Corps de l’article

1. Introduction

Translators clearly deploy a range of skills and draw on different types of knowledge in the exercise of their profession, but which skills or knowledge types are more important than others? Is extensive knowledge of languages more important than the ability to use dictionaries? Is the ability to punctuate sentences correctly more important than solid grounding in economic theory? Greater understanding of the skill-set required would be of benefit not only to translators themselves but also to those who train and employ them, and ultimately to Translation Studies as a discipline and to the profession itself.

Research into what it takes to translate well, which has generally come to be referred to as “translation competence,” has so far concentrated on broadly categorizing skills and knowledge types in order to create a framework for translator training. The study presented here approaches the subject from a different angle and in a particular context: that of recruitment at inter-governmental organizations (IGOs). Rather than contemplating the broad needs of translation graduates about to embark on their careers, the study aims to identify the specific needs of IGO translation services. The first step was to identify the relative importance of the skills and knowledge that IGO translators use (through an “impact questionnaire”) and to employ this as a point of departure for drawing up the candidate profile that organizations should be looking for in their selection procedures. Complementary data was obtained on the relative scarcity of the various components of that profile among new recruits (through a “recruits questionnaire”), and then the findings of the two questionnaires were correlated to produce a weighted listed of skills and knowledge that has implications for recruitment and training at IGOs. Over 300 translators, revisers and heads of service working at 20 IGOs participated in the survey, and the preliminary findings are presented here. The article closes with a brief discussion of the implications of the study and suggestions for further research.

1.1. Translation competence and translation at inter-governmental organizations: a review of the literature

Translation Studies scholars have mainly limited their exploration of translation competence to multi-componential models on which to base translator training. These range from traditional three-part competence models that distinguish a source-language processing component, a target-language processing component, and a strategic component of some kind, such as those of Bell (1991), Cao (1996), Hatim and Mason (1997) and Nord (2005), through to complex models involving 48 separate skills and knowledge types grouped into six “competences” (EMT 2009). The differing conceptualizations, definitions and levels of specification of these models, as well as their proliferation, have been commented on by Orozco and Hurtado Albir (2002), Pym (2003), Arango-Keeth and Koby (2003), Way (2008) and Angelleli (2009). The lack of empirical study devoted to these models has been noted by Waddington (2001) and bemoaned by PACTE (2008), who, together with Campbell (1991), have begun to make some inroads in this regard by producing hard data on the relations between certain components of their own models.

No empirical research seems to have been done, however, into the relative importance of the components of translation competence. Nor, apparently, has a methodology been devised for doing so. And that applies to any context, let alone the rather hermetic world of IGO translation. This is probably because the underlying driver of the discussion of translation competence in Translation Studies has been the need or desire to design and justify translator training programmes. As Pym notes, “[i]n most cases, the complex models of competence coincide more or less with the things taught in the institutions where the theorists work” (Pym 2003: 487). For syllabus planning or justification purposes, the catalogues of competencies loosely grouped under various headings serve as frames of reference. The concerns and interests of academic institutions are not necessarily the same as those of IGO human resources officers, however, and from the IGO recruitment perspective, what is needed is not a catalogue, but a shopping list in which skills and knowledge types are prioritized on the basis of the empirical analysis of the organization’s needs.

So what are the needs of IGOs as far as the profile of new recruits is concerned? Considering the enormous volume of translation work that IGOs handle and the hordes of translators they employ, it is rather surprising that so few studies have been made of IGO translation. This could be the result of lack of interest in the subject among translation scholars, whose reputation for ivory-tower attitudes is such that whole books are published on the issue. Consider, for example, Chesterman and Wagner’s Can Theory Help Translators? A dialogue between the ivory tower and the wordface (2002) and Baer and Koby’s Beyond the Ivory Tower. Rethinking translation pedagogy (2003). It could also be the result of reluctance on the part of IGO translation services to encourage enquiry into their workings. This might in turn stem from the confidential nature of some of the documents IGOs handle, time limitations and other practical constraints, the obligation of staff to protect the image of the IGO, a fear of tarnishing a reputation for high standards of excellence, or a combination of all these and other, unknown factors. In any case, the literature on IGO translation seems to have been produced mainly by those who are or were insiders, such as Koskinen (2000; 2008) and Wagner, Bech and Martinez (2002) from the European Commission’s Directorate-General for Translation, and by those working, if not with IGOs, then with large institutions: Mossop from the Canadian Government’s Translation Bureau (1988; 2001), and Williams (1989) who has worked with the same Bureau as a consultant. To supplement the scholarly literature, we have also examined publications and statements issued by IGOs (e.g., EMT 2009; European Commission 2009; Sekel 2008; Hindle 1965/1984). These works provide insights into the nature of translation work at IGOs, what shapes their approach to translation and the expectations of professional translators but, as in the case of the literature on translation competence, these works do not attempt to establish the relative importance of the components of the set of skills and knowledge required.

