131LANGUAGE Attitudes towards Euro-English in a European Union Institution ABSTRAcT This study deals with the attitudinal aspect of Euro-English, denoting a specific form of the English language that is frequently used within the institutions of the European Union. A questionnaire survey was conducted among 285 representatives who work for one of these institutions in Brussels. The respondents were asked to rate several deviations from Standard English, identified in a corpus-based analysis of EU texts, as either ‘acceptable’ or ‘unacceptable’ English usage. The findings reveal that the high acceptability rates of the proposed features among the non-native English-speaking respondents were mainly related to their proficiency in English and/or mother tongue interference. Moreover, since native speakers of English also accepted most of the proposed deviations, it follows that the participants did not seem to be aware of non-standardness in the test sentences. Euro-English must be regarded as EU jargon due to its technical, administrative or legal nature and not as a separate non-standard form of English for EU institutional settings. Keywords: European Union institution; Euro-English; corpus linguistics; deviations from Standard English; attitudes towards language; Eurojargon Odnos do evroangleščine v opazovani inštituciji Evropske unije POvZETEK Ta raziskava analizira odnos 285 predstavnikov ene izmed inštitucij Evropske unije do t.  i. evroangleščine kot oblike angleščine, kakršno pogosto uporabljajo v svojem večjezičnem delovnem okolju. Raziskava je bila usmerjena na ugotavljanje sprejemljivosti določenih jezikovnih vzorcev, ki odstopajo od modela standardne angleščine. Z analizo pisnih besedil Evropske unije v angleškem jeziku smo določili, katere izmed le-teh vključiti v testne stavke, s katerimi smo anketirance soočili. Analiza podatkov za nerojene govorce kaže na to, da obstaja določena povezanost med sprejetjem predlaganih odstopanj od standardne angleščine in znanjem ciljnega jezika in/ali prepletanjem maternih jezikov. Ker so tudi materni govorci sprejeli večino predlaganih značilnosti, ugotavljamo, da anketiranci evroangleščine ne dojemajo kot nestandardno obliko angleščine znotraj inštitucij Evropske unije, temveč kot EU žargon zaradi svoje tehnične, administrativne ali zakonodajne narave. Ključne besede: inštitucija Evropske unije; evroangleščina; korpusna lingvistika; odstopanja od standardne angleščine; odnos do jezika; EU žargon Tina Balič University of Maribor Slovenia 2016, vol. 13 (2), 131-152(246) revije.ff.uni-lj.si/elope doi: 10.4312/elope.13.2.131-152 UDc: 811.111'27(4) 132 Tina Balič Attitudes towards Euro-English in a European Union Institution Attitudes towards Euro-English in a European Union Institution1 1 Introduction: English as the Primary ‘In-House’ Working Language Although the European Union (hereafter the EU) currently has twenty-four legally established official and working languages (enshrined in the council Regulation No 1 of 1958), the picture that emerges in the everyday practice of its various institutions and bodies, where communication is typically conducted with no or very few native speakers (hereafter NSs) present, is one in which their number is drastically reduced to only a few, i.e., principally English, French, and German, whereby the source language of the working documents is predominantly English. Particularly the 2004–2007 enlargements – when central and Eastern-European representatives and their staff joined the EU institutions (whose common language of communication was English) – have significantly strengthened the role of English. This is mainly justified by time and budgetary constraints as well as the need for a mutually intelligible language within the EU’s administration that manages the day-to-day business of the EU as a whole (Graddol 2001, 47; Seidlhofer 2011, 55; Berns 1995, 3–11). That English is indeed the primary working language of the surveyed EU institution is evident from the most recent figures provided by its Translation Service (personal communication, December 11, 2014), according to which as many as 81.3% of original documents were drafted in English in 2014 (as compared to 77.6% in 2012 and 62% in 2004), followed by French, with only 5%. consequently, the EU’s commitment and efforts to ensuring and promoting language equality are often overshadowed by the pre-eminence of English as the ‘de facto’ lingua franca of the EU (Seidlhofer 2011, 55; Graddol 2001, 47), giving rise to the concept of Euro-English (hereafter E-E). In the literature the term E-E in its broader sense commonly denotes English as spoken by its non-native speakers (hereafter NNSs) within continental Europe (Strevens 1992; Jenkins, Modiano, and Seidlhofer 2001; McArthur 2003; crystal 1999; Mollin 2006; Modiano 2003; Seildhofer 2011; Berns 1995). More narrowly, it is often associated with a specific usage of English within the EU institutions (Jenkins et al. 2001, 13; Modiano 2006; crystal 1999, 15; Seidlhofer 2011, 17; Mollin 2006, 92; Trebits 2008, 2009; Jablonkai 2009) that their numerous representatives, who predominantly come from mother tongue (hereafter L1) backgrounds other than English, use for conducting the day-to-day business of the EU. For the purposes of this study, whenever the term E-E is used, it should be understood as denoting a ‘nativised’ (i.e., non- standard) form of English as used in EU institutional settings that is different from any standard model of the English language (hereafter SE). Our analysis was limited to lexico-grammar as the more promising area in terms of obtaining potentially interesting results of E-E, which is assumed to be in its infancy (Jenkins et al. 2001). In this context, Modiano (2006, 223) argued that the predominant role of English within the EU will inevitably have a significant influence on its forms (and functions) within continental Europe. 1 In order to ensure the desired anonymity, the EU institution is not disclosed in this article. In addition, the views expressed are exclusively those of the author based on the findings obtained in this study and do not in any way reflect the opinions of the surveyed institution. 133LANGUAGE However, from a sociolinguistic point of view, as Seidlhofer (2011, 83) highlighted, variety status depends “considerably less on objectively identifiable linguistic features than on the existence of a relatively stable community of speakers who themselves acknowledge that they speak a variety of their own”. Besides, important remarks of great relevance for this study were made by Llamzon (1969) and Petzold (2002), when they both emphasised the importance of the ‘educated elite’ in defining linguistic standards within a speech community. Likewise, Schneider (2007) pointed out that future research into NNS varieties should be based on a careful usage of the ‘educated’ users as representatives of the society as a whole. Accordingly, given that the attitudinal aspect is crucial for the acceptance of any non-standard model, this study was aimed at investigating attitudes of diverse representatives within the multicultural and multilingual communicative contexts of the EU institutions towards those uses of English that deviate from SE. 2 Methods 2.1 The Research Question The research question was formulated based on the aforementioned supposed importance of the ‘educated elite’ in defining linguistic standards within a speech community (see Llamzon 1969; Petzold 2002; Schneider 2007). Accordingly, we examined what attitudes (i.e., positive, negative or ambivalent) the representatives from different EU member states who work for the surveyed EU institution have towards the kind of English that they frequently use for their work in business matters. In this situation, the tens of thousands of EU representatives are regarded as a ‘European’ community of practice that use English as a working language in EU institutional settings and are unrestricted to a Member State (hereafter MS) where English is a L1, i.e., the United Kingdom and the Republic of Ireland (Berns 1995, 7). In fact, several studies in sociolinguistics, but also in anthropology and social psychology, conducted since the mid-20th century have shown how attitudes influence the perception of languages at all linguistic levels (e.g., accent and pronunciation, spelling and grammar, vocabulary) and can lead to different evaluations of linguistic standards, depending on who the speakers are (Seidlhofer 2011, 35). 