IJIE Volume 5(2)

Getting to the Core of Role: Defining Interpreters’ Role-Space

IJIE Logo

Volume 5 (2) ~ November 2013

ISSN # 2150-5772 – This article is the intellectual property of the authors and CIT. If you wish to use this article in your teaching or in another format, please credit the authors and the CIT International Journal of Interpreter Education.

Getting to the Core of Role: Defining Interpreters’ Role-Space

Peter Llewellyn-Jones
SLI Ltd., UK

Robert G. Lee
University of Central Lancashire, UK

Correspondence to: p.llewellyn-jones@slilimited.co.uk

Download PDF (422 KB)

1.       Introduction

Much of the writing on community or dialogue interpreting concentrates—perhaps disproportionately compared with other areas of interpreting—on “role” and “role boundaries.” Hale (2007) in Community Interpreting devoted a whole chapter to interpreters’ codes of “ethics” or, more precisely, codes of practice (pp. 101–136); of note, the 16 such codes that she described are all based on prescriptivism and proscriptivism, that is, on what interpreters must or must not do.

Tate and Turner (1997, 2001) set signed language interpreters a variety of problematic interpreting situations and asked them whether they would abide by the letter of the Code of Practice then in force in the U.K., that is, if they would act “mechanistically,” or in a way that they felt was more beneficial to the interlocutors or the interaction, even if it meant not abiding by the strictures of the code, in their terms, “non-mechanistically.” Reacting to the first scenario, a case of a patient clearly misunderstanding the nature (and dangers) of a prescribed drug because the patient and doctor had been talking at cross-purposes, 99% of Tate and Turner’s participants said that they would act non-mechanistically and intervene. Their second scenario was as follows:

 

You are interpreting with a Deaf mother-to-be when she goes for a scan. You know that she doesn’t want to know the sex of her baby, but the gynaecologist suddenly comes out with the information that it’s a boy. What would you do and why? (Tate & Turner, 2001, pp. 57

In this case, 77% of interpreters stated that they would make some kind of non-mechanistic intervention. That those interpreters who opted for the strategy that resulted in the least harm, either for the individual or for the interaction as a whole, described their behavior as “stepping out of role” would be construed as surprising by practitioners of any other field. If the “right” action in the eyes of many of the interpreters required them to step out of role, there is, surely, something wrong with the role as described in, and prescribed/proscribed by, the codes.

Many of the current interpreters’ codes of ethics are simply refinements of those first introduced some 30 or more years ago. They are firmly rooted in the notion that dialogues are to do with thoughts being expressed linguistically by one interlocutor that are then responded to by the other interlocutor(s). In other words, dialogues are collections of short monological contributions that invite or require short monological responses. The intervening 30 years since the introduction of these codes, however, have seen the blossoming of sociolinguistic and psycholinguistic approaches to the study of human interaction, how people behave when communicating and how people jointly construct meaning when interacting. Both Wadensjö (1998) and Roy (2000) drew attention to these developments and their significance for interpreters, but the implications have not permeated the professional and regulating bodies.

In two earlier conference papers (Lee & Llewellyn-Jones, 2011; Llewellyn-Jones & Lee, 2009), we argued for a wholly new approach to the role of the community interpreter, one that reflects the work the interpreter actually does—the interpretation of dialogues—rather than the performance of “language transfer” traditionally associated with the (seemingly) monological work of simultaneous conference interpreters. We proposed that the only way to determine what sort of roles might be appropriate for the dialogue or community interpreter would be to first examine what interactions look like when they do not need interpreter intervention. What happens when two interlocutors who share a common language and culture interact and converse?  Our experience as interpreters and our observations of other interpreters at work convinced us that many of the “dos and don’ts” of the prescriptive/proscriptive codes merely serve to inhibit or denormalize interactions, either by deskilling one of the interlocutors or by introducing, in effect, a third interlocutor who doesn’t “play by the rules.” If we can identify the norms of face-to-face interactions, we argued, we can begin to discuss how an interpreter can be introduced without having an inhibiting or negative impact. Our original hypothesis was that dialogue/community interpreters can only help to normalize, or “oil the wheels,” of dysfunctional interactions between interlocutors who either can’t or can’t easily communicate with each other, by acting “normally.” If, instead, the interpreter who is meant to be the enabler begins by replacing expected, culturally required, introductions with long descriptions of what he or she is about to do or by pretending he or she is not actually there (donning the “cloak of invisibility”), then this only further denormalizes and, hence, inhibits the interaction.

We do not question that the community interpreter, to be effective, must be functionally fluent in the languages he or she is working between. It also goes without saying that the community interpreter must also have a good knowledge of the cultures, worldviews, and likely world knowledge (shared “scripts”) of the interlocutors and an appreciation of the goals of the participants and of the consequences of whether the interaction is meaningful/successful or not—as well as knowledge of the settings/domains they find themselves in. We are convinced that the interpreting field must be regulated so that it can be taken for granted by the users of interpreting services that the interpreter will be competent and act with integrity, that is, offer a professional service that allows the interlocutors to interact in the way that they would if they shared a language. Instead, what we are investigating here is the room the interpreter has to maneuver and the extent of the freedom he or she has to make professional decisions.

To begin to examine these, we looked again at what is known about face-to-face human conversational interactions. As leaders of university postgraduate programs in interpreting in the U.K., we have taught many of the sociolinguistic theories relating to interaction and conversation, for example, the presentation of “self” (Goffman, 1959/1990; 1981), the cooperative principle (Grice, 1975), communication accommodation theory (Giles, Coupland, & Coupland, 1991), politeness theory (Brown & Levinson, 1978/1987) and, with our students, we have discussed how these might influence the way an interpreter approaches a range of dialogic encounters. What we have not done, until now, is put these together with the research findings of psycholinguists— conversational alignment (Garrod & Pickering, 2007) and others—and then overlaid these with the observations and findings of interpreting researchers such as Mikkelson, Wadensjö, Roy, and Hale, among others. In other words, we haven’t yet put all of these facets of human communication together to arrive at a coherent theory or account of the role of the community interpreter.