2. Methodology

2.1. Main Hypothesis and Definition of Terms

The aim of the study is to identify the translator profile components that IGOs need to look for and to rank them by their importance, particularly from the recruitment perspective. The main hypotheses tested were quite simple:

H1: Some components of translation competence are more important than others in IGO translation;

and

H2: Some components of translation competence are more often lacking than others among new recruits at IGOs.

where:

  1. Components of translation competence refers to the skills and knowledge needed to produce translations that meet required standards. Knowledge is understood as information assimilated through learning, and skills are understood to be abilities that have been developed through training or experience.

  2. Important refers to two vital aspects of the work of IGO translation services: their productivity and their reputation. Translators with the right set of skills and knowledge ideally produce more error-free translations, errors being features of translations that need to be corrected because they undermine the communicative aims of the organization or harm its reputation. Errors may be produced by defective judgment, deficient knowledge or carelessness. The number of errors translators produce has an impact on the productivity of the translation service: the fewer errors there are in the translations, the less revision work they generate and the greater the productivity of the service. Revisers are not infallible, however, and not all translations get revised: the seriousness of errors, defined in terms of the degree to which they harm the prestige of the organization, also needs to be taken into account. IGOs therefore want to find translators who have the skills and knowledge needed to minimize both the quantity and the seriousness of the errors that can be generated in the translation process, in other words, the translators who will have the least negative (or the most positive) impact on the productivity and the reputation of the IGO.

  3. IGOs refers to inter-governmental organizations, which are understood to be organizations comprised mainly of sovereign States, usually referred to as member States. NATO, the United Nations and all its bodies, the various entities of the European Union, the Organization of American States, the OECD and the International Committee of the Red Cross are all examples of IGOs.

  4. New recruits are translators who have been working for the organization for less than one year.

2.2. The design of the survey

Competence cannot be measured directly, but only through product and process analysis, i.e., through the manifestations and displays of that competence. Text analysis, however, can only suggest the skills and knowledge that translators apply when they translate; it cannot rank them in order of importance. Process analysis, meanwhile, is not practical on a large scale as a research method. The remaining way to find out about the skills and knowledge used in IGO translation is to consult the people with the greatest first-hand experience in using and observing their application and repercussions and hence the greatest awareness of their relative importance to the organization. A survey was therefore chosen as the data collection tool.

First, a questionnaire was designed for translators, revisers and heads of IGO translation services in which they would be asked to rate, on a five-point Likert scale, the impact that different skills and knowledge types have on the effectiveness of the translations produced by the organization. “Effective translations” were defined as those that achieve the communicative aims of the organization and protect its image. This will be referred to as the impact questionnaire and it can be found in Annex 1.

As input for translator training or preparation for work at IGOs, the ranking of skills and knowledge types by respondents to the impact questionnaire is useful. To identify recruitment priorities, however, the findings would need to be correlated with data on the relative scarcity of these same skills and knowledge types among new recruits. A high-impact skill or knowledge type that was already being found in abundance through current recruitment procedures (e.g., knowledge of French vocabulary in the case of English translators) would not have the same implications for recruitment as a high-impact one that was not being found in great quantities (e.g., knowledge of unconventional weapons). A second questionnaire was therefore designed for revisers and heads of IGO translation services, as those most familiar with the skills and knowledge of new recruits (self-report data on supposedly former shortcomings being interesting but less valuable). It asks them to rate, also on a five-point Likert scale, the frequency with which new recruits lack different skills and knowledge types. This will be referred to as the recruits questionnaire and it can be found in Annex 2.

The list of skills and types of knowledge presented for rating in both questionnaires was drawn up on the basis of the broad categories identified in the literature on translation competence and the specific skills and knowledge suggested by the review of the literature on IGO translation, as well as on the basis of the feedback from the two pilots that were conducted at one IGO prior to the survey. Two revisers participated in the first pilot, and three translators and four revisers in the second.

