“The goal is to enable students to communicate”: Communicative competence and target varieties in TEFL practices in Sweden and Germany

Abstract This paper compares teaching practices of English as a foreign language in Sweden and Germany based on a questionnaire targeted at investigating teachers’ experiences and views on language use, giving a snapshot of teaching practices in classrooms in the two countries. In this regard, the focus is on the following questions: Which target varieties are used in TEFL, and to what extent? What is the status of different target varieties among teachers of English in the two countries? The results show that the use of target varieties is still in play in the TEFL classroom, despite the recent move towards communicative competence as the goal. The main target varieties used are AmE, BrE, a mixture of the two, and some neutral variety; with a slight preference among German teachers for BrE and among Swedish for the neutral variety alongside AmE or BrE. Further, teachers are in conflict between two ideals: learned (where they teach an English belonging to the native speakers) and didactic (where the goal is communicative fluency). In application, this conflict should be addressed by teacher training programmes in order to make (future) teachers aware of it and provide possible ways to cope with it.


Introduction
There is a long tradition of teaching English in European countries, and similarities across teaching practices of English as a foreign language (EFL) are observable. Both Sweden and Germany, the countries under scrutiny, have traditionally been viewed as EFL countries, where English is acquired for practical reasons but lacking official status (Moag 1982). Teachers are given different directives to work with, sometimes including reference to a certain variety of English, depending on the country and region they work in.
The unequivocal distinction of English as foreign (EFL), second (ESL) and first (ENL) language varieties has been criticized, especially in the era of globalization, as being too rigid and not accounting for the heterogeneity of ever more fluid speech communities (e. g. Bruthiaux 2003;Schneider 2007;Buschfeld and Kautzsch 2016). Further, it has been shown that some varieties are moving between categories, such as Cyprus English from ESL to EFL (Buschfeld 2013) or English in the Netherlands from EFL to ESL (Edwards 2016). Similarly, it has been suggested that due to the increased societal use of English (for example in the media), speakers in Scandinavia in particular exhibit proficiencies comparable to those of ESL speakers (cf. Sundqvist and Sylvén 2014: 5). In Sweden, English proficiency is high among almost 90 % of the population (Simensen 2010), in contrast to Germany where only 60 % are reported to speak English (Mollin 2006: 175;Kautzsch 2014: 203). This difference, as well as differences in curricula that have a major impact on the transmission of English to young speakers, makes a comparison between Teaching EFL (TEFL) strategies and practices in these two countries particularly interesting.
Guidelines in Sweden and Germany have in recent years changed to target communicative competence (cf. Sekretariat der Ständigen Konferenz der Kultusminister der Länder in der Bundesrepublik Deutschland 2014 for Germany; Swedish Government 2010 a, 2010 b for Sweden), which seems at first glance very much opposed to the acquisition of a certain target variety associated with native speakers. However, using target varieties as learning goals for TEFL still seems prevalent in both countries. One objective of this study is to find out if, to what extent, how and why target varieties are employed in TEFL in Germany and Sweden. Comparing English language practices in Sweden and Germany by asking teachers about their experiences and views on their teaching gives a snapshot of actual practices in the English language classroom in the two countries.
Altogether, we aim to investigate teacher experiences, views and practices concerning communicative competence and target varieties in the TEFL classroom in Germany and Sweden, given that a theory-practice divide has been reported with respect to orientation towards communicative competence and the inclusion of different varieties of English in the TEFL classroom (Galloway 2017

English in Europe
Across Europe an array of varieties of English with varied forms and functions can be found. English serves as an official language in Malta, Ireland and de facto in the UK, and is relatively prominent in the linguistic landscape of Cyprus. In the rest of Europe, it is usually acquired as a foreign language in school (Modiano 2003) but is very much present in the everyday lives of many people, and has become a "language of socialisation" (Jenkins 2003: 38) and a lingua franca for international communication (Mollin 2006).
The status of several Englishes across Europe as second or foreign language varieties has been critically discussed recently (cf. Bongartz and Buschfeld 2011;Buschfeld 2013;Edwards 2016), and Modiano (2017b: 314) argues that English in the EU can probably no longer be viewed as EFL, but rather as a second language or English within a "developing outer circle context".
With respect to target varieties in continental Europe, Simon (2005) reports that 96 % of the Flanders students of English in her study stated that it was important for them to acquire a native-like accent, since most partakers wanted to become teachers (2005: 6) and a native-like accent seemingly felt important in this profession. 96 % of the participants aimed for RP while only 4 % aimed for GA, a reflection of RP being the target accent in Flanders' schools. Interestingly, the lecturers who took part in the study mentioned intelligibility for native speakers and other interlocutors, consistency in the target accent, but not the degree of nativeness as the most important criteria for their students' pronunciation (Simon 2005: 5). Similar to the Flemish students, Carrie's (2017) Spanish participants showed clear preferences for native varieties of English. In a verbal guise experiment, pupils perceived RP as standard and prestigious as opposed to GA, which was rated as a non-standard variety. This again emphasises the strong position of RP as a target among European speakers of English. Similarly, Henderson et al. (2012: 20) showed in their survey that RP was generally the target preferred by teachers in many European countries, while GA was slightly preferred by pupils. Aiello (2018: 119) discusses how her Italian participants position the Englishes they have access to within or somewhere in between either native or non-native varieties, and within or between British English (BrE) or American English (AmE). 1 These adolescent participants were aware of the existence of multiple Englishes and they drew resources from their feature pools in different ways depending on their audience. Just as Aiello's participants challenged the assumptions of L2 norms of English (2018: 125), Rindal and Piercy's (2013: 220) research, using language data from adolescent speakers, suggests that the target model used by participants may not be that of a native speaker, but rather that of a fluent bilingual speaker.
