Shifting ground: exploring the backdrop to translating and interpreting

ABSTRACT Previous scholarship has assumed a monolingual norm. This norm has been noted across disciplines and societies, where it demands assimilation and language shift. However, socially responsible language scholars have increasingly rejected this norm in favour of renewed interest in language contact, and associated complexity. Many social institutions in which interpreters and translators work in modernist nation states view languages as separate, separable units which come into contact in highly regularised ways and can therefore be highly regulated. I argue instead that a renewed focus on ethics in the study and practice of translation and interpreting involves recognising forms of communication which are increasingly common and diversified due to superdiversity. This paper describes an investigation into multilingual language practices in four changing UK cities, using Linguistic Ethnography. The study’s own approach to ethical practice is first evaluated. I then ask how increased study of multilingualism can contribute to interpreters’ and translators’ work and how this work fits with contemporary patterns of language use, drawing on distinctions between the ethical and moral. I suggest that, whilst translating and interpreting can offer an ethicised approach to language contact, a truly moral approach may require rich understands of contemporary superdiverse societies.


Introduction
It is frequently noted that a good deal of past scholarship and large swathes of social life have worked from the assumption that there exists a monolingual norm (Brutt-Griffler and Varghese 2004, 93). This norm follows from an assumption that the world is monolingual and that even when more than one language is made relevant, interaction should still be modelled on a monolingual situation as far as possible (Canagarajah 2012, 3) entailing that the two or more languages in play should be functionally separated (Cummins 2008). This norm has been noted by scholars across disciplines (e.g. language teaching and testing (Jessner and Cenoz 2007, 159); psycholinguistics (Zurer-Pearson 2010) and sociolinguistics (Rojo 2010, 60)) and occurs within many societies where it demands, from speakers, assimilation (Bartha and Anna 2006) and language shift (Li Wei 1994). The norm is enforced through such devices as expectations and policy and through language activities like interpreting and translating. These activities, for all that they bring two languages into play, would usually keep them demarcated and separated.
In recent years, the monolingual norm has been seriously problematised. As the commitment to this norm has been critiqued, so it is coming to be dismantled. The potential value of this re-appraisal of the social order has provided a new impetus for a socially engaged language scholarship with a renewed and reconceived interest in language contact. As a result, the complexity and diversity of forms of contact which can result when languages, and indeed people and practices, interact have come to be investigated using new analytic and theoretical frameworks. One of these, translanguaging, a form of linguistic fluidity which arises 'as speakers juggle the limits of face-toface intelligibility . . . with new styles of expression made up of ever-changing linguistic resources' (Parkin 2016, 73, see also, for example, Creese and Blackledge 2010), is a focus of this paper. Beyond the academy, late modern conditions have pushed modernist nation states onto 'shifting ground' which witness 'rapid contemporary transformation' in social arrangements and composition (Stroud and Wee 2012, 26). Nonetheless, whilst interpreters and translators are likely to be familiar with a rich multilingual tapestry which may run throughout their lives, many of the social institutions in which they work still view languages as separate and separable units which come into contact in highly regularised ways and can therefore be highly regulated whenever they meet. In other words, many social institutions still operate on the assumption of a monolingual norm even though many social actors within them do not. In this paper, I argue that a major component of a return to ethics in the study of translation and interpreting (Pym 2001) involves recognising the complexity of forms of communication which are now common as a result of the superdiversity of contemporary social life (Vertovec 2007) and responding to that spread (Angermeyer 2015). Similarly, if we are to consider social responsibility, or accountability to others, in and surrounding translation and interpreting, we need to recognise that responsibility, inevitably social, is 'fluid, negotiated and constructed in interaction' (Solin and Jan-Ola 2014, 288). I suggest that social institutions should at least be sympathetic to the language practices which contemporary social conditions seemingly precipitate and explore how close observation of social encounters could enable them to move beyond only sympathy.
To that end, this paper describes a large research study which is investigating multilingual language practices in four UK cities using the detailed gaze offered by Linguistic Ethnography. The paper presents the research as part of a move to increase knowledge of multilingualism in the UK in this time of change in order to encourage institutions to reflect on the ethical dimensions of their assumptions about multilingual people and communities. The paper in turn considers how increased study of the multilingualism which is part of people's daily lives can contribute to the work of translators and interpreters and how translating and interpreting fit with contemporary patterns of language use in England and Wales. In doing so, the paper examines a distinction between the ethical and moral, proposed by Brecher (2010) and suggests that whilst translating and interpreting can offer an ethicised approach to language contact, a truly moral approach may require an enriched understanding of contemporary superdiverse societies. It also takes interpreting in legal contexts as a case study and, drawing on ideas from Angermeyer (2015), considers the specific interpreting activities which might be developed through a detailed understanding of translanguaging. Tusting remarks that, for language scholars, 'one of the key ways in which social responsibility can be carried out is through speaking out to others in the social world ' (2014, 440). Pursuing this theme, the paper closes by considering a socially responsible position for language researchers, given their potential to speak.