2.2 Data collection and Analysis For the purposes of this attitudinal study, two methods of data collection were used. While the corpus-based approach was applied as the technique better qualified for studying language usage within EU-specific contexts (i.e., EU institutions) as well as comparing it against SE (Biber, conrad, and Reppen 1998; Gries 2009), a questionnaire was used as a less time-consuming and intrusive method as compared to face-to-face or telephone interviews for surveying attitudes of EU representatives. This gave the respondents adequate time to complete the questionnaire within their own time-tables, while respecting their right to anonymity. The pilot study revealed that EU representatives are generally not so easily approachable when using the interview method and actually favour self-completion surveys over interviews, mainly due to their busy schedules. As well, respondents such as high-ranking officials, people in high income groups or the highly educated tend to be more conscious of socially acceptable behaviour and may feel greater pressure to conform to social norms (Karp and Brockington 2005; Kothari 2004, 99). 134 Tina Balič Attitudes towards Euro-English in a European Union Institution The data elicited from the questionnaire were analysed using the statistics software SPSS 14.0. correlation analyses were performed in order to examine the relationships between stated attitudes and relevant socio-demographic variables on the one hand and between all acceptability judgements and relevant socio-demographic variables on the other. A dependency was considered as significant when the probability of rejecting the null hypothesis in error was less than 5% (i.e., significance level α = 0.05). Great significance was assigned to a significance level α = 0.01. In addition to the SPSS, certain charts and calculations were performed in Microsoft Office Word and Excel. 2.3 The Research Setting The main EU institutions were chosen as the target setting of this study mainly because they represent truly EU-centred multicultural and multilingual professional contexts, in which English-mediated communication is dominated by NNSs who interact predominantly among themselves. Also, since they are mostly located in one city (i.e., Brussels), there are unlimited opportunities for face-to-face communication and discussions in English. Last but not least, English is used for specific purposes within the EU institutions (Jablonkai 2009; Trebits 2008, 2009), i.e., to legislate and conduct the daily EU business. Hence, there is a constant and an ever- increasing need to use the language in order to enable the mutual understanding and successful functioning of the EU as whole. considering that previous research of E-E did not provide any convincing systematic evidence to support its existence within continental Europe (Dalton-Puffer et al. 1997; Murray 2003; Sing 2004; Mollin 2006; Simeone 2011; Forche 2012), we are at this point in time only able to predict that if a non-standard ‘European’ form of English was indeed used within the EU, it would show its signs first in a relatively restricted community of speakers who deal with specific subjects and use the English language substantially more frequently than other European speakers (carstensen 1986; Jenkins at al. 2001, 13; Berns 1995, 6–7; crystal 1999, 15; Jablonkai 2009; Mollin 2006, 92; Trebits 2009, 200). 3 Empirical Analysis of E-E Linguistic Features Representatives of EU institutions regularly produce different types of texts in all EU official languages but predominantly in English, and these represent a fertile ground for linguists analysing naturally occurring language. The main aim of the corpus-based analysis in this study was to determine which of the more or less systematic uses of E-E that deviate from SE should be integrated into the acceptability test described in Section 4. Euro-English was analysed based on the EUROPARL7 corpus (Koehn 2005), which is available in the Sketch Engine concordancing programme (Kilgarriff et al. 2004). The respective corpus is extracted from large numbers of European Parliament (hereafter EP) discussions in English (i.e., it contains 2,218,201 sentences and 53,974,751 words) and includes material from various subject areas, ranging from economy to social issues, education and training, science, food and health industry, environment, trade, etc. Our decision to choose a specialised corpus, as opposed to a more general one, was further justified by the fact that we did not anticipate, at least in the short-term, to find genuine ‘nativised’ tendencies within all domains of Europe where English is used. Importantly, it can be argued that the authors of the corpus material are, as anticipated, congruent as much as possible with the survey participants. Namely, they all work for one of the main EU institutions where its culturally and linguistically diverse representatives regularly use English as the most useful ‘instrument’ for communication that cannot be conducted in their L1s. 135LANGUAGE 3.1 Results and Interpretation Several pre-defined formal features identified in the literature as potentially distinctive of E-E (and of NNS varieties of English in general) were searched for their frequencies of occurrence within the EUROPARL7 corpus (James 2000; crystal 2003; vOIcE 2009; Seidlhofer 2011; Mollin 2006), but also in the publicly available database of EU legislative texts, i.e., EUR- Lex, where appropriate. Moreover, they were compared against the British National corpus (hereafter BNc), which represents a wide range of spoken and written British English (hereafter BrE). However, the European court of Auditors report of 2013 (hereafter EcA 2013) titled “Misused English words and expressions in EU publications” served as the primary basis for our study of E-E, accompanied by EU glossaries. These sources offer extensive lists of precisely those lexical and grammatical features that were identified as distinctive of the kind of English that is frequently used in EU institutional settings. All in all, according to the results obtained in the linguistic analysis of EU texts, it was decided to study respondents’ attitudes towards those deviations from SE that were deemed more likely to be among the affected areas due to their rather systematic usages within the E-E corpus, as compared to the general BNc. Some of the more interesting findings are presented subsequently. With regard to grammar, ‘aid’ meaning “money, food, etc. that is sent to help countries in difficult situations”2 and ‘competence’ meaning “the power of a court, an organisation or a person to deal with something”3 were frequently used as if they were countable with the designated meanings in the EUROPARL7 corpus: whereas the former occurred 415 times out of 2,308 cases referring to financial or economic assistance that is given to help or support the member states (e.g., 166 times referring specifically to ‘state aid’ to mean “government assistance to enterprises on selective basis”), the latter was used as many as 2,109 times out of 5,901 cases denoting the ability of the various EU institutions and its other bodies to influence or control. In addition, a common pattern observed in the EU material were phrases with ‘of ’ instead of the possessive ’s constructions or noun-noun compounds, as in “I very much appreciate the report of Mr Di Pietro, which shows his excellent work in such a short period of time” (instead of: Mr Di Pietro’s report) or “The aim of the communication of the commission is to introduce a series of measures which allow the potential which electronic commerce offers the Internal Market to have full effect” (instead of: commission communication). Although such usage is not grammatically incorrect in SE, it may lead to ambiguity, especially since EU texts are already too technical and complex to be understood by the average reader. certainly, the most obvious differences were observed at the level of lexis (see Annexes 1 and 2). We found many examples of Euro-specific vocabulary that only occurred in the EUROPARL7, i.e., ‘Euro area’ (2,076), ‘Neighbourhood policy’ (1,519), ‘comitology’ (1,131), ‘Euro zone’ (1,052) or ‘Euroscepticism’ (95), as well as those that were significantly over-represented there, as compared to the texts of the BNc, i.e., ‘Member State’ (16,165 vs. 2,287), ‘subsidiarity’ (4,374 vs. 184), ‘additionality’ (236 vs. 44) or ‘Acquis communautaire’ (611 vs. 8). In addition, the following words or phrases were commonly used with the designated E-E meanings in the EU texts examined: ‘third country’ to denote “any country that is not part of the EU”; ‘to attribute’ used as a verb to mean “to allocate, grant, give or award”; ‘coherent’ to mean “in agreement with something or somebody”; ‘fiche’ to denote anything from a document, paper, and sheet to entry, 2 Hornby, A. S. 2015. “Aid” Def. 1. Oxford Advanced Learner’s Dictionary of current English. 9th edition. See also Def. 2. 3 Hornby, A. S. 2015. “competence” Def. 2. Oxford Advanced Learner’s Dictionary of current English. 9th edition. 136 Tina Balič Attitudes towards Euro-English in a European Union Institution record, and file, as well as ‘perspective’ to mean “an expectation for the future”. Similarly, we found several instances of certain words or phrases which occurred with a more or less correct meaning there but in different contexts than in SE, these being: ‘semester’ to refer to periods of six months in contexts other than education; ‘homogenise’ as a verb meaning “to make something uniform or standardised”; ‘hierarchical superior’ instead of a boss or superior, as well as ‘transmit’ used as a synonym for “to send something off by letter or email”. One control feature that did not occur at all in the EUROPARL7, i.e., the French word ‘delay’ to mean “deadline” in English was included in the acceptability test in order to verify if the respondents, who are generally proficient in English, recognise the more obvious deviations from SE.4 Last but not least, an analysis of four-word5 lexical bundles showed that all of the identified examples meet the criteria of occurring more than 40 times per million in the EUROPARL7 corpus – the first one ‘I would like to’ [do something] being as frequent as 679.3 per million.6 A similar structure, i.e., ‘I should like to’ [do something], ranked fourth on the top-ten list, occurring as many as 325.7 times per million. Accordingly, considering that EP discussions from which E-E is extracted in this study are conducted orally, the first observation we can make is that sequences that are centred on pronouns and clause fragments (i.e., a verb) seem to be more typical of spoken than written registers (i.e., primarily speeches); especially if compared against the BNc, where respective lexical bundles occurred less than 40 times per million (see Annex 3). With regard to the structure of the lexical bundles, the findings moreover reveal that the two most common patterns in the EU texts involve noun phrases (mainly with of-phrase fragments) or prepositional phrases, as in ‘countries of the European Union’, ‘Members of the European’, ‘on behalf of ’ or ‘on the basis of ’. These results are highly consistent with previous findings, as written registers, such as EU official documents and EU online news (Jablonkai 2009), as well as textbooks (Biber, conrad, and cortes 2004), were also found to be operating with bundles of this structural type. To conclude, the findings from the corpus-based analysis of EU texts have not revealed any serious tendencies towards a possible existence of a non-standard form of English for EU institutional settings. We can therefore only speak of a kind of EU jargon rather than of a separate ‘European’ form of English within the EU institutions. The identified words and phrases are mainly used within EU institutional settings with the aim of enhancing collaboration and shared understanding. They were developed for the (smooth) running and managing of the daily business of the EU, but particularly for the purposes of those working in its various institutions and bodies. It is clear that the new forms of lexical expression mostly come from the interactive contexts of the EU as a primarily political and economic union rather than a cultural entity. consequently, EU vocabulary is generally very technical, complex, and not easily understood by the general public. 4 In English, ‘delay’ always means that something is being late or taking longer than needed, while the French near-homophone “délai” denotes a specific date or time by which something needs to accomplished. 5 For an analysis of lexical bundles we focused on four-word collocations mainly because we noted that three-word lexical bundles are frequently part of four-word bundles and tend to be less common. 6 According to Biber and Barbieri (2007), a four-word lexical item must recur at least 40 times per million in a corpus to be considered a lexical bundle. This requirement has been adopted for examining lexical bundles in this study in order to ensure representative results as much as possible. Biber et al. (1999, 990) defined lexical bundles as “recurrent expressions, regardless of their idiomaticity or structural status”, which follow each other more frequently than expected by chance, usually as words in sets of two, three, four, five or more. 137LANGUAGE 4 Empirical Analyses of Acceptability Judgments of E-E Features Most scholars define an ‘attitude’ as a tendency to react ‘favourably’ or ‘unfavourably’ to a social object of certain kind, i.e., in our study a linguistic form (Gardner 1985; Baker 1992; Yule 1996). The status of a language is mainly determined by the distribution of power, welfare, and prestige in a particular society during a specific time period and under certain social, cultural, political, and economic circumstances. Therefore, social attitudes generally determine which linguistic variety, feature or style speakers will choose in a given situation, whereby a rich linguistic repertoire certainly helps them achieve communicative competence (Hymes 1974), i.e., the ability to successfully convey information while expressing individual or group identity and making use of it. Accordingly, languages are both powerful ‘instruments’ of communication and important ‘markers’ of social and cultural identity (Yule 1996). 4.1 Methodological Difficulties The main methodological difficulty of using an anonymous questionnaire was that we were unable to obtain feedback or additional information from the respondents on certain questions, which may have provided us with further insights into our research question (Mouton and Marais 1996; Kothari 2004). However, since persons within a specific organisation were surveyed, we were still able to use a more complex questionnaire than would normally be the case when targeting a general community, aimed at enriching data and attempting to balance out the aforementioned weakness. On the whole, both sampling and collecting relevant data were conducted smoothly, whereby the latter was facilitated with the incentives for timely return of the questionnaire and by a follow-up of those persons who had not responded initially. For the purposes of this study, it was decided that the advantages of using a questionnaire (e.g., they are practical, cost effective, relatively easy to analyse, familiar to people, reduce bias, etc.) can outweigh its potential disadvantages. 