ENGLISH AS A LINGUA FRANCA: FORMS AND FEATURES IN A SWISS CONTEXT

This paper considers how the way that English is learned and used in Switzerland may affect the form it takes. Focussing particularly on features demonstrating sociolinguistic competence, it aims to present the different ways that English as a lingua franca speakers may, and may not, acquire the patterns found in native speakers. Although there is no way to predict which features will and will not be natively acquired, the paper demonstrates how sociolinguistic competence is a valuable tool in understanding how English might develop in future in Switzerland and elsewhere as different features have different outcomes.


INTRODUCTION
It is relatively uncontested that English is used as a lingua franca in Switzerland and has been for a number of years (Cheshire and Moser 1994, Droeschel 2011, Dürmüller 2001, 2002, Durham 2003, 2014, Rosenberger 2010, Watts and Murray 2001).The main development in this lingua franca use has been in terms of Swiss speakers' interlocutors: initially English was primarily used with tourists (Dürmüller 2001), but in the past two decades it has been increasingly used by Swiss speakers with one another as well, making it an intranational lingua franca and making English a de facto Swiss language.Despite this second shift, it is also quite clear that English continues to be conceptualized and taught in Switzerland as a foreign language rather than a second or third language.This means that the teaching models are native speaker ones; with material focussing on British English or sometimes American English varieties.This paper will consider the implications of these two partially opposing facets and discuss the consequences this may have in terms of the form(s) that English spoken in Switzerland may take.The focus will be on features which demonstrate sociolinguistic competence (Adamson andRegan 1991, Regan 1995) and the extent to which non-native speakers match (or don't match) native patterns of usage.Such features, where there are often two or more variants which are equally acceptable, grammatically at least, can provide a Cahiers de l 'ILSL 48, 2016 108 clearer understanding of how underlying and unconscious patterns are transmitted.While this will be obviously useful to gain greater insight into how English as used in Switzerland may be changing, it is also relevant beyond Switzerland: English is a lingua franca in an increasing number of countries.Many, like Switzerland, have so far made no changes in the way that English is taught and perceived.This discrepancy is not anodyne, as the way that speakers use English and acquire (or do not acquire) specific aspects of it are key to understanding the forms it takes (Seidlhofer 2011).As English is used more and more widely across the world, it is vital to re-examine the ways it is classified because the native vs. ESL vs. EFL model has long been thought to no longer adequately reflect the reality in many countries.

ENGLISH ACROSS THE WORLD
The current situation of English across the world makes it quite clear that many older models classifying the types of English speakers in different countries no longer fully portray the entire situation and there are almost as many ways of categorizing world Englishes as there are varieties of English spoken.As noted by Cogo and Dewey (2012), Jenkins (2009) and Meierkord (2013), some groupings focus on the functions of English in different countries, others focus on the mode/manner of instruction (Modiano 1999) and yet others on the historical background of English use (Kachru 1982).None so far fully encompasses the multitude of options that exist in terms of how/why/where English is used today however.How does one deal with cases such as the Swiss one where English is still taught as a foreign language, but which is, in some situations, used on an everyday basis across the country?While this paper does not aim to resolve the issue of how to group varieties of English, by discussing some of the outcomes different uses of English may have, it hopes to demonstrate why the models need rethinking at a time when English use as an inter-and intranational language is increasing world-wide.

ENGLISH IN SWITZERLAND
The difficulty in classifying the different purposes for which English is used is not surprising given that this can change substantially from country to country, even within in the same area.The way that English is used in Switzerland is very different from how it is used in nearby France, Germany and Italy, for example.Many aspects of Switzerland's use of English are tied more broadly to its multilingualism, but this does not change the fact that, unlike surrounding countries, English is regularly used by Swiss speakers amongst each other nowadays.

109
As mentioned above, this situation is a relatively new one: English was initially used primarily for tourism purposes and thus was spoken primarily with people from outside of Switzerland.While this use continues, it has also broadened and an increasing number of people in Switzerland use it with each other -primarily as a lingua franca in cases where they do not share the same mother tongue.Recent census results confirm this: the results from 2013 (FSO 2014) show that out of a population of 8 million, nearly 13% of the Swiss population (i.e. a million people) regularly use English at work and 4.6% usually use it at home.It is worth noting that the 'language at work' numbers do not include the figures for the non-working population of students for whom English is practically always one of the required languages in schooling, so the number of speakers of English in the country is most likely even higher.The much lower rate outside of work contexts underlines the practical purposes to which English is used in Switzerland.These numbers of English users have increased in the past 15 years: the use of English in the 2000 census was lower, although even then it was clearly used sufficiently to be considered a lingua franca (Durham 2003, Lüdi & Werlen 2005).This brief discussion of the changing place of English in Switzerland serves to underline the extent of its use in the country and make it clear that we must consider what form it takes, but has this shift affected the way that it is taught in any way?