Finally, a focus group study with teachers from around the world (Johnstone Young and Walsh 2010), shows that a majority did not want to use an English as a Lingua Franca (ELF) model for teaching English as it was seen as "not standard", "broken" or "simplified". (Many) native varieties of English carry prestige, and while many participants agreed with this citation by one participant: "Learners need a REAL target, not a made-up one" (2010: 133), they continuously highlighted the importance of successful communication as the goal of teaching English. Thus, simultaneous orientation towards native accents and communicative competence seem to be important goals in the EFL classroom. Berns (1995: 9) assigns the Englishes of Europe (including Germany) to an "overlap area" between Kachru's (1985) Outer and Expanding Circles, and Hilgendorf (2007) classifies it as an additional rather than a foreign language. While not an official language, Erling (2007: 120) reports that nowadays most university students are proficient in English, and this proficiency seems to be a characteristic criterion of "Germanness". Children are taught a foreign language from year 1 or 3 in primary school, depending on the federal state they live in (Grau 2009: 161), and English is usually the first foreign language (Hilgendorf 2005). In higher education, the importance of English is also steadily growing, as it is increasingly being used as the medium of instruction (Erling and Hilgendorf 2006). Henderson et al. (2012: 21) show in their survey that RP is preferred over GA for both reception (91.19 % vs. 80.08 %) and production (91.19 % vs. 67.82 %).

English in Germany
Grau uses the terms "English from above" and "English from below" (as used by Preisler 1999in Grau 2009 to distinguish English taught in the classroom and English "playfully taken up by German youths and integrated into their own language" (Grau 2009: 163). German adolescents use English in a number of situations as it is ubiquitous in the media and an integral part of (European) youth culture (James 2000). Indeed, Erling (2007: 117) reports that "the most prevalent domain in which students are likely to read and write English is not at university [...] but [on] the internet". This is a consequence of the internet having overtaken TV as the most frequently-used digital medium in the 14-49 age bracket in Germany (Statista 2018). Germans do not come into contact with English much when watching TV as, along with Italy and Spain, Germany is the leading dubbing nation in Europe (Goethe Institut 2018) and English-language TV programmes are usually only available with English-language audio on Netflix or via other streaming services (Fischl 2017). However, it is unclear whether and to what extent these original English versions are actually watched, since Germans are the European people that disfavour subtitled films the most (Eurobarometer 2006: 59).
While the importance of English as a lingua franca is also emphasised by Gnutzmann, Jakis and Rabe (2015: 181) in their study on attitudes towards multi-lingualism and English as a lingua franca among German university students, the participants in their study still felt the "ownership" of the English language to be with its native speakers. Thus, their participants clearly preferred native speaker norms in the form of BrE and AmE over a suggested new variety "Euro-English" (Gnutzmann, Jakis and Rabe 2015: 178). Euro-English was felt to be a bad imitation of "real" English and a superfluous artificial construct.

English in Sweden
In the 1990 s, fears were expressed in the media and by linguists that a continually increasing use of English would have negative effects on the Swedish language (see Boyd and Huss 2001;Bolton and Meierkord 2013). As a response, the government issued a report in 2002, giving suggestions for how the two languages can or should be used. For higher education, Swedish is maintained alongside English in order to ensure that speakers continue to use both, and that Swedish continues to develop in the domains of teaching and research (Swedish Government 2002a).
English became the primary foreign language in schools in 1962 (Sundqvist and Sylvén 2014: 4), and since 2011 children start learning it at age 7, in their first year of compulsory school (Swedish Government 2010a). It is the only compulsory language subject apart from Swedish, and passing English is a requirement for enrolment in upper secondary school (Sundqvist and Sylvén 2014: 5). The English syllabi in compulsory (years 1-9) and upper secondary (years 10-12) schools do not specify a variety to be taught, but rather focus on the pupil being able to communicate with a range of interlocutors in different social situations (Swedish 2010b). Prior to 1994, the curricula stated that BrE should be the taught variety (Modiano 2009: 66).
Swedish year 4 pupils use and are exposed to English outside of school more than during school hours (Sundqvist and Sylvén 2014: 17), through activities such as gaming, television and social networking. TV channels subtitle rather than dub imported television shows (apart from those aimed at very small children), and Swedes are thus exposed to English through their media consumption (Swedish Government 2002b). Some researchers, such as Hyltenstam (1999: 212) and Melander (2001: 13), argue that English in Sweden should be seen as a second rather than foreign language, due to its widespread use.