4.2 Research Sample In order to investigate the research question outlined in Section 2.1, an empirical study was designed. Instead of conducting an EU institution-wide analysis, it was decided to study the attitudes towards E-E within one EU institution, whereby a limited number of its representatives were surveyed. All questionnaire participants were randomly selected from the totality of representatives working for the surveyed EU institution, whereby two samples were obtained. The first one included 19 NSs and the second 266 NNSs of English from diverse MSs of the EU (see Annex 4). All participants were between 20 and 70 years old. Moreover, the survey involved a rather balanced proportion of women and men, i.e., 56.7% and 43.3%, respectively. Interestingly, the percentage of participants who are proficient users of English was surprisingly high, i.e., 86.6% claimed to have either c1 or c2 English proficiency levels according to cEFR.7 As anticipated, the majority of the respondents indeed use English every day at work (93.9%) and in their free time (85.2%). In essence, information about respondents’ gender, age, and self- assessed proficiency in English helped determine whether there is a possible relationship between these socio-demographic variables and their stated attitudes towards the use of English in EU institutional settings. 7 common European Framework of Reference for Languages 138 Tina Balič Attitudes towards Euro-English in a European Union Institution 4.3 Data Analysis Apart from gathering certain general information about the respondents, two questions were considered crucial. Firstly, the question that elicited the respondents’ target form when using English revealed that an almost equal percentage of them aim for English as used by speakers of English regardless of nativeness for communication in international settings such as business, diplomacy, advertisement, travel, etc. (hereafter EIL), and BrE (39.2% vs. 38.4%, respectively), whereas only 14.2% selected ‘English as spoken in continental Europe’ as their target model. While it is clear that E-E is not the desired form in the respondents’ minds, the fact that so many of them still chose EIL may indicate that English is becoming increasingly less ‘foreign’ within the EU and its institutions, thus acknowledging its importance and resulting advantages that proficiency in English may bring to its users world-wide, but particularly outside NS communities. Nevertheless, Pearson’s correlation examining the relationship between the respondents’ competence in English and the form chosen revealed that the more competent they claim to be in English, the less likely they are to select EIL or E-E (r=-0.155; p<0.05, 2-tailed) as their target form. One explanation for this may be due, in part, to a certain stigma attached to the notion ‘non-native’ (i.e., non-standard) within continental Europe in that only NSs of English are said to be able to speak the language properly and thus ‘own’ it. EU institutions generally follow SE that reflects the language use of its two MSs where English is a L1, i.e., the UK and the Rep. of Ireland, as their target model for English (EcA 2013, 4). Secondly, the question eliciting respondents’ attitudes towards English as frequently used within the EU institutions showed that as many as 81.5% of them agree that English there differs from SE. Furthermore, Pearson’s correlation revealed that the more competent respondents claim to be in English, the less likely they are to disagree with the assumption that the use of English within the EU institutions is different from SE (r=-0.203, p<0.01, 2-tailed). Likewise, the question concerning the ‘quality’ of written and spoken English there is particularly relevant. The results indicate that slightly less than half (44.9%) of the respondents agree that English as used in EU institutional settings is ‘good’ English. Nonetheless, a notable proportion of 26.9% disagree with this proposition and 28.2% are indecisive about it. Interestingly, Pearson’s correlation revealed that the more respondents recognise that E-E is different from SE, the less likely they are to agree that it is ‘good’ English (r=-0.256, p<0.01, 2-tailed). One of the reasons for the different responses about the ‘quality’ or ‘accuracy’ of English within the EU institutions may be that what is considered as ‘good’ English generally depends a lot on the attitudes of the people making the judgments. According to Gardner and Lambert’s (1972) definition of integrative and instrumental motivation, it could be argued that more integratively oriented speakers of English commonly define ‘good’ English with regard to strictly defined rules of linguistic correctness, whereas people who are more instrumentally oriented in using English would rather define it in terms of successful accomplishments such as strong communication skills in the target language leading to successful communication. As regards the term E-E, 63.9% of the respondents indicated they were familiar with the term, whereas 36.1% were not. Interestingly, Pearson’s correlation again revealed that the more respondents recognise that English as used in EU institutional settings deviates from SE, the more likely they are to associate it with E-E (r=+0.223, p<0.01, 2-tailed). At this point, it is important to remember that, in order to regard E-E as a specific form of English for EU institutional settings, there would have to be a community of speakers who are 139LANGUAGE not only developing and regularly using distinctly non-standard patterns of English but who are, moreover, genuinely and openly accepting them as ‘appropriate’ in their minds. In order to verify whether the EU representatives constitute such a community of practice, the results of the acceptability test in which the respondents had to rate eight test sentences that deviated from SE according to their perceptions of what is ‘acceptable’ English usage and what is not were crucial for answering the research question at hand.8 The main aim was to find out what form of English the surveyed NNSs aim at approximating when producing English, although the NSs were considered as well; after all, it is their L1 that has been appropriated for the purposes of conducting the daily EU business within the EU institutions. 4.4 Evaluation criteria The approach adopted can be summarised as follows. On the one hand, if a respondent rejected a sentence because he or she identified a proposed E-E feature and at the same time gave a correct solution for it in line with SE, the sentence was coded as ‘unacceptable’. In this way we assumed that respondents generally follow the approved norms of linguistic correctness when using English, although it needs to be considered that acceptability judgements on written language will more likely tap into speakers’ awareness of SE norms. In this respect, informal EU discussions would have been a more promising source of English usage for this study, as speech produced spontaneously in a conversation is considered more natural than when texts are read out in more formal settings. This was unfortunately impossible to do because EU institutions prohibit observing and recording conversations of any kind due to confidential information that may be discussed there. However, if a sentence was rejected and a proposed deviation identified, but a correction was not made in line with SE, it was decided item-by-item how to code it. In other words, its rejection due to rather trivial or irrelevant reasons for the purposes of this study (e.g., punctuation marks were added or