Now, after a period of focused research and thinking, we propose that the interpreter’s role not be rules-based but be governed instead by the “role-space” the interpreter creates and inhabits in any given situation. This “space” is determined by a range of factors that can be represented along three main axes:

x:       the axis of participant/conversational alignment (sociolinguistic and psycholinguistic)

y:       the axis of interaction management

z:       the axis of “presentation of self”

 

We argue that these three axes define and delimit those areas in which interpreters make decisions and employ strategies to enable successful interactions. By plotting the interpreter’s anticipated/actual positioning on these three axes, we can generate a three-dimensional shape, or space, that delineates the appropriate role-space of the interpreter in any particular interaction. Discourse and conversation analysis have demonstrated that the relationships and roles of interlocutors change during interactions, that interactions are dynamic phenomena. Similarly, interpreter role-spaces adapt and change; they are as dynamic as the interactions they describe. Although most of the examples we draw on to illustrate  our arguments relate to signed–spoken language interpreted interactions, our consultations with interpreters with other language combinations indicate that many of the issues are equally relevant to spoken language community interpreters.

2.       Presentation of “Self”

In his seminal work, The Presentation of Self in Everyday Life, Erving Goffman {Goffman, 1990 #524}(1959/1990) claimed that in all interactions with others, individuals are engaged in a type of performance. He stated that we are constantly involved in “impression management,” that is, acting in such a way as to attempt to control how other people perceive one’s self, with the (obvious) goal of coming across as best one can. This can include how one speaks, dresses, moves, walks, and so forth.

Interpreters are present in interactions to allow people who do not share a language to interact effectively.  There have been concerns that the presence of an extra person in any given interaction (even one who is present so that others can communicate) can have a negative effect. Typically, interpreter training, the interpreting codes of conduct/practice, and even guidelines for working with interpreters have stressed that interpreters must not behave in such a way that their presence as a person affects the interaction, specifically by not expressing opinions or, more strictly, not speaking for themselves. Interpreters have traditionally been taught that, to minimize their impact, they should maintain an impersonal, professional distance; that is, they should have a low presentation of self and should not interact with the interlocutors other than to interpret the meaning of their utterances.

Expressed in Goffman’s (1981) terms, interpreters should only enact the role of the animator of language, not the roles of author or principal. Interpreters are advised to be always neutral, impartial, or in extreme cases—according to the “machine model” (McIntire & Sanderson, 1993)—to not initiate communication nor respond to questions directed to them. These types of behavior exemplify what Goffman (1959/1990) called disruptive events; times when one or more of the parties acts contrary to expected behavior (e.g., by not responding to a direct question).

 

When these disruptive events occur, the interaction itself may come to a confused and embarrassing halt. Some of the assumptions upon which the responses of the participants had been predicated become untenable . . . and all the participants may come to feel ill at ease, nonplussed, out of countenance, embarrassed, experiencing the kind of anomaly that is generated when the minute social system of face-to-face interaction breaks down. (Goffman, 1990, pp. 23–24)

This restriction on which speaker roles interpreters can enact (and, by extension, how much of themselves they can present in a given interaction) is problematic in many ways. The first is that all successful communicative interactions, even interpreted ones, involve the crucial factor of trust among all the participants. (Indeed, even in adversarial interactions, participants can still follow the rules if they are respectful of the other person’s right to communicate, even if they do not agree with the content of the communication. It is this mutual respect that allows people to communicate, yet still disagree.) If an interpreter in an interaction does not present himself or herself in an expected way but, as a demonstration of impartiality, projects a false “professional” self, this can result in inhibiting and deskilling both interlocutors. The interpreter’s unwillingness to cooperate and converge may be read as disinterest in both the participants and their contributions. Worse, the interpreter’s inhibition of conversational responses and feedback—which are seen as essential in the joint creation of meaning—can lead to the interlocutors never being quite sure that they have been fully understood.

These behaviors also increase the likelihood that the interpreter will be viewed as an “auditor” (Bell, 1984), that is, perceived by the participants as someone who is judging the individuals and/or the content of their contributions. Interpreters often state that they must be impartial (nonparticipatory) in order be fair to both sides of the interaction. Rather than be impartial to be fair, we suggest that an interpreter must instead be “bi-partial”; understanding, overtly valuing (through appropriate back-channelling, etc.), and being able to articulate (through faithful interpretation) both points of view.

In simultaneous conference interpreting, the participants may never meet nor, indeed, be able to see the interpreter, so the interpreter’s presentation of self, other than through tone of voice or linguistic choices, will be minimal. However, in face-to-face interactions, the interpreter’s presentation of self takes on much greater importance. When individuals meet each other for the first time, as soon as they begin to communicate, they start to make judgments about each other. These might, initially, be based on accent (where they might be from), use of vocabulary (level of education), use of politeness markers (their awareness of communicative norms), and their willingness to converge (i.e., demonstrate social solidarity). Successful interpreting depends on trust. Interlocutors have to trust that what they are saying (and meaning) is being portrayed in a way that they would communicate it if they could speak the language and, hence, communicate directly with the person(s) they are interacting with. However, whether the interpreter will faithfully interpret what is said is more than merely a matter of trust; it rests on whether the interpreter, as a person, values the interlocutors and their contributions. This is akin to what Mead (1934) and Turner (1956, 1962) referred to as role-taking, that is, the ability to understand and relate to the other person in an interaction. In other words, the interpreter has to be seen as having empathy for the principal participants.

We have stated previously that “interpreters will often speak of ‘stepping out of role’ to rationalize behaviors which, we would argue, are an integral part of the remit of the interpreter (for example in seeking clarification from one or more of the interactants)” (Lee & Llewellyn-Jones, 2011, p. 1), and this is precisely because role is erroneously seen as something one has, rather than something one does; “one may enact a role but cannot occupy a role” (Turner, 1956, p. 317).

We know from our own experience of conversation that apparent disinterest or, worse, ambivalence on the part of the person we are talking to is inhibiting. Wadensjö (1998) highlights this in her example of a midwife and pregnant woman losing their enthusiasm for continuing their conversation because the interpreter did not laugh when they did. They read the interpreter’s failure to engage not as a sign of formality or professionalism, but as a lack of interest in what they were saying. Interlocutors’ contributions to any talk exchange are tokens of their self-representation; they represent who the indivduals are rather than merely what they are saying at that point in time. The instant response of the receiver signals, through words or some other semiotic expression (facial expression, nod, etc.), both that the utterance is understood and that the receiver approves not merely of the spoken utterance but also of the self that it represents. Face-to-face communication requires cooperation. The responses to the contributions of the parties are as important as the contributions themselves. No response is not seen a sign of neutrality, but rather as a sign of disapproval or lack of understanding, an ambiguous but powerful reaction. When interpreting a face-to-face conversation, the interpreter is clearly participating in the interaction and, consequently, must also cooperate. It should be noted that in contrast to spoken language interactions, where signals of cooperation (e.g., back-channelling) can be performed both vocally and visually (e.g., saying “uh-huh” as well as nodding), interactions involving sign language use only visual signals of cooperation—thus, it is imperative for the deaf or hard of hearing participant that an interpreter use such signals.