ENGLISH TEACHING IN SWITZERLAND
To understand the teaching of English in Switzerland and how it may influence the form it takes, we must consider two separate aspects.How early and for how long English is taught, and the way in which it is taught and presented.
First of all, it must be said that English is taught earlier in Switzerland than previouslyin some cantons at least.In several of the German cantons there has been a push to make English rather than French the first non-native language children learn (Busslinger 2005, Cossy 2004) and most children start learning English by the time they are 11.This obviously will affect what the English spoken by younger Swiss speakers is like as it raises the likelihood of high levels of competence in the language, although see Pfenninger (this volume) for a discussion of how motivation levels and type of instruction can be stronger predictors of language competence than starting age.
Secondly, it is important to note that all language teaching (of English, but also of French, German and Italian) is oriented towards a model outside of the country: students who learn French are focussed towards France and not French-speaking Switzerland, those who learn German have textbooks with locations in Germany and those who learn English learn about British and American culture alongside the language.Of course, this is less surprising for English than for the national languages, but from the perspective of the teaching of English in Switzerland it could almost be said that the lingua franca use that might come post schooling is a bonus, rather than one of the main aims.Related to the external focus of English teaching, the lack of direct transmission of 'Swiss English' from generation to generation bears underlining: English is almost always initially learnt at school, which means that each generation is likely to have similar features to the previous one but not directly from them.There is nonetheless a small possibility of a founder effect (Mufwene 1996), at least at the point where learners of English become lingua franca users, in that any tendencies towards the use of one form over another will be strengthened by contact with others who also have those features.
Taking the increased lingua franca use and the fact that English remains taught as a foreign language, what can we expect the language to be like and to what extent could this help us better understand the changing faces of English more broadly?

FEATURES OF LINGUA FRANCA ENGLISH
Because of the way that it is taught and transmitted (in the classroom and almost never from parent to child), the English spoken across Switzerland is extremely unlikely to be identical across different linguistic areas, although there may be some shared features nonetheless (Droeschel 2011, Durham 2007, 2014, Rosenberger 2010) but it is still possible to examine how features with a range of variants, all of which are acceptable, are acquired by Swiss speakers as it can help us better understand how English as a Lingua Franca, ELF, is likely to change and develop more generally.It is also a chance to reflect on what features might change in cases, such as the one in Switzerland, where English is taught in one way and is used in a very different way.Previous research (Durham 2007, Durham 2014) has demonstrated that the reasons underlying the use of English can affect what it is like.For example, when considering the concept of sociolinguistic competence, it is clear that different functions and types of use can help predict whether learners will be able to match native speakers' patterns (Mougeon, Nadasdi & Rehner 2010, Howard, Mougeon & Dewaele 2013, Regan, Howard & Lemee 2009).This is important because the way that underlying features are transmitted to non-native speakers has potential implications for the future direction a language may take and can help us understand language change more broadly.Features that are 'lost' in a lingua franca situation may be recovered subsequently but if communicative urgency is key then they may not.