removed, word order modified or a sentence shortened but the structure of that sentence was not significantly influenced, etc.) was still coded as ‘acceptable’. On the other hand, for those respondents who judged a sentence to be ‘acceptable’, it was considered crucial to examine the relationship between the accepted item and the participant’s proficiency in English. Again, it has to be considered that accepting a form does not automatically mean that speakers are striving towards a non-SE model, since they might accept the form but still not use it (frequently) in their English. In order to quantitatively examine whether the respondents whose L1 is not English accept or reject the test sentences, we calculated the value that appeared most frequently in the responses given (i.e., ‘the mode’) based on the nominal and dichotomous variables (i.e., two values), whereby we anticipated that ‘value 1’ for indicating ‘acceptance’ of a proposed deviation will appear the most frequently. As demonstrated in Table 1 below, in all cases (i.e., 88%), except in one (i.e., Q20), the surveyed NNSs accepted the proposed sentences even though they deviated from SE. 8 Instructions for the participants as regards the acceptability test were as follows: In this part of the questionnaire you are presented with the so-called acceptability test. It involves sentences for which you are asked to judge whether or not you find them to be acceptable English usage as quickly as possible. For the purpose of this questionnaire, a sentence is to be judged acceptable if you believe that it appears natural and appropriate in a context in which you would typically use it, even if it is grammatically incorrect. When you find a sentence to be unacceptable, please give the corrected version. 140 Tina Balič Attitudes towards Euro-English in a European Union Institution Table 1. Mode for the acceptance of deviations from SE by NNSs of English. Test sentences Mode % Acceptable Q18 The European commission verifies state aids in the European Union. 1 66.8 Q20 In order to respect the delay for transmission, the European commission can make small changes to the notification. 2 49.3 Q22 Policy fiche on external action can be found on page 43. 1 67 Q26 Any absence during core time must be approved from the hierarchical superior. 1 72.9 Q28 The 2011-2013 multiannual strategy paper was approved in the first semester of 2011. 1 82.2 Q30 commission fixes the standard values for imports from third countries, in respect of the products and periods stipulated in Annex XvI. 1 55.3 Q32 Each Member State shall transmit its Partnership contract to the commission within 3 months of the adoption of the cSF. 1 74.6 Q34 The EU has exclusive competences in the area of common foreign and security policy. 1 73.6 In order to verify with what level of significance it can be stated that the majority of the NNSs accepted most of the deviations from SE, the non-parametric chi-Square probability test was performed. The main result is seen from Table 2 subsequently, indicating that the significance level is less than 0.05 (see last column). What this illustrates is that the differences between the acceptance and rejection of the sentences as evaluated by the surveyed NNSs of English are statistically significant. Table 2. chi-Square Probability Test. Chi-Square df Asymp. Sig. Q18 The European commission verifies state aids in the EU 24.558 a 1 .000 Q20 In order to respect the delay for transmission, the commission can make small changes to the notification .043b 1 .835 Q22 Policy fiche on external action can be found on page 43 23.453 c 1 .000 Q26 Any absence during core must be approved from the hierarchical superior 41.613 d 1 .000 Q28 The 2011-2013 multiannual strategy paper was approved in the first semester of 2011 81.873 e 1 .000 141LANGUAGE Q30 commission fixes the standard values for imports from third countries, in respect of the products and periods stipulated in Annex XvI 2.239e 1 .135 Q32 Each Member State shall transmit its Partnership contract to the commission within 3 months of the adoption of the common Strategic Framework 47.761e 1 .000 Q34 The EU has exclusive competences in the area of common foreign and security policy 43.904 e 1 .000 a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 108.5. b. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 103.5. c. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 101.5. d. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 99.5. e. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 98.5. However, before implying that the non-native English-speaking EU representatives are indeed developing and using a non-standard form of English for their work-related purposes, the reasons for the high acceptance rates needed to be studied in more detail. After all, previous research of E-E suggests that deviations from SE are merely ‘errors’ or ‘mistakes’ due to L1 interference or a lack of proficiency in the target language rather than features of a ‘nativised’ European English. For this purpose, it was deemed crucial to examine the relationships between all acceptability rates and respondents’ levels of proficiency in English and a possible L1 influence on their acceptability judgements. In addition, the alternatives, i.e., the solutions, provided by the respondents who rejected the sentences were examined for their correctness in some detail, too. The calculation of Pearson’s correlation coefficient between the respondents’ proficiency in English and the acceptance of E-E features revealed that there is a statistically significant relationship between the two variables in five out of eight test sentences, these being (see Annex 5): • Q18: The European commission verifies state aids in the EU (r=0.161; p<0.05, 2-tailed); • Q20: In order to respect the delay for transmission, the European commission can make small changes to the notification (r=0.150; p<0.05, 2-tailed); • Q22: Policy fiche on external action can be found on page 43 (r=0.199; p<0.01, 2-tailed); • Q26: Any absence during core must be approved from the hierarchical superior (r=0.160; p<0.05, 2-tailed); • Q30: commission fixes the standard values for imports from third countries, in respect of the products and periods stipulated in Annex XvI (r=0.180; p<0.05, 2-tailed). The correlation is in all cases positive, which means that the better knowledge of English a respondent has, the more likely he or she is to reject the proposed E-E sentence. What this implies is that the acceptance of most of the deviations (e.g., ‘aids’ used countably to mean financial or economic assistance, ‘fiche’ used instead of a document or the preposition ‘from’ used in passive constructions, where the agent of the action should normally be introduced with the preposition ‘by’) is somewhat related to participants’ knowledge of English. Yet, considering that the participants were proficient speakers of English (i.e., most of them indicated to have c1 or c2 levels according to cEFR, as put forward in Section 4.2), the obtained results seem to indicate that their self-perceived attitudes do not necessarily correspond with their actual knowledge of 142 Tina Balič Attitudes towards Euro-English in a European Union Institution English. According to the author’s first-hand experience, most EU representatives have a B2 level of English, which is also a general requirement to qualify for a job within the EU institutions. In addition, as regards respondents’ L1s, quantitative and qualitative analyses seem to suggest at least some influences on the acceptability judgements, in that those NNSs of English who have a similar structure in their L1s were somewhat more likely to accept than reject the deviations from SE, these being: • ‘aids’ to mean financial or material assistance was accepted by the respondents of the following EU languages: both Finno-Ugrian languages (i.e., Estonian and Hungarian), all Slavic languages except Slovak (i.e., Bulgarian, czech, Polish, Slovenian, and croatian), all Germanic languages except Dutch (i.e., German, Danish, and Swedish), both Baltic languages (i.e., Latvian and Lithuanian), all Romance languages except Spanish (i.e., Portuguese, Italian, Romanian, and French), and by most of the Greek speakers. Several of these languages indeed seem to use ‘aid’ in the plural form for the designated meaning: e.g., ‘ajudas estatais’ or ‘auxílios estatais’ in Portuguese, ‘aiuti di stato’ in Italian, ‘ajutoare de stat’ in Romanian, ‘aides d’état’ in French, ‘κρατικές ενισχύσεις’ in Greek, ‘die Staatshilfen’ in German, ‘valsts atbalsts’ in Latvian, ‘državne pomoči’ in Slovenian or ‘državne pomoći’ in croatian. In contrast, in Dutch ‘state aid’ is typically used only in the singular form (‘staatsteun’), which could explain why none of the Dutch respondents accepted ‘state aid’ in the plural form. • ‘competences’ to denote the abilities/responsibilities of the EU institutions to influence or control was accepted by all of the surveyed NNSs except the czech speakers. In several anaylsed languages the noun ‘competence’ indeed seems to be used with the plural ending -s, these being: all Romance languages, e.g., as in ‘competências da UE’ in Portuguese, ‘les compétences de l’UE’ in French, ‘competenţele UE’ in Romanian, ‘competenze dell’UE’ in Italian or ‘competencias de la UE’ in Spanish. A similar case applies to certain Slavic languages, as illustrated by ‘kompetencje UE’ in Polish or ‘kompetence/pristojnosti EU’ in Slovenian or ‘nadležnosti EU-a’ in croatian. Likewise, in the examined Germanic languages, i.e., German (‘die Zuständigkeiten der EU’) and Dutch (‘bevoegdheden van de EU’), the word for ‘competence’ also takes the plural form for the respective meaning. On the contrary, in the czech language there seem to be two separate words for ‘competence’, i.e., ‘pravomoc’, which is used in the plural form ‘pravomoci’, and ‘kompetence’, which seems to be typically used in the singular only. • ‘fiche’ to mean either a document, paper, sheet or file, etc., in English, whereby L1 influence is, as expected, the strongest within the Romance language group, considering that it is a borrowing from French. Apart from the fact that 90% of the French-speaking respondents accepted it, we found that Romanian speakers, who also use the word ‘fişier’ to refer to a file, dossier or index card, for example, also accepted it by 83.3%. On the contrary, the majority of the Portuguese speakers, who only use the word ‘fisha’ to refer to the rather outdated ‘microfiche’, rejected it by 66.7%. considering that more than half of the surveyed German-speaking respondents also accepted the word ‘fiche’ (52.2%), it is particularly interesting to note that in the German Leo Online Dictionary the respective word is featured as a ‘new’ entry to denote a ‘fact sheet’, i.e., ‘Datenblatt’ in German. • ‘third country’ to denote “a country outside the EU”, since this term is translated and used with the designated meaning in other EU languages as well, e.g., ‘tretje države” in Slovenian, “Drittstaaten” in German or “pays tiers” in French, etc. 143LANGUAGE Last but not least, we analysed the acceptability rates for the surveyed NSs of English, who are greatly outnumbered by its NNSs within the EU institutions.9 The mode for the data collected was again calculated, whereby ‘value 2’ for implying ‘rejection’ was assumed to appear the most frequently. According to Table 3 shown below, out of eight test sentences the surveyed NSs of English rejected three (i.e., Q18, Q20, and Q22), whereas four were accepted as ‘appropriate’ English usage (i.e., Q26, Q28, Q32, and Q34). In one case (i.e., Q30) half of them accepted the test sentence and half of them rejected it. However, the non-parametric chi-Square probability test showed that the rejection of the test sentences as judged by the native English-speaking participants is not statistically significant (see Annex 7). Therefore, in the end, it appears that the NSs only rejected the semantic transfer of ‘delay’ (délai) meaning “deadline”, motivated from a French model. Table 3. Mode for the rejection of deviations from SE by NSs of English. Test sentences Mode % Rejected Q18 The European commission verifies state aids in the European Union. 2 56.3 Q20 In order to respect the delay for transmission, the European commission can make small changes to the notification. 2 73.3 Q22 Policy fiche on external action can be found on page 43. 2 53.3 Q26 Any absence during core must be approved from the hierarchical superior. 1 35.7 Q28 The 2011-2013 multiannual strategy paper was approved in the first semester of 2011. 1 28.6 Q30 commission fixes the standard values for imports from third countries, in respect of the products and periods stipulated in Annex XvI. 1 50.0 Q32 Each Member State shall transmit its Partnership contract to the commission within 3 months of the adoption of the cSF. 1 28.6 Q34 The EU has exclusive competences in the area of common foreign and security policy. 1 28.6 All in all, due to the high acceptability rates among respondents as a whole, the tested features must be regarded as part of EU jargon used primarily for specific purposes in EU institutional settings. While certain EU concepts are examples of borrowing from French into E-E and appear totally unadapted in various MSs’ texts with reference to EU matters (e.g., fiche), some others are translated and used in other EU languages as well (e.g., third country, semester). These results correspond with the findings of previous research into the attitudes towards E-E within continental Europe (i.e., Dalton-Puffer et al. 1997; Murray 2003; Sing 2004; Mollin 2006; Simeone 2011; Lesničar 2011; Forche 2012). 9 For a brief summary on the main grammatical and lexical features from the acceptability test discussed in Sections 3 and 4, please see Annex 6. 144 Tina Balič Attitudes towards Euro-English in a European Union Institution 5 Overall Discussion and Conclusion This study aimed at identifying EU representatives’ attitudes towards E-E formal features based on the data gained from the corpus-based analysis of EU texts examined, including questions of perceptions about the frequency and quality of English usage within a specific EU institution where English is the predominant working language. As anticipated, the findings reveal that the majority of the respondents use English every day both at work and in their free time in order to communicate with people of diverse cultural and linguistic backgrounds and not specifically to identify with NSs or their culture. However, as already indicated throughout this article, it is obvious from the study results that the identified words, phrases and/or their meanings are predominantly of administrative, technical or legal nature. They are used primarily for the purposes of its community members, i.e., the numerous EU representatives, who communicate with other members intra-community in the language of that community, despite of their geographical, cultural, and linguistic backgrounds, about topics of common interest and concern. Although the surveyed NNSs accepted most of the deviations from SE within the acceptability test, the performed correlation analyses revealed that their acceptance appears to be related (at least to some degree) to their proficiency in English and/or L1 interference in certain cases. In other words, those respondents who have a comparatively lower level of competence in English and/or a similar structure in their L1 are somewhat more likely to accept than reject the proposed E-E features. The fact that NSs also accepted most of the deviations from SE furthermore endorses our assumption that English within the EU institutions is learnt not as a language of cultural expression but as a functional variety