A common notion of the machine model of interpreting was that interpreters should be “invisible.” This problem has been explored by a variety of researchers (see, e.g., Angelelli, 2001; Nakane 2009; Roy, 1993). In discussing the role of interpreters in workplace settings, Dickinson and Turner (2009, pp. 175–176) reported, “It is still common for SLIs [sign language interpreters] to encourage their clients to ‘imagine that I am not here’ and to ‘just ignore me,’ covering themselves with an imaginary ‘Cloak of Invisibility.’ ” They go on to provide a quote from an interpreter:

 

I was recently accused of being “off- hand” and “frosty” in my Deaf client’s workplace. I was deep in (signed) conversation with the Deaf client, the hearing staff had walked in and I voiced over the Deaf client’s greeting. Because I personally did not say hello to the staff or make eye contact they said I was stand-offish. (Dickinson & Turner 2009, p. 176)

 

This comment is an example of the low end of the presentation-of-self axis; only in very marked social situations are participants said to be “invisible.” Even though the point was for the interpreter’s presence to have a minimal effect, such ways of behaving are alienating to other participants in an interaction.

Interpreters must behave in ways that are consonant with, rather than counter to, the expectations of the participants. By normalizing their own communicative behaviors, and by acting in ways that are similar to the other participants, interpreters can be more effective in facilitating successful interactions. Thus, the first dimension in the role-space model is the presentation of self, the z axis in the 3-D model. The range of behaviors in this dimension range from minimal presentation of self (those behaviors associated with the machine model or “invisible” interpreter, e.g., not interacting with the participants except through interpreted renditions, not answering direct questions, referring to one’s self as “the interpreter,” etc.) to actual, expected presentation of self in ways that are consistent with the situation (e.g., introducing one’s self, referring to one’s self in the first person, etc.).

 

Figure 1: The axis of Presentation of Self

 

 

 

 

 

 

 

 

 

 

 

 

As will be shown with the axis of alignment and the axis of interaction management, the amount of self-presentation by the interpreter is rarely the same throughout the entire interaction. For example, at the beginning of an interaction among people who have not met before, there might be a requirement for more overt self-presentation by the interpreter, because it may be necessary to gain the trust of the participants. By presenting one’s self in a way that follows the expected norms of the interaction, the interpreter allows the participants to become familiar with him or her and this, then, starts to engender trust. This is, in effect, the same strategy that any other professional meeting people for the first time would employ.

3.       Alignment

One of the first “rules” student interpreters learn is that they must be impartial. Hale’s (2007) comparison of  codes of ethics (or, more accurately, codes of conduct/practice) from a range of countries, interpreter organizations, and service commissioners found that the third most frequently included tenet, after confidentiality and accuracy, was impartiality. The problem, of course, is that impartiality or neutrality is rarely, if ever, possible (in addition to Hale, 2007, see, e.g., Metzger, 1999). Hale  argued that, instead, interpreters should strive for “objectivity,” that is, they should not allow their own ideas or religious or philosophical beliefs to color or otherwise impact on their interpretations. From our experience as teachers of postgraduate courses, a significant percentage of students who come to us from vocational and undergraduate interpreting courses are instilled with notions that they have to demonstrate their impartiality by exhibiting behaviors that preclude any action that might be perceived as aligning with the interlocutors. In the normal course of events, this type of “un-giving” behavior would appear extraordinary, so much so that it would inhibit rather than encourage communicative interaction, and it is these behaviors that we need to question.

Wadensjö (1998, p. 8) notes that the conduit/machine model of interpreting is monological, that is, “the meaning of the words and utterances is seen as resulting from the speaker’s intentions or strategies alone” (her emphasis). “The dialogical model, in contrast, implies that the meaning conveyed in and by talk is partly a joint product . . . [that] presupposes a reciprocity between the people involved.” To illustrate, when two people who can speak the same language begin to communicate with one another, they start (albeit tentatively at first) to align. On a sociolinguistic level, the interlocutors will start to accommodate each other (Giles et al., 1991); that is, they will signal their desire to lessen, maintain, or increase social distance. The desire to lessen social distance, or converge, is signalled by the interlocutors beginning to emulate or reflect each other in terms of speech style and, even, body posture. The extent to which they converge is often dictated by the context and the relationship of the interlocutors, and there will come a point, in any interaction, when the level of social distance feels appropriate to one or other of the participants.

It is at this point that convergence becomes maintenance. Typical of this type of maintenance is the appropriateness of the social distance a teacher might have with his or her students. As teachers of experienced (if so far unqualified) interpreters, we are, more often than not, teaching fellow professionals, yet we are able to more or less converge—usually appropriately—depending on the context of the interactions we are engaged in. For example, we tend to converge quite naturally during informal, social, out-of-the-classroom interactions, yet we are able to maintain an appropriate level of social distance that is significantly greater during teaching sessions. Just as the language style (level of formality) of interlocutors adapts to different contexts and topics, so too does the level of convergence. Communication accommodation theory (Giles et al., 1991) is now regarded as a central sociolinguistic tenet. We can’t choose not to accommodate through convergence, maintenance, or divergence; we do so naturally as soon as we begin to interact with another person, and not converging may well be perceived as choosing, instead, to diverge, that is, to act in a potentially face-threatening way (Brown & Levinson, 1987).

On a psycholinguistic level, as interlocutors start to converse they begin to create a novel, shared mental reality, a conceptual construct that conversation analysts refer to as a situation. They check and reaffirm their understanding of this joint situation by giving each other signals, or simultaneous feedback, which assure the person whose turn it is that his or her contribution is being understood in the way that it is intended it to, and this adds to and develops their shared understanding. They are, to use Wadensjö’s (1998) terminology, jointly constructing meaning. This reciprocity manifests itself through the use of nonverbal signals (e.g., head nods), phatics (e.g., uh-huh), and reinforcing contributions (right, really?); the imitation of each other’s vocabulary choices; and, as the joint meaning becomes more apparent or shared, more obvious features such as overlapping talk and the finishing of each other’s sentences (see, among others, Garrod & Pickering, 2007; Malone, 1997).

