When roles within interpreter-mediated psychiatric consultations speak louder than words

Abstract

In South Africa, clinicians working in public psychiatric hospitals are mainly fluent in English and Afrikaans, while the majority of patients are not proficient in these languages. Due to a lack of professional interpreting services, informal, ad hoc interpreters are commonly employed in public psychiatric hospitals. We collected data on language practices in public psychiatric care in South Africa, and provide a detailed account of what happens when bilingual health care workers and cleaners haphazardly take on the additional role of interpreter. Data were collected during 2010 at a public psychiatric hospital in the Western Cape, South Africa. Thirteen interpreter-mediated psychiatric consultations were video-recorded, and 18 audio-recorded semi-structured interviews were conducted with the interpreters and clinicians who participated in the interpreter-mediated psychiatric consultations. Patients were proficient in isiXhosa (one of the 11 official languages of South Africa), the clinicians (all registrars) were first language English or Afrikaans speakers, while the health care workers (nurses and social workers) and cleaners were fluent in both the patients’ and clinician’s language. Our findings suggest that interpreters took on the following four roles during the interpreter-mediated psychiatric consultations: regulating turn-taking, cultural broker, gatekeeper and advocate. Our findings suggest that, despite interpreters and clinicians having the patient’s best interests at heart, it is the patient’s voice that becomes lost while the clinician and interpreter negotiate the roles played by each party.

Keywords

cultural broker interpreting language limited English proficiency (LEP)psychiatry

Introduction

The challenges associated with language barriers to health care access are a global health problem (Bowen, 2001; Flores, 2006; Jacobs et al., 2004). For many countries with high rates of migrant workers and refugees, such as the United States and those in mainland Europe, the challenge is not new (Flores, 2006). However, of late, there has been a renewed call for the global health community to pay greater attention to language access and diversity in health care, not only for the sake of immigrants and refugees, but also for indigenous communities (Flood & Rohloff, 2018). Indigenous communities are known to experience language barriers to health care due to histories of marginalization and colonialism (Flood & Rohloff, 2018), and, in the case of South Africa, due to Apartheid (Levin, 2006).

When South Africa became a democratic country in 1994, a new constitution, regarded as one of the most progressive constitutions globally, was adopted under the leadership of Nelson Mandela. It promotes equal access to services and non-discrimination on the basis of language, class, or race. However, the implementation of the constitution at the grassroots level is limited in some settings. Over 20 years after the end of Apartheid, many South Africans face language discrimination when they engage with public mental health care services. Although formal arrangements are made for interpreter services in the courts and in the legislative sector, and even to some extent in the general health care sector, this is not the case when it comes to public mental health care in the country. Informal interpreters are commonly employed in public psychiatric hospitals. Informal interpreting falls within the category of Community Interpreting (Bot, 2005; Hale, 2007)—a branch of interpreting that takes place in community settings where people communicate in spontaneous face-to-face interactions (Mason, 2014).

The lack of formal interpreter services can be ascribed to various factors. Financial constraints constitute one of the contributing factors. The provision of language services falls by the wayside in a country facing one of the world’s highest incidences of human immunodeficiency virus infection and acquired immune deficiency syndrome and tuberculosis. Furthermore, the majority of clinicians working in public psychiatric hospitals are mainly fluent in English and Afrikaans, while the majority of patients are not proficient in these languages (Levin, 2006). Another compelling explanation for language discrimination is that, as seen from the outside, the health care system seems to work, despite the lack of official interpreter services (Drennan & Swartz, 2002). Drennan and Swartz (1999) explain that the problem of language access and routinized strategies to work around it have, to a great extent, become institutionalized aspects of the everyday practice.

In the Western Cape (one of the nine provinces of South Africa), a few informal interpreters are primarily employed as administrative staff and are therefore not always available to fulfil the additional role of interpreter. This often results in ad hoc arrangements being made with bilingual health care workers, cleaners, or security guards, who take on the role of interpreter when the need for language services arises. There is a need for empirical research studies that explore the roles taken on by interpreters in different settings (Leanza, 2005), as such work will allow for a better understanding of the complexities associated with the multifunctional role of the interpreter (Hsieh, 2008; Leanza, 2005).

Interpreters actively take on different—and often overlapping—roles within the same session. Hsieh (2008) found that interpreters take on four roles, namely that of conduit, advocate, manager, and professional. The role of conduit prescribes that interpreters should be ‘invisible’ language agents, while the role of advocate entails empowering mainly the patients during the interpreter-mediated session. Interpreters perceive themselves as managers, since they regard themselves as responsible for various aspects of the interaction in which they are involved, such as monitoring the conversation between the patient and clinician and managing the exchange of information. The role of professional relates to interpreters’ status as individuals providing a professional service, and is evidenced in interpreters’ active efforts to exert control during the talking session. In addition, Leanza (2005) proposes the role of integration agent, which takes place outside of the interpreter-mediated consultation. This role entails assisting the patient in finding resources that will aid them to make sense of their world and to negotiate meaning (Leanza, 2005). Davidson (2000) and others (Shiu-Thornton et al., 2007) have noted the need for an exploration of the patterned ways in which interpreters act as gatekeepers of culture and information, and how these socially positioned agents function in a biomedical setting, given the potential impact that these factors may have on the goals and the process of psychiatric consultation.