that is primarily used for transactional rather than interactional purposes, such as for various business agreements or exchange. consequently, it appears that the participants, regardless of ‘nativeness,’ did not seem to be aware of non-standardness in the test sentences, implying that certain English structures which are deviant from SE are by now well-ingrained into the English usage within the EU institutions, but are largely incomprehensible outside their formal settings or mainly familiar to specialists in particular fields that the EU is concerned with. In this regard, the linguistic analysis of EU texts revealed that most of the identified features and their meanings must be classified as part of EU jargon due to their administrative, technical or legal nature rather than features of a specific non- standard usage in EU institutional settings. Accordingly, the examples of institutionally specific use of terminology should not be considered as ‘mistakes,’ but rather as particular ‘performance codes’ as stated by Kachru (1992, 55). Accordingly, the features tested cannot be counted as ‘nativised’ uses of English that would already indicate a separate form of English for EU institutional settings, much less for continental Europe at large, apart from some possibility of grammaticality of ‘aids’ and ‘competences’, since these two features were less fine-grained as compared to other items tested. Instead, the phenomenon at hand relates to the administrative terminology that is used in EU institutions, which draws on semantic shifts of general vocabulary, some loans, and the creation of new English terms. These observations correspond with the findings of previous research towards E-E in continental Europe (Dalton-Puffer et al. 1997; Murray 2003; Sing 2004; Mollin 2006; Simeone 2011; Lesničar 2011; Forche 2012). Taking the above-mentioned results into consideration, we argue that the key trait that distinguishes this type of vocabulary from the rest of the English language are the numerous newly coined words and phrases denoting concepts relating in the first place to the EU as a 145LANGUAGE political, economic, and legal community of its MSs and are untypical outside EU institutional settings. consequently, since these uses essentially mirror the different functions and multiple activities of the EU institutions and its other bodies (in close cooperation with the MSs), they should be regarded as technical or administrative jargon (e.g., transmit, fiche, perspective, planification). Euro-English also includes less obvious jargon mainly influenced by legal or diplomatic terminology (e.g., third country, internal market, member state, mission, etc.). In addition, while certain EU concepts have been borrowed from French into E-E completely unadapted (e.g., fiche), others are translated and used in other EU languages as well (e.g., third country). To conclude, it is clear that the English language is used in EU institutional settings primarily for successful communication among linguistically diverse speakers who do not share a common L1. References Baker, colin. 1992. Attitudes and Language. clevedon: Multilingual Matters. Berns, Margie. 1995. “English in the European Union.” English Today 43: 3–11. Accessed February 23, 2011. doi: 10.1017/S0266078400008348. Biber, Douglas, Susan conrad, and Randi Reppen.1998. Corpus Linguistics: Investigating Language Structure and Use. cambridge: cambridge University Press. BNc. The British National corpus. 2010. Accessed September 14, 2016. https://the.sketchengine.co.uk/. crystal, David. 2003. English as a Global Language. cambridge: cambridge University Press. Dalton-Puffer, christiane, Gunther Kaltenböck, and Ute Smit. 1997. “Learner Attitudes and L2 Pronunciation in Austria.” World Englishes 16 (1): 115–128. doi:10.1111/1467-971X.00052. European commission. 2015. “Translation”. Accessed July, 19. http://ec.europa.eu/dgs/translation/whoweare/ index_en.htm. European court of Auditors. 2013. Misused English Words and Expressions in EU Publications. Accessed January 5, 2014. http://ec.europa.eu/translation/english/guidelines/documents/misused_english_terminology_eu_ publications_en.pdf. Forche, christian R. 2012. “On the Emergence of Euro-English as a Potential European variety of English: Attitudes and Interpretations.” Jezikoslovlje 13 (2): 447–78. Gardner, Robert c, and Wallace E. Lambert. 1972. Attitudes and Motivations in Second Language Learning. Rowley, MA: Newbury House. Graddol, David. 2001. “The future of English as a European language”. The European English Messenger X (2): 47–55. Gries, Stefan Th. 2009. Quantitative Corpus Linguistics with R: A Practical Introduction. New York and London: Routledge. Hornby, Albert Sidney. 2015. Oxford Advanced Learner’s Dictionary. 9th edn. Oxford: Oxford University Press. Hymes, Dell H. 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Philadephia: University of Pennsylvania Press. Jablonkai, Reka. 2009. “In the Light of: A corpus-Based Analysis of Lexical Bundles in Two EU-related Registers.” WoPaLP 3: 1–27. Accessed May 3, 2013. http://langped.elte.hu/WoPaLParticles/W3Jablonkai.pdf. James, Alan. 2000. “English as a European Lingua Franca: current Realities and Existing Dichtomies.” In English in Europe: The Acquisition of a Third Language, edited by Jasone cenoz and Ulrike Jessner, 22–38. clevedon: Multilingual Matters. 146 Tina Balič Attitudes towards Euro-English in a European Union Institution Jenkins, Jennifer, Marko Modiano, and Barbara Seidlhofer. 2001. “Euro-English.” English Today 17 (4): 13– 19. doi:10.1017/S0266078401004023. Kachru, Braj B. 1985. “Standards, codification and Sociolingusitic Realism: The English Language in the Outer circle.” In English in the World: Teaching and Learning the Language and Literatures, edited by Randolf Quirk and Henry G. Widdowson, 11–30. cambridge: cambridge University Press. —, ed. 1992. The Other Tongue. English Across Cultures. 2nd edn. University of Illinois Press: Urbana and chicago. Karp, Jeffrey A., and David Brockington. 2005. “Social Desirability and Response validity: A comparative Analysis of Overreporting voter Turnout in Five countries.” The Journal of Politics 67: 825–840. doi:10.1111/j.1468-2508.2005.00341.x. Kilgarriff, Adam, Pavel Rychly, Pavel Smrz, and David Tugwell. 2014. “The Sketch Engine: Ten Years on.” Lexicography 1–30. Accessed March 15, 2014. http://www.sketchengine.co.uk. Koehn, Philipp. 2005. Europarl: European Parliament Proceedings Parallel Corpus. Accessed 6 March, 2014. http://homepages.inf.ed.ac.uk/pkoehn/publications/europarl-mtsummit05.pdf. Kothari, c. R. 2004. Research Methodology: Methods and Techniques. New Delhi: New Age International. Lesničar, Barbara. 2011. “večjezičnost v Evropi in angleščina kot jezik sporazumevanja v institucijah Evropske unije.” MA Thesis, University of Maribor. Llamzon, Teodoro A. 1969. Standard Filipino English. Manila: Ateneo de Manily University Press. McArthur, Tom. 2003. “World English, Euro-English, Nordic English.” English Today 19 (1): 54–58. Modiano, Marko. 2006. “Euro-Englishes”. In The Handbook of World Englishes, edited by B. B. Kachru, Y. Kachru, and c. L. Nelson, 223–39. Malden: Blackwell Publishing. Mollin, Sandra. 2006. Euro-English: Assessing Variety Status. Tübingen: Gunter Narr. Mouton, Johann, and H. c. Marais. 1996. Basic Concepts in the Methodology of the Social Sciences. Pretoria: van Schaik. Murray, Heather. 2003. “Swiss English Teacher and Euro-English: Attitudes to a Non-Native variety.” Bulletin suisse de linguistique appliquée 77: 147–65. Petzold, Ruth. 2002. “Towards a Pedagogical Model for ELT.” World Englishes 21 (3): 422–26. Schneider, Edgar W. 2007. Postcolonial English: Varieties around the World. cambridge: cambridge University Press. Seidlhofer, Barbara. 2011. Understanding English as a Lingua Franca. Oxford: Oxford University Press. Simeone, veronica. 