The base list of 40 skills and knowledge types presented for rating in the questionnaires is given below. It should be borne in mind that the purpose of the study was to obtain a hierarchy of skills and knowledge types as input for the design of text-based translation tests. Consequently, the level of specificity for the text-production skills is greater than for other types of skill, and interpersonal and other skills and knowledge types that do not lend themselves to text-based testing were excluded. The numbers alongside each skill or knowledge type are those used to identify them in the research. Short descriptions are used for the sake of space. For the full descriptions used in the questionnaires, see Annex 1 or 2. Also for convenience, the following acronyms have been used: SL for source language, TL for target language, SC for source culture, TC for target culture, ST for source text and TT for target text.

  • Knowledge for comprehension: Knowledge of: (1) the SL, (2) different varieties of the SL, (3) the SC, (4) the subject, and (5) the organization;

  • Analytical skills: The ability to: (6) understand complex topics, (7) master new subjects quickly, (8) work out the meaning of obscure texts, (9) detect inconsistencies contradictions, nonsense, etc. (10) detect logical, mathematical or factual errors, and (28) ensure the coherence of the TT;

  • Target-text production skills and knowledge: Knowledge of: (12) TL spelling, (14) TL punctuation rules, (13) the finer points of TL grammar, (23) TL varieties and (24) the TC. (11) An extensive vocabulary in the TL. The ability to (15) produce idiomatic translations, (16) produce translations that flow smoothly even when the ST does not, (17) select and combine words in the target language to capture the exact and detailed meanings (nuances) of the ST, (20) convey the ST message clearly, (21) convey the intended effect of the ST, (25) tailor language to the reader, (22) achieve the right tone and register, (19) produce an elegantly written target text regardless of how elegantly written the source text is, (18) recast sentences in the target language (to say the same thing in different ways), and (27) ensure the completeness of the TT;

  • Research skills: The ability to track down sources to: (30) understand the topic, (29) check facts, (31) mine reference material for accepted phrasing and terminology, and to (32) judge the reliability of sources;

  • Computer skills: The ability to: (33) type accurately and fast, work with (39) more than basic Word functions, (37) translation memory software, and (40) Excel and PowerPoint, and to (38) make effective use of terminology tools;

  • Other skills: The ability to: (34) maintain quality under pressure, (35) explain translation decisions, (36) follow complicated instructions about what to do with a text, and (26) adhere to in-house style conventions.

Respondents were also asked in both questionnaires to report and rate any other skills or types of knowledge not presented in the base list. Questions on text-processing responsibilities, editing and revision practice were included to contextualize findings about some of the skills and knowledge types. Being able to create pie charts in Excel is of little importance, for example, if translation services have a graphic design department that does it for them. And the importance of being able to translate poorly written originals depends on how systematically source texts are edited. Direct questions to obtain information on the expectations of new recruits at the IGOs were also included to contextualize the findings for the purposes of recruitment test design. A high frequency rating in the recruits questionnaire (i.e., the suggestion that a skill or type of knowledge is often lacking) might not be relevant if the skill or knowledge type in question can be developed after recruitment through training or in-house experience. Knowledge of the organization and in-house spelling rules, for example, can usually be mastered on the job.

2.3. The survey respondents

To reach the largest possible number of translators and revisers at IGOs, the backing of the International Annual Meeting on Language Arrangements, Documentation and Publication (IAMLADP) was sought and obtained. IAMLADP is a forum and network of managers of international organizations employing conference and language service providers, mainly translators and interpreters. The questionnaires were prepared in electronic format and sent out in February 2010. Respondents received an e-mail inviting them to complete the questionnaires online, and data were collected over a period of three months. The sample population was relatively homogeneous inasmuch as the respondents worked in the same domain, had similar academic profiles and had all passed similar recruitment examinations. The final realized sample included 165 completed impact questionnaires, of which 163 were usable, and 157 completed recruits questionnaires, of which 153 were usable. The answers from all the usable questionnaires were included in the analysis.

The impact questionnaire was completed by 7 heads of unit, 27 revisers and 129 translators from over 24 bodies of the United Nations (UN), the European Union (EU) and other organizations.[1] Translators working in a wide range of language combinations (24 different languages) participated. The recruits questionnaire was completed by 128 revisers and 25 heads of service from over 20 bodies of the United Nations, the European Union and other IGOs.[2] The results are based on the work of new recruits revised by revisers working with 22 languages among them.

3. The results of the questionnaires

3.1. The results of the impact questionnaire: the relative impact of skills and knowledge types

The mean impact ratings for the skills and knowledge types presented in the impact questionnaire are shown in Table 1. Ratings were awarded in response to the question:

For the purposes of this survey, effective translations are those that achieve the communicative aims of the organization and protect its image. How large is the impact of the following skills and knowledge types on the effectiveness of translations at your organization? 1 = extremely small; 5 = extremely large; N/A = not applicable because the skill or knowledge is not required.

