RSS feed

Training translators to work for the EU institutions: luxury or necessity?

Vilelmini Sosoni, Ionian University of Corfu, Department of Foreign Languages, Translation and Interpreting


The European Union (EU), a sui generis geopolitical entity, non-comparable to any other due to its supranational character, unquestionably gives rise to a multicultural and multilingual activity which is unique and as such quite challenging for all those involved.
The idiosyncrasy of the EU and of the text production and translation that take place within its institutions poses particular problems to translators, interfering with their competence and ability to succeed in the EU’s vigorous and painstaking translator selection process. Similarly, the EU has trouble recruiting translators who meet its quality requirements and are able to deal successfully with the specificities of its official texts. Our claim is that the problems which EU translators face cannot be solved ‘on the job,’ without prior special training. This training has to be offered by universities in close collaboration with the EU itself. It needs to focus on the socio-political reality of the EU and its unique linguistic status quo and include extensive hands-on practical translation work with a clear focus on the linguistic conventions and demands of EU texts.


Translator training, translation teaching, linguistic equality, translation quality, EU texts, translation norms, intertextuality, standardisation of terminology, the translation profession.


The European Union (EU) is a democratic federation of 27 equal nations and around 498 million people which was founded almost sixty years ago and whose aim is the unification of Europe, a task that is based on a very simple principle: uniting the nations of Europe while fully respecting their cultural and linguistic diversity. Since its inception, in 1952, Europe has been developing at high speed at various levels; it has integrated national currencies and taxes and has established a common body of law, the acquis communautaire, which governs aspects varying from environmental pollution to bacteria found in cheese. Still, no such ‘integration’ occurs with respect to language, and that is owed to the fact that the founding fathers of the EU, the authors of the Treaties of Rome, recognised right from the beginning the importance of language as the bearer of the cultural identity of a people (Brackeniers 1995: 13) and on April 15 1958 they agreed on a policy of multilingualism. In particular, they adopted Council Regulation No 1 which guarantees that the official languages of all the member states are both official and working languages of the EU institutions and are considered to be equal (Šarčević 2001: 314).

Therefore, instead of using just a couple of languages like other supranational organisations do, like for instance the OECD, the EU currently uses 23 languages, i.e. the official languages of its 27 member states. The reason for that is inextricably related to the EU’s nature, the struggle for unity in diversity. In addition, and as Koskinen (2000: 53) points out, most Europeans consider it to be essential. Babiniotis (1996: 393) points out that a multicultural and multinational union like the EU, whose aim is the synthesis of European countries and not their fusion—which in fact would lead to nothing but a political utopia—is by definition multilingual.

Besides the reasons mentioned above, the policy of multilingualism is essential because the EU is very different from a traditional intergovernmental organisation. Its decisions and legislation affect directly the lives of its citizens and therefore these have to be published in all the member states’ official languages. In addition, the EU’s proposals must be aired for debate at all levels—European, national and local—in forms accessible to non-linguists and non-diplomats. It would be very hard to justify a Union using a lingua franca that most of the people subject to its laws and regulations could not understand (Koskinen 2000: 50). Moreover, EU institutions, despite their complicated procedures, aim at being “as open as possible” (Brackeniers 1995: 13) to the general public as well as to government departments and official and unofficial interest groups of all kinds. This too is reflected in Council Regulation No 1: Articles 2 and 3 lay down that the residents of the member states have the right to communicate with the EU institutions in their official language.

1.1. Translation organisation in the EU

Translation and interpreting have played an essential role in maintaining the delicate balance between supranational and national institutions and have enabled the principle of multilingualism to survive throughout the successive phases of enlargement despite the ever-increasing cost. Every new addition of official languages has been accompanied by an increase in staffing and resources for the translation and interpreting services.

Since EU institutions are responsible for deciding on a wide range of policies and for lawmaking, the volume of language work they produce is considerable. In Brussels, three EU bodies have a permanent ongoing translation activity; these are the European Commission, the European Council and the Economic and Social Committee. In Luxembourg, the European Parliament, the Court of Auditors, the Court of Justice and the European Investment Bank each has its own translation service. In addition, in Luxembourg, the Translation Centre for the Bodies of the EU (CdT) is a Community agency which provides translation services required by the specialised decentralised agencies of the EU. It also provides services to the Community institutions and bodies which have their own translation service in order to absorb any peaks in their workload. Finally, the European Central Bank in Frankfurt also has its own translation service. However, the Commission, which according to Dollerup (1996: 297) is and will remain in the foreseeable future the largest and most powerful institution in the EU, has the largest and most complex translation service in the world. Indicatively, according to Commission’s Directorate General of Translation (DGT), i.e. its translation service, there are 1,750 translators working full-time on translating documents and on other language-related tasks (European Commission 2009a), accompanied by some 600 support staff in management, secretarial, communication, information technology and training functions (ibid.).

To cope with a level of demand that keeps increasing, especially under the light of successive enlargements, most EU institutions call on external translation providers (translation companies or freelance translators) to help with the translation of official EU documents. In particular, the Commission’s DGT, which has always called on external translation providers, indicates that recourse to this option has increased considerably from 1998 to 2008, with the number of pages translated externally reaching 475,000 in 2008, the equivalent of approximately 26 % of total output.

Despite the workload, high standards are considered vital in the translation of EU texts not only by in-house members of staff but by external contractors as well. At this point, it should be noted that the texts that the translators are asked to translate vary from purely legal texts to speeches and speaking notes, briefings and press releases, international agreements, policy statements, reports, green and white papers, answers to written and oral parliamentary questions, technical studies, minutes, internal administrative matters and staff information, scripts and captions for films and other promotional material, correspondence with ministries, firms, pressure groups and individuals, and publications of every size and format on a huge range of topics for opinion-formers and the general public (Wagner, Bech and Martinez 2002). Many of these texts very often have intricate political, legal or financial implications, so mistakes can be very expensive. Other texts aim at explaining the European project to the general public in the EU's 27 member states, a rather complex task, when we consider how difficult European leaders find it to get the message across to the citizens of their own countries. In such cases, faulty or unsuccessful translations can have even more costly repercussions.

In order to ensure high standards, the translator selection process is vigorous and painstaking, and the task of the translators who work for the EU institutions is demanding and challenging. In line with the above, the working conditions are also quite attractive, at least in terms of remuneration and especially in the case of staff translators. It is indicative that in the Commission an entry level gross salary for a translator is 4,267.72 euros per month, not including extra entitlements and allowances1 (European Commission 2010 a). For contractors, remuneration may be less attractive, but it is still very competitive especially in comparison to other translation genres. One would therefore expect EU institutions and EU contractors to be spoilt for choice when it comes to translator selection. Unfortunately, the situation is very different from the one anticipated.

1.2. The thorny issue of translator selection

I will outline the state of play by providing a personal experience. In 2003, I was recruited by Attica-Attimedia, an Athens-based translation company which was a contractor of the European Parliament, the European Central Bank (ECB) and the European Commission. One of my first missions as a Translation Manager was to expand the database of translators that the company could use for its EU projects. “Easy enough,” I thought and I embarked on the hunt confident and optimistic that I would carry out the task quickly and effectively. I was wrong.

The process was straight-forward: I would advertise for freelance and in-house translators and I would then go through the CVs, make a first selection on the basis of the candidates’ studies (a Master’s degree in Translation was desirable) and experience (at least 3 years of proven experience). I would then ask shortlisted candidates to carry out a test at home which involved the translation into their mother tongue, i.e. Greek, of a 300-word Source Text (ST) in the language or languages of their choice (mainly English, French, German and Spanish). The ST was not particularly demanding, although it included specialised EU terminology. Yet given that candidates had plenty of time and a plethora of resources (Eurodicautom, TIS, Eur-Lex, Eurovoc, etc.) to carry out the translation, a satisfactory outcome was expected. All translation tests were assessed by very experienced translators and proofreaders who had been working for the company for many years and who also had experience of working as freelance translators for the European Commission, the European Parliament or/and the ECB. The quality values used were the ones also widely used by the EU institutions for the assessment of all contractors’ work: i.e. Unacceptable, Below Standard, Good, Very Good.2  Good and Very Good were the only two values accepted by the EU and consequently by the Attica-Attimedia. Given that around 73% of EU documents are originally drafted in English (European Commission, 2009 a), the results of the candidates’ assessment who chose to translate from English into Greek constitute the focus of this paper and are provided in Figure 1 below.

