“When I speak people look at me”

Deaf people’s lives are predicated to some extent on working with sign language interpreters. The self is translated on a regular basis and is a long-term state of being. Identity becomes known and performed through the translated self in many interactions, especially at work. (Hearing) others’ experience of deaf people, largely formed indirectly through the use of sign language interpreters, is rarely understood as intercultural or from a sociocultural linguistic perspective. This study positions itself at the cross-roads of translation studies, sociolinguistics and deaf studies, to specifically discuss findings from a scoping study that sought, for the first time, to explore whether the experience of being ‘known’ through translation is a pertinent issue for deaf signers. Through interviews with three deaf signers, we examine how they draw upon their linguistic repertoires and adopt bimodal translanguaging strategies in their work to assert or maintain their professional identity, including bypassing their representation through interpreters. This group we refer to as ‘Deaf Contextual Speakers’ (DCS). The DCS revealed the tensions they experienced as deaf signers in reinforcing, contravening or perpetuating language ideologies, with respect to assumptions that hearing people make about them as deaf people, their language use in differing contexts; the status of sign language; as well as the perceptions of other deaf signers about their translanguaging choices. deaf people draw upon their multilingual and multimodal repertoires, contingent and situational influences on these choices, and extend our understanding of the relationship between language use,


Introduction
Deaf signers' lives are frequently intertwined with interactions via sign language interpreters. The 'deaf self' (Young, Napier and Oram forthcoming) is translated on a regular basis 1 and is a long-term state of being. In many professional and public contexts where interpreters are provided, identity becomes known to others and performed through the translated self in many interactions with hearing people where those hearing people do not sign, especially at work. Yet (hearing) others' experience of deaf signers, largely formed indirectly through the use of interpreters, is rarely understood as intercultural. Interactional, situational and performative understandings of deaf culture(s) 2 have been explored, where relationships that deaf people develop with each other are based on mutual understanding of 'same-ness' (Ladd 2003;Friedner and Kusters 2015). However, being translated as a regular feature of deaf signers' lives, whether by choice or not, has not previously been considered as a component of the shared experience of deaf cultural identity nor cultural formation.
The translated self creates an epistemic dependency -deaf people are known to others through another person, even though they are present -and an epistemic unknowability -they are being represented, seen and known through another (Young, et al, forthcoming, in press) but does that match their sense of who they are? And how do they feel about their translated self?
The Translating the Deaf Self project 3 was a qualitative, multi-method study that explored the real-life experiences of deaf signers, interpreters and hearing colleagues about deaf people being known through translation, and specifically through sign language interpreters. The current study highlights three deaf professionals' perceptions of the strategies they use to ensure that they are represented in the workplace, and when and how they choose to work with interpreters. This is part of a broader, complex nexus of topics, including: how interpreters feel a sense of responsibility to represent deaf signers' identities and that this is predicated on trust (Napier, et al. in press); whether hearing professionals actually feel that they know their deaf signing colleagues through interpreters (Young, et al, in press); and the ontological insecurities that deaf signers can feel about only being known through translation (Young, et al. forthcoming).
The ready availability of a workforce of sign language interpreters that is much larger than ever before in some countries (de Wit 2016), combined with legislation supporting deaf people's rights of access to interpreting and translation services (Napier 2011) and the emergence of the deaf professional class in some contexts, combine to put on the agenda a new question, that is the concern of this paper. Namely, how do deaf professionals perceive the significance of being translated, as it impacts how their professional identity is represented and enacted? What does it mean to be represented well?
What is the interaction between representation of professional identity and representation of deaf identity in the workplace? And how do deaf professionals account for the complexities of their translated selves in emergent, unpredictable professional settings?