Table 1

Impact ratings of selected skills and knowledge types[a]

Impact ratings of selected skills and knowledge typesa

(Response count and mean rating)

a

where TT = target text; TL = target languages; SL = source language; ST = source text. Descriptions of skills and knowledge have been abbreviated for the sake of space. For the full descriptions, see the impact questionnaire in Annex 1.

-> Voir la liste des tableaux

Being able to make sure that all the content of the original is relayed in the translation is apparently the most important skill translators need to have at IGOs. Although a rather fundamental aspect of the translator’s task in any context, it may assume particular importance in certain IGO settings such as committee meetings, in which participants are working from parallel language versions and any omissions (or unnecessary additions for that matter) may incur comment, confusion and delays and hence embarrassment and costs. Being able to express ideas clearly in the target language and having sound knowledge of the source language are also unsurprisingly rated as highly important assets. Ensuring coherence, i.e., internal consistency within each translation and all the documents produced by their translation services, is another understandable priority in IGO translation services. Translations are the written records of the institution’s workings. Possibly more unexpected is the fact that analytical skills are among those rated as being highly important (mean rating of 4.5 or above). The ability to work out obscure passages in the source text and the ability to detect inconsistencies, contradictions, nonsense, unintended ambiguities, misleading headings, etc., are ranked 7th (mean rating 4.64) and 14th (mean rating 4.54), respectively. This may be explained by the need to translate unedited source texts. The importance of being able to maintain quality even when working under pressure of time (ranked 12th) reflects another common aspect of IGO translation work, especially when servicing meetings.

It should be noted that even the skills and knowledge types at the bottom end of the ranking produced by the impact questionnaire are not considered unimportant. On a scale of 1-5, a rating of 3 or more suggests the skill or knowledge type is at least of some importance. This applies to 38 of the 40 included in the list. The only two to which it does not apply are the ability to detect mathematical errors and the ability to work with Excel and PowerPoint, which were also reported as not required at all by about 12% of respondents. In other words the vast majority of skills and knowledge types presented for rating were considered to have an impact of the effectiveness of IGO translation work.

The hierarchy reflects the opinions of 163 translators, revisers and heads of unit working in over 20 translation services. As such it provides an indication of the relative importance that professional translators working for inter-governmental organizations collectively attach to the skills and knowledge required on the job. As Kelly notes, professional considerations are one of the factors that should define the learning objectives of training programmes (Kelly 2005: 22-28). This information could therefore be of interest to translators thinking of working for IGOs and translator training institutions. The fact that knowing how to spell correctly is apparently more important than being able to work with translation memory software, for example, may come as a slight surprise, as might the low ranking of cultural knowledge.

3.1.1. The additional skills and knowledge reported in the impact questionnaire as being needed on the job: an insight into translation at IGOs

The optional question in the impact questionnaire on other skills and knowledge required was answered by 48 people (29% of respondents), and general observations were made by 50 (31%). The responses provide several insights into the nature of translation work at IGOs today. The skills and knowledge types that, according to the survey participants, are required in addition to the 40 included in the ranking can be loosely grouped as shown below. The number of times each one was mentioned is presented in parentheses, and the respondents’ own wording is used as much as possible:

  • Communication skills to be able to elicit assistance or answers from others in the organization, especially authors of source texts (11)

  • Ability to work with revisers: openness to feedback and ability to learn from it (10)

  • Team work skills, including for working on large translation projects and with shared translation memories (8)

  • Language skills: ability to learn new languages, translate from several source languages, ability to create new terminology for new concepts, to adapt to the style used in the unit (8)

  • Knowledge of translation theories and practices (6)

  • Critical thinking skills (ability to “reason through” the translation, a “second channel” in the brain to think beyond just choosing the right word) (5)

  • Organization and time management skills (5)

  • General knowledge, awareness of current and world events (5)

  • Interpersonal skills: patience with authors, requesters and colleagues (4)

  • Flexibility, adaptability to cope with unpredictable workload, changes in procedures, etc. (4)

  • Experience in other international organizations, ability to cope with life in another country, awareness of cultural differences (3)

  • Ability to work independently (3)

  • Ability to deal with a wide range of specialized subjects (3)

  • Self-revision skills (1)

  • Willingness to search for information (1)

As shown in the above list, interpersonal skills such as the ability to work as a member of a team and communicate with “clients” are mentioned more than any other type. Several comments on the need to be able to seek clarification from authors were accompanied by remarks on the poor quality of source texts and the ability to argue points diplomatically and “not to be rattled by criticisms of one’s English from people not really qualified to make them,” as one translator put it.