Sosoni image
Figure 1: Translation test results 1/Attica-Attimedia 2003.

It is apparent that the outcome of the hunt was far from satisfactory, given that only 17% of the candidates passed the test, i.e. scored Good or Very Good. Yet I did not give up, since I thought that the result might be coincidental. A couple of months later I gave it another shot. I advertised for the positions, followed the same selection procedure and, surprisingly, came up with similar results, which can be seen in Figure 2 below.

Sosoni image
Figure 2: Translation test results 2/Attica-Attimedia 2003.

What is worse is that around 50% of the top candidates, those who scored Good or Very Good at the preliminary test, failed to produce translations of similar quality when asked to carry out an actual EU project. In other words, only 9%, and in real numbers 9 translators, lived up to the expectations and could actually work for an EU project.

Seven years later, in 2010, I asked the same translation company I worked for in 2003, i.e. Attica-Attimedia, and ECI Athens, a translation and captioning company with offices in Athens and London and a contractor of EU institutions, to provide me with information about the success rates of their translation candidates. The candidates were selected in a similar manner to the one described earlier and were subsequently tested in the translation into Greek from English. The success rates were once again discouraging. In particular, according to the data provided by Attica-Attimedia fewer than 10% of its applicants managed to score Very Good or Good at the translation test, while for ECI that rate was just short of 14%. What is also interesting is that for most non-EU projects (legal texts, technical manuals, even subtitling projects for various clients ranging from Ministries and public services to software and pharmaceutical companies) both Attica-Attimedia and ECI had no particular problems recruiting translators who passed the preliminary test (a test similar to the one used for EU texts in terms of length and level of difficulty). In fact, according to the data provided by ECI,3 around 60% of the translation candidates scored Good or Very Good at the translation test, while according to the records kept by Attica-Attimedia4 the success rate was around 55%.

Yet it is not only translation companies which face recruitment problems. EU institutions, which recruit translators directly through a vigorous selection procedure, share similar concerns since only “30% of those applying are successful” (Euractiv 2010). If we also take into account the fact that many staff members in the EU institutions’ language departments are approaching retirement, it is easy to understand why fears of a succession crisis are growing.

In the same spirit, students at both undergraduate and postgraduate levels express their frustration at not being able to cope with the specialised EU texts and at failing open EU competitions as well as translation tests set by contractors of EU institutions.

A number of questions naturally arise: What are the reasons for that lamentable success rate? What is being done and what should be done to address the problem swiftly and effectively?

2. Specificities of EU texts and their translation

Texts, as Stoddard (1991: 3) claims, cannot be considered as entities independent of the producer, processor, and environment in which they are generated and received. On the contrary, they are tied up with cultural, social and political realities (Dollerup 1996: 312) or, in the case of multilingual texts such as the ones produced by the EU institutions, with “socially developed intercultures” (Pym 2004: 1). Therefore, the broader societal and political as well as intercultural and linguistic framework in which such texts are embedded cannot be overlooked. In the case of the EU, it is widely accepted that its activities are very complex. In fact, Kennedy (1998: 57) points out that getting to grips with the activities of the EU institutions is rather like wrestling with jelly, while Dollerup (2001: 284) observes that “unless one is superhumanly gifted,” it is difficult to understand the complexities of EU work and its language work.

As pointed out in the introduction, EU translation is shaped under the EU’s overwhelming cultural and linguistic diversity and the constraints of European multilingualism, while according to Gibová (2009: 145) it “operates within a distinctive conceptual structure at Community level.” Thus, it is not surprising that both text and translation production in the EU institutions are complex and idiosyncratic.

2.1. Translation quality: a term revisited

One key notion which is closely tied to the translation of EU texts and which contributes to its idiosyncrasy is that of quality. Earlier it was indicated that translators fail to produce quality translation work, i.e. translations which can qualify as Good or Very Good in EU terms. But what exactly is quality and how can it be defined? First of all, it should be stressed that, according to the European Commission’s DGT and the European Parliament’s DGTRAD, quality is a key concern. In particular, one of the main tasks of the European Parliament’s DGTRAD is in its own words “supplying a translation service which ensures both quality and efficiency” (European Parliament 2010). Similarly, the Commission’s DGT claims that ensuring quality is not only a main concern, but also its duty as part of the European Public Service, and a requirement, since it must ensure legal certainty of texts that provide rights and obligations for citizens and stakeholders (European Commission 2009b). In fact, EU institutions when issuing tenders for translator contractors, weigh quality over price in a ratio of 60% over 40%.

Although quality appears to be a sine qua non in the translation produced by and for the EU institutions, a clear definition is not provided. Thankfully, the Commission’s DGT has published a Guide for External Translators which casts some light into its quality requirements. In particular, it states that:

Each assignment shall be carried out in full, in accordance with the instructions given on the Information sheet, order form and in the tender specifications (if applicable). The work delivered must have been thoroughly revised and reviewed by the contractor, so as to be usable as it stands, without further amendment by the Commission. The contractor shall pay particular attention to ensuring, inter alia, that:

  • all specific instructions from the requesting department are followed (Information sheet);
  • the delivered target text is complete (no omissions nor additions are permitted);
  • the target text is a faithful, accurate and consistent translation of the source text;
  • references to documents already published have been checked and quoted correctly;
  • the terminology and lexis used are consistent throughout the text and with any relevant reference material;
  • sufficient attention has been paid to the clarity and register of the target text;
  • the target text contains no syntactical, spelling, punctuation, typographical or other grammatical errors;
  • the formatting of the original has been maintained (Legiswrite, including codes and tags if applicable); and
  • the agreed deadline is met.

The contractor may be asked to provide a final version of an assignment incorporating corrections requested by the Commission.
We strongly advise the contractors to take note of any terminology or stylistic preferences communicated by the DGT, with a view to adjusting future translation assignments to the Commission’s requirements (European Commission 2008).

Translation quality is a very relative notion dependent on the theoretical framework adopted by the people who assess it and on the general sociocultural environment within which a translation is commissioned and produced. As House (1997: 1) aptly observes, “[e]valuating the quality of a translation presupposes a theory of translation. Thus different views of translation lead to different views of translation quality, and hence different ways of assessing it.” Therefore, it is apparent why the DGT’s guidelines are particularly enlightening; they specify what its translators need to pay attention to in order to produce a quality translation.

Let us look now more closely at these requirements. First of all, it is only natural that the client, i.e. the DGT, demands that the assignment is carried out in full, in accordance with the instructions on the Information sheet, and that it is delivered on time and ready-to-use without further amendments. It is also only natural that the TT should not contain syntactical, spelling, punctuation and typographical or other grammatical errors. Yet requirements 2, 3, 4, 5, 6 and 8 are not straight-forward and need to be discussed and understood under the light of the particularities of EU texts and the principle of linguistic equality.

2.1.1. Linguistic equality, norms and appropriateness

The principle of linguistic equality which states that all languages are equal, or ‘equally authentic’ (Wagner, Bech and Martinez, 2002: 7), and that translations are not really translations but language versions is inherently related to the policy of multilingualism and clearly affects the nature of EU texts. In simple terms, it means that EU texts are not merely translated but drafted in all languages simultaneously, and that none of the versions is derivative from any other. Although this principle of equality applies first and foremost to legal EU texts (Šarčević 2001: 319), it seems to have extended to almost all genres in the EU institutions, like press releases, public statements, regulations, etc. (Dollerup 2001: 289).

This notion of ‘equivalence’ might sound illusory. Nonetheless, it constitutes the cornerstone of translation practice in the EU. Indeed, irrespective of any qualitative characteristics, the versions are assumed to be ‘equivalent’. In other words, once the translations are completed, the ST actually ceases to exist as such, since none of the 23 ‘equivalent’ documents is supposed to carry any sign which distinguishes its status from the others.