As part of a bigger project on language access to mental health care, we collected data on language practices in public psychiatric care in South Africa. Here, we report on a smaller exploratory sub-study nested within the larger project. We provide a detailed account of what happens when bilingual health care workers and cleaners haphazardly take on the additional role of interpreter.

Methods

Data were collected during 2010 at a particular public psychiatric hospital in the Western Cape, South Africa. The hospital is one of only three public psychiatric hospitals in the Western Cape (one of the nine provinces in South Africa) and is located in the Cape Town Metropole. The data were collected at two of the hospital’s inpatient wards, one for females and one for males, with approximately 30 beds in each ward. Patients are referred to the wards once they are stabilized at admissions. We present information pertaining to the study participants in Table 1 (also reported elsewhere—see Kilian et al., 2014). At the time of data collection, the hospital did not employ formal interpreters. As such, bilingual health care workers, cleaners, or security guards were used to provide language services. The first author video-recorded 13 interpreter-mediated psychiatric consultations and conducted 18 semi-structured interviews (audio-recorded) with the interpreters and clinicians who participated in the abovementioned 13 interpreter-mediated psychiatric consultations. The clinicians were all registrars (also known as residents) training to become psychiatrists. In 11 of the interpreter-mediated psychiatric consultations the registrars were female, and in two of the consultations one male registrar was involved. All the interpreters were female, while six patients were female and seven were male. All the female registrars were white, while the male registrar was of mixed ancestry. The patients and interpreters were all black. We did not have information pertaining to participants’ age or educational background. Patients were proficient in isiXhosa (one of the 11 official languages of South Africa), the clinicians (all registrars) were English or Afrikaans speakers (as their first language), while the health care workers (nurses and social worker) and cleaners were fluent in both the patients’ and clinician’s language. We provide more detailed information on the sampling of the hospital in question as well as the research participants elsewhere (Kilian et al., 2014, 2015).

Table 1.

Study participants’ official occupation and sex.

Interview	Registrar	Patient	Interpreter
1. Interview	Female (Dr. B)	Female (P1)	Female cleaner (Interpreter R)
2. Interview	Male (Dr. D)	Female (P2)	Female nurse (Interpreter Y)
3. Interview	Female (Dr. F)	Male (P8)	Female nurse (Interpreter Z)
4. Interview	Female (Dr. F)	Male (P9)	Female nurse (Interpreter Z)
5. Interview	Female (Dr. F)	Male (P10)	Female nurse (Interpreter Z)
6. Interview	Female (Dr. E)	Female (P11)	Female nurse (Interpreter Y)
7. Interview	Female (Dr. E)	Female (P12)	Female nurse (Interpreter Y)
8. Interview	Female (Dr. E)	Female (P13)	Female nurse (Interpreter Y)
9. Interview	Male (Dr. C)	Female (P19)	Female cleaner (Interpreter V)
10. Interview	Female (Dr. D)	Male (P20)	Female social worker (Interpreter P)
11. Interview	Female (Dr. D)	Male (P8)	Female social worker (Interpreter P)
12. Interview	Female (Dr. D)	Male (P3)	Female social worker (Interpreter P)
13. Interview	Female (Dr. D)	Male (P21)	Female social worker (Interpreter P)

Video recordings of the interpreter-mediated psychiatric consultations and audio recordings of the semi-structured interviews were transcribed verbatim. The first author, fluent in English and Afrikaans, was responsible for forward- and back-translating the Afrikaans dialogue. A bilingual, first-language isiXhosa speaker and part-time translator were responsible for forward- and back-translating the isiXhosa dialogue. The forward and back translations of the isiXhosa and Afrikaans dialogue were cross-checked by two independent bilingual individuals and no discrepancies were found. We used Conversation Analysis (CA) to analyze the interpreter-mediated sessions that were video-recorded. Conversation Analysis, commonly used in the field of community interpreting research, provides a micro-analysis of the features associated with spoken conversations (Hale, 2007). It is aimed at identifying the actions used by speakers and their influence on the conversation, as well as the organization, stages, structure, and goals of the conversation (Peräkylä, 2005). We followed a conventional approach to CA, which is aimed at providing a rich description of the conversation. However, our aim is not to develop a theory. The advantage of this approach is that the researcher is less likely to form a biased interpretation of the themes emerging from the data (Hsieh & Shannon, 2005). We employed the CA techniques, as adopted by Friedland and Penn (2003), to analyze the video recordings. This involves identifying facilitators and inhibitors. A facilitator is any action used by any of the three parties that makes an overall positive contribution to the goals of the psychiatric consultation, while an inhibitor consists of any action that has a negative impact on the goals of the consultation (Friedland & Penn, 2003). Content analysis was used to analyze the audio recordings. The study was approved by the Committee for Human Research at Stellenbosch University and by the board of the hospital (Ref. no. N09/05/162). All participants gave their consent to participate in the study. Prior to the interpreter-mediated psychiatric consultations, the first author obtained consent from the registrars and interpreters. The same interpreter that participated in the psychiatric consultation assisted the first author to obtain consent from the patient. In the presence of the first author and registrar, the interpreter explained the purpose of the study and provided information pertaining to patient confidentiality. The patient was provided with the opportunity to ask questions, and if they were interested in participating, verbal consent was obtained. None of the participants declined to participate. For the audio-recorded semi-structured interviews with registrars and interpreters, the first author obtained written consent.