2011. “The Role of English as a New Potential variety in the Multilingual European Institutions: A Social and Linguistic Study.” Thesis. Università degli Studi di Udine. Accessed May 5, 2015. http://www.tesionline.it. Sing, christine Simone. 2004. “curriculum Ideals and classroom Reality in the Process of Empowerment through Euro-English.” LAUD Paper 625: 1–22. Essen: Universität Duisburg. Strevens, Peter. 1992. “English as an International Language: Directions in the 1990s.” In The Other Tongue: English across Cultures, edited by Braj B. Kachru, 27–47. Urbana and chicago: University of Illinois Press. Trebits, Anna. 2008 “English Lexis in the Documents of the European Union: A corpus-based Exploratory Study.” WoPaLP 2: 38–54. Accessed May 9, 2013. http://langped.elte.hu/WoPaLParticles/W2Trebits.pdf. —. 2009. “conjunctive cohesion in English Language Documents: A corpus-based Analysis and its Implications.” English for Specific Purposes 28 (3): 199–210. doi:10.1016/j.esp.2009.04.004. 147LANGUAGE vOIcE. vienna-Oxford International corpus of English. 2005–2013. Accesed July 5, 2013. http://voice. univie.ac.at. Yule, George. 1996. The Study of Language. New York: cambridge University Press. Appendices Annex 1. Specific terms derived at from the prefix Euro- according to their frequencies of occurrence in the EUROPARL7 and the BNc. TARGET WORDS EUROPARL7 BNC European Union 101,176 311 Euro area 2,076 0 Euro zone 1,052 0 Eurobarometer 269 2 Euro notes 143 0 Euroscepticism 95 0 Euroland 73 0 Annex 2. Specific terms describing EU concepts according to their frequencies of occurrence in the EUROPARL7 and the BNc. TARGET WORDS EUROPARL7 BNC Member State 16,165 2,287 Internal market 10,611 305 Subsidiarity 4,374 148 candidate country 3,822 1 Neighbourhood policy 1,519 0 comitology 1,131 0 Proportionality 763 140 Acquis communautaire 611 8 community method 526 3 conditionality 389 13 Flexicurity 342 0 148 Tina Balič Attitudes towards Euro-English in a European Union Institution Additionality 236 44 communitisation 104 0 Delocalisation 40 3 Rendez-vous clause 6 0 Annex 3. Frequencies of four-word lexical bundles in the EUROPARL7 compared with the BNc. TOP 10 LEXICAL BUNDLES IN EUROPARL7 FREQUENCIES OF OCCURRENCES (PER MILLION) EUROPARL7 BNC I would like to 679.3 31.9 of the European Union 376.4 0.1 on behalf of the 346.8 12.0 I should like to 325.7 2.8 of the committee on 217.4 0.3 in the European Union 188.5 0.0 of the European Parliament 180.0 1.2 on the basis of 149.5 26.7 at the same time 226.2 61.7 of the Member States 122.9 1.1 Annex 4. Number of participants per EU Member State. EU MS Frequency Austria 11 Belgium 10 Bulgaria 9 cyprus 5 czech Rep. 11 Denmark 6 Estonia 6 Finland 5 France 15 Germany 24 Greece 5 Hungary 7 149LANGUAGE Ireland 4 Italy 14 Latvia 5 Lithuania 6 Luxembourg 3 Malta 15 Netherlands 9 Poland 8 Portugal 12 Romania 9 Slovakia 8 Slovenia 42 Spain 10 Sweden 11 UK 15 Total 285 Annex 5. Pearson’s correlation between NNSs’ English language proficiency and acceptability of test sentences.     Q4 English language proficiency level  Q4 Your English language proficiency level  Pearson correlation 1   Sig. (2-tailed)     N 249 Q18 The European commission verifies state aids in the European Union Pearson correlation .161*   Sig. (2-tailed) .018   N 217 Q20 In order to respect the delay for transmission, the European commission can make small changes to the notification Pearson correlation .150*   Sig. (2-tailed) .032   N 207 Q22 Policy fiche on external action can be found on page 43 Pearson correlation .199** 150 Tina Balič Attitudes towards Euro-English in a European Union Institution   Sig. (2-tailed) .004   N 203 Q26 Any absence during core must be approved from the hierarchical superior Pearson correlation .160*   Sig. (2-tailed) .024   N 199 Q28 The 2011-2013 multiannual strategy paper was approved in the first semester of 2011 Pearson correlation -.104   Sig. (2-tailed) .145   N 197 Q30 commission fixes the standard values for imports from third countries, in respect of the products and periods stipulated in Annex XvI Pearson correlation .180*   Sig. (2-tailed) .012   N 197 Q32 Each Member State shall transmit its Partnership contract to the commission within 3 months of the adoption of the cSF Pearson correlation .035   Sig. (2-tailed) .621   N 197 Q34 The EU has exclusive competences in the area of common foreign and security policy Pearson correlation .128   Sig. (2-tailed) .073   N 197 * correlation is significant at the 0.05 level (2-tailed). ** correlation is significant at the 0.01 level (2-tailed). 151LANGUAGE N N S ac ce pt Pr ofi ci en cy co rr el at io n L1 - in te rf . N N S ac ce pt N on -S E fe at ur e C om m en t ai ds Y (6 6. 8% ) Y (r =0 .1 61 ,p <0 .0 5) Y Y N /Y Ac ce pt ed b y N N Ss (o ve r: 65 % ). D es pi te p ro f. co rr. (a nd L 1- in te rf. ), th e le ve l o f s ig ni fic an ce is lo w er if co m pa re d to E U ja rg on , e .g ., ‘th ird c ou nt ry ’& ‘fi ch e’. D es pi te n on -S E us ag e, N Ss a cc ep te d. N ot a s fin e- gr ai ne d as o th er te ste d fe at ur es . co m pe te nc es Y (7 3. 6% ) N Y Y N /Y Ac ce pt ed b y N N Ss (o ve r: 70 % ) a nd n o sig ni fic an t pr of . c or r. D es pi te n on -S E us ag e, N Ss a cc ep te d. N ot a s fi ne -g ra in ed a s o th er te ste d fe at ur es . th ir d co un tr y Y (5 5. 3% ) Y (r =0 .1 80 , p <0 .0 5) Y Y N EU ja rg on (l eg al ) d es pi te p ro f. co rr. & L 1 in te rf. N N Ss & N Ss a cc ep te d it. W el l-i ng ra in ed in E U us ag e. fic he Y (6 7% ) Y (r =0 .1 99 , p <0 .0 1) Y Y N Bo rr ow in g fro m F re nc h. E U ja rg on (a dm in ist ra tiv e) d es pi te p ro f. co rr. & L 1 in te rf. N N Ss & N Ss a cc ep te d it. W el l-i ng ra in ed in E U us ag e. se m es te r Y (8 2. 2% ) N N Y N EU ja rg on (a dm in ist ra tiv e) d es pi te n o pr of . c or r. & n o L1 -in te rf. N N Ss & N Ss a cc ep te d it. W el l- in gr ai ne d in E U u sa ge . hi er ar ch ic al / fr om Y (7 2. 9% ) Y (r =0 .1 60 , p <0 .0 5) N Y N EU ja rg on (a dm in ist ra tiv e) d es pi te p ro f. co rr. N N Ss & N Ss a cc ep te d it. W el l-i ng ra in ed in E U us ag e. W ro ng u se o f p re po sit io n, e ve n by N Ss . Tr an sm it / of Y (7 4. 6% ) N N Y N EU ja rg on (a dm in ist ra tiv e or te ch ni ca l) de sp ite n o pr of . c or r. & n o L1 -in te rf. N N Ss & N Ss a cc ep te d it. W el l-i ng ra in ed in E U u sa ge . W ro ng p re po sit io n. D el ay (d él ai ) N (4 9. 3% ) Y (r =0 .1 50 , p <0 .0 5) N N N Fr en ch w or d to m ea n ‘d ea dl in e’. Th e on ly fe at ur e re je ct ed b y N N Ss & N Ss o f E ng lis h. Annex 6. Summary of E-E features from the acceptability test (see Sections 3 & 4 for a detailed explanation). 152 Tina Balič Attitudes towards Euro-English in a European Union Institution Annex 7. chi-Square Probability Test for NSs of English. Chi-Square df Asymp. Sig. Q18 The European commission verifies state aids in the EU .250a 1 .617 Q20 In order to respect the delay for transmission. the commission can make small changes to the notification 3.267b 1 .071 Q22 Policy fiche on external action can be found on page 43 .067b 1 .796 Q26 Any absence during core must be approved from the hierarchical superior 1.143c 1 .285 Q28 The 2011–2013 multiannual strategy paper was approved in the first semester of 2011 2.571c 1 .109 Q30 commission fixes the standard values for imports from third countries. in respect of the products and periods stipulated in Annex XvI .000c 1 1.000 Q32 Each Member State shall transmit its Partnership contract to the commission within 3 months of the adoption of the common Strategic Framework 2.571c 1 .109 Q34 The EU has exclusive competences in the area of common foreign and security policy 2.571c 1 .109 a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 8.0. b. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 7.5. c. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 7.0.