Interpreting and multilingualism in the EU: Leave or Remain? = Interpretación y multilingüismo en la UE: ¿irse o quedarse?

This paper seeks to give an overview of the current status of multilingualism and interpreting in the EU institutions at a time when English is increasingly used as a lingua franca. It examines the use of English in the different EU bodies and the extent to which ELF (English as a lingua franca) has an impact on the use of other languages. After evaluating the advantages and disadvantages of an English-only EU, the reactions of interpreters and interpreting services to the rise of ELF are analysed. In closing, the article reviews possible strategic actions the interpreting services could endorse in order to facilitate a more realistic and less resentful response to an inevitable phenomenon.


INTRODUCTION
The last years were undoubtedly some of the most demanding in the history of the European Union. Once established as a coalition for enduring peace on our continent, the EU is currently facing enormous challenges such as the largest migratory movement in recent European history, rising Euroscepticism, and the breaking away of one of its biggest member states, the UK. At the same time, the EU's language and thus interpreting policy is confronted with numerous demanding developments. This article sets out to analyse the most prominent one -the clash between multilingualism and English as a lingua franca.

Karin Reithofer
Interpreting and multilingualism in the EU: Leave or Remain? CLINA vol. 4-1, June 2018, 115-136 eISSN: 2444-1961 Ediciones Universidad de Salamancacc by-nc-nd (Europäische Kommission 2012). It is to be expected that the prominent role of English will continue to rise, considering that the number of European secondary school pupils who are currently learning English is as high as 90% (EACEA 2012).

Working languages in the EU institutions
As mentioned above, the EU does not guarantee perfect multilingualism in all institutional settings. A full language regime with interpretation into and out of all official languages is usually limited to events with a high symbolic value such as the formal sittings of the Council or the plenary sitting of the European Parliament (He 2006, 27).
While 24 languages are recognised as official EU languages, the different EU bodies can choose their own working languages. What at first seems to be in contradiction with the principle of multilingualism makes perfect sense in the Union's day-to-day business, since not all institutions and bodies are in direct contact with citizens. As there is no legal definition of which working languages have to be used in which institution, the working languages are not officially specified as such, but are a matter of practice (Gazzola 2006, 397) and vary depending on the institutions' different tasks.

European Commission
The Commission staff is exclusively made up of civil servants who have to speak at least English and French. Internal communication primarily takes place in English, French and German.

European Parliament
The language policy of the European Parliament (EP) differs greatly from the one employed by the Commission. It is a multilingual body that aims to guarantee true democratic representativeness, in that persons from all social backgrounds should be able to stand for election. They should not be elected because of their language skills, but because of their political programme (Gazzola 2006, 404, He 2006. Thus, for all plenary sittings, interpretation is provided into and out of all 24 official languages. The preparatory work for the plenary session in the parliamentary committees and political groups usually takes place with a reduced language regime (He 2006, 28) that depends on their composition. Languages like Gaelic and Maltese are hardly ever used outside the plenary sitting, one reason being the lack of qualified interpreters for these languages.
Unofficial communication, so-called corridor talk, between MEPs of different countries mainly takes place in English (Wright 2007

. Council of the European Union
The situation in the Council is rather similar to the one in Parliament. For official meetings of the heads of state or government at the European Council summits or the specialised ministers' meetings in the Council of the EU, a full interpretation regime is usually provided.
The thematic working groups and the member states' permanent representatives generally meet with a restricted language regime. Similarly to the EP, delegates usually resort to English for unofficial chats or bilateral exchanges during the breaks.
Even at the highest level, however, English is often spoken even in meetings with interpretation. There seem to be two main schools of thought in this regard: on the one hand, there are politicians like the former German vice-chancellor Hans-Dietrich Genscher who famously said: «In a foreign language I say what I can. In my own language I say what I want.» They are of the opinion that using a foreign language might mean that they are in a disadvantaged position in respect to native speakers.
On the other hand, some politicians seem to believe that direct communication -mostly in English-is more efficient or practical and hence accept the risk of exposing themselves to ridicule such as the German EU commissioner Öttinger who was mocked on the Internet for speaking heavily accented English. In this context, the former German finance minister Wolfgang Schäuble was criticised for insisting on speaking English even when interpretation was provided and thus compromising all attempts to credibly establish German as a strong working language in the EU (Littger 2013). It appears that the official policy to use German whenever possible (Auswärtiges Amt 2015) is ignored by several German politicians. Some are quite aware of the deficiencies of their language skills: Schäuble once apologised for his English saying that he pitied everyone who had to endure it, but that after all, badly spoken English was one of the most widely spoken languages in the world (Littger 2013).
Hence, the status attached to language and language proficiency seems to be judged differently by these two groups. The quote also goes to show that the acceptance of non-standard English appears to be growing.