Diversifying understandings of language contact in the UK An example from the work of Marco Jacquemet (2014) helps to lay the ground for an exploration of the central concept of this paper, translanguaging, and to indicate its relevance to translating and, particularly, interpreting. Jacquemet examines interpreted interviews in the asylum process in which people seeking asylum must state their case and on which decisions about their future are made. In the example below, the young man seeking asylum knows Pashtun, the interviewer knows Italian and the interpreter knows Farsi, Pashtun and Italian. A translation is provided in italics on the right of the page: Interviewer [to interpreter]: l'udienza si terrà a Roma o a the hearing will be held Ca = serta? = in Rome or Caserta? Man seeking asylum: = Roma = Rome From Jacquemet (2014) Jacquemet's 'equals signs' indicate that the word 'Roma' is said, by the young man seeking asylum, in overlap with the interviewer's question. Jacquemet explains that the young man 'eager to show his awareness of the exchange and his ability to answer' draws on his knowledge of context and recognition of the proper noun 'Roma' (Rome) to answer the interviewer's question without the interpreter even getting the opportunity to speak. This despite the man being unable to 'comprehend a whole Italian sentence'. Jacquemet observes that in interviews like this where central participants do not share a language, proper names, particularly personal names and place names, take on particular significance because of their powerful denotational referentiality. He points out that proper names may be seen as linguistic resources which are available to be understood and used by all participants because of the way that they might survive in some forms across languages. He explains that they can become 'anaphoric cairns that allow listeners to follow, albeit approximately the turns they do not understand' and that they can allow applicants to make sense of the otherwise incomprehensible flow of talk. How might such exchanges be received by participants present? To answer this question and introduce lay understandings of multilingual practices and interpreting, I turn to a different context, the police station. I present a short vignette which illustrates how institutional actors can have difficulties recognising subtleties of the multilingual experience.
Despite speakers' confidence in their abilities to evaluate their interlocutors' bilingual capabilities, research suggests that a strong performance of communicative competence can be misleading. Eades (e.g. 1982;1984) has shown that apparent proficiency in an additional language is not necessarily a good indicator of likely comprehension of that language. She exemplifies, specifically, that shared grammar between speakers might be accompanied by 'significant pragmatic differences' in other words, a lack of fit between the way that language intended and used. Indeed, it has been suggested that those who speak acrolectal varieties of a language (varieties which are close to the standard form) can be placed at more of a disadvantage when they face the law than those who speak basilectal varieties (those more obviously different from the standard) because their different communicative frameworks can be masked (Mildren 1999). Thus, institutional actors who use their observations to calculate that a speaker can get by in the institutionally dominant language might be on very shaky ground.
In stark relief to the monolingual norm mentioned in the first section of this paper and the expectations and beliefs mentioned above, contemporary scholarship is pushing the boundaries of our understanding of language practices in ways that unpick these ideas about languages and dialects in social contexts. This scholarship illustrates that people do not use languages in isolation from other languages. It shows that people have, indeed, evolved ways of making meaning that call in wide arrays of linguistic resources not even confined to spoken languages. This section illustrates the development of this research trajectory and further indicates its relevance to interpreting.

Vignette: The all-or-nothing of interpreting in police interviews
Over the last two decades, I have undertaken a good deal of research with the police (e.g. Rock 2001;Rock 2007;Rock 2015). This has enabled me to have many conversations with police officers and police trainers about language as part of their job and as part of the activity of policing. I have found many police personnel to be thoughtful about language, sensitive to its nuances and power and creative in their use of language to do their work conscientiously and judiciously. Yet in view of this, officers often profess surprising views of multilingualism. On multiple occasions, I have heard officers saying that some suspects or witnesses ask for or come to be speaking through an interpreter despite 'obviously knowing English'. This is a problem, they say, because such people 'get the benefit of hearing everything twice' during interviews and other investigative processes (quotations from fieldnotes). This position is rooted in a monolingual norm. It suggests that languages are discrete from one another so that they do not spontaneously come to be used simultaneously or in combination. It suggests that people will either know a language in full or not know it at all (even the language of the place where one happens to be). It suggests that if people are able to use a language in one context, such as an informal chat in a police custody block or in undertaking study such as a degree, they will automatically be able to use that language just as effectively in any context, including the stressful, complex often somewhat formal and technical police interview. It suggests that languages are portable across contexts and constant in that portability. The position shows, in short, a lack of awareness of the complexities of the experience of language for the bilingual individual. These views are notable, though perhaps not surprising when placed in their wider context. The monolingual norm is deeply rooted across social life in many parts of the world and the UK is just one example despite its history of language contact involving both 'native' and 'migrant' languages including English. In schools, students are likely to have learned to compartmentalise languages according to the separate class sessions in which they were taught (Byram 2008, 21). The expectation seems to persist that single languages will be seen and heard in isolation from one another and that anything else is 'broken' language at best. There is no necessary reason for those employed in public service, local government, and even the third sector to develop a different philosophy of language, even if they know multiple languages themselves. Despite living amongst a rich tapestry of language practices, it can be easy to be drawn into the monolingual norm, 'norm' being the key word, so that other realities are veiled or problematised.