There were also complaints about the restrictions imposed by tradition and conventions. One respondent commented that “[c]omplying with the in-house rules and conventions is sometimes more important than the ability to reproduce a text elegantly or even accurately,” and another stated that “translators have to ‘forget’ their general knowledge and use the terms agreed on at the inter-institutional level even if they don’t agree with them or have a better suggestion.”

In addition to suggesting other skills and knowledge types, several respondents to the impact questionnaire took the opportunity to emphasize the importance of expert subject knowledge (of economics and legal systems for instance). This was ranked only 29th by respondents across all organizations, possibly revealing the prime importance of this knowledge in some organizations (rated 5 by 70 respondents) but not in others.

A number of respondents said they had nothing to add and commented on the comprehensiveness of the list of skills and knowledge types presented for rating. Some took the opportunity to sum up what they thought translators at IGOs need. One EU translator wrote: “Basically, we need a thorough knowledge of source languages and the ability to efficiently find whatever info/terminology we need and convey the message in the target language clearly while adhering to in-house rules, often [translating] from source texts of poor linguistic quality.” What would appear to be a seasoned reviser at the ILO noted: “The skills required haven’t changed much: thorough knowledge of source and target languages, highly developed analytical skills, and the ability to assimilate new knowledge quickly.” Both, incidentally, mention two not wholly unrelated aspects of IGO translation that are not usually given much attention in translator training programmes or recruitment: the translation of poorly written source texts and the need for superior analytical skills.

3.2. The results of the recruits questionnaire: the relative scarcity of different skills and types of knowledge among new recruits

The mean frequency with which each skill or knowledge type was found lacking among new recruits is presented in Table 2. Ratings were awarded in response to the question:

Please think about the mistakes you usually correct when going over translations done by new recruits. How often do you think those mistakes are due to a lack of the following? 1 = Almost never and 5 = Almost always.

Table 2

Frequency with which selected skills and knowledge types are found lacking among new recruits[a]

Frequency with which selected skills and knowledge types are found lacking among new recruitsa

(Number of responses and mean rating)

a

Where TT = target text; TL = target languages; SL = source language; ST = source text. Descriptions of skills and knowledge have been abbreviated for the sake of space. For the full descriptions, see the recruits questionnaire in Annex 2.

-> Voir la liste des tableaux

New recruits at least sometimes lack 18 of the 40 skills and knowledge types that revisers were asked to rate on the five-point scale (these were all rated above 3). Of these, almost half (8) are related to target-language writing skills; 3 are research skills, 2 refer to analytical skills (working out obscure passages and detecting inconsistencies); 2 to specialized knowledge (of subject areas and of the organization), 1 to capturing detailed and exact meanings (nuances), 1 to in-house style and 1 to productivity. In other words, in addition to linguistic skills traditionally associated with translation and deficient TL writing skills usually associated with revision work, analytical skills (the ability to detect inconsistencies in the source text and to judge the reliability of sources) are among those rated, on average, as more often than not lacking among new recruits. Indeed, the relative inability of new recruits to figure out meaning is generating almost as much revision work as their (expected) ignorance of in-house style. And this shortcoming is not due to their lack of language skills since their SL knowledge is only relatively rarely found to be deficient (ranked 22nd with a mean frequency rating of only 2.78).

3.2.1. The additional skills and knowledge types reported as lacking among new recruits

The optional question on other skills and knowledge types that new recruits often lack and need was answered by 26 people (17% of respondents), and general observations were made by 48 (31%). The answers have been grouped as follows (the number of times each one was mentioned is presented in parenthesis) and respondents’ own wording is used as far as possible:

  • General knowledge, awareness of current affairs (10)

  • Organization and time management skills, including ability to attain the right balance between speed and quality (7)

  • Quality management skills: how and when to adjust criteria, ability to check for internal consistency and read for sense (7)

  • Knowledge of translation and terminology theories, strategies and practices (6)

  • Target language skills: thorough grasp of own language, correct grammar; the ability to write idiomatically, avoid “apishly reproducing the structure of the source text,” and achieve the right tone and register (6)

  • Ability to work with computer-based translation memory, glossary-building, machine-translation and voice-recognition tools and to judge the results obtained (5)

  • Client orientation, professional conduct (3)

  • Awareness of the SL structures that pose translation challenges (2)

  • Sensitivity to the subtext and implications of what the text is saying (2)

  • Ability to approach and work effectively with authors of source texts (2)

  • Team work skills (2)

  • Ability to work with poorly written source texts (1)

  • Ability to work with revisers: openness to feedback and ability to learn from it (1)

  • Willingness to take the time needed to search properly for information (2)

Several revisers also took the opportunity in the comments section to stress the shortage of certain components of the skill and knowledge set: subject or technical knowledge (ranked 7th in the frequency ratings), for example, was mentioned again eight times, and knowledge of the organization and its activities (ranked 13th) was referred to again three times. There were in general more remarks about the lack of target-language writing skills than about deficient source-language comprehension skills.