While 32.4 % of programmes aired on Sweden's 11 television channels are produced in Sweden, 16.3 % are produced in the UK and 37.8 % in the USA (Asp 2016: 24). In addition, approximately half of the population report having access to and using on-demand streaming services; in the age group 9-19 that number is 81 % (The Swedish Press and Broadcasting Authority 2016 a; 2016b). The majority of the English that Swedish pupils hear in their spare time is closer to AmE, while their English teachers likely use something closer to BrE (Mobärg 2002: 129).
In Ribbås (2016), 54 upper secondary school pupils self-rated their proficiency for English reading, writing, speaking and understanding. Participants tended to self-report their ratings as 4 or 5 on a scale from 1 (very bad) to 5 (very good); and 12 of them indicated that their English proficiency is just as high, or almost as high as their Swedish. Mobärg (2002) reports that Swedish pupils (years 7-9) prefer using BrE when speaking English, but that a large minority of tokens are produced using AmE features. There is some indication in Mobärg's work that pupils tend towards BrE in more formal contexts; while AmE is associated with more informal speaking styles.
In Söderlund and Modiano's (2002: 162) survey of 15 teachers, 57.1 % reported using BrE, and 28.5 % reported Mid-Atlantic English (MAE) when teaching; however, privately 71.4 % report using BrE, 14.2 % reported using MAE, and 14.2 % "modified RP". Westergren Axelsson (2002: 142) further concludes that English teachers in 2000 seemed more open to pupils learning other varieties of English than they were in 1992.

Summary of English in Germany and Sweden
Both countries are Western European countries where speakers start learning English at an early age (before or by age 10). The countries also have in common the discussions on whether English should be considered a foreign or a second language. English proficiencies are rather high, somewhat more so in Sweden than in Germany, and English is an important language in higher education as well as in other contexts.
One of the main cultural differences in relation to the use of and exposure to English is through media and entertainment, whereby the German audience will experience most of the content produced in the English-speaking world dubbed into German; whereas the Swedish audience will be used to hearing the original English, and seeing subtitles in Swedish.

Method
3.1 Questionnaire design and data collection procedure TEFL strategies and practices among German and Swedish teachers were assessed using an online questionnaire 2 . Part one of the questionnaire was aimed at determining possible institutional and informal influences gained before starting the profession (for example through schooling, including their teacher education), which may influence the teachers' reported language use and language teaching practices. Teacher education and teaching experience questions were thus included in order to assess the amount of institutional training the participants had received, which might in turn influence the impact of official guidelines on their language teaching. Questions on the teachers' language learning history were included in order to evaluate the influence of non-institutional factors, such as stays abroad and informal language acquisition, on the teachers' language use in-and outside of the classroom. This has been shown to be influential, especially in terms of fluency (Trenchs-Parera 2009).
Part two of the questionnaire was intended to assess the current teaching practices of the participants in terms of target varieties. Thus, the language teaching practice questions enquired about official instructions concerning target varieties specifically, but also investigated different aspects of language varieties and their importance in the teachers' language classrooms. The teachers' private language use was assessed here, in order to find out whether and to what extent it plays a role for their language teaching.
The questionnaire did not include distractors, deemed unnecessary in a research design employing a direct method, i. e. asking for language attitudes and ideologies directly instead of employing a guised technique. Further, the participants were informed of the purpose of the study beforehand. The questionnaire was, prior to use, validated by colleagues and subsequently in a pilot phase (cf. also Artino et al. 2014). The pre-test contained 38 questions and was performed with 16 participants from Sweden and Germany, using the friend-of-a-friend approach (cf. Milroy 1980;Milroy and Gordon 2003).
The final version of the questionnaire contained 35 questions and several sub-questions. It was distributed via a link shared on social media sites (Facebook, Twitter, Instagram) and via e-mail to teachers known to the authors through collaborations with secondary schools, as well as to contacts working in connection to school environments. This method of distribution is similar to the friend-of-a-friend approach, with the possibility of recruiting a larger and more diverse participant pool than a traditional hard-copy questionnaire permits (Sue and Ritter 2012), but without control over the representativeness of the sample. The link was left open for 2 months and all participants who completed the questionnaire could participate in a lottery to win amazon.com gift vouchers. As shown in Table 1, the 80-participant sample was relatively balanced with regard to nationality (34 German, 46 Swedish participants) and the participants were from different regions of Germany and Sweden. There are more female than male participants, especially in the German data set; this reflects the population of teachers. It should be noted that the Swedish participants had more teaching experience than the German ones, and that the Swedish teachers started learning English earlier than the Germans. The latter is attributable to the different educational policies with regard to English (see Section 2).