This notion of meaning as a joint construct can be modelled, albeit simplistically, as two overlapping circles. As two individuals who don’t already know each other begin to communicate, this shared situation starts off as being rather tentative. The level of shared meaning represented by the overlap could be achieved with simple gestures and smiles, the type of contact two people might have if they didn’t share a language. They would, however, in any face-to-face interaction, share a context (that in itself is laden with meaning) and the unspoken expectation that whatever the other person does is intended to mean something. If the two people share a language, the shared meaning (overlap) they would be able to construct (on top of the already present meaning of place, context, and intent) would grow very quickly.

Many student texts on interpreting (e.g., Corsellis 2008; Stewart, Schein, & Cartwright 1998) tend, no doubt unwittingly, to reinforce the idea that when an interpreter assists two parties who wish to communicate with each other, the task is approached as if the interlocutors are producing short monological utterances, rather than participating in a dialogue, and the notion of a jointly constructed meaning, or situation, is lost.

 

 

 

 

 

 

Figure 2: Traditional monological model of interpreting

 

 

 

 

 

 

 

 

 

 

 

 

Bélanger (2004) argued convincingly that this model, even when illustrated as a triangle—as it is in much of the interpreting literature—is monological, and the arrows reinforce this by portraying utterances as causal. She points out that interaction cannot be represented by the sequential utterances of the interlocutors. Interaction is instead—like the psycholinguists’ situation—what is created by all three.

An example of the temptation to represent interpreted interactions as monological is shown in Napier, McKee, and Goswell (2006). Although they recognize that the interpreter is a “participant in constructing the interaction” (p. 19), when they discuss interpreter positioning in face-to-face signed–spoken language interpreting settings, they echo Stewart et al. (1998) in suggesting that the interpreter should position him- or herself as closely as possible to the non-deaf interlocutor and, ideally, slightly behind, so that the hearing person is not tempted to address the interpreter directly and looks only at the deaf interlocutor. This is just one example of taught behaviors that are, in our view, counterproductive and are construed, by interlocutors who haven’t worked with interpreters before as very odd. Worse, such behavior signals to the non-deaf interlocutor the interpreter’s lack of willingness to align, either by converging or by participating in creating a shared situation.

3.1.     Case Study: An Example of Nonparticipation

We have described elsewhere (Llewellyn-Jones & Lee, 2009) how, during a mock interpreted interaction between a student playing the role of a deaf person and a naïve non-deaf person (that is, someone who had never communicated with a deaf person via an interpreter), the naïve interlocutor wasn’t sure that his contributions had been fully understood because there had been no back-channelling (simultaneous feedback) from the interpreter, and the deaf interlocutor had, for the most part, looked at the interpreter and rarely at him. He was, in effect, attempting to have a conversation with two people who couldn’t or wouldn’t look at him. The interpreter was following the guidelines she had been taught in a previous course and described her behavior as “more empowering” for the deaf person. This type of interaction deserves some detailed scrutiny because it is a product of the strategies routinely taught in signed language interpreter training courses.

If we look at the interaction from the point of view of accommodation theory, the sign-language-using interlocutor appeared to be converging, to some extent at least, with the interpreter; she was looking at and engaging with her by nodding, smiling and, through the utterance of the occasional signed phatic, signalling understanding. It would appear, then, to the naïve non-deaf interlocutor, that the two of them were communicating and, therefore, interacting effectively. Because the interpreter couldn’t be seen by the non-deaf interlocutor, he had to assume from the communicative behavior of the deaf person that there was some level of reciprocity, that is, that the interpreter was responding appropriately to the contributions of the deaf interlocutor. The Deaf interlocutor, however, did not appear to be making any attempt to converge with the non-deaf interlocutor: There was the occasional nod and smile when there was a pause in the interpretation but, all the time that the non-deaf interlocutor was speaking, his conversational “partner” was looking just over his shoulder (at the interpreter). The assumption on the part of the non-deaf interlocutor that the interpreter was reciprocating and actively engaging with the deaf interlocutor is only to be expected because that is the norm, the default, when two people are interacting.

Unbeknownst to the non-deaf interlocutor, the interpreter was in fact not cooperating with the sign-language-using interlocutor as one would expect. Although in constant eye contact, she was not back-channelling but, instead, just delivering a spoken rendition of that interlocutor’s contributions. There was no attempt to signal convergence through body posture or by reacting in any way to what was being said or signed. When asked about this later, she explained that she used to gesture (back-channel) when she voiced a Deaf person’s signed utterance, but was told by her previous tutors that this was distracting and unprofessional. Instead, when interpreting from signed to spoken language, she should clasp her hands in her lap and not react, in case the deaf interlocutor thought that she was agreeing with the contribution, and this would be a breach of impartiality. The sign-language-using interlocutor, then, was attempting to converge with the interpreter but her attempts weren’t being reciprocated. From the viewpoint of communication accommodation theory (Giles et al., 1999), the interaction was not normal—far from it, it was, in fact, dysfunctional—and neither was it dialogic. The interpreter’s behavior, or learned strategy, precluded any possibility of the interlocutors jointly constructing meaning by creating a shared situation.

Wadensjö (1998, p. 236) noted that, in analyses of interpreted interactions, “a number of studies have shown that feedback parts of utterances tend to be reduced or omitted (cf. Englund Dimitrova, 1991)” and several studies have found that “the amount of feedback and other checking of shared understanding is also strikingly less compared to ‘ordinary’ exchanges between people interacting.” This, we are sure, accounts for the numerous reports from students that they aren’t able to reformulate sufficiently when interpreting from British Sign Language (BSL) to spoken English because the deaf signer insists on trying to lip-read them to ensure that they are, according the students, “using the right words.” We posit that this is as a direct result of the lack of back-channelling on the part of the interpreter. Without appropriate simultaneous feedback, the deaf person has no idea whether the interpreter has understood or not, and the only way of monitoring this is to try to lip-read the interpreter’s renditions. This breaking of the “rules” of interaction only serves to condition those who regularly rely on interpreters into accepting that they will rarely, if ever, have a truly dialogical interaction with a person who speaks a different language. In the above example, the interpreter’s attempts to convey nonalignment with either participant were perceived—at least by the non-deaf participant—as biased in favor of the minority-language-using interlocutor. One cannot profess impartiality when one’s behavior is perceived as the opposite.