Representation and professional identities
We consider the notion of identity through sociocultural and sociolinguistic theoretical lenses.
Due to the interdisciplinary nature of our work, we draw specifically on sociocultural linguistics to inform our understanding of representation and identity, which sees identity as something that is dynamic and constantly evolving, and is context bound (Bucholtz and Hall 2004). The ways we see and represent ourselves influences how we communicate, what we communicate about, how we communicate with others and how we communicate about others. Hence, identity, representation, culture and difference are all central to communication (Tajfel 1978;Howarth 2002). So essentially we enact and present different identities in different contexts, and others perceive our identities, in part, through the sociolinguistic choices we make on both macro and micro levels (Marra and Angouri 2011).
Professional identity is defined as a person's professional self-concept based on attributes, beliefs, values, motives and experiences (Ibarra 1999). Slay and Smith (2011) suggest that professional roles are typically defined as prestigious and provide the role holder with autonomy, so people whose racial, gender, or other identities are stigmatized are not accorded the same level of prestige and/or privilege because their identities are 'tainted'. In sociolinguistic terms, professional roles and aligned identities are actively constructed and socially produced (Sarangi 2010;Marra and Angouri 2011;Jones and Sin 2013).
Professionals represent their work-related identity through their language choice, whether that is speaking with the right amount of authority, asserting power, using politeness, using appropriate terminology or jargon, and generally adhering to principles for culturally acceptable ways of talking in a specific workplace (Holmes and Stubbe 2015).
Deaf professionals may have a 'tainted' identity if their hearing counterparts focus more on their deaf identity, rather than their professional identity; and this will also be influenced by language choice. Although deaf professionals who are signers may utilize a range of strategies to communicate directly with their hearing counterparts (cf. Kusters 2017), this is more likely to happen in informal contexts (i.e., water cooler talk). For more formal work-related communication, such as meetings, it is more likely that they will work with interpreters, and therefore have to represent their professional identity to non-deaf, nonsigning colleagues through an 'other'. The extent to which deaf professionals may feel that their professional identity is compromised by the fact that they are deaf is under-researched, in part because of the only recent emergence of a deaf professional class.

Deaf (professional) identities
Earlier notions of development of a single undifferentiated 'deaf identity' (Glickman and Carey 1993) are changing. More sophisticated discussions now recognise the 'fuzziness' of deaf identities (Leigh 2009;Young and Temple 2014;Napier and Leeson 2016). The complex intersectional nature of deaf identities means that deaf people interact also based on their gender, race, sexuality, socio-economic status (Valentine 2007). They acquire sign language in different ways. Some deaf people are more transnationally mobile than others. Linguistic and cultural diversity exists within signing communities. And deaf people like other populations may enact and be ascribed more than one identity depending on the context in which they are interacting with others (Bat-Chava 2000; Young and Ackerman 2001;Skelton and Valentine 2003;Breivik 2006;De Clerck 2010;Friedner and Kusters 2015). This same principle is also being applied to discussion of deaf signing communities rather than one deaf community (Kusters, O'Brien and de Meulder 2017).
Another aspect of deaf identities that is changing, particularly in the Global North because of greater opportunities to access higher education, is synonymous with the emergence of a deaf 'middle class' (Padden and Humphries 2005); a sub-group within signing communities who work in professional roles (de Meulder 2007). These deaf professionals work in a range of different roles, such as managers, academics, and lawyers (Hauser, Finch and Hauser 2008) and regularly work with sign language interpreters in the workplace to translate their interactions with their hearing counterparts (Dickinson 2014;Miner 2017). Thus, in these contexts, sign language interpreters represent their deaf and professional identities or 'deafand-professional' identity, depending on your point of view. This is an argument wellrehearsed in the literature on being deaf and from a minority ethnic or indigenous community, and whether identities are separated ('deaf' and 'ethnic/ indigenous') or combined ('deaf-andethnic/indigenous') (see for example Atkin et al. 2002;Smiler and McKee 2007).
In his study of deaf ideologies and perceptions among deaf people of identity and belonging in signing communities, Ladd (2003) proposes a self-actualising process of 'becoming deaf'; what he terms 'Deafhood'. Drawing upon critical theory and Gramsci's (1971) concept of the 'subaltern' as social groups who are at the margins of a society, Ladd defines the 'deaf subaltern': the community of 'grassroots' deaf people who have sign language as their first or preferred language who are on the margins of the majority society of people who hear and speak. He also identified another group: the 'subaltern-elite', who are typically deaf people who have greater facility with signed, spoken and/ or written languages and take on professional roles, and draw on their subaltern experiences as signers, in order to develop new career opportunities, such as teaching and researching sign language; essentially a deaf professional middle class. One of the central ideological elements of Deafhood is embracing the collectivist values of deaf communities that rely on shared experience and people 'not sticking out' (de Meulder 2017). Thus, the emergence of a subaltern-elite has not necessarily been welcomed with enthusiasm by subaltern members of deaf communities (de Meulder 2017), as there is a perception of disloyalty and that people joining the middle class are distancing themselves from the values of the subaltern group. This notion of 'DEAF-SAME' Kusters 2014, 2015;Kusters, de Meulder and O'Brien 2017) leads to disappointment when the deaf subaltern perceive that the subaltern-elite are behaving differently and further can lead to divides between deaf professionals and their wider community caused by an "increasingly competitive environment in which deaf professionals had to 'compete' against their hearing peers" (de Meulder 2017, 116). Therefore, we see a tension between deaf people taking on professional roles, wanting to succeed at work and having to work with interpreters to navigate mainstream (hearing) workplaces, and the perceptions of other deaf people about them and how they manage their professional roles and their interactions via interpreters in order to ensure that their deaf-andprofessional identities are appropriately represented. This tension led us to consider the following specific research questions: • How do deaf professionals perceive that they are represented through interpreters?
• What strategies do deaf professionals use to enact their deaf-and-professional identities?
• How do deaf professionals work with interpreters in order to project their deaf-andprofessional identities?
• How do deaf professionals feel they are perceived by others in language choices and projection of their deaf-and-professional identities?
Until our own study, there had been only one study that attempted to delve deeper into the question of representation and how the development of one's identity is understood through interpreted discourse. To measure the impact of the interpreter's presence, Feyne (2015) looked at how museum personnel rate and evaluate interpreted gallery talks that were presented by deaf guides. The deaf guides were well established in their practice and had developed experience in delivering museum talks in American Sign Language (ASL) to deaf members of the public. All of the evaluators were hearing, had no knowledge of ASL and were selected based on their background as trainers for various museums. The evaluators could only make a judgement of the deaf guides' abilities based on the quality of the spoken English interpretation. The guides, using ASL, used appropriate terminology in that language, but Feyne found a misalignment between an understanding of the interpreter's task and the recipient's assumptions about the way interpreters' function.
Essentially, Feyne suggests that no matter how much a person tries to convey their own identity, it is the recipient of the message who assigns identity (credible or not, genuine or not, knowledgeable or not). It is not clear in Feyne's study whether the interpreters were chosen by the deaf guides themselves. As we know that choice of interpreter and familiarity with the interpreter are important factors from the perspective of deaf professionals (Haug et al. 2017;Miner 2017), any lack of choice or familiarity between the guide-interpreter pairs may have contributed to the misalignment.
This study positions itself at the cross-roads of translation studies, sociolinguistics and deaf studies, to specifically discuss findings from our scoping study that sought, for the first time, to explore whether the experience of being 'known' through translation is a pertinent issue for deaf signers. Ours is the first study to examine how deaf professionals feel about projecting their identities through interpreters, and therefore how they feel they are represented through translation, whether they feel that their deaf-and-professional identities are known by others through translation; and how they feel that other deaf people may perceive them in their use of interpreters in order to project their identities. However, our focus is on deaf persons' understanding of their experiences, and not on an interactional analysis of interpreted situations. 4 In order to examine how deaf professionals might seek to ensure that their identities are represented as they would wish through interpreters, we can consider the sociolinguistic theory of translanguaging as a framework for analysis.