It goes without saying that the relationships between texts produced in different languages are by no means straight-forward. Sager (1998: 325-6) identifies five types of dependence between ST and TT and points out that the relationship between ST and TT is very important mainly because it indicates how closely the TT has to be modelled on a previously existent translation and how closely it has to follow linguistic and /or pragmatic features of the ST. According to his classification, EU texts can be defined as equal documents, i.e. “documents characterised by a high degree of reciprocal dependence between the text in the source language (SL) and TL, to the extent that each document fulfils exactly the same function in its culture and, from the point of view of its function, it is no longer possible to qualify one document as a translation of the other” (Sager 1998: 326).

In the case of EU texts, however, this ‘equality’ seems to be located mainly on the notional—equal value—level discussed above and on the surface level. Surface similarity, which mainly consists of ‘sameness’ (Koskinen 2000; Zauberga 2001) between texts in the official languages, is considered to guarantee that readers of the various translations all get the same message. In particular, the translation of EU texts is strictly regulated in order to arrive at identical texts in 23 different languages. The two main requirements for that are the ‘sameness format’ and the ‘full-stop rule.’ The ‘sameness format’ means that there is literal rendering and the closest possible syntax and lexis (Dollerup 1996: 306), while the ‘full-stop rule’ prescribes an equal number of full stops in the ST and the TT (Trosborg 1997; Koskinen 2000: 55). Specifically, one sentence in the ST must correspond to one sentence in the TT; the number of paragraphs has to match, and headings and subheadings have to be located in the same place as in the original Trosborg 1997: 152). These two requirements for surface similarity apply first and foremost to legal texts and are deemed necessary because in oral and written negotiations that are based on a text, the negotiators must be able to refer to a particular article, paragraph, sentence, etc. However, as mentioned before, the constraint of surface similarity seems to have extended to almost all the official EU texts (Koskinen 2000) with obvious consequences. For example, Karker (1993: 60) mentions an example of a long and complicated sentence in Danish, where no fewer than 57 words come between the subject and the corresponding verb!

It is apparent that this principle of linguistic equality can account for requirements 2, 3 and 8 set by the Commission’s DGT, which specify that the delivered TT is complete, without any omissions or additions, that it is a faithful, accurate and consistent translation of the ST and that it unswervingly follows its formatting. These requirements clearly set particular restrictions to translators, who have to follow closely the text that they use as a ST – whether it exists in theory or not – often deviating from the Target Language (TL) and Target Culture (TC) rules, conventions or norms (Sosoni 2003: 206). In particular, they are asked to ignore what they are widely taught as students, i.e. to forget that that their ultimate aim is the production of a TT which is pragmatically accurate, “in conformity with the text-typological conventions in the target language and culture and thus fulfils the expectations of the TT addressees in a particular communicative situation” (Schäffner 1997a: 2). Instead, they are asked to conform to the conventions and norms of the genre of EU texts and fulfill the expectations not necessarily of the TT readers, i.e. the Danes in the example provided above, but of the translation commissioner, i.e. the EU Institution commissioning the job. In other words, translators need to produce translations which are ‘appropriate’ for a particular communicative situation even if that entails violating TL and TC rules, conventions and norms.

2.1.2. Clarity and register

Another quality requirement which is set by the DGT and is worthy of analysis and discussion is requirement 6, which demands that translators pay sufficient attention to the clarity and register of the TT. Clarity, vagueness and the multilingual character of the EU

In translation practice, under normal circumstances, translations should be clear, easy for readers to process and understand. It is, therefore, no surprise that DGT asks for clarity. Of course, this requirement presupposes that STs are also clear and easy to process and understand. But are they?

A common belief about language is that good usage involves clarity and precision. As a result, vagueness, imprecision and ambiguities should be avoided. Such a view, as Channell (1994: 1) observes, is too simple. In her study of vague language, she has shown that it forms a considerable part of language use, it occurs in a wide range of contexts and therefore cannot be treated as the exception but as the rule (1994: 196). In fact, her findings also corroborate the view that vagueness is inherent in communication for many reasons, but above all because, as Deese claims (1974), it is inherent in the structure of our ideas. What is more, vague language can be used deliberately, especially when information is withheld for rhetorical purposes, i.e. when a writer wants to persuade an audience, or when precision, mainly in legal writing, may end up interfering with the intelligibility of the text or limiting the scope of a particular clause or provision (Charrow, Crandall and Charrow 1982). It is thus apparent that texts are almost always vague or at least not 100% precise.

In the case of EU texts, things are far more complicated. Vague language is present not only because it is inherent in the structure of ideas and often deployed on purpose for rhetorical or legal purposes, but also because EU texts have to deal with complex and elusive concepts (c.f. 2.1.4). Moreover, they are drafted, in their majority, in English by non-native speakers (NNS) of English or, even worse, collectively in committees, working groups and teams (Koskinen 2000: 59) which are made up of different speakers of different languages. As the European Commission’s DGT points out in its English Style Guide:

Writing in clear language can be difficult at the Commission, since much of the subject matter is complex and more and more is written in English by (and for) non-native speakers, or by native speakers who are beginning to lose touch with their language after years of working in a multilingual environment ( European Commission 2010e ).

In fact, a survey carried out in the European Commission in 2009 showed that 95 % of Commission drafters wrote mainly in English, although only 13% of them had English as their mother tongue (Wagner, 2010: 4).  What is more, according to the same survey, 54 % of such drafters, that is more than half of the entire Commission population drafting documents, rarely or never have their documents checked by a native speaker. Since writers seem to construct texts according to text patterns learned when acquiring their native tongue even when they write in a foreign language (Lindholm 2008: 34), it is apparent why texts produced by non-native speakers can often be vague and hard to process. In fact, Goddard (2009: 180-181) claims that such “drafts are ‘infected’ by considerable ‘hidden translation’ of inference from the native tongue of the NNS drafter” and that the terminology used “fails to correspond to the intended content.” Unfortunately, in the rare occasion that drafters have their documents checked by a native speaker, clarity is not always guaranteed; native speakers working for the EU institutions, living and working away from their mother land, in a multicultural and multilingual environment often lose touch with their mother tongue. It is, therefore, apparent that EU texts can be ‘unnaturally’ vague and perplexing and can pose challenges for readers and translators alike.

Generally speaking, when a ST is vague, the translator needs first to determine whether the text is vague on purpose. If it is, then the translator has to retain the obscurity and vagueness, since this was the original intention of the author/s; if it is not, then the translator can try to interpret it and render it as clearly as possible in the TL. In case the translator has trouble interpreting the text and does not have the means available to clarify the contested points (e.g. the possibility to consult the author/s and/or commissioner), it is preferable to retain the vagueness rather than misinterpret the ST. According to Mossop (2010: 107-108), translators have the following options: they may report all and only what they see as the meaning of the source text, or they may report what they think the source writer should have written (i.e. they correct errors) or might have written (i.e. they add or subtract material). In the latter case, Mossop (ibid.) claims that they become the ‘motivator’ behind the ideas expressed in the translation, but they may be either ‘loyal’ motivators (adding or subtracting in the spirit of the source as they see it) or 'disloyal' motivators (engaging in their own writing project).

In the context of the EU and on the basis of the guidelines provided by the DGT, it appears that translators do not really have all the options put forward by Mossop. In particular, requirement 2 which states that no omissions or additions are permitted in the TT clashes with the option which gives the translator the freedom to add or subtract material when reporting what they think the ST author might have written, i.e. when they interpret the ST text and provide TT readers with their interpretation. In addition, translators in most cases, and especially in legal texts, cannot report what they think the ST author should have written; in other words, they cannot correct errors because they are bound by the ST as per requirement 3. Consequently, it can be assumed that in cases of vague texts, EU translators, when unable to consult the author/s or commissioner, may report all and only what they see as the meaning of the ST, retaining the vagueness if necessary. Clarity, therefore, as required by the DGT, is not always feasible. Register

The second feature mentioned under quality requirement 6, i.e. register, is equally problematic. Register, as defined by Halliday and Hasan (1985: 8), is the “configuration of meanings that are typically associated with a particular situational configuration of field, mode and tenor.” Halliday, based on the theories of Malinowski and Firth, tried to relate the text to the context of situation. He pointed out that readers/hearers reconstruct from the text, written or spoken, certain aspects of the situation, i.e. the field, the tenor and the mode. Hatim and Mason (1990: 51) argue that these three variables are in a way interdependent: a given level of formality (tenor) influences and is influenced by a particular level of technicality (field) in an appropriate channel of communication (mode).