Private language use
Participants were asked to self-assess their English speaking competence on a scale from 1 (disfluent) to 8 (native-like). 43 Swedish participants rated them-selves between 5 and 8, with an average of 7.23, and a mode of 7.30. German participants rated themselves between 4 and 8, with an average of 7.0, and a mode of 8.0 Responses to a combination of questions are needed to give a full picture of how the teachers acquired English. Participants were asked to estimate what percentage of the spoken and written English they had acquired in a formal education setting (cf. Grau 2009 "English from above"), as opposed to informal situations (cf. Grau 2009 "English from below"). Figure 1 shows written language self-ratings on the x-axis and spoken language ratings on the y-axis, with trend lines showing the correlation between the two modalities. Both participant groups report that their written language skills are more influenced by formal education than their spoken skills; unpaired t-tests show no statistically significant difference in responses between the groups, meaning that they respond in similar ways. When testing the correlation between estimations of spoken and written English learned in a formal setting however, a Pearson's r test shows that while there is only a weak correlation for the Swedish group (r=0.2614, p=0.0793), the correlation for the German group is moderate (r=0.5622, p=0.000665).
The German outlier in the bottom left corner rated their acquisition of English in formal settings fairly low. At the same time, this participant rates their speaking competence as very high. Looking at the other answers reveals that this speaker had lived in English-speaking countries for some time. The low formal acquisition ratings in combination with the high speaking competence rating most likely means that the speaker thinks they acquired most of their English in non-formal contexts when living abroad.
Responses to the multiple-choice question "Where or how did you learn English?" are shown in Figure 2. Multiple responses were possible, and are shown in percentages by country. The largest differences in Figure 2 are the categories "in an English-speaking country" and "in the classroom". The latter corresponds to the results of the Pearson's r test on formal/informal settings for the acquisition of English (Figure 1), whereas the former can be related to the fact that 27 Swedish (58.7 %) and 25 German (73.5 %) participants report having spent more than two months continuously in an English-speaking country, mainly in the UK (4 Swedes, 11 Germans) and the US (10 Swedes, 5 Germans). The number of responses for audio-visual media is slightly higher for the Swedish participants, however the difference between the two groups is fairly small. Three Swedish participants responded that they had learned English from an English-speaking family member while none of the German participants chose this option.
There is some variation shown in the responses to the question "Do you use a specific variety of English when speaking, privately?". Other 3 (6.5 %) 3 (6.5 %) No answer 0 (0.0 %) 2 (4.3 %) While the majority report using BrE or AmE, 9 Swedish and 1 German participant report not using a specific variety. One Swedish participant reports using MAE, and one of the 17 Swedish respondents favouring AmE in fact report "According to me I have an American dialect, according to native-speaking friends I have a 'Scandinavian academic dialect'." (Swe27m).
Teachers were asked "Which aspects of your own pronunciation of English do you consider to be the most important?" (Table 3). Sticking to one accent/consistency 3 3 Sounding native/authentic/natural/credible/fluent 2 6 Segmental features/articulation/speaking clearly 10 2 Participants appear to regard intonation as most important in terms of their own pronunciation. Articulation and segmental features are mentioned by 10 Swedes but only by two of the German participants, while 6 Germans and 2 Swedes mention authenticity, fluency, credibility or sounding natural or native. They were able to give as few or as many answers as they wanted, and as such no further analysis as to the frequencies is possible here.

Language use in the classroom
In terms of their teaching practices, 89.1 % of the Swedish and 73.5 % of the German participants reported that they had not been given official instructions to use a specific variety of English in their teaching. Three Swedish participants highlighted that the curriculum and the Department of Education require pupils to learn about different varieties of English. Two of the Swedish participants appear to limit the varieties available to them to two, BrE and AmE: (1) "No, but I've heard that only British or American standardized varieties are accepted." Two German participants state that BrE is the target variety, and one that it is recommended to use BrE, while two report that school books mainly use BrE. One reports that there are no longer instructions about the use of a particular variety, implying that there used to be at one point. In general, AmE and BrE are the varieties most often named in response to this question by the German participants (7 times).
Investigating the teachers' own reported language use in the classroom, Table 4 shows responses to the questions "Which variety of English do you use when teaching speaking skills/writing skills?".  The answers shown in Table 4 are comparable to results from Henderson et al. (2012), whereby teachers in Germany display a preference for RP, with GA rated slightly lower. While BrE and AmE are the most commonly-used varieties when teaching both speaking and writing skills, there is some variation: teachers are not always consistent across their teaching of different modalities; a fair few teachers from both countries use a mixture of BrE and AmE for both modalities; and quite a large proportion of the Swedish teachers report using a neutral or non-specific variety of English in teaching, while none of the Germans mention this. This is an issue that is particularly intriguing, as in private as well as in school, teachers state that they are able to switch between different varieties of English with ease. Following up on the use of different varieties in teaching, participants were asked if they use native-speaker audio files in class, and if so, which varieties of English the speakers in the files use. Responses to this question showed some variation between teachers' practices, with most stating that they used BrE (Sweden 20, Germany 19) or AmE (Sweden 20, Germany 17) speakers, but with some regional variation from these countries, including "not only standard varieties", showing awareness of regional and/or social variation or not only interpreting "variety" as something connected to a specific country (cf. Milroy 2001). Among the Swedish participants, one included "immigrant backgrounds" and two mentioned "World Englishes", and among the German participants there is mention of "non-native", "immigrants to English-speaking countries" and French speakers, indicating that representations of non-native varieties of English feature in their teaching.