Sociolinguistics has long had an interest in arrays of languages and in investigating what happens when languages co-habit in social spaces (e.g. Blom and Gumperz 1972;papers in Milroy and Muysken 1995). For many years, the dominant paradigms to investigate such co-habitation were ones which stressed the separation and distinctiveness of languages, even when they sat alongside one another and jostled together within conversations and even utterances. Contemporary paradigms, however, add attention not to the patchwork of 'languages' which is evidenced by conversations in which languages become entwined with one another, but to the resulting forms, the new, often transient, ways of making meaning which arise in such social interactions. In other words, attention has shifted from the combinations of languages and other varieties in evidence in social contacts to the forms of interaction which are created from those combinations.
The translanguaging approach to the use of languages evidences this shift. It 'considers the language practices of bilinguals not as two autonomous language systems as has traditionally been the case, but as one linguistic repertoire with features that have been societally constructed as belonging to two separate languages' (García and Li Wei 2014, 2). These ideas of societal construction being the determining factor in views of language and their interrelationships clearly relate to the point of Jacquemet's example and the vignette, above. The term 'translanguaging' arose out of understandings of 'languaging', itself intended to loosen the fixity of the notion of 'language' and to move to viewing language as 'something we do rather than a system we draw on' (Pennycook 2010, 2). As Becker puts it 'try thinking of it this way: assume that there is no such thing as Language, only continual languaging, an activity of human beings in the world.' (1991,34). His emphasis is on the constant formation and reformation of language rather than giving attention merely to its structures. This enables an expansive view of language. The notion of 'trans-' is added to the concept of 'languaging', as García and Li Wei note, to indicate the additional complexities and diversities of experience of multilinguals engaging with languaging when compared to monolinguals (2014, 18). Translanguaging is not merely about language contact but incorporates a discursive element. The framework sees codeswitching as unmarked and usual so that the analyst can focus on the way that multilinguals deploy resources (Noguerón-Liu and Warriner 2014, 182). Translanguaging stamps on the monolingual ideology which otherwise keeps any speaker from 'performing their linguistic identity' (Hélot 2014, 218). In this sense, the concept has sometimes come to be used as part of a rallying cry to valorise multilingual language practices. This is because translanguaging is not only about conceptualising language practices. The concept also implies social consequences. As García and Leiva explain, with reference to classroom contexts, translanguaging can enact processes of social transformation by releasing learners from the constraints of monolingualism (2014,205). Their point about the relevance of translanguaging to social justice and social responsibility might appear to be confined to the educational setting. However, I want to bring it out of the classroom by considering how understanding translanguaging as it happens across social contexts can inform the practice of those who work with multilingual individuals elsewhere in society. I suggest that as in educational contexts, by modifying the parts, in other words, by changing orientations to languages and languaging, 'a new reality emerges' (García and Leiva 2014, 205).

Linguistic ethnography
I now turn to an approach to studying multilingual and other language arrangements which produces rich understandings of social life by examining that life closely. I suggest that research which uses this approach, Linguistic Ethnography, can contribute to an ethics of interpreting and translation.
Linguistic Ethnography (LE) developed as a branch of sociolinguistics during the late twentieth century. As its name suggests, it draws together ethnography and language study. Through this particular combination, it aims to enrich the study of language in social life. Crucially, it entails studying 'the local and immediate actions of actors from their point of view and considers how these interactions are embedded in wider social contexts and structures' (Copland and Creese 2015, 1). The effort to connect the local and particular to the structural which Copland and Creese describe indicates LE's utility for drawing connections between individual multilingual interactions and the institutional arrangements through which and in which translating and interpreting happen. The emphasis on taking the emic point of view, in Copland and Creese's gloss, is important to how linguistic ethnography gets done and the way that a linguistic ethnography of a multilingual context needs to be embedded in and open to that context in the fullest sense.