Other institutions and bodies
The other bodies' language policies reflect their functioning and composition in a similar way to the one described for the three most important institutions.
The European Economic and Social Committee (EESC) is composed of representatives of the member states' social partners, and the Committee of the Regions (CoR) is a forum for the different local and regional authorities. Hence, multilingualism is held in high regard for the same reasons as in the EP (He 2006, 25).
Other institutions, however, such as the European Central Bank, the Court of Auditors and the many technical agencies work with supranational civil servants and not Karin Reithofer Interpreting  nationally elected representatives and thus use only between one and five languages in their everyday work (Ammon 2006, 269, He 2006.

English vs. French
English is a working language in 96% of EU institutions and the only working language in eight of them (He 2006, 26). This highlights the prominent role of English once again. Quite clearly, it has taken the place originally occupied by French, which continues to loose influence and prevalence, a trend that has increased since the big rounds of enlargement in 2004 and 2007. In most of the mainly Central and Eastern European countries that joined then, French does not enjoy a privileged status, while English does -just like in the fields of business and science worldwide (Reithofer 2010, 145f.). Dollerup (2002) sees a further possible explanation for English winning over French in the English native speakers' greater tolerance towards non-native speakers: The French are much more squeamish about 'purism' than the English who have long had to accept that most foreign speakers of English use the lingo to speak to other foreigners rather than to Englishmen born and bred. But the English tolerance is definitely a factor which pushes more of the ground work into English. (2002,197) While the reasons behind the rise of English in the EU institutions are manifold, numbers serve to substantiate the extent to which it is happening.

English in written texts
Although this article focuses on interpreting, written communication in the EU institutions shall be examined briefly, since data from this field confirms the growing significance attached to the English language.
In the area of written communication, the principle of multilingualism is not fully applied either. Only documents of high importance, such as legislative texts, parliamentary reports, the official journal or white books are available in all 24 official languages. Other documents are translated only into the languages needed in each case.
Examining the languages of source texts of translations yields a striking result: 81% of original texts are drafted in English, with French lagging far behind with only 4.5%. In 1997, these two languages still accounted for roughly the same amount of source texts. Since then, the influence of English has constantly been on the rise as shown by the percentage of source texts in English (European Commission/DG Translation 2014, 7; cf. This reflects inter alia the drafting procedure in Council working groups where the national delegates have the possibility to speak their mother tongue, but usually work on the basis of a text in English. English is thus the common procedural language, the «de facto drafting language» (Phillipson 2003, 120).
In the Commission, 95% of drafters write mainly in English, with only 13% of them being native speakers (Wagner 2010). Hence, most texts are written by non-native speakers.
Furthermore, when considering the target languages into which texts are translated, English is by far the most common one with over 250,000 pages (followed by French with over 100,000 less pages), confirming once more the language's privileged status (European Commission/DG Translation 2014, 8).

English in oral communication and interpreting
As mentioned above, a full interpreting regime is not provided in all official EU meetings. This is partly because it would not be possible to find enough interpreters to cover the enormous number of language combinations resulting from 24 official languages in as many as 60-80 meetings per day in the Council alone. In some meetings, a scheme called interpreting upon request is offered, meaning that interpreting is guaranteed exclusively into and from the languages of member states who request it (Gazzola 2006, 394