The component most often mentioned as lacking among new recruits (in addition to those presented for rating in the recruits questionnaire) was general knowledge. This is considered vital because, as one reviser put it, “[t]he more understanding you have of the general background, the better are your translations.” There were several comments on how younger translators in particular rarely read newspapers or news magazines to keep themselves informed of current affairs.

There were a number of observations about the inferior quality of the translations produced by new recruits in general: they “tend to translate words and not meanings”; “[s]ince most documents are written by non-native speakers, the translations can be meaningless”; they lack the ability to “step back” and read their own work for “internal consistency.” Also, reaffirming the high ranking of the abilities to produce idiomatic translations and to produce translations that flow smoothly, some respondents reported that there is “too much tendency to stick to the [original language]” and “translations are too literal.” A number of revisers also expressed concern over an apparent lack of translation strategies and techniques (mentioned six times in the comments), which is possibly reflected in the relatively high ranking of the ability to recast sentences.

Several comments were made about time and workload management and having to work to increasingly shorter deadlines. Some lamented new recruits’ inability to keep up; others lamented the lack of room afforded under the more “market-oriented approach” for “continuous learning by (new) translators, which is a vital prerequisite to staying abreast with new knowledge.”

There were also comments about working with poorly written originals, which require the “ability to understand, navigate through and decode the various types of ‘pidgin’ English […which] is becoming more and more usual within international organizations.” These came from revisers working at organizations in which documents are only rarely sent for editing prior to translation.

The number of comments on the need for translators with advanced computer skills (5) seems to point to two views of training needs among revisers. There are those who seem to think translators should have mastered the relevant technology before recruitment, and others who think there is a need to get “back to basics,” instead of worrying about how to use new technologies, which they maintain can usually be learned quickly on the job.

4. Using the findings of the two questionnaires to identify recruitment and training priorities

The findings of both the impact questionnaire and the recruits questionnaires have implications for recruitment at IGOs. Adjusting selection procedures to identify candidates with the profile that emerges from the findings of the impact questionnaire should help organizations find translators who will have the most positive impact on the effectiveness of their translations. Adjusting selection procedures according to the findings of the recruits questionnaire, meanwhile, should help reduce the revision workload and hence enhance the quality and efficiency of the translation service. The findings of the two questionnaires therefore need to be correlated. The relative impact of a component of the skill and knowledge set has a bearing on how much attention should be paid to its relative scarcity among new recruits. Or to put it the other way round, the relative scarcity of a skill or knowledge type determines how much attention should be paid to its impact. Subject knowledge, for example, is often found to be lacking (ranked as the 7th most frequent cause of revision work) but is rated as relatively less important than 28 other components of the skill and knowledge set in the impact questionnaire. Conversely, knowledge of the source language, which is ranked as the 4th most important component, seems to be relatively easy to find among new translators or is being well tested for, inasmuch as it only rarely or sometimes accounts for the errors that revisers have to correct (mean frequency rating of 2.74). In order to improve the quality of translation work and reduce the revision load, organizations need to recruit more translators who, in addition to the more important skills and knowledge types already being successfully assessed in recruitment, have the important ones that are currently not being found in sufficient quantities. These skills and knowledge types can be pinpointed by mapping the impact and frequency ratings of the components of the skill and knowledge set on a scatter chart. Priorities can then be determined by applying the logic of qualitative risk analysis, as described by Tusler (1996) for example, in order to classify the components into four categories: tigers, foxes, anteaters and hedgehogs (an adaptation of Tusler’s equally whimsical labels), each with their own implications for training and recruitment.