The first problem regarding the register of EU texts is related to tenor and in particular to the fact that the texts’ readership consists of readers of various educational levels who read the texts for totally different reasons. Roughly speaking, EU texts can be categorised as follows: documents going out or for external use, documents coming in or for internal use and documents going nowhere (Harris 1999; Schäffner 1997b: 193). Documents going out are addressed, by nature, to a wide readership ranging from the expert to the general educated layman and the uninformed. Documents coming in are in theory addressed to experts, to EU officials, politicians and negotiators, but in reality they sometimes have a wider readership in that they become available for EU officials and the public alike. Therefore, both the readership and the purpose or function of many texts are variable or even mixed, a fact that adds to the difficulty of the task that writers and translators are asked to carry out. Similarly, texts going nowhere are a whole range of documents written by the EU for the EU. Examples include the minutes of meetings of EU committees, administrative boards, budget committees and draft decisions or recommendations. Such texts which are drafted or translated but not really used add to the frustration of writers and translators who have to create documents that have a mere symbolic function.

When the DGT calls translators to pay attention to the register used, it in fact calls them to identify which particular register is appropriate for the particular sub-genre they are working on. In other words, there is no uniform register that should be used across EU texts. This task is not straight-forward, first, because translators, especially freelance translators, do not always have sufficient information regarding the skopos and readership of the TT (if any) and, second, because EU texts are characterised by intertextuality (cf. 2.1.3) and interdiscursivity (Lindholm 2008) to the degree that it is not always possible to isolate and use a uniform register throughout a text. What is more, the register of EU texts, which to a large extent are legal texts, is mostly formal. As Alcaraz and Hughes (2002: 7) observe, there is a universal tendency in legal discourse “towards stiffness and formality [...] a tendency heightened by the unusual density of old-fashioned syntax and antiquated vocabulary.” Yet sometimes, when non-legal texts are involved, the register used is very different. According to the DGT’s Greek Style Guide, in cases of more ‘journalistic texts’, for instance press releases, texts produced by the DG Communication and texts intended for the Europa portal, the register used should not be formal, but more direct and colloquial (European Commission 2010d). In fact, the Greek Style Guide gives specific instructions for the translation of Internet texts, emphasising the need for attractive and transparent writing, the use of second person singular or plural when addressing readers, the avoidance of highly specialised terminology and especially acronyms and abbreviations as well as the use of examples and references to which Greek readers can relate (localisation). These instructions appear to clash with the general instructions provided by the translation services of the EU institutions and can confuse translators who are accustomed to using a formal register rather than a journalistic and informal one.

2.1.3. Intertextuality

Translators can be further confused in their effort to comply with requirement 4 set by the DGT which demands that they check and quote correctly any references to documents that have already been published, i.e. that they respond successfully to the strong intertextuality of EU texts.

Intertextuality, which has been described as an “all pervasive textual phenomenon” (Hatim 1997a: 29), is a mechanism through which “texts are recognised in terms of their dependence on other relevant texts” (Hatim and Mason 1990: 120). In other words, a text may refer backward or forward to previous or future texts. Foucault (1972: 98) claims that “there can be no statement that in one way or another does not reactualize others.” This ‘reactualisation’ can operate at any level of text organisation involving phonology, morphology, syntax or semantics (Hatim 1997b: 201).  It can manifest itself as the literal presence of one text in another text, often by a reference to the document title, or by a direct quotation from the actual text. It can also refer to the use of single words or phrases that have special cultural significance in a given linguistic community at a given time, to macro-textual conventions and constraints associated with genre, register and discourse. Intertextuality, therefore, refers to any macro- or micro-element that enables readers to identify and derive meaning from the surface features of the text in question by reference to other texts or text features they have come across.

Although intertextuality as a concept originates in literary theory (Kristeva 1980), EU texts are in fact intertextually more tightly knit than most literary texts (Gibová 2009: 148). As mentioned earlier, a strong intertextual interconnection of EU texts results from the linguistic equality principle and their co-drafting which lead to 23 simultaneously elaborated and equal translations. More importantly, intertextuality is manifested in the interconnection of the newly drafted documents with the previous ones through the use of innumerable references, direct or indirect quotations, language clichés and the consistency of terminology. An example of the intertextuality of EU texts is provided below in an excerpt from the European Commission’s 2009 Report on Competition Policy (European Commission  2009c):

1.1 State aid control
1.1.1. Shaping and applying the rules
The implementation of the State Aid Action Plan (SAAP)1 continued in 2009, with the adoption of guidance papers on training aid2 and aid to disabled and disadvantaged workers3. Guidance on the in-depth assessment of regional aid to large investment projects4 was also adopted. The principles detailed in these guidelines were applied for the first time in the Dell Poland case5, where the Commission concluded that the investment project by Dell to set-up a manufacturing plant in Łódź would significantly contribute to regional development and that these benefits outweigh any potential negative effects on competition.

1 State Aid Action Plan – Less and better targeted State aid: a roadmap for State aid reform 2005-2009 (COM(2005)107 final, 7.6.2005).
2 Communication from the Commission - Criteria for the compatibility analysis of training State aid cases subject to individual notification (OJ C 188, 11.8.2009, p. 1).
3 Communication from the Commission - Criteria for the compatibility analysis of State aid to disadvantaged and disabled workers subject to individual notification (OJ C 188, 11.8.2009, p. 6).
4 Communication from the Commission concerning the criteria for an in-depth assessment of regional aid to large investment projects (OJ C 223, 16.9.2009, p. 3).
5 C 46/2008.

As can be seen from the footnotes, there are five references to five different documents in just 205 words!  It is notable that the particular Report on Competition Policy is 13,283 words long and there are 225 footnotes referring to various official EU documents and EU case-law. Most official EU documents – especially reports and legal acts, such as regulations and opinions – follow a similar pattern. Lindholm (2008: 37), who studied the European Commission’s press releases, concluded that the particular genre is similarly characterised by a strong intertextuality. 

[…] the Commission press release is part of a communicative sequence. The Commission press release is issued in connection to the daily press conference of the Commission, the Midday briefing. In fact, the press releases are often formulated with respect to possible questions from the journalists gathered in the press room of the Commission. Also, a central feature of the Midday briefing is the Midday Express. The Midday Express is the Commission’s summary of the key announcements of the day and is made up of the introductions of the press releases issued, as well as references to speeches and other material issued. In addition, the press release is only one of several texts on the same issue. Reports, underlying legislation and additional information material to go with the press release are other texts which are part of the same textual universe. Many of them are also explicitly drawn upon in the press releases […]

It is obvious that intertextuality in EU texts appears to be the norm rather than the exception. This has tremendous repercussions for translators who need to be aware of the importance of such references in order to look up any terminology and/or phraseology or direct quotations pertaining to the texts in question. Translators need to be constantly alert, because it is only natural that, in long texts, such references, which are not always marked by specific footnotes, are numerous and can easily go unnoticed. 

2.1.4. Terminology

When translating LSP (Language for Special Purposes) texts, terminological accuracy is of utmost importance. Since EU texts are LSP texts, DGT’s quality requirement 5, which demands that the terminology and lexis used in translation are consistent throughout the text and with any relevant reference material, makes sense. Yet in practical terms there are many problems in satisfying this requirement.

First of all, it should be noted that when DGT calls for consistency with any relevant material, it in fact refers to the demands imposed by the intertextuality of EU texts discussed above. In other words, it goes without saying that when a ST (Text A) makes reference to another already translated text (Text B), the terminology to be used in the TT (Text C) shall be the one used in Text B. On the one hand, this can be demanding for translators as they need to carry out extensive research and often consult numerous reference documents, but on the other hand it facilitates their task, since they have to look for terminology in the specific reference document(s) rather than in termbanks and scores of parallel texts. Yet the terminological situation in the EU is very complex and translators are very often faced with significant problems which cannot be solved easily.