Listening to English in meetings
Interpretation into English is provided in almost all meetings (98%), with 11,158 interpreting days per year, followed by French and German, which are provided in 76% and 60% of meetings respectively. Thus, the English booth has become virtually indispensable. According to the 2015 customer satisfaction survey 1 by the European Commission's Directorate General Interpretation (DG SCIC 2015a), the vast majority of those for whom interpretation into their first language (L1) was not available listened to (the interpretation into) English, followed by German at 5% and French at 3%. In this respect, there also seems to be a trend towards more English as can be seen in Furthermore, the survey showed that delegates with interpretation into their L1 nevertheless listened to another language sometimes (31%) or even exclusively (4%). One can assume that this other language was often English. The main explanations for this given by the respondents were that they were more familiar with the subject in another language (38%) or that they needed to report in another language (10%) (DG SCIC 2015a).
One could assume that those who had to listen to an interpretation into English instead of into their L1 were at ease with the situation, since their authorities had not requested interpretation for the meeting in question. However, the satisfaction rates do not corroborate this hypothesis. The surveys clearly showed that the satisfaction rate was higher when delegates listened to interpretation into their L1 than when interpretation into their L1 was not provided (cf.  55% of delegates stated that it would have been easier to participate in the meeting if they had been able to listen to interpretation into their L1. Nevertheless, this figure has decreased over the years (cf . Table 4). Thus, the number of those who are comfortable with no interpretation into their L1 seemed to be growing. In the 2017 customer satisfaction survey however, this trend seems to have reversed: Majcen (2018) reports that 75% of the participants would have found it easier to participate in the meeting if it had been possible to listen to their L1. Since not all data have been published yet, the result cannot be fully interpreted.

Karin Reithofer
One can only speculate as to why certain delegations did not ask for interpretation into their L1. A few cases might also be due to a lack of interpreters, since DG SCIC cannot satisfy 100% of the demand. Others might be similar to the reasons described in section 2.1.3.
Another striking outcome of the 2015 SCIC survey is that non-native speakers of English were markedly less satisfied with English booth interpreters than native listeners (cf. The question arises as to why this is the case. One reason might be that the nonnative listeners' command of English is not as good as they or their superiors -who do not request interpreting for them -might think. Following a meeting in a foreign language might be more tiring and thus lead to lower satisfaction with the interpreters. Possible reactions to this are discussed in section 5.

Speaking English in meetings
There is no empirical data on which language delegates speak when they have not requested interpretation from their mother tongue and thus cannot use their L1 in a meeting. In my personal experience as an EU freelance interpreter, the overwhelming majority of those who do not speak their mother tongue speak English. Non-native speakers of French or German do sometimes take the floor, but it is extremely rare. Data from other conference interpreting settings show that non-native English is indeed omnipresent. Some examples come from as early as 1994, when a study on a conference showed that 67 of 96 English presentations were given by non-native speakers (Pöchhacker 1994, 154). Another conference study in 2002 found that 19 out of 28 English speakers were non-natives (Basel 2002). In a survey among 58 AIIC interpreters (Katikos 2015, 100), the respondents estimated that around 80% of English speakers in conference settings are non-native speakers.

Karin Reithofer
In the EU context, some delegates deliberately choose not to speak their mother tongue in meetings where interpretation from their L1 is provided. In the 2015 SCIC customer satisfaction survey, 7% stated they always spoke another language than their L1 and 11% said they sometimes used another language. The survey did not ask what the other language(s) were. Considering the above data, English is again a very likely option. 41% of those respondents chose to do so because they were more familiar with the subject of the meeting in another language. 10% did not know in advance whether they would be able to speak their mother tongue and presumably had already prepared their statements in English. 34% thought that the message was better conveyed in a more widely spoken language and 14% were worried that interpretation was not accurate.
In my experience, many delegates speak English because of an asymmetric interpreting regime, i.e. they can speak their L1, e.g. Danish, since interpreters in other booths may have Danish as a C language, but cannot listen to it, since no Danish booth is provided or was requested. Thus, they often listen to the English booth and then proceed to speak English when they take the floor, since they have been listening to English all day long.
Even though many delegates use English as a foreign language in meetings, not all respondents were happy with a situation in which they were not able to speak their L1. In the 2015 survey, 52% said it would have been easier to participate in the meeting if they had been able to speak their L1. Yet, similarly to the above figures this number shrunk over the years: as can be seen in Table 6, the percentage was still significantly higher in 2013 and 2010. Again, the 2017 customer satisfaction survey, showed a reversed trend: 76% of the participants would have found it easier to participate in the meeting if it had been possible to speak their L1 (Majcen 2018