Results

Our findings suggest that interpreters took on the following four roles during the interpreter-mediated psychiatric consultations:

Regulating turn-taking

Cultural broker

Gatekeeper

Advocate

Here we provide a detailed account of each of these roles, drawing on illustrative extracts from the interviews.

Regulating turn-taking

Our findings suggest that interpreters were relatively successful in managing turn-taking in the interaction by employing different turn-taking techniques. Interjection was most frequently used to regulate turn-taking, exemplified by the following example from the data. The interpreter interjected the patient’s sentence to indicate to the patient to stop talking in order for the interpreter to convey to the clinician the content of the patient’s response. The interpreter said to the patient, “Wait, wait let me narrate this to the doctor because your story is very long.” Interpreters also used the quotatives “he says” and “she says” to regulate turn-taking. The phrase “and then” (in English) was another turn-taking technique used by the interpreters to signal to the patient to continue with their turn. Interestingly, the clinicians did not leave the task of regulating turn-taking completely to the interpreters, and of their own accord took on some of the responsibility by using interjection as well as other innovative techniques. For example, clinicians often used non-verbal techniques such as physically touching the patient’s arm to indicate to the patient that he or she should stop talking. Also, one of the clinicians used her limited proficiency in the patient’s first language to regulate turn-taking. The particular clinician used the phrase “Wait mama” to stop the patient talking whenever needed.

Although interpreters and clinicians were largely successful in sharing the responsibilities of turn-taking and managing the consultations, this was not without its problems. There were a few instances where interpreters seemed confused about their responsibilities during the consultations. For example, as evident in the extracts below, the interpreter regulated turn-taking and provided renditions of utterances only when explicitly asked to do this by the clinician or patient, while in other instances the interpreter regulated turn-taking without being asked to do so. In the extract below (Extract 1), the interpreter interjects to provide a rendition without the patient or clinician requesting her to do so:

Extract 1:

Dr. C: Mm, I understand, but I want to know what was difficult, stressful. I get the feeling that you say there wasn’t enough support.

Patient 19: Mm.

Dr. C: From your family members.

Patient 19: Yes. I didn’t sleep.

Dr. C: Yes, I understand, I …

[Interpreter interjects]

Interpreter V: The doctor wants to ask you if when you think, you see when you think.

Later in the same consultation, the clinician explicitly asked the interpreter to assist him (see Extract 2):

Extract 2:

Dr. C: Too much was happening.

Patient 19: Yes.

Dr. C: Your plate was too full. You had to do too many things.

Patient 19: Mm, then I say to her, no don’t want to do the adopted because at Eastern Cape, they are gonna say where are they [the children]. And it’s a little, little girl so I can’t, it’s my sister, so.

Dr. C: Ok. Can you just help me [looking at the interpreter]. Specifically what I would like is if you can find out if there was a lot of praying, not sleeping and cleaning at night at strange hours. Feeling that there is too much energy and too much power.

Furthermore, clinicians seemed frustrated at times by interpreters’ management of the flow of the consultation, particularly when interpreters did not provide clinicians with a turn to talk or allowed for very long turns. In the extract below (Extract 3), the clinician tries to interject but to no avail. Instead, the interpreter continues to talk to the patient:

Extract 3:

Interpreter Y: About muthi stuff.

Dr. C: Can I say …

[Interpreter interjects]

Interpreter Y: And at night?

Patient 2: I am beaten by things I do not know at night, I am beaten after I prayed.

Interpreter Y: Where?

In another consultation the clinician raised her voice when the interpreter allowed long turns instead of shorts turns (see Extract 4 below):

Extract 4:

Dr. E: Um, quickly ask her was she eventually admitted to hospital and for how long?

Interpreter Y: You were admitted at that time to the hospital?

Patient 13: No, no sister, at the hospital, I called, I called this Mrs Cohen, it was already night-time and I was really scared, shaking, and again I called the ADT and then they came.

Interpreter Y: At that time.

Dr. E: [In a raised voice] Was she eventually admitted to hospital?

Although this is purely speculative, clinicians’ frustrations are likely due to the time constraints under which they work in a busy hospital setting. The theme of time constraints was a common one that emerged throughout the data. Below, we present examples of instances in which the theme of time constraints featured prominently in the video recordings. In one of the consultations, the clinician told the interpreter “We should hurry up,” and in a raised voice asked the interpreter what the patient was saying. In response, the interpreter attempted to get the patient to give a concise response instead of a more detailed one. When the patient tried to explain that her story was detailed and that she did not want to leave out information, the interpreter interjected by saying, “Oh God” and pressurized the patient to only answer questions directly related to the clinician’s questions due to time constraints. It is also evident from the extract below that the interpreter herself became frustrated by the patient’s long turns and in the end told the patient, “I said to you, you will only answer questions that are being asked by the doctor otherwise we won’t finish.” The following dialogue arose:

Extract 5:

Interpreter Y: What healed you? My question is what healed you?

Patient 13: I was saying sister, I …

[Interpreter interjects]

Interpreter Y: Now, before you arrived here, you became sick after some time? You became sick again while using pills.

Patient 13: Sister my illness is long. I went to San Marco again.

Interpreter Y: [With emphasis] Oh God!

Patient 13: I went to sleep at San Marco, do you remember?