An accepted part of the signed language interpreter’s preparation is for the interpreter to meet the deaf participant in advance. For example, before interpreting for a medical appointment, the interpreter will approach the deaf patient in the waiting room to, ostensibly, make sure that they can understand each other. This initial conversation will involve presentation of self (“I’m so and so . . . I’m your interpreter . . . no, my parents aren’t deaf,” etc.), situation alignment (the joint construct of meaning through appropriate back-channelling), and an appropriate level of convergence. When confronting the doctor, the interpreter (according to some of our students) should simply say, “I’m just here to interpret for Mrs Scroggins” and not give a name but, instead, ask if it is permissible to move a chair. No other kind of professional that we know of refrains from introducing him- or herself when meeting new people, but some interpreting students have been taught that not giving a name—that is, not presenting oneself as a real person—somehow minimizes the interpreter’s presence. On seeing two people who by their behavior obviously already know each other, the doctor quite naturally assumes that the non-deaf person is there to support (i.e., is already aligned with) the deaf person, not to act as a neutral intermediary.

 

Figure 3: Doctor’s perception of interpreter alignment

 

 

Patient

Doctor

 

Leeds (2009) surveyed doctors’ surgeries (clinics), located in one of the U.K.’s largest cities, that she knew had contracts with an interpreting agency that provided only fully qualified signed–spoken language interpreters for medical appointments. She created a simple questionnaire for the doctors, designed to find out who they thought was providing the interpreting for their deaf patients. Given the way that many interpreters are told to introduce themselves, it is perhaps unsurprising that 54% of doctors thought the person accompanying the patient was a friend, 15% thought a caregiver, and 8% thought a social worker. Corsellis (2008) described how the spoken-language public service interpreter in the U.K. is advised to introduce him- or herself. Although to our minds (but not necessarily to the intended readership) a little over-prescriptive, it does include the interpreter introducing him or herself by name, as a real person, and, if required, by level of qualification. For the interpreter to truly participate rather than pretend he or she isn’t there is in fact a more realistic way of minimizing the interpreter’s impact (or footprint), is. If, as Metzger (1999), Hale (2007), and others agree, impartiality and neutrality are unachievable goals, then surely it is far more productive to aim, instead, for equality of alignment, especially if the consequence is more successful interactions.

So, how does the interpreter align appropriately, and fairly, with both interlocutors? We have no problem with the interpreter “getting to know” the interlocutor with whom they share a language (for the interpreter, a second language) but, we propose, this initial perception of imbalance must be repaired by the interpreter introducing him or herself by name to the other interlocutor and explaining that he or she has just met the other interlocutor and has spent a few minutes making sure that they can understand each other. By introducing him- or herself, and explaining briefly his or her relationship with the minority-language user, the interpreter is acting as any professional would be expected to. Just as the best place to hide a book is in a library, by acting as people would expect, the interpreter can “blend” (Napier et al., 2006) or minimize his or her “footprint” far more effectively. This then paves the way for the interpreter to align sociolinguistically and psycholinguistically with both interlocutors.

Alignment cannot be construed as equal if the interpreter, as soon as he or she arrives, behaves unexpectedly or gives instructions that deskill one of the interlocutors by asking him or her to behave unnaturally. It is rarely necessary to instruct the majority-language interlocutor to only look at the minority-language user. If the interpreter takes up the position that would be expected in any triadic interaction, more or less equidistant from the interlocutors, this can be done, quite simply, by the interpreter directing his or her eye gaze at the interlocutor who is being addressed. There is rarely a need for the interpreter to say, “Speak to her/him, not to me.” If the majority-language user looks at and addresses the interpreter, the interpreter, by looking at the minority-language user, naturally redirects the gaze of the other interlocutor. Very few people will continue to talk to the side of a person’s head; they will, quite unconsciously, direct their eye gaze, and hence their utterance, to the focus of attention at that point in the interaction, the other principal interlocutor. By positioning him- or herself equidistantly, the interpreter can both accommodate (typically converge with) and conversationally align with both interlocutors. The result is a genuine opportunity for a triadic communication event, one in which there is a shared situation, a shared construction of meaning. The by-product is that both interlocutors feel that they are accommodating (usually converging) appropriately, if only through the interpreter.

 

Figure 4: Dialogical model showing the potential for accommodation and conversational alignment

 

 Interpreter

 

 

 

Participant A                                                 Participant B

 

The extent of this shared situation/accommodation does, of course, depend on the type of interpreted event. If we map expected/appropriate accommodation and conversational alignment against the range of situations an interpreter might come across in day-to-day work, we arrive at some interesting differences.

4.       Interaction Management

Often discussed in the literature on community interpreting is the interpreter’s responsibility for regulating or managing the interaction between the two (or more) principal interlocutors. Wadensjö (1998, p. 266) noted that the interpreter is both a “relayer” and a “coordinator.” Controlling the interaction, as well as regulating the flow of the interaction through, for example, managing turn-taking and overlapping talk, can be, and often is, demonstrated in other ways. Corsellis (2008, p. 147) prescribed how an interpreter should handle introductions, including the order in which they occur (interpreter last), and above we have shown how signed language–spoken language interpreters, through seating arrangements and verbal, pre-interaction instructions (“don’t look at me, look at the deaf person”) often take it upon themselves to decide who is allowed to look at whom.

Corsellis (2008), after emphasizing the requirement for absolute “accuracy” in the interpretation of every utterance (turn) of the principal participants, listed four circumstances in which it is appropriate for the interpreter to interject with self-initiated /authored utterances:

 

  1. To ask for accommodation to the interpreting process, for example, when someone is speaking inaudibly or too quickly
  2. To ask for clarification of an utterance that is ambiguous, unclear or contains a term that is not understood by the interpreter. This may be a technical term or slang
  3. To alert the participants that someone may not have understood what has been said although the interpreting was correct. Interpreters have a good feel for comprehension and can sense when this is happening. The causes may be various but they can, for example, include either party using formal registers or technical terms that are not comprehensible to the other in either language
  4. To alert the participants that a relevant cultural inference may have been missed. This refers to an item of information that is not in the cultural frame of reference of one or more of the participants. An example could be the significance of religious festivals and why it is important that a family member should be going home for Christmas, Divali and so on. (p. 48)