EU texts, as already pointed out, are produced in a multilingual and multicultural environment and aim at expressing new and pan-European concepts. No matter how demanding the task, all of these supranational and new concepts need to be expressed in all the official languages of the EU. This is achieved primarily with the use of Eurospeak. Trosborg refers to Eurospeak as the language used by Eurocrats, which is “often blurred, complicated and hard to understand” (1997: 152), but Wagner (1999) points out that it is in fact a useful language coined to describe EU inventions and supranational concepts which have no exact parallel at national level. Eurospeak is characterised by neologisms and borrowings. According to Crystal (1997: 132), neologisms appear in order to bear new meanings and can be defined as a) newly coined lexical units or b) existing lexical units that acquire a new sense (Newmark 1988). EFSF (European Financial Stability Facility) and ESM (European Stability Mechanism) in English are examples of the first type of neologism, while comitologie in French, and directive, cohabitation and own resources in Englishare examples of the second type of neologism in that they are restricted to the sphere of the EU but “marked by an extension of their semantic content” (Goffin 1994: 639). It is notable that neologisms can subsequently be borrowed. For instance, comitology in English and επιτροπολογία in Greek are borrowed from French, and cohabitation in French and cohabitación in Spanish are borrowed from English. As Crystal (1997) observes, borrowings involve the introduction of words from one language or dialect into another. If we want to make a distinction between different borrowings we can refer to a) loanwords with virtually no change in written form, like for instance acquis communautaire (body of EU law) in English from French, Schadenfreude (happiness at someone else’s misfortune) in English and French from German, stagiaire (trainee) in English from French and Statute (Staff Regulations) in English from French; b) loanwords which are morphologically adapted to the norms of the recipient language, like for instance νεποτισμόςin Greek from the French nepotism and c) calques or loan translations, namely loanwords in which the individual elements of a SL item are translated literally to produce a TL equivalent, like for instance third countries in English and τρίτεςχώρες in Greek from the French pays tiers (Crystal 1997:12; Shuttleworth and Cowie, 1997: 17-18; Goffin, 1994: 639-640).

Eurospeak with its neologisms and borrowings poses significant problems to translators, who first need to understand the concepts behind the terms and then attempt to transfer them into the TL. The understanding of these terms and concepts is a demanding task, requiring a profound knowledge of EU history and law. With new terms and concepts literally appearing every day, translators are constantly faced with new challenges. What is more, some of these concepts, according to Wagner, Bech and Martínez (2002: 62), are difficult to express in different languages simply because they are culture-specific, i.e. specific to certain countries for reasons of geography, climate, culture, etc. For instance, the Mediterranean countries have a great variety of terms relating to olive growing. Finland, Sweden and Denmark, on the other hand, have no climatic conditions for growing olives themselves and little tradition of trade in olives. Yet EU directives and reports on olive-growing have to be translated in Finnish, Swedish and Danish. Similar considerations apply to fishing in the Baltic, coalmining in Germany and tolerated child labour (newspaper boys) in the UK and Denmark. What is more, the EU has a common legal framework, the acquis communautaire; this in essence means that legislative instruments are produced within a single system, but they are applied in each of the 27 domestic legal systems of the member states. Consequently, in EU legal texts there seems to be an inherent tension between the common EU law on the one hand and the national legal systems in which the laws are implemented, on the other. According to Gibová (2009: 150) another factor contributing to the idiosyncrasy of EU translation is the dynamics of EU law. The EU is a legal entity in fieri and EU law is still under development. As a result, there are legal concepts lacking a deep semantic structure otherwise typical of legal semantics. Therefore, many EU terms might be fuzzy and vague.

A possible solution for this problem could be the standardisation of terms, which according to Cabré (1996: 26) is essential in LSP communication if we want to exclude the risk of naming the same concepts with variants of terms and achieve “accurate, modern and unambiguous professional communication”. However, in case of legal translation this may be very problematic because the meaning of legal terms is hardly ever fixed and is liable to being redefined by lawmakers (Gibová 2009: 150). Unfortunately, non-standardisation of terminology in the EU is very common and does not only affect legal terms but various technical, scientific, economic as well as political terms. For instance avian influenza is translated in various official EU texts in Greek as γρίπητωνπουλερικών, γρίπητωνορνίθων, γρίπητωνπτηνών; intellectual property appears in Greek EU texts both as πνευματικήιδιοκτησία and as διανοητικήιδιοκτησία, while Community industry is translated both as κοινοτικήβιομηχανία και κοινοτικόςκλάδοςπαραγωγής.  The list of such double or triple renderings is endless. If we add to the list cases of different spelling of proper names or lexical items, such as the translation of Lisbon in Greek which has four (!) different spellings, i.e. Λισσαβώνα, Λισαβώνα, Λισαβόναand Λισσαβόνα, things get even more complicated. In such cases, translators use the translation which is either used in the reference documents or appears more frequently in parallel texts found in EUR-lex, i.e. the EU’s site which provides access to EU law and related official documents.

Yet sometimes a ST makes reference in the form of direct quotations to official legal instruments which use different translations for a given term. For instance, the term flavouring substances is translated as αρτυματικέςύλες in the 1999/217/EC: Commission Decision of 23 February 1999 and as αρωματικέςύλες in Regulation (EC) No 2232/96 of the European Parliament and of the Council of 28 October 1996:

1999/217/EC: Commission Decision of 23 February 1999 adopting a register of flavouring substances used in or on foodstuffs drawn up in application of Regulation (EC) No 2232/96 of the European Parliament and of the Council of 28 October 1996.

1999/217/ΕΚ: Απόφαση της Επιτροπής, της 23ης Φεβρουαρίου 1999, για τη θέσπιση του ευρετηρίου των αρτυματικών υλών που χρησιμοποιούνται εντός ή επί των τροφίμων, το οποίο καταρτίστηκε κατ’ εφαρμογή του κανονισμού (ΕΚ) αριθ. 2232/96 του Ευρωπαϊκού Κοινοβουλίου και του Συμβουλίου.


Regulation (EC) No 2232/96 of the European Parliament and of the Council of 28 October 1996 laying down a Community procedure for flavouring substances used or intended for use in or on foodstuffs.

Κανονισμός (ΕΚ) αριθ. 2232/96 του Ευρωπαϊκού Κοινοβουλίου και του Συμβουλίου, της 28ης Οκτωβρίου 1996, σχετικά με τη θέσπιση κοινοτικής διαδικασίας για τις αρωματικές ύλες που χρησιμοποιούνται ή προορίζονται να χρησιμοποιηθούν εντός ή επί των τροφίμων.

In such cases, translators have to use the quotations unedited, without changes, and then choose one translation to use throughout the rest of the text.
It emerges from the discussion that term standardisation is a complex process and that the terminological consistency, which is sine qua non in EU texts, is often hard to achieve and requires painstaking research, implementation of the guidelines provided by the EU institutions (Interinstitutional Style Guide, official language Style Guides, etc.) and extensive practice.

2.2. Translators vs EU texts: 0-1

In light of the above, it is safe to conclude that EU texts are indeed idiosyncratic texts, very different from LSP texts as a whole. In fact, they are so different that the validity of many classic concepts ofTranslation Studies (TS), such as ST, TT, equivalence and translation quality, is questioned in terms of applicability in the EU framework. This uniqueness undoubtedly challenges translators who have to reconsider many of the traditional translation theories and steer away from widely-practiced methods and strategies. In particular, translators who wish to live up to the expectations of EU institutions and be able to produce ‘quality’ translation working either as in-house members of staff or as contractors need to be made aware of the specificities and perplexities of EU texts, as well as of the best possible ways to deal with them. This, of course, cannot be achieved ad hoc, without specialised training in an institutional setting. Universities across the EU need to offer translation students the chance to familiarise themselves with the translation and text production that take place within the EU institutions. In their effort, they cannot afford to be alone; they need the unwavering support of the EU itself.

3. Action Taken by the EU institutions and the state of play as regards translator training
3.1. Notable EU actions and initiatives

The EU has already taken significant steps to raise awareness regarding the demands of EU texts and help with the training of text drafters and translators. One of the actions worth mentioning is the set up in 1995 of the Interinstitutional Committee for Translation and Interpreting (ICTI) which followed an initiative of the heads of the various bodies to bring together representatives of all the translation and interpretation services of EU institutions and other bodies. Since its inception, the committee has been coordinating and implementing interinstitutional projects and managing the Inter-Active Terminology for Europe (IATE), i.e. the interinstitutional terminologydatabase of the European Union which brings together all the EU terminology resources thus making information more easily available and ensuring its standardisation. The ICTI submits an annual report to the EU institutions, where it reviews the progress in interinstitutional cooperation in the field of translation and interpretation and suggests steps for further development (Pariente 2010: 49).