ENGLISH ONLY IN THE EU?
Given that a vast amount of oral and written communication in the different EU institutions already takes place in English and that with 24 official languages, the costs for interpreting and translation have risen exponentially2, some argue that English should be the only working language in the EU (i.a. Van Els 2005, Van Parijs 2011). In the 2015 SCIC survey, 5% of the 250 suggestions received by delegates proposed just that. One argument put forward by critics is that the EU language regime «has become economically unsustainable» (Cogo/Jenkins 2010, 272). Some argue that a monolingual regime would dispense with translation and thus save a lot of money (Ozvalda 2005, 76). This conclusion, however, does not take into account that in many cases the translated texts are legal texts that -in case of a regulation -directly enter the member states' body of law. Thus, they have to be translated anyway, the only difference being that the cost would have to be borne by the member states themselves. Hence, it would not lead to a reduction of costs, but rather a reallocation of costs from the European to the national budgets. This scenario would actually be more unfair, as a small member state with a lower GDP, such as Malta, would have to bear the same costs as a big member state with a higher GDP, such as France (Reithofer 2014, 53). Furthermore, the question of legal certainty of those texts would arise (Volz 1994, 91). Gazzola/Grin (2013) also highlight that the costs of multilingualism are not limited to the primary costs, that is direct costs for translation and interpreting, because one must not forget secondary costs such as reduced productivity of staff or mental fatigue when having to work in a foreign language (ibid., 99). Furthermore, implicit costs can arise, that is «costs borne by natural or legal persons […] who cannot interact in their first language with a public authority […], because it is not included in the set of official languages of the organisation considered.» (ibid., 100). These comprise translation and interpreting costs as well as alienation or loss of prestige of languages, that is, effects that are hard to quantify. The authors' analysis of a possible «English-only» situation in the EU also comes to the conclusion that «a multilingual, translation-based language regime is both more effective and fair than a unilingual regime based on English […].» (ibid., 104), since about 49% of EU citizens do not speak English at all and only few citizens have language skills that actually grant them access to all EU documents and debates in English. Moreover, the knowledge of English is not equally distributed among citizens of different member states.
As for the alleged economic unsustainability of translation and interpreting one has to bear in mind that the costs amount to less than 1% of the EU budget (ibid., 100). The annual expenditure per EU citizen for interpreting is EUR 0.23 (DG SCIC 2015b). Another argument put forward by those who are against multilingualism and a broad interpreting regime in the EU are the supposed quality flaws in communication. Van Els (2005, 274f.) claims that in interpreted meetings, delegates use short sentences, avoid idioms and jokes and often renounce the use of linguistic nuances in order to facilitate the interpreters' work, yet he fails to deliver empirical evidence for it. Since many meetings -i.a. of the European Parliament -are webstreamed, anyone can decide for himself or herself whether the speakers there speak in a simplified way when using their mother tongue or if this is rather the case for non-native speakers.

Karin Reithofer
Furthermore, Van Els claims that often delegates do not listen to interventions because they are interpreted: [M]eetings may last a long time and it may quickly become tiring to listen to a large number of interventions in translated form; it is by no means unusual that participants only listen to interventions in those languages of which they have a passive knowledge and do not bother with the translations of the other languages. (Van Els 2005, 274) He describes his source as «relatively anecdotal evidence» (ibid., 275). While listening to interpretations can without doubt be tiring occasionally, it has to be underlined that in many of these meetings legal acts are being drafted that apply directly to the EU member states. Hence, it is rather unlikely that delegates would be content with following only part of the debate on the texts. Furthermore, the SCIC customer satisfaction surveys show the opposite of what Van Els claims: the respondents rate the quality of interpreting very high with around 90% of the respondents stating they are satisfied or very satisfied with interpretation (DG SCIC 2015a).
Apart from the many advocates of a unilingual EU, there are also numerous voices who criticise the growing use of English in the EU (i.a. Wagner 2001, Wooding 2002, Gazzola/Grin 2013, Luccarelli 2016. It is often -but not exclusively -the multilingual staff of the language services who see themselves as guardians of multilingualism and want to limit an ever-growing influence of English (Phillipson 2003, 130).
Another argument against an English-only EU is that this would lead to unfair advantages for native speakers of English (Dollerup 1996, 34, Phillipson 2003 who would be able to speak their mother tongue while others might suffer a reduction of their political weight due to not being able to negotiate in their preferred language (Gazzola 2006, 398).