As has been shown in the analyses of real interpreted events, however, there are frequently times when a wider range of interpreter-initiated interventions (or noninterventions) are appropriate and/or necessary. Rosenberg (2002) undertook a quantitative analysis of his own interpretations of 11 medical consultations involving physicians and the Spanish-speaking parents of the child patients. Having transcribed a total of 1,334 interpreter utterances, he found that only 40.8% of them could be categorized as “close renditions” or “accurate” interpretations of the participants’ utterances. At first sight, this seems a remarkably low percentage considering that, traditionally, quality in interpreted dialogues has been judged by the accurateness and completeness of the interpretation of every participant turn. It transpires, however, that a further 39.7% were examples of interpreter decisions that could, even through the lens of “talk as text,” (Wadensjö, 1998) be viewed as legitimate. The  largest group of utterances in this category, 26.9%, were categorized as “zero renditions,” resulting from the interlocutors speaking to each other directly via their—albeit sometimes limited—knowledge of each other’s language, and a further 12.8% comprised either “expanded” or “reduced” renditions.

We are still left, however, with 19.5% of the interpreter’s utterances that were something other than the close/zero/expanded/reduced utterances already listed. Of these, the breakdown was recorded as 56.9% phatic, 16.5%          clarifications, 7.7% understood (without intervention from the interpreter), 6.9% off task (the interpreter speaks to either party about something not related to the conversation), 6.2% banter, and 5.8% repetitions (Rosenberg, 2002).

Overlapping talk is a natural part of conversation and occurs for several distinct reasons, some unintentional—for example, when two people coincidentally start to talk at the same time—and some intentional, for example to signal the listeners’ group solidarity and support of the speaker (see, e.g., Tannen, 1984, 1989). It is also a necessary feature in the joint creation of meaning as back-channelling often requires (as the alternative term simultaneous feedback indicates), the utterance of phatics, single words and short phrases signalling understanding, agreement, confusion, and so on, while the conversational partner is speaking. The words or short phrases uttered can also demonstrate prediction, in that one interlocutor might finish another’s sentence.

The interpreting literature almost invariably points out that it is impossible to interpret for two people at the same time, especially as the complete and “accurate” rendition of each turn has been viewed as a necessary aspect of a “quality” interpretation. Roy (2000) suggested four possible ways that the interpreter can deal with overlapping talk:

  1. Stop one (or both) speakers and allow the other speaker to continue. If an interpreter stops both speakers, then either the interpreter indicates who speaks next or one of the primary speakers decides who talks next.
  2. Momentarily ignore one speaker’s overlapping talk, hold the segment of talk in memory, continue interpreting the other speaker, and then produce the “held” talk immediately following the end of a speaker’s turn. Decisions about holding talk in one’s memory lie within the interpreter’s ability to do so and the interpreter’s judgment regarding the importance or impact of the talk to be held in memory.
  3. Ignore overlapping talk completely.
  4. Momentarily ignore overlapping talk and upon finishing the interpretation of one speaker, offer the next turn to the other speaker, or indicate in some way that a turn was attempted. (p. 85)

 

In our experience, interpreting students are typically taught that the first option (above) is the default way of dealing with overlapping talk should it arise or, better still, that the interpreter should ensure from the outset that overlapping talk does not occur. Instructions that interlocutors should speak one at a time are commonplace and, should one or more of the participants forget, the usual overt management strategy employed is to stop the interaction to remind the participants of the turn-taking “rule.” The problem with this, of course, is that it militates against the possibility of the two (or more) principal interlocutors creating a shared situation. Options 2 and 4 are often an appropriate way of allowing limited overlapping talk, but any back-channelling loses its simultaneity if the interpreter deals with it as a next turn.

A fifth option that is appropriate in many face-to-face interpreted interactions is for the interpreter to momentarily interrupt the participant he or she is interpreting for, interpret the feedback utterance, then switch attention back to the first participant to indicate that he or she still has the floor. This strategy works well for phatics, words, and short feedback utterances and one can typically recognize, from intonation and other paralinguistic cues, whether an interlocutor is simply engaging in simultaneous feedback or attempting to take the turn.

Overt management of turn-taking can be particularly inhibiting for participants in nonantagonistic small group discussions and meetings. Here, instructions from the interpreter that participants must speak one at a time are commonplace and in some circumstances militate against the discussion proceeding as it would without an interpreter. These interactions, when all participants speak the same language, are characterized by overlapping talk, interruptions, and attempts to gain the floor by speaking over other participants. In certain cultures and types of interaction, for example, those involving participants of unequal status, these turn attempts might be regarded as impolite and as breaking conversational conventions. When the participants are close colleagues or friends, however, these moves can still be classed as cooperative overlap (Tannen, 2000) because they are intended to contribute to the group’s shared understanding and, as such, are seen as being creative rather than intrusive or counterproductive. Strict enforcement of sequential turns inhibits this spontaneity and creativity.

4.1.     Mapping the Axis of Management to Different Settings and Contexts

The most obvious example of the interpreter having little or no control over the management of the interaction is in the simultaneous interpretation of formal addresses or presentations, whether in conference or formal lecture settings. The interpreter has no control over the pace, complexity, or propositional density of the presentation and is unable to intervene to ask the speaker to modulate her pace, pause to allow her to catch up, or ask for clarification if something is not completely understood. On the axis of management this would be represented as in Figure 5.

 

Figure 5: Simultaneous conference interpreting

 

High management

Low management

 

 

 

 

 

 

 

 

 

 

 

 

The reality, of course, is that most community interpreting assignments involve a minority-language user accessing public services, and it is in the participants’ interests that the communication is clear and understood. Consequently, the interlocutors are typically very tolerant of the interpreter checking understanding, asking for more time to make sure something has been understood, and so forth.

 

Figure 6: Interpreting public service consultations

High management

Low management

 

 

 

 

 

 

 

 

 

 

4.2.     Interaction Among the Axes

The diversity of the assignments undertaken by community interpreters means that there is an indeterminate range of role-spaces that might be appropriate at some time or another. As has been seen in the scenarios presented above, these spaces are not, however, defined solely by the setting. If we put the three axes together we begin to see patterns emerge that reflect the type and goals of the interactions rather than the specific settings or modes of interpreting chosen (e.g., simultaneous, consecutive, chuchotage, etc.). Just as the dynamics of interactions change as the interlocutors adjust their social distance and agree new goals, so too do the role-spaces.