Translanguaging
Earlier models of bilingualism referred to a form of language contact known as 'code-switching', where linguistic features from one language are transferred to another at an intraor inter-sentential level, and the bilingual in question makes conscious and deliberate decisions to do so according to the context they are in and who they are conversing with (Clyne 2003).
'Translanguaging' is a newer framework that proposes a different sociolinguistic understanding of this phenomenon through the lens of multilingualism (Creese and Blackledge 2010, García and Li Wei 2014), and is defined as: "using one's idiolect or linguistic repertoire without regard for socially and politically defined language labels or boundaries-in order to make sense, solve problems, articulate one's thought, and gain knowledge" (Li Wei 2016, 4, our italic emphasis).
Translanguaging practices are fluid: individuals have one linguistic repertoire rather than switching between or mixing different languages (or dialects), because previous references to code-switching (or code-mixing) have been based on the assumption of full knowledge of named languages that are then mixed. Translanguaging "encompasses codeswitching, but entails a wider set of practices and use of resources" (Kusters et al. 2017, 4), but it is fundamentally different from code-switching because it embraces both the conscious and unconscious language strategies used by people when they communicate with others. Creese and Blackledge (2010) argue that every interaction creates a new linguistic reality, and some language users, for at least some of the time, hold beliefs about the importance of language use in relation to their sense of identity. So people may use translanguaging in order to project a particular identity, or to ensure that their identity is represented or to respond to particular contextual features of the interaction, for example when someone might switch from formal spoken language to local dialect of the same language in order to solve a problem of miscommunication or misunderstanding An older body of sociolinguistic research has explored how deaf signers code-mix between a spoken and a signed language (e.g. Lucas and Valli 1992;Napier 2006;Emmorey, Borenstein and Thompson 2008), where English words are mouthed on the lips or manually coded (fingerspelled) while the signer is still using linguistic features of the signed language (e.g. spatial mapping and visual metaphor).
However, explorations of multimodal, multilingual languaging practices draw attention to the wide range of linguistic and semiotic repertoires available to both deaf and hearing signers . Multimodal translanguaging has been explored as a strategy used to scaffold learning among young deaf students (Swanwick 2016(Swanwick , 2017; between adult deaf and hearing people in university lectures (Napier 2016a;Holmström and Schönström 2017); and between spoken and written forms of Hindi, Marathi, English, Gujarati, sign language, and gesture between deaf and hearing people in Mumbai (Kusters 2017). To our knowledge, there has never been any previous consideration of how and why deaf signers engage in translanguaging when working with sign language interpreters, and how that is perceived by other deaf people. Thus this study provides a unique perspective on translanguaging, deaf studies and translation studies. This paper specifically discusses findings from a set of interview data with three deaf professionals who draw upon their linguistic repertoires and adopt bimodal translanguaging strategies in their work to assert or maintain their professional identity including bypassing their representation through interpreters. This discussion of deaf professionals' management of the representation and projection of their deaf-and-professional identities examines how this group of deaf people draw upon their multilingual and multimodal repertoires, contingent and situational influences on these choices, and extends our understanding of the relationship between language use, ideology, identity and representation.