Teachers were further asked a number of questions concerning their pupils' language skills and use. Just over half of the Swedish teachers (n=23) agreed that pupils are expected to choose and be consistent with any one variety of English for their own use, while 20 participants disagree. Some said it is encouraged but they do not strictly enforce it, some said they require consistency in writing but not in speaking, and one said they require it within a single text but not across different texts. 15 German teachers agreed that pupils should choose one variety and be consistent in its use while six participants disagreed, two said they would require consistency from older pupils but not at lower levels, one said pupils should be consistent but are not, and one that the teachers are not consistent either. There are indications that teachers view BrE and AmE as the only two varieties available to choose from, or at least, these are the only two specifically mentioned. Teachers were also asked to give an estimate of which varieties are the most popular among pupils (see Table 5). Again, these results are comparable to those in Henderson et al. (2012), in which German teachers report pupils' preferences to be slightly higher for GA than RP. One of the German participants stated that the pupils hardly think about which variety they use, but that they are influenced by exposure (Ger18f). Three Swedish participants mentioned contradicting patterns in variety usage among the pupils: older pupils use BrE, and younger ones AmE (Swe4o); older pupils use AmE (because of exposure outside the classroom), and younger ones BrE (because they are taught it at earlier levels) (Swe12f); and "When speaking, it is definitely American. The same pupil, however, might use British when writing." (Swe7f). Furthermore, three Swedish participants stated that AmE is more popular than BrE, in line with the results (cf. Table 5).
To investigate which aspects of their teaching the teachers think are most important, they were asked to rank four skills by importance, where those ranked as #1 are "most important" and those ranked as #4 are "least important". Figure 3 shows the responses for each skill: Communicative skills, vocabulary, grammar, and pronunciation. Figure 3 shows some minor differences between the teachers in the two countries: while the German participants ranked grammar and pronunciation as slightly more important than the Swedish participants did (grammar: on average 1.91 for Germany and 2.13 for Sweden; pronunciation: 1.82 for Germany and 2.13 for Sweden), the rankings for communicative skills and vocabulary indicated greater importance for all participants (communicative skills: average 1.24 for Germany and 1.22 for Sweden; vocabulary: 1.42 for Germany and 1.39 for Sweden).
The mode values for all aspects were the same across both groups of participants: 1 for communicative skills and vocabulary, and 2 for grammar and pronunciation. Using a Mann-Whitney U test, the differences between responses by individuals in the two participant groups are shown not to be statistically significant for any of the four skills investigated here. Participants were further asked which approaches/methods they use for teaching these four skills, and responses were similar across the groups. Starting with pronunciation, most responses concerned letting pupils listen to native speakers (and sometimes to different accents), and many teachers used repetition exercises in class. A few reported using feedback/correcting pupils (sometimes specified as given individually), and especially the Swedish teachers reported that pupils are instructed to speak English as much as possible. Five Swedish and seven German teachers reported using the International Phonetic Alphabet or articulatory/phonetic instruction when teaching pupils about pronunciation.
With respect to grammar, most responses related to the use of worksheets or grammar textbooks containing exercises; for Sweden these seemed to be exclusively online resources. Many teachers mentioned using and practising grammatical constructions, and especially Swedish teachers specifically mentioned writing. Input in different forms seemed to be equally important to the two groups, with examples of listening, reading, films, and lectures being mentioned. A number of German teachers specified using inductive vs. deductive methods for the teaching of grammar.
For vocabulary, Swedish teachers seemed more focused on textual and audio input than German teachers, who reported using lexical chunks or different types of networking techniques, as well as games. Both groups reported using word lists; Germans report using exams/quizzes and flashcards more than the Swedes, who used online resources more.
For communicative skills, most respondents reported focusing on group/pair work and/or discussions in the classroom. Many German teachers especially reported using drama, role play or the acting out of real-life scenarios. A number of teachers in both groups reported using games, exercises and writing, as well as speeches and presentation type exercises in order to practice communicative skills with pupils.
Teachers were asked "Which aspects of their [the pupils'] pronunciation do you consider to be most difficult for students to learn and why? E.g. consonants, vowels, rhythm, intonation..." (see Table 6). Intonation 22 "Struggle with intonation" "Tend to copy American TV"

-
Rhythm --10 -Further responses include five Swedish teachers stating that pupils do not have many problems regarding pronunciation in English, and five participants specifically state that pupils with other L1 backgrounds struggle more. This is echoed in the German results, where one participant writes: "Depends on the learner's native language. The most difficult things to pronounce are always those outside their repertoire of native phonemes" (Ger32m). This question was in an openanswer format, and occurrences of mentions have been counted; this means that table 6 only gives an overview. The participants were then asked in what ways they work with pupils on the aspects of pronunciation they deem are the most difficult (Table 7). Exposure to native/model speakers 10 2 Pupils record their speech 3 2 Anatomical/articulatory info 2 4 Feedback/comments on pronunciation 4 3 Most participants report using listen-and-repeat type exercises with their pupils, or letting pupils practice in different ways, such as by self-recordings. While 10 Swedish participants report using exposure to native or model speakers as a way of allowing pupils to practice, this is rarely practiced by German teachers. Two German participants further report that there is very little time to work on these things, and that it is the least important aspect of their teaching. In addition to pronunciation, participants were asked whether they introduce pupils to grammatical structures from varieties other than BrE or AmE, and whether these would be allowed as correct variants on a test. The question seemed to trigger some reflection on what variation is and its role in the classroom, as well as what is accepted from an L2 speaker of English in a Swedish or German learning environment.