Figure 1 shows the correlation of the impact and frequency ratings for the skills and knowledge types on the base list presented for consideration in the corresponding questionnaires. The other skills and knowledge types mentioned by respondents as being important in IGO translation or as sometimes lacking among new recruits (mainly general knowledge and interpersonal skills) have not been included because the focus of this research is on obtaining input for text-based translation skills testing. These additional types of skills and knowledge are obviously highly relevant, however, and should be factored into candidate selection. To identify recruitment priorities, impact ratings of 4.5 or more and frequency ratings of 3 or more are considered high. These cut-off points are not completely arbitrary: they isolate the high-impact skills and knowledge types that at least sometimes account for the errors that revisers have to correct. The resulting groupings and the implications of classification in each grouping are presented below. The items within each category are listed by cluster from top right to bottom left, i.e., from the highest priority to the lowest from the recruitment perspective. The numbers in brackets correspond to the numbering used in Figure 1.

Figure 1

Inter-governmental organizations: skills and knowledge types by frequency found lacking among new recruits and impact on the effectiveness of translations

Inter-governmental organizations: skills and knowledge types by frequency found lacking among new recruits and impact on the effectiveness of translations

Key to figure

  1. Knowledge of SL

  2. Knowledge of SL varieties

  3. Knowledge of SL culture(s)

  4. Subject knowledge

  5. Knowledge of the organization

  6. Understand complex topics

  7. Master new subjects quickly

  8. Work out the meaning of obscure passages

  9. Detect inconsistencies, contradictions, etc.

  10. Detect mathematical errors in the ST

  11. An extensive TL vocabulary

  12. Knowledge of TL spelling rules

  13. Knowledge of TL grammar

  14. Knowledge of TL punctuation rules

  15. Produce idiomatic translations

  16. Produce translations that flow smoothly

  17. Capture nuances of ST

  18. Recast sentences in the TL

  19. Write elegantly regardless of the ST

  20. Convey the ST message clearly

  21. Convey the intended effect of the ST

  22. Achieve the right tone and register

  23. Knowledge of TL varieties

  24. Knowledge of TL culture(s)

  25. Tailor language to the readers’ needs

  26. Adhere to in-house style conventions

  27. Ensure the completeness of the TT

  28. Ensure the coherence of the TT

  29. Track down sources to check facts

  30. Track down sources to understand the topic

  31. Mine reference material for phrasing

  32. Judge the reliability of information sources

  33. Type accurately and fast

  34. Maintain quality even under time pressure

  35. Explain translation decisions and problems

  36. Follow complicated instructions

  37. Work with translation memory software

  38. Work with electronic terminology tools

  39. Handle more than basic Word functions

  40. Work with Excel and/or PowerPoint

-> Voir la liste des figures

4.1. The hedgehogs: low-impact commonly found skills and types of knowledge

  • Understand complex topics (6), Work with electronic terminology tools (38), Follow complicated instructions (36), Master new subjects quickly (7), Tailor language to the readers’ needs (25)

  • Explain translation decisions and problems (35), Knowledge of TL culture(s) (24)

  • Work with translation memory software (37)

  • Knowledge of SL culture(s) (3)

  • Knowledge of TL varieties (23)

  • Handle more than basic Word functions (39), Knowledge of SL varieties (2)

  • Type accurately and fast (33)

  • Detect mathematical errors in the ST (10)

  • Work with Excel and/or PowerPoint (40)

The hedgehogs are the lowest recruitment priorities. New recruits usually have these skills and types of knowledge, and their impact on the effectiveness of translations at IGOs is relatively small. They are scattered quite widely within the group. At one end of the spectrum, the ability to understand complex topics is very close to being classified as a fox. At the other end, typing, mathematical and word-processing skills are rated so low that they probably do not need to be factored into assessment at all. The presence of cultural knowledge in this group is somewhat surprising considering the amount of comments made on the need for general knowledge.

4.2. The foxes: high-impact commonly found skills and types of knowledge

  • Knowledge of SL (1)

  • Ensure the completeness of the TT (27)

  • Knowledge of TL grammar (13), An extensive TL vocabulary (11), Knowledge of TL punctuation rules (14)

  • Knowledge of TL spelling rules (12)

The foxes are needed because of the impact they have on the quality and productivity of the translation service. They include the basic prerequisite of source language knowledge and the ability to ensure a fundamental quality of a good translation: that it transmits all the information contained in the source text. These skills and types of knowledge are not often found lacking among new recruits and are therefore not generating much revision work, which means they are either relatively easy to acquire or are being found in relative abundance through current selection procedures. They are certainly all routinely tested in recruitment examinations.