Method
communicative strategy for them (which became apparent from the interview content).
Data in all cases were kept in the source language of BSL for the purposes of data analysis and coding. The BSL was not translated for analysis, as there are inherent problems in representing the deaf 'voice' in BSL to written English translations where nuances may not be adequately represented (Stone andWest 2012, Temple andYoung 2004), but excerpts were necessarily translated as illustrative quotes for the presentation of results in this article.
The data was thematically analysed using the N-Vivo 10 sort and retrieve CAQDAS programme. Only excerpts required for quotes in the write-up of the results were translated.
The broader codes focused on general themes of representation and identity, with subthemes coding for translanguaging practices and deaf ideological stances. All the data were co-analysed by two of the interpreter researchers from the team together (Napier and Skinner) and then confirmed by a third from a deaf perspective (Oram).

Results
Given that we are only reporting on interviews with three deaf professionals, and the British deaf community is intimately interconnected through social networks, we have removed any identifying information about the participants when presenting excerpts from interviews, in order to preserve their identities and avoid any potential identification of our participants, and all quotes are presented anonymously. First we present the broader themes of representation and identity, before we concentrate on the themes of translanguaging and deaf ideological stances.

Representation through interpreters
For all of the participants, the role the interpreter plays in ensuring that they are represented well to others in the workplace was important as well as the role of self-representation. For them, the emphasis was less on the interpreter as the conduit to full participation and more on the skills of the interpreter to represent them well, and also how difficult it is to secure an interpreter that does know them well. An example can be seen in the excerpt below.
With skilled and experienced interpreters, you can't always get them because you have to book one year in advance, you feel confident. They know you and they know your background.
Furthermore, the participants confirmed that the benefits of working with an interpreter that is familiar with them is that things go more smoothly. The participant in the excerpt below alludes to the fact that interpreters who do not know them as well may cover up if they do not understand something. Knowledge of who the interpreter is, and knowing how familiar the interpreter is with them, was linked to levels of confidence about how they are represented. The participants also commented on the benefits of working regularly with the same interpreter, as it makes them more comfortable: If the identity of the interpreter was unknown or out of the control of the individual then s/he was less likely to feel confident in the situation where an interpreter was needed.