(2) "Probably not, but that's more because I would not recognize them (due to limits in my own experience of all the different varieties of English). Also, I feel that as second language speakers it's probably more relevant to learn a more 'general' English. It would be odd for a Swedish student to learn Jamaican Patoisunless of course that student planned to move there or something." (Swe25m) (3) to my mind students should be consistent in the variant they use, which is usually either British or American (it'd be different if I knew the student was part Irish or Nigerian or whatever and I could trust that their English is a native speaker's English)" Here, some teachers also reported that even if the structure in question was unknown to the teacher, they would allow it if the pupil could show it is correct. They thereby show a willingness to bring other, more non-standard, forms into the classroom. This willingness is further supported by the similar responses given to questions surrounding the teaching of vocabulary; on the question whether they teach both BrE and AmE variants an overwhelming majority respond affirmatively. Only two Swedish teachers said "no" or "not applicable", without expanding on their response.
Concerning a European English variety (Euro-English) to be used in their own teaching, neither Swedes nor Germans seem in favour of the idea (participants responding "no": Sweden 25, Germany 16). The main concern among teachers seems to be that this is a variety they do not know and would have to learn prior to being able to teach itthis in fact comes up both among those who are opposed to the idea, and those who are not. The arguments against using Euro-English are similar for the two countries: participants say it is artificial or constructed, and that authentic or native anglophone targets should be used instead. Some of the teachers' arguments are listed below: (4) "That would be another example of things being intentionally dumbing things down [sic] to accommodate incompetence rather than using proper means of making the incompetent competent, such as proper training and demanding effort" (Ger32m) (5) "English should be as close to native speaking varieties as possible and selling a few new structures, specific vocabulary or proverbs as a new variety will simply water down the standard of language teaching and acquisition" (Ger17m) (6) "Politicized 'management' of language has largely failed for the last 3000 years, and inventing/teaching any such variety is for people living in a decontextualized realm of short-term historical memory." (Swe26m) (7) "No. (This is a largely outdated concept. Global English took over long ago in the real world.)" (Swe22f) At the end of the questionnaire, participants were given space to comment on the questionnaire or any aspect of it in case they wanted to provide feedback. This following comment illustrates a recurrent theme in the results viz., that participants are apparently aware of a change in teaching practices and agree that variety-centered approaches are no longer up to date: (8) "This questionnaire seems to center on varieties of English which seems odd to me. This is (by far) the least important aspect of teaching English. The goal is to enable students to communicate. Even a total confusion of all varieties of English would be acceptable as long as it gets the job (successful communication) done. I am aware that this has not always been the case in German educational doctrine but there has been a push in that direction in recent years and for once, that's a change that I agree with." (Ger26m)

Discussion
In this section, we summarise the findings and discuss them in terms of similarities and differences between the two countries. Despite the focus (by education departments and teachers alike) on the communicative approach, teaching of specific native varieties is still shown to be widespread in our sample. Given that the acquaintance with different varieties of English might in fact further commu-nicative competence in real life situations with speakers of different varieties of English (e. g. Bieswanger 2008), teachers' reports of using recordings from different varieties of English, as well as the willingness of the sampled teachers to consider different (native and non-native) varieties in their teaching is promising. However, there are other issues in our results that require further discussion. Generally, results concerning varieties of English used by the participants in different situations seem to suggest that some teachers feel they are able to switch between varieties with ease; the responses across use in private and in teaching, as well as across the teaching of writing and speaking skills are similar, but there are some differences. Swe27m reports that he would describe his spoken English as AmE but that his native English-speaking friends would describe it as "Scandinavian academic dialect". His comment indicates that selfperception, perception by others (and, ultimately, actual use which was not determined here) may vary considerably. In this study, we have relied on selfreports, and a comparison with recordings from the same participants would shed further light on this issue.