BREXIT -ALSO THE EXIT OF THE ENGLISH LANGUAGE?
When the majority of the UK's population voted for the country to leave the European Union, a shock wave went through Europe. It was not only the political and business world, however, that was deeply shaken. The international press heralded dramatic changes for Union's language regime as well saying that with the UK leaving, English could disappear as an official EU language. Some French MEPs advocated a ban of English in the institutions. The Commission's president Juncker -famous for his multilingual speeches -chose not to speak English when addressing the European Parliament after the Brexit referendum. There were rumours that the Commission would use French and German more often in its external communications to make a clear statement towards the British. The aforementioned Danuta Hübner even stated in a press conference that English was not going to be an official EU language anymore (Goulard 2016). She referred to the rule in the regulation on the EU's official languages, which stipulates that every member state shall nominate one official language. Since English was already an official language, requested by the UK, when the two other EU members who use English as one of their official languages, Ireland and Malta, joined, they nominated their other official languages, Gaelic 3 and Maltese, respectively. Thus, with the UK's exit, English -the language the Brits had nominated and for which they currently pay the lion share in interpretation in the Council -would cease to be official. The European Commission representation in Ireland reacted to the agitated reports by pointing out that any change to the EU institutions' language regime had to be voted on unanimously in the Council, implying that Ireland would for obvious reasons not vote for a ban of English (EC Ireland 2016).

Karin Reithofer
Regardless of whether these reports were deliberate scare mongering or were based on a lack of information, most insiders knew that English would not simply disappear from the EU institutions and will agree with the following opinion on the functioning of English as a lingua franca in Europe: That's not how lingua franca's [sic] work. They're just not something which is bureaucratically decided. They're absolutely whatever people find useful as a method of communication, just as with the myriad of pidgins and creoles around the world. The use of English in Brussels is, as with its use across much of Europe, a market driven outcome. (Worstall 2016, n.p.) One might even argue that once the UK has left the EU, it will be less contentious to use English as a lingua franca, because it will no longer be the language of one of the large member states that might benefit from this language regime, but that of two rather small members. to many new developments such as the growing demand for remote interpreting, the most frequent complaint -by civil servants and freelance interpreters alike -is the increasing use of English, often spoken by non-native speakers. It has been shown elsewhere (Reithofer 2010) that the negative stance of interpreters towards the rise of English and non-native speakers is a result of work-related as well as economic and psychosocial reasons.

Non-native speakers
Rants about «BSE» -bad simple English -have become as frequent as complaints about the infamous Brussels weather. Interpreting blogs and EU-internal interpreter fora are full of criticism about non-native speakers of English (e.g. De Rioja 2013).
As far as work-related issues are concerned, it has been shown that non-native speakers and their accents and deviations from standard English are a major stress factor for interpreters and can have adverse effects on interpreting delivery (for an overview see Reithofer 2010Reithofer , 2011. While research in ELF studies suggest that ELF communication is mostly successful (e.g. Seidlhofer 2001, 137), interpreting occurs in a unidirectional communication setting that does not allow for common ELF strategies such as repetition, paraphrasing, follow-up questions or the negotiation of meaning (Reithofer 2010(Reithofer , 2014. In scholarly literature, Dollerup (1996, 35) claimed already 20 years ago that the delegates' English skills are often insufficient and lead to misunderstandings, often because speakers overestimate their own mastery of languages (Dollerup 2001, 13). Phillipson (2003, 134) refers to accounts of funny situations, but also real communication problems due to non-native speakers.
Jones (2014) -himself a staff EU interpreter -describes the use of «globish» as one of the three main obstacles to communication in international meetings. To him, most non-native speakers misjudge their own language skills: The vast majority of speakers who choose to speak English as a foreign language in international meetings overestimate their competence. They may have a good accent, they may have a reasonable grasp of syntax and grammar, often they will have an excellent knowledge of the relevant technical vocabulary. But still they are not able to express themselves clearly and precisely in English. (Jones 2014, n.p.) Albl- Mikasa et al. (2017) found in their recent study with student interpreters that the biggest problem trigger for interpreters of non-native speakers of English is their accent in combination with incorrect use of words in relation to context. Because of the «speakers' constrained pragmatic fluency and restricted ability to express their key point in precise terms […]  logic behind the phrases and expressions.» (ibid, 232) This seems to primarily lead to omissions. Grammatical irregularities in non-natives' use of English, on the other hand, appear to be less of a problem. More research is needed for an evidence-based overview of the main problem triggers.