… if it's with an interpreter I'm familiar with, they know how to interact and
This was anxiety before the fact, influenced by previous experiences, not judgments made afterward. It was fundamentally linked to the anticipation of the unknown (who is the interpreter, will they be good, and how well do they know me?) and to the lack of control over the means of how they might be represented in the given communicative context. The feeling might be considered akin to someone else choosing the clothes you are going to wear, knowing in advance that people will make assumptions and form judgments on how you look when you have not been able to control how you wish your identity to be portrayed through your clothes in any given context.
The deaf professionals expressed anxieties about individual interpreters, the quality of interpreters, whether interpreters had been booked and who booked the interpreters. When asked about their general experience of communicating with hearing people via an interpreter each of the participants typically referred to workplace interactions. Each participant has regular access to interpreters through the UK government sponsored Access to Work (AtW) scheme 7 . Managing their own AtW budgets and personally arranging their interpreting needs meant they reported a considerable level of control, autonomy and choice about with who acts as their interpreter. Nevertheless, the interviews still touched on instances where they had no control over who the interpreter was, and the potential anxiety this created (based on previous experiences), fundamentally not just about whether they would be represented well but whether they would have the opportunity to be represented well, as in the example below about managing turn taking and participation: Participants described "relief" when they knew the interpreter who had been booked based on their positive pre-knowledge of his/her ability to match their requirements. In discussing how anxiety might be provoked when interpreter identity was unknown and how confidence might be impacted by the identity/skills of the interpreter, participants were likely drawing on previous experiences of feeling let down, and were not discussing just whether the interpreter was adequately representing them in linguistic terms. Rather they were pointing out that anxiety and lack of confidence in the interpreted situation would impact on how they as individuals were able to represent themselves -would they feel confident to discuss subtle, complex or distressing matters if they were unsure of the interpreter? Would they feel able to challenge, discuss or debate with the hearing other if the anxiety they had experienced prior to the appointment had robbed them of self-belief or confidence? In other words, the translated self was also about creating the conditions in which the individual was able to portray themselves with confidence, it was not just about the skills of the interpreter, and the interpreter's familiarity with them, but the conditions surrounding interpretation could impact on that confidence, as noted by one participant:

Professional identity
The professional workplace was often seen as a space where one must be able to demonstrate ability and independently fulfil what was professionally expected. Working with an interpreter to facilitate communication was seen to potentially threaten this image. As professionals they sought to assure their own full participation and took responsibility for that, part of which was exercising control over the choice of interpreter who knew them well enough to represent their language, characteristics, identity and message. The ability to manage individual AtW budgets and choose who acts as your interpreter was an influential factor in this regard. One participant explained how they need to decide what is important and how they (and not the interpreter) need to be involved in order to ensure that they achieve full participation: Thinking about different interpreters, it really gets to me when a video, meetings or in the minutes you find the interpreter has omitted something, they have decided it's not relevant enough to interpret. It may be relevant to me but not to you (the interpreter).

Please interpret what the person says in full, don't reduce it. With interpreters who do
give what I need, it makes me happy because I can be treated the same as everyone else and I'm involved in the meeting.
The participants valued and were passionate about their careers, but they felt they have expertise that is not always properly understood by their colleagues, and therefore they did not feel valued in return. They felt that their interlocutors' attitudes towards deaf people generally meant that their deaf identity seems to obfuscate their professional identity, but the deaf identity is not necessarily valued, i.e., their identity is tainted: Participants also noted that interpreters often gain more attention than they do as the professional in any given context, which makes them feel that they are not recognised, and certainly not known, as shown below: As noted by another interviewee:

When I attend training its the same experience, teachers chatting to the interpreter. It often depends on the interpreter, not all interpreters will be aware of their boundaries and some know not to exclude me. Some interpreters do exclude me. I've become so used to it… I feel like people see me as some "moaner." Sometimes I think "let it go."
If it is about a client then there I make absolute certain they are included. If its just the three of us... well… What usually happens is the interpreter will explain "oh they're asking me about how to become interpreter, how long have I know or how did I get in to..." I think well include me! If you include me then I can be part of this conversation.

It's difficult.
In short, the ability of interpreter provision indirectly and unintentionally to obscure the identity, participation, and contribution of a deaf individual is apparent. It was also largely accepted by deaf professional participants as a fact of life. It just happens. Whilst it may be annoying, it came with the territory of communicating with hearing people with an interpreter.