Interpreter Y: I’m asking a question, answer my question only.

Later during the same psychiatric consultation, this exchange occurred:

Interpreter Y: Did you beat Themba with your hand or a stick?

Patient 13: Themba in the roof.

Interpreter Y: [In a raised voice] Mama answer the question.

Patient 13: I beat him.

Interpreter Y: Answer the question.

Patient 13: I beat and strangled him.

Interpreter Y: I said to you, you will only answer questions that are being asked by the doctor otherwise we won’t finish.

From the semi-structured consultations, it emerged that interpreters were very conscious of the issue of time constraints. In fact, interpreters regarded this as one of the main reasons why they had reservations about the additional role of interpreter. Interpreters explained that interpreting was time-consuming and involved more than simply translating words from one language into another. It also involved establishing a relationship between all three parties. The extract below exemplifies one of the interpreters’ comments regarding the theme of time constraints:

Extract 6:

Researcher: You are a cleaner [i.e., household aide], ok. So I’d like to know whether this, because this is an additional role, being an interpreter, right?

Interpreter V: Yes.

Researcher: It’s over and above what you must do. I want to know whether this interferes with your official duties being as, um, a cleaner.

Interpreter V: Sometimes, erh, it’s difficult to me because I must leave my job and come to help the doctor. Sometimes other patient is so difficult, sometimes other patient don’t want even to speak out to the doctor, to talk. I must try to make that patient come to talk to the doctor. To explain to the doctor what’s going on, because sometimes other doctors, they don’t know how to talk to the patient—the isiXhosa patient. It’s so difficult because that patient is not even hear what the doctor say. Now first of all, I must try to make a friendship with me and the doctor and the patient. That time my work is waiting for me.

Researcher: Because it takes time, you say?

Interpreter V: It takes time, because other patient is so difficult. Others, they answer the doctor and talk long. Now, I must try to help that patient. Ok, doctor don’t want to ask you this question. He answer the question in the wrong way.

Cultural broker

Some of the interpreters took on the role of cultural broker, which involved building bridges between cultures, in this case psychiatry and the patient’s respective cultures. This is not surprising, given that they shared a first language as well as a cultural background with patients. Interpreters were, at times, assigned this role by the clinicians, who would explicitly ask interpreters about their opinions regarding cultural practices. For example, one of the male patients mentioned that he had not yet undergone the initiation process. The initiation process is a process that young men undergo which involves circumcision, and is their rite of passage into adulthood. During the psychiatric consultation, the clinician asked the interpreter whether the patient’s preoccupation with his genitals could be ascribed to the fact that he had yet to undergo initiation. The interpreter explained to the clinician that it was unusual for a man at the age of 23 years (i.e., the patient’s age) not to have yet taken part in the initiation process. At other times interpreters took on the role of cultural broker without being asked to do so. For example, interpreters offered information about other cultural practices, such as gooi-gooi (i.e., a community-based savings club). During the semi-structured consultations, interpreters spoke about the importance of the role of cultural broker. In particular, interpreters referred to the prominent role that the ancestors and amafufunyana (i.e., a type of spirit possession) played in patients’ narratives. Below (Extract 7), the interpreter explained that she is able to explain to clinicians that when a patient is called by the ancestors to be a sangoma (i.e., a traditional healer), it does not mean they are ill. Instead it should be regarded as part of their culture:

Extract 7:

Researcher: Do you think that culture plays a role during the interpreting session?

Interpreter: Ja, like sometimes it can be because the patient he can talk things like amamfufunyana, the ancestors, okay that man want me to go for sangomas, I’m not crazy because they called me. It’s a calling. And I can explain now. Okay, she said no, it’s not so sick, it’s a calling, it’s a culture. All those things. Hmmm.

Gatekeeper

Interpreters at times took on the role of gatekeeper by controlling what information is conveyed to the clinician and patient. Interpreters did this by either responding to clinicians’ questions on behalf of patients or by responding to patients’ questions on behalf of clinicians (see Extract 8 below). Interestingly, this happened primarily in instances where interpreters were also health care workers. Clinicians and patients responded to these actions in no uncertain terms. Clinicians made it clear that they wanted patients to personally respond, while patients encouraged interpreters to convey their messages to clinicians:

Extract 8:

Dr. D: What are his plans for when he is going to go home?

Interpreter P: What did we say? I can’t remember. What did we say but …

[Clinician interjects]

Dr. D: No, no, I want to know from him [the patient] what is he planning.

Patient 11: I say my dear; I have a request to the doctor. Can she ask for a loan on my behalf from the government or whoever? They can make a request for me, then after completion of my traditional healer ritual, I can return the money.

Interpreter Y: They won’t be able to.

Patient 11: Tell her.

Dr. E: What is she saying?

Patient 11: Which brother was it that she spoke with, I don’t have a brother.

Interpreter Y: I don’t know.

Patient 11: Ask her.

Dr. E: What is she saying?

The role of gatekeeper did not feature as a theme during the semi-structured consultations; however, clinicians mentioned that they felt that the presence of the interpreter made them feel excluded and alienated (see Extract 9 below). Clinicians may be exasperated by interpreters’ actions as gatekeepers:

Extract 9:

Dr. C: Well for instance, the patient, because the translator is speaking her mother tongue she, or there is a tendency for her [the patient] to trust or believe the translator and place more trust with her. And she can be more comfortable with that person. So at times I felt that I was a party to their discussion, although I am leading this discussion technically. So there is a small bit of alienation there as an interviewer that I feel that I am a bit removed there.