In terms of data collection, ethnography brings to Linguistic Ethnographic work, close observation of social life and the systematic recording of those observations through such activities as taking fieldnotes, making drawings, and holding interviews and other conversations with participants and recorded research discussions. The particular attention to language in Linguistic Ethnography comes through audio-recording naturally occurring spoken data and photographing and otherwise collecting written language. Because LE work typically also attends to the context beyond only strictly 'linguistic' communication video-recording, photographing people in interactions and consideration of modes of communication which are often seen as extra-linguistic such as gesture and visual images also contribute bringing attention to the multimodal complexity of social life (illustrated in the work of Bezemer e.g. 2015). By combining forms of data, LE provides for analysts to amalgamate their observations of social encounters with participants' own perspectives. This happens ethically in ways which respond to what we might call a 'participant turn' in social research. In this way, LE attempts to work with and for research participants rather than only on them (Cameron et al. 1993). LE's analytic method is rich because researchers might choose to come at their analytic task through the door of any of the individual forms of data they have collected or through a deliberate combination of forms of data or, indeed, through theoretical or applied questions. Ultimately, the various ways of looking which are implied in LE are brought into dialogue with one another when the forms of data are combined and recombined throughout the analytic exercise. The theoretical basis of LE is not fixed although there is a notable influence from the work of Bakhtin (e.g. 1986) which draws attention to various forms of multiplicity and flow in communication. The work of Gumperz has also contributed to LE (e.g.1972) bringing its focus on language contact and context. Copland and Creese (2015, 17-22) add to this mix the work of Goffman, Hymes and Erickson as having laid foundations in providing ways of looking at the social world through the lens of language. In general, discourse analytic approaches are key to LE analytic work and its theoretical orientation. LE itself has been criticised for failing to 'fully engage with its social responsibility in making the connection between small scale findings and wider social implications' as Creese (2008) notes. However, Linguistic Ethnography can contribute to an ethics of interpreting and translation. It can do this in at least four ways: (1) Raising the profile of multilingual language practices for policy-makers and the public by simply providing a reminder of the existence of these practices, in order to encourage movement away from the monolingual norm (2) Increasing understanding of multilingual language practices by demonstrating their diversity to a range of practitioners (3) Providing tools for professionals such as interpreters to appraise their own work using scrutiny of that work from LE (4) Providing case studies for use in learning and other contexts which exemplify good practice Rock (2007) illustrates similar interventions in non-interpreted legal settings. The next section picks up some of these points by showing just one realisation of the LE approach in a large study. This indicates the study's relevance to contributing to an ethics of interpreting and translating.

Developing understandings of translanguaging in england and wales
The project 'Translation and Translanguaging: Language and superdiversity in four UK cities' (Tlang) began in April 2014 running for four years. It involves linguistic ethnographic research in four cities, Birmingham, Cardiff, Leeds and London. In each city, a multilingual team of researchers has conducted periods of fieldwork in sites in four spheres of social life: business (e.g. shops, market stalls and selfemployed working); heritage (e.g. libraries, actors, artists); sports (including football, basketball and karate) and law (with a particular focus on legal advice). Each fieldwork period lasted for four months and centred on the activities of a Key Participant, who was also designated as part of the research team yet who was a practitioner in, and member of, the research site without any previous connection to the academic field of study or the research team. In each site, each team's activities involved making observations and taking fieldnotes in order to record the researchers' understandings of events and to record as broad a context as possible for the other forms of data; taking photographs within the site and of the surrounding linguistic landscape in order to enrich the more traditional forms of language data and develop a situated understanding of each site over time; making audio-recordings and where possible video-recordings of both front-stage and back-stage talk in order to build a coherent picture of the key participants' language practices; collecting social media data in order to understand each research site beyond its immediate physical location.
Many aspects of the research design exhibit what we might call an ethics of multilingualism in the form of a commitment to multilingual working in its diverse forms. For example, the project's research assistants, one or two based in each city, were hired because they spoke multiple languages and brought a knowledge of local contexts rather than for traditional academic qualifications (although several of them had these too). The project's success is absolutely dependent on these Research Assistants in translating data but also in exploring those data. The project's design simply does not work without a multilingual team to bring to bear not only their language knowledge but their ways of making sense of the meaning-making they encounter. Similarly, the project design affords a special status to the key participants who were all either multilingual themselves or worked in a multilingual environment. They were given this status through research training which was intended to position them as valid contributors to the intellectual work of the project, not mere research subjects. In reality, some participants took up this position more than others and all took it up in different ways however, at the very least, they gained some research training which is a potential resource for the future. The process of making sense of the data was ideally a dialogue between the university-based researchers and the key participants. These integral aspects of the research design meant that even where multiple languages were not immediately evident in the data, there was an orientation to translanguaging in the form of awareness of individual participants' repertoires of resources (after Blommaert and Bakus 2011).

Considering translanguaging data
This section presents just a fragment of data from the Tlang project to illustrate the above points a little. It is difficult to discuss translanguaging without essentialising the languages present in order to accomplish description. Labels for languages are so pervasive that they can seem to drag the analyst back to separation of 'codes'. However, the research design rested on the notion that translanguaging in its fullest sense would be recognised as such so the labels used below are simply a shorthand for the indexicalities of the forms they denote.