Translanguaging as a strategy
Our data revealed that signers draw upon their linguistic repertoires and adopt bimodal translanguaging strategies in their work to assert or maintain their professional identity, including bypassing their representation through interpreters.
Lack of trust in the interpreter, either because they have a lack of technical (linguistic) skills, and/or because they lack adequate interpersonal, social (affective) skills, or lack of familiarity with them, was seen by participants as contributing to poor quality interpreted situations. Participants gave examples of where they felt they had to monitor the interpreter, if they could, by lipreading the spoken English interpretation of what had been signed in BSL.
The participants tended to combine the two aspects (linguistic and affective), talking about monitoring the interpreter's skills and ensuring they are represented well, which sometimes means speaking for themselves.
Having to 'battle for attention' was a reason cited by deaf professionals that they would choose to use their own voice, instead of relying on the interpreter to render their utterance in BSL into spoken English. One participant explained how using their own voice at meetings assisted with maintaining focus, keeping those involved in the meeting focused on them in their role as chair of the meeting:

When I speak people look at me. When the interpreter speaks they look at them… I feel like saying "hello! I'm still here!". I'm trying to chair a meeting but I don't have people looking [at me].
The rationale to speak can also be about relaxing and not having to monitor the interpreter (through lipreading) to see if they are producing a correct rendition into spoken English:

It can feel a bit disruptive if people are not concentrating their eyes on me. Also, if I'm honest, I have to be honest here, you don't need to monitor the interpreter for accuracy.
It was also noted that the decision to switch into English can play to the advantage of the deaf person and redress any imbalance. In the example below, the sudden switch to English by the deaf professional was unexpected by hearing others and the dramatic impact helped with getting the point across and becoming heard:

Like, if someone interrupts and says "I'm not happy with that!" I might respond [in BSL] 'I disagree' but the message doesn't get through to them in the spoken interpretation. I can then repeat myself 'I disagree' using my own voice and then it has a stronger impact. I use my voice politically and strategically to focus attention on the key issues. Sometimes through the interpreter and sometimes not.
These examples from deaf professional participants reveal how critically they think about how their self is represented, either as a person or as a professional (or both), and how they have the advantage of their bimodal bilingualism by either using their voice or lipreading (or both) to control the way they are perceived. Therefore the process of translanguaging, and choosing to work with the interpreter or not, becomes a strategic tool for them to solve potential communicative problems, and overcome any tainted perceptions of their identity.
For example, at times it may be more important for them to be seen as a professional, or to be perceived in role, and the fact that they are a deaf signer needs to be disregarded. The fact that they can choose to speak or lipread gives them an advantage that is not afforded to all signers, which they recognised.
This translanguaging strategy can also assist in building relationships with hearing people when perceptions of their 'otherness' (as a person) is seen as a potential barrier: Translanguaging certainly seems to be an effective strategy used by deaf professionals to mitigate for any potential negative perceptions of how they think their translated self is perceived by hearing others. But even with these strategies in place, the participants still report a negative impact on their wellbeing, stress, anxiety or confidence; as shown in the following example:

Ideology, deaf-and-professional identities and translanguaging
The deaf professionals revealed the tensions they experienced as deaf signers in reinforcing, contravening or perpetuating sign language ideologies, with respect to assumptions that hearing people make about them as deaf people; their language use in differing contexts; the status of sign language; as well as the perceptions of other deaf signers about their translanguaging choices.
Interestingly, but not surprisingly, the participants revealed tensions that they experience because of perceptions that other signers have about their translanguaging choices.
The ability to translanguage is not always considered to be an advantage by other deaf people in the same workplace who cannot or choose not to use their voice, or who cannot lipread. In the above quote, it is clear that this deaf professional takes the ideological stance that bilingualism can be equated with his ability to speak, and he thinks that this is valuable.

Sometimes, in some meetings I might be bilingual. Speak and sign
The same participant goes on to say that other deaf people, however, have a different ideology: Unfortunately, I get negative criticisms from other BSL users. I get told "why did you use your own voice?!"" I explain "I used my voice to get the message across". But they still think "sign and use the interpreter to translate!" Then I have to break it to them that the interpreter doesn't always get everything that is said.
This point goes back to the issue of 'full participation' and 'technical' skills. So although evaluation of how interpreters represent professional identity goes far beyond the technical, the technical linguistic skills still very much matter in ensuring full participation.

