta, otázky, perspektivy. Praha: Karolinum. Hirci, N. (2005). Prevajanje v nematerni jezik: Tabu ali nuja? In N. Pokorn, E. Prunč & A. Riccardi (Eds.), Beyond DAVID MRAčEK Equivalence (pp. 87–104). Graz: Institut für Translationswissenschaft, Universität Graz. Kelly, D., Martin, A., Nobs, M.-L., Sanchez, D., & Way, C. (Eds.) (2003). La direccionalidad en traducción e interpretación. Granada: Atrio. Ladmiral, J.-R. (1979). Traduire: théorèmes pour la traduction. Paris: Payot. Levý, J. (2011). The Art of Translation. Transl. Patrick Corness. Amsterdam: John Benjamins. Liashuk, X. (2014). Translation into a foreign language: theoretical aspect. Prekladateľské listy 3: teória, kritika, prax prekladu (pp. 82–89). Bratislava: Univerzita Komenského. Ličko, R. (2014). Translation into English as a Foreign Language. A Slovak Survey. Banská Bystrica: Belianum. Lise, W. (1997). Machine translation: Is it working in Japan? Language International, 9(1), 27–35. McAlester, G. (1992). Teaching translation into a foreign language — status, scope and aims. In C. Dollerup & A. Loddegaard (Eds.), Teaching Translation and Interpreting. Amsterdam: John Benjamins, 291–297. Miko, F. (1970). Text a štýl: K problematike literárnej komunikácie. Bratislava: Smena. Mraček, D. (2017). Gramatické a stylistické chyby v překladu do angličtiny jako nemateřského jazyka. In Sborník příspěvků z konference a soutěže eLearning 2017 (116–126). Hradec Králové: Gaudeamus. Newmark, P. (1988). A Textbook of Translation. London: Prentice Hall. Pavlović, N. (2007). Directionality in translation and interpreting practice. Report on a questionnaire survey in Croatia. Forum, 5(2), 79–99. Pedersen, V. H. (2000). Translation into L2 — In practice, and in the classroom. In M. Grosman, M. Kadric, I. Kovačič & M. SnellHornby (Eds.), Translation into Non-Mother 25 Tongues: In Professional Practice and Training (pp. 109–116). Tübingen: Stauffenburg. Pietrzak, P. (2013). Divergent Goals: Teaching Language for General and Translation Purposes in Contrast. In K. Piątkowska & E. Kościałkowska-Okońska (Eds.), Correspondences and Contrasts in Foreign Language Pedagogy and Translation Studies, Second Language Learning and Teaching (pp. 233–240). Cham: Springer. Pokorn, N. (2005). Challenging the Traditional Axioms. Translation into a Non-Mother Tongue. Amsterdam: John Benjamins. Pošta, M. (1999). Explicitation processes in translation between Czech and English. Folia Translatologica 7/1999 (Issues of Translation into a Non-Native Language) (pp. 84–109). Prague: Charles University. Pym, A. (2016). Translation Solutions for Many Languages — Histories of a Flawed Dream. London: Bloomsbury. Schmitt, P. A. (1998). Marktsituation der Übersetzer. In M. Snell-Hornby, H. Hönig, P. Kussmaul & P. A. Schmitt (Eds.), Handbuch Translation (pp. 84–109). Tübingen: Stauffenburg. Snell-Hornby, M. (2000). ‘McLanguage’: The identity of English as an issue in translation today. In M. Grosman, M. Kadric, I. Kovačič & M. Snell-Hornby (Eds.), Translation into NonMother Tongues: In Professional Practice and Training (pp. 35–44). Tübingen: Stauffenburg. St. John, E. (2003). Translating into L2 during Translator Training. Project Papers of the Consortium for Training Translation Teachers (CTTT) Available at WWW < https://web. archive.org/web/20120701075858/http:/isg. urv.es/cttt/cttt/research/stjohn.doc Svoboda, T. (2011). Anketa — průzkum překladatelského trhu. Tlumočení a překlad (ToP), 99, 15–17. David Mraček | Institute of Translation Studies, Charles University, Prague <[email protected]> OPEN ACCESS