One similarity between the sample groups from the two countries is that the majority of participants report that there are no official instructions to use one specific variety of English in their teaching; however, where varieties are mentioned, both AmE and BrE occur. An indication throughout the results is that the participants view BrE (i. e. RP), and/or AmE as the "correct" forms of speaking or using English. This surfaces partly through responses concerning which varieties the respondents use themselves or expose pupils to, and partly through responses to the question concerning Euro-English, as well as the questions regarding which grammatical variants are allowed as correct answers on a test. This stands in contrast to the sometimes very reflective and insightful responses that deal with non-native or non-standard varieties. In addition, the respondents themselves do not claim that the goal is native-like, "correct English" but rather the communicative aspect of being able to use the language. Perhaps these contrasting views are due to a mismatch between learned speaking ideals and didactic ideals concerning the communicative goal of English as a school subject in both countries, a theory-practice divide also mentioned by Galloway (2017). The present study shows that a dissociation of these two concepts is difficult for teachers in the sample, probably due to the difficulty of actually implementing teaching strategies targeted at communicative competence instead of target varieties (cf. Galloway and Rose 2018;Rose and Montakantiwong 2018). This is closely linked to a lack of appropriate practical materials, as well (Matsuda 2012). One solution could be an increased emphasis on these topics in academic and practical teacher training, as described by Selvi (2017) and Mora and Golovátina-Mora (2017), for instance. Modiano's solution is the introduction and acknowledgement of Euro-English "as a complement to traditional norms and practices" (2017b), and by doing so, "liberat[ing] continental European L2 users of English from the tyranny of standard language ideology" (2017a). The present study shows that teachers are becoming more flexible in their use and acceptance of varieties of English, which could be an indication that the Englishes spoken in Europe do not (any longer) fit neatly into any category of English along the lines of ENL, ESL or EFL, and are often difficult to define in terms of national varieties. Pupils picking features from a feature pool consisting of more than one variety instead is increasingly accepted by teachers. Still, the introduction of a new standard such as Euro-English is, judging by the results of the present study, not something teachers of English would welcome. Many of the participants in the present study would potentially, however, be interested in teaching it as an additional variety or style.
A very detailed picture of the use of different varieties emerges when comparing the ideas expressed regarding which variety of English the teachers consider should be used, to their reports of which variety they themselves use (in teaching). As a comparison, 96 % of teachers in training in Simon's (2005) Flanders study felt it was important to acquire a native-like accent. There seems to be a tendency among our participants to be less concerned with the ownership of English residing with the native English speakers, such as was the case in Gnutzmann et al. (2015). Our results are more in line with the views of the university lecturers in Simon's study who focused on intelligibility of English in relation to interlocutors (2005: 6) and the idea that exposure to different varieties of English in teaching might ultimately further communicative competence (Bieswanger 2008). On the other hand, Modiano's (2017a: 320) statement that "The idea of keeping the two varieties [BrE and AmE] separate has almost completely died out, and instead we see greater approval of a more general competence where the emphasis is now on the use of the language as a tool in communication." is only partially supported by our results: our respondents see a clear division and they relate to and adjust their knowledge of the separation throughout their (working) life. The second part of Modiano's statement is, however, upheld: there is indeed a greater approval of more "general competence" and the focus is "on the use of the language as a tool in communication" in the present sample. This is where the aforementioned dissociation between learned and didactic ideals emerges most strongly: while our participants generally seem to focus on fluent communication, the question of Euro-English triggers many responses implying that only "real", "authentic", native English should be taught. Responses such as these are signs that the learned ideals, or standard language ideologies in relation to English, are prevalent to some extent.
Further, looking at the preferred varieties of English in a number of the previous studies we can conclude that: 1) pupils view RP as prestigious and professional and tend toward BrE in formal contexts, while considering AmE as modern and non-standard, and leaning towards using it in informal contexts (Mobärg 2002;Carrie 2017); 2) some pupils prefer BrE (Westergren Axelsson 2002); while 3) other pupils prefer AmE (Söderlund and Modiano 2002;Henderson et al. 2012); 4) teachers prefer using BrE, both in teaching and in private (Westergren Axelsson 2002; Henderson et al. 2012).
In the present sample, when a variety is named, BrE and AmE are favoured by TEFL teachers in Germany and Sweden, with a slight preference among the German teachers for BrE, and for AmE among the Swedish teachers, but with the Swedish teachers being almost as likely to report using a 'neutral' or 'non-specific' variety of English as AmE when teaching. This is one of the main differences between the data reported by the German and the Swedish participants. It also ties in with Rindal and Piercy's (2013) and Aiello's (2018) reports of speakers aiming for a fluent nonnativeness when speaking English. These results are thus not in line with 2) and 4) above. As for the English used by pupils, there are indications that our participants agree with 1) and 3) above, and they report that AmE and mixtures or combinations of both BrE and AmE are most popular. A quantitative analysis of read and free speech by German English university students confirms these observations (cf. Jansen and Langstrof forthcoming). While the results from previous studies detailed above seem to stand in contrast to our findings concerning Euro-English, Gnutzmann, Jakis and Rabe (2015) report a preference for native targets over Euro-English by participants, which is considered a bad imitation of "real" English. This statement is reproduced in our results, namely that Euro-English is not something "real native speakers" would use, that it would constitute "dumbing things down to accommodate incompetence". It is important to remember, however, that we had no control over whether participants in the present study had any knowledge of Euro-English; for all we know some might have never heard the term prior to participating in the survey, and their unfamiliarity may well taint these results. Nevertheless, there appears to be some interest among the participants in learning about Euro-English. This interest does not (necessarily) mean that their regular target variety is replaced by this "new" variety, but it might be included when teaching non-native or nonstandard varieties, or indeed, any varieties teachers themselves may not use.