Advocate

Our findings suggest that interpreters had conflicting desires when it came to the role of advocate. On the one hand they advocated for the clinicians, while on the other hand they advocated for patients, and often this happened within the same consultation. For example, in one of the consultations the interpreter, advocating for the patient, conveyed to the clinician that she was concerned about the patient’s change in mood. Later in the same consultation, the interpreter was abrupt with the patient and pressurized the patient to respond to the clinician’s question about his use of illegal substances. More specifically, the patient denied that he ever used illegal substances. Nevertheless, the interpreter (unknown to the clinician) told the patient that he was lying and that the clinician had the right to do a urine test, which would reveal the truth about his substance abuse. It may be that, in this instance, the interpreter wanted to provide the clinician with a “desirable” response (i.e., that the patient admitted he had a substance abuse problem).

The video recordings suggest that interpreters were more likely to advocate for clinicians than for patients. Below, we present a typical example of an interpreter advocating for the clinician. During the consultation, the clinician asked the patient to talk about her miscarriage. The patient, on three occasions, refused to talk about the topic. However, without the clinician’s knowledge, the interpreter persisted and urged the patient to talk about the topic. Only once the patient had become frustrated and raised her voice did the interpreter abandon the questioning and inform the clinician that the patient did not want to talk about the matter. See below for the dialogue that arose:

Extract 10:

Dr. B: I had a quick look at her file and I see that towards the end of last year she was pregnant. Ask her if this is true.

Interpreter R: She said she heard that you were pregnant last year?

Patient 1: That’s a past story and I don’t want to go over it again.

Interpreter R: Pardon sister.

Patient 1: [In a raised voice] That’s a past story and I don’t want to talk about it.

Interpreter R: The doctor wants to understand.

Patient 1: I said that’s a past story and [in a raised voice] I don’t want to repeat it.

Interpreter R: She said she can’t keep on saying the same thing. She already told you about this [about the pregnancy].

However, during the semi-structured discussions, interpreters emphasized their role as patient advocate, without reference to any desire to advocate for the clinician. One of the interpreters explained that when patients are admitted they are anxious and that it was her responsibility to comfort patients and provide them with reassurance. Another interpreter reported that she went out of her way to help patients and that she would offer a sandwich or cigarette to those patients that she perceived to be lonely. See below for the dialogue that arose:

Extract 11:

Researcher: So the more you explain to them why they are here, and why you are asking them [questions], the more they talk?

Interpreter R: Mm. They are open and free to talk. Most of the patients when they arrive are fearful. It’s part of their illness. They feel lonely here. Then we talk to them and inform them that they are safe here [in the hospital], “You came to rest here and once you are well you will go home. And your people can visit you, so don’t be scared.” And the more we interpret for them the more they become free. They came to love me and are fond of me. Sometimes, they will ask me, “Sisi bring me some tobacco tomorrow.” Then I say, “Ok, I will bring you some tobacco tomorrow.” Or maybe, since I work in the kitchen, then I will give the patient a sandwich. Then I will give it to that patient seems so lonely and quiet. And maybe I see the patient is improving, then I will ask the patient, “Come help me fold the laundry.” Then he will do this and while we are busy with the laundry we chat with one another. Then tomorrow when the doctor arrives then I will tell the doctor what the patient told me. This way, the doctor finds it easier to treat the patient. No, the doctors ask us how we see the patients, since the doctors know that we are around the patients while we clean and so. And because we are very fond of our patients. We have to keep an eye on them. Like the other day, I saw this patient playing with the water in the bathroom. So I went to check that the warm water doesn’t burn him or that he doesn’t drown.

Discussion

Our findings suggest that interpreters took on four roles during the interpreter-mediated psychiatric consultation: regulators of turn-taking, cultural brokers, gatekeepers, and advocates. The roles that informal interpreters took on in our study are largely similar to those reported in other studies conducted around the globe (Bischoff et al., 2012; Hsieh, 2008; Leanza, 2005).Our findings seem to suggest that informal interpreters, without any formal training, largely take on similar roles to those of formal interpreters who are trained (Hsieh, 2008).However, unlike the study conducted by Hsieh (2008), we did not find that interpreters in our study took on the role of professional, which is not surprising given that interpreters in our study were not professional interpreters. Below, we discuss the roles interpreters took on in our study in greater detail.

The task of regulating turn-taking is at the heart of community interpreting. Community interpreting is generally done in the consecutive mode (Bot, 2005). Unlike in the simultaneous mode of interpreting, where the interpreter listens to one speaker and interprets at the same time, in the consecutive mode the interpreter waits for one party to finish speaking, and then translates what they said in a separate turn. In this mode, the interpreter is required to interpret a dialogue between different parties and therefore shorter turns are helpful (Hale, 2007). Although it is common for the clinician to manage the flow of the conversation and regulate turn-taking during psychiatric consultations, this role is largely taken over by the interpreter during interpreter-mediated consultations, the interpreter being the only party that understands both the patient and clinician. Our findings suggest that interpreters were relatively successful in taking over this task by employing different turn-taking techniques, predominantly interjection.