The example below is from our first research site, a shop in a suburb of the city of Cardiff, Wales. The shop was owned by an Iraqi married couple who employed a butcher to run a halal meat counter within the premises. The owners, Mr and Mrs B spoke both English and Arabic in the shop and in their home and online lives. The butcher, Anil (all names are pseudonyms), was an avid language learner who as well as his Urdu and Panjabi brought from home in Pakistan, spoke English, Arabic and even some Malay which he had recently been learning from friends. In the fragment below, Mr B is on the telephone, placing an order for chicken to be sold in the butcher's area. The chicken company is another local, migrant business, located on the other side of the city. It is owned by a family from Pakistan. During the call, Mr B moves between his call and a conversation with Anil as he firms up his order. In the transcription below, translations of words originally said in Arabic appear in italics with the original Arabic alongside. A full transcription key, detailing the other conventions used is provided at the end of the article. Turns from the chicken vendor were not audible on the recording but were all short: Excerpt 1: the telephone call [CarBusTr_20xx1022_010] The excerpt opens as Mr B answers the telephone. He begins by greeting his interlocutor and, in turn 5, checking who he is speaking to, which enables him to establish whether the caller is the same person he discussed orders with earlier in the day. Through this process, he discovers how much contextualising detail he needs to give in his call and how much he needs to undertake discursive translanguaging, moving across discourses rather than languages, for this purpose. Once he has found that the caller is familiar with his business of the day, he begins his order by specifying that he would like a quotation for Freeman's chickens (turn 7), Freeman's being a relatively local halal slaughterhouse with whom his interlocutor trades. He knows that he and the caller will 'speak the same language' around chicken so he does not explain the terms 'Freeman' (turn 7) or 'baby chicken' (turn 17) and he does not specify how many chickens he would expect to find in a 'box' (turn 9, 15 and 17). From turn 15 onwards he begins checking details with Anil, the butcher in his shop. Here, Mr B becomes a mediator between the chicken butcher and the shop butcher. He frames this shift in activity, moving to including Anil in the conversation by explicitly marking it with the words 'let me ask Anil'. In doing so, he assumes either that the caller will know who Anil is or that the caller will be able to infer that Anil is a local butcher or stock-control staffmember. Alternatively, this framing shows a lack of orientation to his interlocutor in that he says words which implicitly the caller cannot 'understand' in the fullest sense. In any case, the strategy appears to be successful in that it enables him to assume a liminal position between the two conversations. This, itself, contains an element of 'trans-' in that he is neither in only one conversation or the other yet simultaneously in neither and both. As Anil joins the conversation, he too shows himself to know the 'language' of chickens, their sizes and supply conventions (boxes). Mr B does what may be a little translation in line 11 as he refers to the chickens as 'medium' rather than 'Freemans'. Anil fluidly joins the conversation able to provide information and to understand and participate in the 'trans-' situation of the double-sided interaction. The requests for information are marshalled by Mr B between the caller and Anil, and the telephone call is rapidly completed in turn 23. This is not an end to the discussion of chicken supplies, however. Instead, Mr B now holds a conversation with Anil only, about 'large chickens' and their preparation before the pair decide to buy two boxes of those, too. After serving a customer in turns 30-33, Mr B calls the chicken supplier once more at the end of turn 33 and places an extra order. On this occasion, he does not undertake any protracted greeting sequence at the beginning of the call but instead greets the remote butcher with the word 'brother' and places his order before expressing concern about the price and hanging up in turn 37.
Regarding language choices themselves, Mr B uses predominantly English when speaking to the caller who does not share his Arabic. However, the analysts allocation of words to 'languages' is not as straightforward as they might seem. Mr B uses the Arabic salutation 'waʿalaykumu s-salaam', responding to the pair-and-response initiation 'as-salaamu ʿalaykum' reportedly said by the caller. In this exchange, the pair can be said to have 'switched' to Arabic yet the caller does not otherwise 'speak' Arabic. Therefore, it is more meaningful to say that the initiation from the caller is part of his repertoire in neither language specifically but in his individual meaningmaking system as a whole. The caller is a Muslim man who owns a halal butchery firm so his ability to 'use Arabic' to open and close telephone calls is highly communicative of his situation, job role, orientation to his customers and to his trade. For him, greeting customers in Arabic is more than just operating in Arabic. It can signal a great deal about his social relations with his customers and his orientation to his work and clientele. Mr B's response 'waʿalaykumu s-salaam' could also be described as being 'in Arabic' yet, having spoken to people from the butchery company on previous occasions, he selects English for the rest of the call so, once more, this greeting is better viewed as holding interactional currency, rather than simply switching languages. Use of 'brother' at the end of the salutation is 'in English' yet, like the Arabic greeting, it signals a connection to a socio-religious way of constructing relations beyond the family with routes in Islam. In this sense the whole salutation, incorporating two languages transcends either in its meaningmaking potential. Furthermore, in turn 11, Mr B tells the caller in English 'let me ask Anil' and then immediately speaks to Anil in Arabic, first nominating him by name. In the transcription, Anil appears in English when the word is addressed to the caller and in Arabic when it is addressed to Anil yet what 'language' is being used here? Such turns confound the notion of separate and separable languages.