 

Figure 7: Interaction between the axes

 

 

 

 

 

Alignment

 

 

 

 

 

 

 

 

 

Relatively recently, the first author was one of two interpreters booked to interpret for the defense team in the trial of a middle-aged man accused of offenses that would, if he were found guilty, have resulted in a long custodial sentence. The morning started with a meeting involving the barrister, the solicitor (who had prepared the original brief) and her assistant, the interpreter who had interpreted all of the preparatory interviews between the solicitor and the defendant, and the defendant. The purpose was to discuss the plea (not guilty), the allegations being presented by the prosecution, and the arguments that were to be used as the defense. Although a formal part of the judicial process, the style and social distance of the interlocutors would best be described as “consultative” (Joos, 1967).

Both interpreters approached the assignment as a formal/consultative dialogic encounter in which the presentation of self was sufficient for the interlocutors to make judgements about, for example, their integrity but limited to observing expected politeness conventions. The level of alignment was (as far as possible) equal and fair but restricted to the level of sociolinguistic and conversational alignment necessary to ensure understanding between the parties (a degree of convergence followed by maintenance). The aims of the participants and of the interaction afforded the interpreters plenty of opportunity for interaction management, because the central concern of both principal interlocutors was clarity and complete understanding of the questions and responses. The interpreters were, therefore, able to use both overt and covert management strategies to check their understanding and, through back-channelling, embedded, and parallel utterances (Bélanger, 2004), assure the interlocutors that their contributions were being understood and that the interpreted renditions were “faithful” and complete. The role-shape for the first part of the meeting could be illustrated as in Figure 8.

 

 

 

 

 

Figure 8: Meeting with barrister present

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The meeting, which lasted some 3 hours, was punctuated by the barrister leaving the room to, initially, seek clarification of a particular allegation and, later, to put the defendant’s responses to the prosecution barrister to see whether there was any possibility of a retraction or negotiation of a particular charge. During his absences, which sometimes lasted for 15 to 20 minutes or more, the other occupants of the room made polite conversation. They talked about holidays, the weather, local places to eat, almost anything other than the reason they were all together. The solicitor was still a solicitor but she was also a person, and social convention required her to talk about herself, her holiday plans or whatever else (within certain limits) came up in conversation. All of the occupants of the room participated in the “banter,” to use Rosenberg’s term (see above), including the interpreters. Taking turns as and when appropriate, if one interpreter was asked whether he had yet been on holiday, the other spontaneously took over the interpretation of that part of the conversation. The role-shape for these episodes was different from the shape during the parts of the meeting when the barrister was present, because the goals were different and social convention demanded that the five people, in a room together for an extended period of time, talk about something.

 

Figure 9: Meeting with barrister absent

Interaction Management

Other participants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Later that day, after a final briefing by the barrister during which the interpreters re-adopted the role-shape of Figure 8, the parties were summoned to court to appear before the judge. The goal of this interaction was very different from the goals of the previous two, as were the interlocutors’ expectations of the role of the interpreter. The interpreter’s alignment with the principal interlocutors—judge, clerk of the court, prosecution barrister, defence barrister and defendant—was limited to sufficient conversational alignment to ensure understanding (but no more), and presentation of self was limited to the interpreter giving his name, his qualifications and, under oath, affirming that he would interpret faithfully and to the best of his ability.  Because the British courts demand that proceedings are transparent, any management of the interactions is, necessarily, overt and typically preceded by a request to the relevant principal (questioning barrister or judge) for permission to seek clarification or for the question to be reframed (see Berk-Seligson, 1990, for an account of judges’ and lawyers’ unrealistic demand that any interpretation should be verbatim). The role-shape was as in Figure 10.

 

Figure 10: The court hearing

Judge, Court officials, etc.

 

 

 

 

 

 

 

 

 

 

 

Defendant

 

 

 

 

 

 

This role-shape is not dissimilar to the shape adopted by the interpreters in Figure 8, but it is more restricted, indicating that the interpreter has less room for maneuvering. The point on the interaction management axis suggests that the interpreter has less opportunity to regulate the flow of the interaction—as, indeed, is the case—but it also reflects that, because of the formality of the setting, the interpreter is likely to be more reticent in halting the proceedings. We could debate whether this reticence does a disservice to the interaction (it could, quite possibly, affect the outcome) but it is, we suggest, because the interpreter is not a machine and is, consequently, influenced by his perception of his status in that setting.

5.       Conclusion

Remember that in any given interaction, it is not the interpreter who decides on the nature and dimensions of the role-space; instead, it is the characteristics of the interaction that determine the appropriateness of the myriad approaches and roles available to the interpreter. We have seen that certain communication events leave the interpreter with little room to maneuver, for example, in the formal meeting with the barrister where the stakes are high and in the adversarial court appearance where the interpreter is tolerated but seen only as a necessary tool by the principal interlocutors. On the other hand, we have described the freedom afforded by less formal, collaborative conversational interactions. With a long tradition of prescriptive and proscriptive descriptions of the role of the interpreter, perhaps the major limiting factor is how comfortable the individual interpreter is with the freedom for professional decision making that many of the role-spaces allow.

The model outlined in this article is, we contend, a useful way of describing the place that community interpreters have in the communicative interactions in which they work. The role-space model takes into account the fact that interpreters do not merely transfer meaning from one language to another.

In this role-space, first, interpreters need, above all else, to be human beings with well-honed social skills, sensitivities, and awareness, as well as excellent linguistic skills. It is important to recognize that how interpreters present themselves in any given situation can have a huge effect on the interaction. Second, interpreters must be keenly aware of how individuals align and accommodate each other to achieve communicative goals. Rather than be a hindrance in the middle of interactions, interpreters have a vital role to play in aligning with the participants and, by extension, allowing them to align with each other.  Last, but by no means least, interpreters must be skilled in a variety of techniques to appropriately manage interactions in order to facilitate successful communication between and among participants.

These points are important for all areas of the profession, from the screening of potential students to curriculum design to the structure of qualifications and professional codes. Recognizing they have a place in the interaction, skilled community interpreters must make sure that the role-spaces they occupy are principled, appropriate, and respectful to all parties involved; and interpreter educators must prepare interpreting students to enact different roles as necessary.

 

References

 

Angelelli, C. V. (2001). Deconstructing the invisible interpreter: A critical study of the interpersonal role of the interpreter in a cross-cultural/linguistic communicative event (Unpublished doctoral thesis). Stanford University, Palo Alto, CA.