Moreover, the European Commission’s DGT, in its effort to enhance translation quality and promote the translation profession, has taken various actions towards this direction especially during the past decade. First and foremost, it launched in 1998 the Fight the Fog campaign, an attempt to train text drafters and translators in the principles of clear writing (Wagner 2005: 224). The campaign was masterminded by English translators and involved just the English language, i.e. the EU’s main drafting language.  Yet in 2010 the European Commission set in motion a new Clear Writing campaign, which, as Wagner (2010: 4) observes, is wider-based, with a steering group drawn from several Commission departments and offers information about clear writing in all languages. The new campaign was born because it had become evident, especially after the 2004 enlargement, that it was necessary to raise awareness and provide practical aids to clear writing, offering more access to more kinds of training, and helping promote the idea that important texts should be revised by a native speaker before they are finalised and published (Wagner 2010: 4). Most of the practical advice appears in the booklet How to Write Clearly (European Commission no date), which draws on guidance from several different European countries and is available in all 23 official languages.

Alongside the booklet, which is intended for EU officials and the wider public alike, the campaign includes many activities directed at in-house members of staff: an intranet website, a special online tutorial, a ‘Tip of the Week’, lunchtime lectures and taster courses. What is more, there is a helpline providing instant advice for drafters in distress! At the end of 2010, the campaign hosted national clear writing specialists at a major conference entitled Clear Writing throughout Europe. More importantly, the campaign is hoped to culminate in a report with recommendations on sustainable action.

Another recent project undertaken by the Commission’s DGT is the development of a Study Strategy, which aims at gathering knowledge and information about translation using a variety of methods such as internal studies, workshops and outsourced study projects (Kauko and Häggman 2011: 4). The DGT then disseminates this information by making it available on the Internet and by organising a major annual event, i.e. Translation Studies Days, where the findings of these studies are presented and discussed by linguists, translators and other stakeholders in European affairs (Häggman 2011: 6).  Indicatively, some of the studies commissioned by the DGT involve web translation as a genre, lawmaking in the EU multilingual environment and the contribution of translation to the multilingual society in Europe. This strategy is particularly important because it raises awareness about translation as an activity and translators as key players in the dissemination of information and implementation of the policy of multilingualism; it educates EU officials and the public about the translation profession; finally, it brings to light mishaps and provides solutions which can facilitate the job of the translator and optimise the translation process.  

In the same spirit, the Commission’s DGT launched another initiative in 2006, labelled Quality Management in Translation, which basically revisited the DGT’s quality strategy and aimed at creating a strong dynamic towards sustained quality management in translation. The basic line of the exercise was the idea that quality must be ensured before, during and after translation, including editing, advisory or localisation processes. This led to the publication in 2009 of theProgramme for Quality Management in Translation – 22 Actions (European Commission 2009b) which described actions and expected outcomes for improving translation quality and efficiency, such as providing training and coaching for assessors, strengthening the relationship with contractors and raising awareness of translators about the nature and purpose of texts sent for translation.

Having realised that it is in its best interest to be actively involved in the training of translators, the European Commission’s DGT recently set two schemes in motion: the Visiting Translator Scheme (VTS) and the European Master’s in Translation (EMT) network of Universities.

The VTS allows Commission staff translators to spend a few weeks at universities around the EU giving classes on translation, the work of DGT and languages in the EU institutions. In that way, they promote the EU and its policy of multilingualism, they offer ‘inside’ information about the job that translators are asked to carry out and they provide details on the specificities of EU language work. Moreover, they encourage students to consider applying for a translation-related post with the EU.

EMT is another project involving the Commission’s DGT and academia; it aims to help raise the standard of translator training in the EU and foster cooperation and exchanges between academic institutions offering translation courses. More specifically, it seeks to produce highly qualified translators and to equip students with the necessary skills in order to compete successfully for jobs on the translation market, including the EU institutions (European Commission 2010b).

With that goal in mind, the EMT expert group established a reference framework for the competences applied to language professions or to translation over a wide semantic or professional range, including various modes of interpreting (see EMT). This framework sets out what is to be achieved, acquired and mastered at the end of training regardless where, when and how. Six interdependent competences are proposed: language competence, intercultural competence, information mining competence, technological competence, thematic competence and translation service provision competence in each of the six areas are interdependent. They all lead to the qualification of experts in multilingual and multimedia communication and together, they comprise the minimum requirement to which other specific competences may be added (for instance in the case of localisation, audiovisual translation or research) (Gambier 2009).

Admittedly, EMT is of significant value primarily because it brings together academia and the largest translation service in the world in a common effort “to enhance the status of the translation profession in the European Union” (European Commission 2010b). Yet this initiative is not genre-oriented, i.e. it does not focus on the translation of EU texts and as a result it does not provide particular guidelines for the design of a postgraduate programme able to meet the specific demands of EU texts translation.

3.2. Academic institutions and training
We cannot expect to have one dominant and comprehensive theory of language and translation. There are too many different kinds of languages, too many different types of texts, and too many different audiences with diverse needs. (Nida, 2003: 142)

It is evident that the EU and especially the European Commission’s DGT have been active in a multitude of ways. Yet their initiatives and actions are not enough to solve the problem. As pointed out earlier, the poor performance of translation candidates can be explained by the fact that they are not competent enough to respond to the demands posed by the EU institutions. As “translation competence is most effectively developed at an academic institution” (Schäffner and Adab 2000: x), we can assume that if we wish to fight the problem at its core, we need to study, evaluate and review the way translator training is offered today across Europe. 

The current picture as regards translator training is one of heterogeneity and lack of consensus both form- and content-wise. An extensive research project regarding training in translation and interpreting in Europe which was completed in 2000 under the Thematic Network Project in the Area of Languages (TNP-D) highlighted the lack of homogeneity of translator and interpreter training in Europe in terms of organisation, responsibility, duration of courses, qualifications and contents. In 2005, Ulrych published the results of a survey she carried out regarding the translator training practices at tertiary level at universities and interpreter institutions. Her findings also underline the heterogeneity of translation courses as regards both structure and content (Ulrych 2005: 8).

Generally speaking, translation is either taught in the framework of a general language programme (Modern Languages and Translation, English Studies and Translation, TESOL and Translation, etc.) or as a stand-alone course, which can be undergraduate or postgraduate. The structure of the programme can also vary depending on the target-end users, i.e. the students of the programme who can be newcomers to translation or practicing translators hoping for particular specialisation in a given field (Gile, 1995: 6-7). Furthermore, teaching can be process-oriented or product-oriented, it can involve theoretical components as well as practical components, practical components with no or limited theory, theory with no or limited practice, familiarisation with technology and CAT tools, training in project-management, etc. It is evident that the landscape in translator training is vague and translation competence hard to delineate, hence the “lack of consensus among the parties involved as to what should be included in the training and what the actual practice of translating is about” (Thelen, 2001: 16).

4. Addressing the problem: tailoring training to practice

Independently of the differences that academic programmes in translation have, the discussion about the poor performance of translation graduates as regards EU texts brings to light the need for specialised training at an academic environment. Universities across Europe have to step up and take action in order to offer students vigorous training in the translation of EU texts. This training is by no means a luxury; it’s an unquestionable necessity, and it constitutes the only way that will help universities respond to the current deficit and live up to the expectations of students and the EU institutions as major stakeholders and employers of translators. Admittedly, some do offer specialist modules in the translation of EU texts and have been doing so for a number of years, but the majority of university programmes at both undergraduate and postgraduate levels offer practical translation courses in related areas, such as Economics and the Law, without focusing specifically on EU texts and their intricacies.

Although it is beyond the scope of this article to describe the ways in which universities can offer their students the best possible training in the translation of EU texts, an attempt is made to provide an overview of the main axes of such training.