Staying with the theme of languages in the excerpt, from Mr B, we see predominantly English addressed to the caller and the Welsh customer who enters in turn 30 and predominantly Arabic addressed to Anil. Thus, the directive in line 14 from Anil 'buy five' is made in Arabic but converted into English by Mr B as it is passed along to the caller. Yet Anil and Mr B sometimes spoke to one another in English beyond this conversation. In this instance, though, Arabic allowed Mr B to discuss matters which are ultimately financial with his colleague without disclosing them to the caller. He also hung up the call in order to discuss further aspects of ordering, rather than asking the caller to hold on whilst he did this. The language choice made for a little privacy in tough financial times for this business. When Mr B and Anil speak on this occasion, some 'English' words are used and these are noticeably technical terms associated with shop-keeping and butchery 'large' (turn 23), 'order' (turn 25) and 'fillet/Philip' (turns 26 and 28), the latter, a little difficult to hear on the recording referring either to a cut of meat or another local meat supplier. These are not words which are inaccessible to either speaker in Arabic yet they appear in English here as part of the way that the colleagues make this particular social context work.
The shop customer converses with Mr B entirely in English. This is not inevitable when 'English speakers' come into the shop as some, like the caller on the phone, would use 'Arabic' greetings, product names or terms of address. The use of this particular English here therefore indexes the shopper's appearance, demeanour and familiarity to the shop keeper. It is not an inevitable choice.
Even this short excerpt has shown the delicate interplay of languages in just one social context which is diversely employed and motivated. We see how speakers draw on multiple languages to undertake their daily activities whilst simultaneously expressing their identities, orientations to others, social needs and expectations. The excerpt has shown how this happens even in mundane and really rather unimportant social interactions. By foregrounding the multilingual language practices in everyday language to those who work with interpreters and translators and to linguists themselves, we can develop more informed and responsive approaches to multilingual language practices in interpreted encounters and use of translated texts.

Ethics: an ethicised or moral approach
The previous section has advocated a more informed and responsive approach to translating and interpreting. However is an informed approach necessarily more ethical? If so, is a more ethical approach enough? In any case, whose ethics are at issue here? This section addresses these questions and uses them to examine the usefulness of increased awareness of multilingualism beyond interpreted encounters to interactions within interpreted encounters.
Brecher makes the useful distinction between an ethical and moral approach to actions and activities in the social world. For him, an ethical approach is concerned with the individual. Therefore, ultimately, 'ethics' serves to apportion blame for professional and political decisions to individual social actors. A moral approach, on the other hand, is concerned with human matters of right and wrong in the form of values and social respect. Through this division, the argument which Brecher constructs identifies a neo-liberal pattern of 'dressing up' all manner of socially disruptive policy decisions as ethically driven. This allows for the true root of social ills to be concealed, in order that, in turn, social actors can be absolved of moral responsibility provided that they act ethically. He labels this the 'ethicisation of public life ' (2010, 353).
Brecher turns his attention to the interplay between professionalisation and this ethicisation saying: the actual function of professional ethicswhatever the intentions of those who decide their contentis increasingly to render individuals being held to account for failures that are in fact structural and/or political; and disproportionately to lay responsibility at the door of those individuals further down, rather than higher up, the managerial chain. Consider triage situations in intensive care units, for example . . . The problem is brought about by political decisions to spend resources on, for example, replacing Britain's submarine nuclear weapons system, or bailing out collapsing banks, rather than on more and better resources for intensive care units. The 'solution', however, together with the responsibility for handling that responsibility responsibly, is devolved from the political decision-makers to the Trust concerned (it's up to them to decide priorities, within the budget they've been allocated) and thence to front-line staff (because it's a matter for professional medical judgement, informed jointly by the ethics of their professional body and of the Trust itself). One rational, and thus moral, responseand arguably the right moral-political responseis thus ruled out. (2010,353) He concludes that such mystification of responsibility can come to suggest that behaving ethically is some sort of 'add on' to professional roles, rather than a morality being embedded in them unavoidably. It may also be impossible for interpreters to work out how to reach a moral conclusion to problems because of ethical requirements. For example, Lee points out a conflict between on the one hand professional ethics and on the other the need to explain to the extent that one is sure that one's interlocutor understands (2007,245). Similarly Ng notes the potential for ethical codes to deter interpreters from clarifying points of confusion with speakers (2013, 602). Norma and Garcia-Caro go further by exploring the potential of ethical codes to disenfranchise women and children through a misguided idea of impartiality (2016,1313). Ultimately, such codes might restrict the interpreter's agency (Määttä 2015, 30) making truly moral conduct impossible. In all of these instances, communicative practices clash with ethical requirements.