Non-native listeners
It is, however, not only non-native speakers of English who seem to present a challenge: non-native listeners also add to the EU interpreters' discontent. As mentioned above, customers who have to or want to listen to the interpretation into English even though it is not their mother tongue are less satisfied with the interpreters' performance than native speakers of English (DG SCIC 2015a).
Many interpreters -such as David Smith, former head of the English interpretation unit at DG SCIC -believe that non-native listeners are often not proficient enough in English: «Whereas thirty years ago very few delegates in continental Europe spoke English, nowadays it seems most of them do -or at least believe they do -and they are in consequence more critical of interpreters' performance.» (Smith 2013, n.p.). Furthermore, this lack of proficiency might mean that following a technical meeting becomes more tiring and thus generally less pleasant.

Psychosocial effects
Apart from concrete work-related problems, interpreters also perceive the predominant role of English and non-native speakers as a threat to multilingualism which ultimately is the raison d'être for the EU's interpreting services. The trend of more and more English-only meetings endangers the interpreters' existence: In many ways translation and ELF […] seem incompatible, mutually exclusive solutions to the perennial human need to communicate across language barriers. The growth of a lingua franca can end a need for translation and for this reason, taken at face value, the growth of ELF seems destined to influence detrimentally the very multilingualism which translation both depends on, and sustains. (Cook 2012, 256) This fear is fuelled by the fact that in the EU, the demand for interpretation has diminished significantly since the Juncker Commission presented a plan to reduce the number of legislative initiatives in favour of a focus on important files (De la Baume 2015). The insecurity is tangible in the EU interpreters' realm, for obvious reasons in a more accentuated way among freelance interpreters who are faced with an actual reduction of income, but also among civil servants who sometimes feel superfluous. Smith describes how the perceived importance of interpreters has diminished over the years: «We feel sometimes we are perhaps almost ornamental. We are being put in there for political reasons but we don't really in every meeting play that important role as a communicator.» (Smith 2013, n.p.) This leads to a situation in which interpreters find it hard to see any good in English being used as a lingua franca.

The interpreting services' reaction
DG SCIC's strategic plan shows that the Commission's interpreting service acknowledges the ELF reality and its impact on the interpreters' work, but believes that not much can be done about it: The quality of interpretation may be negatively impacted by different factors (e.g. a limited language coverage in meetings due to budget restrictions in Member States, the quality of the speakers' contributions, speakers not using their mother tongue or reading out speeches), which are out of DG Interpretation's control. (DG SCIC 2016b, 7) One of the concrete reactions to the rise of English as a lingua franca is to include the topic of non-native English in awareness-raising seminars organised by DG SCIC for newcomers in the European Commission and national delegates of an upcoming Council presidency. Apart from preparing these future speakers for the work with interpreters in general, they are strongly advised to speak their mother tongue and are given reasons as to why this is in their own interest. One of the main issues with the seminars is the relatively low turnout, since future Commission officials are not obliged to attend them (Iturri 2013).
Furthermore, it seems to be rather difficult to convince people that speaking nonnative English might have a negative effect on their communicative goals. According to the awareness-raising coordinator for DG SCIC José Iturri, many argue -like in the satisfaction surveys -that they are not used to presenting certain topics in a language other than English, not even in their mother tongue (Iturri 2013 and personal communication).
The EP's interpreting service, DG LINC formerly DG INTE, has long been communicating the rules of working with interpreters, mainly to MEPs and their assistants. However, it appears to be quite challenging to convincingly describe the disadvantages of using ELF to non-linguists or to explain that sticking to one's native tongue when giving a speech is not a favour to the interpreters, but might actually increase the communicative impact, for direct listeners as well. An awareness-raising campaign at the EP to that effect does not seem to have been understood as intended, since the BBC news online reported that «[t]he European Parliament's chief of staff has urged MEPs to speak slower and stick to their native language, to help the interpreters.» (BBC 2016). This again gave the impression that speakers had to support the interpreters and not the other way round. To avoid such misunderstandings, any type of encouragement to speakers to speak their mother tongue must be substantiated by Karin Reithofer Interpreting and multilingualism in the EU: Leave or Remain? CLINA vol. 4-1, June 2018, 115-136 eISSN: 2444-1961 Ediciones Universidad de Salamancacc by-nc-nd facts that prove the added value for them. Our profession will not benefit from being seen as an additional difficulty rather than as facilitators of communication. DG LINC is well aware of this and is currently working on a new awareness raising campaign that promotes a shift in focus from the interpreter to effective multilingual communication (internal report and personal communication by DG LINC Strategy Unit).