6. SOCIOLINGUISTIC COMPETENCE
What exactly is sociolinguistic competence?It is related to Dell Hymes' notion of communicative competence (1972), at least in the sense that it forms part of the performance and communicative aspects of language learning and socialization.Not only is it important to know when and how to communicate, but speakers also must know which form (from a range of registers or styles) is appropriate to use at what time.For example, English permits several relative pronouns within the same structure in some cases (example 1-3) and while they are all grammatically correct, there are nonetheless underlying patterns of use which native speaker follow with respect to formality, age, origin, etc.
1) The document we sent him 2) The document which we sent him 3) The document that we sent him (modified from Durham, 2014: 83) Some analyses, on English in Switzerland, but in non-native lingua franca varieties more broadly, have focused on the features where native targets were missed: words misused, prepositions misplaced, grammatical structures misanalysed (see Meierkord 2013 for a discussion of the issues with having this as the main focus).While useful, this seems to me only a partial view of what is going on: the focus on the salient and on the different belies the fact that for the most part the language used is a close approximation to what native speakers would have.The interest then lies in the more subtle side of things: the cases where two forms are used by native speakers and really either is acceptable in most contexts, but they still are constrained in some way.This makes it possible to look beyond surface similarity and establish whether the underlying processes are the same as well, which allows for deeper insight into how closely the patterns are shared.
Given that a prime focus for English as a Lingua Franca is communication rather than matching native speaker use and patterns, it might seem odd to examine sociolinguistic competence as it is clearly highly linked to native speaker norms.But it is useful as it can better pinpoint how and when transmission is uncomplicated and where it is less so.Additionally, because of the way English is taught in Switzerland and because a native-like competence is assumed to be the aim, it can help us establish to what extent this is in fact the case.
Sociolinguistic competence and how it is acquired has received increasing attention in the 21 st century and this focus can be found in language acquisition contexts of different types: native children (Foulkes, Docherty & Watt 2005, Roberts 2012, Smith, Durham & Richards 2013), students in classroom contexts (Mougeon, Nadasdi & Rehner 2010), students in study abroad contexts (Dewaele andRegan 2002, Regan, Howard &Lemee 2009), immigrants to the United Kingdom and Ireland (as children or as adults) (Drummond 2011, 2012, Meyerhoff & Schleef 2012, Nestor, Ní Chasaide & Regan 2012, Sharma 2011), new speakers of minority languages (Nance 2015) and lingua franca speakers (Durham 2014).These different contexts have revealed that sociolinguistic competence is first of all not necessarily straightforward to acquire and in a number of instances learners/users never match native patterns, but also that different types of features, different uses of the variants (e.g. is one variant more stylistically constrained than the other) and of course the contexts themselves can help influence whether or not native patterns are replicated.
In terms of the potential outcomes of lingua franca use with respect to variable features, there are three main options which can each help us understand the ways sociolinguistic competence can be acquired.
A. Variation fully acquired.

B.
Variation not acquired due to learning related aspects.

C. Variation not acquired natively but new patterns visible.
Outcome A represents cases where native sociolinguistic patterns are found to be replicated in the non-native speakers.These would be instances where it would be possible to demonstrate that the non-native speakers had been able to pick up on the subtle, underlying patterns and use them in the same way as native speakers.Outcome B represents those cases where the native patterns are not replicated, i.e. those where the non-native speakers were not able to acquire the underlying patterns and produced something markedly different from native speakers.The contrast with this outcome and outcome C lies in whether the patterns found are shared across several groups (the French, Italian and German speakers in this instance) (outcome C) or whether they appear to be primarily due to language transfer and related learning issues (outcome B).
Outcome C is not always considered in research on the acquisition of sociolinguistic competence (but see Nance, 2015, Nance, McLeod, O'Rourke & Dunmore, 2015), but it is important in that it represents cases where the divergence from native patterns is not due to native language transfer or language learning difficulties, but rather where it is due to the fact that the non-native group has modified the existing patterns to their own purposes.It is especially important to allow for this option in lingua franca cases where language use is not necessarily modelled towards the native speakers.

7. MAIN FINDINGS OF PREVIOUS RESEARCH ON ELF IN SWITZERLAND
This section will briefly discuss examples of all three outcomes taken from previous research (for a full discussion of the features and the findings see Durham 2014).The data it comes from is a prime example of English as a lingua franca in Switzerland use and comprises a set of emails collected from 2001 to 2005 sent by members of a medical students' association (see Durham 2003Durham , 2007Durham , 2014 for a further discussion of this).In the period examined, English was the main language used by members from across Switzerland and whose native languages were French, German and Italian.This represents a natural and unforced use of English as the members decided themselves to make English the main language of their emails following a realisation that a mix of French and German (according to home university) was not sufficient to ensure understanding throughout.While these medical students were no longer learning English, they had of course all taken English during their school years.This is precisely the situation discussed above, in that although their learning had prepared them for English use with native speakers, they in fact used it with other non-native speakers and in a quite different way than it was taught.
In the case of relative pronouns (examples 1-3 above) and complementizers (examples 4-5), the results clearly demonstrated that the Swiss speakers had acquired the native patterns (outcome A).This was despite the fact that the full variability was not taught at school.4. I think you'll understand why.
5. I think that you'll understand why.(adapted from Durham 2014:116) In both cases, the Swiss speakers furthermore showed very low rates of instances where the features were used in a way that would not be acceptable for native speakers (for example, using who with inanimate objects).This demonstrates that in some cases ELF speakers are able to match the unconscious patterns found in native speakers.In terms of the forms that ELF can take, this underlines that even some supposedly complicated aspects may be transferred over to non-native speakers and maintained.
The variation found in the use of the additive adverbials also, as well and too on the other hand was purely constrained by the speakers' native language and none of the patterns found were comparable to the native speakers (examples 6-8) (outcome B).
6.They also have to go for two months.