As it was shown, translation quality is a very relative notion which depends on the theoretical framework adopted by the people who assess it and on the general sociocultural environment within which a translation is commissioned and produced. Given that translation quality in the framework of the EU is measured in relation to the elusive concept of linguistic equality and on the basis of conformity to the very specific guidelines presented in 2.1., it goes without saying that training should focus on the familiarisation of students with those particular aspects.

As pointed out earlier, the problem imposed by the implementation of the principle of linguistic equality is that it goes against what translators are taught at universities and against the large volume of translation theory that advocates a TT which respects the ST, but at the same time respects the TL and TC conventions and norms. In the case of EU texts, TL and TC conventions and norms are sacrificed in favour of surface similarity of texts in the official EU languages and, by extension, in favour of conventions and norms of the genre of EU texts, which are neither SC- and SL-oriented nor TC- and TL-oriented, but occupy the ground in-between, a sort of no-man’s land. Consequently, universities need to draw their students’ attention to this paradox by offering them specialist translation theory classes which place particular emphasis on functionalist approaches to translation and on the notion of the TT purpose as the decisive criterion in translation. Modules on the history of the EU, its policies, body of law and Institutions should also be offered as they can help students grasp the idea of European unification, the notions of ‘integration’ and ‘unity in diversity’, and consequently the EU’s complex policy of multilingualism.  

Moreover, since translation quality is inextricably related to the reliable implementation of the guidelines set out by the EU institutions, students need to be vigorously trained in the strict, unswerving use of the Interinstitutional Style Guide, the Guide for External Translators and the Style Guide in their respective TL. Particular emphasis must be placed on the importance of reference material as a result of the intertextuality and interdiscursivity of EU texts and the need for terminological and register consistency. In addition, students need to learn how to reconcile the vague language often present in EU texts with the need for clarity as a translation requirement.

To achieve their goal, universities have to include practical translation modules in their curricula, where students are given the chance to follow the aforementioned guidelines while translating different EU genres—ranging from European Commission regulations to European Parliament verbatim reports of plenary proceedings. Yet the best way for students to familiarise themselves with the nuts and bolts of translating for the EU is to translate under real-life conditions. It is, thus, suggested that universities work closely with the EU in order to give students and/or graduates the chance to do a traineeship in a formal EU setting so that they can gain professional translation experience. Although at present a graduate traineeship is in place in the Commission, the Parliament and the Translation Centre for the Bodies of the EU (CdT), it is suggested here that this is expanded and redesigned in collaboration with academic institutions with a view to offering trainees an intensive and constructive translation experience tailored to their needs. Traineeships could also be offered by EU contractors in the individual member states in collaboration with academic institutions.

The European Commission’s DGT should also work closely with academic institutions to expand the EMT network and give it a more clearly-defined focus, with EU texts at its core. It should also expand its Visiting Translator Scheme so that more translators can spend time at academic institutions, offering specialised training and inside information to Translation students and working with academics to help enhance the status of the translation profession in the EU. It is suggested that more EU institutions participate in the VTS and that they periodically organise translation seminars with the participation of academics, EU translators and officials. These seminars can work as a forum for the exchange of best practices in translation training and problem-solving. It goes without saying that all the current initiatives of the EU institutions, such as the Study Strategy and the Clear Writing campaign, need to be sustained and enhanced.

6. Concluding remarks

It is commonly accepted that a translator is a very versatile character, who is expected to adapt to various situations and requirements. For that reason, most trainers maintain the view that translation programmes should be general enough to enable their graduates to embark on a wide range of activities and at the same time specific enough to lay the foundations for a fast and effective acquisition of any special skills required by a particular job. According to Cronin (2005: 211), the possession of a particular expertise confers legitimacy on the work of translators and justifies the existence of a specific training. Although it is not argued here that we as trainers should interpret “professional expertise” in a narrow sense and “limit the cultural impact of translators” (Cronin 2005: 211), it is claimed that academic institutions offering translation programmes at undergraduate and postgraduate level should include in their curricula specialised courses in the translation of EU texts.

The benefits are considered to be manifold. Training institutions will establish their reputation in the education market by producing self-confident, more qualified translators ready to respond to the ever-increasing and ever-demanding job of the professional translator. In turn, they will attract better-equipped and more determined students. Once these graduate, they will be sought-after in the translation market, able to negotiate better terms of employment. Translation companies will operate more effectively, optimising their services and thus maximising their profit. What is more, the translation services of the EU will operate more smoothly. Translations will be carried out more efficiently and this will unquestionably help the smoother and more effective functioning of the EU institutions, since translation is an activity which can be found at their core. Most importantly, it is envisaged that these courses will contribute to the transformation of attitudes towards translation and translators both in the professional world and in academia. Ulrych (2005: 23) acknowledges that there is a long way to go before translators receive full recognition for their work. She adds that “much depends on how translator training is able to cope with the changes that the internationalisation of economies and cultures will increasingly bring about and to meet the challenges that the profession faces today.” It is hoped that specialised training in EU texts translation will help achieve just that.