There is, I suggest, a useful extrapolation to be made here. Interpreters' codes of ethics undoubtedly individualise responsibility with the interpreter whilst ruling out critique of certain assumptions such as those around translanguaging. Thus, for example, the interpreter must take responsibility for some measurable and observable form of communicative success whilst predictable 'oddities' such as translanguaging are to be explained away, rather than recognised as inevitable or even discussed and perhaps ultimately valued. Additionally, 'provision of an interpreter' is itself an ethical step for other professionals, such as police officers and doctors, which individualises responsibility and limits response: First, it places responsibility for both facilitating and accomplishing communication into the hands of specific, other individuals thus divesting responsibility from themselves. This is evidenced by its redress, training materials for those who work with interpreters aiming to improve their skills in that task (e.g. Cambridgeshire Police 2017); secondly, it removes the potential for the non-linguist professional to undertake forms of communication which might serve additional social functions and to provide alternative forms of support or intervention. In this way, the interpreter can be positioned as the whole and only adjustment needed to facilitate communication; the ethical box of compliance with procedure has been ticked and the other professional might see no further attention as being needed from them. Rudvin too makes the distinction between moral and ethical in discussing interpreting specifically when she observes 'being truly ethical is beyond the call of duty and is related to a wider moral sphere of human conduct ' (2007, 51). For her, this dimension is crucial to understanding what it means to be a professional interpreter. The ethical traps outlined in this paragraph are only addressed through moral frameworks.
García and Li Wei note that 'all languaging is enmeshed in systems of power, and thus, can be oppressive or liberating, depending on the positioning of speakers and their agency ' (2014, 8). Translation and interpreting are rooted in particular positionings of speakers and agencies, and the resulting systems have effects on all participants. They are part of systems which give speakers 'differential access to forms, to linguistic/communicative resources, resulting in differential capacities to accomplish certain functions' and 'differential access to contextual spaces, that is, spaces of meaning ratification where specific forms conventionally receive specific functions, resulting in differential capacity to map forms onto functions, in other words, differential capacity to interpret' (Blommaert 2005, 76). In the case of interpreting, these differentials relate to who gets to speak, read and write, in what language, how, when and to whom.
Whilst incorporating translanguaging into interpreting practice would not be something to advocate without extensive research, recognition of the communicative activities which occur beyond interpreted encounters would only seem liable to improve those encounters for all involved.

Legal applications
Turner problematises the judicial system's use of interpreting saying that 'the system is full of holes ' (2001, 126). He notes that some critics of such systems believe that 'the presence of an interpreter is actively prejudicial to the interests of the minority language-using defendant' and should be replaced by trials conducted in the language of the defendant (2001,126). This would be one logical conclusion of work on multilingualism in society, if an impractical one, certainly in England and Wales. It would not, however, work for trials involving multiple defendants who bring different languages nor for the obligations of the court in relation to witnesses who may not speak the language of the court or the defendant, for example.
The relevance of the study of translanguaging in superdiversity to interpreting is illustrated by the insightful work of Angermeyer (2015) although he largely avoids these terms themselves. He observes that in the face of the increasing diversity of languages and peoples participating in legal systems in the global north, English-based courts have turned to interpreting as the default solution. This is based, he says, 'on an assumption that persons who speak another language can participate in proceedings to the same extent as English speakers, as long as they are assisted by a qualified court interpreter who translates accurately between the languages ' (2015, 4). He notes that this has led to forms of research which prioritise better training for interpreters and examine good practice to inform that training. He suggests that there has been a relative paucity of research on sociolinguistic contexts of interpreting and the pragmatic contrasts between talk mediated by an interpreter and that which is not.
As Gibbons has noted, in legal contexts where people who do not speak the language of the court need to be heard and to hear alongside those who do not, 'providing the same treatment does not equate with providing just treatment ' (2003, 202). The question of what, in a moral framework, amounts to just treatment are continuing to crystallise. It appears that the answers are not going to be the same for all speakers. The findings of research on sociolinguistic, discursive and pragmatic aspects of multilingual practice are continually adding to the picture of interpreter practice and that of wider societal multilingualism as well as their relation to one another. In 1995, Cooke described an inquest in Elcho Island, Darwin, in which many Aboriginal witnesses were called on to testify. Cooke points out that the witnesses were given the assistance of an interpreter on an essentially arbitrary basis. He catalogued the circumstances which led to an interpreter being permitted to act at that inquest. The circumstances predominantly centred on a breakdown of communication from the point of view of courtroom personnel, specifically counsel and the coroner's assistant. However, sometimes interpreters would begin to do their work simply because they happened to remain in the witness box after a previous witness and, occasionally, at the witness' request. Cooke's point is that this unmotivated disarray of reasons for the interpreter presence is unacceptable. The point is reinforced by others such as Berk-Seligson who problematise the use of interpreters whose involvement is essentially chance (e.g. 2009, 2011). However, what is disarray in one instance can become principled, selective, responsive intervention in another. Angermeyer points out that interpreting which incorporates a 'standby mode', in which the interpreter can stop interpreting when appropriate, offers many advantages. He notes that interpreting can make testimony 'fragmented' and concludes that: 'the rights of litigants would be better served if courts adopted a more flexible approach to court interpreting rather than taking an all-ornothing approach in which litigants are discouraged from testifying in English' once they have used a different language (2015,139). He illustrates that such an interpreting mode can be sensitive to participants' needs and facilitates communication by providing language advice as litigants testify for themselves rather than necessarily interpreting on their behalf throughout. Whilst, as Angermeyer notes, such interpreting can seem reminiscent of nonprofessional interpreting such as that involving language brokering (2015, 140), he suggests training in knowing when and how assistance is needed would reduce this association. He cautions that the use of such a practice depends on context and the interactional proficiency of those involved (ibid., 141). Gottleib is critical of what she calls an 'ideology of proficiency', in other words an assumption that speakers are proficient, which is embedded in standby interpreting. She illustrates that this mode of interpreting is a complex construct ripe for further study (2012). My point is not that interpreting practice should necessarily change to a standby mode or to a mode driven by translanguaging. It is instead that as interpreters review their ethical and indeed moral position, consideration of the character of multilingual talk, as it is currently understood is an important component both for them and, particularly, for the institutions and individuals who employ them.