POSSIBLE STRATEGIES FOR AN ELF REALITY
A factor that must not be overlooked in times of ELF is professional excellence, even when working from a non-standard source text. The added value of interpreting in respect to English-only has to be apparent or as David Smith (2013, n.p.) puts it: «Unless we can demonstrate that we are measurably better than the people listening to us in our linguistic skills, then we are out of our jobs.» It is not only about getting the message across, but doing it with an extraordinary degree of skill (Wooding 2002).

Training
To be able to deliver such a service under adverse circumstances, experienced professionals as well as novices would have to be trained accordingly as already demanded by scholars as early as 1952 (Herbert) and even more so since the marked rise of English (among others Albl- Mikasa 2013, Jones 2014, Kurz/Basel 2009, Reithofer 2010. While the provision of in-house language and thematic trainings for the EU services' interpreters have long become standard, a systematic preparation of interpreters for the work with non-native speakers is not on the training agenda (yet).
However, further research is needed to determine what this preparation might look like. In Zurich, a pilot project with a promising didactic programme using ELF/ESL source texts has been launched (Albl-Mikasa et al. 2017). Such programmes would have to be further tested and adapted to professional interpreters' needs and circumstances.
Furthermore, experiments have already shown that familiarity with non-native speech and contact with non-native speakers (NNS) are the most relevant factors for NNS intelligibility, with a statistically significant difference between listeners of NNS with and without this type of experience (Reithofer 2013, 67, Reithofer 2014.
Since the variations in non-native English largely depend on the speakers' first language, the exposure to different NNS has to be maximised. Studies (Chang/Wu 2014, Katikos 2015 show that interpreters understand ELF speakers with whom they share an L1 best and find speakers with significantly different language backgrounds most difficult to interpret. While EU interpreters are constantly exposed to different European ELF varieties, they usually interpret NNS under stress. It might be beneficial for them to work from NNS source texts in a learning environment, in which appropriate coping strategies could be consciously applied. Special research-based trainings with different ELF varieties would allow interpreters to prepare for this challenging task in a more relaxed setting that allows for specific preparation as well as reflection and analysis of their performance.

Karin Reithofer
As a basis for these seminars, the different language units might be able to supply information on main L1 transfer features in «their» ELF variety and provide colleagues from other booths with data on most recurrent false friends and the like. It might be beneficial to create info sheets on the different ELF varieties that could be regularly updated as a work in progress.
Any didactic model and strategy would have to be shaped in close cooperation with the interpreting services.

Preparation for assignments
Thorough preparation for assignments is of even higher priority when interpreting NNS with low proficiency levels or strong accents (see also Albl-Mikasa 2013, 9). When consolidated interpreting strategies such as anticipation are hampered by the typical features of non-native speech as described above, interpreters need to be able to increasingly resort to very specific background knowledge on the subject to be able to infer the intended meaning. This additional preparation effort would have to be recognised by the services and added to the calculation of interpreters' working hours.