Angelelli, C. V. (2004) Medical interpreting and cross-cultural communication. Cambridge, UK: Cambridge University Press.

Bélanger, D.-C. (2004). Interactional patterns in dialogue interpreting. Journal of Interpretation, 17, 1–18.

Bell, A. (1984). Language style as audience design.  Language in Society, 13, 145–204.

Berk-Seligson, S. (1990). The bilingual courtroom, Chicago, IL: Chicago University Press.

Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage. Cambridge, UK: Cambridge University Press.

Corsellis, A. (2008). Public service interpreting: The first steps. Basingstoke, UK: Palgrave Macmillan.

Dickinson, J. (2010). Interpreting in a community of practice: A sociolinguistic study of the signed language interpreter’s role in workplace discourse (Unpublished doctoral thesis). Heriot Watt University, Edinburgh, Scotland.

Dickinson, J., & Turner, G. H. (2009). Forging alliances: The role of the sign language interpreter in workplace discourse. In R. de Pedro Ricoy, I. Perez, & C. Wilson (Eds.), Interpreting and translating in public service settings (pp. 171-183). Manchester: St Jerome.

 

Englund Dimitrova, B. (1997). Degree of interpreter responsibility in the interaction process in community interpreting. In S. E Carr, R. Roberts, A. Dufour, & D. Steyn (Eds.), The Critical Link: Interpreters in the community: Papers from the first international conference on iterpreting in legal, health and social services settings 1995 (pp. 147–164). Amsterdam, the Netherlands: John Benjamins.

Frishberg, N. (1990). Interpreting: An introduction. Alexandria, VA: RID Publications.

Garrod, S., & Pickering, M. J. (2007). Alignment in dialogue. In M. G. Gaskell (Ed.), The Oxford handbook of psycholinguistics (pp. 443–451). Oxford, UK: Oxford University Press.

Gile, D. (1995). Basic concepts and models for translator and interpreter training. Amsterdam, the Netherlands: John Benjamins.

Giles, H., Coupland, N., & Coupland, J. (1991). Accommodation theory: Communication, context, and consequence. In H. Giles, J. Coupland, & N. Coupland (Eds.), Contexts of accommodation (pp. 1–68). Cambridge, UK: Cambridge University Press.

Goffman, E. (1981). Forms of talk. Philadelphia: University of Pennsylvania Press.

Goffman, E. (1990). The presentation of self in everyday life. London, UK: Penguin. Originally published 1959.

Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics (pp.41-58). New York, NY: Academic Press.

Hale, S. B. (2007). Community interpreting. Basingstoke, UK: Palgrave Macmillan.

Joos, M. (1962). The five clocks. Bloomington, IN: Indiana University Research Center in Anthropology, Folklore, and Linguistics. Reprinted in 1967 by Harcourt, Brace & World.

Lee, R. G. (1997). Roles, models and world views: A view from the states. Deaf Worlds, 13(3), 40-44.

Lee, R. G., & Llewellyn-Jones, P. (2011). Re-visiting role: Arguing for a multi-dimensional analysis of interpreter behaviour. Paper presented at the Supporting Deaf People Online Conference. Retrieved from http://www.online-conference.net/previous.htm

Leeds, R. (2009). Who do they think we are? The view of the interpreter, as seen by medical practicioners (Unpublished master’s dissertation). Univerisity of Leeds, Leeds, U.K.

Llewellyn-Jones, P., & Lee, R. G. (2009). The “role” of the community/public service interpreter. Paper presented at the Supporting Deaf People Online Conference. Retrieved from http://www.online-conference.net/sdp2009.htm

Malone, M. J. (1997). Worlds of talk: The presentation of self in everyday conversation. Cambridge, UK: Polity.

McIntire, M., & Sanderson, G. (1993). Bye-bye! Bi-bi!: Questions of empowerment and role—A confluence of diverse relationships. In Proceedings of the Thirteenth National Convention of the Registry of Interpreters for the Deaf (pp. 94–118). Silver Spring, MD: RID Publications.

Mead, G. H. (1934). Mind, self, and society from the standpoint of a social behaviorist. Chicago, IL: University of Chicago Press.

Metzger, M. (1999). Sign language interpreting: Deconstructing the myth of neutrality. Washington, DC: Gallaudet University Press.

Nakane, I. (2009). The myth of an “invisible mediator”: An Australian case study of English–Japanese police interpreting.  PORTAL Journal of Multidisciplinary International Studies, 6(1), 1-16.

Napier, J., McKee, R. L., & Goswell, D. (2010). Sign language interpreting: Theory and practice in Australia and New Zealand. Annandale, New South Wales, Australia: Federation Press.

Pöchhacker, F. (2004). Introducing interpreting studies. London, UK: Routledge.

Rosenberg, B. A. (2002). A quantitative discourse analysis of community interpreting. Translation, 16, 222–226.

Roy, C. B. (1993). The problem with definitions, descriptions and the role metaphors of interpreters. Journal of Interpretation, 6, 127-154.

Roy, C. B. (2000). Interpreting as a discourse process. New York, NY: Oxford University Press.

Stewart, D. A., Schein J. D., & Cartwright B. E. (2004). Sign language interpreting: Exploring its art and science. Boston, MA: Pearson/Allyn & Bacon.

Tannen, D. (1984). Conversational style: Analyzing talk among friends. Norwood, NJ: Ablex Publishing Corporation.

Tannen, D. (1989). Talking voices: Repetition, dialogue, and imagery in conversational discourse. Cambridge, UK: Cambridge University Press.

Tate, G., & Turner, G. H. (1997). The code and culture: Sign language interpreting—In search of the new breeds’s ethics. Deaf Worlds, 13, 27–34.

Tate, G., & Turner, G. (2001). The code and the culture: Sign language interpreting – in search of the new breed’s ethics. In F.J. Harrington & G. H. Turner (Eds.), Interpreting Interpreting: Studies and Reflections on Sign Language Interpreting (pp.53-66). Gloucestershire: Douglas McLean.

Turner, R. H. (1956). Role-taking, role standpoint, and reference-group behavior.  American Journal of Sociology, 61, 316–328.

Turner, R. H. (1962). Role taking: Process versus conformity. In A. M. Rose (Ed.), Human behavior and social processes: An interactionist approach. Boston, MA: Houghton Mifflin.

Wadensjö, C. (1998). Interpreting as interaction. London, UK: Longman.