However, confusion arose at times over who should regulate turn-taking, and time constraints—which characterize public mental health settings in South Africa—created frustration on the part of clinicians. Tribe and Thompson (2009) have observed that the slowed pace of communication in interpreter-assisted interactions in mental health settings could be advantageous, providing the mental health professional with time to think between conversational turns. However, their work was conducted in relation to psychotherapy. In this latter therapeutic mode, time constraints may not be as acute (or present at all), which is quite different to a hospital context and to clinical interviews. With the necessary training, informal interpreters may be more inclined to take on the role of professional within the interpreter-mediated psychiatric consultation. As suggested by Hsieh (2008), interpreters that take on the role of professional are more inclined to control the conversation. This may lead to more effective turn-taking, which may ultimately save time.

Our finding that interpreters took on the role of cultural broker is not surprising, given that they shared a first language as well as a cultural background with patients. Hunt and Swartz (2017), as well as Miller et al. (2005) and Penn and Watermeyer (2012), have acknowledged the relevance of the cultural broker model of interpreting in a multicultural context such as South Africa.

Cultural brokering involves the employment of interpreters to act not only as translators of language, but also as explainers of culture. Interpreters who act as cultural brokers are familiar with the culture of the client as well as the culture of the medical establishment (Westermeyer, 1993).

However, as Hunt and Swartz (2017) observe, along with authors such as Kaufert and Koolage (1984) and Rechtman (1997) before them, there are disadvantages to the cultural broker model. The interpreter’s interpretations of the client’s purportedly culturally informed meanings take precedence. The client has no way of conveying whether the interpreter’s interpretation of their utterances—their use of metaphor, for instance—is accurate. Heterogeneity within cultures is generally a positive thing, and so it is not reasonable to expect that interpreters and clients will share the same expressive nuances and culturally-informed frame of reference, even if they hail from the same cultural background (Hunt and Swartz, 2017; Palmer, 2004; Rechtman, 1997).

Hunt and Swartz write that “if the interpreter is supposed to be both ‘of the world’ of the client and the therapist, then they run the risk of diluting the client’s explanatory model when trying to make it appropriate to that of the therapist” (Hunt & Swartz, 2017, p. 4). Penn and Watermeyer (2012) have shown that when interpreters try to make the client’s story more acceptable to those working from a biomedical understanding of mental health or towards the biomedical ends of the consultation, the patient’s information is often lost and their lifeworld is only partially explained. In our study, we observed that this does take place, as interpreters act as gatekeepers of information, distilling or condensing meanings between parties. We discuss this in more depth presently.

A particularly notable contribution of the present work concerns the gatekeeping role which interpreters may play in South African medical settings. Gatekeepers are individuals who act as an interface between their own lifeworld and the lifeworld of others (Cranefield & Yoong, 2007; Tribe & Thompson, 2009). The gatekeeper is a boundary-spanning individual (Allen, 1967; Katz & Tushman, 1981). In our study, interpreters sometimes took on the role of gatekeeper by controlling what information was conveyed to the clinician from the patient, and vice versa. Interpreters did this by either responding to clinicians’ questions on behalf of patients or responding to patients’ questions on behalf of clinicians.

In a rare in-depth exploration of the role of interpreters as institutional gatekeepers, Davidson (2000) analyzed interpreting against the historical and institutional context within which it takes place, noting the institutional goals that frame patient–physician discourse. He concluded that interpreters do not simply act as “neutral” machines of semantic conversion, nor simply as cultural brokers, but are active participants in the process of diagnosis. Specifically, as diagnosis depends on the evaluation of social and medical relevance of patient utterances, the interpreter—in their deciding what to translate and what to omit, where to probe and where to condense—becomes a further institutional gatekeeper between patients with limited English proficiency and the health care service. It is interesting, then, that such gatekeeping by interpreters mainly occurred in our study in instances where interpreters were also health care workers. This perhaps indicates these health care workers’ greater inclination—in comparison to lay interpreters—is to “aid” the diagnostic process. It could be that the health care workers felt more empowered to aid the diagnostic process, compared to cleaners and security guards. It is possible that the latter group of lay interpreters do feel that they are part of the treating team, but are not empowered enough to try and aid the diagnostic process during the interviews. Another, perhaps more likely, explanation for this phenomenon, however, could be that, as health care workers, the interpreters are more acutely aware of the time constraints. They may therefore take on the role of institutional gatekeeper as a means to save time.

Davidson (2000) makes an interesting point in this regard. If we accept that “the measure of the interpreter’s success may not be an abstract count of how ‘accurate’ they are, but rather the degree to which she allows, through her actions, the speakers first to negotiate and then to achieve their goals for the speech event in question,” then such gatekeeping is understandable, and—possibly—necessary (Davidson, 2000, p. 380). However, as Davidson (2000) is quick to point out, in a context like the medical one, the goals of the speech event—the diagnostic consultation or clinical interview—are not easily determined. We note that, in our study, these goals are often determined from the top down. As interpreters may be expected to be both of the world of the patient (culturally), but also aligned with the goals of the medical professional (who holds a position of power, with a rank above theirs in the hospital hierarchy), determining the goals of the exchange between patient, clinician, and interpreter is manifestly complex. Thus, gatekeeping will inevitably misalign with one individual’s goals for the exchange. We see in our data that interpreters—in the real life setting of the clinical interview—align themselves with the goals of the clinician, whilst explicitly—in interviews—aligning themselves with the patient. In his study, Davidson (2000) noted that the addition of the interpreter increased the patient’s difficulty in making themselves, and their agenda, the priority of the clinical interaction. In our findings, as two people were representing the world of the medical establishment (the clinician and the interpreter), and only one the world of the patients, this dynamic is likely to pervade.