The majority of participants report using BrE and/or AmE both in private and when teaching, and reports are similar when asked about the pupils' varieties.
The main difference between respondents from Sweden and Germany in this sample appears to be that a fairly large number of the Swedish participants report using a neutral or non-specific variety of English, which is not the case in Germany. The term "Mid-Atlantic" is used by one Swedish teacher in response to which variety he/she uses for speaking in private settings. The differences and similarities between MAE and "a mixture between AmE and BrE" can, of course, be discussed at length, but would exceed the scope of this paper. Responses to questions on which variety the teacher uses for teaching also include responses by Swedish teachers such as "Watered-out [sic] European", "Scandinavian academic dialect". These responses, together with the large number of participants who report not using any specific variety when teaching or speaking privately (cf. the "neutral" variety in Rindal and Piercy 2013) further indicate that these speakers may not be aiming for a native-like production of English. Given the situation outlined in the beginning of this paper, it might be the case that the Swedes, who to a greater extent are exposed to English through the media, are more aware of their English not being native-like than the Germans. Another possible explanation could be that native varieties carry more prestige in Germany; however, this issue requires further investigation.
The German participants report learning more spoken English in a formal setting in relation to their written English compared to the ratio reported by the Swedish participants. This means that the Swedish participants report learning more spoken English in an informal setting, which could be argued to follow an expected pattern in respect of how speakers from these two speaker groups might acquire English, when considering the spread of English in for example the media (cf. Eurobarometer 2006). With Swedish TV programmes mostly being available in the English original with subtitles (Swedish Government 2002b) and programmes in Germany generally being dubbed (Goethe Institut 2018), Swedes are indeed much more exposed to English in their everyday private activities than Germans are. How recent and ongoing changes in media consumption towards more internet usage (for Germany: Statista 2018), could affect varieties of English used in the two countries is impossible to tell without further analysis of the language in the media in question.
However, no participant reports that pupils aim to use any "non-native" varieties of English, or that the pupil does not have a specific aim (although six Swedish teachers do not respond to this question, and one German teacher explicitly states that pupils do not think about this issue). There are parallels which could be drawn here between these neutral, "watered-down", MAE or "mixture" varieties and Euro-English, at least in the reports from teachers: they consider these as something widely different from native speaker varieties. They are, indeed, different; but it remains a mission for future research to determine to what extent they are different, and in what way.
Interesting as the results obtained in the present study might be, they are subject to certain limitations. First and foremost, they are not generalisable given the relatively small number of participants. However, they offer a comparable view into teachers' practices and provide a point of departure for future studies. Further, the insights gained from this comparative study apply to secondary schools only, it might be interesting to expand the study to include views from primary school teachers as well. Finally, as mentioned above, the present study is based on selfreports, which are always subjective and do not represent what teachers actually do in the classroom. However, the views of practitionersin particular from a comparative point of viewremain underrepresented in this area and provide valuable insight into the implementation of language policies and curricula in the classroom (Rose and Montakantiwong 2018). Thus, our results are certainly valuable and could, in combination with observational data, for instance, support the development of teacher training or new material for the TEFL classroom.

Conclusions
Despite official regulations for both German and Swedish schools stipulating an orientation towards communicative competence rather than the use of target varieties in TEFL, the concept of target varieties still seems to be very much prevalent in German and Swedish classrooms. The main varieties used as targets in both countries are AmE, BrE, a mix of the two, and some (sometimes unspecified) neutral variety. There are slight preferences for BrE among German teachers, and for a neutral variety along with AmE or BrE among the Swedish teachers. German pupils are reported to prefer a mixture of AmE and BrE, while Swedish pupils are reported to prefer either a mixture, or AmE.
Teachers appear to consider BrE and AmE as the most "correct" forms to teach in the TEFL classroom. There are, however, strong indications that teachers are in conflict between learned and didactic ideals, with the former being what they themselves may have been instructed to use (an English belonging to the native speaker), and the latter being what is currently in the curricula for both countries: the goal of enabling effective communication and allowing pupils to experience and learn about many different Englishes. Awareness of, and tools for managing the balance between these conflicting ideals should be part of teacher training across Europe.
In terms of the status of non-standard varieties among the teachers, there are ideals present regarding the use of the more mainstream varieties in teaching, but most teachers seem interested in the possibility of using or including other varieties (such as Euro-English) in their teaching.
Whether or not the situation in Sweden and Germany will change in the coming years remains to be seen. We envisage three main factors contributing to long term effects which will have to be assessed in the future: 1) increased internet usage and availability of media produced in different varieties of English, 2) an increased focus on communication rather than variety-focused education, and 3) whether the effects of Brexit on (as Modiano (2017a) puts it) the guardianship of Britain on the Englishes in Europe will be stronger than plain geographical proximity, on the issues studied here.