Conclusion
This paper has problematised the monolingual norm, the expectation that speakers will be monolingual and that even when they are not, society should create ways of handling situations as if they were. In exploring the notions of translanguaging and superdiversity as they relate to social change in England and Wales, this paper has considered the challenges these notions pose to the monolingual norm. The paper then introduced Linguistic Ethnography as a means of looking at these changes and considering their functionality in social contexts. The paper described one study of multilingualism using Linguistic Ethnography and showed how the work provides insights into multilingualism beyond interpreted encounters. It then presented a sample of data from the project to show how a view of languages as separate codes can be unhelpful to our understanding of the ways in which people make meanings together in conversations. The final sections of the paper introduced the contrast between a moral and ethical approach to multilingualism by noting that current systems for managing multilingual encounters in institutions can be more ethical than moral. This may appear to be simply an inevitable matter of practicalities but nothing is inevitable here. The discussion closed by reviewing some recent scholarship on modes of interpreting other than the dominant one, not necessarily to recommend those modes but to explore the way that the monolingual norm permeates interpreting and to consider just one example of a form of professional practice which does not do this, standby interpreting. The recommendation is not that interpreter practice should change, moving towards more translanguaging, for example, but that those who work in fields of law, business, healthcare and so on might usefully be made more aware of the reality of multilingual talk as a means to help interpreters and translators to do their work in a more honest and thus ethico-moral environment. As Rudvin observes, the 'norms and ethical guidelines' which govern such activities as interpreting 'are often created by whatever the centre of decision-making power happens to be' (2007,48). If the centre of power holds a monolingual norm that will influence practice very directly. Ozolins adds that 'interpreters are not always masters [sic.] of their own fate, and it is important to see the influence of institutions and enterprises in defining their sphere of action ' (2014, 37). As well as seeing this influence, we might usefully seek to change it through education and information. Giampapa (2016) takes up Ortega's (2005, 433) notion that research is 'always value laden' meaning that researchers should 'think about the social responsibility and the impact of their work'. She specifies that the researcher influences 'what stories get told' and how, as well as how they are interpreted (2016,292). As Marselis notes, the re-telling of private stories in such public fora requires care to avoid, for example, co-opting individuals' narratives (2013,381). If we are to examine the social responsibility of translators, interpreters and those they interact with, we must begin by considering the social responsibility of the academic disciplines which (seek to) inform and support them. In this paper, I have argued that in order to renew the focus on ethics in translating and interpreting, scholars and practitioners need, together, to raise the profile of the complexity of forms of communication which take place, authentically and spontaneously in superdiverse society. By recognising and displaying this complexity amongst those who may not have looked beyond the monolingual norm themselves, the need for, place of and significance of translation and interpreting can be better appreciated. Recognition of translanguaging fundamentally changes our understanding of what happens when communication relies on more than one language. Even if professionals who work with interpreters and translators know several languages, I have suggested that the monolingual norm might constrain their appreciation of the tasks taking place. This process of recognising and informing others about the complexity of multilingual language practices is not to initiate any change to translating and interpreting themselves, necessarily. Rather this turn can facilitate better understanding of the complexity of the role of the linguist in institutional context for those who work with them in order to produce interpreting contexts which are rooted in an honesty about the tasks at hand. To draw an analogy, Rudvin (2007) has lobbied for increased awareness of the different understandings of what makes a professional interpreter which are in play in different places. Such increased awareness might help the Norwegian police officer to understand that the Chinese witness or suspect may not share their expectations of the professional interpreter being an 'impartial helper' instead seeing professional practice as entailing them being 'an aid' who might help them to reach their goals in negotiation (2007,64). At this higher level too, awareness can transform practice.
Assumptions about the benefits of language separation can unnecessarily stigmatise those for whom translanguaging is a normal part of everyday life and can lead to erroneous assumptions about their capacity for full and active participation in encounters across the social scene. The process of considering language use beyond institutions in order to review language use, and assumptions about language use, in institutions is potentially valuable. The aim of such a review might not necessarily be to change practice but to reposition it. In this way, lay knowledge in legal systems, such as that indicated in the excerpt from Jacquemet at the beginning of this chapter and illustrated in the vignette there, might be addressed.