As noted, we found that interpreters had conflicting desires when it came to the role of advocate, advocating for the clinicians and for patients. Despite the fact that clinical transcripts revealed interpreters as advocating for the clinicians, and at times attempting to persuade patients to provide information that the interpreter believed valuable to the clinician, during the semi-structured discussions, interpreters emphasized their role as patient advocate, without reference to any desire to advocate for the clinician. This can be attributed to the above-discussed power dynamics, which are brought to bear on interpreters in medical settings. Nevertheless, as mentioned by one of the interpreters, much of the work of the interpreter occurs outside of the psychiatric consultation, where the interpreter assists the patient in making sense of his world. This supports Leanza’s (2005) suggestion that interpreters take on the role of integration agent.

It should be noted that it is not within the scope of our study to assess the influence of clinician and interpreter demographic characteristics on the roles taken on by interpreters. Nevertheless, our study provides evidence that interpreter roles are multifunctional interpreter within real-life interpreter-mediated psychiatric consultations.

Conclusion

The use of informal interpreters is a reality, and will likely remain common practice in resource-constrained mental health care settings. Therefore, clinicians must be aware of the impact it has on psychiatric consultation and patient-centered care. Our findings suggest that during the interpreter-mediated psychiatric consultations, despite interpreters and clinicians having the patient’s best interests at heart, it is the patient’s voice that becomes lost while the clinician and interpreter negotiate the roles played by each party, with the latter more likely to take center stage when lay interpreters are used. However, perhaps the voice of the patient is likely to take prominence outside the psychiatric consultation, where interpreters continue their work as patient advocates.

Although informal interpreters and clinicians work under immense time pressure and high workloads, we recommend that they meet with interpreters prior to each interpreter-mediated psychiatric consultation to discuss roles, responsibilities, and objectives. We also advise that brief debriefing sessions be held after each interpreter-mediated psychiatric consultation. Furthermore, it is essential for clinicians and interpreters to receive some form of basic training relating to interpreting techniques and models to prevent the patient’s voice from being lost.

Future evidence-based studies should assess the effectiveness of training programs focused on informal interpreters and clinicians working in multi-linguistic resource constrained settings. More specifically, we propose a randomized control trial exploring the effectiveness of interpreter training programs on improving patient and services provider satisfaction as well as clinical outcomes.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Sanja Kilian, PhD, is a research scientist at the Department of Psychiatry in the Faculty of Medicine and Health Sciences at Stellenbosch University, South Africa. Dr. Kilian’s doctoral studies focused on language barriers and their impact on mental health service delivery. After her doctoral studies, she completed a postdoctoral fellowship in psychosis studies. She is currently researching environmental risk factors for schizophrenia. Her published works focus on diagnostic accuracy in multi-linguistic mental health settings, the impact of childhood trauma on biological and clinical outcomes, cognitive neuroscience, treatment outcomes, and remission and relapse.

Leslie Swartz is a clinical psychologist and a distinguished professor of psychology at Stellenbosch University, South Africa. His work focuses on transcultural psychiatry and on questions of disability rights and access in low-income contexts. He has been researching questions of language and access to mental health care in South Africa since the 1980s. His most recent book (coedited with Brian Watermeyer and Judith McKenzie) is The Palgrave Handbook of Disability and Citizenship in the Global South (Palgrave Macmillan, 2018).

Xanthe Hunt has a PhD in psychology, and is a Senior Researcher at the Institute for Life Course Health Research in the Department of Global Health at Stellenbosch University, South Africa. Her work, across five African countries, focuses on the systemic influences on child health, as well as the sexual and reproductive health of persons with disabilities. Her research spans disability studies, public health, monitoring and evaluation of early childhood interventions, and academic communication.

Ereshia Benjamin, MA Clin Psyc, is a health care practitioner, specializing as a clinical psychologist. She is also a senior lecturer at the University of Cape Town in South Africa, and is an active member of the research and teaching faculty of the Department of Psychiatry and Mental Health.

Bonginkosi Chiliza is an associate professor/chief specialist and head of the Department of Psychiatry at the University of KwaZulu-Natal, South Africa. His research interests include schizophrenia, consultation-liaison psychiatry, health services, and medical education. He is the president of the South African Society of Psychiatrists.

ORCID iDs

Leslie Swartz

Xanthe Hunt

References

Allen

(1967). Communications in the research and development laboratory. Technology Review, 70(1), 31–37.

Bot

(2005). Dialogue interpreting in mental health. Rodopi.

Bowen

(2001). Language barriers in access to health care. Health Canada. https://www.canada.ca/en/health-canada/services/health-care-system/reports-publications/health-care-accessibility/language-barriers.html

Cranefield

Yoong

(2007). The role of the translator/interpreter in knowledge transfer environments. Knowledge and Process Management, 14(2), 95–103. https://onlinelibrary-wiley-com-443.web.bisu.edu.cn/doi/abs/10.1002/kpm.271

Bischoff, A., Kurth, E., & Henley, A. (2012). Staying in the middle: A qualitative study of health care interpreters’ perceptions of their work. Interpreting, 14(1), 1–22.