Accommodation of NNS listeners
As described in sections 2.3.1. and 5.1.2. non-native listeners are less satisfied with their interpreters which raises the question what possible implications this finding might have. If in the near future the English booth's main audience are nonnative speakers and one of the service's main objectives is still «meeting the clients' needs» (DG SCIC 2016a, 3), then one might have to think about how to achieve this objective. If we assume that many non-native listeners are not that satisfied with English booth interpreters because of insufficient language skills in English, then there are two possibilities: either they would have to be convinced to request interpreting into their own mother tongue or, in theory, the English booth output would have to be adapted to the imperfect non-native listeners. Adjusting their output to a more heterogeneous audience would most likely be regarded a sacrilege by most English booth interpreters. Nevertheless, in foreign language communication, it is common practice for native speakers to adjust to or accommodate non-native speakers (e.g. Mauranen  48ff.). This does not necessarily imply that interpreters would have to speak an artificial language like Ogden's «Basic English», but it would surely mean that the typically English cricket metaphors would not be approved of (see also Albl-Mikasa 2013, 10). Such an undoubtedly contentious decision is not likely to be endorsed without a policy decision that would have to be initiated by the EU interpreting services. It would certainly have to encompass extensive explanatory work among interpreters as well as clear, evidence-based indications as to what kind of accommodation might be useful. I doubt, however, that professionals would greet such a decision with enthusiasm. In a broader context, this would also lead to a discussion on how we interpreters define our role: Do we see ourselves as guardians of the (English) language or rather as communication facilitators?

Changing attitudes
As explained elsewhere (Reithofer 2010(Reithofer , 2013(Reithofer , 2014, interpreters should accept that ELF does serve its purpose in many settings and recognise that interpreting and ELF are complementary modes of communication: i.e. plenary sessions or webstreamed events with interpreters vs. informal communication settings with ELF. A very negative stance or even resentments towards NNS might be detrimental when having to interpret them. Research shows that intrinsic motivation and attitude towards a task are important performance determinants (e.g. Cerasoli et al. 2014). Thus, a negative attitude might have an adverse effect on interpreting quality as it might stop us from fully exploiting our analytical potential. Interpreters should be made aware of this and informed about findings on ELF and interpreting. A first step was taken in September 2017 when the ELF expert David Graddol and I were invited by DG LINC to give an overview of the field as part of the interpreters' training programme.

More evidence-based and targeted information campaigns
My empirical study showed that while ELF often does serve its purpose in less formal, dialogic interactions which allow strategies such as repetition or clarification, in monologic interactions such as technical presentations and lectures, interpreting provides a significant communicative advantage (Reithofer 2013, 68). Surprisingly, many speakers use ELF instead of their mother tongue especially in such instances of monologic communication, e.g. at EP hearings or in technical Commission meetings. I believe that one reason for this is that these speakers often come from outside the institutions and are not as used to interpreting as delegates or MEPs. This could be an area where the interpreting services might want to increase their effort to raise awareness and make interpreting a standard feature in the briefing for external experts at the time of their invitation. It seems that the possibility of relying on interpreters is not Karin Reithofer Interpreting  communicated in a way that convinces speakers sufficiently. Possibly, the information on the language regime often arrives too late, when the speakers have already prepared their statements or speeches. DG LINC's strategy unit is well aware of these shortcomings that are often beyond their influence and is working on a new awareness raising campaign to circumvent them (internal report and personal communication).
Another alleged disadvantage of the use of ELF is that communication generally seems to decline in monolingual situations. Luccarelli (2016) laments that Englishonly conferences often lead to one-directional communication and a decline in active participation. There is, however, no scientific evidence that confirms this hypothesis. This might be an area that the interpreting services might want to help further explore to be able to present arguments in favour of multilingualism.

CONCLUSIONS
In conclusion, the data assembled above goes to show that ELF is not a passing trend in the EU, but has come to stay. It poses a real and growing challenge to the principle of multilingualism and thus interpreting in the institutions. Hence, there is an urgent need for an integrated, coordinated strategy on the handling of the phenomenon of ELF by the EU's interpreting services. It would have to be based on academic findings and ascertain that a comprehensive picture of the disadvantages of ELF as well as the evident advantages is presented and explained to users and interpreters alike. Settings where interpretation presents an added value would have to be identified and contrasted with situations in which interpreting might indeed be less needed. Users and professionals alike would have to be made aware of the pros and cons of ELF and interpreting respectively (Reithofer 2013(Reithofer , 2014 in order to be able to react appropriately to this new work reality.