4.3. The anteaters: low-impact oft-lacking skills and types of knowledge

  • Produce translations that flow smoothly (16), Achieve the right tone and register (22)

  • Recast sentences in the TL (18), Knowledge of the organization (5), Judge the reliability of information sources (32), Write elegantly regardless of the ST (19), Track down sources to understand the topic (30), Subject knowledge (4)

The anteaters also matter: although they do not have such a great impact on the effectiveness of translations at IGOs, their absence generates a large proportion of revision work. The first two in this category, the abilities to produce translations that flow smoothly and to achieve the right tone and register, came close to being classified as tigers.

4.4. The tigers: high-impact oft-lacking skills and types of knowledge

  • Work out the meaning of obscure passages (8), Capture nuances of ST (17)

  • Convey the ST message clearly (20), Ensure the coherence of the TT (28), Adhere to in-house style conventions (26), Detect inconsistencies, contradictions, etc. (9)

  • Produce idiomatic translations (15), Maintain quality even under time pressure (34), Convey the intended effect of the ST (21)

  • Track down sources to check facts (29)

  • Mine reference material for phrasing (31)

Tigers should be top recruitment priorities. Not only is their absence generating a lot of revision work, but also they have a strong impact on the effectiveness of the translations that IGO translation services produce. These are the skills that IGOs need to find more of, in addition to the skills and knowledge in the fox and anteater categories that they already successfully assess. Alternatively they need to develop these skills in-house soon after recruitment.

4.5. The ideal candidate

The profile of the ideal candidate for IGO translation work that emerges from this study therefore is of a translator who, in addition to strong target-language writing abilities, sound source-language knowledge and good research skills, has superior analytical skills and a keen eye for detail to work out obscure meanings, to rectify or flag inconsistencies and to capture the nuances of the original. Furthermore, this translator can produce clearly written idiomatic target texts that are internally consistent and convey the intended effect of the source text. And they can do so fast. Ideally they also have thorough knowledge of in-house style conventions, good research skills and experience of working with background documents. Translators with that specific profile are the ones that IGOs, based on the views of their translators and revisers, need to find in greater quantities.

5. Discussion

These findings correspond to IGOs as a whole, and the hierarchy presented above suggests the profile of the candidates that IGO should in general be looking for if they wish to improve the output of their translation services in terms of both quality and productivity. It could therefore be of interest to potential applicants for IGO translation work and their trainers.

To be of value to IGO recruitment officers, however, these findings need to be broken down further, by individual organization and possibly even by language service since the expectations of new staff translators and the opportunities for post-recruitment skills development might vary considerably from one organization to the next. The profile of the ideal candidate at NATO, for example, might not be the same as at the EU, and the skill set required in the Arabic translation service at the United Nations Office in Geneva may not be the same as the set required in the Russian service. The composition and weighting of the set of skills and knowledge required by in-house staff might also differ from that of external freelance translators. A more focused analysis could also enhance decisions about where the cut-off should be when classifying skills and knowledge types as high or low impact and high or low frequency when correlating the results of the two questionnaires.

Once broken down by organization or language service, the candidate profile can be adjusted according to expectations of new recruits. High-priority skills and knowledge that translators are not expected to have when they first join the organization can become on-the-job training priorities. This might apply to mastery of in-house style guidelines, the ability to work with reference documents and knowledge of the organization, for example. The weighting of the remaining components of the skill and knowledge set can then be used to identify adjustments that need to be made to current testing practice. The high-impact, oft-lacking components (the tigers) that are not being weighted heavily or assessed at all in recruitment examinations should be incorporated and weighted heavily in grading schemes, certainly more heavily than the skills and knowledge types in the other categories. By the same logic, the components in the fox and anteater categories need to be weighted between the hedgehogs and the tigers. Such adjustments should help IGOs find more candidates who match the particular profile they seek.

6. Conclusions

The study confirms that translators need more than language skills: in addition to general and sometimes specialized knowledge, they need analytical, research, technological, interpersonal and organization skills. It also shows that some skills and knowledge types are more important than others and ranks them by their importance in the context of IGO translation. The study also identifies the most common shortcomings of new recruits at IGOs. Whether these are the result of current IGO selection procedures or a general characteristic of novice translators is a matter for further research. The weighted list of skills and knowledge types obtained by correlating the impact and frequency ratings indicates the profile that IGOs need to look for and, by extension, the skills and knowledge that translators and translator trainers interested in this market niche may wish to develop. The high ranking of analytical skills in both questionnaires poses an interesting challenge to translator training. When considered in light of current testing practice and the expectations of new recruits, the skills and knowledge profiles obtained through the methodology presented here can be used to identify recruitment and/or on-the-job training priorities not only in specialized branches of translation, but possibly also in other fields in which new recruits are closely monitored, such as interpreting.