Davidson

(2000). The interpreter as institutional gatekeeper: The social‐linguistic role of interpreters in Spanish‐English medical discourse. Journal of Sociolinguistics, 4(3), 379–405. https://onlinelibrary-wiley-com-443.web.bisu.edu.cn/doi/abs/10.1111/1467-9481.00121

Drennan

Swartz

(1999). A concept over-burdened: Institutional roles for psychiatric interpreters in post-apartheid South Africa. Interpreting, 4(2), 169–198.

Drennan

Swartz

(2002). The paradoxical use of interpreting in psychiatry. Social Science & Medicine, 54(12), 1853–1866.

Flood

Rohloff

(2018). Indigenous languages and global health. The Lancet Global Health, 6(2), e134–e5. http://doi.org/10.1016/S2214-109X(17)30493-X

10.

Flores

(2006). Language barriers to health care in the United States. New England Journal of Medicine, 355(3), 229–231. https://doi.org/10.1056/NEJMp058316

11.

Friedland

Penn

(2003). Conversation analysis as a technique for exploring the dynamics of a mediated interview. International Journal of Language and Communication Disorders, 38(1), 95–111. https://doi.org/10.1080/13682820304811

12.

Hale

(2007). Community interpreting. Springer.

13.

Hsieh

(2008). “I am not a robot!”: Interpreters’ views of their roles in health care settings. Qualitative Health Research, 18(10), 1367–1383.

14.

Hsieh

H. F.

Shannon

S. E.

(2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277–1288.

15.

Hunt

Swartz

(2017). Psychotherapy with a language interpreter: Considerations and cautions for practice. South African Journal of Psychology, 47(1), 97–109.

16.

Jacobs

E. A.

Shepard

D. S.

Suaya

J. A.

Stone

E. L.

(2004). Overcoming language barriers in health care: Costs and benefits of interpreter services. American Journal of Public Health, 94(5), 866–869. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1448350/

17.

Katz

Tushman

(1981). An investigation into the managerial roles and career paths of gatekeepers and project supervisors in a major R & D facility. R&D Management, 11(3), 103–110.

18.

Kaufert

J. M.

Koolage

W. W.

(1984). Role conflict among ‘culture brokers’: The experience of native Canadian medical interpreters. Social Science & Medicine, 8(3), 283–286.

19.

Kilian

Swartz

Chiliza

(2014). The potential consequences of informal interpreting practices for assessment of patients in a South African psychiatric hospital. Social Science & Medicine, 106, 159–167.

20.

Kilian

Swartz

Chiliza

B. (

2015). Doing their best: Strategies used by South African clinicians in working with psychiatric inpatients across a language barrier. Global Health Action, 8, 28155.

21.

Leanza

(2005). Roles of community interpreters in pediatrics as seen by interpreters, physicians and researchers. Interpreting, 7(2), 167–192.

22.

Levin

M. E.

(2006). Language as a barrier to care for Xhosa-speaking patients at a South African paediatric teaching hospital. South African Medical Journal, 96(10), 1076–1079. http://hdl.handle.net/10520/EJC68595

23.

Mason

(2014). Dialogue interpreting (Vol. 5). Routledge.

24.

Miller

K. E.

Martell

Z. L.

Pazdirek

Caruth

Lopez

(2005). The role of interpreters in psychotherapy with refugees: An exploratory study. American Journal of Orthopsychiatry, 75(1), 27. https://doi.org/10.1037/0002-9432.75.1.27

25.

Palmer

T. G.

(2004). Globalization and culture: Homogeneity, diversity, identity, liberty. Liberal Institute of the Friedrich Naumann Foundation. http://www.fnf.org.ph/downloadables/Globalization%20and%20Culture.pdf

26.

Peräkylä

(2005). Analyzing talk and text. In Denzin

N.K.

Lincoln

Y.S.

(Eds.), Handbook of qualitative research (3rd ed., pp. 869–886). Sage Publications Inc.

27.

Penn

Watermeyer

(2012). Cultural brokerage and overcoming communication barriers: A case study from aphasia. In Gavio

C. B. L.

(Ed.), Coordinating participation in dialogue interpreting (pp. 269–298). John Benjamins.

28.

Rechtman

(1997). Transcultural psychotherapy with Cambodian refugees in Paris. Transcultural Psychiatry, 34(3), 359–375. https://doi.org/10.1177/136346159703400305

29.

Shiu-Thornton

Balabis

Senturia

Tamayo

Oberle

(2007). Disaster preparedness for limited English proficient communities: Medical interpreters as cultural brokers and gatekeepers. Public Health Reports, 122(4), 466–471. doi:10.1177/003335490712200407

30.

Tribe

Thompson

(2009). Opportunity for development or necessary nuisance? The case for viewing working with interpreters as a bonus in therapeutic work. International Journal of Migration Health and Social Care, 5(2), 4–12. http://hdl.handle.net/10552/550

31.

Westermeyer

J. J.

(1993). Cross-cultural psychiatric assessment. In Gaw

A. C.

(Ed.), Culture, ethnicity and mental health (pp. 125–134). American Psychiatric Press.