  • Alcaraz, Enrique and Hughes, Brian (2002). Legal Translation Explained. Manchester: St Jerome.
  • Anderman, Gunilla and Rogers, Margaret (eds) (2005). In and Out of English: For Better, For Worse? Clevedon: Multilingual Matters.
  • Babiniotis, Georgios D. (1996). “Plurilinguisme, Oligolinguisme et Hégémonie Linguistique”. Eleni Karantzola (ed.) (1996). La Langue Grecque dans l’Union Européenne Élargie.Athènes: Centre de la Traduction Littéraire.
  • Baker, Mona, Olohan, Maeve and Calzada Pérez María (Eds) (2010). Text and Context. Essays on Translation and Interpreting in Honour of Ian Mason. Manchester: St Jerome
  • Brackeniers, Eduard (1995). “Europe Without Frontiers and the Language Challenge.” Terminologie et Traduction, 2, 13-17.
  • Cabré, Teresa (1996). “Quality Assurance of Terminology Available on the International Computer Networks.” Somers, Harold (ed.) (1996). Terminology, LSP and Translation. Studies in Language Engineering in Honour of Juan C. Sager.Amsterdam/Philadelphia: John Benjamins.
  • Channell, Joanna (1994). Vague Language. Oxford: Oxford University Press.
  • Charrow, Veda R., Crandall, Jo Ann and Charrow, Robert P. (1982). “Characteristics and functions in legal language.” Richard Kittredge and John Lehrberg (Eds) (1982). Sublanguage. Berlin: Walter de Gruyter.
  • Cronin, Michael (2005). “Deschooling Transaltion.” Martha Tennent (Ed.) (2005) Training for the New Millennium. Amsterdam/Philadelphia: John Benjamins.
  • Crystal, David (1997). The Cambridge Encyclopedia of the English Language. Cambridge: Cambridge University Press.
  • Deese, James (1974). “Towards a Psychological Theory of the Meaning of Sentences.” Albert Silverstein (Ed.) (1974) Human Communication: Theoretical Explorations. Hillsdale NJ: Lawrence Erlbaum Associates.
  • Desblache, Lucile (Ed.) (2001). Aspects of Specialised Translation. Paris: La Maison du Dictionnaire.
  • Dollerup, Cay (1996). “Language Work at the European Union.” Marilyn Gaddis Rose (ed.) Translation Perspectives IX. New York: State University of New York at Binghamton.
  • (2001) “Complexities of EU Language Work”. Perspectives, 9(4), 271-292.
  • Ferenčík, Milan and Horváth, Juraj (Eds) (2009)Language, Literature and Culture in a Changing Transatlantic World: International Conference Proceedings. Prešov: Prešovská Univerzita v Prešove.
  • Foucault, Michel (1972). The Archaeology of Knowledge (tr. A.M. Sheridan Smith). London: Tavistock.
  • Gaddis-Rose, Marilyn (Ed.) (1996). Translation Perspectives IX. New York: State University of New York at Binghamton.
  • Gambier, Yves (2009). “Competences for Professional Translators, Experts in Multilingual and Multimedia Communication.” On-line at: (consulted 20.12. 2010)
  • Gibová, Klaudia (2009). “EU Translation as the Language of a Reunited Europe Reconsidered.Milan Ferenčík and Juraj Horváth (eds)Language, Literature and Culture in a Changing Transatlantic World: International Conference Proceedings. Prešov: Prešovská Univerzita v Prešove.
  • Gile, Daniel (1995). Basic Concepts and Models for Interpreter and Translator Training. Amsterdam/Philadelphia: John Benjamins.
  • Goddard, Christopher (2009). “Where Legal Cultures Meet: Translating Confrontation into Existence.” Investigationes Linguisticae, XVII, 168-205.
  • Goffin, Roger (1994). “L’ Eurolecte: Oui, Jargon Communautaire: Non. Meta, XXXIX(4), 636-642.
  • Häggman, Johan (2011). “Translation Studies Days.” Languages and Translation, 2, 6.
  • Halliday, Michael A. K and Hasan, Rugaiva (1985). Language, Context and Text: Aspects of Language in a Social-Semiotic Perspective. Oxford: Oxford University Press.
  • Harris, David. (1999). “Experiences of Translation for the EU.” Paper presented at the Seminar Translating for the European Institutions (ITI, 6 May 1999).
  • Hatim, Basil (1997a). “Intertextual intrusions: towards a framework for harnessing the power of the absent text in translation.” Karl Simms (Ed.) Translating Sensitive texts: Linguistic Aspects. Amsterdam: Rodopi.
  • (1997b). Communication Across Cultures: Translation Theory and Contrastive Text Linguistics. Exeter: University of Exeter Press.
  • Hatim, Basil and Mason, Ian (1990). Discourse and the Translator. London: Longman.
  • House, Juliane (1997). Translation Quality Assessment. A Model Revisited. Tübingen: Narr.
  • Karantzola, Eleni (Ed.) (1996). La Langue Grecque dans l’Union Européenne Élargie.Athènes: Centre de la Traduction Littéraire.
  • Karker, Allan (1993). Dansk i EF - en Situationsrapport om sproget. København: Gads Forlag.
  • Kauko, Silvia and Häggman, Johan (2011). “DGT Study Strategy: Through Translation to Multilingualism.” Languages and Translation, 2, 4-5.
  • Kennedy, Tom (1998). Learning European Law. London: Sweet and Maxwell.
  • Kittredge, Richard and Lehrberg, John (Eds) (1982) Sublanguage. Berlin: Walter de Gruyter.
  • Koskinen, Kaisa (2000). “Institutional Illusions: Translating in the EU Commission.” The Translator, 6(1), 49-65.
  • Kristeva, Julia (1980). Desire in Language: A Semiotic Approach to Literature and Art. New York: Columbia University Press.
  • Lindholm, Maria (2008). “A Community Text Pattern in the European Commission Press Release? A Generic and Genetic View.” Pragmatics, 18(1), 33-58.
  • Mossop, Brian (2010). “Translating What Might Have Been Written.” Mona Baker, Maeve Olohan and María Calzada Pérez (Eds) (2010) Text and Context. Essays on Translation and Interpreting in Honour of Ian Mason. Manchester: St Jerome, 95-113.
  • Newmark, Peter (1988). A Textbook of Translation.London: Prentice Hall.
  • Nida, Eugene (2003). Fascinated by Languages. Amsterdam/Philadelphia: John Benjamins.
  • Pariente, Audrey (Ed.) (2010). Translation at the European Commission – a History. Luxembourg: Office for Official Publications of the European Communities.
  • Pym, Anthony (2004). “On the Pragmatics of Translating Multilingual Texts. JoSTrans, The Journal of Specialised Translation, 01, 14-28. (consulted 13.12.2010).
  • Sager, Juan C. (1998). “What Distinguishes Major Types of Translation.” The Translator, 4(1), 69-89.
  • Šarčević, Susan (2001). “Multilingualism - The Challenge of Enlargement.” Perspectives, 9(4), 313-324.
  • Schäffner, Christina (1997a). “From Good to Functionally-Appropriate: Assessing Translation Quality.” Current Issues in Language and Society, 4(1), 1-5.
  • — (1997b). “Strategies of Translating Political Texts.” In Anna Trosborg (Ed.) (1997) Text Typology and Translation. Amsterdam: John Benjamins.
  • — (2000). “Discourse Analysis for Translation and Translator Training: Status, Needs, Methods.” Current Issues in Language and Society, 7(3), 177-184.
  • Schäffner, Christina and Adab, Beverley (Eds) (2000). Developing Translation Competence. Amsterdam/Philadelphia: John Benjamins.
  • Shuttleworth, Mark and Cowie, Moira (Eds) (1997). Dictionary of Translation Studies. Manchester: St Jerome.
  • Silverstein, Albert (Ed.) (1974). Human Communication: Theoretical Explorations. Hillsdale NJ: Lawrence Erlbaum Associates.
  • Simms, Karl (Ed.) (1997). Translating Sensitive texts: Linguistic Aspects. Amsterdam: Rodopi.
  • Sosoni, Vilelmini (2003). Aspects of Lexical Cohesion in EU Texts: A Critical Study of Greek Translations and English Hybrid Texts. PhD Thesis. University of Surrey.
  • Stoddard, Sally (1991). Text and Texture: Patterns of Cohesion. Norwood: Ablex.
  • Tennent, Martha (Ed.) (2005). Training for the New Millennium. Amsterdam/Philadelphia: John Benjamins.
  • Thelen, Marcel (2001). “Present Issues in Teaching Specialised Translation: a Practice-oriented MA Programme.” Lucile Desblache (Ed.) (2001) Aspects of Specialised Translation. Paris: La Maison du Dictionnaire.
  • Trosborg, Anna (Ed.) (1997). Text Typology and Translation. Amsterdam/Philadelphia: John Benjamins.
  • — (1997) “Translating Hybrid Political Texts.” Anna Trosborg (Ed.) Text Typology and Translation. Amsterdam/Philadelphia: John Benjamins.
  • Ulrych, Margherita (2005). “Training Translators: Programmes, Curricula, Practices.” Martha Tennent (ed.) (2005). Training for the New Millennium. Amsterdam/Philadelphia: John Benjamins.
  • Wagner, Emma, Bech, Svend and Martínez, Jesus M. (2002). Translating for the European Institutions. Manchester: St Jerome. 
  • Wagner, Emma (1999) “Does Eurojargon Exist and What is the Fight the Fog Campaign?” Paper presented at the Seminar Translating for the European Institutions (ITI, 6 May 1999).
  • — (2005). “Translation and/or Editing: The Way Forward?” G. Anderman and M. Rogers (eds.) (2005). In and Out of English: For Better, For Worse? Clevedon: Multilingual Matters.
  • — (2010). “Why Does the Commission Need a Clear Writing Campaign?” Languages and Translation, 1, 4-5.
  • Zauberga, Ieva (2001). “Discourse Interference in Translation.” Across Languages and Cultures, 2(2), 265-276.

Sosoni imageVilelmini Sosoni, PhD, is Lecturer in Translation at the Ionian University in Greece. She also works as a freelance translator and subtitler. Her research interests lie in the areas of the translation of EU texts, text hybridity, text linguistics and translation, norms in translation, translation quality, language diversity and multilingualism, language ideologies, translation and political discourse. She has published papers in international journals and in edited volumes.


Note 1:
As from 01.07.2009. Subject to salary adjustments according to the Staff Regulation..
Return to this point in the text

Note 2:
As from 16.06.2008, a new quality mark (Acceptable) has been added in the DG Translation system. The quality values are now as follows: Very Good, Good, Acceptable (new), Below Standard, Unacceptable (European Commission 2008). It is notable, though, that only Very Good and Good are considered satisfactory in the terms of EU assessment. The ‘acceptable’ mark should be understood as "just sufficient," as it refers “to translations that are not ready to be sent as such to the requesting service and will possibly require significant revision or upgrading by DGT services or the contractor. It is worth pointing out that this new quality mark will undoubtedly have an impact on the position of the contractor in the ranking system and may lead in the long term to a re-assessment of his / her quality provided in the lot in question” (Ibid.)..
Return to this point in the text

Note 3:
Private communication with the Managing Director of the company, 18 November 2010.<.
Return to this point in the text

Note 4:
Private communication with the CEO of the company, 30 November 2010.
Return to this point in the text