Discussion
For the deaf professionals involved in this study, the parameters of the translated deaf self are actively negotiated as they work to assert their professional as well as their deaf identity.
Their bilingual status is a key aspect of their identity and they make bimodal linguistic choices strategically in order to manage how their self is projected, translated, represented and perceived.
The relationship that they have with interpreters is a critical component of their perceptions of how their deaf-and-professional identity is represented, and also has the potential to impact on their wellbeing given the concerns they may have as to whether their identities are being (or are able to be) represented. The anxiety experienced prior to, during and after, an interpreted event has recently been documented from a personal narrative perspective by a deaf academic (Burke 2017), and the impact she felt that the lack of a deafcentred approach had on her autonomy as a deaf person and as a professional when attending a conference. This was also clearly evidenced in our own data, but our participants also talked about how they then drew on their linguistic repertoires to make language choices that asserted their agency, and the influence on how they then used the interpretation, and the subsequent impact on how they are perceived by hearing and deaf others. However, it should be remembered that the three participants in our study were reflective about, and conscious of, such strategies.
Given that the ability to translanguage in the case of this population also encompasses an ability to use and monitor spoken language that is not universal amongst deaf signers, the association between the societally imbued power of the human voice and strategies of representation through translanguaging is an important avenue of future inquiry (see also Young et al. forthcoming).
In acknowledging that identity is context bound (Bucholtz and Hall 2004), and that people employ their 'translanguaging instinct' to go beyond linguistically defined cues and culturally defined language boundaries in order to achieve successful communication (Li Wei 2018), we have coined the term 'Deaf Contextual Speakers' (DCS) to refer to this group of deaf professionals as their language choices are clearly linked to the context and deliberately go beyond language boundaries (i.e speaking as a deaf sign language user). As the data revealed, these deaf professionals strategically use their bimodal linguistic repertoires to move between signed and spoken language usage, and their translanguaging practice is dependent on the context: who is present, the purpose of the interaction, who the interpreter is, and how familiar the interpreter is with them. They use what García (2009) would term 'responsible code-switching' in that they either plan to deliberately, or spontaneously, switch between languages to maximise the communicative interaction or the impact of their message. At the time of writing this article, Australia has a national disability discrimination commissioner who is deaf: The concept of translanguaging is particularly apt for DCS, because the data reveals that the DCS made decisions based on who the interpreter is, the nature of the context, and to solve problems, and so might make different decisions every time; thus revealing their own ideologies and how they get put into practice and sometimes re-worked in concrete situations.
Thus each interaction creates a 'new linguistic reality' between BSL and English, by recombining resources and transforming repertoires, and deciding when and how to have the interpreter be their 'voice'.
The DCS in our study also revealed the tensions experienced as signers in reinforcing, contravening or perpetuating language ideologies, with respect to assumptions that hearing people make about deaf people, their language use and the status of sign language; as well as the perceptions of other deaf signers about their translanguaging choices. This relates back to Ladd (2003) and de Meulder's (2017) discussions of the relationship between deaf subaltern and subaltern elite groups and the clear tensions with regards to deaf professional status. Ladd and de Meulder both discuss the perceptions of deaf people becoming professional and distancing themselves from signing community collective values. In our data, however, the tension was evident because the DCS chose to speak, make bimodal translanguaging choices, and change the way they work with interpreters dependent on what they want to portray and to whom, which may not be an option for other deaf signers.

Limitations of the study
Before we conclude, we acknowledge that only having three interviewees involved in the study does not provide satisfactory evidence to make an extensive and in-depth investigation.
Thus we consider this a scoping study, with presentation of preliminary findings. The findings may not be generalisable to all deaf professionals who may choose to speak in some situations, however we are content that at a conceptual level, the broader considerations we have highlighted have strong relevance that requires further investigation.
We also acknowledge that the interviews focussed on DCS who are deaf professionals, and other deaf signers who choose to use their voice in other contexts may do so for different reasons. And deaf professionals who are not DCS, and do not have the same translanguaging options available to them, might feel differently about how their deaf-andprofessional identities are represented through interpreters, or how they manage the projection of their identities.
As the interviews were conducted by a hearing member of the team who is also a qualified and practicing interpreter, we recognise that this may have influenced what the participants felt they could say about interpreters and interpreting. In hindsight, it may have been more appropriate to have a deaf member of the team conduct the interviews with this group, or a deaf researcher who is also a contextual speaker. Given the nature of the themes that emerged from the data, one of the merits in having a bilingual hearing person conduct the interviews may have been that the DCS participants perceived that there would be less judgment about their translanguaging choices (i.e. using their voice and perceptions of that from other deaf people). This is difficult to know, but any future interviews will be planned carefully to take these issues in to account.
Providing a follow-up opportunity to discuss the questions after more reflection was also identified as something that we should include in future for any individual interviews with DCS. This became particularly apparent when we received a post-interview email from one DCS participant with the following comment:

Conclusions
The title of this paper and the reference to "When I speak people look at me" captures the essential element of three Deaf Contextual Speakers' rationale for why they engage in translanguaging strategies in order to ensure that their professional identity is perceived as intended, and they are ascribed with the attention they feel they deserve in their professional role.
In revisiting our research questions, this scoping study confirms the following; that Deaf Contextual Speakers who work in professional roles: • perceive that their deaf-and-professional identities are not always adequately represented through interpreters, and this can create feelings of stress as it can impact on the way they are perceived by hearing others in the workplace; • clearly engage in bimodal translanguaging practices to manage perceptions of their deafand-professional identity; • strategically decide how and when to utilise spoken English interpretation of their BSL utterances or use their own voice in order to ensure that their deaf-and-professional identities are adequately represented; and • feel that other deaf people do not always understand why, or approve of, their bimodal translanguage practices and their working relationship with interpreters.
Discussion of DCS' translanguaging and deaf-and-professional identities contributes to theoretical understandings of how people draw upon their multilingual and multimodal repertoires, contingent and situational influences on these choices, and extends our understanding of the relationship between language use, identity and representation.
In different respects, this data source has contributed to an emerging conceptualization of the impact of the translated self on wellbeing. The discussion with deaf professionals who are Deaf Contextual Speakers has introduced the potential relationship between expressive communication choices in specific situational circumstances as anticipatory strategies for the preservation of a preferred representation of self. This may be linked to a protective and/or preservative assumption of its power to maintain control of interactive and communicative participation in a manner that does not threaten the person's role, agency and self-esteem.
This preliminary observation requires further examination.
In addition to the research questions outlined in this study, we are keen to continue to explore the notion of 'oral' situational identities amongst culturally deaf people, as we recognize that for Deaf Contextual Speakers, choosing to speak or lipread may be to do with comfort levels not just about power and control and asserting deaf and/or professional identity. Their choice to speak may be predicated on the development of trust and relationships (discussed elsewhere in Napier, et al, under review). We believe that more exploration of language choices and work with interpreters of DCS, and deaf signers who cannot or will not choose to speak is needed as a largely untapped area of discussion.
We also believe that further exploration of notions of the translated deaf self, and the relationship to deaf cultural practices is needed, in order to develop a deeper understanding of the deaf lived experience, and the role of interpreters in the everyday lives of deaf signers. Notes 1. We use the term translation in the broadest conceptual sense, in relation to the process of translating between two languages and cultures in different modalities, whether that be through the process of written or signed text-to-text edited translation, or via spoken or signed live interpreting (Munday, 2013). However, for the purposes of this paper, when we refer to translation or a person being translated, we specifically mean sign language interpreting, or being interpreted through a sign language interpreter.
2. Previously it has been conventional in the deaf studies, sign linguistics and sign language interpreting literature to distinguish between deaf people who use a sign language and identify with other sign language users (using a capital 'D', Deaf) and those who have a hearing loss but do not use a signed language or identify themselves with a community of signed language users ('deaf'). Napier and Leeson (2016) and Kusters, O'Brien and de Meulder (2017) note that due to complexities of sign language transmission and the evolving nature of deaf communities due to medical interventions and changes in educational policy, greater numbers of deaf people come to the community as late learners of sign language. As a result, they note that definitions of deaf community membership are changing and they use 'deaf' as a generic term, making no judgment about the hearing and linguistic identity or status of people who use a sign language. Following this principle, the only time we retain the 'Deaf' convention is in relation to theoretical concepts.
3. The Translating the Deaf Self project was funded through a UK Arts and Humanities Council Translating Cultures Theme Research and Innovation Grant (Ref: AH/M003426/1). 4. We are exploring the concept of translation from a theoretical perspective, and not examining specific interpreter-mediated interactions or the role of the interpreter. But we acknowledge that there is an increasingly significant body of work in spoken and signed language interpreting studies. See for example Angelelli (2004), Hale (2007), Valero-Garcés and Martin (2008) and Napier (2016b). 5. In the UK, Sign Supported English (SSE) is a common term used to describe how deaf people use signs from BSL but they are produced in English word order (rather than following BSL grammar), sometimes using their voice at the same time.
6. The research team was made up of one deaf social researcher (Oram), one hearing social researcher (Young), and two hearing interpreting studies researchers who are also qualified, practising interpreters (Napier and Skinner).
7. Access to Work (AtW) is a UK Government funded grant that people with disabilities can apply for to cover the costs of practical support in the workplace. For deaf people, this typically means costs to cover sign language interpreting services.