114
The data also showed high rates (20%) of the adverbials used in positions that would not be grammatical for native speakers.This feature underlines that in some cases variability will not be acquired and ELF forms will be markedly different from the ones of native speakers.
Finally, the variation between will and going to demonstrates a case of outcome C: where the variation is not acquired but where new patterns are visible (examples 9-10).9. I'll be quiet.10.I'm going to be quiet.(adapted from Durham 2014:68) Here, the Swiss speakers did not have the variable patterns found in native speakers (Durham 2014, Tagliamonte, Durham andSmith 2014).In fact, their use of going to (and the shorter and more colloquial form gonna) was severely restricted, to the point that it could almost be claimed that it was not part of their repertoire.This is despite the fact that the use of will and going to is a topic covered in English language classes in Switzerland (and elsewhere).The lingua franca speakers had unconsciously discarded a second variant which was too similar, and restricted their use to a single form.However, the French, German and Italian groups patterned very similarly, leading to the conclusion that this might be a case of focussing or an instance of a lingua franca specific form.

DISCUSSION
What does it mean that features are not all learnt the same way and that sociolinguistic competence is matched in some cases and not others?In terms of language acquisition, it might signal to us which ones are most likely to be 'easy' to acquire and which might contain aspects which make them more transparent for speakers to match.In terms of the forms English may take in Switzerland and in other countries where it is used a lingua franca, it underlines that sociolinguistic competence is potentially a valuable place to look for innovation and focussing.
Using English as a lingua franca allows speakers to be perhaps less precise or at least less concerned with the more formal rules and structures which would come into play in a situation with native speakers or where the emphasis was on correctness.This merely serves to underline the fact that English as a lingua franca is not an institutionalised, formalized, 'stuffy' form of language use, but a more living and breathing one.The students in the study discussed above joke with each other in English, engage in word play and so on.All these are things which are not formally taught in class.It is this change, no doubt, with gives rise to the various patterns which are found in terms of acquisition of sociolinguistic competence.Some 115 features remain more closely tied to the way they were taught (overtly or not) and others less so.
When, as in Switzerland, English moves beyond the classroom into the workplace (and in some instances the home), the form that it takes cannot be but different from how it was taught: even if it is not necessarily possible to posit the existence of a single 'Swiss English' shared across the country, at an individual level it is not equivalent to the taught version.Some features will be lost completely (such as the variation between as well, too and also), others will be maintained (relative pronoun choice and complementizer deletion), and finally some will be transformed in new ways (such as variation in the future tense).In the examples presented above, frequency and lexical effects are very likely to play a role in which outcome is found: with items that are both frequent and which have a restricted set of lexical variants (or occur with a restricted set of lexical items) are more likely to be acquired in a way that matches native speakers.Other features, instead, are more likely to follow their own paths: the formal variant might be the main one used in some cases, while the informal one might be chosen in others.It is difficult to predict which outcome will be found a priori and thus underlines the importance of considering a range to get a better idea of the overall situation.The results show that some features related to sociolinguistic competence are fully acquired, while others are not so.Because the variants of the features examined are all generally acceptable (except in terms of positioning in sentence for the additive adverbials), this is only visible when looking beyond the surface at the underlying patterns.Without establishing whether the unconscious patterns are matched, it is not possible to know whether the ELF speakers are modifying the patterns in their own way.

CONCLUSION
This paper has shown how even features which on the surface seem to be used natively may demonstrate underlying differences and may help pinpoint which features are restricted to a specific country or context.Conversely, some features can be shown to be used very similarly to native speakers even in cases where the teaching of the variation is not overt.This underlines that non-native speakers are able to fully match native speaker patterns and acquire sociolinguistic competence in some cases despite the fact that their main use of English is not with native speakers.Overall, these findings demonstrate the importance of considering such features when examining what forms ELF may take in different countries as they can show what is different and what is similar beneath the surface.To fully understand how ELF is developing worldwide, we need to look both at what is immediately visible and what is less so.
7. They have to go for two months as well.8.They have to go for two months too.(adapted from Durham 2014:135) Cahiers de l'ILSL 48, 2016